Supplement of
he Gardens’ Bulletin
Singapore
VOL. 70 (Supplement 1) 2018
ISSN 0374-7859
Hydrology and biodiversity
of
Nee Soon freshwater swamp forest
Singapore Botanic Gardens
THE GARDENS’ BULLETIN SINGAPORE
The Gardens 1 Bulletin Singapore is a peer-reviewed journal publishing original papers and
reviews on a wide range of subjects: plant taxonomy (including revisions), phytogeography,
floristics, morphology, anatomy, as well as horticulture and related fields, such as ecology and
conservation, with emphasis on the plant life of the Southeast Asian-Pacific region.
Dr Geoffrey Davison Dr Cai Yixiong
(Supplement Editor) (Supplement Editor)
Dr David J. Middleton Dr Jana Leong-Skornickova
(Editor-in-Chief) (Managing Editor)
Felicia Tay Dr Daniel C. Thomas Christina Soh
(Graphics Editor) (Copy Editor) (Business Manager)
Editorial Advisory Board
Professor Sir Peter Crane
Yale University
U.S.A.
Dr Rogier P.J. de Kok
Royal Botanic Gardens, Kew
U.K.
Dr W. John Kress
National Museum of Natural History
Smithsonian Institution
U.S.A.
Dr Mark Hughes
Royal Botanic Garden Edinburgh
U.K.
Dr Kiat W. Tan
Gardens By The Bay
Singapore
Dr Nigel P. Taylor
Singapore Botanic Gardens
National Parks Board , Singapore
Dr Ian M. Turner
Royal Botanic Gardens Kew
U.K.
Dr Peter van Welzen
National Biodiversity Center
The Netherlands
Dr Jun Wen
National Museum of Natural History
Smithsonian Institution
U.S.A.
Professor Nianhe Xia
South China Institute of Botany
P.R. China
Instructions for contributing authors
(more details in http://www.sbg.org.sg/research/instructions-for-authors.pdf)
Manuscripts should be sent by e-mail to <gardbullsing@gmail.com> or posted to
The Editor, Gardens’ Bulletin Singapore, Singapore Botanic Gardens, 1 Cluny Road,
Singapore 259569.
The language used is English. Prior to submission, it may be advisable to have manuscripts checked by
someone fluent or proficient in that language. Contributions must not have been published or be under
consideration elsewhere. There are no page charges. Submitted material cannot be returned.
Authors will receive a pdf reproduction of their contribution, gratis.
Cover letter ; submission and copyright In your cover letter or message, state clearly you are submitting
your manuscript (state title) for publication in the Gardens’ Bulletin Singapore. By submitting the
manuscript(s), authors agree that upon acceptance, copyright of the entire work becomes the property
of the Singapore Botanic Gardens, National Parks Board, Singapore. Authors must ensure all required
permission has been obtained in writing to publish any material they present; proof may be required.
Manuscripts are best submitted in electronic form (Microsoft Word 97-2003 preferred). Use Times New
Roman, 10-point font, 1.5-line spacing. Do not send original artwork unless requested by the Editor.
Number all pages serially including the title and abstract on the first.
(continued on inside back cover)
Supplement of
The Gardens’ Bulletin
Singapore
VOL. 70 (Supplement 1) 2018 ISSN 0374-7859
CONTENTS
G.W.H. Davison, Y. Cai, T.J. Li & W.H. Lim
Integrated research, conservation and management
of Nee Soon freshwater swamp forest, Singapore:
hydrology and biodiversity.1
E. Clews, R.T. Corlett, J.K.I. Ho, D.E. Kim, C.Y. Koh, S.Y. Liong,
R. Meier, A. Memory, S. J. Ramchunder, T.M. Sin, H.J.M.P. Siow,
Y. Sun, H.H. Tan, S.Y. Tan, H.T.W. Tan, M.T.Y. Theng,
R.J. Wasson, D.C.J. Yeo & A.D. Ziegler
The biological, ecological and conservation significance
of freshwater swamp forest in Singapore.9
C.T.T. Nguyen, R.J. Wasson & A.D. Ziegler
The hydro-geomorphic status of Nee Soon freshwater
swamp forest catchment of Singapore.33
K.Y. Chong, R.C.J. Lim, J.W. Loh, L. Neo,
W.W. Seah, S.Y. Tan & H.T.W. Tan
Rediscoveries, new records, and the floristic value
of the Nee Soon freshwater swamp forest, Singapore.49
J.K.I. Ho, R.F. Quek, S.J. Ramchunder, A. Memory,
M.T.Y Theng, D.C.J. Yeo & E. Clews
Aquatic macroinvertebrate richness, abundance and distribution
in the Nee Soon freshwater swamp forest, Singapore
71
W.H. Lim, T.J. Li & Y. Cai
Diversity of terrestrial snails and slugs
in Nee Soon freshwater swamp forest, Singapore
109
Y. Cai, C. Y. Ng & R.W. J. Ngiam
Diversity, distribution and habitat characteristics of dragonflies
in Nee Soon freshwater swamp forest, Singapore.123
S.N. Kutty, W. Wang, Y. Ang, Y.C. Tay, J.K.I. Ho & R. Meier
Next-Generation identification tools for Nee Soon
freshwater swamp forest, Singapore.155
Y. Sun, D.E. Kim, D. Wendi, D.C. Doan,
S.V. Raghavan, Z. Jiang & S.Y. Liong
Projected impacts of climate change on stream flow and groundwater
of Nee Soon freshwater swamp forest, Singapore.175
Y. Cai, G.W.H. Davison, L. Chan & S.Y. Liong
Conservation outputs and recommendations
for Nee Soon freshwater swamp forest, Singapore.191
Date of publication: 20 March 2018
Copyright ©
National Parks Board
Singapore Botanic Gardens
1 Cluny Road
Singapore 259569
Printed by Oxford Graphic Printers Pte Ltd
Gardens’ Bulletin Singapore 70 (Suppl. 1): 1-7. 2018
doi: 10.26492/gbs70(suppl.l). 2018-01
1
Integrated research, conservation and management
of Nee Soon freshwater swamp forest, Singapore:
hydrology and biodiversity
G.W.H. Davison 1 - 2 , Y. Cai 1 , T.J. Li 1 & W.H. Lim 1
National Biodiversity Centre, National Parks Board,
1 Cluny Road, 259569 Singapore
cai_yixiong @ nparks. gov. sg
2 Orchard House, Church Lane, Goodworth Clatford,
SP11 7HL, United Kingdom
ABSTRACT. The current paper acts as an introduction to nine following papers concerning the
hydrology and biodiversity of Nee Soon freshwater swamp forest. Freshwater swamp forest
is a threatened and overlooked ecosystem in the Southeast Asian region and in Singapore.
Characterised by predominantly mineral soils supporting forest that contains a subset of the flora
and fauna of lowland forest, but with the addition of important habitat specialists, freshwater
swamp forest is fed by an array of hydrological processes. As conservation management depends
on good hydrological and biological understanding, a research programme was designed to
tease out the roles of the various hydrological components. The background, management
concerns, and aims of the project are detailed.
Keywords . Habitat management, project design, project overview, site management
Introduction
Water relations are critical to the global occurrence of tropical evergreen rain forest
(Richards, 1952) and to the occurrence, zonation, species composition and ecology
of the various rain forest formations such as cloud forest, mangroves and peat swamp
forest (e.g. Whitmore, 1984). One of the lesser known tropical forest formations
occurring in all three major tropical regions (the Neotropics, Africa, and Asia) is
freshwater swamp forest.
Freshwater swamp forest is characterised as forest growing on mineral soils
and periodically flooded by fresh water that originates not only from rain; additional
sources can include groundwater fluctuations, spill-over of floodwater from adjacent
rivers and streams, backflow and a range of other hydrological processes. The forest
occurs on soils with an organic content that results in less than 65% loss on combustion
(Whitmore, 1984), but this is a somewhat arbitrary as well as generalised distinction
from peat swamp forest. In practice, organic content varies through the soil profile
with highest levels in the surface humus.
2
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
The global and regional importance of freshwater swamp forest, and the national
significance of Nee Soon freshwater swamp forest in particular, are reviewed by
Clews et al. (2018). Historically, small areas of freshwater swamp forest are thought
to have been present in Singapore on the lower reaches of various small river and
stream systems, upstream of mangroves, and downstream of dryland forest (Corlett,
1991, 1992). The freshwater swamp forest within each catchment would have been
physically and biologically partly isolated from that in other catchments, resulting in
local differences in floral and possibly faunal composition.
Corner (1978) began studies of freshwater swamp forest in the Malay Peninsula,
including Singapore, around 1932. His work demonstrated the major ecological and
floristic features of this forest formation, and indicated floristic differences between
the forest at Jurong (now gone, in the vicinity of Jurong Lake Park) and that at Mandai
(now gone, but for the currently studied fragment at Nee Soon, in the vicinity of Upper
Seletar Reservoir). Turner et al. (1996), Ng & Lim (1992), and Lim et al. (2011) have
provided further information on freshwater swamp forest in Singapore and at Nee
Soon.
The area within Singapore that is likely to have been primevally under
freshwater swamp forest has been various estimated as 65 km 2 (O’Dempsey, 2014) to
74 km 2 (Corlett, 1991). Of this possibly half was in the catchment of the Kallang and
Singapore rivers, one third in the catchment of the Jurong and Pandan rivers, and the
remainder scattered in many tiny fragments along the middle courses of small streams
prior to them debouching into coastal mangroves.
The relatively intact patch of freshwater swamp forest in Nee Soon has long
been considered the most important area in Singapore for native aquatic fauna and
flora (Ng & Lim, 1992). Early estimates of its richness suggested that it contains 48%
of the primary freshwater fish, 71% of the amphibians, 28% of the reptiles and 34% of
the avian fauna of Singapore. Nee Soon is particularly well known for its importance to
crustaceans (Ng, 1997; Ng & Yeo, 2005); the freshwater crab Parathelphusa reticulata
is a global endemic to Nee Soon. The swamp also has the highest percentage of native
and threatened freshwater fish species on the island, as well as being the main (if not
the only) habitat of the aquatic plant Barclay a motleyi Hook.f. By 1992 it was also the
last refuge within Singapore of two mammals, the Raffles Banded Langur, Presbytis
femoralis femoralis, and the Cream-coloured Giant Squirrel, Ratufa affinis affinis. The
banded langur has gradually increased in numbers and expanded beyond Nee Soon
into other parts of the Central Catchment Nature Reserve, whereas the giant squirrel is
now thought to have become locally extirpated (Davison et al., 2008).
Study of the swamp forest has until recently been at a survey and discovery
phase. Research priorities have been primarily to establish detailed species lists, the
status of endangered species and the extent of buffer zones. There has been little
documentation of spatial differentiation within the swamp forest.
Nee Soon project overview
3
Management concerns
The Nee Soon freshwater swamp forest constitutes part of the Central Catchment
Nature Reserve, administered primarily by the National Parks Board. The Central
Catchment Nature Reserve covers approximately 3,100 hectares (31 km 2 ), of which
approximately 2,600 hectares are land area and 500 hectares are made up of the surfaces
of the MacRitchie, Upper Peirce, Lower Peirce, and Upper Seletar Reservoirs. The
land owner is the Public Utilities Board and the land manager is the National Parks
Board, but several other government organisations also have limited jurisdiction,
causing some complexity in the management of the reserve and the swamp.
Part of the lower catchment is occupied by two firing ranges under the
management of the Ministry of Defence and an old disused firing range, now reverted
to secondary forest, once existed 0.8 km to the southeast. A water supply pipeline,
partly above ground and partly below, runs through the forest, with a grassy side-table
for maintenance.
Public access to the nature reserve is limited to designated trails, none of
which intrudes into the area of freshwater swamp forest. Visits to the freshwater
swamp forest, whether by scientists, educational groups or individuals, are managed
by permits. Continuous patrols are not feasible, but legal action against those who
infringe regulations can be taken under the Parks and Trees Act (2006).
Past research as well as management have tended to treat the Nee Soon freshwater
swamp forest catchment as a single unit without internal differentiation. In fact the Nee
Soon stream catchment covers approximately 479 hectares (4.79 km 2 ), but the area
exhibiting swampy conditions is much smaller. The swampy area is approximately 50
hectares (0.5 km 2 ) but cannot be defined exactly because every flood and every dry
period differs in extent and duration, streams may become silted, or change course.
Criteria for differentiating swampy areas do not exist at this microgeographical scale.
Singapore is deeply conscious of the potential impacts of climate change,
including its impacts on biodiversity, and of the role of vegetation as a first line of defence
in mitigating impacts (National Climate Change Secretariat, 2016). The National Parks
Board therefore has an important part in climate and microclimate mitigation through
the management of natural and planted vegetation. This is complementary to its role
in the conservation of biological diversity at ecosystem, community, species and
population levels, including national obligations under the Convention on Biological
Diversity. Recognising the significance of Nee Soon freshwater swamp forest as the
home of a large proportion of Singapore’s native flora and fauna, managers of the
Central Catchment Nature Reserve have been deeply committed to the conservation
of this unique ecosystem.
Management initiatives and responses are constrained by shortage of technical
information and the granularity of the information. A biodiversity survey of the Bukit
Timah and Central Catchment Nature Reserves (covering Nee Soon) in 1993-1997
4
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
(Chan & Corlett, 1997; Ng, 1997) included detailed stream mapping (the original
hand-plotted maps are kept on file), but longitudinal information on water quality
and hydrological changes, and detailed spatial differentiation of plant and animal
communities within the Nee Soon stream catchment have been very limited. This
has resulted in uncertainties about the existence, magnitude, extent and rate of any
changes.
In order to obtain high quality information that would enable the National
Parks Board to fulfil its statutory obligations of biodiversity conservation, funds were
secured to carry out a multidisciplinary study on the hydrology and biodiversity of Nee
Soon freshwater swamp forest.
Scoping and project design
An initial Phase 1 of the project was conducted from January 2011 to March 2012.
The aims were:
i) To establish what we know about Nee Soon freshwater swamp forest in terms of
its ground and surface water environment and ecology;
ii) To characterise the hydrology, geology, topography and flora of the freshwater
swamp forest using measurements appropriate for the subsequent development
of maps and models;
iii) To develop a preliminary hydrological model (surface and groundwater flow);
iv) To conceptualise and test an ecohydrological model characterising
interdependence between groundwater flow and vegetation growth.
Phase 2 of the project was conducted from February 2013 to August 2016. The aims
were:
i) To establish the status of Nee Soon freshwater swamp forest in terms of
vegetation hydrology and aquatic biodiversity;
ii) To identify periodic flux in hydrology and key components of the aquatic
biodiversity;
iii) To develop more refined models that can confirm current conditions (water
balance, nutrient balance, acid flux, faunal distribution) and then test-trial
various management scenarios;
iv) To identify and assess the root causes of impacts, potential issues that may
threaten the hydrological and ecological integrity of the swamp, and management
elements to be addressed;
v) To propose recommendations for possible mitigation of long-term negative
impacts;
vi) To establish a viable, long-term monitoring programme and develop sampling
protocols to ensure continued protection and good management;
Nee Soon project overview
5
vii) To train agency staff in modelling, sampling methods and tools for monitoring;
viii) To deliver workshops on development and interpretation of the models’ outputs;
ix) To publish work on swamp forest ecology and the development of eco-
hydrologic models in international, peer-reviewed scientific journals
Seven teams were formed to conduct the work. They were:
i) Mapping and geospatial imagery team
ii) Field hydrology and geomorphology team
iii) Vegetation ecology team
iv) Faunal ecology team
v) Genomics team
vi) Ecohydrological modelling team
vii) NParks faunal team and project administration
Delivery of results
Each team produced a three-monthly technical report, consolidated by the project
leader into a full quarterly project report. Submission of the draft report to the National
Parks Board was followed by a quarterly meeting to discuss past progress and future
tasks. After agreement had been reached, each report was accepted with revisions.
Clews et al. (2018) provide a comprehensive review of the global, regional and
national significance of freshwater swamp forest. The fundamental geomorphological
characteristics and processes are described by Nguyen et al. (2018), as they have
profound implications for the hydrology and biodiversity of Nee Soon. Chong et al.
(2018) outline some floristic and taxonomic outputs from the project, based on surveys
in and beyond 40 quadrats distributed through the drier and wetter areas of Nee Soon.
Ho et al. (2018) describe the aquatic macrofauna. Lim et al. (2018) list the molluscs of
Nee Soon, and Cai et al. (2018a) the odonates, in the terrestrial and aquatic domains.
Kutty et al. (2018) describe the use of next generation sequencing to forge the links
between field identifications and image databases of the freshwater swamp forest flora
and fauna. Sun et al. (2018) use numerical modelling to describe some of the projected
impacts of climate change on stream flow and groundwater conditions. Finally, Cai et
al. (2018b) summarise some of the main findings and reco mm endations arising from
the entire project.
ACKNOWLEDGEMENTS. We would like to thank staff and students, past and present, at the
Tropical Marine Sciences Institute and the Departments of Biological Sciences and Geography,
National University of Singapore, and colleagues in the National Parks Board who have been
very helpful in providing support, helping to acquire permits, determining work schedules, and
have worked tirelessly from the beginning of the project. Dr Tan Puay Yok was instrumental
6
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
in beginning Phase 1 of the project. We thank the Ministry of Defence for permitting research
teams to access areas around the firing ranges. Colleagues from the Public Utilities Board
have been very helpful and cooperative in providing essential hydrological data required for
the numerical eco-hydrological model, as well as practical information on water management.
Our gratitude goes to the Ministry of Finance and the National Parks Board for funding,
without which it would have been impossible to carry through this project.
References
Cai, Y., Ng, C.Y. & Ngiam, R.W.J. (2018a). Diversity, distribution and habitat characteristics
of dragonflies in Nee Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore
70 (Suppl. 1): 123-153.
Cai, Y., Davison, G.W.H., Chan, L. & Liong, S.Y. (2018b). Conservation outputs and
recommendations for Nee Soon freshwater swamp forest, Singapore. Gard. Bull.
Singapore 70 (Suppl. 1): 191-217.
Chan, L. & Corlett, R.T. (eds) (1997). Biodiversity in the Nature Reserves of Singapore.
Proceedings of the Nature Reserves Survey Seminar, Singapore, December 1997. Gard.
Bull. Singapore 49(2): i-iv, 147-425.
Chong, K.Y., Lim, R.C.J., Loh, J.W., Neo, L., Seah, W.W., Tan, S.Y. & Tan, H.T.W. (2018).
Rediscoveries, new records, and the floristic value of the Nee Soon freshwater swamp
forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 49-69.
Clews, E., Corlett, R.T., Ho, J.K.I., Koh, C.Y., Liong, S.Y., Memory, A., Ramchunder, S.,
Siow, H.J.M.P., Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W., Theng, M.T.Y. & Yeo,
D. C.J. (2018). The biological, ecological and conservation significance of freshwater
swamp forest in Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
Corlett, R.T. (1991). Vegetation. In: Chia, L.S., Ausafur Rahman & Tay, D.B.H. (eds) The
Biophysical Environment of Singapore, pp. 134-154. Singapore: Singapore University
Press.
Corlett, R.T (1992). The Ecological Transformation of Singapore, 1819-1990. J. Biogeogr.
19(4): 411-420.
Corner, E.J.H. (1978). The freshwater swamp-forest of south Johore and Singapore. Gard. Bull.
Singapore, Suppl. 1. Singapore: Botanic Gardens, Parks & Recreation Department.
Davison, G.W.H., Ng, P.K.L. &Ho, H.C. (eds) 2008. The Singapore Red Data Book- Threatened
Plants & Animals of Singapore, 2nd ed. Singapore: Nature Society (Singapore).
Ho, J.K.I., Quek, R.F., Ramchunder, S.J., Memory, A., Theng, M.T.Y., Yeo, D.C.J. & Clews,
E. (2018). Aquatic macro invertebrate richness, abundance and distribution in the Nee
Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 71-108.
Kutty, S.N., Wang, W., Ang, Y., Tay, Y.C., Ho, J.K.I. & Meier, R. (2018). Next-Generation
identification tools for Nee Soon freshwater swamp forest, Singapore. Gard. Bull.
Singapore 70 (Suppl. 1): 155-173.
Lim, K.K.P, Yeo, D.C.J. & Ng, P.K.L. (2011). Nee Soon Swamp Forest. In: Ng, P.K.L., Corlett,
R.T. & Tan, H.T.W. (eds) Singapore Biodiversity: an encyclopedia of the natural
environment and sustainable development , pp. 54-55. Singapore: Editions Didier Millet
in association with Raffles Museum of Biodiversity Research.
Lim, W.H., Li, T.J. & Cai, Y. (2018). Diversity of terrestrial snails and slugs in Nee Soon
freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 109-121.
Nee Soon project overview
7
National Climate Change Secretariat (2016). Singapore’s Climate Action Plan: Take action
today, for a carbon-efficient Singapore. Singapore: NCCS, Prime Minister’s Office,
Singapore. Available at www.nccs.gov.sg.
Ng, P.K.L. (1997). The conservation status of freshwater prawns and crabs in Singapore with
emphasis on the Nature Reserves. Gard. Bull. Singapore 49: 267-272.
Ng, P.K.L. & Lim, K.K.P. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquat. Conserv. Mar. Freshw. Ecosyst. 2(3): 255-266.
Ng, P.K.L. & Yeo, D.C.J. (2005). Malaysian freshwater crabs: conservation prospects and
challenges. In: Chua, L.S.L., Kirton, L.G. & Shaw, L.G. (eds) Status of biological
diversity in Malaysia and threat assessment of plant species in Malaysia. Proceedings
of the Seminar and Workshop, 28-30 June 2005, pp. 95-120. Kepong: Forest Research
Institute Malaysia.
Nguyen, C.T.T., Wasson, R.J. & Ziegler, A.D. (2018). The hydro-geomorphic status of Nee
Soon freshwater swamp forest catchment of Singapore. Gard. Bull. Singapore 70
(Suppl. 1): 33-48.
O’Dempsey, T. (2014). Singapore’s changing landscape since c. 1800. In: Barnard, T.P. (ed.)
Nature Contained: Environmental Histories of Singapore, pp. 17-48. Singapore: NUS
Press.
Richards, P.W. (1952). The Tropical Rain Forest. Cambridge: Cambridge University Press.
Sun, Y., Kim, D.Y., Wendi, D., Doan, D.C., Raghavan, S.V., Jiang, Z. & Liong, S.Y. (2018).
Projected impacts of climate change on stream flow and groundwater of Nee Soon
freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 175-190.
Turner, I.M., Boo, C.M., Wong, Y.K., Chew, P.T. & Ibrahim, A. (1996). Freshwater swamp
forest in Singapore, with particular reference to that found around the Nee Soon Firing
Ranges. Gard. Bull. Singapore 48(1): 129-157.
Whitmore, T.C. (1984). Tropical Rain Forests of the Far East, 2 nd ed. Oxford and London:
Clarendon Press.
Gardens’ Bulletin Singapore 70 (Suppl. 1): 9-31. 2018
doi: 10.26492/gbs70(suppl.l). 2018-02
9
The biological, ecological and conservation significance
of freshwater swamp forest in Singapore
E. Clews', R.T. Corlett 2 , J.K.I. Ho 3 , D.E. Kim 1 , C.Y. Koh 3 , S.Y. Liong 1 , R. Meier 3 ,
A. Memory 1 , S.J. Ramchundei 11 , T.M. Sin 1 , H.J.M.P. Siow 5 , Y. Sun 1 , H.H. Tan 6 , S.Y.
Tan 6 , H.T.W. Tan 3 , M.T.Y. Theng 1 , R.J. Wasson 7 , D.C.J. Yeo 3 & A.D. Ziegler 4
‘Tropical Marine Science Institute, National University of Singapore,
18 Kent Ridge Road, 117229 Singapore
tmsec@nus.edu.sg
2 Xishuanbanna Tropical Botanical Garden, Chinese Academy of Sciences,
Menglun, Yunnan, RR. China
^Department of Biological Sciences, National University of Singapore,
14 Science Drive 4, 117543 Singapore
“^Department of Geography, National University of Singapore,
1 Arts Link, 117570 Singapore
5 National Biodiversity Centre, National Parks Board,
1 Cluny Road, 259569 Singapore
6 Lee Kong Chian Natural History Museum, National University of Singapore,
2 Conservatory Drive, 117377 Singapore
7 Lee Kuan Yew School of Public Policy, National University of Singapore,
469C Bukit Timah Road, 259772 Singapore
ABSTRACT. The Nee Soon stream drainage in the Central Catchment Nature Reserve is
virtually the last remaining fragment of primary freshwater swamp forest in Singapore. The
forest type has been poorly studied in the Southeast Asia. The hydrology, water quality, as well
as aquatic flora and fauna all have great theoretical and practical significance. The ecology and
management of the Nee Soon freshwater swamp forest are reviewed, with remarks on their
national, regional and global contexts. This review sets the scene for a three-year integrated
conservation and management study completed in 2016.
Keywords. Biodiversity, climatology, hydrology, nutrient cycle, research gaps, tropical wetlands
Introduction
Globally, freshwater swamp forests occur in Southeast Asia, Africa, and South America,
with the largest proportion in the Amazon basin (Richards, 1996). In Southeast Asia,
they are located throughout the region, often near large rivers such as the Mekong
and Chao Phraya in Thailand, and the Irrawaddy in Myanmar, and in many smaller
systems such as the Sedili rivers in Johor (Corner, 1978; Whitmore, 1984).
This unique forest formation is mostly restricted to the alluvial soil of flood
plains, often on the landward side of mangrove forests or in areas with a high water
table (Goltenboth et al., 2006). While freshwater swamp forests are often located in
areas with a wet climate, they are also found in seasonally drier regions such as in west
New Guinea and east Java (Whitmore, 1984).
10
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
In Southeast Asia, freshwater swamp forest is a rather understudied forest type,
mainly owing to its inaccessibility and the occurrence of insect-borne diseases within
them (Yamada, 1997). Other wetland habitats, such as mangrove and peat swamp forests,
have tended to receive more attention (Dudgeon, 2000). Nevertheless, the freshwater
swamp forests of Peninsular Malaysia were surveyed relatively comprehensively by
Corner (1978), and additional work has been done in Cambodia (Theilade et al., 2011)
and Singapore (Ng & Lim, 1992; Turner et al., 1996; Lim et al., 2011).
Freshwater swamp forest can be broadly characterised as forest that is subjected
to flooding with relatively mineral-rich fresh water (Whitmore, 1984). As tropical
freshwater swamp forest is a formation of tropical rainforest, several environmental
conditions are common between freshwater swamp forests and other tropical rainforest
formations (e.g. high humidity levels and solar irradiance; see Richards, 1996).
However, beyond these commonalities, conditions found within freshwater swamp
forests and other tropical rain forest formations can differ greatly. Flooding in inland,
freshwater swamp forests is usually semi-permanent, irregular, or seasonal. Water
depth also varies tremendously, ranging from a few centimetres to several metres.
These physical factors, acting individually or synergistically, impact the ecology of
freshwater swamp forests. Indeed, previous work by de Padua Teixeira et al. (2011)
concluded that drainage patterns were the most prominent factor in the spatial
organisation of plants in the swamp forests of Brazil. Despite the variable nature
of freshwater swamp forests, sufficient water is always present during the growing
season, which ensures that organisms adapted to living in water or waterlogged soils
flourish (Junk et al., 2011).
Freshwater swamp forests have several sources of water, including rain, rivers,
and groundwater, whereas peat swamp forests obtain their water solely from rain
(Richards, 1996; Goltenboth et al., 2006). The colour of the water in the freshwater
swamp forest is often an indication of the levels of plant matter present in the water
and soil.
Management of freshwater swamp forests
Freshwater swamp forest soils are relatively nutrient-rich, unlike ombrotrophic swamp
forests, which receive nutrients solely via rain (Yule & Gomez, 2009). In freshwater
swamp forests, nutrients and alluvial soils are subsequently deposited within the forest
via rain and water table fluctuations (Whitmore, 1984; Richards, 1996; Whitten et al.,
2000; Goltenboth et al., 2006). The nutrient-rich soils in freshwater swamp forests
have resulted in over-exploitation for agriculture, such as wetland rice cultivation
(Richards, 1996; Whitten et al., 2000; Corlett, 2009) and oil palm plantations (Yule &
Gomez, 2009). Indeed, Chokkalingam et al. (2007) reported that in southern Sumatra,
fire of varying intensities was utilized to clear the swamp forest for agricultural
purposes. The widespread and repeated fires there transformed a diverse and complex
habitat into a habitat consisting of uniform stands of fire-resistant Melaleuca L. species
thickets. Additionally, mismanagement of these ecosystems via extensive logging and
Global significance of freshwater swamp forest
11
conversion to agriculture has led to severe degradation and loss of ecological and
biological diversity (Rijksen & Peerson, 1991; Hansen et al., 2009; Yule, 2010).
The conceptual model presented in this review explores the direct and indirect
effects of hydrology, physico-chemistry, stream morphology, and vegetation on the
macroinvertebrate and fish co mm unities.
The following sections will explore and review the hydrology, physico-
chemistry, as well as both the aquatic flora and fauna in freshwater swamp forests in
Southeast Asia. Owing to the dearth of studies investigating freshwater swamp forests
specifically, the literature concerning peat swamp forests was also incorporated, as
peat swamps are the habitat most similar to freshwater swamp forests in Southeast
Asia.
Hydrology in freshwater swamp forests
How water flows is a major determinant of geomorphological, biological and bio-
geomorphological processes and functions within aquatic ecosystems (Poff et al.,
1997; Bunn & Arthington, 2002; Davidson et al., 2012; Fig. 1). By influencing
geomorphology, flow plays a major role in determining spatial and temporal benthic
community structure (Poff & Allan, 1995; Bunn & Arthington, 2002; Mim s & Olden,
2013). For example, Leigh & Sheldon (2009) found that hydrological connectivity
had a major effect on macroinvertebrate assemblages, with highly connected water
bodies displaying greater macroinvertebrate diversity than isolated water bodies,
which tended to have fewer diverse assemblages and were dominated by a handful
of taxa. Similar effects have been noted in tropical systems where wet or monsoon
season flooding has resulted in greater proportions of migratory species and changes in
community assemblages (da Silva et al., 2010). On a smaller scale, changes in physical
habitat caused by alterations in flow regime can increase habitat heterogeneity and
thereby increase species diversity (Downes et al., 1998; Bunn & Arthington, 2002).
Furthermore, a highly heterogeneous habitat is able to provide refugia for species
during periods of disturbance such as flooding or drought (Bunn & Arthington, 2002;
Negishi et al., 2002). In ecosystems where water availability can vary dramatically
between seasons, such refugia can become extremely important (Brown, 2003; Leigh
et al., 2010).
In some tropical systems there is pronounced seasonal variation in the hydrologic
regime between the monsoon and dry seasons (Douglas et al., 2005; Mitsch et al.,
2010), though with caveats for Singapore’s meteorology as described below. When
the monsoon season arrives, heavy rains can create flood pulses which integrate
terrestrial and aquatic systems (Davidson et al., 2012). Junk et al. (1989) stressed the
importance of seasonal flood pulses through effects such as over-bank flooding, which
strongly influences biological community structure through allochthonous inputs and
changing of the physical habitat (Douglas et al., 2005; Davidson et al., 2012). For
example, in Brazilian floodplains da Silva et al. (2010) found that turbid and anoxic
water conditions dominated in the high water season compared to low or falling water
12
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
seasons, which had relatively well oxygenated and low turbid waters. The differences
in seasonal hydrology impacted fish diversity, abundance and biomass, which were
significantly higher in falling and low water seasons than during the high water
season (da Silva et al., 2010). In tropical wetland forests similar changes have been
documented in macroinvertebrate communities with stream physico-chemistry as well
as insect assemblages, density and biomass reflecting fluctuations in seasonal rainfall
(Ramirez et al., 2006). How rainfall impacts tropical wetlands, however, is varied and
can depend on both regional and local biotic and abiotic factors.
The effect of flood pulses on tropical wetlands can vary because catchment
hydrology is influenced by the amount of rainfall and speed of the run-off (Page et
al., 2009; Yule, 2010). In tropical peat swamp forests the topography, biodiversity of
the forest and its microtopography can buffer the speed of run-off and thus mitigate
potential impacts further downstream (Page et al., 2009). On a microtopographical
or local scale buffering is partly caused by the presence of hollows, roots, buttresses
and pneumatophores (Page et al., 1999). However, what primarily influences rainfall
run-off and hydrology in a tropical peat swamp forest is how water flows laterally
through the peat and the hydraulic connectivity of the peat near to its surface (Page
et al., 1999). If the water table is consistently high, the amount of peat able to be
sequestered increases as well as the system’s water holding capacity (Page et al.,
1999). Alternatively, if the water table is low or if the swamp forest is being drained,
the speed of the run-off can become faster because of peat oxidation and subsidence
processes (Page et al., 1999, 2009). The impact that faster run-off can have on the
catchment varies from flooding downstream to scouring of the soil, which can have an
impact on water chemistry.
Water quality in freshwater swamp forests
Gasim et al. (2007) conducted a detailed physico-chemistry study in the Bebar river in
the Pahang peat swamp forests in Peninsular Malaysia where the authors documented
low dissolved oxygen (DO) and pH, whilst the average stream flow was estimated at
5 x 10 5 m 3 daily. Indeed, the pH recorded ranged from 3.53 to 4.55 whilst DO ranged
from 0.54 to 1.76 mg L b The high organic content and high biological decomposition
resulted in a high deoxygenation rate when compared with the reoxygenation rate (Das
6 Acharya, 2003; Gasim et al., 2007). Similarly, low levels of DO have been observed
at the Beriah swamp forest in Perak (1.21-2.14 mg L 1 ; Mashhor et al., 2004). Arecent
study by Gandaseca et al. (2015), determined the water quality of four rivers in peat
swamp forests in Sarawak, Malaysia. The authors also documented low levels of DO
(4.98-5.02 mg L' 1 ) and attributed this to the high levels of organic matter in the rivers.
Reported ranges of total dissolved solids (TDS) varied from 0.75 to 15.75
mg L 1 , whilst turbidity ranged from 1.5 to 17.15 NTU. Temporal investigations by
Ramirez et al. (2006) on a tropical wet forest documented a decrease in streamwater
pH throughout the year, from near neutral (pH >6.0) to near acidic (pH <4.5), while
N0 3 -N concentrations were high throughout the year and were independent of
Global significance of freshwater swamp forest
13
Fig. 1 . Conceptual diagram showing the effects of stream morphology, hydrology, vegetation
and physico-chemistry on each other and on the macroinvertebrate and fish community in
freshwater swamp forests, specifically in the Nee Soon freshwater swamp forest.
discharge. The authors identified that changes in pH were related to both streamwater
level and monthly rainfall. Indeed, the decrease in pH was related to the inundation
of the streams and an increase in overland flow (Ramirez et al., 2006). Furthermore,
Ramirez et al. (2006) noted that discharge and pH changed most during the year and
whilst pH decreased, stream discharge increased. The temporal variation in pH was
attributed to the increases in the concentration of humic acids suspended in the water
column during the wet season. Indeed, previous work by Winterbourn & Collier (1987)
observed low pH levels in many New Zealand streams and attributed this to the large
inputs of humic acids from surrounding watersheds. Alternatively, temporal variations
in conductivity were small and were independent of discharge but were related to
groundwater influence (Ramirez et al., 2006).
Aquatic flora in freshwater swamp forests
The environmental conditions of a tropical freshwater swamp forest are similar to those
of a tropical dryland rain forest, but in general, the tree canopy of the freshwater swamp
forest is lower than that in the lowland dipterocarp forest (Corner, 1978; Theilade et
al., 2011). The vegetation structure in freshwater swamp forests is often dependent on
14
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
the nutrients present in the water source and the flooding regime of the forest (Junk
et al., 2011). For example, the nutrient-rich varzea forests in the lower Amazonian
floodplains experience flooding caused by swollen rivers and have the highest species
richness among wetland forests in the world (Junk et al., 2011). Alternatively, the
vegetation of Brazilian blackwater igapo forests is dependent on heavy rains that
determine the forests’ flood height and duration. Igapo forests generally have fewer
tree species than varzea forests and few herbaceous plant species (Junk et al., 2011).
Furthermore, floating plants are rare or absent in this forest type. Freshwater swamp
forests are generally less diverse floristically if compared to dryland forests and, as a
result, are dominated by one or a few tree species (Corlett, 2009). Often the dominant
plant types are used to categorise freshwater swamp forests and the four main types
recognized in Southeast Asia are (1) mixed swamp forest; (2) Melaleuca species
(Myrtaceae) swamp forest; (3) Terminalia L. species (Combretaceae) swamp forest;
and (4) Campnosperma Thwaites species (Anacardiaceae) swamp forest (Goltenboth
et al., 2006).
Floristically, freshwater swamp forests are not easily distinguishable from
dryland forests at the taxonomic levels of family and genus (Whitmore, 1984). There
are few plant species restricted only to freshwater swamp forest ecosystems and
usually they tend to form clusters and have species-poor associations (Whitten et al.,
2000). The diverse assemblage of forest types are influenced by several environmental
factors, including the wide variation of soil content as well as the degree of water
inundation (Yamada, 1997).
Depending on the degree of water inundation, a thin layer of peat may be
present on the ground surface of freshwater swamp forests. However, the limited
decomposition rate of organic matter often results in the development of a peat layer
only a few centimetres thick (Whitmore, 1984). The slow decomposition rate is
due to high phenolic concentrations in the leaves, which can be up to three times
greater than those found in temperate forests (Coley & Barone, 1996). The high
concentration of phenols is thought to be a response to the high levels of mammalian
and invertebrate predation as well as fungal pathogens (Coley & Barone, 1996). The
slow decomposition of organic matter under anoxic conditions causes the release of
humic acid, which greatly lowers the pH of the water (Page et al., 1999; Goltenboth
et al., 2006; Yule, 2010; Posa et al., 2011). This then affects the floral composition
found in swamp forest streams and leads to specialisation of characteristics such as
pneumatophores (Whitmore, 1984).
The waterlogged nature of freshwater swamp forests creates soft, unstable, and
anoxic waterlogged soil that may have led to the evolution of special root adaptations
in freshwater swamp forest trees that are morphologically similar to those found in a
true mangrove forest (Corlett, 1986). Special root adaptations such as pneumatophores
are common within freshwater swamp forests and work by Corner (1978) noted that
pneumatophores typically occur in five forms. For example, in the genus Sonneratia
L.f., they grow as upright, elongated, conical pegs whereas in the species Lophopetalum
multinervium Ridl., they develop as erect planks (Corlett, 1986). Pneumatophores help
provide stability and aid in gas exchange in the anoxic soil conditions (Corner, 1978).
Global significance of freshwater swamp forest
15
Other adaptations include buttress roots, that provide stability in the unstable and soft
substrates. Furthermore, many trees have lenticellate bark which aids in gas diffusion
in anaerobic conditions (Whitten et al., 2000). Many of these root adaptations can be
found in Southeast Asia’s freshwater swamp forests including those in Singapore.
Aquatic fauna in freshwater swamp forests
Previous work by Whitten et al. (2000) has documented that the fauna of freshwater
swamp forests is as diverse as that found in lowland terra firma forests. However,
research in this area is still very lacking (Goltenboth et al., 2006). Nonetheless, Posa et
al. (2011) documented that approximately 23-32% of all species of mammals and birds
in Peninsular Malaysia and Borneo have been recorded from peat swamp habitats. The
proportions of snakes (7-18%) and amphibians (19-23%) are somewhat lower, but
nevertheless, the results collated by Posa et al. (2011) do show that peat swamp forests
provide habitats for a considerable proportion of the region’s fauna.
Additionally, freshwater swamp forests support a number of rare, specialized
and threatened species. Posa et al. (2011) found that 45% of mammals and 33% of birds
recorded in freshwater swamp forests had an IUCN Red List status of near threatened,
vulnerable or endangered. Additionally, Phillips (1998) documented the importance of
swamp forests in conserving primates such as proboscis monkeys (Nasalis larvatus )
and the Bornean banded langur ( Presbytis chrysomelas). Previous work from Johnson
et al. (2005) at Gunung Palung National Park in western Kalimantan, documented a
higher density of Bornean orang-utan nests and individuals than in lowland forest.
In addition, Cheyne et al. (2009) observed a number of endangered felids (e.g. the
flat-headed cat, Prionailurus planiceps; the Sunda clouded leopard, Neofelis diardi;
and the marbled cat, Pardofelis marmorata) within swamp forests, whilst Bezuijen et
al. (2001) documented swamp forests to be favoured habitat for the endangered false
gharial ( Tomistoma schlegelii).
These results already suggest that freshwater swamp forests are extremely
important for conservation, but it must also be remembered that there has been a
general bias towards charismatic mammalian species in biodiversity research (Clark
& May, 2002). Conversely, invertebrate, fish, amphibian and reptilian research has
usually been under-represented (see Wells & Yule, 2008; Yule, 2010), despite the fact
that these groups are often much more diverse than mammals, making up some of
the dominant animal groups in the forest (Clark & May, 2002). Thus, the importance
of freshwater swamp forests in preserving overall faunal diversity has in fact been
understated thus far.
Freshwater fish serve as a good example of the importance of freshwater swamp
forests to less well known groups. They have been documented as exhibiting extremely
high endemicity to swamp forests, up to the point that 33% of known freshwater
fish species are associated with peat swamps (Ng et al., 1994; Kottelat et al., 2006).
Additionally, Posa et al. (2011) found that out of 219 fish species collated from peat
swamps, 80 species are restricted to this ecosystem, while 31 species are point endemic
16
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
species found only in single locations. Furthermore, the critically endangered Betta
persephone, B. miniopinna and B. spilotogena are listed on the IUCN Red List as
highly threatened by extinction as a result of declines in area of occupancy, extent of
occurrence and quality of habitat. Additionally, there are 17 species of swamp forest
fish on the IUCN Red List classified as vulnerable or endangered, 12 of which are
point endemics (Posa et al., 2011). Comprehensive studies of freshwater swamp forest
fish have been carried out in only a few regions such as Thailand, central Sumatra,
Peninsular Malaysia and the Riau Archipelago (e.g. Ng et al., 1992; Tan & Tan, 1994;
Vidthayanon, 2002; Tan & Kottelat, 2009), and all indications point to even greater
fish diversity in unexplored locations.
Far less is known about the invertebrates of swamp forests, as most swamp
invertebrates have not been extensively studied and furthermore, organisms are rarely
identified to the species level. Johnson (1968) did note that Coleoptera, Hemiptera
and Diptera may be abundant and diversified in small blackwater pools but in most
blackwater habitats, freshwater macroinvertebrates are rather scarce and have poor
diversity, with Cheng & Fernando (1969) listing only six hemipteran species from
Malaysian blackwaters. Although there seem to be few if any invertebrate species
specific to or characteristic of swamp forest (e.g. Ng et al., 1992; Abang & Hill, 2006),
rotifer and decapod crustacean species have been found in these habitats (see Ng et
al., 1992). For instance, a total of 133 rotifer species were identified from five coastal
peat swamps on Phuket Island, Thailand (Chittapun et al., 2007), whilst Wowor et
al. (2009) found four species of the freshwater Macrobrachium prawns occurring in
acidic peat habitats.
General climatology of Singapore
Historical trends in precipitation
Nine rain gauges in Singapore (Fig. 2; Liong & Raghavan, 2004), selected for their
data comprehensiveness and relevance to Nee Soon freshwater swamp forest, provided
the daily rainfall data from 1961 until 2007 used for the present study. From some
of the stations the time series was incomplete. Table 1 provides the periods of data
availability for the nine stations considered.
Historical trends in temperature
The trends in daily temperature observations for Singapore reveal evidence of a
warming trend since 1970 that is consistent with broader evidence of global warming
and other temperature trend analyses in the region. Daily maximum and minimum
temperature data are available from four observation stations in Singapore, for varying
periods (shown in Table 4). The station-averaged absolute change values compared
to baseline 1961-1990 show the rise in surface temperatures since 1961-1990, as
recorded at all stations.
Global significance of freshwater swamp forest
17
IRE ORCHID
TZWGAH MET. STATION
EAYALE&AFIWET STATIOT
MACRfTCHIE RESEHVi
J.URONG INDUSTRIAL WATERWORKS
Fig. 2. Geographical locations of selected meteorological stations in Singapore.
Historical trends of wind
The strength and direction of the winds over Singapore are influenced by the monsoon
(Nieuwolt, 1981; National Environment Agency, 2007). Northeasterly winds prevail
during the Northeast Monsoon which occurs from December to early March, with
wind speeds sometimes reaching 30 to 40 km/h in January and February. During the
Southwest Monsoon (June to September), southeast to southwest winds prevail over
Singapore (with more consistent southwest winds over the Indian continent). Periods
between monsoon seasons (Pre-Southwest Monsoon: April and May, Post-Southwest
Monsoon: October and November) receive less wind. The highest extreme gust speeds
come from thunderstorms, whereas higher extreme mean wind speeds come from non¬
thunderstorm events (Choi, 1999).
The source of Fig. 3 (monthly mean wind speed (m/s) in Singapore) is the
National Enviromnent Agency (2012); see also Meterological Services Division
(2016). Overall, the wind speed at Changi is weaker than at the other three stations.
The wind speed during the Northeast Monsoon is stronger (by 1-3 m/s, according to
the site) than in other seasons. Maxima in the monthly mean wind speed tend to occur
in January and February during the Northeast Monsoon, while minima tend to occur
in April and May during the pre-Southwest Monsoon.
18
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
3,5
Month
Fig. 3. Monthly mean wind speeds (m sec 1 ) in Singapore (derived from National Environment
Agency, 2012).
Freshwater swamp forests: a Singapore perspective
Background
Prior to the establishment of modern Singapore, freshwater swamp forests were
estimated to have occurred at all upper river reaches of Singapore, covering about
74 km 2 or about 13% of the island (Yee et al., 2011; O’Dempsey, 2014). However,
with rapid urbanisation and industrialisation from the 1900s, vast forested areas were
cleared for agricultural and industrial purposes. In 1932, Corner(1978) studied one such
area—a patch of original swamp-forest in Jurong, which was completely transformed
into a pineapple plantation the year after. In this patch of more than 6 ha, Corner
(1978) found many plant species that only existed in Jurong but not in Mandai, or any
other Singapore swamp forests. This suggested that Jurong could have represented a
unique phytogeographical area (Corner, 1978; Yamada, 1997). However, no extensive
surveys of faunal species were conducted in the freshwater swamp forest in Jurong
before its clearance. Like this patch in Jurong, vast natural areas in Malaysia and
Singapore were or are often cleared for development before surveys can be conducted
on their existing flora and fauna (Ng & Lim, 1992).
The clearing of Jurong, Mandai, and Pulau Tekong swamp forests has left the
Nee Soon freshwater swamp forest as the last remnant of primary freshwater swamp
forests on Singapore Island (Corlett, 1992; Ng & Lim, 1992; Turner, 1996; Turner et
al., 1996; Yeo & Lim, 2011). Nee Soon is relatively intact but dominated by primary
and old secondary vegetation, covering about 87 ha of land (Corner, 1978; Ng &
Lim, 1992; Turner et al., 1996). Named after a wealthy Chinese-Peranakan merchant.
Table 1. Total Annual Precipitation for nine stations in Singapore and percentage change relative to 1961-1990.
Global significance of freshwater swamp forest
£
£
W5
a
# o
"’M
C3
xn
a
# o
"’M
©
o
g
Oh
DU
43 ^
U w
q ©
O
cm
I
? S H O
i 5 <u os ©
Ph -J ^ cm
cc
©
S ^H O
SO
©
DAD
SO
OS
o
*N
o
©
.2 _1
43 S
rj s©
ON
5 w
U
©
I
»ss
<2 ©3
Of I
£ <"■
© r-
*3 os
© i-H
H"S w
43
© -
fl C
Eh ^
OJD
S3
a*
44
©
CO I /-v
OHO
S3 © jH
© © ©
H H, o
©
OJD _
2 §
£ o ©
S3 £ £
•2 ^
© d
xn
nS
©
’C
©
Ph
CM
in
CM
in
CM
o
r-
Os
O
H
CM
CM
CM
CM
CM
CM
CO
©f
©r
©r
r-
CO
Os
H
CM
©f
©|-
CM
CM
CM
CM
CM
CM
oo
CO
SO
Os
SO
t"
H
CM
©r
in
CM
CM
CM
CM
co
co
OO
o
so
in
o
©r
CM
©r
r~
oo
CM
CM
CM
CM
Os
H
in
CM
O
in
oo
co
©f
in
r~
CM
CM
CM
CM
Os
O
CM
CM
SO
CM
Os
©r
in
SO
r~
CM
CM
CM
CM
Os
r-
in
SO
in
-—i
1—H
O
CO
in
Os
CM
CM
CM
CM
Os
CM
SO
Os
1—H
OO
Os
r~
CM
CO
in
t"
CM
CM
CM
CM
CO
in
Os
O
o
in
O
1—H
©1-
in
O'-
OO
CM
CM
CM
CM
in
r~
Os
CO
r-
in
OO
CM
co
in
SO
CM
CM
CM
CM
O
o
o
O
Os
s—H
s—H
y—i
Os
O
O
O
1
CM
1
CM
1
CM
1
i
l
O
l
SO
SO
Os
O
Os
Os
Os
o
CM
of
r-^
so
so
CO
o
so
o6 2
S3
O
Os
Os
<N
SO
OS
SO
Os
<D
©
• H
7j
cw
©
-O
O
00
in
-a
(0
©
a
o
o
(0
on
©
©
43
u
SO
©r
A
©r
oc
o
in
o
r-^
co
so
t ;
oi
co
so
©r
A
o
o
CM
I
SO
Os
O
o
CM
I
O
Os
Os
2001-2010 20.3 16.9 25.2 23.2 15.4 17.5 25.5 19.0 16.4 13.5
Table 2. Average Maximum Daily Precipitation for nine stations in Singapore and percentage change relative to 1961-1990.
20
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
in
in
t ;
d
A
in
m
cn
(N
in
y—i
y—i
in
p
cn
cn
A
A
in
On
(N
(N
(N
cn
1—H
y—i
y—i
so
in
<N
so
OS
i—H
SO
s— H
y—i
(N
cn
s- H
s - H
s— H
s— H
"xf
OS
OO
(N
SO
On
cn
Os
y—i
s— H
<N
T -f
y—i
s- H
y—i
i—i
OO
Os
Os
r-
o
in
o
d
(N
T -f
t"
OS
r-
On
<N
On
d
oo
d
y—i
(N
y—i
cn
s- H
s- H
y—i
—H
in
OO
CN
cn
ri
Os
CN
(N
(N
T -f
s—H
.—i
in
(N
in
OO
in
OO
o
<N
o
o
cn
y—i
cn
(N
<N
1-H
d
d
d
ON
y—i
(N
cn
in
s- H
y—i
y—i
<N
cn
CN
CN
SO
<N
A
A
y—i
O
(N
SO
s - H
y—i
o
o
o
o
Os
s—H
s—H
y—i
Os
O
O
O
i
(N
i
(N
i
<N
i
1
1
O
1
SO
SO
Os
O
Os
Os
Os
o
i
(N
oo o
OO 1-H
© cn
I I
r-
cn
oo
in so
o r~
o -h'
■ i
<N
<N
Os
in
(N
in
oo
CN
ri
A
so
in
os cn
cn so
CN
oo
<N
O
OS
Os
SO
Os
(U
c
• t-H
O
cd
X)
-a
<U
c3
a
o
o
(U
OO
e
cd
X
u
OS
in
d
A
(N
o
d
oo
r-
d
i
os oo
^1- SO
d ri
O Os
so SO
cn A
r-
o
d
in
o
so
oo
so
in
in
oo
oo
in
(N
oo
OS
oo
o
p
p
oo
oo
d
r-
Os
p
p
OO
cn
SO
in
(N
o
o
o
s—H
y—i
s—H
O
O
O
(N
i
(N
i
(N
i
1
O
1
SO
Os
O
Os
Os
o
—i
<N
Global significance of freshwater swamp forest
21
Table 3. List of stations and the availability of observed daily rainfall data.
Station name
Station
code
Availability
Periods with missing data
Paya Lebar
S6
1961-2007
MacRitchie Reservoir
S7
1961-2007
Ama Keng Telecom
Sll
1961-2007
Tengah
S23
1961-2007
Changi
S24
1967-2007
December 1969, October 1971,
November 1971
Seletar
S25
1967-2007
April 1, 1969 to March 31, 1970
St. James Complex
S31
1961-2007
February 1983
Jurong Industrial
Waterworks
S39
1964-2007
Singapore Orchids
Mandai
S40
1966-2007
Table 4. Average Surface Temperature for four stations in Singapore during the recent past and
absolute changes relative to 1961-1990. (Source: Liong, S.Y. & Raghavan, V.S., 2014)
Period
Station
Average
(°C)
1961-2010
Meteorological Stations (°C)
Tengah
(1971-2010)
Seletar
(1971-2011)
PayaLebar
(1961-2010)
Changi
(1984-2010)
1961-1990
27.3
27.7
27.7
27.3
25.8
1961-2010
27.5
27.7
27.9
27.6
27.2
1990-2010
27.7
27.7
28.1
28.1
27.6
2001-2010
27.7
27.6
28.1
28.4
27.8
Change compared to baseline 1961-1990 (°C)
1961-2010
0.17
0.02
0.21
0.33
1.43
1990-2010
0.42
0.05
0.42
0.80
1.86
2001-2010
0.33
-0.04
0.39
1.07
2.02
22
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Lim Nee Soon (1879-1936), the Nee Soon freshwater swamp forest is located in the
Central Catchment Nature Reserve, surrounded by the Executive Golf Course in the
north, Seletar Expressway and Old Upper Thompson Road in the east. Upper and
Lower Peirce reservoirs in the south, and the southern-most tributary of the Upper
Seletar Reservoir and the northern-most tributary of the Upper Peirce Reservoir in the
west (Yeo & Lim, 2011). This area incorporates areas that were previously part of the
Chan Chu Kang forest reserve (Corlett, 1992), which has allowed it to preserve much
old-growth vegetation. The Nee Soon freshwater swamp forest is thus the last remnant
of a larger swamp forest which was previously found along the entire Seletar drainage,
including the areas surveyed by Corner in Mandai (Comer, 1978; Ng & Lim, 1992;
O’Dempsey & Chew, 2013). It has probably only survived to the present day by virtue
of being both included in the Central Catchment Nature Reserve as well as being used
as a training area for the Singapore Armed Lorces and the presence of military shooting
ranges nearby (Ng & Lim, 1992). According to Corlett (2011), the freshwater swamp
forest is concentrated within several shallow valleys that drain towards the Seletar
River while the elevated areas between the valleys support patches of dryland forest.
Physical environment of Nee Soon freshwater swamp forest
The ground surface of most places in Nee Soon freshwater swamp forest is covered
by a shallow layer of peat and there is some variation in the microtopography of the
forest (Taylor et al., 2001; Corlett, 2011). Because the water table is so close to the soil
surface, periodic to semi-permanent flooding can be observed (Yeo & Lim, 2011). In
addition, depressions in the topography are often saturated with water, forming small
pools and slow-flowing streams (Turner et al., 1996; Taylor et al., 2001). Usually,
the clear water in the swamp forest is stained a dark-tea colour, a result of tannin
leaching from slowly decomposing plant matter under waterlogged soil conditions
(Yeo & Lim, 2011). Some of the larger streams have bed deposits of coarse-grained
sand that probably originates from the Triassic Period Bukit Timah granite formation
that constitutes most of the basement rock below the swamp (Taylor et al., 2001).
Soil analysis conducted at the Nee Soon freshwater swamp forest in the
mid-1990s revealed that the first 5 cm of the soil layer was rich in organic matter
(approaching 80% to 90% loss-on-ignition by mass), though the content decreased
rapidly beyond that depth to less than 50% loss-on-ignition by mass (Turner et al.,
1996). Hence, the lower layer of soil is not considered to be peat, which has more
than 90% loss-on-ignition by mass. The anaerobic and waterlogged conditions found
in the soil of the Nee Soon freshwater swamp forest reduce its rate of decomposition,
contributing to the richness in its organic matter (Turner et al., 1996).
Additionally, leaf litter in the Nee Soon freshwater swamp forest was found to
have a lower level of nutrients such as nitrogen, potassium and phosphorus compared
to the top 5 cm of soil. This could be caused by the draining of nutrients from the leaf
litter into the soil (Turner et al., 1996).
The pH of swamp forest streams and soil water is between 4.6 and 5.5, more
acidic compared to typical forest streams, and could become more acidic 5 cm below
the surface of the soil (Turner et al., 1996; Yeo & Lim, 2011). This could be a result of
Global significance of freshwater swamp forest
23
the thin layer of peat present in the swamp forest causing the release of humic acid, the
mechanisms of which were explained above.
Ecology of Nee Soon freshwater swamp forest
Through the collation of data from published articles—mostly from the Gardens’
Bulletin Singapore and The Raffles Bulletin of Zoology , as well as unpublished data
retrieved from the National Parks Board (NParks), the number of faunal species
recorded in the Nee Soon freshwater swamp forest was determined to be at least 346.
However, this is probably a highly conservative figure, and it is likely that this is only
a small proportion of the species present in the forest.
A series of field surveys was conducted in the 1990s, and their results published
in 1997 in the Gardens’ Bulletin Singapore (Chan & Corlett, 1997). This contributed
immensely to the inventory of biodiversity knowledge in Singapore, especially in the
Bukit Timah and Central Catchment Nature reserves. Among the groups surveyed were
vascular plants, fish, prawns, crabs, butterflies, stick and leaf insects, semi-aquatic
bugs, dragonflies and damselflies, and water beetles.
Based on our compilation of known faunal species in the Nee Soon freshwater
swamp forest, insects and birds make up the largest proportion of animals recorded.
Molluscs and annelids, on the other hand, make up only 3% of the total number of
species which have been documented in this forest patch.
While the Nee Soon freshwater swamp forest has lost much of its original
vertebrate fauna, it is still a very important site for the conservation of Singapore’s
remaining forests (Corlett, 1992). In fact, Yeo & Lim (2011) commented that Nee
Soon freshwater swamp forest supports the highest diversity of native freshwater
organisms in the country, reflecting its high conservation value and in particular
supporting freshwater fish, amphibians, reptiles, freshwater prawns and crabs and
bird species (Ng & Lim, 1992; Yeo & Lim, 2011). Many species, especially primary
freshwater fish, have their main populations in Singapore located in the Nee Soon
freshwater swamp forest and some others can be found nowhere else in Singapore, or
even globally, such as the swamp forest crab, Parathelphusa reticulata (Davison et al.,
2008; Cumberlidge et al., 2009; Lim et al., 2011). This makes the Nee Soon freshwater
swamp forest a vital refugium for many forest or swamp-adapted species in Singapore,
and gives it a very high conservation value (Ng & Lim, 1992; Lim et al., 2011).
Freshwater flora of the Nee Soon freshwater swamp forest
Parts of the Nee Soon forest were cleared during the late 19 th and early 20 th century
for rubber and pepper plantations (Turner et al., 1996). Other parts have been drained
or turned into reservoirs (Ng & Lim, 1992). The remaining vegetation consists of a
mixture of primary and secondary forest (Corner, 1978; Ng & Lim, 1992). Given its
history, the species composition and structure differ across the forest, suggesting a
mixed swamp forest. Many of the tree species in parts of the forest, such as Palaquium
xanthochymum (de Vriese) Pierre ex Burck and Xylopia fusca Maingay ex Hook.f. &
Thomson, exhibit adaptations to flooding, including buttress roots, prop roots, and
pneumatophores whereas some areas are dominated by plant species similar to those
24
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
of a dryland forest (Corner, 1978). In addition, Corner (1978) noted that the vegetation
of the Mandai Swamp Forest, of which the Nee Soon freshwater swamp forest is the
largest surviving remnant, was intermediate between a freshwater swamp forest and
a peat swamp forest. Corner (1978) further suggested that patches of substrate that
contain peat mats and decaying organic matter may transition to a full peat swamp
forest at some point in the future (Ng & Lim, 1992).
Studies are lacking on the aquatic flora of the Nee Soon freshwater swamp
forest; nevertheless, the investigations of Lok et al. (2009) documented the occurrence
of the locally rare forest water lily, which they called Barclaya kunstleri Ridl.
(Nymphaeaceae) but which is treated as a synonym of Barclaya motleyi Hook.f. by
Kiew (2015). This was noted as occurring within the forest at least until 1954 but now
it appears to be extinct at Nee Soon and only present in Singapore at Bukit Timah.
Aquatic fauna structure of Nee Soon freshwater swamp forest
The fauna of the Nee Soon freshwater swamp forest is rich and highly diverse. Insects
and birds make up the largest proportion of animals, whilst molluscs and annelids
make up a total of 3% of the species documented from Nee Soon. Additionally, Ng &
Lim (1992) and Yeo & Lim (2011) have documented that 71% of the amphibians, 28%
of the reptiles, 47% of freshwater prawn and 57% of freshwater crab species known
in Singapore still exist in the Nee Soon freshwater swamp forest. The forest also
contains the highest proportion of threatened native freshwater fish and crustaceans in
Singapore (Ng, 1997; Ng & Lim, 1997).
It is also the only area in Singapore where 11 out of 26 known species of native
freshwater fish can be found, including species such as the dwarf snalcehead ( Channa
gachua ), which was thought to be extinct in Singapore for 20 years before being
rediscovered in 1989 (Ng & Lim, 1989), as well as the black snakehead ( Channa
melasoma ), which was first recorded from Singapore only in 1990 (Ng & Lim, 1990).
Some other fish species in the Nee Soon freshwater swamp forest that cannot be found
elsewhere in Singapore include the spotted eel-loach (. Pangio muraeniformis ) and the
grey-banded loach (. Nemacheilus selangoricus ), while many other fish species known
from the Nee Soon freshwater swamp forest are also forest specialists that are only
found within such habitats.
The Nee Soon freshwater swamp forest is also a vital area for conservation of
freshwater invertebrates (Ng & Lim, 1992) in Singapore, as it was found to have the
highest diversity of water beetles in the country (Balke et al., 1997), whilst odonate
diversity was also found to be very high with eight species being exclusively found
in the Nee Soon freshwater swamp forest (Murphy, 1997). Additionally, the forest
also has the highest diversity of semi-aquatic bugs (Gerromorpha) in Singapore, with
83% of Singapore’s Gerromorpha species having been recorded in the swamp forest
(Yang et al., 1997). The forest is also a stronghold for freshwater decapods, with
multiple species of freshwater shrimp, such as Macrobrachium platycheles (which
was originally described from the area), having thriving populations there. Perhaps
the most important decapod found within the Nee Soon freshwater swamp forest is
the endemic swamp forest crab, Parathelphusa reticulata. It was described from the
Global significance of freshwater swamp forest
25
swamp almost thirty years ago (Ng, 1990) and today, the entire world’s population
of P. reticulata is still confined to the Nee Soon freshwater swamp forest. Hence,
if Nee Soon freshwater swamp forest were to be lost or modified, this species could
potentially be rendered globally extinct.
Conclusions
Nee Soon freshwater swamp forest constitutes Singapore’s last remaining patch of
primary freshwater swamp forest. From the viewpoint of species richness alone, this
makes the conservation of the Nee Soon freshwater swamp forest a top priority. Its
large number of plant and animal taxa, currently found nowhere else in Singapore, only
emphasises its conservation value. Finally, given that Nee Soon freshwater swamp
forest houses a large proportion of Singapore’s overall flora and fauna, conservation of
this habitat undoubtedly has larger-scale, positive effects for biodiversity conservation
in Singapore (Ng & Lim, 1992; Turner et al., 1996), accomplishing conservation of
biodiversity from species to landscape scales.
Owing to the nature of its ecosystem and drainage, the Nee Soon freshwater
swamp forest is extremely sensitive to external disturbances (Ng & Lim, 1992).
Furthermore, many of the species or groups of species found here are rather specialised
and thus, disturbance of the Nee Soon freshwater swamp forest and its surrounding
areas would pose a great threat to these unique groups. Therefore, it is important to
maintain Nee Soon freshwater swamp forest in its current state, as well as ensure
that it is not affected adversely by development. These are amongst the reasons for
conducting a long-term research project to intensify knowledge of the freshwater
swamp forest system (Davison et al., 2018).
Gaps in research
Singapore has been gradually building up knowledge on freshwater biodiversity,
which includes the area in the Nee Soon freshwater swamp forest (Kottelat & Whitten,
1996; Ng & Lim, 1997). The several biodiversity surveys conducted during the 1990s
in the nature reserves of Singapore further contributed to this store of information.
However, the knowledge of wildlife in Singapore is still inconsistent between groups.
Several surveys have been conducted to record vascular plant species in
Singapore, including specifically from freshwater swamp forests such as Nee Soon
(Corner, 1978; Turner et al., 1996). As such, vascular plants are comparatively well
studied. Among the different groups of plants, angiosperms are by far the most
extensively studied plant group. Nevertheless, information on diversity of plants is
probably still incomplete as demonstrated by continuing discoveries in Nee Soon and
its vicinity (Chong et al., 2018).
In terms of the understanding of fauna in Singapore, several groups have
been covered extensively. For example, much is known about the freshwater fish in
Singapore, and our understanding of this group is among the best in the region (Ng &
Lim, 1997). The surveys of the 1990s also improved our knowledge of several other
groups of animals.
26
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
However, even in highly studied areas, many species, such as freshwater
organisms, have only been found in recent years. This shows that streams in Singapore
are still insufficiently surveyed (Kottelat & Whitten, 1996).
Also, several groups of organisms present in Nee Soon freshwater swamp
forest are still underrepresented. While much more is known of the angiosperms of
Singapore, other groups including the macrofungi, fresh water algae and lichens are
understudied (Ng et al., 2011). Amongst the fauna, many forest and soil arthopods, as
well as various protozoan species, are highly unstudied (Ng et al., 2011).
Extinction in Singapore has proceeded at an alarming rate largely due to habitat
loss (Brook et al., 2003). Currently, only about 0.25% of land area in Singapore is
designated as protected nature reserves and 50% of our native species are harboured
in this area. Between 36% and 78% extinction has been inferred amongst various
taxonomic groups (Brook et al., 2003).
A forest cannot be conserved without deep knowledge of the majority of
its species, and more work should be devoted to surveying these groups. A deeper
understanding of the organisms present in the Nee Soon freshwater swamp forest
could greatly aid its conservation. Singapore has many specialist species of animals
and plants and a large proportion of which occur or occurred within the Nee Soon
freshwater swamp forest. We have lost some of these species already, and in order to
protect the rest, we require an improved understanding of their needs and interactions
within the ecosystem.
ACKNOWLEDGEMENTS. We would like to thank Dr Geoffrey Davison, Dr Cai Yixiong,
Sharon Chan, Chew Ping Ting, Li Tianjiao, Lim Weihao, and other colleagues of the National
Parks Board (NParks). We are also grateful to NParks for the permits to conduct fieldwork
in the Nee Soon Freshwater Swamp Forest (Permit no. NP/RP13-068-1), and the Ministry
of Defence (MINDEF) for permitting us to access areas around the firing ranges. Colleagues
from the Public Utilities Board (PUB) have been very helpful and cooperative in providing
essential hydrological data (e.g. rainfall, reservoirs’ water levels) required for the numerical
eco-hydrological model. The PUB’s participation in this project is greatly acknowledged and
appreciated. We are also appreciative of the help from our collegues, past and present, at the
Tropical Marine Sciences Institute and the Departments of Biological Sciences and Geography,
National University of Singapore, who have contributed to the body of work and knowledge¬
base on which this study was developed as well as for providing practical, logistical, and
technical support for the project team. This study forms part of the research project “Nee Soon
Swamp Forest Biodiversity and Hydrology Baseline Studies—Phase 2” funded by the Ministry
of Finance and NParks (National University of Singapore grant number R-347-000-198-490).
References
Abang, F. & Hill, D.S. (2006). The invertebrate fauna. In: Abang, D. & Das, I. (eds) The
Biodiversity of a Peat Swamp Forest in Sarawak, pp. 87-98. Kota Samarahan: Universiti
Malaysia Sarawak.
Global significance of freshwater swamp forest
27
Balke, M., Hendrich, L. & Yang, C.M. (1997). Water beetles (Insecta: Coleoptera) in the nature
reserves of Singapore. Gard. Bull. Singapore 49(2): 321-331.
Bezuijen, M.R., Webb, G.J.W., Hartoyo, P. & Samedi. (2001). Peat swamp forest and the
false gharial Tomistoma schlegelii (Crocodilia, Reptilia) in the Merang River, eastern
Sumatra, Indonesia. Oryx 35: 301-307.
Brook, B.W., Sodhi, N.S. & Ng, P.K.L. (2003). Catastrophic extinctions follow deforestation in
Singapore. Nature 424: 420-424.
Brown, B.L. (2003). Spatial heterogeneity reduces temporal variability in stream insect
communities. Ecol. Lett. 6(4): 316-325.
Bunn, S.E. & Arthington, A.H. (2002). Basic principles and ecological consequences of altered
flow regimes for aquatic biodiversity. Environ. Manage. 30(4): 492-507.
Chan, L. & Corlett, R.T. (eds) (1997). Biodiversity in the Nature Reserves of Singapore.
Proceedings of the Nature Reserves Survey Seminar, Singapore, December 1997. Gard.
Bull. Singapore 49(2): i-iv, 147-425.
Cheng, L. & Fernando, C.H. (1969). A taxonomic study of the Malayan Gerridae (Hemiptera:
Heteroptera) with notes on their biology and distribution. Orient. Insects 3: 97-160.
Cheyne, S.M., Morrogh-Bernard, H. & MacDonald, D.W. (2009). First flat-headed cat photo
from Sabangau peat-swamp forest, Indonesian Borneo. Cat News 51: 16.
Chittapun, S., Pholpunthin, P. & Segers, H. (2007). Diversity of rotifer fauna from five coastal
peat swamps on Phuket Island, southern Thailand. ScienceAsia 33: 383-387.
Choi, E.C.C. (1999). Extreme wind characteristics over Singapore - an area in the equatorial
belt. J. Wind Eng. Ind. Aerod. 83: 61-69.
Chokkalingam, U., Suyanto, Permana, R.P., Kumiawan, I., Mannes, J., Darmawan, A.,
Khususyiah, N. & Susanto, R.H. (2007). Community fire use, resource change, and
livelihood impacts - The downward spiral in the wetlands of southern Sumatra. Mitig.
Adapt. Strat. Gl. 12(1): 75-100.
Chong, K.Y., Lim, R.C.J., Loh, J.W., Neo, L., Seah, W.W., Tan, S.Y. & Tan, H.T.W. (2018).
Rediscoveries, new records, and the floristic value of the Nee Soon freshwater swamp
forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 49-69.
Clark, J.A. & May, R.M. (2002). Taxonomic bias in conservation research. Science 297: 191—
192.
Coley, P.D. & Barone, J.A. (1996). Herbivory and plant defenses in tropical forests. Annu. Rev.
Ecol. Syst. 27: 305-335.
Corlett, R.T. (1986). The mangrove understory - some additional observations. J. Trop. Ecol.
20(1): 93-94.
Corlett, R.T. (1992). The Ecological Transformation of Singapore, 1819-1990. J. Biogeogr.
19(4): 411-420.
Corlett, R.T. (2009). The Ecology of Tropical East Asia. New York: Oxford University Press.
Corlett, R.T. (2011). Terrestrial ecosystems. In: Ng, P.K.L., Corlett, R.T. & Tan, H.T. (eds)
Singapore Biodiversity: An Encyclopedia of the Natural Environment and Sustainable
Development , pp. 44-51. Singapore: Raffles Museum of Biodiversity Research,
Department of Biological Sciences, National University of Singapore.
Corner, E.J.H. (1978). The freshwater swamp-forest of south Johore and Singapore. Gard. Bull.
Singapore, Suppl. 1. Singapore: Botanic Gardens, Parks & Recreation Department.
Cumberlidge, N., Ng, P.K.L., Yeo, D.C.J., Magalhaes, C., Campos, M.R., Alvarez, F., Naruse,
T., Daniels, S.R., Esser, L.J., Attipoe, F.Y.K., Clotilde-Ba, F.-L., Darwall, W., Mclvor,
A., Baillie, J.E.M., Collen, B. & Ram, M. (2009). Freshwater crabs and the biodiversity
crisis - Importance, threats, status, and conservation challenges. Biol. Conserv. 142(8):
1665-1673.
28
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
da Silva, H.P., Petry, A.C. & da Silva, C.J. (2010). Fish communities of the Pantanal wetland in
Brazil - evaluating the effects of the upper Paraguay river flood pulse on baia Caiqara
fish fauna. Aquat. Ecol. 44(1): 275-288.
Das, J. & Acharya, B.C. (2003). Hydrology and assessment of lotic water quality in Cuttack
City, India. Water Air Soil Poll. 150: 163-175.
Davidson, T.A., Mackay, A.W., Wolski, P., Mazebedi, R., Murray-Hudson, M. & Todd, M.
(2012). Seasonal and spatial hydrological variability drives aquatic biodiversity in a
flood-pulsed, sub-tropical wetland. Freshwater Biol 57(6): 1253-1265.
Davison, G.W.H., Ng, P.K.L. & Ho, H.C. (eds). (2008). The Singapore Red Data Book-
Threatened Plants & Animals of Singapore, 2 nd ed. Singapore: Nature Society
(Singapore).
Davison, G.W.H., Cai, Y., Li, T.J. & Lim, W.H. (2018). Integrated research, conservation
and management of Nee Soon freshwater swamp forest, Singapore: hydrology and
biodiversity. Gard. Bull. Singapore 70 (Suppl. 1): 1-7.
de Padua Teixeira, A., Assis, M.A. & Luize, B.G. (2011). Vegetation and environmental
heterogeneity relationships in a Neotropical swamp forest in southeastern Brazil
(Itirapina, SP). Aquat. Bot. 94: 17-23.
Douglas, M.M., Bunn, S.E. & Davies, P.M. (2005). River and wetland food webs in Australia’s
wet-dry tropics: general principles and implications for management. Mar. Freshwater
Res. 56(3): 329-342.
Downes, B.J., Lake, P.S., Schreiber, E.S.G. & Glaister, A. (1998). Habitat structure and
regulation of local species diversity in a stony, upland stream. Ecol. Monogr. 68(2):
237-257.
Dudgeon, D. (2000). Riverine biodiversity in Asia - a challenge for conservation biology.
Hydrobiologia 418(1): 1-13.
Gandaseca, S., Rosli, N., Idris, M.H., Ahmed, O.H. & Pazi, A.M.M. (2015). Effects of
converting tropical peat swamp forest into oil palm plantation on water quality. Am. J.
Appl. Sci. 12: 525-532.
Gasim, M.B., Ismail, B.S., Toriman, E., Islam Mir, S. & Chek, T.C. (2007). A physico-chemical
assessment of the Bebar River, Pahang, Malaysia. Global J. Environ. Res. 1: 7-11.
Goltenboth, F., Langenberger, G. & Widmann, P. (2006). 17 - Special forest ecosystems. In:
Goltenboth, F., Timotius, K., Milan, P. & Margraf, J. (eds) Ecology of Insular Southeast
Asia, pp. 385-399. Amsterdam: Elsevier.
Hansen, M.C., Stehman, S.V., Potapov, P.V., Arunarwati, B., Stolle, F. & Pittman, K. (2009).
Quantifying changes in the rates of forest clearing in Indonesia from 1990 to 2005 using
remotely sensed data sets. Environ. Res. Lett. 4(3): 034001.
Johnson, A.E., Knott, C.D., Pamungkas, B., Pasaribu, M., & Marshall, A.J. (2005). A survey
of the orangutan ( Pongo pygmaeus wurmbii ) population in and around Gunung Palung
National Park, West Kalimantan, Indonesia based on nest counts. Biol. Conserv. 121(4):
495-507.
Johnson, D.S. (1968). Malayan blackwaters. In: Misra, R. & Gopal, B. (eds) Proceedings of
Symposium on Recent Advances in Tropical Ecology, part 1, pp. 303-310. Varanasi:
International Society for Tropical Ecology.
Junk, W.J., Bayley, P.B. & Sparks, R.E. (1989). The flood pulse concept in river-floodplain
systems. Can. Spec. Publ. Fish. Aquat. Sci. 106: 110-127.
Junk, W.J., Piedade, M.T.F., Schongart, J., Cohn-Haft, M., Adeney, J.M. & Wittmann, F.
(2011). A Classification of Major Naturally-Occurring Amazonian Lowland Wetlands.
Wetlands 31(4): 623-640.
Global significance of freshwater swamp forest
29
Kiew, R. (2015). Nymphaeaceae. In: Kiew, R., Chung, R.C.K., Saw, L.G. & Soepadmo, E.
(eds) Flora of Peninsular Malaysia, ser. 2, Seed Plants, vol. 5, pp. 207-219. Kepong:
Forest Research Institute Malaysia.
Kottelat, M. & Whitten, T. (1996). Freshwater biodiversity in Asia, with special reference to
fish. World Bank Technical Paper 343. Washington, D.C.: The World Bank.
Kottelat, M., Britz, R., Hui, T.H. & Witte, K.-E. (2006). Paedocypris, a new genus of Southeast
Asian cyprinid fish with a remarkable sexual dimorphism, comprises the world’s
smallest vertebrate. P. Roy. Soc. B. 273(1589): 895-899.
Leigh, C. & Sheldon, F. (2009). Hydrological connectivity drives patterns of macroinvertebrate
biodiversity in floodplain rivers of the Australian wet dry tropics. Freshwater Biol.
54(3): 549-571.
Leigh, C., Burford, M.A., Sheldon, F. & Bunn, S.E. (2010). Dynamic stability in dry season
food webs within tropical floodplain rivers. Mar. Freshwater Res. 61(3): 357-368.
Lim, K.K.P., Yeo, D.C.J. & Ng, P.K.L. (2011). Nee Soon freshwater swamp forest. In: Ng,
P.K.L., Corlett, R.T. & Tan, H.T.W. (eds) Singapore biodiversity: an encyclopedia of
the natural environment and sustainable development, pp. 52-63. Singapore: Editions
Didier Millet in association with Raffles Museum of Biodiversity Research.
Liong, S.Y. & Raghavan, V.S. (2014). Quarterly Reports on High Resolution Regional Climate
Change Impact Study, Vol I, as a part of the Coastal Inundation Risk Map Study for
Singapore. Submitted to the Building and Construction Authority, Govt, of Singapore.
Confidential (non-disclosure agreement).
Lok, A.F.S.L., Ang, W.F., Lee, S.M.L., Tan, H.H. & Tan, H.T.W. (2009). The status and
distribution of Barclay a (Nymphaeaceae) in Singapore. Nat. Singapore 2: 237-245.
Mashhor, M., Ahyaudin, A., Riley, J., Abu Hassan, A. & Asyraf, M. (2004). Tropical Peat
Swamps - Safe-Guarding a Global Natural Resource, 1 st ed. Penerbit Universiti Sains
Malaysia, Pulau Pinang.
Meteorological Services Division (2016). Climate of Singapore, http://www.weather.gov.sg/
climate-climate-of-singapore/. Accessed 17 Feb. 2016.
Mims, M.C. & Olden, J.D. (2013). Fish assemblages respond to altered flow regimes via
ecological filtering of life history strategies. Freshwater Biol. 58(1): 50-62.
Mitsch, W.J., Nahlik, A., Wolski, R, Bernal, B., Zhang, L. & Ramberg, L. (2010). Tropical
wetlands: seasonal hydrologic pulsing, carbon sequestration, and methane emissions.
Wetlands Ecol. Manag. 18(5): 573-586.
Murphy, D.H. (1997). Odonata biodiversity in the Nature Reserves of Singapore. Gard. Bull.
Singapore 49: 333-352.
National Environment Agency (2007). Climatology of Singapore. 2http://app.nea.gov.sg/cms/
htdocs/article.asp?pid=1088. Accessed 2 Dec. 2015 (now defunct).
National Environment Agency (2012). The weather and climate of Singapore, 2012. Singapore:
National Environment Agency.
Negishi, J.N., Inoue, M. & Nunokawa, M. (2002). Effects of channelisation on stream habitat in
relation to a spate and flow refugia for macroinvertebrates in northern Japan. Freshwater
Biol. 47(8): 1515-1529.
Ng, P.K.L. (1990). Parathelphusa reticulata spec, nov., a new species of freshwater crab from
blackwater swamps in Singapore (Crustacea: Decapoda: Brachyura: Gecarcinucoidea).
Zool. Meded. 63(18): 241-254.
Ng, P.K.L. (1997). The conservation status of freshwater prawns and crabs in Singapore with
emphasis on the nature reserves. Gard. Bull. Singapore 49: 267-272.
30
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Ng, P.K.L. & Lim, K.K.P. (1989). Rediscovery of the Dwarf Snakehead, Channel gachua
(Hamilton, 1822) (Channidae) in Singapore. Raffles B. Zool. 37(1): 172-174.
Ng, P.K.L. & Lim, K.K.P. (1990). The Black Snakehead, Channa melasoma (Bleeker, 1851)
(Channidae): First record from Singapore. Raffles B. Zool. 38(1): 21-24.
Ng, P.K.L. & Lim, K.K.P. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquatic Conserv. 2(3): 255-266.
Ng, P.K.L. & Lim, K.K.P. (1997). The diversity and conservation status of fishes in the nature
reserves of Singapore. Gard. Bull. Singapore 49(2): 245-265.
Ng, P.K.L., Tay, J.B. & Lim, K.K.P. (1994). Diversity and conservation of blackwater
fishes in Peninsular Malaysia, particularly in the North Selangor peat swamp forest.
Hydrobiologia 285(1-3): 203-218.
Ng, P.K.L., Tay, J.B., Lim, K.K.P. & Yang, C.M. (1992). The conservation of the fish and other
aquatic fauna of the North Selangor Peat Swamp Forest and adjacent areas. Kuala
Lumpur: Asian Wetland Bureau and World Wildlife Fund.
Ng, P.K.L., Corlett, R.T. & Tan, H.T.W. (eds) (2011). Singapore biodiversity: an encyclopedia
of the natural environment and sustainable development. Singapore: Editions Didier
Millet.
Nieuwolt, S. 1981. The climates of continental Southeast Asia. In: Takahashi, K. & Arakawa,
H. (eds) World Survey of Climatology: The Climates of Southern and Western Asia, vol.
9, pp. 1 - 66. Oxford & New York: Elsevier Scientific Publishing Co.
O’Dempsey, T. (2014). Singapore’s changing landscape since c. 1800. In: Barnard, T.P. (ed.),
Nature Contained: Environmental Histories of Singapore, pp. 17-48. Singapore: NUS
Press.
O’Dempsey, T. & Chew, P.T. (2013). The freshwater swamp forests of Sungei Seletar
Catchment: a status report. In: Leong, T.M. & Ho, H.C. (eds) Proceedings of Nature
Society, Singapore’s Conference on “Nature Conservation for a Sustainable Singapore”
- 16 October 2011, pp. 121-166. Singapore: Nature Society (Singapore).
Page, S.E., Rieley, J.O., Shotylc, O.W. & Weiss, D. (1999). Interdependence of peat and
vegetation in a tropical peat swamp forest. Philos. T. Roy. Soc. B 354(1391): 1885-1897.
Page, S.E., Hoscilo, A., Wosten, H., Jauhiainen, J., Silvius, M., Rieley, J.O., Ritzema, H.,
Tansey, K., Graham, L., Vasander, H. & Limin, S. (2009). Restoration ecology of
lowland tropical peatlands in Southeast Asia: current knowledge and future research
directions. Ecosystems 12: 888-905.
Phillips, V.D. (1998). Peatswamp ecology and sustainable development in Borneo. Biodivers.
Conserv. 7: 651-671.
Poff, N.L. & Allan, J.D. (1995). Functional organization of stream fish assemblages in relation
to hydrological variability. Ecology 76(2): 606-627.
Poff, N.L., Allan, J.D., Bain, M.B., Karr, J.R., Prestegaard, K.L., Richter, B.D., Sparks, R.E. &
Stromberg, J.C. (1997). The natural flow regime. BioScience 47(11): 769-784.
Posa, M.R.C., Wijedasa, L.S. & Corlett, R.T. (2011). Biodiversity and conservation of tropical
peat swamp forests. BioScience 61(1): 49-57.
Ramirez, A., Pringle, C.M. & Douglas, M. (2006). Temporal and spatial patterns in stream
physicochemistry and insect assemblages in tropical lowland streams. J. N. Am. Benthol.
Soc. 25(1): 108-125.
Richards, P.W. (1996). The Tropical Rain Forest - An Ecological Study, 2 nd ed. Cambridge:
Cambridge University Press.
Rijksen, H.D. & Peerson, G. (1991). Food from Indonesia’s swamp forest - Ideology or
rationality. Landscape Urban Plan. 20: 95-102.
Global significance of freshwater swamp forest
31
Tan, H.H. & Kottelat, M. (2009). The fishes of the Batang Hari drainage, Sumatra, with
description of six new species. Ichthyol. Explor. Freshwat. 20: 13-69.
Tan, H.S. & Tan, H.H. (1994). The freshwater fishes of Pulau Bintan, Riau Archipelago,
Sumatra, Indonesia. Tropical Biodiversity 2: 351-361.
Taylor, D., Yen, O.H., Sanderson, RG. & Dodson, J. (2001). Late Quaternary peat formation and
vegetation dynamics in a lowland tropical swamp, Nee Soon, Singapore. Palaeogeogr.
Palaeocl. 171(3-4): 269-287.
Theilade, I., Schmidt, L., Chhang, R & McDonald, J.A. (2011). Evergreen swamp forest in
Cambodia - floristic composition, ecological characteristics, and conservation status.
Nord. J. Bot. 29(1): 71-80.
Turner, I.M. (1996). Species loss in fragments of tropical rain forests. A Review of the Evidence.
J. Appl. Ecol. 33(2): 200-209.
Turner, I.M., Boo, C.M., Wong, Y.K., Chew, P.T. & Ibrahim, A. (1996). Freshwater swamp
forest in Singapore, with particular reference to that found around the Nee Soon Firing
Ranges. Gard. Bull. Singapore 48(1): 129-157.
Vidthayanon, C. (2002). Peat Swamp Fishes of Thailand. Bankok: Office of Environmental
Policy and Planning.
Wells, A. & Yule, C.M. (2008). The caddisflies (Trichoptera) from a tropical peat swamp in
Selangor, Peninsular Malaysia, including two new species. Aquat. Insect. 30: 69-76.
Whitmore, T.C. (1984). Tropical Rain Forests of the Far East, 2 nd ed. Oxford: Oxford University
Press.
Whitten, T., Damanik, S.J., Anwar, J. & Hisyam, N. (2000). Freshwater-Swamp Forests. Hong
Kong: Periplus Editions.
Winterbourn, M.J. & Collier, K.J. (1987). Distribution of benthic invertebrates in acid, brown
water streams in the South Island of New Zealand. Hydrobiologia 153: 277-286.
Wowor, D., Muthu, V., Meier, R., Bailee, M., Cai, Y. & Ng, P.K.L. (2009). Evolution of life
history traits in Asian freshwater prawns of the genus Macrobrachium (Crustacea:
Decapoda: Palaemonidae) based on multilocus molecular phylogenetic analysis. Mol.
Phylogenet. Evol. 52: 340-350.
Yamada, I. (1997). Tropical Rain Forests of Southeast Asia - A Forest Ecologist’s View.
Honolulu: University of Hawaii Press.
Yang, C.M., Lua, H.K. & Yeo, K. (1997). Semi-aquatic bug (Heteroptera: Gerromorpha)
fauna in the Nature Reserves of Singapore. Proceedings of the Nature Reserves Survey
Seminar. Gard. Bull. Singapore 49: 313-319.
Yee, A.T., Corlett, R.T., Liew, S.C. & Tan, H.T.W. (2011). The vegetation of Singapore - an
updated map. Gard. Bull. Singapore 63(1-2): 205-212.
Yeo, D.C.J. & Lim, K.K.P. (2011). Freshwater ecosystems. In: Ng, P.K.L., Corlett, R.T. &
Tan, H.T. (eds) Singapore Biodiversity - An Encyclopedia of the Natural Environment
and Sustainable Development, pp. 52-63. Singapore: Raffles Museum of Biodiversity
Research, Department of Biological Sciences, National University of Singapore,
Singapore.
Yule, C.M. (2010). Loss of biodiversity and ecosystem functioning in Indo-Malayan peat
swamp forests. Biodivers. Conserv. 19(2): 393-409.
Yule, C.M. & Gomez, L.N. (2009). Leaf litter decomposition in a tropical peat swamp forest in
peninsular Malaysia. Wetlands Ecol. Manag. 17(3): 231-241.
Gardens’ Bulletin Singapore 70 (Suppl. 1): 33-48. 2018
doi: 10.26492/gbs70(suppl.l). 2018-03
33
The hydro-geomorphic status of Nee Soon freshwater
swamp forest catchment of Singapore
C.T.T. Nguyen 1 , R.J. Wasson 2 & A.D. Ziegler 3
‘Tropical Marine Science Institute, National University of Singapore,
18 Kent Ridge Road, 119227 Singapore
2 Lee Kuan Yew School of Public Policy, National University of Singapore,
469C Bukit Timah Road, 259772 Singapore
department of Geography, National University of Singapore,
1 Arts Link, 117570 Singapore
geoadz @ nus.edu. sg
ABSTRACT. This paper presents initial findings from research on the hydro-geomorphic status
of Nee Soon freshwater swamp forest catchment in Singapore. The hydrological system of
Nee Soon contains a swamp that is best described as an organic-rich wetland, with organic
matter content as high as 40% near the surface (too low to be classified as peat). Total long¬
term denudation rate in the catchment is an estimated 23.4 ± 2.08 Mg knr 2 yr 1 , with physical
erosion (5.6 + 0.5 Mg knr 2 yr 1 ) and chemical weathering (17.8 + 1.58 Mg knr 2 yr 1 ) accounting
for 24% and 76% of the totals, respectively. Age dating of a 1.95-m sediment core from the
lower swamp indicates several distinct periods of variable sediment deposition (0.04 to 0.009
cm y 1 ) since 15,000 BCE, across a variety of climate regimes. A missing layer, representing
more than a 7000 year period, verifies substantial channel erosion in the swamp occurring
since 1950. Accelerated erosion associated with forest conversion to agriculture in the upper
catchment could not be verified through examination of sediment cores. High concentrations
of several heavy metals (e.g. As, Cr, Mn, Ni, Sr, V) in the lower catchment, compared with the
upper catchment, appear to be natural (e.g. related to differences in the underlying bedrock),
rather than contamination. The very high concentrations of lead, copper, and zinc associated
with firing activities in the military range in the lower catchment are spatially isolated (e.g.
shooting berms), and currently not posing a threat to the swamp environment. Other hydro-
geomorphic degradation processes/activities now include disruption to hillslope soils and
streams by trampling and mountain biking, back-flow of reservoir release water into the lower
swamp area, and atmospheric deposition of contaminants.
Keywords. Conservation, erosion, heavy metals, hydrology, Pleistocene, vegetation history
Introduction
The transformation of Singapore from a forested island to a modern first-world city
over the last two centuries has been rapid and has been termed environmentally
“catastrophic” (Sodhi et al., 2004), leaving the island nation with less than 1% of
its original forest cover. Historically, the 582 km 2 island was covered by three types
of forest ecosystems (Corlett, 1991; O’Dempsey, 2014): lowland dipterocarp forest
(80-82%), mangroves (13%), and freshwater swamp forests (5%). Between the arrival
34
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
of Stamford Raffles in 1819 and the turn of the 21 st century, nearly all forests had been
converted to other land covers. Today, only about 0.2% of the total area (719 km 2 ) of
Singapore is considered to be primary forest (Brook et al., 2003).
The 7.55 km 2 Nee Soon area, in the heart of Singapore’s Central Catchment
Nature Reserve (Fig. 1), contains virtually the last freshwater swamp forest in
Singapore. Nee Soon freshwater swamp forest catchment occupies 4.8 km 2 (boundary
in Fig. 1). The catchment, which is bounded by the Upper Seletar Reservoir to the
northeast and the Upper and Lower Pierce Reservoirs to the south, is recognised for
its conservation value (e.g. Ng & Lim, 1992; Wee & Ng, 1994; Briffett & Ho, 1999;
O’Dempsey & Chew, 2013; Li et al., 2016; Clews et al., 2018), yet very little has been
written about its physical nature—which is the goal of this paper.
Herein we draw from the results of a recently conducted project entitled “Nee
Soon Swamp Forest Biodiversity and Hydrology Baseline Studies” Phase II (Tropical
Marine Science Institute, 2016). The background and objectives of the project, to
guide future management of the catchment in anticipation of increasing urbanisation,
are described by Davison et al. (2018). We present initial findings related to the hydro-
geomorphological status of the catchment. The interpretations of the findings may
change as new data are collected.
Geology
Nearly all of the Nee Soon catchment is underlain by the Triassic Bukit Timah Granite
Formation, which varies from granite through adamellite, granodirite and several
hybrid granitoids (Ives, 1977). Few core stones or outcrops are present in Nee Soon
catchment, except on the hilltops in the southwest, and therefore variations in the
bedrock can only be inferred from scattered outcrops outside the Nee Soon catchment.
A ground penetrating radar survey, conducted along a publicly restricted walking trail,
the Woodcutter’s Trail, indicated the maximum depth to bedrock is about 9 m (Tropical
Marine Science Institute, 2016). However, because of the difficulty in distinguishing
the interface between the residual soil and the moderately weathered granite, solid
unweathered bedrock may be as deep as 20 m to 70 m in some locations.
A granite rock sample we tested from the upper catchment has medium-
to-coarse grains and contains about 76% SiCf, 13% A1 2 0 3 , and 1.5% Fe 2 0. It also
contains substantial Ba (797 ppm), Mn (197 ppm), and Sr (83 ppm), relative to other
minor elements. Aluminum (6.3%), Na (2.9%) and K (3.5%) are the most abundant
major elements.
Soils
Ives (1977) identified two dominant soil types within the catchment (Fig. 2a): (1)
Rengam Series, developed on igneous rock; and (2) Tengah Series developed on
alluvium. The Rengam Series is generally a clayey, kaolinitic, isohyperthermic, Typic
Geomorphology and hydrology of Nee Soon
35
Topography Elevation (meter)
1.30 -11.23 4, .05 -50.98
£3 11.24-21.17 04 50.99-60 92
21.18-31.11 00 60.93-70,85
C3 31.12-41.04 04 70.86-30.79
Fig. 1 . Location (inset) and topography of the Nee Soon Catchment (defined by thick boundary)
in the Nee Soon Forest Reserve in Singapore.
Legend
■--— Street
- Stream
Nee Soon forest boundary
| Firing range
Catchment boundary
36
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Paleudult that forms on highly weathered Bukit Timah Granite (Chia et al., 1991;
Fauziah et al., 1997). We cannot find a description of the Tengah Series, which in Nee
Soon is associated with the freshwater swamp in the centre of the lower catchment,
where mineral sediments and organic material have accumulated over time (Fig. 2a).
Based on sampling this material to 2m depth, we do not believe it is a peat soil, which
was reported in the past (Taylor et al., 2001), because the organic matter content is less
than the 65% threshold defined by the FAO (Andriesse, 1988). The highest percentage
of organic material we measured was about 40-50%. We therefore recommend
referring to the Nee Soon freshwater swamp forest as an organic-rich wetland, not a
peat swamp. Further, the Tengah Series appears to be a depositional material consisting
of substantial clay with small layers of sand and varying contents of organic matter.
Ives (1977) demarcated a “high ground” zone within the Rengam soils in Nee
Soon (Fig. 2a). In agreement, we find the upper and lower portion of the catchment
to have different soil geochemical signatures. The differences are present in the
mineralogy of the soil in two pits in the upper and lower catchment (Fig. 2a). For
example, the soil in the upper-catchment pit (at 250 cm depth) is composed largely
of Quartz (79%), followed by Kaolin (17%) and small amounts of Gibbsite (3%) and
Goethite (1%). The soil in the lower pit (at 250 cm depth) has much more Gibbsite
(42%), and comparatively less Quartz (33%). The Kaolin content is similar (18%).
Goethite is slightly higher (5%) and a small amount of Illite is present (1%). This
mineralogy is typical of a highly weathered residual soil.
Soil in the upper pit is slightly more acidic (all horizons in the 2 m profile): pH
(determined in water) ranges from 3.4 to 4.2 versus 4.3 to 4.6 in the lower pit. Soil
organic carbon in the 20-30 cm A horizons ranges from 2 to 5% and 2 to 3% for the
upper versus lower pits, respectively, and discussed further by Rahman (2016). The
texture of the B horizon in the upper pit is a sandy clay and sandy clay loam, whereas
the B horizon in the lower pit is mostly clay (upper 1 m) and sand clay loam (lower 1
m). The upper pit has Si0 2 concentrations of 55-69 ppm within the profile, whereas
the ranges of concentrations in the lower pit profile varies between the upper (58-65
ppm) and lower (35-52 ppm) 1 m halves. The lower pit soil also contains more Fe 2 0 3
(4-10%) and Ti0 2 (0.40-0.5%) than the upper pit (2-3% and 0.1-0.2%, respectively).
The geochemical zonation in the catchment soil is apparent in the spatial
distribution of several elements in 227 surface and 30 subsurface samples. Several
elements have significantly higher concentrations in the lower catchment, compared
with the upper catchment (Mann-Whitney U-test; a = 0.05): As, Ba, Cr, Cu, Fe, Mn,
Pb, Sr, Ti, V, and Zn (data not shown). Subsurface samples tend to corroborate the
geochemical patterns found in the surface samples, indicating that enrichment in
the forested lower catchment is natural. The enrichment in some heavy metals gives
the impression of contamination in the lower catchment (shown for Cr in Fig. 3a).
However, we believe the enric hm ent is natural, reflecting a zonation in the underlying
granite bedrock, or some topographically controlled hydro-geomorphological process
affecting soil chemistry occurring over very long time scales (i.e., not anthropogenic).
The higher concentration of Fe 2 0 3 and Ti0 2 in the lower soil pit provides corroborating
evidence that the enrichment of some associated metals is natural (assuming that oxide
Geomorphology and hydrology of Nee Soon
37
Legend
Catchment boundary
SOIL ON ALLUVIUM, on Recent Alluvium Tengah (TgJ Sell
SOIL ON ALLUVIUM, on Recent Alluvium. Junonfl 5en« (Jfl
SOIL ON IGNEOUS ROCK, m Granite. Rengam Senas (Run) Son
High gtnuntl
Street
A
Legend
-&s™*t
- Streim
* We* Soon to rest boun^ar,
^ CilcWwi i h j rt Mw i iy
Fig. 2. (a) Major soil types in the Nee Soon Catchment (based on Ives, 1977). Locations of
the soil pits in the upper and lower catchment are indicated with circles. Locations of cores
collected in the mid and lower catchment are indicated with crosses, (b) Stream network and
the hydrological “operational units” (based on Murphy, 1997) within the Nee Soon Catchment.
concentrations are not associated with contamination). Further, higher concentrations
of many elements to depths below 6 m in the lower pit, compared with the upper pit,
support this interpretation (shown for Cr in Fig. 3b).
With respect to anthropogenic disturbance, we find very high concentrations of
some elements, which are associated with human disturbance, in the lower part of the
catchment—i.e., military lands (ML) or variably disturbed lands (VDL) containing
roads and golf courses. For example, maximum values of some elements greatly
exceed those measured on lands in the forested upper catchment: As (252 versus 70
ppm); Cr (224 versus 55 ppm); Cu (632 versus 16 ppm); Fe (17.32 versus 3.35%); Mn
(1362 versus 110 ppm); Na (1.32 versus 0.16%); Pb (>10,000 versus 188 ppm); Sn
(21.5 versus 8 ppm); Sr (336 versus 11 ppm); Ti (3.83 versus 0.45%); V (521 versus
65 ppm); and Zn (431 versus 55 ppm).
The general similarity between the Nee Soon forested upper catchment and
forested lower catchment soil chemistry and concentrations determined in the nearby
MacRitchie Catchment give support to the reliability of the values we have determined
in Nee Soon, despite the wide range found within such a small area (Table 1). Further,
the maximum values associated with disturbed lands at Nee Soon (372, 679, and 1926
38
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Legend
Element* fft iffl-sw
cr [—| c™ ewnr, E3.<f*ct_Cf_r1
1 ^ U
lt<4
44tN
S»»*#rp
Vatu*
£
u
a
ai
a
Cf (ppm)
S 10
-100
-200
-BOO
-400
-500
-600
-700
800
•i
• _ *
o
o
*
«
*
*
*
*
*
o
o
o
G
O
O
O
o
Q
o
o
o
o
o
Q
IS
« UFC Pit
o LFC Pit
Fig. 3. (a) Near-total (mixed acid digestion; ICP-MS; see footnote in Table 1). Chromium
concentrations (ppm) at 227 surface samples within the Nee Soon Catchment. Shading indicates
spatial distribution determined by ordinary kriging on the samples, (b) Near-total chromium
concentrations within the soil profile of pits located in the upper and lower catchment (see Fig.
2a for pit locations).
ppm for Cu, Pb, and Zn, respectively) demonstrate the likelihood that human activities
have elevated concentrations in a manner similar to that reported by Chen et al. (1996)
for industrial surface soils in Singapore (485, 235, and 3594 ppm, for Cu, Pb, and
Zn). For Nee Soon, the high Cu and Pb concentrations associated with the military
lands (Cu = 632 ppm; Pb > 10,000 ppm) indicate potent sources of anthropogenic
enrichment in isolated areas (e.g. near the berms on firing ranges) that may require
remediation in the future.
Some elements that are considered to be pollutants, including As, Cu, Pb, can
potentially move into the subsoil (Teo, 2016). Measurements of saturated hydraulic
conductivity (and indicating property for infiltrability) in the soil profiles indicates a
sharp decrease from about 485 mm h 1 to 23 mm h 1 by a depth of 50 cm (Kho, 2014,
2016). This decrease, which is typical of a natural tropical forest hillslope with a thin,
organic-rich and permeable A horizon formed over a clay-rich B horizon (cf. Ziegler
et al., 2006), may prevent the movement of enriched elements to great depths in the
soil profile, possibly preventing them from interacting with the groundwater. In 20
water samples tested for heavy metals (ICP-MS), we found very low concentrations,
suggesting limited (if any) movement of potentially harmful elements from the soil to
the stream system (data not shown), or our failure to detect a synoptic signal. From
Table 1. Comparison of the concentrations of selected elements in Nee Soon with those from others studies at MacRitchie and other locations in
Singapore.
Geomorphology and hydrology of Nee Soon
39
c
N
ft
ft
-a
ft
ft
P &
U w
la
^ &
U w
&
o
u
CJ
CZ5
a>
O
-M
as
-ft
a
at
'O
<D
-a
u
3
-m
as
■5
s
3
a
o
o
C/5
<u
a>
z
05
a )
3
CJ
'(-i
s
a
vo vo
oo in
OO CO 00
(N
in cd
h in h
r-
a
2
<
"a
u
B
a
Z
y u
D J
ft ft
Z Z
00
o o a-
(N a
CO
cd^ cd^ cd
vo <n a-
r--- in co cd
_ a-
CO ^S
r- a-
a- a-
<N
i—H
cn
r-‘
in
^^
i-H
^^
o
vo
OO
OO
OO
OO
vo
3t
m
OO
in
CO
t—C
Cd
oc
Cl
H
<D
in
Cf
CO
N—^
oc
r-
in
OO
CO
3t
CO
cd
cd
05
a
<U
42
a
a
-m
05
a
o
o
a
<d
Sh
<
"a
05
a
t/3 13 J
0 ) ^
0 ) ^
Z
05
(►
CD
CD
■e
3
t/5
T3
<D
X
s s
ft
Q
>
g
.Oj
7h
7g
B
C
N
c
4h
tD
fl
S
"d -
a
o
O
o
w
CD
o
CD
o
■M
• i-H
O
o
■M
• »-H
o
4—i
• i-H
o
0^
CO
m
Pi
DO
DO
o
CD
CD
CD
CD
o
C3
o
a
CD
CD
CD
CD
CD
S z
in co t— vo o
cd '-h vo oo in a-
■T"
cd
cd
P
CO
a
ON
in
'—^
co"
'—^
cd"
ON
1
CO
CO
1
OO
'—^
3t
1
in
CO
CO
vo
r-
Cd
CO
ON
in
vo
cd
CO
cd
r-
3t
OO
vo
I
r-
OO CO
ov a-
l l
r-H r-
o
in
CO
vo
OO
r-
o
O
o
cd
cd
OO
r-
o"
c-
1
1
1
^|-
1—H
1
co
1
"7
vo
in
CO
CO
OO
in
cd
—H
CO
CO
OO
vo
vo
co
'—^
s —^
ON
Ol-
OO
vo
CO
vo
1
1
"7
1
1
00
CO
in
CO
c-
I I I
a - o i
OO vo
t-h CO
h 7 7
I O OO
■H CO CO
I I I
>n a -
in r- cd
m, — cl
I I I
Cd rH in
CD g
N- O
o
3 o
a
o
o
m
CD
CD
£
05
k
05
a
o
Cm
a
CD
u
CD
ft
O
03
d
a
CD
T3
• i-H
C/D
CD
P£h
o
03
'a
03
a
as
a
o
03
"0
a
a
a
e a
S °
o o
do
CD qy
O OD
£ £
y u
p j
PH ft
a
o
o
DO
CD
CD
Z
Cd Cd
-O TD
CD CD
ca ca
-a
a
a
CD
CD
ft
03
a
P
CD ™
OD
Oh a
03 --p
a 23
P S
c
a
c
o
o
o
o
o
o
DO
DO
DO
CD
CD
CD
CD
CD
CD
z;
z
z
jq c «
a cd p
o £ a
c3 3
a c 3 <n
8 u 3
S fe 3
c/5 Q- —
<D CL T?
pt <d
’S D -o
§ -o S
S3 | P
2 o S
.ft O CU
« tv ^
(D - DO
3
ft CO
3 * £
> £ -3
03 P 03
a a a>
O to £2
m a kj
% > 'O'
3 3 8
S % §
a cs o
g 3
V 'o
to -g txD
C3
CD W) on
b C X3
.3 3 8
s bD
u o
a • - ^
3 «
2 3 a
- ° S
cd a 3
to 3 —
b ° a
5 c aj
8 © -e
a 3 3
O to c/3
Cd
co « a?
3 3
a h .S
<u § b
co 2 cd
3 O >
a 1/3 o
s <p
3 a> to
-- - 7
<D ^
3£ £
O oi J
■s 8 Q
3 >
> to'
o -a
e
3
Vh ,_i
,o ■ •
V-t re
l-c -1
a £
<u
£ 3
3 o
o '*-*
a a
o a
a a
t3
(U
& v=
§ s
a dh
3a
8 ”c
a a
3
I £
i§
< M .
C/5
a
3
CD
0J5
a.
a
o
33
OJ
5 S
o
CS
a
CD
x
o
- g
o
03
CD
£
C3
a
c3
a
c
3 p
c a,
o rti
'—'
Vi ■ >.
cd
G3 a
a o
S _r
' 3
+-•
O ID
§ 3
O
a n
al
a te
• • a
to p
a cd
o a
a c
<D C
! £
S 2
o' ^
S £
a o
Ǥ
8 u
O o
« O
CD 00
CD
a
CD
P<
CD
3
a
3
3
8 &
§ .3
a a
f-H
3 £
a
c
3
t/3
CD
£
3
O
to
3
3
Q
C3
G 2
CD CD
> a
O
CD G
* |
2 c
CD O
£ M
S 2
o ^
G O
S3 O
0J3
.£ te
3 U
~o °
CD O
to
CD ’— 1
0J3 O
=3 a
CO rg
<U C
"S. S
£ cd
cd
CO
c2
<5
CD
£
o
o
a
cv
. o
to
cd a
3 So
co ©
a ^
OD >, S
3^2
3 a £
to
G
G
O
•e
P
CD
8 ^
8 2
CD
• s u
3
CD
o
a
a _
* g
• 2 «
te cd
S—i r — 1
(D
H
.2 ^
cd
u
’£
o
cd
<D
o
cd
P
CD
r" 1
M S -g
ci a
G 3
O U
CD W
a
CD
« 2 a
« g-
cd
CD
D D
a 8 o
£ G ft
2 « a
3 -G 3
3
a
">
a
s
1 ^
ie
3 w
a 3
^ Cf3
£ CD
3 a
Q aj
^ ft
t-H 3
cd a
o OX)
P 3
3 "3
s S
to
ft 3
£ .2
8 2
■a 2
CD
Z G
i- O
CD O
OX) rt
3 "ob
■£ a
a
ON
CD
a
CD
>
3
C
‘g
i
o
CN
D' i
a _
a p
CD G
52 a
cd 32
\0 cd
O''- a
si
4-*
b
CD
X)
g
C/3
P
m
<D
P
>
-2
""pH
*— <
P
£
00
p
C/1
C/5
2
OO
o-
' ^
£
2
_ 3
i-T
a
OD
£
CD
W)
<u
_p
4-*
P
d—*
a>
C/3
>
CD
"O
cd
CD
CD
C/3
P
jp
<D
13
c-
4—*
cd
O
p
4-*
■
T3
Uh
CD
s
Td
p
bn
p
O
o
c
a-)
<u
<D
>
cd
oo n
SO Hpr CO
(D
cr®ii
Z E >
® ON >>
a
a
a
U
ffi
a
o 3
CD to
b rn
3
X
£
3
a
H
40
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
these results we conclude that polluted areas do not currently pose a threat to the
freshwater aquatic ecosystem, but this issue should be monitored and some of the
polluted areas may require remediation.
Stream hydrology
The drainage system in the 4.8 km 2 Nee Soon catchment includes a third order stream
that flows into the Lower Seletar River (Fig. 2b). Comparison with old maps and field
observations shows that the stream has been altered drastically in the lower part of the
catchment. For example, the main channel was straightened above where it joins the
outflow from the spillway of the Upper Seletar Reservoir (Fig. 2b; Lower 3 section).
Because the stream system has been so greatly altered, Murphy (1997) divided it into
hydrological “operational units” (Fig. 2b), rather than dividing the drainage network it
into meaningful hydrological sub-catchments (discussed below).
Streams draining from the upper catchment typically have slope gradients less
than or equal to about 5° and coarse sandy or sandy-loam streambeds. Streams on low-
gradient terrain accumulate dense mats of organic material. Where the streams reach
the swamp, they are shallow and narrow, and they move across the swamp, as evidenced
in cores by thin layers of channel sand separated by organic-rich clay. Streams in the
upper catchment tend to maintain flow year round, indicating a significant groundwater
contribution that returns to the surface via springs.
The water balance of the Nee Soon catchment is primarily driven by rainfall
associated with two monsoon seasons, the Northeast monsoon (October to early
January) and the Southwest monsoon (late March to May), yet rainfall is typically
plentiful even in the inter-monsoon seasons—except in occasional droughts that
are, for example, associated with phenomena such as El Nino (Ziegler et al., 2014).
Mean annual rainfall at Nee Soon is 2330 mm; monthly means range from 159 to
288 mm (http://www.weather.gov.sg/climate-climate-of-singapore/). In the simulation
of the water balance for the catchment, modelled evapotranspiration (1100 mm) and
stream runoff (2200 mm) represented 33% and 66% respectively of the total rainfall
input (Sun et al., 2018; Liong Shui-Yui, Tropical Marine Science Institute, personal
communication).
We find that the main stream draining the upper part of the catchment has a mean
pH value of 6.29 ± 0.38 (n = 27). Two tributary streams to the main channel have very
similar mean values of 6.05 ± 0.27 and 6.03 + 0.44. The springs in the upper part of
the catchment are the most acidic, with values ranging from 4.49 ± 0.18 to 4.88 ± 0.33.
Other surface water and ground water samples have pH values less than 6.0, including
the swamp in the mid part of the catchment (5.69 ± 0.48). We believe stream pH in
Singapore is low because of both natural (low buffering related to acidic granite) and
anthropogenic (influenced through acid rain) phenomena. During the last two years of
the study, we measured rainfall pH values ranging from 3.41 to 6.35 (collected at the
National University of Singapore). Our initial analysis of rainfall pH does not reveal
a strong relationship between low pH and indicators of acid rainfall, S0 4 2 and N0 3 ,
Geomorphology and hydrology of Nee Soon
41
but this is expected, as other acid-related constituents, sea-spray, dust, and aerosols
from biomass burning all contribute to acid rain in Singapore (Balasubramanian et al.,
2001 ).
The ranges of specific conductivity values determined at five sites (n=64 samples)
tend to be lower in the upper catchment streams (19-30 pS cm 1 ), springs (16-38 pS
cm 1 ), and groundwater (23-73 pS cm 1 ), than in the lower catchment main channel
(16^-04 pS cm 4 ), streams (27-207 pS cm 4 ), and ground water (25-149 pS cm 4 ).
The lower catchment waters tend to have higher concentrations of Cl (6 vs 2 ppm)
S0 4 2 (6 vs 1 ppm), Na + (5 vs 2 ppm), and Ca 2+ (4 vs 2 ppm). Values vary depending
on rainfall conditions (with respect to depth and acid rain associated with the urban
environment). As with the case of pH, however, insufficient data have been collected
to determine controls, natural versus anthropogenic, of the differences among streams
in the catchment.
Hydrological resilience
The Nee Soon freshwater swamp forest appears to be somewhat resilient to weather/
climatic fluctuations, including the 2014-2016 drought, which contained the second
driest year recorded for Singapore (Meteorological Service Singapore, 2015). In
the simulated water balance for the 3-year study period, mean annual water storage
loss in the catchment was an estimated 70 mm (Sun et al., 2018), a depth that surely
taxed groundwater reserves feeding the swamp. Nevertheless, while many Singapore
streams dried during the drought, much of the swamp area in Nee Soon remained wet
or moist (observations by the authors).
Evidence of long-term resilience can be gleaned from our preliminary pollen
analyses of a 1.95 m core taken from the lower swamp at Nee Soon. For example, old
growth forest pollen was only about 20% at about 600 BCE, increasing to about 60%
by 1000 CE. During this period, increasing numbers of fern spores and palm pollen
provide evidence of increasing rainfall, because the other key variable for vegetation,
temperature, can be assumed to have varied little at the equator in this time period. An
abundance of rainfall is also suggested by the low values of charcoal found in the cores
(i.e., few fires). Pollen and spores within a third sediment layer suggests gradually
increasing cover of old growth forest and ferns, and decreasing grass. This period
may be associated with increasing rainfall following the Little Ice Age (-1300-1870
CE). In recent times (the last 70 to 100 years), little change in the vegetation can be
observed. Given that we have dated the deposits to more than 15,000 years, our initial
pollen analyses indicate the swamp has remained despite the vegetation being quite
different (e.g. from grasslands to forest) in response to changing climates.
Another form of resilience is associated with the persistence of the swamp
despite inundation of the lower stream system when excess water is released over the
slipway of the Upper Seletar Reservoir and released water flows backwards up the
main stream channel, flooding the lower part of the swamp by several centimetres.
The reverse flow transports sediments, nutrients, and biota into the swamp and up the
42
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
stream network for short periods (typically less than a day). This artificial flooding
occurs in response to weather conditions when water is exchanged between reservoirs,
to maximise storage, relative to use, rainfall, and evaporative loss. Model results
provide estimates of the magnitude of these flows: the reservoir contribution to total
modelled catchment outflow was about 55% higher than that of natural stream flow in
the lower part of the catchment (Liong Shui-Yui, personal communication).
Despite this hydrological resilience, we found evidence that substantial erosion
has taken place within the stream network. Our analysis of the 1.95 m sediment core
and the corresponding sedimentation rates, together with the observations of Murphy
(1997), suggest that stream channel erosion has been significant in the past. In the
sediment core we examined, dates derived from 14 C analysis (on pollen) corresponded
to 1950 CE and 5500 BCE at depths of 50 cm and 51 cm, respectively (also discussed
below). This large, abrupt age gap is best explained by the erosion of sediment layers
that had accumulated over about 7000 years.
Murphy (1997) indicated incision of the stream by as much as 1 to 2 m may have
resulted from runoff and accelerated erosion related to construction and maintenance
of a water pipeline that bisects the catchment. During our recent surveys, disturbance
by the pipeline, roads, and trails on the stream channel are apparent, but their
geomorphological impacts are now less severe than indicated by Murphy (1997). The
greatest disruption to the stream system is trampling during military training activities,
by unauthorised personnel during hikes, as well as unauthorised mountain biking on
sloping trails and stream crossings.
Soil erosion and denudation
By measuring the radionuclides 10 Be, 137 Cs, and 210 Pb and major elements in rocks and
soils, we are able to assess the following: (a) element accumulation/loss related to
physical and chemical weathering processes; and (b) catchment short- and long-term
erosion and sedimentation rates (in accordance with human influenced and natural
processes). Several elements, including Ca, K, Mg, Ba, Na, Mn, Ti, Co, Sr, Ni, Zn,
Cr, Fe, and V, are depleted throughout the soil catena, relative to stable Zr. Loss in
concentrations of elements of low mobility (e.g. K, Ti, Cr, and V) suggests intense
weathering has occurred in the catchment. However, element depletion/enrichment
is variable within soil profiles in the catchment. Alternating zones of enrichment
and depletion of selected elements along catenas are not associated with commonly
reported micro- or macro-environmental forcing variables (gradient, organic matter,
soil texture, infiltration) but may be a result of pulses of surface erosion and deposition
along the slopes.
Total denudation rate, determined from the cosmogenic nuclide 10 Be in Nee
Soon stream sand (recalculated using the correction for quartz enrichment) is -9 + 0.8
m Ma _1 , or about 23.4 + 2.08 Mg km 2 yr 1 (assuming a rock density of 2.6g cm 3 ). The
estimated physical erosion rate (5.6 + 0.5 Mg km 2 yr 1 ) and the chemical weathering
rate (17.8 + 1.58 Mg knr 2 yr 1 ) for soil are calculated from the total denudation rate
Geomorphology and hydrology of Nee Soon
43
based on a mass balance approach (Nguyen, 2017). These calculations suggest physical
erosion accounts for only about 24% of total denudation, compared with 76% for
chemical weathering and loss from the soils.
Low denudation rates are not unexpected in the Nee Soon catchment, given the
gently sloping terrain (steepest stream slope is ~5°) and dominance of forest vegetation.
The unusually high contribution of chemical weathering (43-84%) is indicative of the
warm temperatures and high rainfall of this tropical locale (NO 1.39017°, E103.80893°).
We believe that the physical erosion rate (5.6 ± 0.5 Mg km 2 yr 1 ) is slow compared
with other studied granite-derived soil environments (e.g. in Riebe et al., 2001, 2004).
However, because chemical weathering is high (accounting for c. 76% of the total
denudation), biochemical processes likely play an important role in soil formation.
Our field observations suggest that tree uprooting, which causes redistribution
downslope of the soil in the root mat, as well as bioturbation by abundant termite and
ant activity extending into the B horizon, are important for soil formation/alteration.
Moreover, volumetric strain calculations suggest that rock deformation and soil
formation has been intense over a period of hundreds to thousands of years (Nguyen,
2017).
Dating of swamp sediments by the use of 137 Cs and 210 Pb(ex) suggests that a
period of accelerated erosion may have occurred during and/or since the 1950s due to
disturbance from the construction and maintenance of a water pipeline in the catchment
and other peripheral activities at the lower catchment. Again, accelerated erosion
related to these disturbances is not substantial today. Accelerated erosion associated
with forest conversion to agriculture in the upper parts of the catchment also cannot be
ascertained with these methods.
Sedimentation rates
We collected cores from the middle and lower swampy areas with a 6.36 cm gouge
auger to determine sediment deposition and to support a variety of analyses (description,
total organic carbon (TOC), bulk density, texture, elemental and oxides concentrations,
radioisotope dating). Core depths vary from 1 to 1.95 m depending on the ability
to penetrate subsurface material and recover an intact core. For the lower reach, we
focus primarily on one core, but use others to provide additional material for analysis
and to examine spatial patterns of deposition. Ages of various layers were determined
from a variety of isotope techniques ( 14 C, 210 Pb and 137 Cs) on pollen, charcoal, and
sediments. Thus, our interpretations reference a composite core, constructed from data
from several cores (Fig. 4).
Within the total length of the composite core for the lower swamp, we demarcate
several distinct layers having highly variable deposition rates (Table 2; see also Fig. 4).
The ages of these layers extend back through the Holocene (0-11,700 BP) to before
the Late Glacial Maximum (10,000-13,000 BP). However, the contemporary periods
of maximum disturbance related to forest conversion to agriculture (from about 1850)
are missing due to channel incision (layer 2), as mentioned above.
Only three dates can be determined for the middle swamp core. Radiocarbon
44
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Eroded layer
Depth
tm
Composite tore
ItmiCE
19S0s CE
w. 5500BCE
tt. 65onset
aJOOOKE
ca, SMQDBCE
ta. 1 lOQEieCE
ca. 1 5Q006CE
m
i
FINE 5 AND IN BROWN: CL At
BROWN CLAY
BROWN GREY CLAY
FINE AND COARSg SAND
ORANGE INCLUSIONS
OREFCUY
SLACRCUW
LAMINATIONS
LEAf REMAINS
PEBBLE
ca.IWOOBCE
Fig. 4. Distinguishable layers and associated ages and deposition rates within the sediment core
collected in the lower swamp of Nee Soon Catchment (location shown in Fig. 2a).
Table 2. Estimated deposition rates within layers revealed by composite soil core for the lower
swamp area of Nee Soon freshwater swamp forest.
Layer
Depth
Age range
Sedimentation rate
(cm y 1 )
1
0 - 50 cm
2016 to 1950 CE
0.76
2
Missing
No data
No data
3
51 - 60 cm
1950 BCE to 5500 BCE
0.009
4
60 - 80 cm
5500 BCE to 6500 BCE
0.04
5
80 - 90 cm
6500 BCE to 9400 BCE
0.004
6
90 - 100 cm
9400 BCE to 10,000 BCE
0.02
7
100- 155 cm
11,000 BCE to 15,000 BCE
0.011
8
155 - 193 cm
15,000 BCE to 19,000 BCE
0.0095
Geomorphology and hydrology of Nee Soon
45
dates on pollen indicate three distinct periods of accumulation: (1) present to 1273 CE
(0-58 cm; 0.08 cmy 1 ); (2) 1273-1241 CE (58-62 cm; 0.125 cmy 1 ); and (3) 1241 CE
to 657 CE (62-115cm; 0.09 cm y' 1 ). However, given the uncertainty in each estimate,
as well as the similarity in accumulation rates, one rate of 0.1 cm y 1 can be assumed
for the entire core, representing about 1360 years of deposition. The sedimentation
rates suggest that the upper catchment swamp was not disturbed substantially by forest
conversion to agriculture in the mid-19th century, whereas disturbance/change in the
lower catchment has been dynamic with respect to climate and human activity. Again,
direct comparison with the lower swamp is impossible, because sediments associated
with the period between 1950 CE and 5500 BCE are missing in the lower core.
Lastly, anthropogenic inputs of Pb, and perhaps Ba and Cu, are detectable in
the upper few centimetres of the cores (Kho, 2014, 2016; Nguyen, 2017). According
to Chen et al. (2016) there are several possible sources contributing Pb to the nearby
MacRitchie Reservoir in Singapore, including power generation-based coal combustion
from nearby Indonesia (Lucarelli, 2010). Our isotope data are less variable than those
of Chen et al. (2016), but also appear to be enriched by external sources, although it is
not possible to calculate a source budget at this time.
Conclusions
The hillslopes and stream network of the Nee Soon catchment have been degraded by
many decades of human impact, but their resilience is being demonstrated by clear
evidence of recovery of the vegetation, soils, swamps and streams. Resilience of the
hydrologic system helps to explain the apparent resilience of the aquatic fauna (e.g.
Ng & Lim, 1992; Ho et al., 2018), including the survival of hyperendemics, despite
disturbance of the associated vegetation. The upper catchment forested hillslopes
were disturbed by anthropogenic activities beginning about 1850 when agriculture
was first established. Some of the greatest disturbances occurred about 70-100 years
later when the firing ranges were established, the lower stream and swamp network
were disturbed by channel straightening, and erosion was accelerated from building
and maintaining a water pipeline. Further, much of the lower catchment vegetation has
been converted to urban land uses, including a golf course and roads, and the existence
and operation of nearby reservoirs alters the catchment hydrology.
With respect to management, the following represent some of the most obvious
current challenges to address: (a) disruption of the natural stream flow by water
release from adjacent reservoirs; (b) high concentrations of some heavy metals in the
soils within the military firing range; (c) inputs of heavy metal contaminants to the
catchment through atmospheric deposition; and (d) disturbance of the forest slopes and
the stream network by hikers, mountain bikers, and military personnel during training.
Given that Nee Soon is the last remaining freshwater swamp forest in Singapore and
an important site of biodiversity, we recommend addressing these management issues
in tandem with conducting additional baseline research and a campaign of educational
awareness.
Beyond identifying negative impacts, the goal of this study was to develop
46
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
a hydro-geomorphological baseline for the Nee Soon catchment from which future
studies and management activities could be developed. The initial findings, summarised
briefly above, reveal shortcomings with respect to establishing such a baseline.
Additional research is needed to improve our understanding of basic hydrological and
geomorphological phenomena in the catchment: e.g. a) water balance components;
b) stream dissolved and particulate fluxes; c) soil formation and denudation rates;
d) hillslope degradation and transport processes; e) hydrological pathways; and f)
anthropogenic contamination; g) and recommendations for management, particularly
remediation of polluted sites.
ACKNOWLEDGEMENTS. This paper was completed as part of the research project titled
“Nee Soon Swamp Forest Biodiversity and Hydrology Baseline Studies” Phase I and II (Project:
TMSI-NS-FR-0816), executed by the Tropical Marine Science Institute, National University of
Singapore, for the client National Parks Board (NParks), Singapore. We thank several students
and researchers who provided insights and input to this work: Charlene Teo, Elvagris Segovia
Estrada, Karisham Pai, Rachel Y.T. Kho, Lim Meng Wee, Muhd Herzad bin Mohd Rahman,
Nisha Ramzdan, Ong Mia Xiang, Sebastian Canterrero, and Win Swe Hlaing.
References
Andriesse, J.P (1988). Nature and management of tropical peat soils. FAO Soils Bull. 59.
Rome: Food & Agriculture Organisation.
Balasubramanian, R., Victor, T. & Chun, N. (2001). Chemical and statistical analysis of
precipitation in Singapore. Water Air Soil Pollut. 130: 451-456.
Briffett, C. & Ho, H.C. (1999). State of the natural environment in Singapore. Singapore:
Nature Society (Singapore).
Brook, B.W., Sodhi, N.S. & Ng, P.K.L. (2003). Catastrophic extinctions follow deforestation in
Singapore. Nature 424: 420-426.
Chen, F. (1999). A study of heavy metal distribution in Singapore. M.Sc. thesis, Nanyang
Technological University, Singapore.
Chen, W., Tan, S.K. & Tay, J.H. (1996). Distribution, fractional composition and release of
sediment-bound heavy metals in tropical reservoirs. Water Air Soil Pollut. 92: 273-287.
Chen, M., Boyle, E.A., Switzer, A.D. & Gouramanis, C. (2016). A century long sedimentary
record of anthropogenic lead (Pb), Pb isotopes and other trace metals in Singapore.
Environ. Pollut. 213: 446-459.
Chia, L.S., Rahman, A. & Tay, D.B.H. (1991). The Biophysical Environment of Singapore.
Singapore: Singapore University Press.
Clews, E., Corlett, R.T., Ho, J.K.I., Koh, C.Y., Liong, S.Y., Memory, A., Ramchunder, S.,
Siow, H.J.M.P., Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W., Theng, M.T.Y. & Yeo, D.C.J.
(2018) The biological, ecological and conservation significance of freshwater swamp
forest in Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
Corlett, R.T. (1991). Plant succession on degraded land in Singapore. J. Trop. Forest Sci. 4(2):
151-161.
Geomorphology and hydrology of Nee Soon
47
Davison, G.W.H., Cai, Y., Li, T.J. & Lim, W.H. (2018). Integrated research, conservation
and management of Nee Soon freshwater swamp forest, Singapore: hydrology and
biodiversity Gard. Bull. Singapore 70 (Suppl. 1): 1-7.
Fauziah, C.I., Jamilah, I. & Syed Omar, S.R. (1997). An evaluation of cation exchange capacity
methods for acid tropical soils. Pertanika J. Trop. Agric. Sci. 20: 113-119.
Ho, J.K.I., Quek, R.F., Ramchunder, S.J., Memory, A., Theng, M.T.Y., Yeo, D.C.J. & Clews,
E. (2018). Aquatic macroinvertebrate richness, abundance and distribution in the Nee
Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 71-108.
Ives, D.W. (1977). Soils of the Republic of Singapore. New Zealand Soil Survey Report 36.
Lower Hutt: New Zealand Soil Bureau, Department of Scientific and Industrial Research.
Kho, R.Y.T. (2014). A sedimentary record of land use change and heavy metal deposition
in Nee Soon Swamp Forest, Singapore. B.Sc. (Hons) thesis, Geography Department,
National University of Singapore, Singapore.
Kho, R.Y.T. (2016). Impacts of different types of vegetation cover on soil hydraulic
characteristics in Nee Soon Swamp Forest, Singapore. B.Sc. (Hons) thesis, Geography
Department, National University of Singapore, Singapore.
Li, T., Chay, C.K., Lim, W.H. & Cai, Y. (2016). The fish fauna of Nee Soon Swamp Forest,
Singapore. Raffles Bull. Zool., Suppl. 32: 56-84.
Lucarelli, B. (2010). The history and future of Indonesia’s coal industry: impact of politics and
regulatory framework on industry structure and performance. Program for Sustainable
Energy Development, Freeman Spogli Institute of International Studies, Stanford
University, Stanford, California, USA. https://pesd.fsi.stanford.edu/sites/default/files/
WP_93_Lucarelli_revised_Oct_2010.pdf. Accessed 12 Dec. 2017.
Meteorological Service Singapore (2015). Annual Climate Assessment 2015. http://
www.weather.gov.sg/wp-content/uploads/2016/03/Annual-Climate-Assessment-
Report-2015.pdf. Accessed 2 Sep. 2016.
Murphy D. H. (1997). Final report on map work and arthropod survey in 1994-96. Singapore:
National Parks Board.
Ng, O.-H., Tan, B.C. & Obbard, J.P. (2006). Lichens as bioindicators of atmospheric heavy
metal pollution in Singapore. Environ. Monit. Assess. 123: 63-74.
Ng, P.K.L. & Lim, K.K. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquat. Conserv. Mar. Freshw. Ecosyst. 2: 255-266.
Nguyen, C.T.T. (2017). Is Nee Soon catchment disturbed? A holistic approach to the catchment’s
elemental distribution, movement and deposition in soils and sediments. Doctoral thesis,
Department of Geography, National University of Singapore, Singapore.
O’Dempsey, T. (2014). Singapore’s changing landscape since c. 1800. In: Barnard, T.P. (ed.)
Nature Contained: Environmental Histories of Singapore, pp. 17-48. Singapore: NUS
Press.
O’Dempsey, T. & Chew, P.T. (2013). The Freshwater Swamp Forests of Sungei Seletar
Catchment: A Status Report. In: Leong, T.M. & Ho, H.C. (eds) Proceedings of Nature
Society, Singapore’s Conference on “Nature Conservation for a Sustainable Singapore ”
- 16th October 2011, pp. 121-166. Singapore: Nature Society (Singapore).
Rahman, M.H.B.M. (2016). Soil Organic Carbon Stocks in the Nee Soon Swamp Forest. B.Sc.
(Hons) thesis, Geography Department, National University of Singapore, Singapore.
Riebe, C.S., Kirchner, J.W., Granger, D.E. & Finkel, R.C. (2001). Strong tectonic and weak
climatic control of long-term chemical weathering rates. Geology 29: 511-514.
Riebe, C.S., Kirchner, J.W. & Finkel, R.C. (2004). Erosional and climatic effects on long¬
term chemical weathering rates in granitic landscapes spanning diverse climate regimes.
Earth Planet. Sci. Lett. 224: 547-562.
48
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Sin, Y.M., Wong, M.K., Chou, L.M. & Alias, N.B. (1991). A study of the heavy metal
concentrations of the Singapore River. Environ. Monit. Assess. 19: 481-494.
Sodhi, N.S., Koh, L.R, Brook, B.W. & Ng, RK.L. (2004). Southeast Asian biodiversity: an
impending disaster. Trends Ecol. Evol. 19: 654-660.
Sun, Y., Kim, D.Y., Wendi, D., Doan, D.C., Raghavan, S.V., Jiang, Z. & Liong, S.Y. (2018).
Projected impacts of climate change on stream flow and groundwater of Nee Soon
freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 175-190.
Taylor, D., Yen, O.H., Sanderson, P.G. & Dodson, J. (2001). Late Quaternary peat formation and
vegetation dynamics in a lowland tropical swamp; Nee Soon, Singapore. Palaeogeogr.
Palaeocl. 171: 269-287.
Teo, W.T.C. (2016). Assessment of Soil Heavy Metal Concentrations within the Nee Soon
Swamp Forest. B.Sc. (Hons) thesis, Geography Department, National University of
Singapore, Singapore.
Tropical Marine Science Institute (2016). Nee Soon Swamp Forest Biodiversity and Hydrology
Baseline Studies Phase 2. Final project report. Singapore: Tropical Marine Science
Institute, National University of Singapore.
Wee, Y.C. & Ng, P.K.L. (1994). A first look at Biodiversity in Singapore. Singapore: National
Council on the Environment.
Zhou, C.Y., Wong, M.K., Koh, L.L. & Wee, Y.C. (1997). Soil lead and other metal levels in
industrial, residential and nature reserve areas in Singapore. Environ. Monit. Assess. 44:
605-615.
Ziegler, A.D., Negishi, J.N., Sidle, R.C., Noguchi, S. & Nik, A.R. (2006). Impacts of logging
disturbance on saturated hydraulic conductivity in a tropical forest in Peninsular
Malaysia. Catena 67: 89-102.
Ziegler, A.D., Terry, J.P., Oliver, G.J.H., Friess, D.A., Chow, T.W.L., Chuah, C.J., Wasson, R.J.
(2014). Increasing Singapore’s Resilience to Drought. Hydrol. Process. 28: 4543-4548.
Gardens’ Bulletin Singapore 70 (Suppl. 1): 49-69. 2018
doi: 10.26492/gbs70(suppl.l). 2018-04
49
Rediscoveries, new records, and the floristic value of the
Nee Soon freshwater swamp forest, Singapore
K.Y. Chong 1 , R.C.J. Lim 1 - 2 , J.W. Loh 1 , L. Neo 1 ,
W.W. Seah 1 - 3 , S.Y. Tan 1 & H.T.W. Tan 1
department of Biological Sciences, National University of Singapore,
14 Science Drive 4, 117543 Singapore
kwek@nus.edu.sg
horticulture and Community Gardening Division, National Parks Board,
100K Pasir Panjang Road, 118526 Singapore
herbarium, Singapore Botanic Gardens, National Parks Board,
1 Cluny Road, 259569 Singapore
ABSTRACT. The unique plant communities of the freshwater swamp forests of southern
Johor (Malaysia) and Singapore attracted the attention of E.J.H. Comer, but there have been no
comprehensive follow-up studies to his seminal work. Meanwhile, freshwater swamp forests in
the region have been mostly lost to logging and in-filling for plantations or urban development.
The Nee Soon catchment contains the last substantial tract of this forest type in Singapore. We
collated the rediscoveries of vascular plant species presumed Nationally Extinct in the 2 nd and
latest edition of the Singapore Red Data Book, and new records for the Singapore vascular
plant flora from the Nee Soon catchment, including those that we found and collected through
the establishment and survey of 40 vegetation plots, each 20 x 20 m. We have identified 672
species from 117 families, of which 288 are trees from 60 families represented by at least one
stem > 5 cm DBH. The catchment is especially species rich and abundant in the Myristicaceae.
In the last ten years, 53 rediscoveries, 11 new species records, and two new varietal records
have been uncovered from (or can be found in) the Nee Soon catchment. The Nee Soon
freshwater swamp forest is one of Singapore’s most valuable botanical areas, and warrants
sustained conservation effort and study.
Keywords. Floristic value, freshwater swamp forest, new records, rediscoveries
“...pockets of vegetation remain and from these botanists may pick up and extend
where I, perforce, have withdrawn.”
E.J.H. Corner (1978: 1)
Introduction
In a supplement of this journal in 1978, the eminent botanist Edred Jo hn Henry Corner
described his observations on the unique freshwater swamp forests in the far south of
Peninsular Malaysia and Singapore. Most of his observations and collections were
made from opportunities provided by the felling of such forest in the 1930s. This
50
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
deforestation continued unabated in the decades that followed, almost to completion.
A wealth of research has been written and published about the Amazonian wetland
forests (e.g. Junk et al., 2010), but in comparison, few studies have been made since
Corner’s seminal work to update our knowledge of tropical freshwater swamp forests
in this part of the world (Turner et al., 1996a) where they are no less special. The
recent literature on freshwater swamp forests in insular Southeast Asia is focused
predominantly on peat swamp forests (swamp forests with a deep layer of ombrogenous
peat) because of interests in preserving the massive below ground carbon stocks and
avoiding transboundary air pollution arising from the burning of this particular swamp
forest type (Yule, 2010; Posa et al., 2011). There is no historical evidence of true peat
swamp forests in Singapore (see also Nguyen et al., 2018), therefore in this article we
exclude peat swamps in our use of the term ‘freshwater swamp forest’.
Freshwater swamp forest was suggested by Corlett (1991) to have once
constituted 5% of the vegetation of Singapore prior to modern human settlement. The
swamp forests at Mandai Road and Jurong were among the main study sites of Corner
(1978). Most of the Mandai swamp was flooded to become part of the Upper Seletar
Reservoir, while the Jurong swamp was in-filled to be used as industrial land. Today,
the freshwater swamp forest at Nee Soon is the country’s last substantial remnant of this
forest type (Turner et al., 1996a). This paper presents a portion of our research focusing
on the vegetation and plant communities in the Nee Soon catchment, conducted as part
of a larger project (Davison et al., 2018) to understand the biodiversity and hydrology
of the Nee Soon freshwater swamp forest that is within this catchment.
Prior to our study, 16 vascular plant species that were presumed Nationally
Extinct in Singapore in the most recent edition of the Singapore Red Data Book
(ferns and fern allies: B.C. Tan et al., 2008; seed plants: H.T.W. Tan et al., 2008)
were rediscovered in the Nee Soon catchment and published in various sources (Table
1). One new species record, Hoy a caudata Hook.f. (Apocynaceae), was reported by
Rodda & Ang (2012). Therefore, the Nee Soon catchment was already a known prime
spot for new records and rediscoveries leading Chong et al. (2012) and the authors of
many of the references listed in Table 1 to call for more botanical exploration of the
patch of freshwater swamp forest in this catchment.
Turner et al. (1996a) compiled a list of freshwater swamp vascular plant species
in Singapore from three of the clusters of plots studied by Wong et al. (1994) for trees,
and Turner et al. (1996b) for herbs, that were in the freshwater swamp forest areas
of Nee Soon, and supplemented this with the species recorded by Corner (1978) at
Mandai and Jurong, as well as with past herbarium specimens. Wong et al. (2013)
produced a preliminary checklist of all the land plant species (i.e., including mosses
and liverworts) of the Nee Soon catchment by adding data from newly surveyed plots,
including ten 15 x 15 m plots established in the earlier Phase 1 of this project, and
recently collected herbarium specimens. During Phase 2 of this project, we were able
to re-assess the vascular plant species in Wong et al. (2013) and correct some of the
nomenclature and identifications (for details, see Chong et al., 2016; Lim et al., 2016;
Neo et al., 2016; Tan et al., 2016). The physical, climatic and ecological context of the
Nee Soon freshwater swamp forest is described by Clews et al. (2018).
Floristic value of Nee Soon swamp
51
Table 1 . Sixteen species presumed Nationally Extinct but rediscovered in the Nee Soon swamp
forest after the publication of the 2 nd edition of the Singapore Red Data Book, and documented
by various sources, prior to our study.
S/No.
Species
Family
Habit
Reference
1 .
Aeschynanthus albidus
(Blume) Steud.
Gesneriaceae
Epiphyte
Lok & Tan (2008)
2.
Bulbophyllum
singaporeanum Schltr.
Orchidaceae
Epiphyte
Yam et al. (2010)
3.
Callostylis pulchella (Lindl.)
S.C.Chen & Z.H.Tsi
Orchidaceae
Epiphyte
Lok et al. (2012)
4.
Coelogyne rochussenii De
Vriese
Orchidaceae
Epiphyte
Lok et al. (2011b)
5.
Dendrobium aloifolium
(Blume) Rchb.f.
Orchidaceae
Epiphyte
Ang et al. (2010b)
6.
Dischidia hirsuta (Blume)
Decne.
Apocynaceae
Epiphyte
Rodda et al. (2012)
7.
Ficus delosyce Corner
Moraceae
Strangler
Ang et al. (2014)
8.
Freycinetia javanica Blume
Pandanaceae
Climber
Ang et al. (2012a)
9.
Hetaeria obliqua Blume
Orchidaceae
Herb
Leong & Yam
(2013)
10.
Liparis barbata Lindl.
Orchidaceae
Herb
Lok et al. (2010)
11.
Pinanga simplicifrons (Miq.)
Becc.
Arecaceae
Clumping
palm
Ang et al. (2010a)
12.
Polystachya concreta (Jacq.)
Garay & H.R. Sweet
Orchidaceae
Epiphyte
Lok et al. (2011a)
13.
Pterisanthes cissioides
Blume
Vitaceae
Climber
Yeo et al. (2012)
14.
Renanthera elongata (Blume)
Lindl.
Orchidaceae
Herb
Ang et al. (2011)
15.
Salacca affinis Griff.
Arecaceae
Clumping
palm
Loo (2011)
16.
Trichotosia velutina (Lodd.
ex Lindl.) Kraenzl.
Orchidaceae
Epiphyte
Ang et al. (2012b)
Methods
Our 20 x 20 m vegetation plots were briefly described by Chong et al. (2016). Nine
of the ten plots from the earlier phase of the project were extended and re-surveyed;
one plot was discarded because it fell outside of the new delineation of the Nee Soon
catchment. Using a preliminary digital elevation model developed during the earlier
project phase, we divided the study area into five elevational strata: 0-20, 20-40,
52
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
40-60, 60-80, and 80-100 m. Within each stratum, we generated 20 random points
as candidate locations for our plots. Our aim was to locate half of our plots of each
stratum in ‘wet’ areas (defined by the presence of surface water such as swamp pools or
streams), and half in ‘dry’ areas (without such surface water), taking also into account
the nine plots from the earlier project phase. We then visited each randomly generated
location in turn and established a plot if the conditions of the location satisfied our
targeted plot type. Our ‘wet’ plot conditions were more difficult to sample by random
chance in the study area, therefore in a few cases where the randomly generated
location was near to but not directly over surface water, we shifted the plot location.
The index numbers of the plots that were shifted are suffixed by an ‘a’ (Table 2).
In each plot, all vascular plant species present were recorded. Where field
identification was not possible, a voucher specimen was collected to be used for
further investigation. All woody stems >5 cm diameter at a height of 1.3 m from the
ground (i.e. diameter at breast height or DBH) were also measured; again, where field
identification was not possible, an attempt was made to collect leafy twigs as voucher
specimens. When branches were too high to be collected even with extendable pruners
(with about 6 m reach), we observed the leaves through a pair of high-powered (10 x 50)
binoculars, or photographed them with a telescopic lens, and then searched for fallen
leaves that matched these on the ground. Sometimes, if shorter, younger individuals
deemed to be of the same species were present nearby and had accessible branches,
we collected these instead. When the tree crown was too high for the leaves to be
viewed clearly even through binoculars or a telescopic lens, or where infestation by
climbers obscured the visibility of the leaves, scrapings of the inner bark and sapwood
were taken and passed to another project team to extract DNA barcodes (see Kutty et
al., 2018) as a last resort to provide a putative identification to family or genus. The
outer bark, inner bark and sapwood characteristics, together with observations of latex,
resin, or odour, were then used to further narrow down the identity to probable species
if possible, or were used to complement the identification with leafy twigs or fallen
leaves.
The fresh voucher specimens were pressed and dried in an oven at 60°C for
a few days. To identify the specimens, we consulted published taxonomic keys and
descriptions, and matched them with other specimens deposited in the Singapore
Botanic Gardens’ Herbarium (SING) which had been determined by visiting experts.
Arising from our investigations with fresh and herbarium specimens, we have started
a series of field guides to various families of trees in the Nee Soon catchment:
Lauraceae (Chong et al., 2016), Cratoxylum Blume (Hypericaceae; Neo et al., 2016),
Myristicaceae (Lim et al., 2016), and Xanthophyllum Roxb. (Polygalaceae; Tan et al.,
2016). In this paper, we will focus on reporting the rediscoveries, new records, and
summary statistics of the flora by family and conservation status. A complete set of
vouchers, i.e., containing at least one specimen of every species identified from our
plots, was deposited with the Herbarium, Lee Kong Chian Natural History Museum,
National University of Singapore (SINU). Collection numbers for the first and second
phases of the project begin with ‘NSSF1’ and ‘NSSF2’ respectively. If the collections
were made from within our plots, this was followed by the plot location; if the specimen
Floristic value of Nee Soon swamp
53
Table 2. The categorisation of plots according to preliminary elevational strata and hydrological
conditions.
Elevation (m) Wet plots
Dry plots
Phase 1
Q3, Q4, Q6, Q9, Q10
Q1,Q2, Q7, Q8
Phase 2
0-20
Q104, Q105, Qlll, Q112
Q101, Q102, Q107, Q109
20-40
Q203a, Q206, Q209, Q220
Q204, Q208, Q213, Q217
40-60
Q301, Q308a, Q311a, Q319a, Q320a
Q302, Q305, Q306, Q307
60-80
Q404a, Q414a
Q405, Q408
80-100
Q504
Q509
Total number
21
19
was collected from a measured woody stem, the collection number contained a ‘T’
followed by the serial number of the stem (e.g., see Appendix 1). A map of the final
plot locations (Fig. 1) was printed on the back of the specimen labels of our voucher
specimens for ease of reference.
Rediscoveries
Of the species that were listed as presumed Nationally Extinct in the 2nd edition of
the Singapore Red Data Book (B.C. Tan et al., 2008; H.T.W. Tan et al., 2008), 61 were
later “rediscovered”, based on herbarium specimens collected after its publication.
These 61 rediscoveries were compiled by Chong et al. (2012). We recorded 21 of these
61 species in our plots (Table 3). We also recorded: Ampelocissus cinnamomea (Wall,
ex M.A.Lawson) Planch., which was recently rediscovered in the Bukit Timah Nature
Reserve (Ng et al., 2014), and Globba pendula Roxb. (Zingiberaceae), which was also
recently reported as rediscovered by Niissalo et al. (2017). Their presence in Nee Soon
adds a new locality for these two species as the earlier records were all from other parts
of Singapore. Another species, Plocoglottis lowii Rchb.f. (Orchidaceae), was first
rediscovered and reported from Upper Seletar Reservoir at a locality just outside of
the Nee Soon catchment (Niissalo et al., 2016), but the lead author later found another
population near one of our plots: Q306 (M.A. Niissalo, personal communication).
Aeschynanthus pulcher (Blume) G.Don ( ^Aeschynanthus parvifolius R.Br.) was also
reported as rediscovered from the Nee Soon swamp forest by Williams (2014). We
did not encounter Plocoglottis lowii and Aeschynanthus pulcher ourselves during our
study.
We provide brief accounts below of the rediscovery of twelve more presumed
Nationally Extinct species from our plots, including recent unreported collections by
others. Collection details are in Appendix 1.
54
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 3. Twenty-one of the 61 rediscoveries compiled by Chong et al. (2012) were collected in
our plots. The table shows their presence in the Nee Soon catchment in this study. Occurrence
refers to the number of plots where the species was present, while the number of stems refers
to the number of woody stems > 5 cm DBH.
S/No.
Species
Family
Habit
Occurrence
No. of
stems
1.
Actinodaphne macrophylla
(Blume) Nees
Lauraceae
tree
4
1
2.
Aglaia elliptica Blume ssp.
elliptica
Meliaceae
tree
2
1
3.
Dalbergia parviflora Roxb.
Fabaceae
climber
1
4.
Dichapetalum sordidum
(Hook.f.) Leenh.
Dichapetalaceae
climber
1
5.
Dioscorea orbiculata Hook.f.
var. tenuifolia (Ridl.) Thapyai
Dioscoreaceae
climber
9
6.
Dracaena singapurensis Ridl.
Asparagaceae
shrub
1
7.
Fagraea splendens Blume
Gentianaceae
epiphyte
4
8.
Friesodielsia glauca (Hook.f.
& Thomson) Steen.
Annonaceae
climber
9
9.
Grenacheria amentacea
(C.B.Clarke) Mez
Primulaceae
climber
9
10.
Helicia excelsa (Roxb.)
Blume
Proteaceae
tree
2
11.
Hypserpa nitida Miers
Menispermaceae
climber
2
12.
Knema glaucescens Jack
Myristicaceae
tree
3
2
13.
Meliosma pinnata (Roxb.)
Maxim, ssp. ridleyi (King)
Beusekom
Sabiaceae
tree
1
14.
Neesia malayana Bakh.
Malvaceae
tree
4
5
15.
Nephelium laurinum Blume
Sapindaceae
tree
6
5
16.
Rourea acutipetala Miq. ssp.
acutipetala
Connaraceae
climber
6
17.
Salacia maingayi
M.A.Lawson
Celastraceae
climber
2
18.
Strychnos axillaris Colebr.
Loganiaceae
climber
1
19.
Syzygium kunstleri (King)
Bahadur & R.C. Gaur
Myrtaceae
tree
1
1
20.
Syzygium scortechinii (King)
Chantar. & J.Pam.
Myrtaceae
tree
1
21.
Uncaria attenuata Korth.
Rubiaceae
climber
1
Floristic value of Nee Soon swamp
55
0 0.1 0.2 0.4 Kilometers
L-U.X. .1 J
Q414a
Fig. 1 . Map of plot locations printed on the back of the labels of our deposited voucher
specimens.
56
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
1 . Aglaia tenuicaulis Hiern (Meliaceae)
Listed as present in Singapore in several treatments (e.g. Pannell, 1989; Panned, 2013)
without any specimens cited. Ridley (1900) had indicated that it was “perhaps an
error”. The only specimen of this species we know of that is collected from Singapore
is deposited in the herbarium of the Royal Botanic Gardens at Kew (. Lobb s.n.; cited
in Panned, 1992). The two collections made in our study were identified as Aglaia
tenuicaulis based on the reddish-brown stellate hairs interspersed with pale-brown
stellate hairs and scales present on the lamina below. The twig apices are also densely
covered in reddish-brown stellate hairs. This species has large compound leaves up to
1.3 m long, with 3-5 large leaflets each up to 45 cm long and 14 cm wide, and 16-19
pairs of lateral veins. According to Panned (1989), this is a small tree common in the
Malay Peninsula, in lowland or hid dipterocarp forest.
2 . Baccaurea macrophylla (Mull.Arg.) Mull.Arg. (Phyllanthaceae)
This species was most recently collected from MacRitchie Reservoir {Corner s.n.,
29 Oct 1944 [SING0012586]). It was also previously collected from Bukit Mandai
{MohdNoor s.n., 28 Nov 1917 [SING0012585]). As suggested by its name, Baccaurea
macrophylla is a large-leaved member of the genus. It can be recognised by a distinct
set of characteristics: prominent stipules up to 11 x 11.5 mm in size, stellate hairs,
discoid glands present in a row between secondary veins on its lower leaf surface,
and a papery leaf texture. It is a medium-sized tree and can be found in primary and
secondary rainforests, as wed as peat swamp forests (Haegans, 2000).
3. Callistopteris superba (Backh. exT.Moore)Ebihara&K.Iwats. (Hymenophyllaceae)
Previously collected only once from Bukit Timah {Ridley s.n., 1897 [SING0031767])
and another time from Seletar Woods {Matthew s.n. [SING0032459]). Terrestrial
fern with erect rhizomes and stout roots, winged stipes, and glandular hairs on the
underside of the lamina. In the Singapore Red Data Book, this was listed under the
name Cephalomanes superbum (Backh. ex T.Moore) I.M.Turner; another synonym
used locally is Trichomanes superbum Backh. ex T.Moore (Turner, 1995; Ebihara et
al., 2006).
4 . Deplanchea bancana (Scheff.) Steenis (Bignoniaceae)
A recent collection of this species was made near the Nee Soon pipeline {Leong et al.
SING2013-100, 20 May 2013). Prior to this, it was collected from Kranji {Goodenough
s.n., 18 Dec 1889 [SING0166223]), from one tree in Bukit Timah (Corner, 1988: 176;
Comer s.n., 8 Jul 1938 [SING0004012, SING0004015]), and listed by Corner (1978)
as found in the now-lost freshwater swamp forest at Mandai. It is easily recognised by
its simple, fairly large, whorled leaves covered with light-coloured hairs, with glands
at the cordate base of the lamina. According to Kochummen (1978a: 37), this is a
tad tree with a fluted base and steep buttresses which occurs especially by swampy
streams.
Floristic value of Nee Soon swamp
57
5. Dioscorea stenomeriflora Prain & Burkill (Dioscoreaceae)
According to Turner et al. (1994), one specimen was collected from the Central
Catchment Nature Reserve near the west end of Upper Peirce Reservoir with the
collection number NRS1632, but we are unable to trace this specimen. Otherwise,
there is one collection from Changi (. Ridley s.n., Feb 1894 [SING0010181]). It is a left-
twining climber, with oblong laminas up to 16 cm long, with sagittate bases in younger
leaves. Secondary veins are distinct and widely spaced.
6. Lindsaea repens (Bory) Thwaites var. pectinata (Blume) Mett. ex Kuhn
(Lindsaeaceae)
This was only collected once previously from Bukit Timah {Ridley s.n., 1893
[SING0031423]). A low-climbing or epiphytic fern with long-creeping rhizomes,
quadrangular rachis, and asymmetrical lamina.
7. Syzygium glabratum (DC.) Veldkamp (Myrtaceae)
The most recent collection of this species was from the Raffles College grounds (Md.
Nur SFN 36293, 13 Jun 1939), but previous collections were all from MacRitchie
Reservoir {Corner SFN 36291, 5 Jun 1939), or the Reservoir Jungle {Corner s.n.,
8 May 1936, [SING 0011827], Corner SFN 29225, 31 Mar 1935; Corner s.n., 20
Apr 1933 [SING 0011825, SING 0011835]). It had previously been collected and
listed for Singapore under various names: Eugenia fusiformis Duthie, Eugenia virens
(Blume) Koord. & Valeton and Syzygium gracile (Korth.) Amshoff (see Turner, 1995;
Veldkamp, 2003). Syzygium glabratum can be identified by well-spaced main lateral
veins raised on both sides of the lamina, and extremely dense, minute, dark glandular
pits or dots over the entire lower lamina surface, which are nested within the cells
of the fine tertiary reticulations, visible even to the naked eye. It is an uncommon
tree found from Singapore to Peninsular Malaysia, Borneo, Java, and the Philippines
(Kochummen, 1978b; Ashton, 2011).
8 . Syzygium leptostemon (Korth.) Merr. & L.M.Perry (Myrtaceae)
This species was most recently collected from Bukit Timah {Ngadiman 34994, 15 Jun
1938; Ridley 11324, 1903), and also from Ang Mo Kio {Ridley 25, Mar 1889). Syzygium
leptostemon has large elliptic-obovate laminas with well-spaced main lateral veins
that are raised on the underside of the lamina. The twigs are often angular, distinctly
orange-brown (sometimes grey-brown), contrasting in colour with the leaves, which
dry very dark. It is known to be locally common where it occurs, but favours wet
environments such as floodplains, river banks, and swamp forest (Kochummen,
1978b; Ashton, 2011).
9. Syzygium pseudocrenulatum (M.R.Hend.) I.M.Turner (Myrtaceae)
This species was most recently collected in Singapore from Mandai (Corner, 1978;
Corner 28090, 25 Apr 1924) and also from an unspecified locality {Ridley 6232, 1894).
Syzygium pseudocrenulatum is distinct in its thickly leathery, elliptic to oblong-elliptic
58
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
lamina with conspicuous glandular pits or dots visible on both sides and slightly crenate
margins. It is endemic to Peninsular Malaysia and Singapore (Kochummen, 1978b).
10. Trigoniastrum hypoleucum Miq. (Trigoniaceae)
This species was most recently collected from Bukit Timah Nature Reserve ( Khoo
KMS 105, 17 Jul 2009), as were previous collections {Ngadiman SFN 36148, 21 May
1940; Ngadiman 34965 , 5 May 1938; Ridley 10379, 1899). The sole member of its
genus, this species is recognised by its simple, alternate leaves, which have whitish
lower lamina surfaces owing to a spiderweb-like covering of appressed hairs, and
distinct, finely-reticulate venation. Also diagnostic are the minute impressed glands
that line and thicken the lamina margins and apex. Trigoniastrum hypoleucum is
also found in lowland rainforests in Peninsular Malaysia, Borneo, and Sumatra (Van
Steenis, 1949: 58-60).
11. Uvaria curtisii King (Annonaceae)
This species was most recently collected from multiple localities: Upper Peirce
Reservoir ( Gwee SING 2009-461, 17 Nov 2009; SING 2010-127, 19 Jan 2010),
MacRitchie Reservoir ( Gwee SING 2009-602, 15 Dec 2009), and Bukit Timah ( Gwee
SING 2010-097, 5 Jan 2010). Prior to this, it was collected in Bukit Timah (, Sinclair SF
39652, 27 May 1953). The oblong or oblanceolate laminas of this large woody climber
are covered with soft, rusty-brown, stellate hairs usually 0.5 mm (and not more than 1
mm) long. Uvaria curtisii is otherwise known only from Peninsular Malaysia (Sinclair,
1955: 206-207.)
12. Uvaria lobbiana Hook.f. & Thomson (Annonaceae)
A recent collection of this species was made from the Nee Soon forest (. Lua SING
2011-237, 22 Jun 2011). Prior to this, it was last collected from the Singapore Botanic
Gardens’ Jungle {Ridley 9211, 1898). Uvaria lobbiana is a large woody climber with
red flowers. Its oblong or oblanceolate laminas have a slightly emarginate base, and are
glabrous above with sparse stellate hairs below. Uvaria lobbiana is otherwise known
from Myanmar, Thailand, Peninsular Malaysia, Sumatra, and Borneo (Sinclair, 1955:
208-210; Turner, 2012).
We also encountered Elaeocarpus griffithii (Wight) A. Gray (Elaeocarpaceae) which
was also presumed Nationally Extinct in the Singapore Red Data Book. However,
there are relatively recent collections from Nee Soon ( Nura et al. NK 207, 26 Jan 1995;
NK 227, 26 Feb 1995) that were previously overlooked because the species name for
these specimens in the database utilised during the listing exercise was “Elaeocarpus
stipularis ”. Hence this should be considered more as an error of the listing rather
than a true rediscovery (see Chong et al., 2012). Like the more common Elaeocarpus
petiolatus (Jack) Wall., the apical shoot tips are coated with resin; however, the
underside of the dried leaf lamina is not visibly covered with scattered black dots,
and the tertiary venation is densely transverse to the midrib, forming rows of neat
cells. Corner (1988) also cautioned that it could be easily confused with Elaeocarpus
Floristic value of Nee Soon swamp
59
polystachyus Wall, ex MiilLBerol., except that its leaves are generally glabrous and
the margins are usually entire or sometimes slightly toothed (rather than hairy and
crenate as in Elaeocarpus polystachyus).
We also collected Alangium ridleyi King, Gynochthodes rigida (Miq.) Razafim.
& B.Bremer (= Morinda rigida Miq.), and Willughbeia coriacea Wall., all of which
were considered by Chong et al. (2012) to be erroneously extinct listings. Alangium
ridleyi was later reassessed formally by Wijedasa et al. (2014) as Critically Endangered.
In the case of Willughbeia coriacea , it was mistakenly considered a synonym of
Willughbeia edulis Roxb. by Turner (1995), and subsequently presumed Nationally
Extinct under this name.
Additions to the flora of Singapore
Securidaca philippinensis Chodat in the Polygalaceae (Tan et al., 2016), and
Cryptocarya nitens (Blume) Koord. & Valeton and Lit sea resinosa Blume, both in
the Lauraceae (Chong et al., 2016), have been recently reported as new or overlooked
records for the flora of Singapore. We also encountered Hopea ferruginea Parijs
(Dipterocarpaceae) and Sindora echinocalyx Prain (Fabaceae), which will be among
the new or overlooked records for Singapore from Bukit Timah soon to be reported
elsewhere (M.S. Khoo, S.C. Chua, personal communications); Aglaia erythrosperma
Pannell (Meliaceae), which was already reported by Pannell (2013) for Singapore,
\/
although not explicitly stated as a new record; and Hanguana neglecta Skornick.
& Niissalo (Hanguanaceae), which was first reported for Singapore by Niissalo et
V
al. (2014). Leong-Skornickova & Boyce (2015) were subsequently able to locate
Hanguana neglecta in MacRitchie and Bukit Timah but not Nee Soon, although Nee
Soon is a historical locality based on re-determined past herbarium records.
The following three additions to the flora of Singapore from our surveys in Nee
Soon have not yet been reported:
1. Aglaia yzermannii Boerl. & Koord. (Meliaceae)
This is a small rheophytic tree to 5 m tall. The species is easily recognised by its
slender twigs and petioles which are sparsely or densely covered in yellowish-brown
stellate scales that appear as shiny coppery dots to the unaided eye, and 3-5 linear,
linear-lanceolate or narrowly lanceolate leaflets which are sometimes irregularly
curved. According to Pannell (2013: 155), this species “appears to be restricted to the
banks of relatively deep stretches of otherwise stony, fast flowing rivers” in Peninsular
Malaysia. However, our collections of this species were not made from such habitats
in Nee Soon.
2. Dacryodes incurvata (Engl.) H.J.Lam (Burseraceae)
A small to medium-sized tree. It was once collected from Mandai Road (Kiah s.n., 29
Jul 1940 [SING0053970]) and identified as Santiria laevigata Blume, to which it bears
60
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
a resemblance except for the leaflet margins that are frequently incurved, terminal
buds that are not covered in resin, and petiolules that do not dry black at either ends.
The specimen was later re-determined as Dacryodes incurvata by K.M. Kochummen.
We collected one specimen from one of our plots.
3. Melanochyla angustifolia Hook.f. (Anacardiaceae)
This species was also recently collected from the back of the Seletar Range near Old
Upper Thomson Road (Ali s.n., 1 Mar 2003 [SING0052434]), which is the eastern
side of the Nee Soon catchment. Distinguishable when dried due to a grooved midrib
on the underside and wrinkled petiole. The lamina base is cuneate and the leaves are
well-spaced and not clustered.
We also collected and identified two varieties that are new to Singapore:
Knema curtisii (King) Warb. var. curtisii (Myristiceae), which was reported by Lim
et al. (2016), and Syzygium claviflorum (Roxb.) Wall, ex A.M.Cowan & Cowan var.
maingayi (Duthie) Chantar. & J.Parn. (Myrtaceae). Syzygium claviflorum var. maingayi
can be distinguished from the typical variety by its four-angled and winged twigs
(Kochummen, 1978b).
Floristics
By the end of the project, we identified 671 species (or sub-specific taxa, referred
to here as ‘species’ for ease of presentation; Table 4), of which almost all are native
(98.4%), and most have been assigned a nationally threatened status (i.e., Nationally
Vulnerable, Endangered or Critically Endangered; 72.5%). This does not include
the rediscoveries or new records which would require new national conservation
assessments. The latter also includes two species which require new assessments due
to a recent clarification in their identification ( Trichosanthes elmeri Merr. had been
previously identified as Trichosanthes celebica Cogn.; de Wilde & Duyfjes, 2010) or
taxonomy ( Utania volubilis (Wall.) Sugumaran; Sugumaran & Wong, 2014). Slightly
fewer than half (43.1%) of the species are trees represented by at least one stem > 5 cm
DBH, with similar percentages in the various status categories (Table 4).
The species are from 117 families. While the distribution of species among
families in the Nee Soon catchment generally reflects that of the extant Singapore flora
(Chong et al., 2011) and especially the forest flora (Turner, 1994), the Myristicaceae is
especially well-represented, with 25 species recorded out of the 36 species considered
to be native to Singapore (Chong et al., 2009; Lim et al., 2016), of which 22 species
occurred within our 40 plots (Table 5). Sixty families are represented by at least one
stem > 5 cm DBH, and Gynotroches axillaris Blume (Rhizophoraceae) is the most
abundant tree species in terms of stem counts, followed by Baccaurea bracteata Mull.
Arg. (Phyllanthaceae) and Oncosperma horridum (Griff.) Scheff. (Arecaceae).
Floristic value of Nee Soon swamp
61
Table 4. Number of species and tree species (> 5 cm DBH) from the vegetation plots in each
national conservation status (natives) or invasive status (non-natives).
Category
No. of species
No. of tree species
Native
660
285
Not threatened
120
50
Vulnerable
144
55
Endangered
111
53
Crit. End.
232
106
Not assessed
—Previously presumed Extinct
38
13
—New records
13
8
—Others
2
0
Exotic
8
3
Cultivated only
1
0
Casual
4
2
Naturalised
3
1
Cryptogenic Weed
3
0
Total
671
288
Discussion
The Nee Soon catchment harbours 53 presumed Nationally Extinct species according
to the most recent Singapore Red Data Book, as well as 11 species records and two
varieties which are new to or had been overlooked in previous checklists of the
Singapore flora (e.g., Chong et al., 2009). Our findings reinforce published opinions
(e.g., Turner et al., 1994; Turner et al., 1996a) that the Nee Soon Swamp Forest merits
the highest conservation priority in Singapore, and is of comparable floristic value to
the more well-known Bukit Timah Nature Reserve.
It does not necessarily follow that one should automatically confer upon
rediscovered species the status of Critically Endangered. One example is Uvaria
curtisii, which has been collected from multiple localities in the nature reserves in
recent years. Another example from previous rediscoveries is Grenacheria amentacea
(C.B.Clarke) Mez, which appears to be fairly widespread in Nee Soon (Table 3) and also
in the part of the Central Catchment Nature Reserve along Mandai Road (K. Y. Chong,
personal observations). It is likely that these species belong to poorly documented
62
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 5. Top ten families of vascular plants in the Nee Soon catchment according to the total
number of species, the number of tree species (> 5 cm DBH), stem abundance, and total basal
area of trees. Ranks are given in superscript next to the counts.
Family
No. of species
No. of tree
species
No. of stems
Total basal
area
Rubiaceae
41 1
15
5
136
3
3.93% 8
Annonaceae
40 2
12
7
32
2.00%
Myrtaceae
31 3
20
3
124
5
6.80% 4
Phyllanthaceae
28 4
21
2
217
1
4.03% 7
Moraceae
28 5
9
9
23
1.70%
Myristicaceae
26 6
23
1
126
4
8.26% 3
Euphorbiaceae
22 7
14
6
145
2
5.18% 6
Lauraceae
22 7
17
4
56
9
2.54%
Apocynaceae
19 9
6
32
2.36%
Fabaceae
17 10
6
35
1.53%
Dipterocarpaceae
12
11
8
33
6.10% 5
Burseraceae
11
9
9
27
1.27%
Clusiaceae
10
8
53
10
1.18%
Anacardiaceae
8
7
73
7
8.95% 1
Arecaceae
8
2
73
7
3.19% 10
Sapindaceae
7
4
41
3.59% 9
Rhizophoraceae
3
3
112
6
2.84%
Combretaceae
1
1
1
8.67% 2
Floristic value of Nee Soon swamp
63
or difficult-to-identify taxa that tend to go unnoticed by generalist collectors. While
they would almost certainly qualify as threatened, a proper nation-wide conservation
assessment can be made only after targeted search efforts with a broader geographic
coverage has been carried out.
Given that most of our collections are sterile, it is possible that some of our
identifications may be erroneous. The collection numbers of most of the voucher
specimens of the rediscovered species or new records deposited in the herbaria (see
Appendix 1) therefore contain the plot numbers so that the respective individuals may
be revisited for flowers or fruits. All stems > 10 cm DBH within our plots were also
mapped to positions of 0.5 m resolution so some of the tree species can also be tracked
down to the individuals we encountered; these maps, the DBH measurements, and the
geographic coordinates of the plots, are available upon request.
The floristics presented in Turner et al. (1996a) were based largely on three
clusters of plots with a total area of 0.6 ha. On the other hand, our study consisted
of 40 plots sampled across the different elevation zones throughout the Nee Soon
catchment, and totalled 1.6 ha. Our sampling is more extensive, and less biased than
past opportunistic botanical collections, and therefore the floristic composition we
present here is the most representative of the catchment so far—with the caveat that we
focused more on the forest understorey and trees, and thus may have under-sampled
canopy lianas (which were incompletely identified due to lack of access to leaves), and
crown epiphytes. More detailed analyses of soil, hydrology, and the plant communities
will be presented in future manuscripts.
In Singapore, the phrase ‘filling of swamps’ is evocative of the city-state’s
determination to survive at all costs during its early post-independence years, and
is associated with its industrial and developmental success today. Singapore is not
unique in its past treatment of swamps; economic growth is a key driver of the loss of
wetlands worldwide (van Asselen et al., 2013), with two-thirds of the world’s wetlands
having been lost since 1900 (Davidson, 2014). On the one hand, the recent cache of
rediscoveries and new records from Singapore’s last substantial tract of freshwater
swamp forest in the Nee Soon catchment provides some relief that some of this
natural heritage is safe within a protected area, and hope that more might be found;
on the other hand, it is sobering to imagine what other biological treasures must have
been lost when swamp forests such as those at Mandai and Jurong were inundated or
reclaimed. Going forward, no effort must be spared to ensure that the Nee Soon swamp
forest will continue to persist, and perhaps some effort should even be made to recreate
this habitat type elsewhere in Singapore.
ACKNOWLEDGEMENTS. This work was conducted as part of the Nee Soon Swamp Forest
Biodiversity and Hydrology Baseline Studies—Phase 2 Project funded by the National Parks
64
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Board, Singapore, under the permit number RP13-009. We thank the other research assistants
(Koh Choon Yen, Ng Wen Qing) and our interns (Jake Gonzales, Fiona Chong, Wong Junpeng,
Ahmad Muqit bin Mohamed Sulaimi, and Soong Le Xuan) for their contribution to the project.
We would also like to thank the herbarium staff of the Singapore Botanic Gardens, especially
V
David Middleton, Ali Ibrahim, Paul Leong, Low Yee Wen, Jana Leong-Skornickova, Wong
Khoon Meng, as well as other NParks staff: Ang Wee Foong, Gwee Aik Teck and Stuart
Lindsay, for discussions and assistance with identifications; Siti Nur Bazilah for facilitating
access to SING; and Chua Keng Soon for facilitating access to and deposition in SINU.
References
Ang, W.F., Lok, A.F.S.L. & Tan, H.T.W. (2010a). Rediscovery in Singapore of Pinanga
simplicifrons (Miq.) Becc. (Arecaceae). Nat. Singapore 3: 83-86.
Ang, W.F., Lok, A.F.S.K., Yeo, C.K., Tan, S.Y. & Tan, H.T.W. (2010b). Rediscovery of
Dendrobium aloifolium (Blume) Rchb.f. (Orchidaceae) in Singapore. Nat. Singapore
3: 321-325.
Ang, W.F., Lok, A.F.S.L., Yeo, C.K., Angkasa, A., Ng, P.X. Ng, B.Y.Q. & Tan, H.T.W. (2011).
Rediscovery of Renanthera elongata (Blume) Lindl. (Orchidaceae) in Singapore. Nat.
Singapore 4: 297-301.
Ang, W.F., Lok, A.F.S.L., Yeo, C.K., Angkasa, A., Ng, P.X. & Tan, H.T.W. (2012a). Rediscovery
of Freycinetia javanica Blume (Pandanaceae) in Singapore. Nat. Singapore 5: 117-122.
Ang, W.F., Yeo, C.K., Lok, A.F.S.L., Angkasa, A., Ng, P.X. & Tan, H.T.W. (2012b). Rediscovery
of Trichotosia velutina (Lodd. ex Lindl.) Kraenzl. (Orchidaceae) in Singapore. Nat.
Singapore 5: 199-204.
Ang, W.F., Yeo, C.K., Ooi, Z.Y., Lok, A.F.S.L., Abdullah, I.H. & Tan, H.T.W. (2014).
Rediscovery in Singapore of Ficus delosyce Corner (Moraceae). Nat. Singapore 7:
167-172.
Ashton, PS. (2011). Myrtaceae. In: Soepadmo, E., Saw, L.G., Chung, R.C.K. & Kiew, R.
(eds) Tree Flora of Sabah and Sarawak , vol. 7, pp. 87-330. Selangor: Forest Research
Institute Malaysia, Sabah Forestry Department & Sarawak Forestry Department,
van Asselen, S., Verburg, P.H., Vermaat, J.E. & Janse, J.H. (2013). Drivers of wetland
conversion: a global meta-analysis. PLoS ONE 8: e81292.
Chong, K.Y., Tan, H.T.W. & Corlett, R.T. (2009). A Checklist of the Total Vascular Plant Flora
of Singapore: Native, Naturalised and Cultivated Species. Singapore: Raffles Museum
of Biodiversity Research, National University of Singapore.
Chong, K.Y., Tan, H.T.W. & Corlett, R.T. (2011). A summary of the total vascular plant flora of
Singapore. Gard. Bull. Singapore 63: 197-204.
Chong, K.Y., Lee, S.M.L., Gwee, A.T., Leong, P.K.F., Samsuri Ahmad, Ang, W.F., Lok,
A.F.S.L., Yeo, C.K., Corlett, R.T. & Tan, H.T.W. (2012). Herbarium records do not
predict rediscovery of presumed nationally extinct species. Biodivers. Conserv. 21:
2589-2599.
Chong, K.Y., Neo, L., Tan, S.Y., Koh, C.Y., Lim, R.C.J., Loh, J.W., Ng, W.Q., Seah, W.W., Yee,
A.T.K. & Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon Swamp
Forest (I): Lauraceae. Nat. Singapore 9: 1-28.
Floristic value of Nee Soon swamp
65
Clews, E., Corlett, R.T., Ho, J.K.I., Koh, C.Y., Liong, S.Y., Memory, A., Ramchunder, S.J.,
Siow, H.J.M.P., Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W., Theng, M.T.Y. & Yeo,
D.C.J. (2018). The biological, ecological and conservation significance of freshwater
swamp forest in Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
Corlett, R.T. (1991). Vegetation. In: Chia, L., Ausafur, R. & Tay, D. (eds) The Biophysical
Environment of Singapore, pp. 134-154. Singapore: Singapore University Press.
Corner, E.J.H. (1978). The freshwater swamp-forest of south Johore and Singapore. Gard. Bull.
Singapore, Suppl. 1. Singapore: Botanic Gardens, Parks & Recreation Department.
Corner, E.J.H. (1988). Bignoniaceae. In: Wayside Trees of Malaya, 3 rd ed., vol. 1, pp. 172-184.
Kuala Lumpur: Malayan Nature Society.
Davidson, N.C. (2014). How much wetland has the world lost? Long-term and recent trends in
global wetland area. Mar. Freshwater Res. 65: 934-941.
Davison, G.W.H., Cai, Y.X., Li, T.J. & Lim, W.H. (2018). Integrated research, conservation and
management of Nee Soon freshwater swamp forest: hydrology and biodiversity. Gard.
Bull. Singapore 70 (Suppl. 1): 1-7.
de Wilde, W.J.J.O. & Duyfjes, B.E.E. (2010). Trichosanthes. In: Nooteboom, H.P. (ed.) Flora
Malesiana Series I - Seed Plants, vol. 19, pp. 231-318. Leiden: Netherlands Centre for
Biodiversity Naturalis.
Ebihara, A., Dubuisson, J.-Y., Iwatsuki, K., Hennequin, S. & Ito, M. (2006). A taxonomic
revision of Hymenophyllaceae. Blumea 51: 221-280.
Haegens, R.M.A.P. (2000). Taxonomy, phylogeny, and biogeography of Baccaurea,
Distichirhops, and Nothobaccaurea (Euphorbiaceae). Blumea, Suppl. 12: 1-216.
Junk, W.J., Piedade, M.T.L., Wittmann, E, Schongart, J. & Parolin, P. (2010). Amazonian
Floodplain Forests: Ecophysiology, Biodiversity and Sustainable Management.
London: Springer.
Kochummen, K.M. (1978a). Bignoniaceae. In: Ng, L.S.P. (ed.) Tree Flora of Malaya, vol. 3,
pp. 36-44. Kuala Lumpur: Longman Malaysia.
Kochummen, K.M. (1978b). Myrtaceae. In: Ng, L.S.P. (ed.) Tree Flora of Malaya, vol. 3, pp.
169-254. Kuala Lumpur: Longman Malaysia.
Kutty, S.N., Wang, W., Ang, Y., Tay, Y.C., Ho, J.K.I. & Meier, R. (2018). Next-generation
identification tools for Nee Soon freshwater swamp forest, Singapore. Gard. Bull.
Singapore 70 (Suppl. 1): 155-173.
Leong, P.K.L. & Yam, T.W. (2013). The rediscovery of Hetaeria obliqua in Singapore. Malesian
Orchid J. 12: 23-28.
Leong-Skornickova, J. & Boyce, PC. (2015). Hanguana in Singapore demystified: an overview
with descriptions of three new species and a new record. Gard. Bull. Singapore 67:
1-28.
Lim, R.C.J., Chong, K.Y., Neo, L., Tan, S.Y., Koh, C.Y., Loh, J.W., Ng, W.Q., Seah, W.W., Yee,
A.T.K. & Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon Swamp
Lorest (III): Myristicaceae. Nat. Singapore 9: 69-138.
Lok, A.L.S.L. & Tan, H.T.W. (2008). Rediscovery of Aeschynanthus albidus (Blume) Steud.
(Gesneriaceae) in Singapore. Nat. Singapore 1: 5-8.
Lok, A.L.S.L., Ang, W.L., Chong, K.Y. & Tan, H.T.W. (2010). Rediscovery of Fiparis barbata
Lindl. (Orchidaceae) in Singapore. Nat. Singapore 3: 277-281.
Lok, A.L.S.L., Ang, W.L. & Tan, H.T.W. (2011a). The rediscovery in Singapore of Polystachya
concreta (Jacq.) Garay & H.R. Sweet (Orchidaceae). Nat. Singapore 4: 19-24.
66
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Lok, A.F.S.K., Ang, W.F., Chong, K.Y., Yeo, C.K. & Tan, H.T.W. (2011b). Rediscovery in
Singapore of Coelogyne rochussenii de Vriese (Orchidaceae). Nat. Singapore 4: 49-53.
Lok, A.F.S.L., Ang, W.F. & Yeo, C.K. (2012). Rediscovery of Callostylis pulchella (Lindl.)
S.C.Chen & Z.H.Tsi (Orchidaceae) in Singapore. Nat. Singapore 5: 205-209.
Loo, A.H.B. (2011). Rediscovery in Singapore of Salacca affinis Griff. (Arecaceae). Nat.
Singapore 4: 123-126.
Neo, L., Chong, K.Y., Tan, S.Y., Koh, C.Y., Lim, R.C.J., Loh, J.W., Ng, W.Q., Seah, W.W., Yee,
A.T.K. & Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon Swamp
Forest (II): Cratoxylum (Hypericaceae). Nat. Singapore 9: 29-39.
Ng, X.Y., Lim, R.C.J., Ang, W.F., Ong, K.H. & Yeo, C.K. (2014). The conservation status in
Singapore of Ampelocissus cinnamomea (Wall, ex M.A.Lawson) Planch. (Vitaceae).
Nat. Singapore 7: 129-134.
Nguyen, C.T.T., Wasson, R.J. & Ziegler, A.D. (2018). The hydro-geomorphic status of the
Nee Soon freshwater swamp forest catchment of Singapore. Gard. Bull. Singapore 70
(Suppl. 1): 33-48.
Niissalo, M.A., Wijedasa, L.S., Boyce, PC. & Leong-Skornickova, J. (2014). Hanguana
neglecta (Hanguanaceae): a new plant species form a heavily collected and visited
reserve in Singapore. Phytotaxa 188: 14-20.
Niissalo, M.A., Williams, C.E., Kurzweil, H. & Thomas, D.C. (2016). Rediscovery in Singapore
of Plocoglottis lowii Rchb.f. (Orchidaceae). Nat. Singapore 9: 41-46.
Niissalo, M.A., Leong-Skomickova, J., Khew, G.S. & Webb, E.L. (2017). Very small relict
populations suggest high extinction debt of gingers in primary forest fragments of a
tropical city. Amer. J. Bot. 104: 1-8.
Pannell, C.M. (1989). Aglaia Lour. In: Ng, F.S.P. (ed.) Tree Flora of Malaya, vol. 4, pp. 207-
230. Selangor: Longman Malaysia.
Panned, C.M. (1992). A taxonomic monograph of the genus Aglaia Lour. (Meliaceae). Kew
Bull, Addit. Ser. 16. United Kingdom: Royal Botanic Gardens, Kew.
Panned, C.M. (2013). Aglaia Lour. In: Kiew, R., Chung, R.C.K., Saw, L.G. & Soepadmo, E.
(eds) Flora of Peninsular Malaysia Series II: Seed Plants, vol. 4, pp. 54-155. Kepong:
Forest Research Institute Malaysia.
Posa, M.R.C., Wijedasa, L.S. & Corlett, R.T. (2011). Biodiversity and conservation of tropical
peat swamp forests. BioScience 61: 49-57.
Ridley, H.N. (1900). The flora of Singapore. J. Straits Branch Roy. Asiat. Soc. 33: 27-196.
Rodda, M. & Ang, W.F. (2012). Hoya caudata Hook.f. (Apocynaceae), a new record for
Singapore, and keys to the Hoya species of Singapore. Nat. Singapore 5: 123-128.
Rodda, M., Tran, H.D., Chew, P.T., Liew, D. & Leong-Skomickova, J. (2012). The rediscovery
of Dischidia hirsuta (Apocynaceae, Asclepiadoideae) in Singapore. Gard. Bull.
Singapore 64: 293-299.
Sinclair, J. (1955). A Revision of the Malayan Annonaceae. Gard. Bull. Singapore 14(2): 149-
516.
Sugumaran, M., & Wong, K.M. (2014). Studies in Malesian Gentianaceae, VI. A revision of
Utania in the Malay Peninsula with two new species. Plant Ecol. Evol. 147: 213-223.
Tan, B.C., Tan, H.T.W., Tan, K.-X., Ad bin Ibrahim, Chew, P.T., Chua, K.S., Gwee, A.T.,
Kiew, R., Lee, S.M.L., Leong, R, Lok, A.F.S.L., Loo, A.H.B., Lum, S.K.Y., Morgany
Thangavelu, Saifuddin bin Suran, Sim, S., Samsuri Ahmad, Wee, Y.C., Yeo, C.K., Yong,
J.W.H. (2008). Ferns and fem allies. In: Davison G.W.H., Ng, P.K.L. & Ho, H.C. (eds)
The Singapore Red Data Book: Threatened plants and animals of Singapore, 2 nd ed., pp.
210-212. Singapore: The Nature Society (Singapore).
Floristic value of Nee Soon swamp
67
Tan, H.T.W., Tan, K.-X., Ali bin Ibrahim, Chew, P.T., Chua, K.S., Duistermaat, H., Ganesan,
S.K., Goh, M.W.K., Gwee, A.T., Kiew, R., Lee, S.M.L., Leong, R, Lim, J., Lok,
A.F.S.L., Loo, A.H.B., Lum, S.K.Y., Morgany Thangavelu, Saifuddin bin Suran, Sim,
S., Samsuri Ahmad, Wee, Y.C., Yap, K.F., Yeo, C.K., Yong, J.W.H. (2008). Seed plants.
In: Davison, G.W.H., Ng, P.K.L. & Ho, H.C. (eds) The Singapore Red Data Book:
Threatened plants and animals of Singapore, 2 nd ed., pp. 213-244. Singapore: The
Nature Society (Singapore).
Tan, S.Y., Chong, K.Y., Neo, L., Koh, C.Y., Lim, R.C.J., Loh, J.W., Ng, W.Q., Seah, W.W., Yee,
A.T.K. & Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon Swamp
Forest (IV): Xanthophyllum (Polygalaceae). Nat. Singapore 9: 139-147.
Turner, I.M. (1994). The taxonomy and ecology of the vascular plant flora of Singapore: a
statistical analysis. Bot. J. Linn. Soc. 114: 215-227.
Turner, I.M. (1995). A catalogue of the vascular plants of Malaya. Gard. Bull. Singapore 47:
1-757.
Turner, I.M. (2012). Annonaceae of Borneo: A review of the climbing species. Gard. Bull.
Singapore 64(2): 371-479.
Turner, I.M., Tan, H.T.W., Chua, K.S. & Metcalfe, D.J. (1994). Recent botanical collections
from the nature reserves of Singapore. Gard. Bull. Singapore 46: 1-36.
Turner, I.M., Boo, C.M., Wong, Y.K., Chew, PT. & Ali bin Ibrahim (1996a). Freshwater swamp
forest in Singapore, with particular reference to that found around the Nee Soon firing
ranges. Gard. Bull. Singapore 48: 129-157.
Turner, I.M., Tan, H.T.W. & Chua, K.S. (1996b). Relationships between herb layer and canopy
composition in a tropical rain forest successional mosaic in Singapore. J. Trop. Ecol.
12: 843-851.
Van Steenis, C.G.G.J. (1949). Trigoniaceae. In: Van Steenis, C.G.G.J. (ed.) Flora Malesiana
Series I - Seed Plants, vol. 4, part 2, pp. 58-60. Groningen: P. Noordhoff Ltd.
Veldkamp, J.F. (2003). Nomenclature of Syzygium gracile (Myrtaceae). Blumea 48: 489-490.
Wijedasa, L., Shee, Z.Q. & Chia, E. (2014). Conservation status and lectotypification of
Alangium ridleyi (Cornaceae) in Singapore. Gard. Bull. Singapore 66(2): 233-239.
Williams, C. (2014) The rediscovery of a presumed Nationally Extinct Aeschynanthus.
Gardenwise 43: 10-11.
Wong, Y.K., Chew, PT. & Ali bin Ibrahim (1994). The tree communities of the Central
Catchment Nature Reserve, Singapore. Gard. Bull. Singapore 46: 37-78.
Wong, H.F., Tan, S.Y., Koh, C.Y., Siow, H.J.M., Li, T., Heyzer, A., Ang, A.H.F., Mirza Rifqi
bin Ismail, A. Srivathsan & Tan, H.T.W. (2013). Checklist of the Plant Species of Nee
Soon Swamp Forest, Singapore: Bryophytes to Angiosperms. Singapore: National Parks
Board and Raffles Museum of Biodiversity Research, National University of Singapore.
Yam, T.W., Leong, P.K.F., Chew, P.T., Liew, D. & Ng, W.K.H. (2010). The re-discovery and
conservation of Bulbophyllum singaporeanum. Gardenwise 35: 15-17.
Yeo, C.K., Ang, W.F. & Lok, A.F.S.L. (2012). Pterisanthes (Vitaceae) of Singapore: with a note
on the rediscovery of Pterisanthes cissioides Blume. Nat. Singapore 5: 185-190.
Yule, C.M. (2010). Loss of biodiversity and ecosystem functioning in Indo-Malayan peat
swamp forests. Biodivers. Conserv. 19: 193-409.
68
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Appendix 1. Collection numbers for rediscoveries or new records mentioned in the text and/
or Table 3, as well as those rediscovered prior to our study (Table 1) that we also found in our
plots. All specimens are deposited at the Herbarium, Lee Kong Chian Natural History Museum,
National University of Singapore (SINU).
Species Collection number(s)
Actinodaphne macrophylla (Blume) Nees
Aglaia elliptica Blume ssp. elliptica
Aglaia erythrosperma Pannell
Aglaia tenuicaulis Hiern
Aglaia yzermannii Boerl. & Koord.
Alangium ridleyi King
Baccaurea macrophylla (Mtill.Arg.) Mull.
Arg.
Callistopteris superba (Backh. ex T.Moore)
Ebihara & K.Iwats,
Cryptocarya nitens (Blume) Koord. &
Valeton
D aery odes incurvata (Engl.) H.J.Lam
Dalbergia parviflora Roxb.
Dendrobium aloifolium (Blume) Rchb.f.
Deplanchea bancana (Scheff.) Steen.
Dichapetalum sordidum (Hook.f.) Leenh.
Dioscorea orbiculata Hook.f. var. tenuifolia
(Ridl.) Thapyai
Dioscorea stenomeriflora Prain & Burk.
Dracaena singapurensis Ridl.
Elaeocarpus griffithii (Wight) Gray
Fagraea splendens Blume
Freycinetia javanica Blume
Friesodielsia glauca (Hook.f. & Thomson)
Steen.
Globba pendula Roxb.
Grenacheria amentacea (C.B.Clarke) Mez
Gynochthodes rigida (Miq.) Razafim. &
B. Bremer
V
Hanguana neglecta Skornick. & Niissalo
Helicia excelsa (Roxb.) Blume
Hopea ferruginea Parijs
NSSF1 - Q1000-5066, NSSF2-Q203a U114,
NSSF2-Q204T27
NSSF2-Q209U80, NSSF2-Q4T59
NSSF2-Q112U61
NSSF2-Q111U132, NSSF2-Q203aU86
NSSF2-Q107U05, NSSF2-Q213U82
NSSF2-Q4T30
NSSF2-Q204T16, NSSF2-Q213U100
NSSF2-Pte32
NSSF2-Q1T54, NSSF2-Q213U72, NSSF2-
Q205U45
NSSF2-Q206T55
NSSF1-Q100-962, NSSF2-Q1U09
NSSF1-Q300-502
NSSF2-Q203aT21
NSSF1-Q1C0-1064
NSSF2-Q102U114, NSSF2-Q204U105,
NSSF2-Q2U04, NSSF2-305U36, NSSF2-
Q308U109
NSSF2-Q101U16
NSSF2-Q101U53, NSSF2-109U99
NSSF2-Q104U54, NSSF2-Q111T58
NSSF2-Q308aU81
NSSF1-Q4C0-866, NSSF2-Q4U129
NSSF2-Q10U126
NSSF2-Q308aU39
NSSF2-Q2U41
NSSF2-Q319aU80
NSSF2-18
NSSF2-Q204U54, NSSF2-Q204U87
NSSF2-Q509U12, NSSF2-Q509T46,
NSSF2-Q509T53
Floristic value of Nee Soon swamp
69
Appendix 1. Continuation.
Species Collection number(s)
Hypserpa nitida Miers
Knema curtisii (King) Warb. var. curtisii
Knema glaucescens Jack
Lindsaea repens (Bory) Thwaites var.
pectinata (Blume) Mett.
Litsea resinosa Blume
Melanochyla angustifolia Hook.f.
Meliosma pinnata (Roxb.) Maxim, ssp.
ridleyi (King) Beusekom
Neesia malayana Bakh.
Nephelium laurinum Blume
Pinanga simplicifrons (Miq.) Becc.
Pterisanthes cissioides Blume
Rourea acutipetala Miq. ssp. acutipetala
Salacia maingayi M.A.Lawson
Securidaca philippinensis Chodat
Sindora echinoccilyx Prain
Strychnos axillaris Colebr.
Syzyginm claviflornm (Roxb.) Wall, ex
A.M.Cowan & Cowan var. maingayi
(Duthie) Chantar. & J.Pam.
Syzygium glabratum (DC.) Veldkamp
Syzygium kunsfieri (King) Bahadur &
R.C.Gaur
Syzygium leptostemon (Korth.) Merr. &
L.M.Perry
Syzygium pseudocrenulatum (M.R.Hend.)
I. M.Turner
Syzygium scortechinii (King) Chantar. &
J. Parn.
Trigoniastrum hypoleucum Miq.
Uncaria attenuata Korth.
Uvaria curtisii King
Uvaria lobbiana Hook.f. & Thomson
Willughbeia coriacea Wall.
NSSF2-Q206U77, NSSF2-Q4U121,
NSSF2-Q4U144
NSSF2-Q111U26, NSSF2-Q203aT40
NSSF2-Q204U50, NSSF2-308aT07
NSSF2-Pte2
NSSF2-Q105U46, NSSF2-Q307U66,
NSSF2-Q311aU120
NSSF2-Q112T31, NSSF2-Q3U155
NSSF2-Q504T46a
NSSF2-Q112T48, NSSF2-Q9T10
NSSF2-Q206U140, NSSF2-Q4U120
NSSF2-94
NSSF2-Q107U82, NSSF2-Q308aU116
NSSF2-Q101U100, NSSF2-Q2U35,
NSSF2-Q404aU43, NSSF2-Q8U96
NSSF1-Q400-826
NSSF2-Q102T27, NSSF2-Q102U124,
NSSF2-Q8U72
NSSF2-Q101 Ulll
NSSF1-Q100-1070
NSSF2-Q217U14, NSSF2-Q302U71
NSSF2-Q101U131, NSSF2-Q111T37,
NSSF2-Q203aT29, NSSF2-Q206U42
NSSF2-Q104T38
NSSF2-Q111T53, NSSF2-Q111U116
NSSF2-Q104U35, NSSF2-Q204U78
NSSF2-Q111U122
NSSF2-Q101T40, NSSF2-Q101U61
NSSF2-Q10U110
NSSF2-Q302U74, NSSF2-Q504U89
NSSF2-Q4U113
NSSF2-Q8U111
Gardens’ Bulletin Singapore 70 (Suppl. 1): 71-108. 2018
doi: 10.26492/gbs70(suppl.l). 2018-05
71
Aquatic macroinvertebrate richness,
abundance and distribution in the Nee Soon
freshwater swamp forest, Singapore
J.K.I. Ho 1 - 2 , R.F. Quek 2 , S.J. Ramchunder 2 - 3 , A. Memory 2 , M.T.YTheng 2 ,
D.C.J. Yeo 1 & E. Clews 2
'Department of Biological Sciences, National University of Singapore,
14 Science Drive 4, 117543 Singapore
jonathanho@nus.edu.sg
2 Tropical Marine Science Institute, National University of Singapore,
18 Kent Ridge Road, 117229 Singapore
3 Department of Geography, National University of Singapore,
1 Arts Link, 117570 Singapore
ABSTRACT. The Nee Soon freshwater swamp forest is a vital area for biodiversity conservation
in Singapore. A survey of the aquatic macroinvertebrates in the streams of the Nee Soon
drainage was carried out to capture a representative sample of the communities present. Here,
we present the different groups of macroinvertebrates sampled as well as their abundance and
distribution within the freshwater swamp forest. An annotated checklist of the orders of the
macroinvertebrates found in the freshwater swamp forest follows, together with information on
their distribution and abundance within the Nee Soon catchment.
Keywords. Aquatic biodiversity, freshwater ecology, invertebrate surveys
Introduction
The Nee Soon freshwater swamp forest is the last substantial area of freshwater swamp
forest found in the island nation of Singapore and serves as a vital conservation area
due to its rich biological diversity (Ng & Lim, 1992; Lim et al., 2011; O’Dempsey &
Chew, 2013; O’Dempsey, 2014). It is also one of the most biologically diverse areas
in Singapore (Lim et al., 2011), especially for freshwater fish and decapod crustaceans
(Ng & Lim, 1992; Ng, 1997; Ng & Lim, 1997; Li et al., 2016). Additionally, the
Nee Soon freshwater swamp forest is also an important conservation area for
other groups of water-dependent (aquatic) macroinvertebrates, also known as the
zoobenthos (Dudgeon, 1999), as these animals are usually found in stream beds.
Among these lesser-known groups are the aquatic larvae of insects such as dragonflies
and damselflies (Odonata), stoneflies (Plecoptera), mayflies (Ephemeroptera), and
caddisflies (Trichoptera). Other groups which can be found among the zoobenthos of
the freshwater swamp forest include leeches and other worms (Annelida) as well as
crustaceans such as the seed shrimps (Ostracoda). Several species from these groups
are found nowhere else in Singapore, such as the potbellied elf dragonfly (Risiophlebia
72
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
dohrni ), and would be extirpated from Singapore should anything happen to the Nee
Soon freshwater swamp forest (Tang et al., 2010).
Currently, information on aquatic macroinvertebrate groups in Singapore is
scarce in the scientific literature, with information existing only for certain groups
(Balkeetal., 1997;Murphy, 1997; Yang etal., 1997; Hendrichetal., 2004). Additionally,
even for relatively charismatic and well-studied groups such as the odonates, available
information deals mostly with the adult stage of these insects (Murphy, 1997; Norma-
Rashid et al., 2008; Tang et al., 2010), with not much being known about their aquatic
larval stages. More obscure groups such as the ostracods and aquatic mites lack even
basic information about their presence, distribution and diversity in Singapore. While
several studies on macroinvertebrates in Singapore’s urbanised water bodies exist
(Loke et al., 2010; Blakely et al., 2014; Clews et al., 2014), the macroinvertebrates
in forested areas are less well-known, with scattered reports on selected insect groups
(e.g. Coleoptera, Hemiptera and Odonata) in the entire Central Catchment Nature
Reserve (Bailee et al., 1997; Murphy, 1997; Yang et al., 1997), as well as a single report
on selected insect groups (Hemiptera, Coleoptera, and Odonata) of the Nee Soon
freshwater swamp forest (Gani, 2013). In recent years, books focusing on specific
insect groups, such as those by Tang et al. (2010) and Tran et al. (2015), also provide
a general overview of their target groups in Singapore, including information on the
populations inside the forests of Singapore.
The lack of information on the aquatic macroinvertebrates of Singapore,
particularly baseline ecological information, is meant to be partially addressed by the
current study, which aimed to undertake representative sampling of macroinvertebrate
communities across the Nee Soon catchment. This enabled the identification of different
groups of macroinvertebrates found in the streams, their abundance and distribution
in the Nee Soon drainage. This is the first dedicated and relatively comprehensive
quantitative survey of the presence of aquatic macroinvertebrates in the Nee Soon
drainage, allowing for the establishment of baseline data regarding the aquatic
macroinvertebrates found in the freshwater swamp forest. This data will allow for
more informed management decisions to be made regarding the Nee Soon freshwater
swamp forest, especially with regards to its conservation.
Methods
Study sites
The Nee Soon freshwater swamp forest is located in the Central Catchment Nature
Reserve, roughly bounded by the Upper Seletar Reservoir, the Upper Peirce Reservoir
and the Seletar Expressway. It is the last remaining freshwater swamp forest found
in Singapore, and is accorded “protected” status as a nature reserve on account of
its inclusion in the Central Catchment Nature Reserve, as well as its use for military
training (Ng & Lim, 1992) with the presence of a military firing range within the Nee
Soon catchment (Fig. 1). The Nee Soon freshwater swamp forest is drained by a
single stream network, originating in central Singapore and draining roughly northeast
Aquatic macroinvertebrates of Nee Soon
73
towards Lower Seletar Reservoir. The streams are typically shallow (8-70 cm in depth)
and slow-flowing (mean velocity 1-18 cm/s). Tannins and other chemicals leaching
from leaf litter stain the water a dark tea-colour, as well as reducing the pH to acidic
levels (pH 4-6). Only a small amount of dissolved minerals can be found in the water.
In periods of heavy rainfall, the streams tend to overflow and flood the surrounding
area, creating small pools and flooded areas which may persist for indeterminate
periods (Ng & Lim, 1992; Lim et al., 2011). In addition to ground and rainwater input,
the Nee Soon drainage also receives periodic input from the Upper Seletar Reservoir
near the northeast edge of the lower catchment. The environment in this area of the
Nee Soon drainage is quite different from the rest of the swamp forest. Here, pH levels
tend to be higher (> 5), reflective of the less acidic waters received as well as the more
open forest canopy.
Additional information on the hydrology and geomorphology of the Nee Soon
catchment relevant to the aquatic macroinvertebrates is given by Nguyen et al. (2018)
and Sun et al. (2018).
Sample collection
The diversity of aquatic freshwater macroinvertebrates in the Nee Soon freshwater
swamp forest was represented by sampling a total of 40 sites throughout the freshwater
swamp forest each on a single occasion between October 2013 and September 2014
(Fig. 1). Additionally, sites 18, 33 and 38 (Fig. 1) were surveyed every two weeks
between December 2013 and January 2015. Macroinvertebrates were sampled
with a qualitative kick sampling method described in Blakely et al. (2010). The
sampling procedure involves kicking the stream substrata to disturb and release
macroinvertebrates into a kick-net (36 x 30 cm, 250 pm mesh size) held downstream.
Kick samples were collected from a wide range of microhabitats (e.g. leaf packs,
cobbles, pools, log jams and stream margin) over a 2-minute period within a 10 m
reach delimiting each site. This procedure facilitated the collection of comparable
representation of macroinvertebrate communities, while maximising the likelihood
of collecting all species present including rare and habitat-specific species. All
macroinvertebrate samples obtained were then immediately preserved in the field with
molecular grade isopropanol.
In the laboratory, samples were rinsed with water on a 250 pm mesh Endecott
sieve, and all aquatic macroinvertebrates were removed and identified under xl00
magnification. Individuals were identified to family-level, except for class Ostracoda,
subclass Acari, Collembola and Oligochaeta, order Araneae, and infraorder Brachyura.
The main references used for taxonomic identification were Merritt & Cummins
(1996), Yule & Yong (2004), and Blakely et al. (2010).
In this paper, the distribution of aquatic biodiversity in the Nee Soon drainage
is represented by a series of maps that show the number of individuals (abundance)
of each taxonomic group from each site during the spatially extensive survey (one
sample per site). Taxonomic abundance is represented at the order level for all groups,
except the Ostracoda that are identified to class-level and the Acari, Collembola and
Oligochaeta that are identified to subclass-level. Biological notes for each taxonomic
74
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
group include morphological characteristics that describe each taxonomic level (up to
order) and feeding habits. Biological notes are made with reference to Yule & Yong
(2004), Tang et al. (2010), Tan et al. (2010), and Tran et al. (2015).
For each taxon, the distribution of the group is supplemented with family-level
information. A checklist indicating the presence of families (or lowest practicable
level of identification) at each site is also presented. This checklist is derived from
all samples collected during the period between October 2013 and January 2015
(encompassing both the spatially extensive survey and higher frequency surveys of
three sites). Organisms from this study of the aquatic fauna were provided to a parallel
team working on genomics and imaging, for a depository of photographic images and
of barcodes (Kutty et al., 2018).
Results
In total, 82 taxonomic groups (76 families and six higher taxa) of aquatic
macroinvertebrates were found during the study throughout the Nee Soon freshwater
swamp forest, with the majority of these families from the phylum Arthropoda (Table
1). The most diverse class of organisms recorded within Arthropoda was Insecta (66
families in total). Other arthropod classes recorded were Malacostraca (two families
and one infraorder), Arachnida (two families) and Entognatha (one family). The
phylum Mollusca was represented by two classes, namely Gastropoda (four families)
and Bivalvia (two families). Finally, the phylum Annelida was represented by two
classes, which were Clitellata (two families) and Oligochaeta (one family).
The macroinvertebrate classes in the freshwater swamp forest appear to
have differing distributions, with both classes of Mollusca only being found in
the northeastern section of the Nee Soon drainage, as well as the class Ostracoda
(Table 2). The class Clitellata is found only at one site, with leeches from the family
Erpobdellidae (order Arhyncobdellida) being recorded at site 11 and leeches from
the family Glossiphoniidae (order Rhynchobdellida) being recorded at site 32. Other
aquatic macroinvertebrate groups are more evenly distributed throughout the Nee
Soon freshwater swamp forest, with Insecta, Malacostraca, and Arachnida being the
most common. Entognatha and Oligochaeta are somewhat less common, but are also
found throughout the Nee Soon drainage.
Class INSECTA
The insects are the most diverse macroinvertebrate group found throughout the Nee
Soon drainage, with a total of nine orders and 66 families recorded. Most insects
found in the streams of the Nee Soon drainage are in their larval stage, with the adult
stage being terrestrial rather than aquatic. Many of the classes recorded in the Nee
Soon drainage such as the stoneflies (Plecoptera) and the mayflies (Ephemeroptera)
Aquatic macroinvertebrates of Nee Soon
75
Fig. 1 . The main drainage of the Nee Soon freshwater swamp forest, with 40 sites sampled
from October 2013 to September 2014 for aquatic macroinvertebrates.
are indicators of good water quality (Blakely et al., 2014). Previous studies of the
insect fauna of the Central Catchment Nature Reserve (which includes the Nee Soon
drainage) covered several different insect orders (Bailee et al., 1997; Murphy, 1997;
Yang et al., 1997), which makes the insects among the better known groups of aquatic
macroinvertebrates in the Nee Soon freshwater swamp forest.
Order Diptera
The order Diptera, or the true flies, is an enormously successful and diverse group
of insects. There are currently 17 families of Diptera recorded from the Nee Soon
drainage. Overall, the Diptera are common throughout the entire Nee Soon drainage,
being found in nearly all 40 sites (Table 2). The family Chironomidae, which occur
naturally throughout the Nee Soon drainage, are the most common family. Other
76
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 1 . Summary of aquatic macroinvertebrates recorded across 40 sites in the Nee Soon
Swamp Forest stream network. Macro invertebrates identified are presented in respective
taxonomic structure, with the number of invertebrate taxa (identified at Family level or higher)
for each order denoted in No. of taxa. Total number of sites (No. of sites recorded) and the
range of individuals (Individuals recorded/site) where each order was surveyed were also
presented, along with the site number where the highest number of individual was recorded
(Site with highest abundance). *In this study, Ostracoda are identified at class-level; Acari,
Collembola and Oligochaeta are identified at subclass-level; and Araneae are identified at order-
level. #Within the Decapoda, three taxa were recognised, two families of shrimp (Atyidae and
Palaemonidae) and one infraorder for crabs (Brachyura).
Phylum
Class
Order
No. of taxa
(Family level
and above)
No. of
sites
recorded
Individuals
recorded/ site
Site with
highest
abundance
Arthropoda
Insecta
Coleoptera
12
26
1 - 15
24
Diptera
17
40
8- 1316
32
Ephemeroptera
5
38
1 - 87
32
Hemiptera
6
24
1 - 123
32
Lepidoptera
1
5
1-4
31
Megaloptera
2
6
1 - 2
9, 33
Odonata
10
32
1 - 119
32
Plecoptera
2
8
1 - 3
15
Trichoptera
11
31
1 - 34
32
Entognatha
Collembola*
1
10
1-4
24
Arachnida
Acari*
1
4
1-4
31
Araneae*
1
22
1 - 6
22
Malacostraca
Decapoda #
3
34
1 - 73
32
Ostracoda*
-
1
2
8-225
32
Mollusca
Gastropoda
Mesogastropoda
2
3
1 - 87
32
B asommatophora
2
2
8-29
32
Bivalvia
Veneroida
2
1
127 - 127
32
Annelida
Clitellata
Arhyncobdellida
1
1
1
11
Rhynchobdellida
1
1
1
32
Oligochaeta*
1
24
1 - 50
29
common dipteran families included Ceratopogonidae and Simuliidae, which were
recorded at sites 33 and 15, respectively (Table 2). Diptera were found in relatively
high numbers throughout the entire drainage (Table 1), but the abundance is not as
high as other sites in Singapore (e.g. Clews et al., 2014). Diptera were most abundant
Aquatic macroinvertebrates of Nee Soon
77
N
Upper Pejrce Map data cj 2017 Google
Sefetar Res*'
springiear ivaiure raw tj
28
Diptera
• 20D
• 400
0 wo
0 TOO
0 1000
0 1200
Fig. 2. Distribution and abundance of order Diptera in the Nee Soon drainage. Solid circles
indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
in the northeastern part of the Nee Soon drainage, with more than 1,300 specimens
recorded in a single site (site 32) (Fig. 2).
Order Coleoptera
The order Coleoptera, also known as the beetles, is probably the most speciose order
of eukaryotic organisms in the world. Almost all beetles possess elytra or wing-cases,
a hardened first set of wings designed to protect the more delicate second set of wings
from damage (Romoser & Stoffolano, 1994). In total, 12 families of beetles were
found in the Nee Soon freshwater swamp forest throughout this study. In many aquatic
species of beetle, both the larval and the adult stages are aquatic, unlike many other
insect orders which only have aquatic larval stages (Yule & Yong, 2004). In the Nee
Table 2. Macroinvertebrate checklist across 40 sites in the Nee Soon freshwater swamp forest. Taxa present at each site are denoted as V. * In this
study, Ostracoda are identified at class-level; Acari, Collembola and Oligochaeta are identified at subclass-level; Araneae are identified at order-
level; and Brachyura are identified at infraorder-level. t denote families that were recorded during biweekly sampling between January 2014 and
January 2015 at sites 11, 33 and 38.
78
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
OP
i
i
i
i
i
i
i
i
i
i
i
l
l
i
6£
i
i
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
l
8£
i
i
i
i
i
i
i
i
i
i
l
l
l
L£
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
p-
9£
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
S£
i
i
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
l
P£
i
i
i
i
i
i
i
i
i
i
i
p-
i
p-
i
l
l
££
i
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
i
Z£
i
i
i
y
i
i
p-
i
i
i
p-
p-
i
p-
l
l£
i
i
i
i
i
i
i
i
i
i
i
i
i
l
p-
i
l
l
0£
i
i
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
l
6Z
i
i
i
i
i
i
i
i
i
i
i
i
i
l
i
i
8Z
i
i
i
i
i
i
i
i
i
i
i
i
i
l
l
l
i
LZ
i
i
i
i
--
i
i
i
i
i
i
i
i
l
p-
i
l
l
9Z
i
i
i
i
i
i
i
i
i
i
i
i
i
i
l
l
sz
i
i
i
i
i
i
i
i
i
i
i
i
i
p-
i
i
l
PZ
i
i
i
i
i
i
i
i
i
p-
i
p-
i
l
l
p-
£Z
i
i
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
i
zz
i
i
i
i
i
i
i
i
i
i
i
i
l
i
l
IZ
i
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
oz
i
i
i
i
i
i
i
i
i
i
i
i
l
l
i
61
i
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
l
81
i
i
i
i
i
i
i
i
i
i
i
i
p-
p-
i
l
l
LI
i
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
91
i
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
i
SI
i
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
p-
PI
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
ex
i
i
i
i
i
i
i
i
i
i
i
i
i
i
l
Zl
i
i
i
i
i
i
i
i
i
i
i
i
p-
i
l
p-
XX
i
i
i
i
i
i
i
i
i
i
i
i
l
l
l
ox
i
i
i
i
i
i
i
i
i
i
p-
i
p-
l
i
l
6
i
i
i
i
i
i
i
i
i
i
i
i
i
l
l
8
i
i
i
i
i
i
i
i
i
i
i
i
>
-p*
i
l
l
L
i
i
i
i
i
i
i
i
i
i
i
p-
p-
l
i
i
9
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
l
S
i
i
i
i
i
i
i
i
i
i
i
i
p-
p-
l
l
l
P
i
i
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
p-
£
i
i
i
y
i
i
i
i
i
i
i
i
p-
l
l
l
Z
i
i
i
i
i
i
i
i
i
i
i
i
i
l
i
l
l
X
i
i
i
i
i
i
i
i
i
i
i
i
i
p-
l
l
i
Family
CanacidaeJ
Chrysomelidae
Curculionidaet
Dystiscidae
Elmidae
Gyrinidae
Hydraenidaet
Hydrophilidae
Hydroscaphidae
Psephenidae
Ptilodactylidae
Scirtidae
Athericidae
Ceratopogonidae
Chironomidae
Corethrellidae
Culicidae
Dolichopodidae
Order
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Coleoptera
Diptera
Diptera
Diptera
Diptera
Diptera
Diptera
Table 2. Continuation.
Aquatic macroinvertebrates of Nee Soon
79
OP
i
i
i
l
i
i
i
i
i
l
p-
i
i
i
l
i
i
6£
i
i
i
i
i
i
i
l
i
l
i
i
i
i
i
l
l
i
i
8£
i
l
i
l
l
i
i
i
V'
i
l
p-
p-
i
i
i
i
i
i
Lt
i
l
i
l
l
i
i
p-
1
i
l
-p*
i
i
i
i
l
i
p
9£
i
l
i
i
i
i
i
i
1
i
l
p-
i
i
p
i
i
~p
i
i
i
$£
i
i
i
l
i
i
i
l
1
i
l
i
i
i
i
i
l
l
i
i
pz
i
i
i
l
l
i
i
i
1
i
i
l
i
i
i
i
l
i
i
i
££
i
p-
i
i
i
l
i
l
i
i
p
~p
l
i
i
ZZ
i
l
i
l
l
i
i
l
-p*
i
i
i
i
i
i
i
i
p
X£
i
l
i
i
l
i
i
l
i
l
p-
~p~
i
i
i
l
i
i
i
0£
i
l
i
l
i
i
i
i
l
i
l
p-
i
i
i
i
l
l
i
i
6z:
i
i
i
l
l
i
i
i
l
i
l
i
i
i
i
i
~p
l
i
i
sz:
i
l
i
l
i
i
i
l
1
i
i
l
i
i
i
i
i
l
l
i
i
LZ
i
l
i
l
i
i
i
l
1
i
l
p-
i
i
i
i
l
l
i
i
9Z
i
l
i
l
l
i
i
l
1
i
l
i
i
i
i
i
i
i
i
sz
i
l
i
l
l
i
i
i
1
i
p-
i
i
i
i
i
l
i
i
i
pz
i
i
i
l
i
i
i
l
1
i
i
i
i
i
i
i
l
i
i
i
zz
i
i
i
l
i
i
i
p-
1
i
l
i
i
i
i
>
l
i
i
zz
i
l
i
l
l
i
i
i
i
l
i
i
i
--
l
l
i
i
IZ
i
l
i
l
l
i
i
l
l
i
l
p-
p
i
i
i
i
l
i
i
oz
i
l
i
l
i
i
i
1
i
i
p
i
i
--
~p
i
i
i
61
i
l
i
l
l
i
i
l
V'
i
p-
~p~
~p
i
i
i
l
l
i
i
81
i
p-
i
i
l
i
i
i
1
i
i
p-
i
i
i
i
p
l
i
i
LI
l
l
l
i
i
1
i
p-
p
i
i
i
l
l
i
i
91
i
l
i
l
l
i
i
1
i
l
i
i
i
l
i
i
SI
i
l
i
l
i
i
i
1
i
p-
p-
i
i
~p
p
i
i
i
n
i
l
i
l
i
i
i
l
1
i
l
i
i
i
l
l
i
i
£1
i
l
i
l
l
i
i
1
i
i
i
i
i
i
i
l
l
i
i
Zl
i
i
i
i
i
i
i
1
i
l
p
i
i
i
V'
l
i
i
XX
i
l
i
p-
i
i
p-
l
1
i
i
i
i
i
i
i
i
p
l
i
i
ox
i
l
i
l
i
i
i
1
i
l
V'
i
i
i
i
l
i
i
6
i
l
i
l
l
i
i
l
1
i
l
p-
i
i
i
i
~p
l
i
i
8
i
l
i
i
i
i
i
l
1
i
l
i
i
i
i
i
1
l
i
i
L
i
l
i
i
i
i
i
1
i
l
~p~
i
i
i
i
1
l
i
i
9
i
l
i
i
l
i
i
1
i
-p*
p
i
i
i
1
i
i
i
S
i
l
i
i
l
i
i
l
1
i
i
i
i
i
i
l
i
i
i
P
i
i
i
l
l
i
i
l
1
i
i
p-
~p~
i
i
i
i
1
l
i
i
£
i
i
i
i
l
i
i
1
i
p-
p-
i
i
1
i
i
i
£
i
i
i
l
l
i
i
l
1
i
v-
p-
i
i
i
i
p
i
X
i
l
i
i
l
i
i
i
1
i
l
p-
p-
i
p
i
i
l
i
i
Family
Dryopidae
Empididaet
Nymphomyiidae
Phoridae
Psychodidaef
Ptychopteridaef
Sciomyzidae
Simulidae
Tabanidae
Tanyderidae
Tipulidae
Baetidae
Caenidae
Heptageniidae
Leptophlebiidae
Siphlonuridaet
Corixidae
Gerridae
Haliplidaet
Hebridae
Mesoveliidae
Order
Diptera
Diptera
Diptera
Diptera
Diptera
Diptera
Diptera
Diptera
Diptera
Diptera
Diptera
Ephemeroptera
Ephemeroptera
Ephemeroptera
Ephemeroptera
Ephemeroptera
Hemiptera
Hemiptera
Hemiptera
Hemiptera
Hemiptera
Table 2. Continuation.
80
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
OP
l
i
i
i
i
i
i
l
i
i
i
~~7'
i
l
i
l
6£
i
i
i
i
l
l
l
i
i
l
i
i
l
i
1
i
l
i
i
l
8£
i
l
i
l
i
i
i
i
i
i
i
i
l
l
i
1
i
l
i
i
l
L£
i
l
i
l
i
l
l
i
i
i
i
i
l
i
l
1
i
i
i
i
9£
i
l
i
l
i
i
l
i
i
l
i
l
l
l
1
i
l
i
i
S£
i
l
i
l
l
l
l
i
i
y
l
i
l
l
i
1
i
l
l
i
l
P£
i
l
i
l
i
l
l
i
i
i
i
i
l
i
i
p-
i
l
l
i
i
££
i
>
i
i
l
l
i
i
i
i
l
i
1
i
i
p-
p-
Z£
i
>
i
l
i
i
~~r
i
y
i
i
1
i
p-
p-
l£
i
i
l
l
i
i
l
i
l
i
1
~pr
l
p-
0£
i
i
i
i
l
i
i
i
i
i
l
i
l
1
i
l
i
l
6Z
i
l
i
l
i
l
i
i
i
i
p-
i
l
l
l
1
i
i
l
i
i
8Z
i
i
i
l
i
i
l
i
i
i
i
i
l
i
l
1
i
i
l
i
l
LZ
i
l
i
i
i
l
l
i
l
i
i
l
l
1
i
l
i
i
i
9Z
i
l
i
l
l
l
l
i
i
i
l
i
l
1
~p-
l
l
i
l
sz
i
l
i
l
l
l
i
i
i
i
l
l
i
1
i
l
i
l
PZ
i
l
i
l
i
i
l
i
l
i
l
i
l
P-
i
l
l
i
i
£Z
i
l
i
i
l
l
l
i
i
i
i
l
l
l
1
i
l
i
i
i
zz
i
i
l
l
i
i
l
i
l
l
i
1
p-
l
i
i
i
IZ
i
l
i
i
y
y
i
i
i
l
l
1
i
i
i
i
oz
i
l
i
i
i
i
i
l
i
l
l
l
1
i
-p*
i
61
y
i
i
l
i
l
i
i
i
i
l
1
~~p~
l
p-
i
81
i
i
i
>
i
i
l
i
i
i
i
l
l
i
1
i
l
i
l
LI
i
i
l
l
l
l
i
i
i
i
l
i
i
1
i
i
l
i
p-
91
i
l
i
l
l
l
i
>
l
i
l
l
l
1
i
l
i
SI
i
l
i
l
l
i
i
i
i
i
i
i
p-
i
l
p-
i
i
PI
i
l
i
l
l
l
y
i
i
i
i
l
1
i
i
l
ex
i
i
l
i
i
i
i
i
l
i
i
l
l
1
i
i
l
i
i
Zl
i
l
i
l
l
i
>
i
i
l
i
1
l
l
i
i
XX
i
l
i
i
i
i
i
i
i
i
i
i
i
l
1
i
l
l
i
i
OX
i
l
i
l
i
l
l
i
i
l
i
i
l
l
1
i
i
i
6
l
i
i
l
i
i
i
i
l
i
i
1
i
l
l
8
i
l
i
l
i
l
l
i
i
i
i
i
l
i
i
1
i
l
l
i
l
L
i
i
i
i
l
l
l
i
i
l
i
i
>
i
1
l
p-
i
9
i
l
i
i
l
i
i
i
l
i
i
l
1
p-
l
i
S
l
i
l
i
l
l
i
i
l
i
i
l
"P'
i
i
i
l
P
i
l
i
l
l
l
i
i
i
i
l
i
l
i
i
l
l
p-
i
£
i
l
i
l
l
l
i
i
l
i
l
i
1
p-
i
p-
i
p-
Z
i
l
i
l
l
i
l
i
l
p-
l
p-
l
1
i
l
p-
i
X
i
l
i
l
i
l
l
i
i
l
i
l
i
l
p-
i
l
i
p-
Family
Veliidae
Crambidae
Corydalidae
Sialidae
Aeshnidae
Calopterygidae
Coenagrionidae
Corduliidae
Euphaeidae
Gomphidae
Libellulidae
Platycnemididae
Platystictidae
Protoneuridae
Leuctridae
Perlidae
Calamoceratidae
Dipseudopsidae
Ecnomidae
Hydropsychidae
Hydroptilidae
Order
Hemiptera
Lepidoptera
Megaloptera
Megaloptera
Odonata
Odonata
Odonata
Odonata
Odonata
Odonata
Odonata
Odonata
Odonata
Odonata
Plecoptera
Plecoptera
Trichoptera
Trichoptera
Trichoptera
Trichoptera
Trichoptera
Table 2. Continuation.
Aquatic macroinvertebrates of Nee Soon
81
OP
i
i
l
l
i
i
l
i
~~p~
>
i
i
l
?-
i
i
l
i
i
6£
i
i
i
i
i
i
p-
i
i
V-
i
i
i
l
l
i
i
l
i
i
p-
8£
i
l
l
l
l
i
l
i
p-
i
i
>
i
i
l
l
i
i
i
l
l
l
L£
i
l
l
l
i
i
i
i
~p-
i
i
l
i
i
i
i
l
i
i
9£
l
l
l
i
i
i
p-
i
p-
i
i
i
i
i
i
i
l
i
i
S£
i
l
p-
l
i
l
i
i
p-
i
i
l
i
i
i
l
l
i
i
P£
i
i
l
l
l
i
l
i
i
i
i
i
l
i
i
i
i
i
i
i
££
i
i
l
i
l
i
l
i
-p*
p-
i
i
l
l
i
i
l
l
l
Z£
y
l
l
p-
i
i
l
i
i
i
i
p-
i
~p-
i
l£
i
i
i
l
i
i
i
i
i
l
i
l
l
l
i
0£
i
l
l
l
i
i
l
i
i
i
i
i
i
i
l
i
i
1
l
i
V'
6Z
i
i
l
p-
l
i
i
l
i
i
i
i
i
l
l
i
i
1
l
i
p-
HZ
i
l
l
l
l
i
i
i
l
i
i
i
i
i
i
i
i
i
l
i
i
1
LZ
i
l
i
l
l
i
i
i
p-
i
i
p-
i
i
l
i
i
i
l
l
l
l
9Z
i
l
i
l
l
i
i
>
i
i
i
l
l
i
i
l
l
l
sz
i
l
l
l
l
i
l
i
>
i
i
p-
i
i
i
i
i
i
l
i
l
p-
pz
i
l
l
p-
l
i
i
i
i
i
i
i
i
l
l
i
i
1
i
l
£Z
i
l
i
i
l
i
i
i
i
i
i
p-
i
i
l
i
i
i
1
i
l
zz
i
l
l
l
i
l
i
i
i
i
i
i
i
i
l
l
l
>
IZ
y
l
i
l
l
i
l
i
l
i
i
i
l
l
i
i
1
l
i
p-
oz
l
l
i
i
i
i
l
p-
i
i
i
l
i
i
l
l
i
61
y
i
l
l
l
i
i
i
i
i
i
i
i
i
1
l
i
V'
81
i
l
p-
p-
i
i
l
i
>
i
i
l
l
i
i
l
l
i
V'
LI
i
i
i
p-
l
p-
l
i
l
i
i
p-
i
i
l
l
i
i
l
l
i
1
91
l
l
l
l
i
l
i
-p*
i
p-
~p-
i
i
l
l
i
i
1
l
l
1
SI
l
l
i
i
i
l
v-
i
i
l
i
i
i
l
i
l
1
PI
>
l
l
i
i
l
i
l
i
V'
i
i
l
i
i
i
l
l
l
£1
i
i
i
l
l
i
l
i
V'
i
i
~p-
i
i
l
l
i
i
l
l
l
Zl
i
V'
i
l
i
i
i
l
p-
p-
i
i
i
i
i
i
1
i
l
p-
XX
i
l
1
l
i
i
i
1
i
i
i
l
i
i
i
l
~p-
i
p-
OX
l
l
l
l
i
l
i
i
i
p-
i
i
l
i
i
i
l
l
i
1
6
i
p-
p-
l
l
i
l
i
l
i
l
p-
i
i
l
i
i
i
l
l
l
1
8
i
l
1
l
i
i
l
i
i
l
i
i
i
i
l
i
i
1
i
i
1
L
y
l
1
l
i
i
1
i
i
l
i
i
i
1
i
i
1
9
~-r
1
>
l
i
i
i
1
i
i
i
i
i
i
i
l
l
i
1
S
1
p-
l
i
l
i
i
i
i
i
i
i
i
l
l
l
p-
P
i
l
l
l
i
l
i
i
i
i
i
i
i
i
l
i
i
£
i
l
l
i
p-
i
i
i
i
l
i
i
i
1
l
l
Z
i
1
p-
i
i
p-
i
i
i
l
l
i
i
1
l
i
X
i
1
l
i
i
i
p-
l
p-
i
i
i
i
i
i
1
l
l
Family
Leptoceridae
Odontoceridae
Philopotamidae
Polycentropodidae
Psychomyiidaet
Xiphocentronidae
i
i
l
Atyidae
Brachyura*
Palaemonidae
i
Physidae
Planorbidae
Assimineidae
Thiaridae
Corbiculidae
Sphaeriidae
Erpobdellidae
Glossiphoniidae
1
Order
Trichoptera
Trichoptera
Trichoptera
Trichoptera
Trichoptera
Trichoptera
Collembola*
Acari*
Araneae*
Decapoda
Decapoda
Decapoda
Ostracoda*
B as ommatophora
B asommatophora
Mesogastropoda
Mesogastropoda
Veneroida
Veneroida
Arhyncobdellida
Rhynchobdellida
Oligochaeta*
82
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
oprmf/jear jvaiure rarx u
28
N
Upper Pejrce Map data 2017 Google
Coleoptera
» o
• 6
• •
Fig. 3. Distribution and abundance of order Coleoptera in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Soon freshwater swamp forest, the Coleoptera are not very evenly distributed, being
found in several clusters throughout the swamp forest, including in the more upstream
portions as well as the outskirts of the Nee Soon drainage (Fig. 3). They are relatively
uncommon, being found in 26 sites. Of the 12 families recorded, the most common is
the family Elmidae, which were recorded from 16 sites across the Nee Soon freshwater
swamp forest (Table 2). The other 11 families were largely rare, occurring at one to
four sites across the Nee Soon freshwater swamp forest. In general, their numbers are
relatively low compared to other groups (Table 1), with the maximum abundance in a
single site (site 24) being 14 specimens.
Aquatic macroinvertebrates of Nee Soon
83
spring tear wawre yam u
28
Okm 0.2 km 0.4km
Upper Pejrce Map data 2017 Google
Ephemeroptera
* o
• *>
• *o
£ 60
Fig. 4. Distribution and abundance of order Ephemeroptera in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Order Ephemeroptera
Insects of the order Ephemeroptera are also known as the mayflies. The presence of
this group of insects is generally an indication of good water quality (Blakely et al.,
2014), as they are sensitive to pollution. Mayflies spend most of their life in the larval
stage. When they reach adulthood, they usually live for a very short period of time
(Romoser & Stoffolano, 1994). There are 5 families from this order found in the Nee
Soon drainage. Mayflies are distributed throughout the Nee Soon drainage in large
numbers, but are most abundant in the northeast of the Nee Soon drainage. They were
found in 38 out of 40 sites (not found at sites 11 and 28), with the most abundant site
having around 80 or so individuals (Fig. 4). Common Ephemeroptera families included
Baetidae and Caenidae, which were recorded at 34 and 29 sites across the Nee Soon
84
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
spring tear waiure raw u
28
Fig. 5. Distribution and abundance of order Hemiptera in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
N
Upper PeUce ^ Map data 2017 Google
Hemiptera
• c
• 20
• a (,
• w
• »
0 100
41 120
drainage, respectively (Table 2). Heptageniidae, Leptophlebiidae and Siphlonuridae
were, however, less common, and were recorded at 12 and five sites across the Nee
Soon drainage, respectively.
Order Hemiptera
Six families of Hemiptera were found in the Nee Soon freshwater swamp forest.
Commonly known as the “true bugs”, hemipterans all possess a rostrum, a hollow
feeding tube which is used to pierce and suck food, whether plant or animal in origin
(Tran et al., 2015). Like the beetles, both the adults and the juveniles of this group
are aquatic, with some living on the surface of the water and others swimming in the
water column (Tran et al., 2015). Hemiptera are relatively uncommon in the Nee Soon
Aquatic macroinvertebrates of Nee Soon
85
Fig. 6. Distribution and abundance of order Lepidoptera in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
drainage (recorded in 24 sites), and are typically found in low numbers (one to five
specimens recorded for each family per site). Large numbers of Corixidae (122) were
recorded at site 32 (Fig. 5). Out of the six families of Hemiptera recorded, Gerridae
was the most common (recorded at 16 sites), while Corixidae (seven sites), Haliplidae
(one site), Hebridae (one site), Mesoveliidae (two sites) and Veliidae (six sites) were
uncommon (Table 2).
Order Lepidoptera
The order Lepidoptera includes the butterflies and the moths. While these insects are an
extremely diverse and abundant group, only one family (Crambidae) has been recorded
within the streams of the Nee Soon freshwater swamp forest, as most lepidopterans
86
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 7. Distribution and abundance of order Megaloptera in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
have a terrestrial lifestyle (Romoser & Stoffolano, 1994). Lepidopterans were mostly
found in the northeast of the Nee Soon drainage, occasionally being found deeper
within the swamp forest. They are relatively uncommon (recorded at six sites) and
were found only in small numbers, with the most abundant site (site 31) having only
four individuals (Fig. 6).
Order Megaloptera
Two families of Megaloptera were recorded from the Nee Soon freshwater swamp
forest. The Megaloptera are also known as the dobsonflies, and are famous for the
enormous mandibles which adult males from some genera possess (Cover & Resh,
2008). However, in most species, the adult form does not feed at all, and it is theorised
Aquatic macroinvertebrates of Nee Soon
87
that the mandibles are a secondary sexual characteristic, meant to attract mates or drive
off rivals (Cover & Resh, 2008). In contrast, the larvae are voracious predators, and
feed on almost anything they can catch (Romoser & Stoffolano, 1994; Cover & Resh,
2008). Megalopterans are rare in the Nee Soon freshwater swamp forest (Fig. 7). Only
one Corydalidae specimen was collected from site 22 and low numbers of Sialidae (1
to 2 individuals) were recorded at five sites (sites 6, 9, 18, 19 and 33) (Table 2).
Order Odonata
The order Odonata, better known as the damselflies (Zygoptera) and dragonflies
(Anisoptera), are among the best known groups of insects, due to the visibility and
bright colours of the adults. Most of the information currently available on the Odonata
of Singapore focuses on their adult stage (Murphy, 1997; Norma-Rashid et al., 2008;
Tang et al., 2010; Ngiam & Cheong, 2016). However, knowledge of the aquatic larval
stages is growing, with a recent study by Yeo et al. (in press) enabling the matching of
adult and larval stages of 59 odonate species found in Singapore. Currently, 10 families
of Odonata were documented from the Nee Soon freshwater swamp forest. The most
commonly recorded dragonfly families were Gomphidae (recorded at 26 sites) and
Corduliidae (recorded at 13 sites), while the most common damselfly family was
Protoneuridae (recorded at 12 sites) (Table 2). The Odonata appear to be distributed
throughout the entire Nee Soon drainage, as they were recorded from 32 out of 40
sites. However, they are most abundant in the northeast of the Nee Soon drainage (Fig.
8). For instance, at Site 32, 119 specimens were documented. These specimens were
largely made up of damselfly larvae (94 Protoneuridae individuals, 1 Platystictidae
individual and 1 Coenagrionidae individual), along with some dragonfly larvae (12
Corduliidae individuals, 6 Libellulidae individuals and 5 Gomphidae individuals).
Further information on the Odonata recorded in the Nee Soon freshwater swamp
forest is given by Cai et al. (2018).
Order Plecoptera
Two families of Plecoptera have been recorded from the Nee Soon freshwater swamp
forest. This group is commonly known as the stoneflies, and is commonly used as
an indicator group, since the aquatic nymphs are generally intolerant of pollution
(Blakely et al., 2014). Some species are herbivorous, while others are predatory
(Romoser & Stoffolano, 1994). In the Nee Soon drainage, they are relatively rare,
with one Leutridae individual recorded at sites 4 and 11, and low numbers of Perlidae
(one to three individuals) recorded at seven sites (sites 1, 4, 5, 15, 24, 34, 40) (Table
2). Unlike other aquatic invertebrates, Plecoptera were absent in the northeast of the
Nee Soon drainage (Fig. 9).
Order Trichoptera
The order Trichoptera is commonly known as the caddisflies. These insects are well-
known for the cases which their aquatic larval stages construct, using detritus and silk.
This group is also commonly used as a bioindicator of aquatic pollution (Yule & Yong,
2004; Blakely et al., 2014). Eleven families of Trichoptera were found in the Nee Soon
88
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 8. Distribution and abundance of order Odonata in the Nee Soon drainage. Solid circles
indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
drainage. Ecnomidae was the most common Trichoptera family recorded (found at
22 sites), while five other families - Calamoceratidae, Hydroptilidae, Leptoceridae,
Odontoceridae and Polycentropodidae, were recorded at more than 10 sites across
the Nee Soon drainage (Table 2) In total, trichopterans were found in 31 out of the
40 sampling sites, with their greatest abundance (30 individuals) recorded at site 32
(Fig. 10). Trichoptera individuals recorded at site 32 consisted of individuals from
Hydroptilidae (fourteen individuals), Leptoceridae (eleven individuals), Ecnomidae
(six individuals), Hydropsychidae (one individual), Calamoceratidae (one individual)
and Polycentropodidae (one individual).
Aquatic macroinvertebrates of Nee Soon
89
Fig. 9. Distribution and abundance of order Plecoptera in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Class ARACHNIDA
The arachnids include not only familiar organisms such as the spiders and the
scorpions, but also several other groups such as mites and harvestmen. Arachnids are
easily identified, as they typically possess four pairs of legs as well as a body divided
into two parts, the cephalothorax and the abdomen. Two taxonomic groups have been
recorded from the Nee Soon drainage in the course of this study.
Subclass Acari
The subclass Acari was recorded from the Nee Soon freshwater swamp forest during
this study. This subclass is more commonly known as the mites, and is extraordinarily
90
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
/K
N
Bpnngiear rvawre name y
28
Upper Pejrce Map data 2 oi 7 Google
ier Selet
Trichoptera
• o
• 5
• 1 1 >
• <a
• *
Fig. 10. Distribution and abundance of order Trichoptera in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
diverse and widespread worldwide, having adapted to many different environments
and modes of life (Yule & Yong, 2004). However, in the Nee Soon catchment they
appear to be quite rare, being only found in a few scattered locations in low numbers
(Fig. 11). Members of this subclass were only found in four sites (sites 4, 7, 31 and
34), with the most abundant site (site 31) having a total of four individuals. Freshwater
mites in both adult and larval stages are usually predators or parasites, feeding off
other organisms such as insect larvae.
Aquatic macroinvertebrates of Nee Soon
91
Fig. 11. Distribution and abundance of subclass Acari in the Nee Soon drainage. Solid circles
indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Order Araneae
The Araneae are the arachnids most familiar to most people, as they comprise the
familiar spiders. The aquatic spiders present in the Nee Soon freshwater swamp
forest tend to remain on the water surface, attacking fish, aquatic insects and small
crustaceans that venture too close to them (Ng et al., 2011). Aquatic spiders in the Nee
Soon drainage are uncommon, and can be found distributed across 22 sites throughout
the entire drainage of the swamp (Fig. 12). The highest number of individuals found
in a single site (site 22) was six.
92
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Upper PejfCG ^ Map data ©2017 Google
spring fear nazure rant u
32
Araneae
Fig. 12. Distribution and abundance of order Araneae in the Nee Soon drainage. Solid circles
indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Class MALACOSTRACA
The malacostracans are a class of crustaceans which include the familiar crabs, shrimp
and lobsters, as well as perhaps less familiar amphipods and woodlice (isopods). They
are extremely widespread, being found in marine, freshwater and even terrestrial
enviro nm ents worldwide. They can be identified by their body plan with 20-21
segments, and possess a head, thorax, and abdomen. Many species of malacostracans
are commercially important.
Aquatic macroinvertebrates of Nee Soon
93
/K
N
(fesgjfe
Okm 0.2km 0.4km
Upper Peirce ^
Map data 2017 Google
spring tear nature narK v
28
Decapoda
■ o
• ZQ
• «D
0 60
Fig. 13. Distribution and abundance of order Decapoda in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Order Decapoda
The decapods are so named due to the ten walking legs all members of this group
possess. The crustaceans which most people are familiar with, such as crayfish, crabs,
and shrimp, belong to this group. In the Nee Soon drainage, three groups of decapods
- Atyidae, Palaemonidae (families of shrimp) and Brachyura (infraorder for crabs,
in two families, Gecarcinucidae and Sesarmidae) can be found. The decapods are
distributed throughout the entire drainage and are relatively common and abundant,
being found in greatest numbers in the northeast and southwest areas of the Nee Soon
94
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
drainage (Fig. 13). Of the three groups recorded, Palaemonidae was the most common
throughout the Nee Soon drainage (recorded at 32 sites), while the Brachyura (recorded
at 18 sites), as well as the Atyidae (recorded at 13 sites), were mostly recorded at the
southwest of the drainage (co-recorded at sites 1 to 13) (Table 2). In total, decapods
were found in 34 sites in the Nee Soon drainage, with the most abundant site (site 32)
having 73 Palaemonidae individuals.
Class OSTRACODA
The tiny Ostracoda, or seed shrimps, are small crustaceans which are found in
freshwater and marine environments worldwide. They can be either free-swimming or
benthos-dwellers. Ostracods possess a carapace made out of a hinged valve or shell,
somewhat resembling that of a bivalve, which protects them from predation. In some
species, the carapace is ornamented with spikes or protrusions, while others have a
smooth, featureless carapace (Yule & Yong, 2004). They have a wide dietary range,
with some being predators, while others are scavengers or filter feeders (Ng et al.,
2011 ).
In the Nee Soon drainage, ostracods were found at three sites in the northeast
areas in large numbers (the highest number of individuals recorded was 225 at site 32)
(Fig. 14). They were almost completely absent from the rest of the swamp forest.
Class ENTOGNATHA
The Entognatha were previously classified as insects, but have recently been separated
and placed into their own class. They are wingless, six-legged arthropods which possess
mouthparts recessed into the head, with only the tips of the mandibles and maxilla
exposed. The subclass Collembola are perhaps the best known of the Entognatha,
and are commonly called the springtails, due to their possession of a two-pronged
appendage known as the furcula (Yule & Yong, 2004). This organ is usually found
under the abdomen and is held under tension most of the time. When the springtail is
alarmed, it releases the furcular, which then impacts against the substrate and propels
the springtail into the air, away from the source of the alarm.
Subclass Collembola
In the Nee Soon freshwater swamp forest, Collembola are not very common, and were
found in several scattered locations throughout the entire drainage. They were only
found in 10 sites, with the most abundant site (site 24), only recording a total of four
individuals (Fig. 15).
Aquatic macroinvertebrates of Nee Soon
95
Fig. 14. Distribution and abundance of class Ostracoda in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Class GASTROPODA
The gastropods are a group of molluscs possessing a large, fleshy organ known as
a foot, which is used to move around. They are commonly referred to as snails or
slugs, depending on whether they possess a shell or not. The gastropods are found
in terrestrial, freshwater and marine environments worldwide. Many gastropods are
hermaphrodites. While most feed on plant matter or are scavengers, some gastropods
are predators instead (Ng et al., 2011). Terrestrial snails and slugs from the Nee Soon
freshwater swamp forest have been surveyed by Lim et al. (2018).
96
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 15. Distribution and abundance of subclass Collembola in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Order Basommatophora
The order Basommatophora is part of a larger group called the Pulmonata, also known
as the air breathing snails and slugs. Together with a modified mantle cavity, which
allows them to breathe atmospheric air, the members of this order have their eyes on
the base of their tentacles (Ruppert & Barnes, 1994). Two families from this order -
Physidae and Planorbidae, were found in the Nee Soon drainage in large numbers,
but only in the northeast of the Nee Soon drainage. Relatively larger numbers of
Planorbidae (8 and 27 individuals) were recorded at sites 31 and 32, while only two
Physidae individuals were recorded at site 32 (Fig. 16). Their absence from the rest of
the swamp forest could possibly be due to the lower pH levels within the rest of the
Nee Soon drainage, as snails are intolerant of acidic conditions (Zischke et al., 1983).
Aquatic macroinvertebrates of Nee Soon
97
Fig. 16. Distribution and abundance of order Basommatophora in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Order Mesogastropoda
The gastropods of the order Mesogastropoda possess gills which are directly attached
to the mantle wall, as well as an operculum, features which the Basommatophora
lack (Ruppert & Barnes, 1994). Representatives from two families from this order -
Assimineidae and Thiaridae were found in the Nee Soon drainage, but again only in
the northeast areas and nowhere else within the Nee Soon freshwater swamp forest.
Snails from the family Thiaridae were the most commonly recorded, with 19 and 87
individuals recorded at sites 31 and 32 respectively, while only one Assimineidae
individual was recorded at site 40 (Fig. 17).
98
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 17. Distribution and abundance of order Mesogastropoda in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Class BIVALVIA
The bivalves are another group of molluscs, named after the two hinged shells all
members of the class possess. Most bivalves are filter feeders (though some are
deposit feeders), extracting food particles from the water they are submerged in using
modified gills (Ruppert & Barnes, 1994). When startled, bivalves close their shells
tightly and can hold them closed with surprising force. Bivalves are most abundant in
the marine environment, but many families have adapted to a freshwater environment.
In fact, some freshwater bivalves are among the most invasive species in the world,
such as the Ponto-Caspian zebra mussel (Dreissena polymorpha), introduced into
North America (Lowe et al., 2004). Ironically, native bivalves in North America are
also among the most imperilled freshwater fauna (Ricciardi et al., 1998).
Aquatic macroinvertebrates of Nee Soon
99
Fig. 18. Distribution and abundance of order Veneroida in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Order Veneroida
The order Veneroida is generally made up of marine bivalves, but with some freshwater
representatives. In the Nee Soon drainage, two families (Corbiculidae and Sphaeriidae)
from the order can be found, and both were recorded at site 32 in relatively high numbers
(Fig. 18). At site 32, 125 Corbiculidae individuals and three Sphaeriidae individuals
were recorded. Like the gastropods, bivalves are intolerant of acidic conditions, which
could affect their survival, growth and shell integrity (Bressan et al., 2014). They are
completely absent from the rest of the Nee Soon drainage, possibly due to the effects
of low pH on their shells.
100
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 19. Distribution and abundance of order Arhynchobdellida in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Class CLITELLATA
The class Clitellata are annelids which possess a “glandular girdle”, also known as
the clitellum, giving them their name. This organ produces a substance which forms
a cocoon in which the juvenile worms develop (Ruppert & Barnes, 1994). They are
found in the terrestrial, freshwater and in marine ecosystems. Among the more familiar
members of this class are the earthworms and leeches.
Order Arhynchobdellida
The leeches found in this order do not possess a proboscis. Some members have jaws,
while others are jawless. They also possess multiple feeding methods, with some
Aquatic macroinvertebrates of Nee Soon
101
Fig. 20. Distribution and abundance of order Rhynchobdellida in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
being ectoparasites which feed on blood, and others being predators which hunt down
and swallow smaller invertebrates whole (Ruppert & Barnes, 1994). In the Nee Soon
drainage, one family (Erpobdellidae) from this order has been found in the course of
this study, and only a single specimen was documented (Site 11) (Fig. 19).
Order Rhynchobdellida
The Rhynchobdellida are also known as the jawless leeches, but this common name
is misleading due to the fact that many leech species from the order Arhynchobdellida
are also jawless. The true mark of a member of the Rhynchobdellida is the possession
of a proboscis, which is used to feed. In the Nee Soon drainage, one individual from
the family Glossiphoniidae was documented (Site 32) (Fig. 20).
102
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 21. Distribution and abundance of subclass Oligochaeta in the Nee Soon drainage. Solid
circles indicate sites where representatives were collected and circle sizes are proportionate to
abundance/numbers of individuals captured as reflected in the accompanying legend.
Subclass Oligochaeta
Oligochaeta contains the familiar terrestrial earthworms as well as more obscure
annelids which live in freshwater habitats. They are distributed throughout the entire
Nee Soon drainage (being found in 24 sites in total), and are typically found in relatively
small numbers (1-12 individuals at 22 sites) (Fig. 21). However, large numbers of
Oligochaeta, 22 and 50 individuals, were recorded at sites 25 and 29 respectively.
Aquatic macroinvertebrates of Nee Soon
103
Discussion
The Nee Soon freshwater swamp forest is a vital refuge for the biodiversity of
Singapore, due to the large number of species which can be found nowhere else in
Singapore (Clews et al., 2018; Davison et al., 2018). Furthermore, some species are
endemic to the Nee Soon freshwater swamp forest and found nowhere else in the
world (Ng & Lim, 1992; Cumberlidge et al., 2009; Lim et al., 2011). These species
would likely be extirpated from Singapore or rendered globally extinct, should their
habitat be damaged or severely altered. In order to conserve Singapore’s native
biodiversity, there is a need to understand the diversity and distribution of the aquatic
macroinvertebrates in the Nee Soon freshwater swamp forest and develop a baseline
dataset. Additionally, further studies can be built on this baseline dataset to inform
policy makers and aid in the conservation efforts of the Nee Soon freshwater swamp
forest.
The results of this study suggest that the conservation of the Nee Soon
freshwater swamp forest is pivotal in protecting the rich aquatic macroinvertebrate
biodiversity found there. Surveys during this study yielded a higher number of aquatic
macroinvertebrate taxonomic groups (76 families and six higher taxa) as compared
to the number of macroinvertebrate taxonomic groups (68 families and 6 higher taxa)
collected across 47 urban concrete canals and natural forest streams within the Central
Catchment Nature Reserve (Blakely et al., 2014). Moreover, only 66% of the taxa (54
out of 82 taxonomic groups) in this study are reported in the study by Blakely et al
(2014). This means that the aquatic biodiversity in Nee Soon freshwater swamp forest
consists of a large number of rare taxa (e.g. Psephenidae or water-penny beetle and the
Xiphocentronidae caddisfhes) that are not found in other habitats within Singapore.
This further highlights the need to prioritise the conservation and the protection of the
natural resources within the Nee Soon drainage.
The results of this study demonstrate variations in the distribution of
macroinvertebrate groups within the forest streams with different communities
represented in different parts of the forest. These communities reflect heterogeneous
environmental conditions across the catchment. Some taxa, including both mollusc
classes and the crustacean class Ostracoda, are confined to only the northeast area of
the Nee Soon drainage. Common taxa, such as the Diptera and the Ephemeroptera,
are also found in extremely large numbers in this area and in reduced numbers deeper
inside (further upstream) the Nee Soon drainage. The northeast area of the catchment
receives additional input of less acidic water from an adjacent reservoir. The decreased
acidity here is more suitable for some of the mollusc taxa, since molluscs are sensitive
to low pH levels due to the corrosive effects on their shells (Clements et al., 2006;
Gazeau et al., 2013; Parker et al., 2013).
In addition to the less acidic conditions in the northeast area of the Nee Soon
drainage, the streams are more open and lacking canopy cover, unlike the densely
forested streams deeper within the freshwater swamp forest. Macroinvertebrate
groups such as Plecoptera, Megaloptera and some of the Ephemeroptera (Family
Heptageniidae and Leptophlebiidae) were absent in these less-forested areas.
104
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Macroinvertebrate communities in forested streams are strongly influenced by debris
input from riparian vegetation. The lack of canopy cover could result in a community
that favours taxa with diverse feeding habits that can deal with the allochthonous debris
(e.g, plecopteran shredders that feed by cutting and tearing large pieces of debris/
organic matter) (Schmidt-Kloiber et al., 2006).
Habitats at the outskirts of the Nee Soon catchment are more exposed to potential
encroachment of non-native taxa from other catchments. For example, a South
American cichlid, Acarichthys heckelii, was recently reported for the first time within
the outskirts of the Nee Soon freshwater swamp forest (Tan & Lim, 2008; Ng & Tan,
2010). Whilst existing environmental conditions may hinder the establishment of non¬
native species, the constant propagule pressure from the reservoir (Yeo & Chia, 2010)
may pose a future threat (Holle & Simberloff, 2005; Lockwood et al., 2005). The Nee
Soon streams also drain into the Lower Seletar Reservoir via Sungai Seletar, which
may be another potential source of non-native taxa, since Lower Seletar Reservoir also
contains many established populations of non-native species (Ng & Tan, 2010).
While the present study captured a representation of forest stream fauna across
the catchment, there is always the possibility that certain groups were not obtained
in the course of sampling due to limitations in the sampling methods. In the future,
additional methods such as deployment of colonisers could be utilised in addition to
traditional kick sampling methods, in order to ensure that a wider range of diversity is
captured. As the life cycle of numerous insects are associated with different types of
habitat (e.g. all aquatic dipteran larvae emerge as terrestrial adults), sampling methods
that capture adult insects can also be useful to supplement the determination of aquatic
macroinvertebrate diversity. Malaise traps (Gressitt & Gressitt, 1962) are an effective
means of collecting many types of adult insects, and have been deployed in various
parts of Singapore (including the Nee Soon freshwater swamp forest) for taxonomic
work with adult insects (e.g. Grootaert, 2006; Ngiam & Cheong, 2016). Emergence
traps have also been adapted for use in tropical streams (e.g. in the Philippines) (Freitag,
2004a, 2004b) and are useful for capturing emerging adults at specific riverine locations.
Alternative sampling methods for larvae collection include electroshocking techniques
(Taylor et al., 2001), which use electrofishing equipment to collect and quantitatively
sample stream invertebrates. Processing of samples collected by electroshocking
techniques has been reported to be 40% faster than traditional sampling methods, due
to a reduction of debris collected in the samples (Taylor et al., 2001).
Unfortunately, knowledge of aquatic macroinvertebrate taxonomy is limited in
the tropical Asian region. Many of the molluscs and decapods can be identified to
the species level (Ng, 1988; Ng, 1990; Tan et al., 2012), and some taxonomic groups
such as the Odonata or the Hemiptera can be identified to the genus level (Yeo, 2012;
Tran et al., 2015). However, many other groups including the Trichoptera and Diptera
cannot be identified beyond the family level without the assistance of taxonomic
experts. This results in a loss of resolution and information about species distribution,
especially with regards to species of conservation value. Identifying specimens down
to the genus and the species levels could be achieved in future studies. For instance,
in South Korea, dedicated studies of aquatic macroinvertebrate species facilitated
Aquatic macroinvertebrates of Nee Soon
105
the identification and the designation of endemic Plecoptera (e.g. Scopura gaya and
Scopura jin ) and Trichoptera species (e.g. Agrypnia pagetana) as threatened species
(Kim et al., 2014). Incorporation of molecular techniques (e.g. DNA barcoding)
into identification workflow in future studies would vastly improve this process, as
molecular barcoding allows for the linkage of adult and juvenile forms (Zhou et al.,
2007; Kutty et al., 2018), as well as allowing for the identification of cryptic species
down to the species level. Indeed, previous work by Ball et al. (2005) and Pfenninger
et al. (2007) has shown that high levels of accuracy from molecular techniques can be
achieved in the identification of Ephemeroptera and Chironomidae, respectively.
ACKNOWLEDGEMENTS. We would like to thank Y. Cai, T. Li and W.H. Lim for their
assistance in the field. This study was approved by the National Parks Board of Singapore
(Permit no. NP/RP13-068-1). Funding for this study was received from the National Parks
Board of Singapore (National University of Singapore grant number R-347-000-198-490).
References
Balke, M., Hendrich, L. & Yang, C.M. (1997). Water beetles (Insecta: Coleoptera) in the Nature
Reserves of Singapore. Gard. Bull. Singapore 49(2): 321-331.
Ball, S.L., Hebert, P.D.N., Burian, S.K. & Webb, J.M. (2005). Biological identifications of
mayflies (Ephemeroptera) using DNA barcodes. J. N. Am. Benthol. Soc. 24(3): 508-524.
Blakely, T.J., Harding, J.S., Clews, E. & Winterbourn, M.J. (2010). An illustrated guide to
the freshwater macroinvertebrates of Singapore. Christchurch: School of Biological
Sciences, University of Canterbury.
Blakely, T.J., Eikaas, H.S. & Harding, J.S. (2014). The SingScore: a macroinvertebrate biotic
index for assessing the health of Singapore’s streams and canals. Raffles Bull. Zool. 62:
540-548.
Bressan, M., Chinellato, A., Munari, M., Matozzo, V., Manci, A., Marceta, T., Finos, L., Moro,
I., Pastore, R, Badocco, D. & Marin, M.G. (2014). Does seawater acidification affect
survival, growth and shell integrity in bivalve juveniles? Mar. Environ. Res. 99: 136—
148.
Cai, Y., Ng, C.Y. & Ngiam, R.W.J. (2018). Diversity, distribution and habitat characteristics
of dragonflies in Nee Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore
70 (Suppl. 1): 123-153.
Clements, R., Koh, L.P, Lee, T.M., Meier, R. & Li, D. (2006). Importance of reservoirs for
the conservation of freshwater molluscs in a tropical urban landscape. Biol. Conserv.
128(1): 136-146.
Clews, E., Low, E., Belle, C.C., Todd, P.A., Eikaas, H.S. & Ng, P.K.L. (2014). A pilot
macroinvertebrate index of the water quality of Singapore’s reservoirs. Ecol. Indie. 38:
90-103.
Clews, E., Corlett, R.T., Ho, J.K.I., Koh, C.Y., Liong, S.Y., Memory, A., Ramchunder, S.J.,
Siow, H.J.M.P., Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W., Theng, M.T.Y. & Yeo, D.C.J.
(2018) The biological, ecological and conservation significance of freshwater swamp
forest in Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
106
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Cover, M.R. & Resh, V.H. (2008). Global diversity of dobsonflies, fishflies, and alderflies
(Megaloptera; Insecta) and spongillaflies, nevrorthids, and osmylids (Neuroptera;
Insecta) in freshwater. Hydrobiologia 595(1): 409-417.
Cumberlidge, N., Ng, P.K.L., Yeo, D.C.J., Magalhaes, C., Campos, M.R., Alvarez, F., Naruse,
T., Daniels, S.R., Esser, L.J., Attipoe, F.Y.K., Clotilde-Ba, F.-L., Darwall, W., Mclvor,
A., Baillie, J.E.M., Collen, B. & Ram, M. (2009). Freshwater crabs and the biodiversity
crisis: Importance, threats, status, and conservation challenges. Biol. Conserv. 142(8):
1665-1673.
Davison, G.W.H., Cai, Y., Li, T.J. & Lim, W.H. (2018). Integrated research, conservation
and management of Nee Soon freshwater swamp forest, Singapore: hydrology and
biodiversity. Gard. Bull. Singapore 70 (Suppl. 1): 1-7.
Dudgeon, D. (1999). Tropical Asian streams: zoobenthos, ecology and conservation. Hong
Kong: Hong Kong University Press.
Freitag, H. (2004a). Adaptations of an emergence trap for use in tropical streams. Int. Rev.
Hydrobiol. 89(4): 363-374.
Freitag, H. (2004b). Composition and longitudinal patterns of aquatic insect emergence in
small rivers of Palawan Island, the Philippines. Int. Rev. Hydrobiol. 89(4): 375-391.
Gani, M.B.A. (2013). Biodiversity assessment of aquatic insects in Nee Soon swamp area,
Singapore. Unpublished B.Sc. (Hons) thesis. Singapore: Nanyang Technological
University.
Gazeau, F., Parker, L.M., Comeau, S., Gattuso, J.-P, O’Connor, W.A., Martin, S., Portner,
H.-O. & Ross, P.M. (2013). Impacts of ocean acidification on marine shelled molluscs.
Mar. Biol. 160(8): 2207-2245.
Gressitt, J. & Gressitt, M. (1962). An improved Malaise trap. Pac. Insects 4: 87-90.
Grootaert, P. (2006). The genus Teuchophorus (Diptera, Dolichopodidae) in Singapore. Raffles
Bull. Zool. 54(1): 59-82.
Hendrich, L., Balke, M. & Yang, C.M. (2004). Aquatic Coleoptera of Singapore - species
richness, ecology and conservation. Raffles Bull. Zool. 52: 97-141.
Holle, B.V. & Simberloff, D. (2005). Ecological resistance to biological invasion overwhelmed
by propagule pressure. Ecology 86(12): 3212-3218.
Kim, S.-B., Suh, M.-H., Lee, B.-Y., Kim, S.T., Park, C.-H., Oh, H.-K., Kim, H.-Y., Lee, J.-H. &
Lee, S.Y. (eds) (2014). Korean Red List of Threatened Species, 2 nd ed. Seoul: National
Institute of Biological Resources.
Kutty, S.N., Wang, W., Ang, Y., Tay, Y.C., Ho, J.K.I. & Meier, R. (2018). Next-generation
identification tools for Nee Soon freshwater swamp forest, Singapore. Gard. Bull.
Singapore 70 (Suppl. 1): 155-173.
Li, T., Chay, C.K., Lim, W.H. & Cai, Y. (2016). The fish fauna of Nee Soon Swamp Forest,
Singapore. Raffles Bull. Zool. Suppl. 32: 56-84.
Lim, K.K.P., Yeo, D.C.J. & Ng, P.K.L. (2011). Nee Soon Swamp Forest. In: Ng, P. K. L.,
Corlett, R. T., & Tan, H. T. W. (eds) Singapore biodiversity: an encyclopedia of the
natural environment and sustainable development, pp. 54-56. Singapore: Editions
Didier Millet.
Lim, W.H., Li, T.J. & Cai, Y. (2018). Terrestrial snails and slugs diversity in Nee Soon freshwater
swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 109-121.
Lockwood, J.L., Cassey, P. & Blackburn, T. (2005). The role of propagule pressure in explaining
species invasions. Trends Ecol. Evol. 20(5): 223-228.
Loke, L.H.L., Clews, E., Low, E., Belle, C.C., Todd, P.A., Eikaas, H.S. & Ng, P.K.L. (2010).
Methods for sampling benthic macroinvertebrates in tropical lentic systems. Aquat.
Biol. 10(2): 119-130.
Aquatic macroinvertebrates of Nee Soon
107
Lowe, S., Browne, M., Boudjelas, S. & De Poorter, M. (2004). 100 of the world’s worst invasive
alien species: A selection from the Global Invasive Species Database. The Invasive
Species Specialist Group, Species Survival Commission, IUCN.
Merritt, R.W. & Cummins, K.W. (eds). (1996). An introduction to the aquatic insects of North
America. Dubuque: Kendall/Hunt Publishing Company.
Murphy, D.H. (1997). Odonata biodiversity in the Nature Reserves of Singapore. Gard. Bull.
Singapore 49(2): 333-352.
Ng, P.K.L. (1988). The freshwater crabs of Peninsular Malaysia and Singapore. Singapore:
Department of Zoology, National University of Singapore.
Ng, P.K.L. (1990). Freshwater crabs and prawns of Singapore. In: Chou, L.M. & Ng, P.K.L.
(eds) Essays in Zoology. Papers commemorating the 40th anniversary of the Department
of Zoology, National University of Singapore, pp. 189-204. Singapore: Department of
Zoology, National University of Singapore.
Ng, P.K.L. (1997). The conservation status of freshwater prawns and crabs in Singapore with
emphasis on the nature reserves. Gard. Bull. Singapore 49: 267-272.
Ng, P.K.L. & Lim, K.K.P. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquat. Conserv. Mar. Freshw. Ecosyst. 2(3): 255-266.
Ng, P.K.L. & Lim, K.K.P. (1997). The diversity and conservation status of fishes in the nature
reserves of Singapore. Gard. Bull. Singapore 49(2): 245-265.
Ng, H.H. & Tan, H.H. (2010). An annotated checklist of the non-native freshwater fish species
in the reservoirs of Singapore. COSMOS 6(1): 95-116.
Ng, P.K.L., Corlett, R.T. & Tan, H.T.W. (eds) (2011). Singapore biodiversity: an encyclopedia
of the natural environment and sustainable development. Singapore: Editions Didier
Millet.
Ngiam, R.W.J. & Cheong, L.F. (2016). The dragonflies of Singapore: An updated checklist and
revision of the national conservation statuses. Nat. Singapore 9: 149-163.
Nguyen, C.T.T., Wasson, R.J. & Ziegler, A.D. (2018). The hydro-geomorphic status of the
Nee Soon freshwater swamp forest catchment of Singapore. Gard. Bull. Singapore 70
(Suppl. 1): 33-48.
Norma-Rashid, Y., Cheong, L.F., Lua, H.K. & Murphy, D.H. (2008). The dragonflies
(Odonata) ofSingapore: Current status records and collections of the Raffles Museum of
Biodiversity Research. Singapore: Raffles Museum of Biological Research, Department
of Biological Sciences, National University of Singapore.
O’Dempsey, T. (2014). Singapore’s changing landscape since c.1800. In: Barnard, T. P. (ed.)
Nature Contained: Environmental Histories of Singapore, pp. 18-48. Singapore: NUS
Press.
O’Dempsey, T. & Chew, PT. (2013). The freshwater swamp forests of Sungei Seletar
Catchment: a status report. In: Leong, T. M. & Ho, H. C. (eds) Proceedings of Nature
Society, Singapore’s Conference on “Nature Conservation for a Sustainable Singapore ”
- 16 October 2011, pp. 121-166. Singapore: Nature Society, Singapore.
Parker, L.M., Ross, P.M., O’Connor, W.A., Portlier, H.O., Scanes, E. & Wright, J.M. (2013).
Predicting the response of molluscs to the impact of ocean acidification. Biology 2(2):
651-692.
Pfenninger, M., Nowak, C., Kley, C., Steinke, D. & Streit, B. (2007). Utility of DNA taxonomy
and barcoding for the inference of larval community structure in morphologically
cryptic Chironomus (Diptera) species. Mol. Ecol. 16(9): 1957-1968.
Ricciardi, A., Neves, R.J. & Rasmussen, J.B. (1998). Impending extinctions of North American
freshwater mussels (Unionoida) following the zebra mussel {Dreissena polymorpha )
invasion. /. Anim. Ecol. 67(4): 613-619.
108
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Romoser, W.S. & Stoffolano, J.G. (1994). The science of entomology. Dubuque: Wm. C.
Brown Publishers.
Ruppert, E.E. & Barnes, R.D. (1994). Invertebrate zoology. Fort Worth: Saunders College
Publishing.
Schmidt-Kloiber, A., Graf, W., Lorenz, A. & Moog, O. (2006). The AQEM/STAR taxalist - a
pan-European macro-invertebrate ecological database and taxa inventory. Hydrobiologia
566(1): 325-342.
Sun, Y., Kim, D.Y., Wendi, D., Doan, D.C., Raghavan, S.V., Jiang, Z. & Liong, S.Y. (2018).
Projected impacts of climate change on stream flow and groundwater of Nee Soon
freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 175-190.
Tan, H.H. & Lim, K.K.P. (2008). Acarichthys heckelii (Muller & Troschel), an introduced
cichlid fish in Singapore. Nat. Singapore 1: 129-133.
Tan, H.T.W., Yeo, D.C.J., Chou, L.M. & Ng, P.K.L. (2010). The natural heritage of Singapore,
3 rd ed. Singapore: Prentice Hah.
Tan, S.K., Chan, S.Y. & Clements, G.R. (2012). A guide to snails and other non-marine
molluscs of Singapore. Singapore: Science Centre Singapore.
Tang, H.B., Wang, L.K. & Hamalainen, M. (2010). A photographic guide to the dragonflies
of Singapore. Singapore: Raffles Museum of Biodiversity Research, Department of
Biological Sciences, National University of Singapore.
Taylor, B.W., McIntosh, A.R. & Peckarsky, B.L. (2001). Sampling stream invertebrates using
electro shocking techniques: implications for basic and applied research. Can. J. Fish.
Aquat. Sci. 58(3): 437-445.
Tran, A.D., Yang, C.M. & Cheng, L. (eds) (2015). Water bugs of Singapore and Peninsular
Malaysia. Singapore: Lee Kong Chian Natural History Museum.
Yang, C.M., Lua, H.K. & Yeo, K.L. (1997). Semi-aquatic bug (Heteromorpha: Gerromorpha)
fauna in the Nature Reserves of Singapore. Gard. Bull. Singapore 49(2): 313-320.
Yeo, D., Puniamoorthy, J., Ngiam, R. W. J. & Meier, R. (in press). Towards holomorphology in
entomology: rapid and cost-effective larval-adult matching using NGS barcodes. Syst.
Entomol.
Yeo, D.C.J. & Chia, C.S.W. (2010). Introduced species in Singapore: an overview. COSMOS
6(1): 23-37.
Yule, C.M. & Yong, H.-S. (eds). (2004). Freshwater invertebrates of the Malaysian region.
Kuala Lumpur: Akademi Sains Malaysia.
Zhou, X., Kjer, K.M. & Morse, J.C. (2007). Associating larvae and adults of Chinese
Hydropsychidae caddisflies (Insecta: Trichoptera) using DNA sequences. J. N. Am.
Benthol. Soc. 26(4): 719-742.
Zischke, J.A., Arthur, J.W., Nordlie, K.J., Hermanutz, R.O., Standen, D.A. & Henry, T.P
(1983). Acidification effects on macro invertebrates and fathead minnows (Pimephales
promelas) in outdoor experimental channels. Water Res. 17(1): 47-63.
Gardens’ Bulletin Singapore 70 (Suppl. 1): 109-121. 2018
doi: 10.26492/gbs70(suppl.l). 2018-06
109
Diversity of terrestrial snails and slugs
in Nee Soon freshwater swamp forest, Singapore
W.H. Lim 1 - 2 , T.J. Li 1 & Y. Cai 1
National Biodiversity Centre, National Parks Board,
1 Cluny Road, 259569 Singapore
caLyixiong @ nparks. go v. sg
2 Current address: The Herbarium, National Parks Board,
1 Cluny Road, 259569 Singapore
ABSTRACT. Nee Soon freshwater swamp forest is the last remaining primary freshwater
swamp forest left in Singapore and it contains a rich diversity of native and locally threatened
fauna. As native terrestrial snails and slugs are poorly studied and understood in Nee Soon
freshwater swamp forest, an extensive survey was conducted to establish their current status.
A total of 19 species was recorded, of which one was recorded for Singapore for the first time.
Amphidromus atricallosus temasek, a recently described subspecies endemic to Singapore, was
found to be more commonly distributed than previously known from the swamp forest. Results
also indicate that despite low overall abundance, Nee Soon freshwater swamp forest harbours a
rich diversity of land snails and slugs. Any future long term changes in climate or topography,
or short term changes in hydrology, might affect their distribution and diversity.
Keywords. Land molluscs, Mollusca, species richness
Introduction
Nee Soon freshwater swamp forest is one of the most important conservation sites
in Singapore and contains a significant number of native, endemic and nationally
threatened flora and fauna (Ng & Lim, 1992; Ng, 1997; Clews et al., 2018). The forest
is located within the Central Catchment Nature Reserve, bordered by the Upper Seletar
Reservoir to the north, Seletar Expressway and Old Upper Thomson Road to the east,
Upper and Lower Peirce Reservoirs to the south, and the southwestern tributary of the
Upper Seletar Reservoir and the northern-most tributary of the Upper Peirce Reservoir
to the west (Yeo & Lim, 2011). The forest consists of primary and old secondary
vegetation, the stream catchment covering an area of approximately 500 hectares.
To date, a total of 51 terrestrial snail and slug species (and subspecies),
representing 33 genera in 15 families, have been recorded in Singapore (Ho, 1995;
Tan & Chan, 2009; Tan & Woo, 2010; Tan et al., 2011; Tan et al., 2012; Tan & Chan,
2013). Many of these species are considered to be forest dependent, recorded mainly
in the Singapore Botanic Gardens rain forest, Central Catchment Nature Reserve
(including Nee Soon freshwater swamp forest) and Bukit Timah Nature Reserve.
Forest dependent species include Cyclophorus perdix aquila, Cyclotus rostellatus,
Ditropis cf. koperbergi, Japonia ciliocinctum, Diplommatina nevilli, Microparmarion
110
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
strubelli, Hemiplecta Humphreysiana, Dyakia kintana, Amphidromus atricallosus
temasek and Amphidromus inversus. There has been no comprehensive information
on the terrestrial snail and slug species found in Nee Soon freshwater swamp forest,
even during the last extensive biodiversity survey of the nature reserves conducted in
1993-1997. Therefore, this study aims to provide baseline information on the biology
and distribution of terrestrial snails and slugs within Nee Soon. This information will
be useful in updating the status of individual species and will contribute to the design
of conservation efforts for the freshwater swamp forest.
Material and Methods
Quantitative hand picking sampling was carried out from September 2013 to October
2015 in 15 plots spread broadly across the Nee Soon catchment in four different
areas: Upper Swamp (Upper 1 & 2 Sub-catchments), Middle Swamp (Mid 1-3 Sub¬
catchments), Lower Swamp (Lower 3 Sub-catchments) and Outskirts (Lower 1 & 2
Sub-catchments) (Fig.l). The plots were sampled repeatedly for six collection cycles.
Each plot consisted of two 10 x 5m quadrats. Each quadrat was searched thoroughly
by two people for half an hour to locate any snails or slugs. Most of the specimens
were found on the forest floor, hidden within the leaf litter or up on the trees, on
the underside of leaves. Shell height, width and chirality were recorded at time of
collection. Shell height was measured from the apex to the lowest part of the outer
lip parallel to the coiling axis and shell width was measured at the widest part of the
body whorl perpendicular to the coiling axis. Body length was recorded for slugs.
Temperature and humidity were also measured at each site and summarised in Table 1.
The Shannon-Weiner Index (EP) was used to study and compare species diversity
at each site. The equation for the Shannon-Weiner Index is as follows:
H’ = -S(Pilog[Pi]),
where Pi = number of individual species / total number of individuals
Results
Family CYCLOPHORIDAE
Cyclophorus perdix aquila (Sowerby, 1843), Fig. 2A
Cyclophorus perdix aquila is one of several forest land snail species that survive in
Singapore’s rainforests. Species from Cyclophoridae require moisture to be secreted
on the surface of the mantle cavity to aid in respiration, thus most of them are restricted
to areas that are wet and moist (Tan et al., 2012). All specimens were found near to
forest streams or swampy areas.
Terrestrial molluscs in Nee Soon swamp forest
111
Upper2
ier Peirce Reservoir
Lower Peirce Reservoir
Upper Seietar Reservoir
0 0.1 5 0.3 0.6 Kilometers
.. .
Legend
A Sampling site
- Stream
Nee Soon forest boundary
I I Catchment boundary
Fig. 1 . Map of sampling plots in Nee Soon freshwater swamp forest.
112
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 1. Mean temperature and humidity of the sampling plots.
Site
Humidity (Mean)
Temperature
(Mean)/°C
Upper Swamp 1
76.08
29.46
Upper Swamp 2
78.58
30.05
Middle Swamp 1
87.00
28.05
Middle Swamp 2
90.13
27.75
Middle Swamp 3
92.00
26.20
Middle Swamp 4
90.83
27.63
Middle Swamp 5
81.00
29.51
Middle Swamp 6
85.38
28.34
Lower Swamp 1
81.00
30.25
Lower Swamp 2
89.00
28.96
Lower Swamp 3
83.75
28.70
Lower Swamp 4
86.13
28.96
Outskirt 1
69.92
32.10
Outskirt 2
80.50
28.50
Outskirt 3
76.63
32.19
Cyclotus rostellatus (Pfeiffer, 1851), Fig. 2B
Cyclotus rostellatus is one of the rarer land snails found in Nee Soon freshwater swamp
forest. Restricted to undisturbed forested areas, this snail has only been recorded in
Nee Soon since 1990. The shell has a short sutural tube present near the aperture which
is believed to aid in respiration in moist environments (Ho, 1995). All specimens were
observed in moist areas with dense canopy cover.
Japonia ciliocinctum (Martens, 1865), Fig. 2C
Japonia ciliocinctum is mostly to be found foraging in the depths of the freshwater
swamp forest. The shell is dull brown in colour with a hairy periostracum along the
periphery. It has been speculated that the hairy periostracum might facilitate movement
in a moist environment by relieving surface tension (Pfenninger et al., 2005). However,
the hairy periostracum falls off easily upon handling. All observed specimens were
found in swampy areas with a dense canopy cover.
Family ACHATINIDAE
Achatina fulica Bowdich, 1822, Fig. 2D
Achatinafulica is an invasive species originating from Africa and is commonly known
as the Giant African Snail. In Nee Soon freshwater swamp forest shell heights of 7.5
cm or more have been observed. It is commonly found in parks and degraded forest
but rarely in undisturbed forest. Unfortunately, several specimens were recorded deep
Terrestrial molluscs in Nee Soon swamp forest
113
Fig. 2. A. Cyclophorus perdix aquila. B. Cyclotus rostellatus. C. Japonia ciliocinctum.
D. Achatina fulica. E. Lamellaxis gracilis. F. Dyakia kintana. G. Quantula striata. H.
Coneuplecta microconus. I. Liardetia convexoconica. J. Liardetia doliolum. K. Geotrochus
lychnia. L. Helicarion perfragilis. M. Hemiplecta Humphreysiana. N. Parmarion martensi. O.
Microparmarion strubelli. P. Damayantia off. simrothi. Q. Amphidromus atricallosus temasek.
R. Amphidromus egg cluster. (Photos: A-H, K-R W.H. Lim; I-J Y.C. Ang)
114
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
in the forest indicating that this species may have already established a population
within Nee Soon.
Family SUBULINIDAE
Lamellaxis gracilis (Hutton, 1834), Fig. 2E
Lamellaxis gracilis is an introduced species found on the outskirts of Nee Soon
freshwater swamp forest. It is likely to have spread from nearby plant nurseries and
gardens. Although the species is easily confused with Subulina octona, it can be
identified by its more slender shell shape and its columella is not truncated (Tan et al.,
2012 ).
Family DYAKIIDAE
Dyakia kintana (de Morgan, 1885), Fig. 2F
Dyakia kintana is mostly found in undisturbed forest. It can be easily recognised by
its sinistral coiled shell and the animal’s body ranges from pink to white in colour
with two black stripes. The periphery of the shell is acutely keeled. It has only been
observed on the ground, hiding among leaf litter and fallen timber.
Quantula striata (Gray, 1834), Fig. 2G
Quantula striata is found commonly in urban areas as well as in the forest. Most of
the specimens were observed in the outskirts of Nee Soon in close proximity to urban
areas. Its shell colour is known to be variable (Tan et al., 2012); however only reddish
brown examples were observed in Nee Soon freshwater swamp forest.
Family EUCONULIDAE
Coneuplecta microconus (Mousson, 1865), Fig. 2H
Coneuplecta microconus is a very tiny snail that was found near the outskirts of Nee
Soon. The shell is conical in shape but unlike Liardetia convexoconica the periphery
is acutely keeled.
Liardetia convexoconica (Moellendorff, 1897), Fig. 21
Liardetia convexoconica is the most abundant species found in Nee Soon freshwater
swamp forest. The shell is conical in shape and ranges from light brown to yellow
in colour. The animal’s body is yellow in colour with a tint of red on its head. It is
arboreal and mostly found in shaded areas, hidden underneath the leaves.
Liardetia doliolum (Pfeiffer, 1846), Fig. 2J
Liardetia doliolum is a common species that can be found in both forested and urban
Terrestrial molluscs in Nee Soon swamp forest
115
areas. It has a tiny shell with distinct ribs on the outer surface. It has a lower spire than
Liardetia convexiconica and its shell is dull brown. It was found near the outskirts of
Nee Soon freshwater swamp forest.
Microcystina sp.
Microcystina is a tiny snail with a shell width of less than 3 mm. The shell is glossy
and brownish in colour. Only one taxon within the genus appears to be present within
Singapore, but it was not attributed a species name by Ho (1995), nor is it here. Only
three specimens were observed throughout the survey.
Family TROCHOMORPHIDAE
Geotrochus lychnia (Benson, 1852), Fig. 2K
Geotrochus lychnia is a very rare species that was only encountered twice during the
survey, in both cases only as empty shells. Its shell has a small umbilical opening and
its periphery is acutely keeled. Singapore is the type locality for Geotrochus lychnia.
Helicarion perfragilis (von Mollendorff, 1897), Fig. 2L
Helicarion perfragilis is an arboreal snail that is commonly found in open vegetation.
The shell is very fragile and breaks easily. It was observed in rather high numbers near
the outskirts of Nee Soon but none was recorded within the forest.
Family ARIOPHANTIDAE
Hemiplecta Humphreysiana (Lea, 1841), Fig. 2M
Hemiplecta humphreysiana is the largest native snail found in Nee Soon freshwater
swamp forest and can grow to a shell width of up to 5 cm. The shell is brown with a
dark brown stripe that runs along the periphery of mature specimens. Although, this
species is commonly known to be found on algae covered logs and the forest floor
(Ho, 1995), several individuals were observed clinging on to trees up to more than 2
metres high. It was observed feeding on fungus on many occasions. Singapore is the
type locality for Hemiplecta humphreysiana.
Parmarion martensi Simroth, 1893, Fig. 2N
Parmarion martensi is a common slug found in many gardens and parks. Only two
individuals were observed during the survey, at the outskirts of Nee Soon.
Microparmarion strubelli Simroth, 1893, Fig. 20
Microparmarion strubelli is restricted to forested areas with a dense canopy cover.
Unlike other snails, its shell is reduced to a thin plate to which it can no longer
retract for protection. Black markings are observed on the visceral hump. Only three
specimens were observed throughout the whole survey.
116
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Damayantia aff. simrothi Collinge, 1903, Fig. 2P
Damayantia aff. simrothi is recorded for the first time in Singapore. However,
available references and comparative material in Singapore are few and inadequate
to provide determination of the species, thus the identification of the slug mentioned
is provisional. Its body is yellow in colour and the mantle is smooth. Its foot is keeled
with a series of ridges (rugae). It is likely to be arboreal as all the specimens were
observed foraging on trees or underneath the leaves.
Family BRADYBAENIDAE
Bradybaena similaris (Ferussac, 1821)
Bradybaena similaris is commonly found throughout the moist tropics in urban areas
such as gardens and plant nurseries. It is an agricultural pest and is most likely to have
been introduced due to the horticultural and agricultural trade (Tan et al., 2012). It was
observed in high numbers near the outskirts of Nee Soon but none was recorded within
the freshwater swamp forest.
Family CAMAENIDAE
Amphidromus atricallosus temasek, Tan, Chan & Panha, 2011, Fig. 2Q
Amphidromus atricallosus temasek is a recently described subspecies endemic to
Singapore (Tan et al., 2011) and is restricted to forest within the Central Catchment
Nature Reserve, Western Catchment (Chua & Tan, 2015), Pulau Ubin (Tan & Xu,
2013) and Pulau Tekong (Tan et al., 2015). Populations include individuals with both
dextral and sinistral chirality. The ratio of dextral and sinistral specimens found is
approximately 1:1 and interestingly both are well distributed within the swamp. This
species is arboreal and spends much of the time up in the trees. It can sometime be
observed on man-made concrete structures. A suspected egg cluster was observed
during the survey (Fig. 2R).
Distribution of land snails
A total of ten families, 18 genera and 19 species of land snails were recorded from the
15 plots in Nee Soon. The data are shown in Table 2. Four species of snails: Dyakia
kintana, Liardetia convexoconica, Hemiplecta humphreysiana and Amphidromus
atricallosus temasek were identified as having a wide distribution within Nee Soon,
being recorded in at least ten of the 15 plots. Four species of snails, Coneuplecta
microconus, Helicarion perfragilis, Parmarion martensi and Bradybaena similaris
were confined to the outskirts of Nee Soon.
Terrestrial molluscs in Nee Soon swamp forest
117
Species diversity and richness
The Shannon-Weiner Index (H’) is commonly used as a measure for species diversity
between habitats (Krebs, 1989). Based on the average H’ calculated above for the 15
sampling plots (Fig. 3), the Lower Swamp 1 had the highest species diversity (H’=
0.53), while the Upper Swamp had the lowest species diversity (H’=0). Many of the
non-native snails are recorded in the Outskirts, reflected in a high H’ but this does not
reflect the true native diversity.
The Species Richness (R) indicates the average number of species present at
each of the survey sites (Fig. 4). Based on the table above, Lower Swamp 1 had the
highest species richness (R=4.5) and Upper Swamp 2 had the lowest species richness
(R=0.33). The number of non-native snails recorded in the Outskirts distorts the
species richness value there.
Mean population abundance
The three most abundant species found in Nee Soon freshwater swamp forest are
Liardetia convexoconica, Helicarion perfragilis and Hemiplecta Humphreysiana. All
three species had a mean frequency of more than ten individuals per cycle of sampling
(Fig. 5). However, six species of snails: Japonia ciliocinctum, Lamellaxis gracilis,
Microcystina sp., Geotrochus lychnia, Parmarion martensi and Microparmarion
strubelli are extremely rare in Nee Soon and had a low mean frequency of fewer than
one per cycle of sampling.
Discussion
Studies on the land snails in Nee Soon freshwater swamp forest revealed a rich diversity
despite the low abundance. Nineteen species of snails and slugs were recorded of
which one species was documented for the first time in Singapore. Of the 19 species,
16 are native and three are introduced. Eight are forest dependent species: Cyclophorus
perdix aquila, Cyclotus rostellatus, Japonia ciliocinctum, Dyakia kintana, Hemiplecta
Humphrey siana, Microparmarion strubelli, Damayantia cf. simrothi and Amphidromus
atricallosus temasek, only observed in forested areas with a dense canopy cover.
Introduced species are restricted to the outskirts of the Nee Soon freshwater swamp
forest, with the exception of Achatina fulica on rare occasions.
It is not surprising that low abundance of land snails was observed due to
the nature of a freshwater swamp forest. As it is well known that land snails have a
high calcium requirement for the formation of their shells and eggs, the abundance
of snails correlates with the soil calcium content (Graveland et al., 1994). However,
Singapore’s tropical soils are generally acidic and poor in calcium (Chia & Foong,
1991), especially in freshwater swamp forest. Thus a shortage of calcium in Nee Soon
soils (Nguyen et al., 2018) may be a limiting factor to the growth and reproduction
Table 2. Distribution of land snails at sites 1-15 in Nee Soon freshwater swamp forest across 6 cycles (* represents introduced species).
Abbreviations: +++: Abundant (>15.00), ++: Common (5.00-14.00), +: Rare (1.00-4.00), -: Not found at site
US: Upper Swamp, MS: Middle Swamp, LS: Lower Swamp, O: Outskirt
118
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Amphidromus atricallosus + + ++ + + + + +++
temcisek
Terrestrial molluscs in Nee Soon swamp forest
119
Fig. 3. Average Shannon-Weiner Index (H’) of terrestrial snails and slugs at all survey sites.
Fig. 4. Average Species Richness (R) of terrestrial snails and slugs at all survey sites.
of these land snails. The sampling method also plays a part in the observed low
abundance. The handpicking sampling technique is limited, focuses sampling only
from the ground level up to vegetation as high as two metres, and might miss very tiny
or well camouflaged specimens. Many of the arboreal species that dwell in the canopy
are not recorded. The lack of soil and leaf litter sieving sampling method also misses
out on species that dwell within the soil.
120
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 5. Mean population abundance of terrestrial snails and slugs at all survey sites.
We observed that the relative moisture level of the area plays an important role
in the distribution of the land snails. Based on the survey results, the Middle Swamp
and Lower Swamp generally exhibit a higher diversity and richness as compared to the
Upper Swamp. This may be due to the wetter and more humid habitat in the Middle
Swamp and Lower Swamp. Many species such as Cyclophorusperdix aquila, Cyclotus
rostellatus and Japonia ciliocinctum were only observed in partially waterlogged
areas, in close proximity to forest streams. Thus, any change in the hydrology of the
Nee Soon catchment may have adverse effects on the diversity and distribution of the
land snails.
Nee Soon freshwater swamp forest is the last remaining freshwater swamp
forest in Singapore. Not only does it harbour a rich diversity of land snails and slugs,
it houses a significant population of the endemic Amphidromus atricallosus temasek.
The discovery of the newly recorded Damayantia cf. simrothi also indicates that there
may be more species waiting to be discovered. Unfortunately, the freshwater swamp
forest ecosystem is extremely fragile, facing threats from environmental changes and
urban development (Clews et al., 2018). Furthermore, the changes in global climate
leading to extreme weather events will put additional stress on the forest (Sun et al.,
2018). Unless measures and precautions are taken to safeguard the Nee Soon freshwater
swamp forest, the rich biodiversity and heritage might be lost forever.
ACKNOWLEDGEMENTS. We would like to thank Drs Lena Chan and Geoffrey Davison
for their support, Tan Siong Kiat for assistance in species identification and colleagues for
helping in field surveys. This study forms part of the Nee Soon Swamp Forest Biodiversity
and Hydrology Baseline Studies Project funded by the National Parks Board (FC 12302501).
Terrestrial molluscs in Nee Soon swamp forest
121
References
Chia, L.S. & Foong S.F. (1991). Climate and weather. In: Chia, L.S., Rhaman, A. & Tay, D.B.H.
(eds) The Biophysical Environment of Singapore, pp. 89-133. Singapore: Singapore
University Press for the Geography Teachers’ Association, Singapore.
Chua, M.A.H. & Tan, S.K. (2015). Tree snail Amphidromus atricallosus temasek in Western
Catchment Area. Singapore Biodivers. Rec. 2015: 9.
Clews, E., Corlett, R.T., Ho, J.K.I., Koh, C.Y., Liong, S.Y., Memory, A., Ramchunder, S.,
Siow, H.J.M.P., Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W., Theng, M.T.Y. & Yeo, D.C.J.
(2018) The biological, ecological and conservation significance of freshwater swamp
forest in Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
Graveland, J., van der Wal, R., van Balen, J.H. & van Noordwijk, A. J. (1994). Poor reproduction
in forest passerines from decline of snail abundance on acidified soils. Nature 368:
446-448.
Ho, W.H. (1995). A Review of the Land-snail Fauna of Singapore. Raffles B. Zool. 43(1):
91-113.
Krebs, C.J. (1989). Ecological methodology, 2 nd ed. Menlo Park: Addison-Wesley Educational
Publishers.
Ng, P.K.L. (1997). The conservation status of freshwater prawns and crabs in Singapore with
emphasis on the Nature Reserves. Gard. Bull. Singapore 49: 267-272.
Ng, P.K.L. & Lim, K.K.P. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquatic Conserv. 2(3): 255-266.
Nguyen, C.T.T., Wasson, R.J. & Ziegler, A.D. (2018). The hydro-geomorphic status of the
Nee Soon freshwater swamp forest catchment of Singapore. Gard. Bull. Singapore 70
(Suppl. 1): 33-48.
Pfenninger, M., Hrabakova, M., Steinke, D. & Depraz, A. (2005). Why do snails have hairs? A
Bayesian inference of character evolution. BMC Evol. Biol. 5: 59.
Sun, Y., Kim, D.Y., Wendi, D., Doan, D.C., Raghavan, S.V., Jiang, Z. & Liong, S.Y. (2018).
Projected impacts of climate change on stream flow and groundwater of Nee Soon
freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 175-190.
Tan, S.K. & Chan, M.K.K. (2013). Land snail Cyclophorus semisulcatus rediscovered at Bukit
Timah. Singapore Biodivers. Rec. 2013: 4.
Tan, S.K. & Chan, S.Y. (2009). New records of predatory slugs from Singapore with notes on
their feeding behaviour. Nat. Singapore 2: 1-7.
Tan, S.K. & Woo, H.P. (2010). A Preliminary Checklist of the Molluscs of Singapore. Singapore:
Raffles Museum of Biodiversity Research, National University of Singapore, Singapore.
Tan, S.K. & Xu, W. (2013). Tree snail Amphidromus atricallosus temasek on Pulau Ubin.
Singapore Biodivers. Rec. 2013: 22.
Tan, S.K., Chan, S.Y. & Panha, S. (2011). Anew subspecies of Amphidromus (Amphidromus)
atricallosus from Singapore (Mollusca: Gastropoda: Camaenidae). Raffles B. Zool.
59: 39-46.
Tan, S.K., Chan, S.Y. & Clements, G.R. (2012). A Guide to Snails and other Non-marine
Molluscs of Singapore. Singapore: Science Centre, Singapore.
Tan, S.K., Lim, K.K.P. & Chua, M.A.H. (2015). Terrestrial snails and slugs (Mollusca:
Gastropoda) of Pulau Tekong, Singapore. Nat. Singapore 8: 25-30.
Yeo, D.C.J. & Lim, K.K.P. (2011) Lreshwater ecosystems. In: Ng, P.K.L., Corlett, R.T. & Tan,
H.T.W. (eds) Singapore Biodiversity: An Encyclopedia of the Natural Environment
and Sustainable Development , pp. 52-63. Singapore: Raffles Museum of Biodiversity
Research, Department of Biological Sciences, National University of Singapore.
Gardens’ Bulletin Singapore 70 (Suppl. 1): 123-153. 2018
doi: 10.26492/gbs70(suppl.l).2018-07
123
Diversity, distribution and habitat characteristics
of dragonflies in Nee Soon freshwater
swamp forest, Singapore
Y. Cai 1 , C. Y. Ng 1 & R.W.J. Ngiam 2
National Biodiversity Centre, National Parks Board,
1 Cluny Road, 259569 Singapore
cai_yixiong @ nparks. go v. sg
2 Central Nature Reserves, National Parks Board,
1 Cluny Road, 259569 Singapore
ABSTRACT. Biodiversity baselines were established for dragonflies of Nee Soon freshwater
swamp forest based on quantitative sampling across the eight sub-catchments. Surveys were
conducted from December 2014 to April 2016. Hydrological, physiochemical parameters and
habitats were analysed to identify the main drivers structuring the dragonfly community. A
total of 1706 odonate specimens were recorded, comprising 49 species of 34 genera in 11
families. The species diversity in each sub-catchment was compared using the Shannon-
Wiener Index (H’). Hierarchical clustering and Detrended Correspondence Analysis (DCA)
indicated that three main groupings of sites existed, each with a distinct community of
associated species. Further analysis by Canonical Correspondence Analysis (CCA) with 12
significant environmental variables showed that these groups were significantly associated with
respective environmental variables. Principal Components Analysis (PCA) was performed to
analyse the full 23 environmental variables. The first four principal components of the PCA
explained 63% of the variation in all the environmental variables. These four axes were input
as independent variables into an Ordinary Least Square (OLS) model to test the significance of
the link between habitat characteristics and diversity of the dragonfly community. Threats to
the odonate fauna of the freshwater swamp forest are identified and conservation management
measures are discussed.
Keywords. Community structure, ecology, Odonata, statistical analysis
Introduction
Nee Soon freshwater swamp forest (Fig. 1) is the only primary freshwater swamp
forest left in Singapore and is a critical refuge for a large number of Red Listed plant
and animal species threatened with national extirpation (Ng & Lim, 1992; Clews et al.,
2018). It represents the remaining section of a larger freshwater swamp forest that once
existed between Upper Seletar Reservoir and Mandai Road (Corner, 1978; Turner,
1996) and may now be threatened by anthropogenic activities in the surrounding
areas. The streams in Nee Soon catchment are shaded and have shallow waters flowing
over clay, sand, or mud. Their pH is more acidic than other forest streams due to
the decomposition of accumulated leaf litter and woody debris (Ng & Lim, 1992).
124
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Additionally, the soil in Nee Soon freshwater swamp forest is anaerobic and unstable
due to periodic flooding (Nguyen et al., 2018). All these factors have resulted in a
diverse flora and fauna that are adapted to these unusual conditions. However, it also
means that these habitat specialists are sensitive to environmental disturbances such as
changes in the drainage system, introduction of non-indigenous species, soil erosion,
and disturbances from development (Ng & Lim, 1992; Yeo & Chia, 2010).
Odonates (Order: Odonata) are of increasing applied research interest as
potential biological indicators and tools for ecological modelling (Bried & Samways,
2015). They are relatively well known taxonomically and the adults are easy to identify
(Simaika & Samways, 2012; Kutcher & Bried, 2014). Hence they are good ecological
indicators for the assessment of aquatic environments, especially for wetland and stream
quality as they occupy the interface between aquatic and terrestrial ecosystems and are
highly sensitive to environmental changes (Carvalho et al., 2013; Monteiro-Junior et
al., 2013; Oliveira-Junior et al., 2015). There is also a major division within the order
in terms of the ecophysiological requirements of different species (De Marco et al.,
2015). Members of the sub-order Anisoptera (dragonflies) are more useful indicators
of degraded environments because they have more efficient homeostatic mechanisms
and are more mobile, enabling them to tolerate a wider range of environmental
conditions. By contrast, members of the sub-order Zygoptera (damselflies) tend to
provide a more useful role as indicators of more preserved environments and higher
levels of environmental heterogeneity because of their smaller body sizes, home
ranges and greater ecophysiological restrictions (Oliveira-Junior et al., 2015). The
structure of odonate communities can shift predictably in response to changes in local
enviromuental conditions (Corbet, 1999; Juen et al., 2007; Juen & De Marco, 2011;
Luke et al., 2017).
To date most studies of odonates in Singapore have been taxonomic accounts.
There is very little available data on their distribution and abundance in Nee Soon
freshwater swamp forest. Murphy (1997) reported Odonata biodiversity in the nature
reserves of Singapore by comprehensively reviewing the historical account and
providing a list of species found in Singapore. He briefly discussed the distribution and
habitat preferences for some of species and, from his list, only 15 species were specified
as occurring within the location of Nee Soon, of which eight species were confined to
Nee Soon freshwater swamp forest (other widely distributed species might have been
found there without being mentioned specifically) Norma-Rashid et al. (2008) updated
the list of dragonflies in Singapore, and identified 35 more species from Nee Soon.
Tang et al. (2010) further updated the list, with 17 more species occurring at Nee Soon.
Cheong et al. (2009) and Dow & Ngiam (2011) added two more species for Nee Soon.
Munirah (2013) conducted a biodiversity assessment of Nee Soon freshwater swamp
forest aquatic insects and recorded 15 species of larvae from eight odonate families.
In this study, we aim to establish the most up-to-date knowledge on odonate
diversity for Nee Soon freshwater swamp forest with an emphasis on abundance.
The study may be read in conjunction with the sampling of aquatic insects by Su
(2016) and aquatic macroinvertebrates (including odonate larvae) by Ho et al. (2018)
concurrent with our study. We also investigate any distinct spatial distribution within
Dragonflies of Nee Soon swamp forest
125
Lower 3
Mid2
Mid3
C D
Upperl
Upper2
Lower Peirce Reservoir
Upper Seletar Reservoir
Upper Petrce Re
0 0,1 S 0.3 0.6 Kilometers
l LJ I i 1 1 I t
Legend
- Street
Stream
Nee Soon forest boundary
Catchment boundary
Fig. 1. Sub-catchments and drainage map with spatial survey sites.
the freshwater swamp forest; and the environmental variables that may influence such
patterns. From the results, we attempt to identify threats to odonates in Nee Soon and
recommend conservation measures.
126
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Methodology
Quantitative environmental and spatial survey
A total of 36 sites were surveyed across all eight sub-catchments of the Nee Soon
catchment (see Nguyen et al., 2018; Sun et al., 2018), together with a separate stream
system that flowed into Lower Peirce Reservoir representing a comparable sub¬
catchment adjacent to Nee Soon freshwater swamp forest. For each sub-catchment,
four sampling stations were selected at intervals from lower stream to upper stream.
The locations were chosen based on how representative of available habitats each was
in that sub-catchment, together with considerations of accessibility. Study sites were
named and ordered alphabetically according to the location of the site within the sub¬
catchments of Nee Soon, starting from the lower to the upper reaches.
Surveys were conducted from December 2014 to April 2016. Hydrological,
physiochemical and habitat characteristics were recorded to analyse environmental
factors that may determine the associated odonate community. Samplings of odonates
were conducted between 10:30 and 13:00 h mostly during sunny weather conditions.
For each location, sightings were recorded over 30 minutes along a 20 m transect of
the stream and 3 m of riparian zone on either side of the stream bank. The transect was
surveyed at a slow pace along the stream channel, counting every individual either
perched or in flight. Species that could not be identified with certainty by sight were
caught with an insect net and released after identification. Occasionally, unidentified
specimens were collected for further examination.
To further investigate the habitat characteristics of the odonate fauna, which has
a life cycle involving an aquatic larval stage and a terrestrial adult stage, a number of
physical and biological characteristics of the stream were quantified. Stream parameters
were measured using measuring tape, a multiparameter and a flow meter. Hydraulic
parameters measured included stream dimension (depth, width), proportion of pools
and riffles, in-stream woody debris, macrophytes, leaf litter, and substrate (sand, silt,
clay and rock). Physiochemical parameters recorded included pH, Dissolved Oxygen
(DO), Oxidation-Reduction Potential (ORP), Total Dissolved Solids (TDS), Salinity
(S), and Temperature (T). Riparian vegetation heterogeneity, bank form, and canopy
cover were also assessed and recorded for all quantitative survey sites. The stream
cross-section of each site was measured multiple times to obtain the average values.
The surrounding habitat types, i.e. swamp, open canopy and distance to forest edges
were also recorded (see appendix 2 for details).
Identifications of adults were primarily based on Tang et al. (2010) and Orr
(2005). Taxonomic classification follows Schorr & Paulson (2017). Local species
updates and conservation status follow Ngiam & Cheong (2016).
Odonate larvae were caught and identified to the family level. Specimens
were caught by tray net sampling, moving upstream and disturbing the water. These
specimens were caught together with other aquatic insect specimens and preserved in
75% ethanol for microscopic examination. These collections were separate from and
supplementary to those of Ho et al. (2018).
Dragonflies of Nee Soon swamp forest
127
Qualitative surveys
Additional species not otherwise recorded during the quantitative surveys were
included to compile a full species list. Species spotted during reconnaissance trips were
also added to provide a comprehensive updated inventory for Nee Soon freshwater
swamp forest.
Data analysis
All data were analysed using Palaeontological Statistics (PAST 3.15) (Hammer, 2017).
Three of the 36 sites were not included in the analysis due to incomplete environmental
data. Habitat parameter data were transformed either by square root or by log (x+1).
The community count data was log (x+1) transformed before analysis.
The Shannon-Weaver Index (H’) was used as a measurement of species diversity.
The index is calculated as , where p. is the proportion of individuals found of species
i and n is the total number of species. Species Richness (R) refers to the number of
species found at each site.
Ward’s hierarchical clustering using Bray-Curtis dissimilarities was carried out
according to abundance and species composition of the odonate assemblage to examine
if there were any natural groupings in the data. The results of the cluster analysis were
overlaid onto other multivariate analysis plots to indicate graphical representation of
groupings in the community. The significance of the groupings was then tested using
Non-parametric Multivariate Analysis of Variance (PERMANOVA).
Detrended correspondence analysis (DCA) was carried out to show the
distribution of odonate communities across all sites. DCA is a multivariate statistical
technique widely used by ecologists to find the main factors or gradients in the
large, species-rich but usually sparse data matrices that typify ecological community
data. Canonical correspondence analysis (CCA) (Legendre & Legendre, 1998) was
performed to determine correlations between odonata abundance and environmental
parameters. CCA is correspondence analysis of a site/species matrix where each site has
given values of environmental variables. The ordination axes are linear combinations
of the environmental variables. CCA is thus an example of direct gradient analysis,
where the gradient in environmental variables is known a priori and the species
abundances are considered to be a response to this gradient (Hammer, 2017). The CCA
model significance was tested using a Monte Carlo permutation test (1000 iterations).
Environmental variables were filtered through a Multivariate Liner Regression model
where each individual environmental variable was treated as one independent variable
and the raw scope of the two DCA Axes (DCA 1 & DCA 2) were loaded as dependent
variables. Results of the regression indicate how relevant the given environmental
variable is in driving the overall gradient order of site and species. Selection of the
variables was then based on the statistical results of R square and p value (F-test).
Principal components analysis (PCA) was applied to analysis of the 23
environmental variables. The results were used to summarise environmental
conditions. The significant axes were determined by eigenvalues which are expected
to be above a random model (Broken Stick) curve (Jackson, 1993; Hammer, 2017).
These significant axes (in this case the first four, which explained 63% of the variation
128
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
in the environmental data) were retained and analysed. The four axes were then input
as independent variables into an Ordinary Least Square (OLS) model to test the
significance of the link between habitat characteristics and diversity of the odonate
community, represented by Shannon-Weaver Index (IT) and Species Richness (R).
Spatial autocorrelation and homoscedasticity of the residuals were investigated with
a Durbin-Watson test and Breusch-Pagan test respectively. The significant PCA
axes were then further examined to understand the major environmental variables
influencing the community diversity.
Results
Abundance and diversity of odonates in Nee Soon
A total of 1706 odonate specimens were sampled, comprising 49 species of 34 genera
in 11 families. The suborder Zygotera was represented by 1014 individuals, distributed
in seven families (Argiolestidae, Calopterygidae, Chlorocyphidae Coenagrionidae,
Euphaeidae, Platycnemididae, and Platystictidae), 14 genera and 22 species. Suborder
Anisoptera contributed 702 individuals, distributed in four families (Aeshnidae,
Gomphidae, Libellulidae and Macromiidae), 20 genera and 27 species. Prodasineura
was the most abundant Zygopteran genus (n=552 specimens) followed by Pseudagrion
(n=lll). Neurothemis was the most abundant Anisopteran genus (n=334) followed by
Orthetrum (n=138). Relative species richness per survey site varied from two to 13,
mean abundance of individuals recorded per survey site varied from three to 106. An
updated species list of Nee Soon freshwater swamp forest odonates, with 68 species of
47 genera in 11 families could be found in appendix 1.
1. Distribution of odonates within Nee Soon freshwater swamp forest
The species found in each of the sub-catchments are summarised in Table 1. The most
widely distributed species was Prodasineura notostigma which was present at 21
of the 32 sites, distributed across all the eight sub-catchment regions. Prodasineura
interrupta and Orthetrum chrysis were found at seven of the eight sub-catchments in
Nee Soon. Fourteen species were only found from one sub-catchment viz, Libellago
lineata (LP), Podolestes orientalis (L2), Argiocnemis rubescens (L2), Pseudagrion
australasiae (L2), Pseudagrion pruinosum (LI), Anax guttatus (LP), Macrogomphus
quadratus (Ml), Acisoma panorpoides (LI), Crocothemis servilia (LI), Nannophya
pygmaea (L3), Orchithemis pulcherrima (LI), Pseudothemis jorina (L2) Rhyothemis
obsolescens (L2), and Trithemis festiva (LI).
2. Spatial variation in odonate abundance, species richness and diversity
The species richness and abundance of odonates in each sub-catchment were summed
up from the four sampling stations and the total results are presented in Fig. 2. Species
richness is further analysed by the average value of the four sampling sites within
each sub-catchment (Fig. 3). Among the eight sub-catchments, the three middle sub¬
catchments all showed low abundance and species richness whereas the three lower
Dragonflies of Nee Soon swamp forest
129
o
£
in
3
o
o
o
3
2
+
in
o
o
U
+
+
A,
c,
o
o
U
b
2
a
o
o
C/2
CD
CD
Z
C/2
CD
03
a
O
to
o
P
O
CD
>
u
3
C/3
CD
>
pH
c3
3
O
ct 1
• t-H
+->
CD
3
P
-O
■i—>
• ^H
S—(
3
• rH
GO
• i-H
TO
TO
3
TO
3
o
3
4h
od
C/3
CD
CD
O
3
od
Ti
• t-H
O
CD
CD
§ *
P> ^
< o
3
e2
a
CD
C/3
2
Oh
Qh
P
P
m
<N
m
P
<N
P
a>
£
S3
S3
S3
O
£
£
o
U
vi
.£
’3
a>
a.
c n
+
ea
c
• »-H
£
td
P
-a
<D
t3
O
Oh
C/3
I
CD
J3
3
+ +
+
+
+
+
+
+
+
+
+
+
bfl
e
£
C/3
jd
P
bfl
3
3
g
too
c
• T—I
£
P
C/3
cd
P
S3
O
O
U
+
+
+
+
+
+
+
+
+
+
+
+ ' +
+ '' +
+
+
+
+
+
+
+
+
+
+
+
+
+
CD
a>
cd
T3
”-£3
C/3
JD
"o
"So
u
<
•2
^3
£
•S2
£
£
•S
.£
£
.£
£
£
sp
CD
3
3
£
£
£
Co
a>
3
-3
£
£
2
2
£
£
4>
3
3
* Kni
£
5
oo
co
o
nr
£
1£
£
.£
”3
.Co
• l*ni
co
wAJ
5
a
£
~£
.o
jg
S:
U
£
£
p
£
£
£
*3
Co
CD
a
.Co
.Co
---
u
2
bo
£
bo
£
bo
£
SJD
3
K
p.
_o
2
2
_o
£
£
£
3
CD
.O
S!
P
£
13
Co
Co
p
~£
P
P
O
^0
g
p
U
P
U
p
P
P
U
TC
CN
m
in
£2
oo
£
.£
.Co
>1
£
P
2
On
CO
£
CD
£
co
CD
P
£
£
CO
CD
£
CD
.£
CD
P
£
£
£
.£
£
CD
CD
£
.g
£
bo
.2
£
8
+
+
+
i i i i i i i i
+
+ '++'++'
+
+ +
+
3
<D
Qh
<D
03
•—
CD
3
CD
a
3
(D
O
3
(D
a
SJD
3
• i-H
CO
s
2
3
Qh
CO
• rH
£
1
o
2
• rH
c-<
CD
C/2
D
•a
p
c n
2
3
13
Vh
o
U
o
1
C/2
<D
M
»
Id
i
aj
PH
• t-H
Sh
P
cn
b
CD
• »-H
P
2
’o
a
c3
CD
u
3
• rH
1
1
1
2
o
>
3
• t-H
1
ctf
Q
u
o
M
O
o
p
CD
J3
S
CD
•2
Co
£
'-■rA
Co
£
£
£
•2
£
bo
-§
£
CD
Co
p
CN
CD
•a
p
C/2
<D
&
£
P
P
£
CD
2
£
£
.£
£
bo
"§
£
£
co
P
CO
£
Co
£
.£
£
£
P
£
•2
£
bo
£
£
Co
p
CD
03
3
*3
S3
JS
a
3
P
Table 1. Continuation.
130
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Ph
P
P
m
m
P
P
+
+
+
+
+
+
+
+
, +
+
I I I I
I I I I
+
+
+
+
+
+
+ +
' +
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
, , , , + +
+
+
+ ' +
+ +
+
+
+ _L +
+ + +
+
+
+
+
+ +
+
+
+
+
+
+
+
+
+
' +
' +
+
+
+
I I I
I I I
I I I
+
I I I
+ +
+
' +
' +
+
60
u
PQ
c
cd
_>
00
<D
C
o
Si
&
<D
E§
VI
60
<D
Th
0)
si
cd
<u
P
£
o
£
cd
"O
cd
0)
H
-a
I
o
U
<u
C/3
cd
-a
£
o
T3
cd
si
00
<D
5h
o
&
60
c
00
60 cd
Sh
<u
C
cd
p
1—
M
£
cd
P
C
O
O
DC
• l-H
£
-4—>
c
CL)
Oh
DC
p
G
o
£
g
'cd
-a
0)
T3
aj
s
e
03
o
•—
o
C
cd
3
S-H
w
o
u
p
PD
H
29 Brachydiplax chalybea Blue Dasher + +++
Dragonflies of Nee Soon swamp forest
131
a
o
• rH
S-^
c3
p
c
• rH
S-^
a
o
U
3
e2
+
+ +
k
<d
D4
00
k
C/3
<D
k
o
CD
-a
<d
a
a
+
+
1
Q
JD
cd
cd
00
d
o
o
U
a
3
s
so
>3
2
o
o
m
3
c-
k
3
=0
SO
3
■s
3
3
g
U
m
3
.3
•2
s
•2
*G
g
■s
a
(N
cn
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
a
3
So
a
r'
£
3
3
:§
cn
m
+
+
u
aj
• rH
"O
o
CO
cd
'ao
C
• rH
cd
C
s
c3
■HH
d
CD
oa
CD
<D
k
a
CD
d
o
00
s
rk
3
cd
c3
S
cd
• rH
k
CD
CD
o
^cd
00
00
U
>
<o
d
• rH
S—>
d
<0
00
<0
J3
s
So
R
a
s
"hi
3
SO
3
|
^r
m
3
k
ik
3
R
3
'■’■ki
3
Cx,
so
3
R
R
g
o
in
m
+
+
+
+
+
+
So
R
.3
s
k
CD
So
3
R
R
g
o
VO
m
+
+
+
k
CD
M
00
-a
<o
<d
d
• rH
CD
00
k
CD
D4
00
(D
PP
d
o
o
O
•2
So
C
R
3
3
k
■Hh
^ 5^
O O
+
+
+
+
+
+
+
+
+
I I I I I I I I
+ ' ' + +
I I I I I I I I
I I I I I I I I
I I I I I I I +
+
+
+ + + +
+
+
+
+
+
+
+
+
+
+
+
+
+
k
<D
M
oo
d
<D
e
a
-a
CD
S—>
c3
W)
<D
• rH
I
$-H
3
00
jiD
cd
CD
00
k
<d
M
00
-a
(D
T3
d
cd
CQ
o
-Q
13
<D
Pi
e
o
o
U
k
(D
k
a
-k
E
CD
N
fi
O
k
PP
k
e
(D
■k
k
-a
CD
s
PP
I
£
o
$
U
<D
<u
-HH
-HH
E
aj
^H
• rH
&
00
bJO
e
• rH
CD
O
k
Q
e
o
co
k
o
bo
c
• rH
CD
O
k
Q
o
bJD
• rH
"O
c
k
(D
bfl
bfl
3
X3
CD
CD
£
aj
C/5
• rH
p
k
u
-a
c
o
CD
3
Cj
3
3
bO
3
k
■k
k>
R
r-
<n
oo
m
3
.3
R
3
So
3
k
■k
<D
■s
O
On
m
3
3
Cj
s
g
■s
CD
■s
*
o
'd -
3
.3
k
3
So
3
■s
"3
3
3
>5
&H
k
k
3
■s
"3
g
PC
<N
^r
So
3
3
Cj
So
3
3
So
R
3
so
3
-s
3
R 1
PC
on
^r
•2
, *^kj
R 1
Pd
so
3
■s
3
R 1
pc
^r
H"
•2
’k
-S
3
bo
3
.3
k
k
So
3
■s
3
R'
Pi
in
H"
g
3
bo
5
g
3
3
3
■s
£
VO
H"
^3
R
k
_k
'd
k
s
3
a>
cd
3
.3
R
•2
Ski
S'
R
1
2
3
■s
R
.3
CD
Ski
k
cd
o
s
s
r-~
H-
oo
■ct
On
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
132
Lower 1 Lower Z Lower 3 Mid 1 Mid 2 Mid 3 Upper 1 Upper 2 LPR
Subcatchment
■■Total species —-Total Individuals
Fig. 2. Species richness and abundance graph of odonates at various sub-catchments.
sub-catchments had high abundance and species richness. The comparative site in
Lower Peirce sub-catchment had the highest abundance and relatively high species
richness. Mean species richness ranged from 3.5 to 9.25 species/site with the highest
mean species richness at Lower 3 and the lowest at Mid 1.
The mean species diversity at each sub-catchment was plotted using Shannon-
Wiener Index (H’) (Fig. 4). Lower 3, with H’ value of 0.88, had the greatest species
diversity among the sub-catchments, while Mid 1, with H’ value of 0.46, had the
lowest species diversity.
Odonate larvae
Larvae were collected together with aquatic fauna surveys. The specimens were
identified to family level. Table 2 shows the mean abundance of larvae collected from
each survey site. The most abundant family was that of Libellulidae, with a total of 33
identified at six of the eight sites surveyed.
Distribution pattern and habitat characteristics
Hierarchical clustering (Fig. 5) and DC A (Fig. 6) indicated that three main groupings
of sites existed, each with a distinct community of associated species. Based on one¬
way PERMANOVAtest, all groups different significantly from each other (p=0.0001).
Further analysis by CCA (Fig. 7, Table 4) with 12 significant environmental variables
which were selected based on a multivariate liner regression model test (Table
3). Results showed that these groups were significantly associated with several
environmental variables. The first group of sites are mostly located at the outskirts
of the Nee Soon freshwater swamp forest, including Lower 1 A, B, C; Lower 2 A, B;
Lower 3 D and LPS. Species representative to this group were mainly Neurothemis
fluctuans, Acisoma parnorpoides, Trithemis festiva, Cratilla metallica, Pseudagrion
Dragonflies of Nee Soon swamp forest
133
12
10
1/1 o
sfl O
u
c
u
£ 6
0J
tJ A
oi 4
D,
<A
" 7,75
9,25
7,00
{
3.50
| 4,25
'■ 4.00
| 5,00 <S ' 2S
7.00
2
0 --
Lower 1 Lower 2 Lower 3 Midi Mid 2 Mid 3 Upper l Upper 2 LPft
Subcatchment
Fig. 3. Mean species richness graph of odonates at various sub-catchments (N=4, Error Bar=±l
SE).
1.2
x
41
-o
g
in
k_
41
>
(A
Q>
W
41
a.
LO
1
as
0,6
0.4
0.2
0
v o,88
-■ 0.64 <■ 0.64
| 0.46
■■ 0,56
I*
.60
SO
0.60
0.65
Lower 1 Lower 2 Lower 3 Midi Mid 2 Mid 3 Upper 1 Upper 2 LPR
Subcatchment
Fig. 4. Mean species diversity index graph of odonates at various sub-catchments (N=4, Error
Bar=±l SE).
microcephalum, Ceriagrion cerinorubellum, Trithemis aurora, Ictinogomphus
decoratus, and Prodasineura humeralis. Most of these species are open canopy
species. Environmental variables that best described this group are locations at or
close to forest edges (-FE) or/and open area (-OA), with less riparian canopy cover
(-RCA), stream water with high temperature (Tern), high pH, and low ORP, stream
with deep water channel. The second group of sites are mostly located along the main
stream (stem channel) at the central or middle part of the Nee Soon catchment. Species
134
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
s ig is :? is S-g'aB
s S' S =
iliSIJlIllIlIlIllllI^IJ!
Fig. 5. Hierarchical clustering of sites according to odonate assemblage. Three clusters were
recognised, colour coded in blue, purple and black.
Table 2. Species composition and mean abundance of odonate larvae at each site in Nee Soon:
+++ Very Common (> 5); ++ Common (1-5); + Uncommon (< 1); - Not present (0);
Family / Species
LowlC
Low2B
Site
Low3C Low3D
Mid2A
Mid3C
UplC
Up2C
Zygoptera
Calopterygidae
-
+
-
-
-
-
-
-
Chlorocyphidae
-
+
-
-
-
-
-
+
Coenagrionidae
+++
+
+
+
-
-
-
-
Euphaeidae
-
-
-
-
-
-
-
-
Platycnemididae
+
-
-
-
-
-
-
-
Platystictidae
-
-
-
-
-
-
-
-
Protoneuridae
+
-
-
-
-
-
-
-
Anisoptera
Aeshnidae
+
+
-
+
+
-
+
-
Gomphidae
-
-
+
-
++
+
+
+
Libellulidae
+++
-
+
+++
+
-
+
+
Dragonflies of Nee Soon swamp forest
135
A^s I
Fig. 6. Detrended correspondence analysis (DCA) results showing the distribution of odonate
communities across all sites. Sites are grouped into group 1 in blue, group 2 in purple and
group 3 in black corresponding to results of cluster analysis. (Please refer to Appendage 2 for
abbreviation and sites and species).
Fig. 7. Canonical correspondence analysis (CCA) results showing correlations between
odonate abundance and the environmental parameters. Sites are grouped into group 1 in blue,
group 2 in purple and group 3 in black corresponding to results of cluster analysis. (Please refer
to Appendage 2 for abbreviation and sites and species).
136
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 3. A summary of the multivariate linear regression model performed on environmental
variables at Nee Soon against the raw scope of the first two DC A axes. Listed here only the
significant environmental variables.
RCA
FE
OA
TEM
ORP
DO
PH
DEP
SI
BSG
BRO
MPW
R squares
0.312
0.321
0.280
0.222
0.159
0.120
0.123
0.093
0.063
0.063
0.055
0.054
p-value
0.000
0.000
0.000
0.000
0.006
0.018
0.020
0.041
0.128
0.169
0.210
0.224
Table 4. A summary of eigenvalues, percentage (%) of total variation represented and p value
of permutation test for the first four axes resulting from the CCA for odonates at Nee Soon
freshwater swamp forest.
Axis
Eigenvalue
Percentage %
Permutation (p )
1
0.6294
26.24
0.003
2
0.4757
19.83
0.018
3
0.3258
13.58
0.091
4
0.2211
9.216
0.315
association included Euphaea impar, Prodasineura interrupta, Libellago aurantiaca,
Vestalis amethystina, Vestalis amoena and Amphicnemis gracilis. All these are typical
forest damselfly species. Environmental variables that best described this group were
stream water with high DO, stream channel with high amount of in-stream macrophytes
and woody debris, stream substrate with high proportion of silt, and with low water
temperature and high ORP. The third group of sites are mostly found at headwater
areas of various stream branches; dominant species in this group are Drepanosticta
quadrata, Prodasineura notostigma, Orchithemis pruinans, Prodasineura collaris,
Libellago hyalina, Copera marginipes. Environmental variables associated with
this group were locations away from forest edges and open areas, with high riparian
canopy cover, stream banks with significant adventitious roots into water, shallow
water depth with low pH. The ordination points for this group of sites were spread out,
with small areas of overlap. The majority of survey sites were in the second and third
group, which covers most of the Nee Soon catchment. In terms of species composition
and associated environmental variables, the three groups were very distinct from each
other.
Principal components analysis (PCA) was performed to analysis the 23
environmental variables (Fig.8. The results were used to summarise environmental
conditions for Nee Soon odonates. The significant axes were determined by percentage
of eigenvalues which are expected to be above a random model (Broken Stick)
curve (Fig. 9). The first four principal components of the PCA explained 63% of the
variation in all the environmental variables (Table 5). PC 1 explained the 26% of the
total environmental variation, was positively correlated to distance from forest edges
Dragonflies of Nee Soon swamp forest
137
Fig. 8. Principal components analysis (PCA) results showing the loading of the 23 environmental
variables to the first two PC axes. Square (blue): sites in group 1; Triangle (purple): sites in
group 2, round (black): sites in group 3.
(FE), riparian canopy cover (RCA), distance to open area (OA), and stream water with
high value of ORP, but negatively correlated to water depth, water with high pH and
stream with steep bank shape. PC 2 explained 17% of the total environmental variation,
and was positively correlated with water DO, stream velocity and stream order, but
negatively correlated with water temperature (Tern) and high levels of leaf litter (LL)
in the substrate. PC 3 explained 11% of the total environmental variation, and was
positively correlated with existence of pools in water channel and stream substrate
with high amount of silt, but negatively correlated with stream substrate with high
amount of sand and stream bank with overhanging tree root. PC 4 explained 9% of the
total environmental variation, and was positively correlated with riparian vegetation
heterogeneity, in-stream macrophytes, woody debris and negatively correlated with
silt substrate.
These four axes were input as independent variables into an Ordinary Least
Square (OLS) model to test the significance of the link between habitat characteristics
and diversity of the odonate community, represented by Shannon-Weaver Index (H’)
and Species Richness (R). Spatial autocorrelation and homoscedasticity of the residuals
were investigated with Durbin-Watson test and Breusch-Pagan test respectively.
OLS results (Table 6, Fig. 10-12) show that only PC 1 and 3 are significantly
correlated with overall odonate diversity in Nee Soon freshwater swamp forest. PC
1 was negatively correlated with both Species Diversity Index and with Species
Richness, while PC3 was positively correlated with only Species Diversity index H’. Its
correlation with Species Richness was rejected due to high spatial autocorrelation. In
summary, distance from forest edge, canopy cover of the riparian vegetation, distance
from nearby open area, shallow water depth, water with high ORP reading, water with
138
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 5. A summary of eigenvectors of each environmental variable, eigenvalues, percentage
(%) of variance explained by the first four axes resulted from the PCA for odonates at Nee Soon
freshwater swamp forest.
Environmental variable
Abbreviation
PC 1
PC 2
PC 3
PC 4
Acidity
PH
-0.277
-0.022
0.050
-0.173
Dissolved oxygen (%)
DO
0.035
0.399
0.124
0.066
Stream velocity
VEL
-0.032
0.373
-0.038
-0.159
Stream width
WID
-0.153
0.134
0.261
-0.235
Stream depth
DEP
-0.288
0.046
0.226
-0.024
Macrophytes and Woody Debris
MPW
0.185
0.052
0.204
0.417
Leaf litter (%)
LL
0.177
-0.275
-0.010
-0.069
Sand (%)
SA
-0.198
0.197
-0.352
0.158
Silt (%)
SI
0.103
-0.070
0.341
-0.402
Bank shape
BSH
-0.276
0.243
0.077
-0.080
Bank with hanging root
BRO
0.157
0.137
-0.330
0.209
Bank covered with shrub and grass
BSG
0.154
0.231
0.174
0.028
Riparian vegetation heterogeneity
RHE
-0.138
-0.001
0.279
0.506
Riparian canopy cover
RCA
0.325
0.211
-0.186
-0.040
Stream order
SO
-0.192
0.314
0.181
-0.027
Distance from forest edge
FE
0.341
0.076
-0.027
-0.018
Distance from nearby open area
OA
0.290
0.148
-0.132
-0.148
Distance from nearby swampy area
SW
0.223
-0.248
0.141
-0.214
Oxidation-Reduction Potential
ORP
0.287
0.163
0.256
0.148
Total Dissolved Solids
TDS
0.008
0.033
0.091
0.220
Salinity
SAL
-0.217
-0.174
-0.131
0.150
Temperature
TEM
-0.092
-0.352
-0.016
0.147
Pool (%)
POL
0.160
-0.118
0.402
0.171
Eigenvalue
6.056
3.823
2.613
1.961
% variance
26.328
16.621
11.363
8.525
low pH value, stream with low angle smooth bank and stream surrounded by swampy
area primarily explained the low diversity and richness of the odonate community in
most of Nee Soon freshwater swamp forest. The presence of pool habitats, substrate
with more silt, less sand, less overhanging root in stream bank, high heterogeneity of
riparian vegetation and the wider channel will partially explain the high diversity level
of odonates in the outskirts and open areas of the Nee Soon freshwater swamp forest.
Dragonflies of Nee Soon swamp forest
139
Fig. 9. Scree plot with a random model (broken stick), indicating that only the first four PC As
are statistically significant.
Table 6. Result of Ordinary Least Square (OLS) test on the significance of the link between
habitat characteristics represented by the four PC axes and diversity of the odonate community
at sampling locations in Nee Soon freshwater swamp forest, represented by Shannon-Weaver
Index (H’) and Species Richness (R).
PCI
PC2
PC3
PC4
H’
R
H’
R
H’
R
H’
R
Permutation p
0.0093
0.0005
0.9271
0.3365
0.0082
0.013
0.2936
0.3787
Durbin-Watson test p
0.32
0.0913
0.3122
0.0995
0.4073
0.0055
0.1725
0.0115
Breusch-Pagan test p
0.649
0.164
0.0642
0.00395
0.2938
0.3069
0.9548
0.6238
140
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
PCI
Fig. 10. Plot results of the ordinary linear square test on environmental variables PCI to species
diversity index (H’), with slope at -0.04, intercept 0.62.
Discussion
Biodiversity baselines for odonates of Nee Soon freshwater swamp forest
Altogether, we recorded 49 species of 34 genera in 11 families for our current study.
Relative species richness per survey site varied from two to 13, and mean abundance
of odonate individuals recorded per survey site varied from three to 106. An updated
species list of Nee Soon dragonflies is provided for future reference, with 67 species
of 47 genera in 11 families, representing about half of all odonates ever recorded in
Singapore. Among the eight sub-catchments, the three mid sub-catchments all show
low abundance and species richness. This is followed by the two upper sub-catchments.
The three low sub-catchments all had high abundance and species richness. Species
richness ranges from 3.5 to 9.25 species/site with the highest species richness at Lower
3 sub-catchment and the lowest at Mid 1 sub-catchment.
Compared to historical records, the change in distribution of Prodasineura is
quite remarkable. Murphy (1997: 343, Fig. 4) plotted records of the three Prodasineura
species and commented that: “none of these have been found outside the CCNR
Dragonflies of Nee Soon swamp forest
141
Fig. 11. Plot results of the ordinary linear square test on environmental variables PCI to species
Richness (R), with slope at -0.79, intercept 5.79.
[Central Catchment Nature Reserve, including Nee Soon freshwater swamp forest]
in Singapore and the genus is not known from Bukit Timah.” “ Prodasineura collaris
is widely scattered and appears associated with still waters choked with leaves.
Prodasineura notostima is common over deeply-shaded open streams. Prodasineura
interrupta occurred together with P. notostigma in the lower part of Nee Soon Swamp
Forest and was widespread in the upper Nee Soon basin where it was the only species
seen. It remains unexplained why this species is, on present evidence, confined to the
Nee Soon catchment, since it is found in riparian galleries not obviously different from
those in other drainage systems. The absence of P. notostigma from the upper Nee
Soon basin is also remarkable.” Two decades later, we found that the genus has spread
beyond the Central Catchment Nature Reserve. Prodasineura notostigma has been
commonly found in forest streams in Bukit Timah and other nature areas in Singapore.
However, statements for Prodasineura collaris and P interrupta remain valid except
that the latter species is no longer the only species found in the upper Nee Soon basin.
Prodasineura notostigma has become the dominant species there. The restriction of
Prodasineura interrupta to Nee Soon freshwater swamp forest is probably related
142
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
PC 3
Fig. 12. Plot results of the ordinary linear square test on environmental variables PC3 to species
diversity index (H’), with slope at 0.06, intercept 0.61.
to the particular environmental variables that it requires either as an adult or larva.
Localised movement behaviour may limit its dispersal capability. This remains a
subject for future investigation.
Prodasineura humeralis was not found in Murphy’s island wide odonate survey
(Murphy, 1997). Murphy et al. (2008: 247), recognising it at the rank of subspecies,
red-listed Prodasineura verticalis humeralis as Critically Endangered (CR) for
Singapore. Tang et al. (2010: 92) stated that it was only first recorded in Singapore in
October 2006, and classified the taxon as uncommon. Lolc (2008) suggested habitat
fragmentation as a possible cause for its limited distribution. Ngiam & Cheong (2016:
161) revised the conservation status of all odonate species found in Singapore and
treated Prodasineura humeralis as being of least concern as it is currently a common and
widespread species. Tang et al (2010: 92) described Prodasineura humeralis as being
“associated with shaded forest streams with fast flowing water.” Observations from the
present study show that the species is abundant in the outskirts and stream stretches
Dragonflies of Nee Soon swamp forest
143
that are associated with an open canopy. It is commonly associated with fast flowing
water, but hardly found in shaded forest streams with high canopy cover. When they
described a new species of freshwater snapping shrimp Potamalpheops amnicus, Yeo
& Ng (1997: 171, 172) commented that “Interestingly, there is a high likelihood that
these shrimps originated from unintentional introductions from Peninsular Malaysia.
This is because the Sedili drainage in Kota Tinggi, Johore, is one of the water sources
for Singapore with water from there being transported via pipeline to the Upper Pierce
Reservoir. The Sedili drainage is where Potamalpheops amnicus is found in far greater
abundance than in Singapore. It is therefore entirely possible for the shrimps to enter
Singapore via the pipeline and become established in certain areas suitable to their
habitat preference like the present stretch of stream.” Prodasineura humeralis has long
been a common and widespread species of forest streams in Peninsular Malaysia (Orr,
2005). There is, therefore, a likelihood of larvae being incidentally transported from
Malaysia via pipeline into Singapore waters. Both its rapid dispersal history, and the
current distribution patterns mainly in open forest streams at the outskirts of the nature
reserves, mirror those of many aquatic introduced species in Singapore (Yeo & Chia,
2010). Its potential impact upon other native damselfly species needs to be closely
monitored.
Community structures and distribution patterns
Odonates of Nee Soon freshwater swamp forest can be grouped into three distinct
community assemblages: 1) headwater, 2) mainstream channel, 3) outskirts and/or lower
stream communities. Each of these community assemblages is linked to a distinctive
suite of riparian vegetation, hydrological and physicochemical characteristics.
Community assembly theory is founded on the premise that the relative
importance of local environmental processes and dispersal shapes the compositional
structure of metacommunities. Four general models describe interesting combinations
of these factors and are frequently used to interpret observed communities: 1) neutral
model, 2) patch dynamics, 3) mass-effect, and 4) species sorting (cf. De Marco et
al., 2015). The species sorting model predicts that assemblages are dominated by the
environmental filtering of species that are readily able to disperse to suitable sites.
De Marco et al. (2015) propose an eco-physiological hypothesis for the mechanism
underlying the organisation of species-sorting odonate metacommunities based on the
interplay of thermoregulation, body size and the degree of sunlight availability in small-
to-medium tropical streams. They considered that narrow streams are more affected by
riparian vegetation than are wide streams, which reduces light input but also generates
a more stable thermal environment. They considered these characteristics favour small
species that are less dependent on exposure to direct sunlight for thermoregulation,
and limit the colonisation of larger, heliothermic species. To quote directly: “The
taxonomic distinction between small streams (dominated by Zygoptera) and larger
rivers (by Anisoptera) is an incidental consequence of differences in body size and
associated thermal properties between these two groups” (De Marco et al., 2015).
The species composition and distribution patterns of odonates observed in Nee Soon
support such a hypothesis. The dominance of small Zygoptera in forested headwaters
144
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
and the increased number of Anisoptera in the lower streams and outskirts of Nee Soon
can be partially explained by species sorting through eco-physiological mediation.
Threats identified and conservation implications
Variations in the hydrology and ecology of the Nee Soon freshwater swamp forest over
the last few decades have made it difficult to determine the conservation actions that are
needed to ensure the long-term sustainability of the ecosystem. Both the forest and the
surface waters of the Nee Soon freshwater swamp forest have changed considerably
over the last two decades, with stream banks experiencing raised water levels in some
parts and drying up in others resulting in a shifting of boundaries between the swampy
and the dryland forests. The changes in hydrological characteristics have altered the
dimensions and profile of stream channels as well as the instream macrophytes, woody
debris and the complexity of the riparian zone in many parts of the drainage system.
The changes pose significant threats to ecological health of the swamp forest, as
well as the aquatic and semi-aquatic life that forms part of this integrated ecosystem.
Recent eco-hydrological modelling conducted by Sun et al. (2018) confirms that
future hydrological conditions of Nee Soon freshwater swamp forest will be further
impacted by global climate change, and they projected 12 scenarios which, according
to the extent of rainfall and operational water level of reservoirs surrounding the
catchment, range from almost total disappearance of surface water to flooding events
that covers most of the swamp forest. The effects of the two inputs (water drawdown
versus flooding) differ by location. A shift in odonate community structure will thus
be inevitable should such hydrological scenarios materialise. As hydrological changes
following climate change may be sudden and drastic, many odonate species may not
be able to find suitable breeding grounds for recruitment and recolonisation even when
conditions have recovered to their original status.
The River Continuum Concept (RCC) (Vannote et al., 1980) is the dominant
concept of how stream ecosystems vary from headwaters downstream to large rivers.
The basic idea is that aquatic communities and ecological processes of the stream
ecosystem change predictably along the downstream gradient of increasing channel
dimensions and canopy opening over the stream. While the River Continuum Concept
is typically viewed as a global stream ecosystem theory, it can be applied to forested
landscapes to depict forest-stream interactions with widening canopy opening over the
stream and shifting geomorphology in the downstream direction. Recognition of the
importance of these linkages between streamside forests and instream communities
has resulted in the creation and protection of riparian buffers as best management
practice in many regions. With a life cycle spanning both aquatic (larvae) and terrestrial
(adult) phases, odonates are more sensitive than purely aquatic or purely terrestrial
invertebrates to the stability of the vegetation at the aquatic and terrestrial interface, be
it as oviposition sites or as perches for both adults and larvae. Changes in distribution
and cover of instream macrophytes, woody debris, as well as aquatic or riparian
vegetation along a river system, either naturally or due to human interference, would
have significant impacts on odonate populations. For example, one of the sampling
sites at Nee Soon, Lower 3D, was at a position along the main stream, adjacent to Mid
Dragonflies of Nee Soon swamp forest
145
2 and Mid 3 sub-catchments. The statistical clustering results indicated that the site’s
odonate community was grouped into group 1, together with most of the sites located
at the Nee Soon outskirts, rather than with those of Mid 2 and Mid 3 sub-catchments.
Close examination shows that the Lower 3D site is right alongside a pipeline service
trail which, although it is near the middle of the freshwater swamp forest, has created
an approximately 10-15 m width gap in the canopy, creating a significant access path
for many heliothermic species penetrating the forest to colonise the site and disturb the
overall RCC distribution pattern. The changes in community assemblage suggest that
the shift in habitats and its long term impact will need to be monitored closely.
There may be more threats posed by the pipeline service trail to the odonate
fauna in Nee Soon. Results from the recent hydrological baseline study of Nee Soon
freshwater swamp forest show that water quality in the stream parallel to the pipeline
has high calcium concentrations (A. D. Ziegler, pers. comm.; Nguyen et al., 2018) as
compared to other sub-catchments. This is probably linked to the construction and
material used for the pipeline. Soil erosion along the pipeline has also been observed to
wash down sediment to the stream during heavy rainfall and occasionally pipe leakage
events. Whether such anthropogenic impacts on water quality have negative effects
on the odonate fauna in Mid 3 sub-catchment (which happened to have low species
richness and low species diversity H’ in our study) should be determined. Additionally,
with the opening of the forest canopy as well as the creation of ephemeral water bodies
along the pipeline, exotic species of odonates may find suitable pathways into the
freshwater swamp forest.
Evaluation of sampling protocol
An updated species list indicated 67 species of odonate have been recorded in the Nee
Soon freshwater swamp forest (Appendix 1), and many species reported previously
were not captured by the current study. These include the majority of rare and
nationally Critically Endangered species so far found in Singapore only within Nee
Soon freshwater swamp forest.
The main focus of the study was on adult odonates, as larvae were collected
as a secondary sampling during the aquatic insect surveys which took place
simultaneously (Su, 2016). When larvae are factored in, the data show the possible
presence of odonates of which the adults were not observed. For example, although
only one adult Tetracanthagyna plagiata was spotted at site 1C, the larvae of this
species (identified as Anisoptera sp.) were found at sites Low 1C, Low 2B, Low 3D,
Mid 2A, and Up 1C. With the combined records, the biodiversity of the survey sites
may be assessed more accurately. Patten et al. (2015), in their study of USA odonates,
pointed out that ecological models for adults were broader geographically and had
a wider, more equitable (higher evenness) balance of contributing environmental
variables (niche dimensions) than did models for larvae, which tended to be more
ecologically specialised. They suggested that surveys of adult dragonflies, which are
relatively easy to conduct because of organised efforts to encourage observations by
citizen scientists, can paint a misleadingly broad picture of a species’ ecological niche.
They recommended that evidence of breeding, especially the presence of tenerals or
146
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
exuviae, be used to outline ecological requirements when questions of conservation or
population monitoring arise. Hence, future studies should dedicate more effort towards
the collection and identification of larvae, including barcoding and environmental
DNA (Kutty et al., 2018), to complement the adult surveys thus presenting a more
representative odonate community in Nee Soon freshwater swamp forest.
Conclusion
This study has provided an overview of the odonate diversity within Nee Soon
freshwater swamp forest, as well as the overall condition of the freshwater streams.
Nee Soon holds the only primary freshwater swamp habitat left in Singapore and
much of its biodiversity, including odonates, is severely threatened with extirpation
at the national level. The continued survival of the indigenous freshwater fauna in the
freshwater swamp forest will require conservation actions at local level as part of the
on-going efforts aimed at sustaining forest biodiversity in the densely populated island
of Singapore. These actions will require in-depth studies of the microhabitat-dependent
distribution of odonates as well as studies of their complex interactions with other
species. Future conservation management actions can include stream rehabilitation
and forest enhancement, the reintroduction of native species to their original habitats,
formulation of long-term monitoring programmes, increased and stricter enforcement
on the protection of Nee Soon freshwater swamp forest.
ACKNOWLEDGEMENTS. We would like to thank Drs Lena Chan, Geoffrey Davison and
Liong Shie-Yui for their support, Dr Liong Shie-Yui for the map, Dr Chong Kwek Yan for
discussion on statistics, and Mr Lim Weihao, Ms Li Tianjiao, Mr Su Weijie and colleagues from
the National Biodiversity Centre for their help with field surveys. This study forms part of the
Nee Soon Swamp Forest Biodiversity and Hydrology Baseline Studies Project funded by the
National Parks Board of Singapore (FC 12302501).
References
Bried, J.T. & Samways, M.J. (2015). A review of odonatology in freshwater applied ecology
and conservation science. Freshw. Sci. 34: 1023-1031.
Carvalho, F.G., Pinto, N.S., Oliveira, J.M.B. & Juen, L. (2013). Effects of marginal vegetation
removal on Odonata communities. Acta Limnol. Brasil. 25(1): 10-18.
Cheong, L.F., Tang H.B. & Ngiam, R.W.J. (2009). New records for Singapore dragonflies.
Agrion 13(1): 8-13.
Clews, E., Corlett, R.T., Ho, J.K.I., Koh, C.Y., Liong, S.Y., Memory, A., Ramchunder, S.,
Siow, H.J.M.P., Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W., Theng, M.T.Y. & Yeo, D.C.J.
(2018) The biological, ecological and conservation significance of freshwater swamp
forest in Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
Corbet PS. (1999). Dragonflies: behaviour and ecology of Odonata. Ithaca: Comstock Publ.
Assoc.
Dragonflies of Nee Soon swamp forest
147
Corner, E.J.H. (1978). The freshwater swamp-forest of south Johore and Singapore. Gard. Bull.
Singapore, Suppl. 1. Singapore: Botanic Gardens, Parks & Recreation Department.
De Marco, R, Batista, J.D., Cabette, H.S.R. (2015). Community assembly of adult odonates in
tropical streams: an ecophysiological hypothesis. PLoS ONE 10(4): e0123023.
Dow, R.A. & Ngiam, R.W.J. (2011). Two damselflies new to Singapore: Amphicnemis bebar
Dow, Choong & Ng and Teinobasis cryptica Dow (Odonata: Zygoptera: Coenagrionidae).
Singapore Biodivers. Rec. 4: 393-396.
Hammer, 0. (2017). PAST: Paleontological Statistics, ver. 3., reference manual. https://folk.
uio.no/ohammer/past/past3manual.pdf. Accessed 14 Dec 2017.
Ho, J.K.I., Quek, R.F., Ramchunder, S.J., Memory, A., Theng, M.T.Y., Yeo, D.C.J. & Clews,
E. (2018). Aquatic macroinvertebrate richness, abundance and distribution in the Nee
Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 71-108.
Jackson, D.A. (1993) Stopping rules in principal components analysis: a comparison of
heuristical and statistical approaches. Ecology 74: 2204-2214.
Juen, L. & De Marco, P. (2011). Odonate biodiversity in terra-hrme forest streamlets in Central
Amazonia: on the relative effects of neutral and niche drivers at small geographical
extents. Insect Conserv Divers. 4(4): 265-274.
Juen, L., Cabette, H.S.R. & De Marco, P. (2007). Odonate assemblage structure in relation to
basin and aquatic habitat structure in Pantanal wetlands. Hydrobiologia 579: 125-134.
Kutcher, T.E. & Bried, J.T. (2014). Adult Odonata conservatism as an indicator of freshwater
wetland condition. Ecol. Indie. 38: 31-39.
Kutty, S.N., Wang, W., Ang, Y., Tay, Y.C., Ho, J.K.I. & Meier, R. (2018). Next-generation
identification tools for Nee Soon freshwater swamp forest, Singapore. Gard. Bull.
Singapore 70 (Suppl. 1): 155-173.
Legendre, P. & Legendre, L. (1998). Numerical Ecology, 2 nd English ed. Amsterdam: Elsevier.
Lok, A.F.S.L. (2008). The biology and distribution in Singapore of Prodasineura humeralis
Selys, 1860. Nat. Singapore 4: 95-99.
Luke, S.H., Dow, R.A., Butler, S., Khen, C.V., Aldridge, D.C., Foster, W.A. & Turner, E.C.
(2017). The impacts of habitat disturbance on adult and larval dragonflies (Odonata) in
rainforest streams in Sabah, Malaysian Borneo. Freshwater Biol. 62: 491-506.
Monteiro-Junior, C.S., Couceiro, S.R.M., Hamada, N. & Juen, L. (2013). Effect of vegetation
removal for road building on richness and composition of Odonata communities in
Amazonia, Brazil, hit. J. Odonatol. 16(2): 135-144.
Munirah, A.G. (2013) Biodiversity Assessment of aquatic insects in Nee Soon Swamp Forest
area, Singapore. B.Sc. (Hons) thesis, Nanyang Technological University, Singapore.
Murphy, D.H. (1997). Odonata biodiversity in the Nature Reserves of Singapore. Gard. Bull.
Singapore 49: 333-352.
Murphy, D.H., Cheong, L.F., Wang, L.K. & Ang, S. (2008). Springtails, peripatus and insects.
In: Davison, G.W.H., Ng, P.K.L. & Ho, H.C. (eds) The Singapore Red Data Book.
Threatened Plants & Animals of Singapore, 2 nd ed., pp. 62-104. Singapore: The Nature
Society (Singapore).
Ng, P.K.L. & Lim, K.K.P. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquatic Conserv. 2: 255-266.
Ngiam, R.W.J. (2009). The biology and distribution of Pseudagrion rubriceps rubriceps Selys,
1876 (Odonata: Zygoptera: Coenagrionidae) in Singapore. Nat. Singapore 2: 209-214.
Ngiam, R.W.J. & Cheong, L.F. (2016). The dragonflies of Singapore: An updated checklist
with revised national status. Nat. Singapore 9: 149-163.
148
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Ngiam, R.W.J., Sun, S.W. & Sek, J.Y. (2011). An update on Heliogomphus cf. retrofiexus
Ris, 1912 with notes on Microgomphus chelifer Selys, 1858 in Singapore (Odonata:
Anisoptera: Gomphidae). Nat. Singapore 4: 95- 99.
Nguyen, C.T.T., Wasson, R.J., Ziegler, A.D. (2018). The hydro-geomorphic status of Nee Soon
forest catchment of Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 33-48.
Norma-Rashid, Y., Cheong, L.F., Lua, H.K. & Murphy, D. H. (2008). The dragonflies
(Odonata) of Singapore: Current status records and collections of the Raffles Museum
of Biodiversity Research. Singapore: Raffles Museum of Biodiversity Research,
Singapore.
Oliveira, J.M.B., Shimano, Y., Gardner, T.A., Hughes, R.M., De Marco, R & Juen, L. (2015).
Neotropical dragonflies (Insecta: Odonata) as indicators of ecological condition of small
streams in the eastern Amazon. Austral Ecol. 40(6): 733-744.
Orr, A.G. (2005). Dragonflies of Peninsular Malaysia and Singapore. Kota Kinabalu: Natural
History Publications (Borneo).
Patten, B.M., Bried, J.T. & Smith-Patten, B.D. (2015). Survey data matter: predicted niche of
adult vs breeding Odonata. Freshw. Sci. 34(3): 1114-1122.
Schorr, M. & Paulson, D. (2017). World Odonata List. Slater Museum, University of Puget
Sound. https://www.pugetsound.edu/academics/academic-resources/slater-museum/
biodiversity-resources/dragonflies/world-odonata-list2. Accessed Sep. 2017.
Simaika, J.P. & Sam ways, M.J. (2012). Using dragonflies to monitor and prioritize lotic
systems: a South African perspective. Org. Divers. Evol. 12(3): 251-259
Su, W.J. (2016). Biodiversity assessment of aquatic insects in Nee Soon swamp forest streams,
Singapore. B.Sc. thesis, Nanyang Technological University, Singapore.
Sun, Y., Kim, D.Y., Wendi, D., Doan, D.C., Raghavan, S.V., Jiang, Z. & Liong, S.Y. (2018).
Projected impacts of climate change on stream flow and groundwater of Nee Soon
freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 175-190.
Tang, H.B., Wang, L.K. & Hamalainen, M. (2010). A photographic guide to the Dragonflies of
Singapore. Singapore: Raffles Museum of Biodiversity Research, Singapore.
Turner, I. M. (1996). Species loss in fragments of tropical rain forest: A review of evidence. J.
Appl. Ecol. 33: 200-209.
Vannote, R.L., Minshall, G.W., Cummings, K.W., Sedell, J.R. & Cushing, C.E. (1980). The
river continuum concept. Can. J. Fish. Aquat. Sci. 37: 130-137.
Yeo, D.C.J. & Chia, C.S.W. (2010). Introduced species in Singapore: An overview. COSMOS
6: 23-37.
Yeo, D.C.J. & Ng, P.K.L. (1997). The alpheid shrimp genus Potamalpheops Powell, 1979
(Crustacea: Decapoda: Caridea: Alpheidae) from Southeast Asia, with descriptions of
three new species. J. Nat. Hist. 31: 163-190.
Appendix 1 . Species checklist of dragonflies in Nee Soon freshwater swamp forest (number under the author column indicates the page number
on which the species was indicated)
Dragonflies of Nee Soon swamp forest
149
Cm SO
O
fN
oi
a
o
QJ
-G
u
68
>
S 3
u H
u ©
t/3
u
QJ
-C
a>
gf>©
S 3 ^
G O
H ot
—
CZ3
C8 00
©
. ©
0 <N
s j
£ a
z ts
-C
a
Sh l>
0 OS
£2
o
U
IK
.4>
'0
V
a
m
E
cs
Uh
D
>
Z
w
Oh
u
Oh
U
D
>
U
-J
D
>
Z
w
u
-J
u
-J
u
-J
D
>
u
-J
u
-J
3 p
> >
Pi
u
u
D
Pi
u
Pi
u
CN
g <N >=y
co On ^
<n
. ON
m
oo;
o oo<
ZmOZ<n
m
m
m
r~-
<n
oo
z
<*s
£
o
Q
■'t
ON
c*i
o
(N
NO
ON
o
m
oc
'fl-
'fl-
'fl-
m
co
CO
o
1>
1>
1—1
CO
CO
CO
T fl-
CO
CO
CO
CO
00
a
to
■a
o
a.
c/3
<d
3
W)
G
PD
C
O
o
u
W)
G
Ph
W)
G
u
s
CD
CD
*Vh
CD
CD
*53
3
3
CD
s
<u
a
a
Gfl
G
S
(D
a
G
CD
(Z3
JJ
3
CD
C/3
Vi
CD
C/3
’sf
o
CO
TD
(D
C/3
O
G
CD
•c
CD
CO
•a
CD
CO
Si
3
3
Vh
o
U
qj
73
CO
(D
.3
3
qj
3
CD
co
qj
•c
CD
co
p
c3
2
c3
ctf
qj
CD
c3
3
cd
S
CD
o
CD
o
’J
X3
qj
C
3
*C3
J0
G
Vi
>>
Vi
o
c
P
CD
Vi
E
U
a
>
PQ
3
>
O
Q
hJ
3
a
•a
CD
co
T3
(D
O
.0
C+H
OJ
W)
c
c3
CD
73
o3
CD
u
W>
G
.s
c3
CO
70
a>
70
CD
3
o
C 2
a>
c3
70
O
’5b
.g
03
£
CO
p
CD
c3
*5b
CD
O
Cj
•2
ss
•C
CD
ss
si
GS
si
si
Q
.si
"■~«i
3
E
CD
O
V.
S
•S
s
Q
-si
"~~1
Co
C3
O
O
o
s
GO
GO
Go
•2
ss
ss
si
"iii
*-«i
Cj
■2
•2
Co
-C3
-Ci
GO
^3
3
^3
JD
Q
-Ci
Co
CD
c
.<D
-S
&
O
g
Go
CD
Si
.<J
C3
g
Cj
o
Si
Q
Co
<3
2
O
Co
S3
s
2
Co
S3
CD
<J
Co
CD
3
D.
CD
Si
Cj
•2
C!
£?
CD
-CJ
3
D.
o
•S
O
Si
•2
■2
C!
3
cj
CD
si
•2
£
g
Co
S
Q
S
•2
GO
•3
s
CD
^3
Cj
g
Cj
si
•2
GO
s
CD
S
co
O
•S
2
si
•2
GO
•§
s
CD
Co
%
•2
-S3
S
D.
si
•2
GO
•§
s
CD
GS
CJ
i
CD
Co
C3
O
Si
CD
c3
TD
CD
O
o
Vi
o
2
U
(D
cd
72
'g
o
•a
W)
cd
C
(D
O
u
C3
2h
C3
CD
C3
-■si
fel
CD
c3
32
’a>
c3
-G
CD
0
w
Platycnemididae Coeliccia albicauda White-tailed Sylvan
150
Card. Bull. Singapore 70 (Suppl. 1) 2018
a
o
• i-H
C3
P
$3
• rH
a
o
U
-3
S3
o
&
a
VC
O
<N
oi
3
©
X
U
es
*3 E
>
Sh ^
Sh "G
tZ3
Sh
a>
X
a>
gf o
3 rH
jS o
H fS
VI
a oo
Cd o
. o
a
£ _•
£ «
Z d
JS
a
Sh I
3
S
o
U
58
‘0
V
a
V3
«
&H
U
D
U
hJ
z
w
u
-J
Pi
u
u
-J
D
>
Pi
u
u
-J
D
>
Pi
u
D
>
X
w
X
w
z
w
u
D
Pi
u
D
>
u
D
u
D
u
D
u
D
K*-* K*-* K*-* K*-* K*-* K*-* K*-*
Cd H
■jPo
Z (N
o
(N
(N
(N
VO
VO
G-
CG
VO
VO
G"
co
ON
H
00
1—H
00
H
OS
o
00
00
ov
G"
CO
G"
CO
G"
CO
G-
CO
CO
co
CO
co
CO
co
CO
co
CO
CO
c
cd
c/5
CD
c
o
X
CD
(D
£
W>
D
cd
CD
Ph
£
o
£
"G
cd
CD
T3
<D
CD
"O
cd
CD
H
-fa
T 3
<D
*cd
T 3
CD
^ ’cd
CD -rt
E
£
H
cd
gj) 3
S
o
X
’o
1 a
C /5
u
O H
3
HH
*G
cd
I
H
s
CD
CD
u
o
G3
cd
X
C/5
o
CD
cd
W)
c
c73
CD
§-
cS
cd
cd
*G
CD
W)
3
cd
E
3
o
6
s
o
U
£
o
Ph
W>
JD
Id
g
E
c/5
c
cd
cd
s
cd
ts
cd
CD
X
C/5
c
x
3
a
*G
(D
>->
cd
E
c/5
<D
*G
cd
3
X)
3
u
-a
CD
cd
E
C/5
X
O
O
E
-a
<D
-a
3
cd
PQ
CD
Oh
s
pp
cd
CD
hP
Sh
CD
%
1
3
cd
’G
p
a
a>
QD
X
cd
E
d
o.
'G 5
<
s
o
CO
C/3
cd
Q
3
S
<D
*G
a>
cd
p
P
W
cd
g
a
H
s
X
s
a
•I
>2
a
Go
g
CJ
a
a
.Sh
.£
£?
a
2
a
5
a
a
a
&
g
a
•2
£?
a
-si
a
S'
o
Sh
a
"Hi
a
a
5
•S
co
-§
2
g
g
2
•S 3
•+H
ss
2
CO
si
CD
si
g
co
S
g
co
a
2
^3
.Co
"Hi
g
Sh
sg,
Co
a
•2
GO
—:
a
a
•2
2
C3
2
3
Sh
Sh
g
•S
53
Co
g
o
Si
53
^3
§
a
-Hi
CD
co
3
o
CD
^3
Co
C3
a
C3-*
co
a
^i
o
S
Co
s
-Hi
CD
CD
Co
a
a
H^
a,
Co
a
-a
Co
a
-a
•H
<D
# si
Sh
3
CD
D.
3
CD
# si
•2
•Si
Co
Qj
-Si
1
-s;
^d
2^
cd
2
o
-a
cd
It
a,
2
o
1
o
‘O
co
Co
si
S5
So
•2
o
^0
o
Si
GO
g
CJ
o
GO
g
§D
2
GO
a
2
GO
a
2
g
&H
g
&H
g
a.
g
Q
-Hi
1
i
1
Cj
§
Sh
a
cq
Sh
a
CQ
Co
g
2
g
a
•2
a
a
a
a
GO
a
-Hi
Co
a
as
g
a
a
•2
a
.a
"Hi
CD
a
a
"a
-a
a
a
-Hi
H
a
Co
Co
a
co
a
a
si
a
a
g
g
-a
2-
5
-a
a
a
•a
a
cd
§
£
a
a
GO
g
Hi.
a
GO
§
Co
a
a
a
.GO
a
a
a
g
a
2
a
.Co
2
•2
a
-g
&
a
•2
a
a
a
-Hi
r-a
'a
a
Go
a.
o
K
C3
CD
C3
§
'—H
"T3
O
S
op
tu
ca
T3
po
cd
CD
cd
-G
O
a
CD
cd
T3
(D
<
CD
cd
TD
CD
X
Appendix 1. Continuation.
Cratilla metallica Dark-tipped Forest
Skimmer
Crocothemis servilia Common Scarlet
Dragonflies of Nee Soon swamp forest
151
u
-J
u
-J
u
-J
u
-J
u
-J
Pi
u
u
-J
u
-J
u
-J
u
-J
u
-J
u
-J
u
-J
u
-J
u
-J
u
-J
u
-J
u
-J
z
w
Di
u
u
-J
u
-J
u
-J
u
-J
K*'*
o
r-
cn
OS
Os
Os
to
c3
C
a
G
o
co
W)
Oh
O
co
g
o
o
u
o
.s
c
CD
CO
CD
25
3
cn
CD
.G
g
(D
CO
X)
c3
>
CO
X
CO
TD
JD
CD
c
’a*
co
•-2
CO
<D
5
a
o
o
U
a
a
2
co
<D
j2
3
5-h
CD
73
G
CD
I &
a I
u g
<D u Gg
U w
Cu c -1
W) C CJ
St:
.a a §
co > co co
<D
;g
o
w)
G
•c
(D
70
C
I
0
0
5-h
E
0
i-H
a
X
CD
U
70
O
a
70
CD
3
5
co
Ol
G
3
E
y
E
<D
70
CD
O
a
CD
N
X
70
G
a
C
O
^0
CD
CD
ctf
0
5-h
rs
cG
PQ
u
PQ
co
m
70
(D
<D
X
O
Oh
a
O
a
a
Cj
2
•S
a
*3
a
a
a
£
&0
a
£
si
si
>}
s:
a
a
'Ki
.«3
s
a
2
a
si
a
.a
s
D.
a
g
3
g
o
D.
a
-a
a
a
CD
a
g
g
O
# >1
2
a
a
.a
a
a
a
a
a
g
03
-a
a
a
a
a
a
K?
s
•S
a
a
g
cn
a
a
a
a
d!
a
^0
•£2
2
a
2
a
2
a
S
a
B
a
2
a
D.
D.
D.
5 ^
5 ^
a
-s
a
a
a
a
a
’a
<3
g
g
g
g
g
Sr
^a
£t
£t
tt
tt
a
a
O
O
O
O
O
a.
a 5
d.
a
g
"a
a
-a
si
a
a
>5
a
—■■o
a
’-Tt
*C>
a
>)
a
g
Q)
£
ce
-£■
o,
.>}
s
a
g
a
ft;
C
a
a
&0
si
a
a
g
a
£
£
•«
"§
-a
—x^
-si
£
w
a
.e
a
y:
a
a
^o
CG
2
W)
G
‘5
Q_
O
G
O
C/3
s
g
W)
c
%
CD
O
O
’S
c
CD
W)
W)
G
X
CD
O
5
£
a
a
g
i£
a
CD
■s
f£
■§
E
o
&
«Q
.O
c
Macromiidae Epophthalmia vittigera Pond Cruiser
152
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Appendix 2. Abbreviation and quantification of environmental variables.
Parameters
Abbreviation
Water parameters pH
PH
Do (%)
DO
Temperature
TEM
ORP
ORP
Mean Velocity
VEL
TDS
TDS
Salinity
SAL
Stream dimension Mean Width
WID
Mean Depth
DEP
Pool (%)
POL
In-stream
Macrophyte+Woody debris (%)
MPW
Substrate
Leaf litter (%)
LL
Sand (%)
SA
Silt (%)
SI
Bank
Shape (smooth 1- vertical 4)
BSH
Root (0-3)
BRO
shrubs/grass (0-3)
BSG
Riparian
Heterogeneity (0-3)
RHE
Canopy (%)
RCA
Connectivity
Stream order
SO
Forest edge (0-3)
FE
Open area (0-3)
OA
Swamp (0-3)
SW
Appendix 3. Abbreviations of dragonfly species.
Abbreviation
Species
Common name
Afe
Agriocnemis femina
Variable Wisp
Agr
Amphicnemis gracilis
Will-o-wisp
Agu
Anax guttatus
Emperor
Apa
Acisoma panorpoides
Trumpet Tail
Aru
Argiocnemis rubescens
Variable Sprite
Avi
Archibasis viola
Violet Sprite
Bch
Brachydiplax chalybea
Blue Dasher
Cce
Ceriagrion cerinorubellum
Ornate Coraltail
Cma
Coper a marginipes
Yellow Featherlegs
Cme
Cratilla metallica
Dark-tipped Forest Skimmer
Dragonflies of Nee Soon swamp forest
153
Appendix 3. Continuation.
Coc
Coeliccia octogesima
Telephone Sylvan
Cse
Crocothemis servilia
Common Scarlet
Dqu
Drepanosticta quadrata
Singapore Shadowdamsel
Eim
Euphaea impar
Blue-sided Satinwing
Evi
Epophthalmia vittigera
Pond Cruiser
Ide
Ictinogomphus decoratus
Common Flangetail
Las
Lathrecista asiatica
Scarlet Grenadier
Lau
Libellago aurantiaca
Fiery Gem
Lhy
Libellago hycilina
Clearwing Gem
Lli
Libellago lineata
Golden Gem
Mqu
Macrogomphus quadrata
Forktail
Nfl
Neurothemis fluctuans
Common Parasol
Npy
Nannophya pygmaea
Scarlet Pygmy
Och
Orthetrum chrysis
Spine-tufted Skimmer
Ogl
Orthetrum glaucum
Common Blue Skimmer
Opr
0 rch i th em is p ru i nans
Blue Sentinel
Opu
0 rch ithem is pulcherrima
Variable Sentinel
Osa
Orthetrum sabina
Variegated Green Skimmer
Pau
Pseudagrion australasiae
Look-alike Sprite
Pea
?Paragomphus capricornis
Banded Hooktail
Pco
Prodasineura collaris
Collared Threadtail
Phu
Prodasineura humeralis
Orange-striped Threadtail
Pin
Prodasineura interrupta
Interrupted Threadtail
Pjo
Pseuclothemis jorina
Banded Skimmer
Pmi
Pseudagrion microcephalum
Blue Sprite
Pno
Prodasineura notostigma
Crescent Threadtail
Por
Podolestes orientalis
Blue-spotted Flatwing
Ppr
Pseudagrion pruinosum
Grey Sprite
Rob
Rhyothemis obsolescens
Bronze Flutterer
Rph
Rhyothemis phyllis
Yellow-Barred Flutterer
Rru
Rhodothemis rufa
Common Redbolt
Rtr
Rhyothemis triangularis
Sapphire Flutterer
Tau
Trithemis aurora
Crimson Dropwing
Tfe
Trithemis festiva
Indigo Dropwing
Tto
Tyriobapta torrida
Treehugger
Vae
Vestalis amethystina
Common Flashwing
Yam
Vestalis amoena
Charming Flashwing
Gardens’ Bulletin Singapore 70 (Suppl. 1): 155-173. 2018
doi: 10.26492/gbs70(suppl.l). 2018-08
155
Next-Generation identification tools for Nee Soon
freshwater swamp forest, Singapore
S.N. Kutty 1 , W. Wang 2 , Y. Ang 2 , Y.C. Tay 1 , J.K.I. Ho 1 & R. Meier 1 - 2
Evolutionary Biology Laboratory, Department of Biological Sciences,
National University of Singapore,
14 Science Drive 4, 117543 Singapore
meier@nus.edu.sg
2 Lee Kong Chian Natural History Museum, National University of Singapore,
2 Conservatory Drive, 117377 Singapore
ABSTRACT. Many invertebrate and plant species are difficult to identify even by taxonomic
experts. This has created a major obstacle for understanding the ecology of tropical
environments. Here we explore the use of new large-scale, cost-effective approaches to species
identification using Next-Generation Sequencing (“DNA barcodes”). Due to the rapid drop
in sequencing cost, such barcodes have the potential to help with many identification tasks
and they will facilitate regular monitoring of habitats. We use this approach to explore the
species diversity of Nee Soon freshwater swamp forest and provide taxonomic identification
tools for the fauna and flora of the forest. DNA-barcode libraries were generated for the flora
(>1000 barcodes; 170 chloroplast genomes) and fauna (ca. 3000 barcodes). In addition, high-
resolution images of 502 animal and 200 plant species were placed on an online image database
(“Biodiversity of Singapore”). These images are available to help experts and non-experts alike
to identify and appreciate these species. The new databases document Nee Soon’s impressive
diversity, but they are also important for in-depth studies of fauna-floral species interactions.
For example, the plant barcodes were used to reconstruct the diet of Raffles’ banded langur
based on faecal samples. Overall, we find that the fauna in Nee Soon freshwater swamp forest
is very diverse and includes many rare species, and that the species composition is very distinct
from those living in surrounding habitats. Animal specimens are readily sequenced, while plant
specimens (especially those represented by sapwood samples) remain a challenge. However,
newer techniques (e.g. based on genome skimming) are starting to help with obtaining plant
DNA-barcodes.
Keywords. DNA-barcoding, Next-Generation sequencing, online image database, species
discovery
Introduction
One of the most irritating and fascinating properties of tropical environments is that
they are species-rich (0degaard, 2000). The large number of species means that there
are potentially a large number of important biological players in a system. This makes
it very difficult to understand critical interactions that sustain the environment. This
problem is exacerbated by the fact that the identification of biological specimens to
species level is far from straightforward and yet important, because species have
156
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
different natural histories and species names are used for filing and retrieving biological
information (Gotelli, 2004). Because the specimens encountered in ecosystems do not
come with species name tags, biologists have to use a variety of techniques for, a)
delimiting (grouping specimens into species) (Wheeler & Meier, 2000), b) describing
species (Winston, 1999), and c) identifying species (Walter & Winterton, 2007).
Not surprisingly, the best techniques for these purposes vary from taxon to taxon. In
addition, the best methods for species identification change over time.
From an identification point of view, the most convenient taxa are those wherein
species diversity is well understood (i.e., species delimitation and description are quite
complete: birds, butterflies), species identification can be accomplished based on
readily-accessible features (e.g. morphology, songs, etc.), and the relevant features
for identification can be obtained without collecting or even disturbing the animals
or plants. Fortunately, many vertebrate species fall into this category. On the other
end of the scale are taxa where many/most species are neither delimited or described,
and therefore not identifiable. Unfortunately, more than 90% of the world’s species
(mostly invertebrates) fall into this category (Meier & Dikow, 2004; Odegaard, 2000).
Intermediate along the spectrum of identification feasibility are taxa for which most
species have been delimited and described by scientists, but identification is difficult for
a variety of reasons. These include the lack of good identification tools (e.g. keys), the
reliance on identification features that can only be used by a few taxonomic experts, and
the use of identification features that are only visible during certain times of a species’
life cycle. Good examples are many insect species that can only be identified based on
minute details of genitalia (Ang & Meier, 2010; Pont & Meier, 2002), species of plants
that can only be identified when they happen to flower, and insects whose aquatic
larval or nymphal stages are unidentifiable because identification tools only exist for
adults (e.g. dragonflies, midges; Cranston et al., 2013). It is often the unidentifiable
stages that are the most important, from an ecological and biomonitoring point of view
(vegetative parts of plants, larval stages of insects).
Identifying most species is a task that can currently only be performed by
experts with extensive training in biology. Indeed, for many invertebrate groups
there are only a handful of experts worldwide who can identify species (Gotelli,
2004). This unfortunately means that many identification needs of society are not
met. An alternative way of identifying species is through the use of so-called “DNA
barcodes” (Hebert et al., 2003a; Meier et al., 2006, 2016; Meier, 2008). For animals,
most biologists use a small piece of the cytochrome oxidase subunit 1 (COT) gene
for species identification (“DNA barcode”; Hebert et al., 2003b). This particular gene
sequence (barcode) is distinctly different between most species (Kwong et al., 2012a;
Hajibabaei et al., 2007; Meier, 2008; but see Kwong et al., 2012b and Meier et al.,
2006). One advantage of using DNA barcodes is that it “democratises” the process of
species identification. Instead of only having a handful of experts worldwide who can
identify species in a particular group, DNA barcodes can be generated by thousands
of laboratories around the world. In addition, the cost of obtaining DNA barcodes
has been dropping rapidly so that the number of biologists with access to these kinds
of data is also increasing rapidly (Wong et al., 2014; Meier et al., 2016). Barcoding
DNA barcoding for Nee Soon flora and fauna
157
all individuals from specimen rich bulk samples with cost-effective high-throughput
pipelines also allows for presorting using DNA barcodes and mitigates downstream
morphological work on presorted units (Wang et al., 2018). DNA barcodes have the
additional advantage of enabling associations between different life history stages
(e.g. larvae and adults; see Yeo et al., in press), and the identification of animal and
plant parts that are otherwise not diagnostic. For example, DNA fragments can be used
to carry out a diet analysis based on DNA remnants in faecal matter (Srivathsan et al.,
2015, 2016), while free-floating DNA in water can be used to assess which animals
were swimming in the water (Lim et al., 2016).
Advantages aside, the use of DNA barcodes in species identification comes with
several caveats that we need to bear in mind; some stem from the nature of species,
while others are essentially technical. For example, DNA barcoding uses genes that
are not functionally related to the origin of species (Kwong et al., 2012b). Instead, the
species-specific signatures in barcode genes are due to the fact that most species pairs
are old enough that sister species are distinguishable based on the genetic differences
that accumulated over evolutionary time through a mixture of genetic drift and natural
selection (Meier, 2008). Predictably, recently diverged species pairs can share DNA
barcodes; i.e., they cannot be distinguished based on these barcodes. Based on ten
years of experience with barcoding, this is fairly rare in animal species and about
90% of all species have their own signature in COI sequences. A bigger problem that
is more technical in nature is that a large proportion of animal species are not yet
barcoded which interferes with the use of DNA barcodes for species identification
(Kwong et al., 2012a). This is unfortunate because many environmental problems can
be diagnosed using DNA barcodes, e.g. the presence of invasive species (Collins et
al., 2012; Ng et al., 2016). As for plants, their genes evolve slower so that there is a
larger proportion of closely-related species that are indistinguishable based on DNA
barcodes (Hollingsworth, 2008; Hollingsworth et al., 2011). This means that DNA
barcodes can often only distinguish plant genera. One solution to this problem - which
was also pursued in this study - is sequencing multiple genes or whole chloroplast
genomes (see “genome skimming”; Straub et al., 2012).
Other technical problems with DNA barcodes are mostly related to cost and
time. In particular, traditional Sanger-based DNA barcodes are very expensive
(consumables and manpower). Fortunately, we recently developed Next-Generation-
Sequencing (NGS)-based DNA barcodes that circumvent these problems (Meier et
al., 2016). This is why we were also able to barcode a large number of Nee Soon
specimens and use this information for species discovery. Another technical problem
is the large amount of Polymerase Chain Reaction (PCR)-inhibitors in DNA extracts of
plants. This interferes with amplifying plant barcodes. We addressed this issue through
the use of different extraction techniques and by using genome skimming for obtaining
chloroplast genomes. The latter has fewer amplification problems and yields more data
at roughly the same cost because the cost per base pair of DNA is much lower for NGS
than Sanger sequencing.
Identification of specimens via DNA barcodes is slower than identification via
morphology for those species with obvious diagnostic morphological features. For
158
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Sanger barcodes, the normal time between collection and obtaining identification is
two working days while it can be several weeks for cost-effective barcoding via NGS.
This is why morphology is the identification technique of choice for all species where
the relevant morphological features are obvious and easily accessible (or can be made
assessable through better imaging). Such data can now be conveniently displayed
online in digital reference collections (Ang et al., 2013a) and modem publishing also
allows for image-rich species descriptions (Ang et al., 2013b). Ideally, morphology
and DNA should be combined in recognising species and providing identification tools
(Tan et al., 2010; Rohner et al., 2014). Such “integrative taxonomy” is most likely to
identify accurate species boundaries and allows for species identification based on
either type of data.
In the Nee Soon hydrology and biodiversity project (Clews et al., 2018; Davison
et al., 2018), our team used a wide variety of techniques to tackle species identification
problems and to create tools for the future. The ultimate goal was to enable and to make
it easier to identify biological specimens from Nee Soon freshwater swamp forest. A
secondary goal was to generate more “democratic” identification tools; i.e., to provide
tools that are less reliant on expensive and rarefy available taxonomic expertise.
Democratisation of species identification can be achieved by generating higher-quality
images that help non-experts identify species (Ang et al., 2013a). This approach was
pursued for many animal and plant taxa and we generated a species database in which
more than 500 species are illustrated. This database is also a colourful celebration of
Nee Soon freshwater swamp forest’s glorious biodiversity.
Main objectives
1) Insect diversity. In order to explore the insect diversity of Nee Soon freshwater
swamp forest, we used NGS-based species discovery techniques for targeting taxa that
belong to different ecological guilds.
2) Faunal identifications: Nee Soon freshwater swamp forest was extensively surveyed
by Faunal Ecology teams (Ho et al., 2018; Tim et al., 2018) who collected a large
number of specimens. We imaged these specimens using specialised digital camera
systems and sequenced the CO/barcode for these specimens using Sanger sequencing
and NGS.
3) Floral identifications: The flora of Nee Soon freshwater swamp forest was studied
by the Vegetation Ecology team (Chong et al., 2018) who provided samples for
DNA barcoding. Initially, we targeted multiple plant barcode genes ( matK , rbcL,
trnL, etc.) but due to PCR-inhibitors this was very expensive in terms of manpower
and consumables. We therefore switched to NGS-based sequencing of chloroplast
genomes via genome skimming. The sequencing efforts concentrated on trees and
lianas because they are most relevant for understanding the vegetation ecology of Nee
Soon freshwater swamp forest. To assist the floral field team with identifying tree
DNA barcoding for Nee Soon flora and fauna
159
species in which taxonomically important parts were not readily accessible, we also
developed a NGS-based technique for identifying trees to genus based on sapwood
samples. This was challenging because such samples contain little DNA and large
amounts of PCR-inhibitors.
Methods
Faunal barcoding
Barcodes for insects were generated through COI amplification using direct-PCR
(dPCR) (Wong et al., 2014), where a small amount of tissue is dissected from each
individual specimen and serves as a template for amplification without prior DNA
extraction. In addition, DNA from many specimens was also extracted using a
novel reagent known as QuickExtract™ DNA Extraction Solution (EPICENTRE
Biotechnologies); DNA extracts obtained with QuickExtract were used directly as
input template for amplification of barcoding genes. A short fragment (313 bp) of COI
was used as the general faunal barcoding gene. Subsequent downstream sequencing
was conducted using a combination of traditional Sanger sequencing and high-
throughput, paired-end NGS on Illumina™ platforms. Most barcodes were generated
with NGS (Srivathsan et al., 2015). For NGS barcodes, each specimen’s amplicons
were tagged with a uniquely labelled primer pair in the PCR step; the use of indexed
primers allowed for barcodes to be traced accurately to their specimen of origin in the
downstream bioinformatic process. Sequences generated from either Sanger or NGS
methods were aligned using MAFFT ver.7 (Katoh & Standley, 2013), before being
grouped into molecular operational taxonomic units (mOTUs) based on objective
clustering, whereby sequences are grouped by similarity based on uncorrected
pairwise (p) distances at specific uncorrected percentage thresholds (Meier et al., 2006;
Srivathsan & Meier, 2012). Ideally, the threshold value set for clustering into mOTUs
should be within a numerical range where the number of clusters remains stable; this
is because stable clusters are likely to represent species. When used appropriately,
intraspecific and interspecific variability can also be compared to further assess the
stability of the species boundaries (Meier et al., 2008).
Floral barcoding
DNA extraction for all floral specimens were carried out on leaf and sapwood samples
using a modified CTAB-chloroform extraction protocol (Doyle & Doyle, 1987).
DNA barcodes for plant species were generated using both Sanger sequencing and
NGS platforms. Initially, sequences for four selected barcode genes were amplified
and sequenced: maturase K ( matK ), ribulose-bisphosphate carboxylase (rbcL ), a
chloroplast tRNA gene ( trnL ), and the internal transcribed spacer 2 ( ITS2 ) region of
nuclear ribosomal DNA. The trnH-psbA intergenic spacer region was also explored.
With the advent and availability of high throughput sequencing, Illumina™
platforms were used for both amplicon sequencing and chloroplast genome skimming.
160
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Using tagged primers similar to the protocol used for COI barcoding, data was
generated on a MiSeq sequencer. To cost-efficiently carry out genome skimming for
~240 species, multiplexing was done by ligating a 20bp species-specific tag to the
DNA of different species using a modified version of the Meyer & Kircher (2010)
protocol. Three such “plant pools” libraries comprising of 75-100 species each with
insert sizes of 400-900bp were prepared and sequenced on a HiSeq 2500 (250PE)
platform. The sequence data were demultiplexed based on species-specific tags using
SABRE (https://github.com/najoshi/sabre) and quality checked using custom scripts.
Trimming of reads were carried out in CLC Genomics Workbench (Limit=0.001,
https://www.qiagenbioinformatics.com) and assembled into chloroplast contigs using
default parameters in MITOBIM (Hahn et al., 2013), by iterative mapping onto a
closely-related reference chloroplast genome. Species reads were mapped to the
MITOBIM-assembled contigs in CLC Genomics Workbench to calculate the average
coverage of each chloroplast genome.
Tree identifications via sapwood samples
The suitability of all four plant barcodes for tree identification via sapwood material
was assessed in preliminary experiments, which showed the ~400bp fragment of the
ITS2 marker to be most effective at identification. However, PCR amplification and
Sanger sequencing successes with this marker were low due to length variants (success
rates of only -30%). Hence, we switched to using the trnL markers (short fragment of
10-50bp) for sapwood-based identifications of the remaining samples unidentifiable
with the previous marker. Between one to five PCR amplifications using tagged
primers for the trnL marker were performed for each sample, and sequenced on an
Illumina™ MiSeq Nano run. Sequence data obtained from the run were demultiplexed
and binned into unique read clusters using PEAR (Zhang et al., 2014) and OBITOOLS
(Boyer et al., 2016) respectively. Consensus sequences of each unique cluster were
then matched against both the global and local plant trnL databases for identifications
via blastn (BLAST 2.2.28+, Camacho et al ., 2009).
Specimen imaging and online database
Photography and image preparation. The specimens are kept at the Lee Kong Chian
Natural History museum (specimens in main collection and DNA in cryo-collection).
One specimen per species was imaged using a high-resolution photomacrography
system (Visionary Digital™ Lab Plus System). Specimens were imaged under high
magnification at different focal depths and exported via Adobe Lightroom. These
images were then digitally stacked into a completely focused composite image using
Helicon Locus Pro. The composite images were then digitally optimised in Photoshop
CS5 Extended by white-balancing, image sharpening, light/shadow adjustments, and
digitally removing impurities from background and specimens. Depending on the
taxon, specimens were imaged in different orientations and magnifications to illustrate
key diagnostic features. These separate images were then digitally stitched into an
image plate.
DNA barcoding for Nee Soon flora and fauna
161
Image plates were exported in a format that allows for online magnification. This
allows the users on internet browsers to view both an overall view of the plate but also
zoom in to high-magnification images of structures that are critical for identification: it
divides an image into a series of smaller-sized picture tiles at different resolutions and
sizes that are presented onto a fixed frame. Because the viewer frame requires only few
picture tiles to be loaded at any time, viewing is fast and smooth.
Online image database. All images are displayed on an online image database
(https://singapore.biodiversity.online/). The website also displays other species from
Singapore. The webpage has a collapsible taxon-based navigation panel on the left,
while the main field displays all imaged specimens (in thumbnails). There are also
filter options. The image database features a three-tiered design; users can select a
taxon group on the left navigation panel, which will show all the available species on
the right panel, segregated by another two taxon levels (usually, order, followed by
family or genus). Clicking on a species thumbnail will direct users to the species page,
which displays the image plate for the specimen, as well as other basic information
such as species name, common name, taxonomic information, image information
as well as other additional species information and links to other websites for more
information on the species (where available). Users can then navigate to other taxon
groups using navigation panels.
Results
Faunal barcoding
Using a combination of traditional Sanger sequencing and high throughput Next
Generation Sequencing (NGS barcoding, we generated 2904 animal barcodes
(predominantly COI, with some COII for Odonata) (Table 1). Overall, we have a total
of 2904 barcodes from the faunal specimens collected from Nee Soon freshwater
swamp forest. The majority of barcodes are for insects: Diptera (1399), ants (652)
and Odonata (347). Insects collected from Nee Soon’s waterways are represented by
128 barcodes. Fishes represent the next largest contribution (201) of barcodes for non¬
insect fauna. However, numerous additional barcodes continue to be generated for the
material that was collected. All barcodes were checked against Genbank in order to
rule out contamination.
DNA barcode database for plants
A total of 1189 barcodes were generated for 503 species representing 98 plant families,
using a combination of traditional Sanger sequencing and high throughput Next
Generation Sequencing. We also generated genome data for 240 species from which
170 chloroplast genomes could be assembled via genome skimming. Coverage of
chloroplast genomes depended on both sequencing depth and amount of chloroplast
162
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 1. Faunal barcodes generated for Nee Soon.
Taxa
Overall no. of barcodes
Fishes
201
Mollusca (Snails)
4
Crabs and Shrimp (Decapoda)
12
Damsel- and Dragonflies (Odonata)
347
Bees (Anthophila)
49
Ants (Formicidae)
652
Termites (Isoptera)
112
Fungus Gnats (Mycetophilidae)
875
Mosquitoes (Culicidae)
320
Horse Flies (Tabanidae)
5
Hover Flies (Syrphidae)
8
Soldier Flies (Stratiomyidae)
19
Chironomidae (Non-biting midges)
170
Ceratopogonidae (Biting Midges)
2
Baetidae (Small Minnow Mayflies)
3
Caenidae ( Squaregill Mayflies)
3
Gerridae (Water Striders)
1
Nepidae (likely Ranatra )
1
Hemipteran (likely leafhopper)
1
Gyrinidae (Diving beetles)
1
Scirtidae (Marsh beetles)
1
Nemouridae (Stoneflies)
3
Perlidae (Stoneflies)
2
Calamoceratidae (Caddisfly)
2
Dipseudopsidae (Caddisfly)
9
Ecnomidae (Caddisfly)
10
Hydropsychidae (Caddisfly)
32
Hydroptilidae (Caddisfly)
6
Leptoceridae (Caddisfly)
26
Polycentropodidae (Caddisfly)
24
Psychomyiidae (Caddisfly)
2
Blattodea (Cockroach)
1
Total: 2904
DNA barcoding for Nee Soon flora and fauna
163
50,000 100,000 150,000
Ficus_gro$sufarioidei --
average coverage: 11SX
Garcinia_ rostra ta
average coverage: 1 3.7SX : EEEE iEEEEEEEEE... —i n £ £^.-4 “ r.^ —:^-cE"EE E EEE EE Ec^e^= E VrE
Fig. 1. Variation in coverage of chloroplast genomes (>150,000 bp) between representative
Nee Soon species.
material in each species. Based on our data, chloroplast reads varied from 0.23% to
12% of total reads depending on sample. These factors contributed to the uneven
genome coverage (see Fig. 1).
Tree identifications via sapwood samples
A total of 360 amplicons from inner bark and sapwood scrapes were sequenced. Based
on the local ITS2 database generated from the leaf samples, Sanger DNA barcodes
allowed identification of 40 of the 87 sapwood samples with high confidence to at least
the genus level (>=95% sequence match), and 8 samples to at least the family level
(90-95% sequence match). Of the 173 sapwood samples that were sent for Illumina
sequencing of the short trnL gene fragment, 56 and 29 samples were identified with
high confidence to the family and genus levels respectively. However, we also find
molecular identifications that conflict with expected IDs that were obtained with
morphological means; i.e., the technique requires more testing.
Specimen imaging and databasing
A total of 502 faunal specimens originating from Nee Soon freshwater swamp
forest were imaged and uploaded onto the Biodiversity of Singapore database: (Fig.
164
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 2. The home page of the Biodiversity of Singapore website.
2: https://singapore.biodiversity.online/) as individual species pages (Fig. 3). This
includes specimens from multiple groups, including Vertebrata, Crustacea, Mollusca,
Coleoptera, Odonata, Blattodea, Ephemeroptera, Trichoptera, Plecoptera, Hemiptera
and Diptera (Fig. 4). Additionally, images for more than 200 plant species were also
added into the database. Table 2 shows the breakdown by taxa. These high resolution
images allow for a close-up view of the finer details of the individual species as well
(Fig. 5).
Discussion
We set out to use a wide variety of techniques to tackle species identification problems
and to create identification tools for the future. As documented in the Results section,
we succeeded to various degrees. In addition, the material that was collected during
the project continues to be studied and new species are found and imaged every week.
The Faunal Ecology teams extensively surveyed the aquatic habitats and
collected a large number of specimens (Ho et al, 2018; Fim et al., 2018). Because
the processing of these samples was time-consuming, we only obtained them fairly
late. Nevertheless, our preliminary results indicate that the species diversity of Nee
Soon is very high and that most species found in the aquatic environments of Nee
Soon freshwater swamp forest are distinctly different from what is found in the nearby
reservoirs. A good example are the chironomid midges where preliminary sampling
revealed more than 250 species in Nee Soon, while only about 40 species were found
DNA barcoding for Nee Soon flora and fauna
165
Fig. 3. An example of a species page from the Biodiversity of Singapore website.
in the surrounding reservoirs - with some species reaching nuisance levels (Cranston
et al., 2013). What is remarkable is that only very few species are shared (unpublished
data). We see similar results emerging for other taxa (Odonata, Trichoptera). However,
a full evaluation will take some time.
We have also carried out species discovery projects on terrestrial insect groups
including ants, termites, fungus gnats (indicators of fungal diversity), stratiomyids
(soldier flies), syrphids (hover flies that are pollinators), and mosquitoes. Most of
these taxa are very species rich in Nee Soon and the fauna is again very distinct from
other habitats for which data exists (NUS campus, mangrove fragments in Singapore).
Particularly remarkable is the very high species diversity in fungus gnats (over 200
species). In order to explore the diversity in so many insect groups, we relied on
DNA barcodes. To date, we have generated more than 3000 barcodes for the fauna of
Nee Soon freshwater swamp forest. The DNA barcodes for different specimens were
compared and grouped based on overall similarity (3% threshold) (Meier et al., 2008;
Srivathsan & Meier, 2012). Afterwards, one specimen for each cluster was selected
for imaging. So far, over 500 images of Nee Soon species have been added to the
publicly accessible database “Biodiversity of Singapore”, while the entire database
includes over 10000 species. However, many additional species are being added every
month. Important recent additions are images for the c. 200 species of fungus gnats.
Most of the species are currently only known from Nee Soon freshwater swamp forest
and many species are only known from a single specimen. Such rarity is a pervasive
pattern of many tropical species (Lim et al., 2011).
We also succeeded in barcoding most of the important plant species in Nee Soon
166
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 4. Some of the diversity of macroinvertebrates found in Nee Soon.
(total of 1189 barcodes). The barcoding gene with the highest sequencing success was
rbcL (321 species). Unfortunately, this gene rarely allows for distinguishing closely
related species. We therefore invested more resources and time into getting sequences
for matK (275 spp), trnL (321 species), ITS2 (190 spp), and TrnH-psbA (86 spp). The
value of these barcodes was immediately illustrated when they were used to analyse
the diet of Singapore’s critically endangered Raffles’ banded langur population (Ang
et al., 2010; Ang et al., 2012) based on their faecal samples (Srivathsan et al., 2015,
2016). The faecal material included DNA signatures for more than 50 plant species
(Ang et al., 2013a) and it will be important for the conservation of Raffles banded
langur to keep healthy populations of food plants. Unfortunately, obtaining these plant
barcodes via PCR and Sanger sequencing was extremely time-consuming, therefore we
also developed a different approach via genome skimming (low coverage sequencing
of whole-genomes). Based on the results, we predict that genome skimming will be
DNA barcoding for Nee Soon flora and fauna
167
Fig. 5. A close-up view of one of the species found in Nee Soon.
the future technique of choice because it is not only cost-effective, but also yields more
data. Genome skimming is a viable alternative to plant barcoding, because vegetative
tissues are naturally enriched with chloroplast genes, thus low coverage sequencing of
whole-genome extractions can yield enough data for obtaining chloroplast genomes.
Whole chloroplast genomes automatically cover most plant barcoding genes that
are located on this genome, while yielding much more information than barcoding
genes, because whole genomes are much longer (150,000 bp) than all barcoding genes
combined (<2000 bp). The work yielded chloroplast genomes for -170 species, but we
hope to eventually cover much of Singapore’s flora. A barcode database for all species
would contribute towards understanding species interactions as illustrated by our work
on the diet of Raffles’ banded langur.
168
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 2. Breakdown for images of species according to taxa featured on the image database.
Taxon group
No. species/MOTUs featured
Vertebrates
(subtotal: 97)
Fishes
53
Anura (Frogs)
16
Aves (Birds)
18
Mammalia (Mammals)
10
Crustacea
(subtotal: 9)
Decapoda (Shrimps)
5
Brachyura (Crabs)
4
Mollusca
(subtotal: 7)
Gastropoda (Terrestrial snails)
7
Diptera (True Flies)
(subtotal: 138)
Dolichopodidae (Long-legged Flies)
15
Chironomidae (Non-biting Midges)
1
Culicidae (Mosquitoes)
35
Mycetophilidae (Fungus Gnats)
78
Stratiomyidae (Soldier Flies)
7
Ceratopogonidae (Biting Midges)
2
Odonata
(subtotal: 35)
Anisoptera (Dragonflies)
19
Zygoptera (Damselflies)
16
Blattodea
(subtotal: 33)
Isoptera (Termites)
32
Cockroach
1
Ephemeroptera (Mayflies)
(subtotal: 2)
Baetidae (Small Minnow Mayflies)
1
Caenidae (Small Squaregill Mayflies)
1
DNA barcoding for Nee Soon flora and fauna
169
Table 2. Continuation.
Taxon group
No. species/MOTUs featured
Trichoptera (Caddisflies)
(subtotal: 7)
Calamoceratidae
2
Ecnomidae
1
Hydropsychidae
1
Leptoceridae
3
Plecoptera (Stoneflies)
(subtotal: 2)
Perlidae
2
Hemiptera (True Bugs)
(subtotal: 3)
Gerridae (Pond Skaters)
1
Nepidae (Water Scorpions)
1
Cicadomorpha (likely leafhopper)
1
Coleoptera (Beetles)
(subtotal: 2)
Gyrinidae (Whirligig beetles)
1
Scirtidae
1
Plants
(subtotal: 164)
Peridophyta (Ferns)
2
Monocots
5
Magnoliids
45
Rosids
78
Asterids
22
‘other’ Eudicots
12
Total: 502
Lastly, we used Sanger barcodes and developed NGS barcodes for identifying
trees to genus based on sap wood samples. This is often needed because it is difficult
to obtain leaves from tall trees; frequently, the only available material is sap wood
sampled from the tree trunk. Unfortunately, obtaining DNA from such samples
is challenging because the DNA content is small and the DNA extractions include
170
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
large amounts of PCR-inhibitors. In order to succeed, we first built an ITS2 database
based on leaf samples and used Sanger barcodes to identify most of the 89 sapwood
samples to at least family level. However, we were not able to use the same approach
for the remaining sapwood samples because there were too many problems with
DNA amplification and sequencing. We therefore switched to NGS barcoding of a
short trnL gene fragment for 169 sapwood samples. The advantage of this approach
is that the shorter fragments are more likely to amplify, but this comes at the cost
of these fragments containing less information. Using this approach, we were able
to identify 44, and 29 samples with high confidence to the family and genus levels
respectively. Compared to barcoding of leaf samples, sapwood samples will remain
very problematic, hence new approaches should continue to be pursued. Particularly
promising may be anchored hybrid enrichment of chloroplast genes.
Conclusions
Being able to identify specimens to species is important for most in-depth study
of biological systems. However, obtaining these identifications is very challenging
in tropical environments. Fortunately, a number of new tools make this task less
daunting. New imaging techniques help with illustrating relevant characters and new
and cheaper DNA barcodes allow for the generation of databases that can be used by
many researchers. Overall, making the fauna and flora of Nee Soon freshwater swamp
forest and Singapore identifiable is achievable. Several hundred, or even thousands of
species may potentially be revealed from samples that have been collected and stored.
By focusing on particular taxa belonging to different ecological guilds, it is feasible
to start understanding species turnover rates across habitats in Singapore and to use
this information for conserving Singapore’s native fauna and flora. A particularly high
priority is obtaining plant barcodes for all of Singapore’s vascular plant species. This
will allow for in-depth studies of species interactions between plants and animals (e.g.
pollination).
ACKNOWLEDGEMENTS. This study forms part of the research project “Nee Soon Swamp
Forest Biodiversity and Hydrology Baseline Studies—Phase 2” funded by the National Parks
Board (NParks), Singapore. The authors are also grateful to the Public Utilities Board (PUB),
Singapore for making this study possible.
References
Ang, Y. & Meier, R. (2010). Five additions to the list of Sepsidae (Diptera) for Vietnam:
Perochaeta cuirassa sp. n., Perochaeta lobo sp. n., Sepsis spura sp. n., Sepsis sepsi
Ozerov, 2003 and Sepsis monostigma Thompson, 1869. Zookeys 70: 41-56.
Ang, A., Ismail, M.R.B. & Meier, R. (2010). Reproduction and infant pelage colouration of the
banded leaf monkey (Mammalia: Primates: Cercopithecidae) in Singapore. Raffles B.
Zool. 58:411-415.
DNA barcoding for Nee Soon flora and fauna
171
Ang, A., Srivasthan, A., Md-Zain, B.M., Ismail, M.R.B. & Meier, R. (2012). Low genetic
variability in the recovering urban banded leaf monkey population of Singapore. Raffles
B. Zool. 60(2): 589-594.
Ang Y., Puniamoorthy, J., Pont, A.C., Bartak, M., Blanckenhorn, W.U., Eberhard, W.G.,
Puniamoorthy, N., Silva, V.C., Munari, L. & Meier, R. (2013a). A plea for digital
reference collections and other science-based digitization initiatives in taxonomy:
Sepsidnet as exemplar. Syst. Entomol. 38: 637-644.
Ang, Y., Wong, L.J. & Meier, R. (2013b). Using seemingly unnecessary illustrations to improve
the diagnostic usefulness of descriptions in taxonomy-a case study on Perochaeta
orientalis (Diptera, Sepsidae). Zookeys 355: 9-27.
Boyer, F., Mercier, C., Bonin, A., Le Bras, Y., Taberlet, P. & Coissac, E. (2016). OBITOOLS:
a UNIX-inspired software package for DNA metabarcoding. Mol. Ecol. Resour. 16:
176-182.
Camacho, C., Coulouris, G., Avagyan, V., Ma, N., Papadopoulos, J., Bealer, K. & Madden, T.L.
(2009). BLAST+: Architecture and applications. BMC Bioinformatics 10: 421.
Chong, K.Y., Lim, R.C.J., Loh, J.W., Neo, L., Seah, W.W., Tan, S.Y. & Tan, H.T.W. (2018).
Rediscoveries, new records, and the floristic value of the Nee Soon freshwater swamp
forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 49-69.
Clews, E., Corlett, R.T., Ho, J.KX, Kim, D.E., Koh, C.Y., Liong, S.Y, Meier, R., Memory, A.,
Ramchunder, S.J., Sin, T.M., Siow, H.J.M.P, Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W.,
Theng, M.T.Y., Wasson, R.J., & Yeo, D.C.J. & Ziegler, A.D. (2018). The biological,
ecological and conservation significance of freshwater swamp forest in Singapore.
Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
Collins, R.A., Armstrong, K.F., Meier, R., Yi, Y., Brown, S.D.J., Cruickshank, R.H., Keeling,
S. & Johnston, C. (2012). Barcoding and border biosecurity: identifying cyprinid fishes
in the aquarium trade. PLoS ONE 7(1): e28381.
Cranston, P, Ang, Y., Heyzer, A., Lim, R.B.H., Wong, W.H., Woodford, J.M. & Meier, R.
(2013). The nuisance midges (Diptera: Chironomidae) of Singapore’s Pandan and
Bedok reservoirs. Raffles B. Zool. 61: 779-793.
Davison, G.W.H., Cai, Y.X., Li, T.J. & Lim, W.H. (2018). Integrated research, conservation and
management of Nee Soon freshwater swamp forest: hydrology and biodiversity. Gard.
Bull. Singapore 70 (Suppl. 1): 1-7.
Doyle, J.J. & Doyle, J.L. (1987). A rapid DNA isolation procedure for small quantities of fresh
leaf tissue. Phytochem. BullA9: 11-15.
Gotelli, N.J. (2004). A taxonomic wish-list for community ecology. Philos. T. Roy. Soc. B 359:
585-597.
Hahn, C., Bachmann, L. & Chevreux, B. (2013). Reconstructing mitochondrial genomes
directly from genomic next-generation sequencing reads-—a baiting and iterative
mapping approach. Nucleic Acids Res. 41: el29.
Hajibabaei, M., Singer, G.A.C., Hebert, P.D.N. & Hickey, D.A. (2007). DNA barcoding: how
it complements taxonomy, molecular phylogenetics and population genetics. Trends
Genet. 23 (4): 167-172.
Hebert, P.D.N., Cywinska, A., Ball, S.L. & deWaard, J.R. (2003a). Biological identifications
through DNA barcodes. Proc. Roy. Soc. London, Ser. B, Biol. Sci. 270: 313-321.
Hebert, P.D.N., Ratnasingham, S. & deWaard, J.R. (2003b). Barcoding animal life: Cytochrome
c oxidase subunit 1 divergences among closely related species. Proc. Roy. Soc. London,
Ser B, Biol. Sci. 270: S96-S99.
172
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Ho, J.K.I., Quek, R.F., Ramchunder, S.J., Memory, A., Theng, M.T.Y., Yeo, D.C.J. & Clews,
E. (2018). Aquatic macroinvertebrate richness, abundance and distribution in the Nee
Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 71-108.
Hollingsworth, RM. (2008). DNA barcoding plants in biodiversity hot spots: Progress and
outstanding questions. Heredity 101: 1-2.
Hollingsworth, P.M., Graham, S.W. & Little, D.P. (2011). Choosing and using a plant DNA
barcode. PLoS ONE 6(5): el9254
Katoh, K., & Standley, D.M. (2013). MAFFT multiple sequence alignment software version 7:
improvements in performance and usability. Mol. Biol. Evol. 30(4): 772-780.
Kwong, S., Srivathsan, A., Meier, R. (2012a). An update on DNA barcoding: low species
coverage and numerous unidentified sequences. Cladistics 28: 639-644.
Kwong, S., Srivathsan, A., Vaidya, G. & Meier, R. (2012b). Is the COI barcoding gene involved
in speciation through intergenomic conflict? Mol. Phylogenet. Evol. 62: 1009-1012.
Lim, G.S., Balke, M. & Meier, R. (2011). Determining Species Boundaries in a World Full of
Rarity: Singletons, Species Delimitation Methods. Syst. Biol. 61(1): 165—169.
Lim, N.K.M., Tay, Y.C., Srivathsan, A., Tan, J.W.T., Kwik, J.T.B., Baloglu, B., Meier, R. & Yeo,
D.C.J. (2016). Next-generation freshwater bioassessment: eDNA metabarcoding with
a conserved metazoan primer reveals species-rich and reservoir-specific communities.
Royal So. Open Sci. 3: 160635.
Lim, W.H., Li, T.J. & Cai, Y. (2018). Terrestrial snails and slugs diversity in Nee Soon freshwater
swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 109-121.
Meier, R. (2008). DNA sequences in taxonomy - Opportunities and challenges. In: Wheeler,
Q.D. (ed) New Taxonomy, pp. 95-127. London: CRC Press.
Meier, R. & Dikow, T. (2004). Significance of specimen databases from taxonomic revisions
for estimating and mapping the global species diversity of invertebrates and repatriating
reliable specimen data. Conserv. Biol. 18: 478-488.
Meier, R., Shiyang, K., Vaidya, G. & Ng, P.K.L. (2006). DNA barcoding and taxonomy in
Diptera: A tale of high intraspecific variability and low identification success. Syst. Biol.
55: 715-728.
Meier, R., Zhang, G.Y. & Ali, F. (2008). The use of mean instead of smallest interspecific
distances exaggerates the size of the “barcoding gap” and leads to misidentification.
Syst. Biol. 57: 809-813.
Meier, R., Wong, W., Srivathsan, A. & Foo, M. (2016). $1 DNA barcodes for reconstructing
complex phenomes and finding rare species in specimen-rich samples. Cladistics 32:
100 - 110 .
Meyer, M. & Kircher, M. (2010). Illumina sequencing library preparation for highly multiplexed
target capture and sequencing. Cold Spring Harb. Protoc. 2010: 5448.
Ng, T.H., Tan, S.K., Wong, W.H., Meier, R., Chan, S.Y., Tan, H.H. & Yeo, D.C.J. (2016).
Molluscs for sale: assessment of freshwater gastropods and bivalves in the ornamental
pet trade. PLoS ONE 11: e0161130.
Gdegaard, F. (2000). How many species of arthropods? Erwin’s estimate revised. Biol. J. Linn.
Soc. 71: 583-597.
Pont, A.C. & Meier, R. (2002). The Sepsidae (Diptera) of Europe. Fauna Entomologica
Scandinavica, vol. 37. Leiden: Brill.
Rohner, P.T., Ang, Y., Lei, Z. & Meier, R. (2014). Genetic data confirm the species status of
Sepsis nigripes Meigen (Diptera: Sepsidae) and adds one species to the Alpine fauna
while questioning the synonymy of Sepsis Helvetica Munari. Invertebr. Syst. 28: 555-
563.
DNA barcoding for Nee Soon flora and fauna
173
Srivathsan, A. & Meier, R. (2012). On the inappropriate use of Kimura-2-parameter (K2P)
divergences in the DNA-barcoding literature. Cladistics 28: 190-194.
Srivathsan, A., Sha, J.C.M., Vogler, A.R & Meier, R. (2015). Comparing the effectiveness of
metagenomics and metabarcoding for diet analysis of a leaf-feeding monkey ( Pygathrix
nemaeus ). Mol. Ecol. Resour. 15: 250-261.
Srivathsan, A., Ang, A., Vogler, A.R & Meier, R. (2016). Fecal metagenomics for the
simultaneous assessment of diet, parasites, and population genetics of an understudied
primate. Front. Zoo!. 13: 1-13.
Straub, S.C.K., Parks, M., Weitemier, K., Fishbein, M., Cronn, R.C. & Liston, A. (2012).
Navigating the tip of the genomic iceberg: Next-generation sequencing for plant
systematics. Am. J. Bot. 99: 349-364.
Tan, D.S.H., Ang, Y., Lim, G.S., Ismail, M.R.B. & Meier, R. (2010). From ‘cryptic species’ to
integrative taxonomy: an iterative process involving DNA sequences, morphology, and
behaviour leads to the resurrection of Sepsis pyrrhosoma (Sepsidae: Diptera). Zool. Scr.
39: 51-61.
Walter, D.E. & Winterton, S. (2007). Keys and the crisis in taxonomy: Extinction or reinvention?
Annu. Rev. Entomol. 52: 193-208.
Wang W.Y., Srivathsan A., Foo M., Yamane S.K., Meier R. (2018). Sorting specimen-rich
invertebrate samples with cost-effective NGS barcodes: Validating a reverse workflow
for specimen processing. Mol. Ecol. Resour, https://doi.org/10.llll/1755-0998.12751
Wheeler, Q.D. & Meier, R. (2000). Species Concepts and Phylogenetic Theory. A Debate. New
York: Columbia University Press.
Winston, J.E. (1999). Describing Species: Practical Taxonomic Procedure for Biologists. New
York: Columbia University Press.
Wong, W.H., Tay, Y.C., Puniamoorthy, J., Balke, M., Cranston, PS. & Meier, R. (2014). ‘Direct
PCR’ optimization yields a rapid, cost-effective, nondestructive and efficient method for
obtaining DNA barcodes without DNA extraction. Mol. Ecol. Resour. 14: 1271-1280.
Yeo, D., Puniamoorthy, J., Ngiam R. W. J., Meier, R. (in press). Towards holomorphology in
entomology: rapid and cost-effective larval-adult matching using NGS barcodes. Sys.
Ent.
Zhang, J., Robert, K., Flouri, T. & Stamatakis, A. (2014). PEAR: a fast and accurate Illumina
Paired-End reAd mergeR. Bioinformatics 30(5): 614-620.
Gardens’ Bulletin Singapore 70 (Suppl. 1): 175-190. 2018
doi: 10.26492/gbs70(suppl.l).2018-09
175
Projected impacts of climate change on stream flow
and groundwater of Nee Soon freshwater
swamp forest, Singapore
Y. Sun 1 - 2 , D.E. Kim 1 , D. Wendi 1 , D.C.Doan, S.V.Raghavan 1 ,
Z. Jiang 1 & S.Y. Liong 1 -- 3
tropical Marine Science Institute, National University of Singapore,
18 Kent Ridge Road, 119227 Singapore
2 Willis Towers Watson, 51 Lime Street,
London, United Kingdom
3 Center for Environmental Sensing and Modeling, SMART,
1 Create Way, 138602 Singapore
tmslsy@nus.edu.sg
ABSTRACT. As Singapore’s only remaining patch of primary freshwater swamp forest, the
good management of the Nee Soon catchment is of utmost importance if a large proportion
of the flora and fauna in Singapore is to be conserved. An integrated eco-hydrological
model is developed for the area, with the objectives to numerically model the hydrological
variations, to assess the possible impacts of future climate change, and to facilitate future eco-
hydrological management. The numerical model considers the hydrological processes in a
holistic manner, including rainfall-runoff, evapotranspiration, the interaction between surface
water and groundwater, etc. The numerical model makes use of a combination of field survey
data and alternative remote sensing data. With climate projection inputs from the Regional
Climate Model (RCM), the numerical model is applied to run future scenarios to assess the
climate change impact. A few management strategies are considered if favourable hydrological
conditions are to be maintained for conserving the local ecosystem.
Keywords. Eco-hydrological model, eco-hydrology management, remote sensing, reservoirs
Introduction
Various studies have emphasised the significance of Nee Soon freshwater swamp forest
to habitat and species conservation in Singapore (Ng & Lim, 1992; Clews et al., 2018).
Nee Soon is the only remaining locality for hundreds of plant and animal species
that have been extirpated elsewhere in the country (Tan et al., 2008), supports habitat
specialists relying on acidic swamp conditions (Turner et al., 1996), contains several
global endemics (Ng & Lim, 1992), and continues to be a locality for discoveries of
new species and of species new to Singapore (e.g. Evenhuis & Grootaert, 2002). The
environment within the Nee Soon catchment depends critically on the local hydrology
(Clews et al., 2018). Changes in surface water and groundwater is likely to affect both
the flora and the fauna wherever they occur, while some species will likely prove to
be more vulnerable than others (Ho et al., 2018). Urbanisation and climate change is
176
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
also likely to affect the surface and groundwater of Nee Soon freshwater swamp forest,
both locally and in its surroundings.
A numerical model, which can anticipate the likely changes in surface and
groundwater, and their impacts on fauna and flora, is highly important to any area,
and to the Nee Soon catchment in particular, considering the national conservation
significance of this, the last pristine freshwater swamp forest in Singapore (Sun et
al., 2015). The Mike-SHE eco-hydrological modelling system, as a multi-physics
modelling package, is well suited to simulate integrated catchment hydrology (DHI,
2014). Mike-SHE simulates water flow over the entire land surface based on different
phases of the hydrological cycle from rainfall to river flow, via various flow processes,
such as overland flow, infiltration into soil, evapotranspiration, and groundwater flow.
It is thus an ideal tool for simulating the hydrology of wetlands. Such a modelling
system, however, requires model inputs. Essential input data include: topography,
geological coverage, soil properties, land use maps, hydro-meteorological data,
evapotranspiration information, vegetation distribution, etc. For the Nee Soon
freshwater swamp forest, the complex hydro-geological characteristics and the strict
requirements for conservation hinder the installation of monitoring stations to acquire
the necessary information. This study, therefore, adopted a combined approach to
develop the numerical model, which makes use of field survey data and the alternative
remote sensing data.
The numerical model is calibrated based on groundwater table and water level
measurements; it is then combined with the future projected rainfall from the Regional
Climate Model (RCM) to assess the hydrological impacts that might result from future
climate change. A few possible management strategies are suggested corresponding to
the severe drought and flood scenarios in order to maintain favourable hydrological
conditions for conserving the local ecosystem.
Modelling Scheme
Study area
Fig. 1 shows the geographical location of the Nee Soon freshwater swamp forest in
Singapore. Details are given by Chong et al. (2018), Clews et al. (2018), Davison et al.
(2018) and Nguyen et al. (2018).
With an estimated area of about 485 ha, the catchment of the freshwater swamp
forest covers the lower parts of shallow valleys with slow-flowing streams and a little
higher ground supporting dryland forest. The elevation of the Nee Soon catchment
ranges between 1 m to 80 m above mean sea level (MSL). The aquifer depth in the
Nee Soon catchment is from 20 m to 40 m, and the major soil type features silty sand
with a hydraulic conductivity of 4.05 x 10 5 m/s. The boundary of the study area on
the east is defined by catchment delineation based on the catchment topography, the
administrative boundary of the Central Catchment Nature Reserve, and the physical
barrier formed by a major highway. The boundary on the west and south is defined by
Stream flow, ground water and climate change
111
reservoirs, the inclusion of which, being an important water source for the catchment,
is crucial for the numerical surface water and groundwater simulations.
Model setup
Table 1 summarises the two scenario runs performed in this study. Scenario 1 serves
to search for the steady state condition of the water balance in the system, which
was modelled beginning with fully saturated conditions and run for many years with
assumed zero rainfall and evapotranspiration (ET). The steady state water distribution
resulting from Scenario 1 is then used as the initial condition for the real simulation in
Scenario 2, which considers the real operational reservoir water levels and observed
rainfall as the driving forces. For tropical swamp forests, evapotranspiration plays an
important role in the entire water balance and hydrological cycle. Scenario 2 utilises
a 2-layer water balance model to simulate the water loss from ET and the unsaturated
zone storage. The 2-layer water balance model is a simplified water balance method
which divides the unsaturated zone into a root zone and a zone below the root zone;
ET can be extracted from the root zone, w hil e it does not occur in the zone below (Yan
& Smith, 1994).
Setting up the 2-layer water balance model essentially requires three inputs, i.e.,
the root depth (RD), the leaf area index (LAI) and the reference ET. The root depth is
calculated based on a linear equation
RD = 0.07624 x DBH + 0.11185
where DBH is the diameter at breast height measured in 40 vegetation plots distributed
across the Nee Soon catchment (Chong et al., 2018). The average value is used as the
representative root depth in each plot; Thiessen polygon is then applied to interpolate
the point values into the entire study area.
The reference ET is the rate of ET with an unlimited amount of water from a
reference surface - a hypothetical grass reference crop with specific characteristics
(Allen et al., 1998). The reference ET data is obtained in this study from MOD 16
Global Terrestrial Evapotranspiration Data Set (Mu et al., 2013). The MOD16
project is part of the National Aeronautics and Space Administration (NASA)/Earth
Observing System (EOS) project to estimate global terrestrial evapotranspiration from
earth land surface by using satellite remote data. The MOD 16 dataset is derived from
an improved ET estimation algorithm with inputs including the GMAO and MODIS
land cover, LAI, FPAR and albedo data (Mu et al, 2011). Fig. 2 (a) shows samples of
the MOD 16 reference ET over the Nee Soon freshwater swamp forest. The missing
data accounts for about 1.2 km 2 within the Nee Soon catchment (less than one third
of the catchment area), which are supplemented with the interpolated values from
neighbouring cells. The 2012 reference ET ranges from 65 to 140 mm /month with an
average of 100 mm/month.
The Leaf Area Index is a dimensionless quantity that characterises plant
canopies. It is defined as the one-sided green leaf area per unit ground surface area in
178
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 1. Geographical location of the Nee Soon freshwater swamp forest in Singapore.
Table 1. Summary of essential information in mode setup.
Input type
Scenario 1
(Search for steady state
condition)
Scenario 2
(Real simulation)
Simulation duration
01/01/2001 -31/12/2012
01/01/2001 - present
Grid size
20x20 m
20x20 m
Soil type
Silty sand
Silty sand
(hydraulic conductivity =
(hydraulic conductivity =
4.05xl0 5 m/s)
4.05X10 5 m/s)
Rainfall
No rainfall
Observed rainfall
Evapotranspiration
Not triggered
2-layer water balance model
Initial condition
Fully saturated soil
Extracted from Scenario 1
(at 01/01/2001)
(condition at 31/12/2012)
Inner boundary condition
(reservoir)
Mean observed reservoir levels
Observed reservoir levels
Outer boundary condition
(land-land)
-5% gradient
-5% gradient
Outer boundary condition
(land-water)
Zero flux
Zero flux
Stream flow, ground water and climate change
179
Fig. 2. (a) Reference ET over Nee Soon freshwater swamp forest catchment in Dec 2012
(Source: MOD 16); (b) Leaf Area Index across Nee Soon freshwater swamp forest in Dec 2012
(Source: GLASS-MODIS).
broadleaf canopies (Watson, 1947). LAI can be determined directly through sample
measuring and indirectly such as hemispherical photography. This study acquires the
LAI information from the Global and Surface Satellite (GLASS)-MODIS LAI dataset,
a global LAI product released by the Center for Global Change Data Processing
and Analysis (CGCDPA) of Beijing Normal University (Liang & Xiao, 2012). The
GLASS-MODIS LAI dataset is retrieved using the general regression neural networks
(GRNNs) trained with the MODIS and CYCLOPES LAI products as well as the
reprocessed MODIS reflectance products (Xiao et al., 2013). Samples of the GLASS-
MODIS LAI over the Nee Soon catchment are plotted in Fig. 2 (b). Typical LAI values
range from 0 to 7, implying areas from no vegetation to dense canopy coverage.
Simulation results
Fig. 3 compares the simulated with the observed water depths at Upper and Mid stream
gauges, whereas Fig. 4 shows the comparison between the simulated and the observed
groundwater tables at stations DP4 and DP9. The numerical model simulates the water
depth reasonably well, with root mean square error (RMSE) respectively being 0.11
m and 0.17 m for these two stations. Both DP4 and DP9 are located upstream, where
the water table varies within 1 m below the ground surface. The numerical simulation
successfully captures the rising and falling trends within the series of observations,
producing insignificant model errors (RMSE respectively being 0.08 m and 0.11 m).
The model errors are mainly caused by the uncertainty in the reference level arising
from the smoothing effect of the 20 m grid.
180
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
15 / 05/2013
15 / 09/2013
15 / 01/2014
Date
15 / 05/2014
15 / 09/2014
250
200
150
S’
£
E
cc
50
B
Fig. 3. Simulated vs. observed water depths (a: Upper stream gauge; b: Mid stream gauge).
Model Application
Impacts of climate change
The numerical model, as shown in the previous section, having been calibrated based
on field measurements, is applied to assess the climate change impacts by simulating
future scenarios. The future scenarios, as defined in Table 2, combine the effect of two
factors: reservoir level and rainfall. Reservoir level is categorised as low, medium or
high, respectively represented by bottom, mean and top operating levels as managed
by the responsible government agency. Rainfall scenarios comprise no rainfall, low,
medium and high rainfall conditions. Future rainfall is projected from the numerical
climate model formulated as
Stream flow, ground water and climate change
181
□j
03
■os
£
£
4 -i *
i
m
3
>i s
-13
-2 8
15/02/2012 15/06/2012 15/1Q/2Q12 16/02/2013 15/06/2013 15/10/2013 15/02/2014 15/06/2014 15/10/2014
Oslo
- 1
-- —
-
Rainfall
—— -Obifrvfd
1 iiil J1 1 ■ i i L In
■» -illi ll 1 .lJll.iUjliFI.id li ii l _ jIJLjIIl J
lliUlll. Mil lIJi 11 [ill iiiiilL 1„
ii Jiid ii
ill i
1 till ii 1 !iiJ 1 !if
l iLkM JihI.I. jLaiJLl J. Jitw.
260
200
SO
0
A
0.2 250
Cite
B
Fig. 4. Simulated vs. observed groundwater tables (a: Piezometer DP4; b: Piezometer DP9).
Future Rainfall = Current Rainfall x Change Factor
where Change Factor is calculated based on the climate projections simulated from
the Regional Climate Model - WRF (Weather Research and Forecasting model, http://
www.wrf-model.org; Liong & Raghavan, 2014). The matrix of three reservoir levels
and four rainfall conditions results in a total of 12 simulated future scenarios.
Fig. 5 and Fig. 6 respectively present the simulated groundwater table maps
and surface water extent maps after 5 years corresponding to the management
scenarios. Scenario 11, based on projected medium rainfall and high reservoir level,
unsurprisingly is the most similar to the current conditions, due to the similar forcing
resulting from (1) similar rainfall amount, and (2) the reservoir levels being kept close
to their maximum capacity.
182
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 2. Scenarios for future eco-hydrology management.
Reservoir
Level
No.
Rainfall
Low Medium
High
Low
1
2
4
4
Medium
5
6
7
8
High
9
10
11
12
Table 3. Proposed drought mitigation management strategies for Scenario 9.
Drought
Mitigation
Management
Strategies
Location
and
Symbols
Discharge
rate at each
point source
(m3/s)
Volume at
each point
source
(m3/day)
Total
Discharge
Volume (m3/
day)
Proposed
System
9A
®in Fig. 7
0.02
1,728
26,000
Pump + Pipe
9B
®in Fig. 7
0.04
3,456
52,000
Pump + Pipe
9C
Ain Fig. 7
0.02
1,728
26,000
Pipe
9D
Ain Fig. 7
0.01
864
13,000
Pipe
Scenario 1 would have been the obvious severe drought case to be studied and
a series of drought mitigation managements would be recommended accordingly.
However, we consider Scenario 9 more appropriate in the Singapore context, as
Singapore is unique in its number of desalination plants and recycled water plants.
During the unusually long dry periods in early 2014 and 2015, many of the desalination
and recycled water plants were operating at their full capacity. With this in mind, in
this study the focus is placed on Scenario 9 as the severe drought scenario. Scenario
12 is selected as the severe flood case; it is analysed and a series of flood mitigation
management strategies is proposed.
Management strategies
Severe drought case
Based on the climate projections for Singapore, the unprecedented five consecutive
severe drought years that occurred in California (Reid, 2015), and the severe droughts
experienced in other places such as North Korea, Brazil and South Africa, Scenario
9 is selected for the severe drought case study; the drought mitigation management
strategies for the Nee Soon freshwater swamp forest are then proposed. Table 3
summarises the proposed drought mitigation management strategies to tackle the dry
situation in Scenario 9. Point sources at strategically selected locations, as shown in
Stream flow, ground water and climate change
183
No Rainfall
Low Rainfall
Medium Rainfall
High Rainfall
Fig. 5. Simulated groundwater table maps, after 5 years, for all 12 Scenarios.
Current Condition
f-tSSF Boundary | 1
ftMetwr H Surtittr WMW
Z
i
CC
a
D
c
i
No Rainfall
Low Rainfall
Medium Rainfall
High Rainfall
Fig. 6. Simulated surface water depth maps, after 5 years, for all 12 Scenarios.
184
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
4
Legend
■-- Stream
9 Po«*t Source ( 9 A 96 )
A Po.fftSou.reo ( 9 C. 9 DI
Elevation
I j 0 9 - is
16 - 26
■§ 29-44
IB 44 - f>2
■ 92-Si
A
0 500 1,400
'-CTftrf?-'
Fig. 7. Locations of the proposed point sources for drought mitigation management strategies
of Scenario 9.
Fig. 7, and discharge rates (volume) are inserted and simulated in the numerical model.
Fig. 8 and Fig. 9 respectively present the simulated groundwater table maps and
surface water maps in their current condition, after 5 years of Scenario 9, and after 5
years of Drought Mitigation Management Strategies 9A, 9B, 9C and 9D. The point
sources of Strategies 9C and 9D are located in the catchment downstream (nearer to
the stream) as compared to Strategies 9A and 9B. Therefore, the piped-in water in
Strategies 9C and 9D has more direct effect in nourishing the swampy area near the
stream than in Strategies 9A and 9B.
Fig. 10 illustrates a suggested drought mitigation system for management of the
point source strategies. A pump and pipe system would be required for Strategies 9 A
and 9B, whereas Strategies 9C and 9D would only need a pipe system due to the lower
elevation of the point sources than the maximum operating reservoir water levels
(thus, a gravity flow system). Despite providing lower coverage, a pipe system would
not only incur lower cost in construction and management, but would also require less
water consumption compared to a pump and pipe system.
Severe flood case
Flooding often results in poor soil aeration, polarisation of soil pH, accumulation of
organic matters, unfavourable sedimentation and/or erosion, etc. It, therefore, hampers
Stream flow, ground water and climate change
185
Drought
Drought
Mitigation
Strategy
$9C
5-vear
Drought
S9
S9B
S9D
Fig. 8. Simulated groundwater table maps: Current vs. Scenarios 9, 9A, 9B, 9C, 9D.
186
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Fig. 9. Simulated surface water maps: Current vs. Scenarios 9, 9A, 9B, 9C, 9D.
Stream flow, ground water and climate change
187
Pump + Pipe
System
Pipe System
(Gravity Flow)
Fig.10. Proposed drought mitigation management systems: Pump+Pipe System vs. Pipe
System.
Fig. 11. Simulated surface water maps: Current vs Scenario 12 (after 2 years).
188
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 4. Proposed flood mitigation management system: retention ponds.
Flood
cluster
Map
Flooded
area (m 2 )
Flood vol¬
ume (m 3 )
Average
depth (m)
Proposed
dimension
1
2
3
9,200
11,200
12,800
198.34
127.77
859.98
0.022
0.011
0.067
iBm
12m
Reduction 97.5 98.7 93.0
of flooded
area (%)
the growth of trees and even leads to death of the root system. Fig. 11 compares the
current surface water area extent with the projected future surface water area extent
(after 2 years) resulting from a model simulation of Scenario 12. The excess surface
water area, after 2 years, is 36.72 ha with an excess water volume of 73,606 m 3 . To
mitigate the extent of flooding, and also to promote habitats for fauna, retention ponds
with a maximum depth of 1 m, various surface areas and different water volumes are
suggested.
To demonstrate the flood mitigation management approach, we focus on three
flooded areas as circled in Fig. 11. Three retention ponds in their respective locations
are indicated. Detailed information on the flooded areas and retention ponds is
summarised in Table 4. The suggested retention ponds appear to reduce the flooding
area by more than 90%. As mentioned above, the retention ponds could also promote
habitats for fauna if properly designed and managed.
Conclusions
An integrated eco-hydrological model for the Nee Soon freshwater swamp forest has
been developed in this study. The surveyed GIS data, including the stream network,
the cross-sections and the Digital Elevation Model (DEM), have been incorporated in
Stream flow, ground water and climate change
189
the model setup. The spatial and temporal variations of LAI and reference ET retrieved
from remote sensing data, i.e. the reference ET from MODIS and LAI from GLASS-
MODIS, with the aid of the surveyed RD, are used to establish a two-layer water
balance model to account for the water loss from evapotranspiration and the amount
of water recharging to the saturated zone. In addition, the field measurements from
piezometers and stream sondes have been processed and integrated to calibrate and
validate the model parameters.
Twelve scenarios were introduced, being the combinations of various reservoir
operating levels and the projected future rainfall resulting from climate change study.
Despite rainfall appearing to be the most influential factor for the overall catchment
water availability, i.e., the spatial average over the catchment, it is interesting to
observe the different contributing factors of both rainfall and reservoir water at sub¬
catchment levels. The effects of the two inputs differ depending on the locations as
can be seen from the hydrological maps. This spatial distribution information is of
importance should eco-hydrological management be approached at sub-catchment
levels or spatially distributed.
Several possible management strategies are put forth to mitigate severe
drought and flood resulting from the projected climate change impacts as simulated
in Scenarios 9 and 12. These have yet to be evaluated in terms of cost, engineering
feasibility, and biological impacts. Introducing water sources (point sources) in the
catchment upstream is a potential strategy to mitigate future drought. Retention ponds
could be a simple and effective solution in mitigating flooding and simultaneously
promoting habitat for aquatic fauna. Discussion of these two possibilities does not
represent a commitment to carry them out, as they must be considered in relation to
recommendations from other teams in the study of Nee Soon freshwater swamp forest.
ACKNOWLEDGEMENTS. This study forms part of the research project “Nee Soon Swamp
Forest Biodiversity and Hydrology Baseline Studies—Phase 2” funded by National Parks Board
(NParks), Singapore. The authors are also grateful for access to data generously provided by
the Public Utilities Board (PUB), Singapore, that helped to make this study possible.
References
Allen, R.G., Pereiro, L.S., Raes, D. & Smith, M. (1998). Crop Evapotranspiration: Guidelines
for Computing Crop Water Requirements. FAO Irrigation and Drainage Paper 56.
Rome: FAO.
Chong, K.Y., Lim, R.C.J., Loh, J.W., Neo, L., Seah, W.W., Tan, S.Y. & Tan, H.T.W. (2018).
Rediscoveries, new records, and the floristic value of the Nee Soon freshwater swamp
forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 49-69.
Clews, E., Corlett, R.T., Ho, J.K.I., Kim, D.E., Koh, C.Y., Liong, S.Y., Meier, R., Memory, A.,
Ramchunder, S.J., Sin, T.M., Siow, H.J.M.P, Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W.,
Theng, M.T.Y., Wasson, R.J., Yeo, D.C.J. & Ziegler, A.D. (2018). The biological,
ecological and conservation significance of freshwater swamp forest in Singapore.
Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
190
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Davison, G.W.H., Cai, Y., Li, T.J. & Lim, W.H. (2018). Integrated research, conservation
and management of Nee Soon Freshwater Swamp Forest, Singapore: hydrology and
biodiversity. Gard. Bull. Singapore 70 (Suppl. 1): 1-7.
DHI (Danish Hydraulic Institute) (2014). MIKE SHE, User Manual, Volume 1: User Guide.
Hprsholm: Danish Hydraulic Institute.
Evenhuis, N.L. & Grootaert, P. (2002). Annotated checklist of the Dolichopodidae (Diptera)
of Singapore, with new records and descriptions of new species. Raffles B. Zool. 50:
301-316.
Ho, J.K.I., Quek, R.F., Ramchunder, S.J., Memory, A., Theng, M.T.Y., Yeo, D.C.J. & Clews,
E. (2018). Aquatic macroinvertebrate richness, abundance and distribution in the Nee
Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 71-108.
Liang, S. & Xiao, Z. (2012). Global Land Surface Products: Leaf Area Index Product
Data Collection (1985-2010). Data set. Beijing Normal University, doi: 10.6050/
glass863.3004.db.
Liong, S.Y. & Raghavan, V.S. (2014). Quarterly Reports on High Resolution Regional Climate
Change Impact Study, Vol I, as a part of the Coastal Inundation Risk Map Study for
Singapore. Submitted to the Building and Construction Authority, Govt, of Singapore.
Mu, Q., Zhao, M. & Running, S.W. (2011). Improvements to a MODIS global terrestrial
evapotranspiration algorithm. Remote Sens. Environ. 115: 1781-1800.
Mu, Q., Zhao, M. & Running, S.W. (2013). Algorithm Theoretical Basis Document: MODIS
Global Terrestrial Evapotranspiration (ET) Product (NASA MOD 16 A2/A3) Collection
5. NASA Headquarters.
Ng, P.K.L. & Lim, K.K.P. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquatic Conserv. 2(3): 255-266.
Nguyen, C.T.T., Wasson, R.J. & Ziegler, A.D. (2018). The hydro-geomorphic status of the
Nee Soon freshwater swamp forest catchment of Singapore. Gard. Bull. Singapore 70
(Suppl. 1): 33-48.
Reid, R.L. (2015). Engineering for every drop. Civil Engineering - the Magazine of the
American Society of Civil Engineers 85(10): 46-55.
Sun, Y., Wendi, D., Kim, D. E. & Liong, S. Y. (2015). Technical Note: Application of artificial
neural networks in groundwater table forecasting - a case study in Singapore swamp
forest. Hydrol. Earth Syst. Sci. Discuss. 12: 9317-9336.
Tan, H.T.W., Tan, K.-X., Ali bin Ibrahim, Chew, P.T., Chua, K.S., Duistermaat, H., Ganesan,
S.K., Goh, M.W.K., Gwee, A.T., Kiew, R., Lee, S.M.L., Leong, P., Lim, J., Lok,
A.F.S.L., Loo, A.H.B., Lum, S.K.Y., Morgany Thangavelu, Saifuddin bin Suran, Sim,
S., Samsuri Ahmad, Wee, Y.C., Yap, K.F., Yeo, C.K. & Yong, J.W.H. (2008). Seed
plants. In: Davison, G.W.H., Ng, P.K.L. & Ho, H.C. (eds) The Singapore Red Data
Book: Threatened plants and animals of Singapore, 2nd ed., pp. 213-244. Singapore:
The Nature Society (Singapore).
Turner, I.M., Boo, C.M., Wong, Y.K., Chew, P.T. & Ali bin Ibrahim (1996). Freshwater swamp
forest in Singapore, with particular reference to that found around the Nee Soon firing
ranges. Gard. Bull. Singapore 48: 129-157.
Watson, D.J. (1947). Comparative physiological studies on the growth of field crops: I.
Variation in net assimilation rate and leaf area between species and varieties, and within
and between years. Ann. Bot. 11: 41-76.
Xiao, Z., Liang S., Wang, J., Chen, P., Yin, X., Zhang, L. & Song, J. (2013). Use of general
regression neural networks for generating the GLASS leaf area index product from time
series MODIS surface reflectance. IEEE Trans. Geosci. Remote Sens. 52(1): 209-223.
Yan, J. & Smith, K.R. (1994). Simulation of integrated surface water and ground water systems
- model formulation. Water Re sour. Bull. 30(5): 1-12.
Gardens’ Bulletin Singapore 70 (Suppl. 1): 191-217. 2018
doi: 10.26492/gbs70(suppl.l).2018-10
191
Conservation outputs and recommendations for Nee Soon
freshwater swamp forest, Singapore
Y. Cai 1 , G.W.H. Davison 1 , L. Chan 12 & S.Y. Liong 3
‘National Biodiversity Centre, National Parks Board,
1 Cluny Road, 259569 Singapore
cai_yixiong@nparks.gov.sg
international Biodiversity Conservation Division, National Parks Board,
1 Cluny Road, 259569 Singapore
3 Tropical Marine Science Institute, National University of Singapore,
18 Kent Ridge Road, 119227 Singapore
ABSTRACT. The current paper acts as a summary to the “Nee Soon Swamp Forest biodiversity
and hydrology baseline studies project”, including results published previously and the results
from papers of the current volume. Overall, flora and fauna surveys indicate healthy and
diverse plant, fish and aquatic macro invertebrate communities in Nee Soon freshwater swamp
forest. There are some concerns over terrestrial and aquatic alien invasive species, loss of big
emergent trees, small population sizes and viability of various native species, and the uncertain
outcomes of changes in water quality and quantity. The findings inform management that Nee
Soon freshwater swamp forest is especially vulnerable to changes in hydrology and there is
much dependency on precipitation for its water budget. Projected climate change effects on
precipitation and statistical analyses of biotic responses to hydrology clearly define drought
as a major, perhaps the foremost, source of vulnerability to the ecosystem functioning of Nee
Soon freshwater swamp forest. Potential management solutions are suggested to address five
issues of concern for the forest: hydrological integrity, erosion and sedimentation, ecological
integrity, the impact of the spillway, and impacts of construction and development.
Keywords . Biodiversity management, hydrological management, monitoring, wetlands
Introduction
Nee Soon constitutes Singapore’s last remaining patch of primary freshwater swamp
forest. From the viewpoint of ecosystem diversity alone, this makes the conservation
of the Nee Soon freshwater swamp forest a priority (Clews et al., 2018). The number
of plant and animal taxa currently found nowhere else in Singapore but Nee Soon
only emphasises its conservation value. Given that Nee Soon freshwater swamp forest
houses a large proportion of Singapore’s overall flora and fauna, conservation of this
habitat undoubtedly has larger-scale, positive effects for biodiversity conservation
in Singapore as a whole (Ng & Lim, 1992; Turner et al., 1996), addressing the
conservation of biodiversity from species to landscape scales.
Owing to the nature of its ecosystem and drainage, the Nee Soon freshwater
swamp forest is extremely sensitive to external disturbances (Ng & Lim, 1992).
Furthermore, many of the species found here are highly specialised and, thus,
192
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
destruction of the Nee Soon freshwater swamp forest and its surrounding areas would
pose a great threat to these unique groups of species. Therefore, it is important to
maintain Nee Soon freshwater swamp forest in approximately its current state, as well
as ensuring that it is not affected adversely by development and other environmental
pressures.
The Nee Soon hydrology and biodiversity project was a response to the need
for better understanding of the forest ecosystem and its dynamic processes, as a basis
for management and monitoring (Davison et al., 2018). A detailed but still in many
respects preliminary review of the results has been given in the accompanying papers
in this same volume (Cai et al., 2018; Chong et al., 2018; Clews et al., 2018; Ho et al.,
2018; Kutty et al., 2018; Lim et al., 2018; Nguyen et al., 2018; Sun et al., 2018).
Project Aims
The project aimed to undertake field and modelling investigations that were required in
order to create maps and collate information for the development of eco-hydrological
models of Nee Soon freshwater swamp forest. The project was carried out in two
phases, from January 2011 to March 2012, and from February 2013 to August 2016.
The detailed aims of each phase are listed by Davison et al. (2018).
Table 1 summarises the achievements of the project and these fully covered the
listed aims.
Results
Mapping and geodatabase
Phase 1 of the “Nee Soon Swamp Forest hydrology and biodiversity project” generated
important baseline datasets on the hydrology, geology, topography and flora of the
area and clearly set out the research path for Phase 2. Key focus areas of relevance
to mapping and imagery included: (i) the development of more refined hydrological
models for systems understanding and for utilisation in scenario modelling; and (ii)
establishment of current ecological status/condition across a number of biotic and
abiotic components. The geospatial team had a major underpinning role in Phase 2
of the project, achieved through two broad tasks, the collection and analysis of spatial
information, and the collation of all such information into a geodatabase.
The environment of Nee Soon freshwater swamp forest represented a challenge
to conventional topography due to limitations in access. Moreover, line-of-sight was
greatly reduced in areas of higher tree density, constraining the efficacy of direct
topographic surveys. Through application of remote sensing techniques, direct
topography effort was optimised, focusing on acquiring high-resolution stream data
and ground control points for validation of remote-sensing models. Photogramme try
of remote sensing imagery involves the use of digital image data in conjunction with
automatic image matching techniques to produce Digital Elevation Models (DEMs)
Conservation recommendations for Nee Soon
193
Table 1. Achievements in Phase 2 of the Nee Soon hydrology and biodiversity project.
#
1
Aims
Achievements
Establish the status of 1.
Nee Soon freshwater
swamp forest in terms
of vegetation hydrology
and aquatic biodiversity
Literature Review of freshwater swamp forest literature and
research on the Nee Soon freshwater swamp forest updated (Tan et
al., 2013; Clews et al., 2018); new results derived from this project
presented on vegetation (Chong et al., 2018), fauna (Cai et al.,
2016, 2018; Li et al., 2016; Ho et al., 2018; Lim et al., 2018) and
cryogenics (Kutty et al., 2018).
Assessment of geomorphology in relation to catchment
hydrological function (Nguyen et al., 2018).
Consolidation and documentation of all spatial data into a geo¬
database (Davison et al., 2018; Sun et al., 2018).
2
Identify periodic flux 1.
in hydrology and key
components of the 2
aquatic biodiversity
Assessment of variation on intra- and inter-annual hydrographs
(project reports).
Stream hydrological regimes examined as well as faunal responses
to hydrology (project reports and Ho et al., 2018).
3
Develop more refined
models that can confirm
current conditions
(water balance, nutrient
balance, acid flux,
faunal distribution) and
then test-trial various
management scenarios
1. Developed an integrated eco-hydrological model using Mike-SHE
(Sun et al., 2018);
2. Simulated and assessed twelve future scenarios (Sun et al., 2018).
3. Developed conceptual models of erosion and elemental
redistribution in the catchment as related to hydrological and
geomorphological processes (project reports; Sun et al., 2018)
4. Elucidated faunal response models (project reports and Ho et al.,
2018).
4
Identify and assess
root causes of impacts,
potential issues that may
threaten the hydrological
and ecological integrity
of the swamp, and
management elements
to be addressed
1. Investigated, using numerical model, effects of rainfall and
reservoir operating levels on the spatial distribution of surface and
ground-water.
2. Impacts of climate change.
3. Investigated the responses of seedlings of six tree species common
in Nee Soon freshwater swamp forest to changes in soil and
hydrology and tree coring for measurements (Neo et al., 2017).
4. Investigated spatio-temporal variation in faunal communities
in association with physicochemical and hydrological data to
identify key issues likely to affect in-stream fauna, Management
recommendations to monitor and maintain the ecological integrity
of the forest stream faunal communities.
5. A synthesis of long-term (historic) and more recent/current
environmental impacts on the Nee Soon freshwater swamp forest
catchment.
194
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Table 1. Continuation.
#
5
Aims
Achievements
Agree on the necessary 1.
recommendations for
possible mitigation
of long-term negative
impacts ^
3.
The Management Recommendations Chapter of the final report
synthesises and consolidates recommendations for management
of the swamp forest drawing on detail provided in preceeding
chapters.
Proposed mitigation management for extreme flooding and drought
Proposed to cany out propagation of rare plant species in Nee
Soon and use the propagules for restoration of potential freshwater
swamp forest sites elsewhere to mitigate the risks of extinction of
these native species over the longer term
Recommendations relevant to aquatic fauna are with respect to i)
extreme flooding and drought; and it) other potential management
issues such as the introduction of non-native taxa and erosion.
6
Establish a viable, 1.
long-term monitoring ^
programme and
sampling protocols
to ensure continued
protection and good
management
Recommendations for ecological monitoring of streams.
Set up and handed over 40 vegetation plots for long-term
monitoring of tree growth, recruitment, and mortality in the wet
and dry areas of Nee Soon. The GPS coordinates and tracks to the
40 plots have also been prepared.
7
Train agency staff in
modelling, sampling
methods and tools for
monitoring
1. A training course for application of Eco-hydrological Modelling
developed from this study was held on 22nd July 2016.
2. On the 11th June 2015, a workshop on the identification of fish,
decapod crustaceans and macroinvertebrates from the Nee Soon
freshwater swamp forest was held. Several theory and practical
sessions for aquatic faunal identification purposes were facilitated
by both NUS (TMSI, DBS, and LKCNHM) and NParks staff.
3. Sampling methodology for faunal collection was also introduced in
the introductory field resource, “A Guide to the Freshwater Fauna
of Nee Soon swamp forest” (Ho et al., 2016) which was reviewed
during the workshop and revised in response to participants’
feedback.
8
4.
5.
Deliver workshops
on development and
interpretation of the
models’ outputs
1 .
2 .
3.
The e-book version of the guidebook is available at http://emid.
nus. edu. s g/ebooks/web/vie wer. j sp
In phase 1 of the Nee Soon freshwater swamp forest project, a
workshop was held to train NParks staff on identification of the
common plants of Nee Soon freshwater swamp forest. One book
was also published to facilitate the identification of these plants.
Conducted workshop for non-governmental organisations and
individuals at Hort Park, Singapore, December 2016
Seminar provided to University of Warwick, UK, March 2017
Other workshops to be provided as required
Conservation recommendations for Nee Soon
195
Table 1. Continuation.
# Aims
Achievements
9 Publish work on swamp
forest ecology and the
development of eco-
hydrological models
in international, peer-
reviewed scientific
journals
So far, one guide book (Ho et al., 2016), one book chapter (Cai et
al., 2016), nine journal papers (Neo et al., 2016, 2017; Sun et al.,
2015, 2016; Chong et al., 2016; Li et al., 2016; Lim et al., 2016;
Tan et al., 2016; Wendi et al., 2016) and 18 conference papers
have been published. Another eight papers have been accepted for
publication (Cat et al., 2018; Chong et al., 2018; Clews et al., 2018;
Ho et al., 2018; Kutty et al., 2018; Lim et al., 2018; Nguyen et al.,
2018; Sun et al., 2018).
and ortho-images to act either as direct input to a GIS system or as the basis for the
production of hardcopy image maps or line maps.
Ecohydraulic models require approximation of evapotranspiration (ET), which
is difficult to measure in the field. For the purposes of the Nee Soon freshwater
swamp forest study, ET data from MODIS satellite were used. These satellite data
compared reasonably well with Leaf Area Index (LAI), which is the ratio of total
upper leaf surface of vegetation divided by the surface area of the land on which the
vegetation occurs. Beyond simple applications for theoretical purposes, hydrological
and hydrodynamic models operate on spatial grids, requiring the different types of
information to be spatially compatible. Under the broad theme of remote sensing and
mapping, the Geospatial team managed and quality-assured the efforts of the direct
topographic survey, conducted remote sensing for additional data acquisition, and
converted all data into spatial formats suitable for utilisation by the hydrological
modelling team.
Surveyors were engaged to conduct direct topographic surveys in Nee Soon
freshwater swamp forest. Following quality assurance checks, the field data were
converted into shape files and compiled for upload to the geodatabase. The data were
then processed to generate i) a 3D drainage network for Nee Soon freshwater swamp
forest; and ii) 1,647 cross-sectional profiles at 5 m spacing along the network. These
outputs were provided to the hydrological modelling team.
A file geodatabase was established using ESREs ArcGIS Platform. The
Geodatabase created forms the Nee Soon Freshwater Swamp - Geographical
Information System (GIS) which helps to organise, manage and analyse data spatially
(geo-referenced) and non-spatially pertaining to the study.
In order to establish the geodatabase, the version of ArcGIS utilised was ArcGIS
10.1 for desktops. ArcGIS is a proprietary GIS suite of systems developed by ESRI
(Environmental Systems Research Institute). The ArcGIS suite’s components include
ArcMap, ArcCatalog and ArcToolbox, which allow users to author, analyse, map,
manage, share, and publish geographic information. ArcGIS works with geographic
information managed in geodatabases as well as in numerous GIS file formats. The
geodatabase is the native data structure for ArcGIS and is the primary data format
196
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
used for editing and data management. Geodatabases work across a range of database
management system (DBMS) architectures and file systems, come in many sizes, and
have varying numbers of users. They can scale from small, single-user databases built
on files up to larger workgroup, department, and enterprise geodatabases accessed by
many users.
The Nee Soon file geodatabase is a single centralised location with metadata
that gives information on the ownership and authorisation of data. Currently, the file
geodatabase contains 68 layers of both raster and feature layers. ArcMap related layers
are categorised as groups shown in Table 2. All the digital information is retained
by the National Biodiversity Centre, National Parks Board, Singapore, and (with
appropriate access restrictions and provisos) on the web-based information platform
named BIOME.
Field hydrology and geomorphology
Effective conservation of the Nee Soon Catchment will require strategies informed by
the dynamic baseline of the system, as well as the sources of current anthropogenic
disturbance. While conservation efforts may be primarily concerned with preserving
the unique ecosystems and highly diverse floral and faunal communities, these systems
are intrinsically linked to the geomorphic and hydrological stability of the catchment.
Thus, the underlying hydrology and sedimentary processes must be considered
when implementing management efforts. This consideration was fundamental to the
rationale, aims, design, conduct and deliverables of the project, linking all the work by
the various teams.
Soil erosion reduces the water holding capacity and results in more rapid water
runoff, causing soil organic matter and nutrients to be transported downslope. This
process can greatly affect species diversity of plants, animals, and microbes by rapidly
exporting water, nutrients, and other biological resources out of the biological system
(Zuzao & Pleguezuelo, 2008). Erosional processes are aggravated by a reduction in
forestland cover (Rahman et al., 1991); the canopy of vegetation stabilises the hillslopes
through the adsorption of rainfall and binding root systems drawing moisture from the
groundwater. This protects the ground surface by shielding it from rain impact as well
as removing water from the soils and reducing the frequency of soil saturation and
subsequent surface runoff. Landscape modifications to facilitate farming, irrigation,
and drainage has altered stream flow and subsequently contributed to sedimentation in
low lying areas (Schumm, 1973).
Stream channel bunding in the lower catchment has significantly deepened the
valley floor, preventing water from spreading towards the expressway. This enhances
erosion along the stream channel and limits the development of the swamp forest
area as well as likely increasing the sediment load to the Lower Seletar Reservoir.
One potential strategy for mitigating these enhanced erosional processes is to fill
in the channel to restore the hydrological functioning of the swamp. However, this
intervention may affect fish communities that have since developed in the channel.
Conservation recommendations for Nee Soon
197
Table 2. Group Layers in Nee Soon freshwater swamp forest file geodatabase
Group Name
Layers
Gauges
Piezometer
Sondes
Water sample points
Hydro Vegetation station
Stream network
Streamline
Stream centrepoints
Stream photo
Profile 2014
Drain network 3D 2014
Cross section line 2014
Cross section point 2014
Restricted area
Firing range
Restricted Area
Vegetation
Swampforest extent (GEO Team)
Swampforest extent (Modelling Team)
Tree photo
Tree 2014
Tree DBH
Tree plot
Vegetation Plots
Leaf Area Index (Hi Resolution)
Leaf Area Index
Fauna
Fauna Plots
FE_SamplingPts
Topographical features
Road 2014
Bollard 2014
Concrete Lining 2014
Corner of Column 2014
Drain Features 2014
Fence 2014
Filter 2014
Guardrail 2014
Inspection Chamber 2014
Line Work 2014
Sump 2014
Wooden Pole 2014
Wall 2014
Water Valve 2014
Water Pipeline 2014
Near Bamboo Veg 2014
Nee Soon boundary
Nee Soon Catchment boundary
Catchment Phase2 (old)
DEM
NS_DTM_ 1 m_Ver2 (new)
Topo drainage DEM
ns2 DEM (older)
198
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Other causes for enhanced erosion rates are the presence of trails and traffic of
trekkers, bikers, and hunters who damage tracks and bridges. The clearing of vegetation
and increased activity along this area increases the soil mobility and transport of
sediments. Further restricting visitors to the catchment and repairing damage to the
tracks and bridges would help serve to stabilise these processes. Forest recovery and
replanting of endemic floral families would aid in not only the ecological resurgence
of the reserve, but also the retention of water, sediments, and nutrients in the system.
The excavation of a soil pit in the upper catchment uncovered ancient charcoal
remains, indicating clear evidence of past fires (Nguyen et al., 2018). Fires along
the hillslopes and dryland forests are not only capable of removing large swathes of
primary and old secondary forest, but may expose the slopes to significant erosion
during subsequent precipitation and inundation. The apparently increasing severity of
El Nino effects (Sun et al., 2018), the spread of development and human activities close
to the edges of Nee Soon freshwater swamp forest, and other human pressures within
the nature reserve may all increase the likelihood of fires. Therefore, the catchment
must be readily accessible for fire fighters appropriately trained for operation in a
nature reserve. The specialised techniques required would differ significantly from fire
fighting in non-forested areas. It is unlikely but conceivable that some plant species
require fire for germination and the relative importance of hillslope stability and
ecological succession must be considered.
The construction of the firing range, golf course, and expressway within or close
to the Nee Soon freshwater swamp forest may be affecting the water and soil quality of
the stream channels (Nguyen et al., 2018; Sun et al., 2018). Analysis of soils for heavy
metals throughout the reserve found that soils in the lower catchment are significantly
more enriched than those in the upper catchment. Due to the compositional homogeneity
of soils in the enriched and normal zones, it has been concluded that some components
of the heavy metal enrichments may be a result of such nearby developments (Nguyen
et al., 2018).
It has been reported that the water supply pipeline cutting across the catchment
from the Upper Seletar Reservoir to the Lower Pierce Reservoir did experience
early problems with line breakages due to defective operation of the pump pipeline
and contributed to significant erosion in the lower catchment (Murphy, 1997). It is
important to ensure that the pipeline does not leak significantly and that the bund along
its side does not erode into the swamp.
Finally, the dam at the mouth of the Upper Seletar outlet when opened has
been observed causing regular back-flooding into the swamp flushing the system with
reservoir water, and facilitating the influx of alien aquatic vertebrates and invertebrates.
Strategies to prevent back-flooding need to be investigated.
Friess et al. (2015) estimated the entire island of Singapore had an average of 60
Mg C , which is more than the mean stocks held by 100 other cities around the world
(Dobbs et al., 2014). Considering the urbanised state of the island, the natural areas
disproportionately hold the majority of the carbon stocks. Sites like the Nee Soon
freshwater swamp forest should thus continue to be protected and managed carefully
in order to maximise its carbon sequestration potential.
Conservation recommendations for Nee Soon
199
Vegetation ecology
Although the field study found that soil conditions appeared to be more important
than the presence of open water in structuring tree communities, experiments with
seedlings showed that for three of the five species, flooding had a significant effect
on growth responses but not the soil type. Taken together, these results suggest that
flooding leads to the formation of the tree community structure and the accompanying
soil properties in the freshwater swamp forest over time. Seedlings growing in the
swamp forest substrate are likely to be more resilient against short-term droughts.
Many nationally uncommon and rare plant species are restricted to the swampy
parts of the Nee Soon catchment. The catchment is also a hotspot of new plant
records and rediscoveries of species that have previously been presumed to be extinct
in Singapore. The Nee Soon freshwater swamp forest is therefore of high floristic
conservation value locally (Chong et al., 2018; Clews et al., 2018).
The vegetation ecology team predicts that non-swamp plant species are more
vulnerable than swamp species to anticipated changes in the hydrology of the Nee
Soon catchment in the future. In some sense, this may be good news, as the swamp
flora is generally more unusual and tends to be locally rarer. It is expected that about
2-4 weeks may elapse from the onset of a severely receded water table in originally
swampy areas following extreme drought, before mass die-offs of seedlings will occur
so this provides some buffer time for action to be taken, for example, for manual
irrigation to be set up. However, the time lag periods for saplings and mature trees
were not investigated during the study. It could be anticipated that the deeper roots
of larger trees would provide them with a longer buffer period, but water stress might
also be amplified.
An ordination analysis was conducted to investigate the relationship between the
tree community and soil and hydrology. It was only a cross-sectional study and should
be viewed as correlative rather than causal. To investigate the impacts of drought or
extreme flooding on mature trees would require either of two approaches:
1) field manipulations of precipitation (e.g. using rain screens) and hydrology
(e.g. digging soil trenches or creating artificial flooding) followed by monitoring of
trees. “Drought-Net” is a global network of study sites that are using standardised
methods such as these to artificially create drought-like conditions for plants. However,
they lack partner sites in forest biomes due to the logistic and operational difficulties
of creating rain screens over a forest canopy. Additionally, given that the Nee Soon
freshwater swamp forest is the last substantial tract of this forest type in Singapore,
there would be concerns whether such large-scale manipulations within the forest
could cause more harm than the resulting science could be beneficial. Tree coring to
provide direct measurements of sap flow could not be carried out during this study for
the same reason. A possible solution is to find alternative patchy freshwater swamp
areas outside nature reserves, which are of lower conservation value, for hydrological
manipulation and sap flow experiments.
2) Long-term monitoring of soil hydrology coupled with tree health is the other
approach, but will require a long time to accrue useable data, and will depend on
whether different hydrological conditions will occur in the sampled areas over the
200
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
monitoring period. The trees mapped in the 40 vegetation plots that have now been
handed over to NParks, will provide baseline data for future monitoring; establishing
more plots would increase the probability of sampling future fluctuating hydrological
conditions.
Furthermore, the plant species that were found to be rare or restricted to the Nee
Soon freshwater swamp forest should be targeted for propagation. These propagules
can then be transplanted elsewhere. For example, they can be used for restoring other
potential freshwater swamp sites, and hence the dilution of conservation threats.
The community assembly patterns within the Nee Soon freshwater swamp forest
may provide further insight on restoration strategies. For example, species such as
Gynotroches axillaris Blume and Pellacalyx axillaris Korth. are commonly found
in other swampy areas in the Central Catchment Nature Reserve, but these sites
seldom have other swamp indicator species such as Baccaurea bracteata Mull.Arg.,
Lophopetalum multinervium Ridl., Mussaendopsis beccariana Baill., Palaquium
xanthochymum (de Vriese) Pierre ex Burck, Pometia pinnata J.R.Forst. & G.Forst.,
etc. and almost never have the rarer swamp Syzygium R.Br. ex Gaertn. spp. or nutmegs
(Myristicaceae family). Two main reasons can be suggested: (1) these species are
dispersal-limited from establishing populations in other swampy sites beyond the Nee
Soon catchment; (2) specific soil or biotic conditions are limiting the establishment of
these species. The first reason can be overcome by transplanting propagules obtained
from the Nee Soon freshwater swamp forest. The second reason would require more
detailed studies and experimentation that involve swampy sites outside of the Nee
Soon catchment.
Regarding concerns about increasing tree falls and diebacks, the vegetation
ecology team considered that these are more likely to occur (1) for large trees (2)
of species with weaker wood, and more often observed (3) along forest edges, such
as around the firing ranges and along the pipeline. Only systematic monitoring and
dedicated data collection can test these hypotheses, and determine the extent of the
issue, together with possible solutions. This could begin with large trees, e.g. >10-
cm DBH, within visual range of the forest edge. These should be measured, mapped,
identified, tagged, and checked for liana infestation of the crown that may cause it to
be at risk of being pulled down by other falling trees. Whenever researchers, ground
staff, or members of the public report that a tree fall has occurred, the manager in
charge of the area could visit the affected site and (1) determine if any of the mapped
trees have been affected, (2) salvage any crown epiphytes and rare climbers that may
have been brought down by the tree fall, (3) log the exact locality, spatial extent of the
affected area, and the estimated time of occurrence of the event for future analyses. The
40 vegetation plots established during this study, and the tree plots set up within the
Central Catchment Nature Reserve previously, can be used as baselines for monitoring
tree fall, dieback, growth and regeneration. These can be supplemented by regular
(e.g. bimonthly to half-yearly) surveys along the forest edges to check the status of
these trees. The field guides being published as a result of this study, in addition to the
voucher specimens that were collected and that will be deposited in the two Singapore
herbaria, should help future researchers and staff to identify most of the large trees.
Conservation recommendations for Nee Soon
201
Floristic exploration of the Nee Soon catchment should continue, with more
emphasis on epiphytes and lianas as these were under-represented in our surveys. In
addition, a catchment-wide survey focused on identifying very large trees, e.g. >30-cm
DBH, may yield more rediscoveries, new records, or rare species that can be targeted
for conservation action.
Faunal ecology
Aquatic communities vary spatially across the Nee Soon freshwater swamp forest.
The variations are responses to the hydrological conditions, principally stream
depth characteristics. Depth is correlated with stream order (the order of branching),
from source to confluence. A higher diversity of benthic macroinvertebrate, pelagic
decapods and fish were observed in larger streams. However, this diversity was not
necessarily representative of the least disturbed fauna for the swamp forest streams,
because non-native invasive species were commoner (and added to the species total)
in the larger streams.
Spatial and temporal studies of the fish and decapods in the Nee Soon
freshwater swamp forest indicate that there exists a healthy community of native fish
and decapods within the Nee Soon freshwater swamp forest. However, the situation is
different towards the lower catchment and the edge of the forest, where the community
contains a large percentage of introduced species. Additionally, the main drivers of
diversity and richness among the fish and decapod community appear to be substrate
type, stream depth and stream order.
With this in mind, an important recommendation is to maintain constant
monitoring of the faunal communities in the Nee Soon freshwater swamp forest,
especially in the more upstream sites, so as to ensure that introduced species do not
successfully establish within such sites. If any individuals from an introduced species
are found they should be immediately removed, as they may have adverse effects on
native species once established, as has occured in many other places (Ogutu-Ohwayo,
1990; Beisner et al., 2003).
Additionally, the temporal results suggest that the Northeast monsoon period
(December to February) is an important breeding period for several species of fish.
Therefore, it is important that any form of disturbance in the streams during this period
be minimised. Otherwise, there may be serious effects on the long-term survival and
viability of these fish populations, something which cannot be afforded as many of the
fish in the Nee Soon freshwater swamp forest are endangered in Singapore.
Again, the lack of introduced species in the middle and upper catchment is
heartening, and it is imperative that monitoring continues to be carried out to maintain
this state of affairs and prevent establishment of introduced species deeper in the Nee
Soon freshwater swamp forest.
Invertebrate communities found at the downstream “edge habitat” sites closer
to the Upper Seletar Reservoir spillway were more diverse (higher richness and
abundance) than in the rest of the Nee Soon freshwater swamp forest. Nevertheless
the aquatic community at the “edge habitat” sites was not indicative of a “healthy”
forest stream system (Blakely et al., 2014). Macroinvertebrates such as chironomids
202
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
and snails that tend to be dominant in less “healthy” systems, and indicative of more
enriched and alkaline conditions, were in abundance at these sites. An abundance
of introduced fish species was also recorded at these sites, in comparison to the
predominantly native fauna in the remainder of Nee Soon.
The downstream “edge habitat” sites were likely influenced by disturbance from
the spillway discharging water from the reservoir periodically, discharging mixing
Nee Soon catchment and reservoir water and backing up into the stream system. The
lack of riparian habitat complexity (sparse tree cover) and dominance of fast-growing
vegetation, symptomatic of frequent disturbances at these sites, as well as more
alkaline water chemistry, provided less suitable niches for the more sensitive native
fish or rare invertebrates (e.g. Megalopterans and Plecopterans).
The hydrological conditions in Nee Soon freshwater swamp forest reflected
the wet and dry seasons in Singapore (Chia & Foong, 1991), with high water levels
recorded during the wet phases of early Northeast monsoon (November to December)
and Southwest monsoon (June to September); and low levels during the dry phase of
the late Northeast monsoon season (January to early March). The macroinvertebrate
composition was reflective of short-term changes in abiotic conditions (Barbour et al.,
1999), where the diversity of community corresponded to dry versus wet seasons (i.e.
high abundance in wet season, low abundance in dry season).
Faunal responses to hydrology were examined for the invertebrate fauna since i)
temporal patterns were evident in invertebrate communities when considered against
high frequency water-level observations, whereas fish communities were generally
not correlated with corresponding intra-annual changes in water level; and ii) the
invertebrates tend to be relatively less mobile than fish and more sensitive to short-term
changes in the aquatic environment (Barbour et al., 1999). Spatio-temporal patterns in
fish co mm unities were most strongly evident in the spatial variation of co mm unities
which points to the importance of factors such as water quality and associated tolerance
and resilience of the more “typical” (native) swamp forest fish community within the
heart of the catchment in contrast to the edges of the catchment and in proximity to the
spillway that discharges water from the reservoir.
Faunal response models reflected the potential effects of drought and disturbance
caused by elevated discharge. They revealed that when water levels were at their
lowest, the richness and abundance of invertebrates were reduced. Conversely, the
richness and abundance of invertebrate fauna was also reduced when maximum stream
water levels were greater.
Model results suggest that effects of minimum water level are more pronounced
than those of high water levels and that smaller streams are likely to be more susceptible
to very low faunal counts and even absence of invertebrate fauna altogether at high
maximum water levels or low minimum water levels. The most pertinent implications
of these results for the long-term management of the swamp forest streams is the
potential loss of individuals but perhaps more crucially, loss of taxa is anticipated from
the smaller streams in particular during periods of extreme, extensive low and high
flows. This is of particular relevance in the context of predictions of more intense and
prolonged dry and wet periods in the face of climate change.
Conservation recommendations for Nee Soon
203
Faunal diversity and conservation
Diversity and abundance changes of the freshwater fish and decapod fauna of the
freshwater swamp forest were investigated with the aim of facilitating future
conservation actions. Surveys were carried out by hand sampling and trap sampling
from February 2013 to May 2014 in 12 sites that represent different microhabitats
from the outskirts to the lower, middle, and upper reaches of the streams draining the
catchment. Stream water parameters (temperature, pH, DO, TDS, etc.) and stream
profile characters (substrates, cross section, canopy cover, and riparian vegetation)
were recorded (Cai et al., 2016; Li et al., 2016). Together with the aquatic survey,
visual-encounter surveys were conducted for terrestrial snails on each side of the stream
banks. The air humidity and temperature were also recorded at each site. Biodiversity
baselines were also established for dragonflies of Nee Soon freshwater swamp forest
based on quantitative sampling across the eight sub-catchments. Hydrological,
physiochemical parameters and habitats were analysed to identify the main drivers
structuring the dragonfly community (Cai et al., 2018).
Species diversity and abundance
Thirty-three species of native freshwater fish and 9 species of decapod crustaceans
(5 shrimps and 4 crabs) have been documented in detail. Analysis of the species
diversity at individual sites showed that one of the sites in the middle of Nee Soon
freshwater swamp forest supported the highest diversity and richness of native
freshwater fish and decapods. This site has a deep open pond with slow water flow
and is directly connected to the forest boundaries at both ends by shallow streams. The
sandy substrate, uneven stream bed flanked with large patches of aquatic vegetation
and leaf litter contained a variety of microhabitats stratified for different species of
fish and crustaceans, which could account for the high species diversity. Sites at the
outskirts of Nee Soon freshwater swamp forest supported the highest species diversity
of introduced fish and shrimp. The periodic release of water from the Upper Seletar
Reservoir might have contributed to the high species diversity of introduced fish and
shrimp. The four most abundant native species of fish found in Nee Soon freshwater
swamp forest were Trigonostigma heteromorpha, Hemirhamphodon pogonognathus,
Rasbora elegans and Betta pugnax, and the most abundant species of shrimp and crab
are Macrobrachium malayanum and Parathelphusa maculata respectively (Cai et al.,
2016; Li et al. 2016). A total of 10 families, 18 genera and 19 species of land snails
were recorded from the Nee Soon freshwater swamp forest. The outskirts had the
highest species diversity, while the Upper Swamp had the lowest species diversity.
The three most abundant species found in Nee Soon freshwater swamp forest are
Liardetia convexoconica, Helicarion perfragilis and Hemiplecta humphreysiana (Lim
et al., 2018). Fourty-nine species of odonates, belonging to 34 genera in 11 families
were recorded in the current study. An updated species list of Nee Soon dragonflies is
provided for future reference, with 67 species belonging to 47 genera in 11 families,
representing about half of all odonates ever recorded in Singapore. Among the eight
sub-catchments, the three mid sub-catchments all show low abundance and species
richness. This is followed by the two upper sub-catchments. The three low sub-
204
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
catchments all had high abundance and species richness. Hierarchical clustering and
Detrended Correspondence Analysis (DCA) indicated that three main groupings of
sites existed, each with a distinct community of associated species. Further analysis
by Canonical Correspondence Analysis (CCA) with 12 significant environmental
variables showed that these groups were significantly associated with respective
environmental variables (Cai et al., 2018).
New records of native species
The current survey recorded a native fish species new to Nee Soon freshwater swamp
forest: Barbodes lateristriga. It was observed mostly in the middle swamps with
moderate canopy coverage and deep, open stretches of water converging to large ponds.
One semi-slug, Damayantia cf. simrothi is believed to be a new record for Singapore
and was found at six sites in Nee Soon. Eighteen native species of odonates were new
to Nee Soon freshwater swamp forest though most of them are widespread common
species in Singapore and only found at the outskirts or in open parts of the study area,
viz. Podolestes orientalis, Libellago lineata, Argriocnemis rubescens, Pseudagrion
australiasiae, Pseudagrion micro cephalum, Copera marginipes, Onychargia
atrocyana, Acisoma panorpoides, Crocothemis servilia, Lathrecista asiatica,
Neurothemis fluctuans, Orthetrum sabina, Orthetrum luzonicum, Psedothemis jorina,
Rhyothemis triangularis , Trithemis aurora, Trithemis festiva and Tyriobapta torrida.
In addition, two specimens of the Blackwater mud snake Phytolopsis punctata were
found at Mid 1 sub-catchment. This species presents a new record for Singapore and is
currently only found in Nee Soon freshwater swamp forest (Tan et al., 2014; Thomas
et al., 2014).
Presence of introduced species
Introduced species can have severe impacts on the native species, causing decline
of population and possibly extinction. Through the course of this study, six
species of introduced fish and one species of shrimp were recorded and all are
common and widespread throughout Singapore. These species include Osteochilus
vittatus, Parambassis siamensis, Poecilia sphenops, Puntigrus tetrazona, Rasbora
borapetensis, Rhinogobius giurinus and Macrobrachium nipponense. Larson & Lim
(2005) documented that Rhinogobius giurinus has been outcompeting the local species
of freshwater goby, Pseudogobiopsis oligactis at locations where the two populations
overlap, a result corroborated by this survey. The less acidic water in the outskirts
could have encouraged propagation of introduced over native species. As the lower
reach of Nee Soon stream is connected to the spillway of the Upper Seletar Reservoir
it is subject to occasional flooding when excess water is released from the Reservoir.
Such events might inadvertently introduce foreign aquatic species that reside in the
water bodies of the reservoir and thus account for the introduced species found.
With the discovery of introduced species, it is of great importance to implement
preventive measures to reduce the possibility of alien species from being introduced.
Possible mitigation procedures might include monitoring and physical removal of the
introduced species on a regular basis, as well as the stepping up of existing enforcement
Conservation recommendations for Nee Soon
205
in the nature reserves. Enhancing riparian vegetation along the banks of streams and
retaining woody debris in the streams helps to preserve the natural habitat of the native
species, providing an advantage over invasive alien species which generally prefer
disturbed habitats. Construction of artificial barriers can also prevent the infiltration of
alien species into the core area of Nee Soon freshwater swamp forest. Prodasineura
humeralis is believed to have been introduced into Singapore within the last decade.
Observations from the present study show that the species is abundant in the outskirts
and stream stretches that are associated with an open canopy. The species is commonly
associated with fast flowing water, but hardly found in shaded forest streams with high
canopy cover. Achatina fulica is an invasive species originating from Africa and is
commonly known as the Giant African Snail. In Nee Soon freshwater swamp forest
shell heights of 7.5 cm or more have been observed. It is commonly found in parks and
degraded forest but rarely in undisturbed forest. Unfortunately, several specimens were
recorded deep in the forest indicating that this species may have already established a
population within Nee Soon. Lamellaxis gracilis is an introduced species found in the
outskirts of Nee Soon freshwater swamp forest. It is likely to have spread from nearby
plant nurseries and gardens. Bradybaena similaris is commonly found throughout the
moist tropics in urban areas such as gardens and plant nurseries. It is an agricultural
pest and is most likely to have been introduced due to the horticultural and agricultural
trade. It was observed in high numbers near the outskirts of Nee Soon but none was
recorded within the freshwater swamp forest (Lim et al., 2018). The potential impact
upon other native damselfly and snail species needs to be closely monitored.
Update on conservation status
Out of the 15 species of nationally threatened freshwater fish listed in the Singapore
Red Data Book (Lim et al., 2008), 13 species have been recorded from Nee Soon
freshwater swamp forest in the current survey. Although Trigonostigma heteromorpha
has been listed as “Endangered” in the Singapore Red Data Book, it was found at
all but one of the sites surveyed in Nee Soon freshwater swamp forest and in a high
abundance of an average 17 individuals per site per sampling occasion, with the
highest being 168 in a single sampling event. Similarly, Nemacheilns selangoricus,
which was previously listed as “Critically Endangered”, had relatively high mean
population abundance and presence at 10 out of 12 sites. Although current results
revealed numerical dominance of these species in Nee Soon freshwater swamp forest,
it remains true that within Singapore they are confined to the Central Catchment
Nature Reserve and (in the case of Nemacheilus selangoricus ) have stringent habitat
requirements. Using IUCN Red Data Book criteria adapted to national level (Davison,
2008), no change in their current conservation status is justified. More in-depth
studies have to be conducted to understand their detailed distribution in the nature
reserves. For the remaining 11 species, Boraras maculatus, Desmopuntius hexazona,
Pangio muraeniformis, Pseudomystus leiacanthus, Silurichthys hasseltii, Parakysis
longirostris, Clarias nieuhofii, Macrognathus maculatus, Luciocephalus pulcher,
Channa gachua and C. melasoma, the results are consistent with their current national
status (Li et al., 2016). No new records of shrimp species were found in this study but
206
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
the absence of Macrobrachium idae, M. neglectum and Caridina gracdirostris implies
that these species may have been extirpated from Nee Soon freshwater swamp forest.
All three species need brackish waters to complete the development of their larvae. It
is likely that the dam built to construct the Lower Seletar Reservoir in 1986 prevented
the downstream breeding migrations of these three species and rendered them unable
to complete their life cycles (Cai et al., 2016). The Green Tree Snail, Amphidromus
atricallosus temasek was listed as “Endangered” in the Singapore Red Data Book and
was found to have a widespread population in Nee Soon freshwater swamp forest being
recorded at eleven sites, although its mean abundance was low. Other rare species may
deserve a national conservation status of Endangered or Vulnerable such as Cyclotus
rostellatus, Japonia ciliocinctum and Microparmarion strubelli. All were extremely
restricted in their distribution and exist only within a few isolated patches of Nee Soon
freshwater swamp forest.
Recommendations
In view of these results in the context of existing scientific knowledge, specific
recommendations for i) long-term monitoring, ii) future research, and iii) management
options to ensure continued protection of the aquatic fauna of the Nee Soon freshwater
swamp forest are as follows:
i) Future monitoring
Long term monitoring and sampling of faunal populations are needed to build on
current knowledge and capture long-term trends. This is to optimise the value of the
knowledge/data of Nee Soon freshwater swamp forest faunal communities gained
from this study. Ecological and water quality monitoring at the downstream “edge
habitat” sites in association with similar monitoring of (and access to physicochemical
data from) the spillway itself as well as in Upper Seletar Reservoir (close to the
spillway outlet) and access will be particularly valuable to assess the medium- to
long-term effect of the spillway transfer on the Nee Soon freshwater swamp forest
faunal communities and/or the stability (e.g. resistance, resilience, adaptation) of these
communities to disturbance. A spatially localised project of this nature (of at least 2
year duration to capture temporal and seasonal patterns) is highly feasible, and would
help to answer important questions pertaining to the spillway and its effects on Nee
Soon freshwater swamp forest.
Adoption of standardised survey techniques, such as those applied here for faunal
as well as habitat and other abiotic factors will enable comparison of long-term changes
as well as spatial comparisons within the forest catchments but also against locations
elsewhere in the context of a national monitoring programme. Ecological monitoring
programmes for inland waters including aspects of these techniques are in various
stages of application across Singapore, supported by NParks and PUB (e.g. Clews
et al., 2012, 2014; Blakely et al., 2014; http://emid.nus.edu.sg/Inland/ecostandards.
jsp) in line with programmes developed internationally (e.g. Barbour et al., 1999;
ANZECC 2000a, 2000b; CEC, 2000). Ideally, at least annual screening for surveillance
monitoring should be conducted at multiple stations (preferably representing a range
Conservation recommendations for Nee Soon
207
of stream orders) alongside higher intensity investigative monitoring to investigate
potential issues (e.g. within sites proximal to the spillway) and to improve system
understanding (see recommendations below for future research).
Monitoring of fish should include as many sampling methods as possible to
maximise coverage spatial (microhabitat) and taxonomic coverage, enabling tracking
of populations of as many species as possible. Utilisation of additional trapping
methods, such as differential trapping with bait, i.e. traps with different mouth sizes
and/or mesh size, to exclude predators or unwanted organisms. This will allow for
more complete/comprehensive sampling.
Constant monitoring of Nee Soon freshwater swamp forest fish and decapod
fauna for introduction, establishment, and impacts of alien species within the catchment
(especially along the edges of Nee Soon freshwater swamp forest).
Establishment of a viable, long-term monitoring programme should also be
relevant for more broad-scale surveying of Singapore’s environment within other
water catchments. Training has been provided for agency staff on faunal sampling and
identification methods as tools for monitoring. In June 2015, a workshop was held on
the identification of fish, decapod crustaceans and macroinvertebrates from the Nee
Soon freshwater swamp forest. Several theory and practical sessions for aquatic faunal
identification purposes were facilitated by both NUS (TMSI, DBS, and LKCNHM)
and NParks staff. Sampling methodology for faunal collection was also introduced
in the introductory field resource, “A Guide to the Freshwater Fauna of Nee Soon
Swamp Forest” which was reviewed during the workshop and revised in response to
participants’ feedback (Ho et al., 2016).
ii) Future research
Aquatic food web and trophic structure studies of Nee Soon freshwater swamp forest
would be beneficial, to (i) serve as a basis for understanding of and further research
on community and ecosystem interactions/ecology; and (ii) inform conservation and
management actions for Nee Soon freshwater swamp forest.
Investigating the effects of release of water from Upper Seletar Reservoir is
required on various aspects (e.g. establishment, survival, interactions, ecology,
distribution) of introduced fish species as well as native fish species found at the edges
and lower reaches of Nee Soon freshwater swamp forest (to as far upstream as the
released water may affect).
Sedimentation and, potentially, sediment transport into and within the forest
streams should be investigated to examine the sources of sediments, effects of erosion
on stream fauna as well as trialing potential mitigation techniques such as “soft
engineering” of stream banks through the planting of appropriate plant species.
Minimum and maximum acceptable water levels need to be identified through
refinement and application of faunal response models, and integrated with the other
physical and ecological aspects of the swamp forest. Without agreed minima and
maxima, it will be difficult to identify triggers that initiate conservation actions and
responses.
208
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Statistical elucidation of acute faunal responses to hydrological events was not
possible for this study based on overt surveying but warrants further investigation.
For example, targeted studies should be aimed specifically at capturing rainfall events
along with a high temporal resolution of faunal responses including passive and active
drifting invertebrates (i.e. at scales over hours to days).
iii) Future management
Communication and cooperation among stakeholders should be enhanced by forming
a working group or committee comprising all relevant stakeholders (ministries,
statutory boards, academia, NGOs, public). Stakeholders should include any group
with potential influence, impact and interest not just in Nee Soon freshwater swamp
forest alone, but its watershed upstream and downstream.
Reduction of the influence of the spill way/discharge from reservoirs would help
to i) mitigate against changes in water quality; ii) reduce input of and local expansion
in the distribution of less desirable (non-native) fish species within the swamp forest
streams; and iii) maintain more “typical” forest stream communities of fish and
invertebrates, notably the rarer taxa that are less prevalent elsewhere.
To impede the spread of alien species into Nee Soon freshwater swamp forest, it
may be possible to create a weir or low head dam downstream of Nee Soon freshwater
swamp forest / upstream of reservoir input. However, this will probably bring about
hydrological issues stemming from flooding or ponding upstream of such a weir or
dam. An electric fish barrier (e.g. see http://www.smith-root.com/barriers/) could
be created across the channel downstream of Nee Soon freshwater swamp forest /
upstream of reservoir input, However, this would require funding, and would raise
development, infrastructure, maintenance, long-term commitment; and safety issues.
Environmental impact assessments (EIAs) would first have to be carried out
to ensure that the faunal communities of Nee Soon freshwater swamp forest are not
inadvertently impacted by the development and operation of any of these approaches.
In addition, consultations with management of multiple agencies and other stakeholders
will also be required prior to deployment of any engineering solutions (i.e. weirs, low
head dam and electric fish barrier) to ensure minimal impact to ongoing operations
(i.e. water transfer operations along spillways).
Reduction of maximum water levels could be considered to avoid unnecessary
disturbance of communities, by reduced input from the spillway, and by riparian and
forest planting to reduce peak flows. Improvement of the current spillway and dam
design, such as having a flap gate to prevent backflow of the reservoirs into the swamp,
could also prevent peak flow disturbances to the Nee Soon freshwater swamp forest
faunal communities.
Maintenance of greater than minimum water levels in small streams could be
considered, in particular to support the diversity of aquatic fauna found within the
freshwater swamp forest, especially the rare taxa such as stoneflies that are generally
not well supported in other catchments in Singapore.
Although not explicitly investigated as part of this study, field observations of
the main (3 rd order) stream channel suggest that the banks of the channel are eroding,
and suspended sediment may be relatively high as well as the settling of sediments
Conservation recommendations for Nee Soon
209
downstream, potentially reducing water depths in larger streams and/or changing
benthic habitats towards more, finer, “softer” sediments further downstream. In concert
with research to investigate this, trails of “soft-engineering” approaches conducted
initially off-site could be used to identify i) suitable plants; and ii) suitable techniques
to mitigate against stream bank erosion.
Cryogenic collection, imaging and barcoding
Being able to identify specimens to species level is important for most in-depth study
of biological systems. However, getting these identifications is particularly challenging
in tropical environments. Fortunately, a number of new tools promise to make this
task less daunting (Kutty et al., 2018). New imaging techniques help with illustrating
relevant characters and new and cheaper DNA barcoding techniques will allow for the
generation of databases that can be used by many researchers.
Making the fauna and flora of Nee Soon freshwater swamp forest and of
Singapore identifiable is achievable. The samples that have been collected and stored
have the potential to reveal the presence of several hundred or even thousands of
species. By focusing on particular taxa belonging to different ecological guilds, it is
feasible to begin understanding species turnover rates across habitats in Singapore and
to use this information for conserving Singapore’s native fauna and flora. A particularly
high priority should be using the newly developed plant barcoding techniques for all
of Singapore’s vascular plant species. This will allow for in-depth studies of species
interactions between plants and animals (e.g. pollination).
Eco-hydrological modelling
An integrated eco-hydrological model was developed in this study for the Nee
Soon freshwater swamp forest using Milce-SHE (Sun et al., 2018). The Mike-SHE
model simulates various water flow processes in the hydrological cycle, such as
rainfall, reservoir water of 3 reservoirs (Upper Seletar, Upper Peirce, Lower Peirce),
evapotranspiration, overland flow, infiltration, and groundwater flow.
The surveyed GIS data, including the stream network, the cross-sections and
the updated DEM, were incorporated in the model setup to make the model more
representative. The spatial and temporal variations of leaf area index (LAI) and
reference evapotranspiration (ET) retrieved from the remote sensing data, with the
aid of the root depth (RD) information from Vegetation Ecology team, were used
to establish a two-layer water balance model to account for the water loss from
evapotranspiration and the amount of water recharging to the saturated zone. In
addition, the field measurements from piezometers and stream sondes were processed
and integrated to calibrate and validate the model parameters.
A swamp forest extent map was derived from the numerical model simulation,
following the definiton of groundwater table shallower than 0.2 m below the surface
level. The model’s simulated swamp forest extent matches rather well with the swamp
forest map resulting from the study of O’Dempsey & Chew (2013), with an increased
area of 140 ha from 111 ha. The Nee Soon freshwater swamp forest was divided into 8
sub-catchments based on catchment delineation according to the topographic features.
210
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Various characteristics were derived for each sub-catchment, such as, elevation,
slope, rainfall, evapotranspiration, leaf area index, root depth, groundwater table,
and surface water area extent. Groundwater table map and surface water depth map
were also produced from the numerical model for the present hydrological condtion.
Both maps show the spatial distribution of surface and groundwater of the present
climate. These serve as references in assessing the hydrological vulnerability of the
catchments towards climate change. In addition, an eco-hydrology model which links
the hydrological conditions with the ecological recordings also helps to assess the
climate change impact on local eco-hydrology.
Twelve scenarios were introduced; they are combinations between various
reservoir water operating levels and the projected future rainfall amounts resulting
from a climate change study. Despite rainfall appearing to be the most influential factor
affecting the overall catchment water, i.e., the spatial average over the catchment, it is
interesting to observe the differing contributing factors of both rainfall and reservoir
level at sub-catchment levels. The effects of the two inputs differ depending on the
locations as it can be seen from hydrological maps (Sun et al., 2018). This spatial
distribution information is of importance should eco-hydrological management be
approached at sub-catchment level or spatially distributed.
Several management strategies were suggested to mitigate severe drought and
flood resulting from projected climate change impacts. These included the possibility
of adding water during droughts, and retaining water during floods.
Introducing additional water from the reservoirs to the upstream points of the
catchment is a conceivable option for the severe drought scenario. Two systems were
suggested for a point source management strategy. A pump and pipe system would be
required to increase the water head if the point sources are located at higher elevations
than the reservoir water levels. A pump system could provide broad coverage, but
would incur higher cost in construction and management. A pipe system alone could
be recommended if the point sources are located at lower elevations (i.e. lower than
the operating reservoir water level) or near the stream. A pipe system would cover
smaller areas, but would be more effective in conserving the swampy area at lower
cost and with less water consumption. There would be severe issues to be considered.
Amongst these are changes in water quality and the accidental introduction of alien
vertebrates and invertebrates. There would also be questions over disturbance to
soils and vegetation through the construction and maintenance of any such systems.
Practicality could also be questioned if low water levels in the stream system were to
coincide with water shortages in the reservoirs and in Singapore generally, as expected
during a severe regional drought. In lieu of the water quality of the reservoirs waters,
to protect the fauna in Nee Soon freshwater swamp forest the introduced reservoir
water to the point sources might have to be first filtered. In addition, the importance
and the highest priority of providing drinking water during drought is fully understood
and acknowledged. The proposed irrigation of the swamp will only take place when
reservoir water yield permits to do so.
Retention ponds are a conceivable option for severe flood scenarios. Rentention
ponds not only can reduce the affected flooding areas, but also promote habitat for
Conservation recommendations for Nee Soon
211
fauna. The study shows that rentention ponds with a fixed depth of 1 m could effectively
reduce the flooding areas by about 90%.
Summary conclusions and recommendations
Management objectives for the Nee Soon freshwater swamp forest are driven by
the need to conserve freshwater swamp forest habitat along with the fauna and flora
supported by this (now) rare habitat in Singapore. The conservation value of the forest
is relatively high since it is the last remaining freshwater swamp forest in Singapore
and a high proportion of fauna found in the Nee Soon freshwater swamp forest are
not found elsewhere in Singapore (e.g. Ng & Tim 1992, 1997; Yeo & Tim 2011).
This study has vastly expanded our understanding of the history and current status of
the Nee Soon freshwater swamp forest. Pressures on the ecology of the forest extend
beyond those identified at the onset of this project, namely susceptibility to flood and
drought, to encompass reservoir effects as well as potential impacts of erosion, land-
use change and water quality. Biological baselines have been extensively updated,
with rediscoveries of plant taxa previously thought to be locally extinct in Singapore,
and discoveries of potentially new taxa. The research has also yielded new records of
both flora and aquatic fauna. Extensive barcoding and imaging work has provided the
tools for more rapid identification of taxa to a higher taxonomic resolution. Habitats
closer to Nee Soon freshwater swamp forest edges, particularly in the vicinity of the
spillway, comprised greater proportions of introduced species and macroinvertebrates
associated with a different environment. Overall, faunal surveys indicate a healthy
and diverse fish and aquatic macroinvertebrate communities in Nee Soon freshwater
swamp forest. Hydrological influences were found to be substantial and potentially
limiting community composition. Soil and water chemistry were also found to have
significant effects. These results aid in identifying bottlenecks and thresholds for the
maintenance of key aspects of Nee Soon freshwater swamp forest biodiversity.
This paper has therefore presented a number of potential management solutions
for the conservation and management of the swamp forest, some of which can be
evaluated in the context of eco-hydrological modelling and others which warrant further
investigation and consideration. For hydrological-engineering management options,
the potential implications of modelled scenarios for in-stream faunal communities
are presented. Proposed solutions are then discussed in the context of management
objectives, cost-benefit and feasibility.
The adoption of potential management solutions should reflect stakeholder
priorities and management objectives and the relative importance of effects on the
hydro-ecology of the system carefully balanced. Implementation of management
efforts must consider ecological, hydrological and sedimentary processes. For
example, management to restore hydrological integrity should be evaluated alongside
ecological impacts. Protection and management of the Nee Soon freshwater swamp
forest should aim to maximise carbon sequestration potential and functionality on the
system as far as possible.
212
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Potential management solutions are subject to review and agreement amongst
primary stakeholders; given the primacy of water management and defence as national
priorities, these would include the Public Utilities Board, Ministry of Defence and
National Parks Board. Management solutions can be broadly classified to address five
issues of concern for the forest: hydrological integrity, erosion and sedimentation,
ecological integrity, the impact of the spillway, and impacts of construction and
development.
Maintaining hydrological integrity
a) Irrigation via the introduction of water sources (point sources) in the catchment
upstream could be an option to mitigate future extreme drought. There would be major
implications over water availability from other sources, cost, physical and ecological
effects, risks of introducing alien aquatic species, and water quality, as well as costs,
all of which would need to be included in an evaluation.
b) Retention ponds could be an option to mitigate flooding and simultaneously
promote additional habitats for flora and fauna. The study showed that retention
ponds with a depth of about 1 m could effectively reduce the flooded areas by about
90% but such steps would require evaluation of many factors similar to those for
irrigation solutions to address drought.
c) Reduction of maximum water-levels by i) reducing disturbance of communities via
reduced input from the spillway; ii) riparian and forest planting to reduce peak flows.
d) Maintenance of minimum water-levels in small streams in particular to support the
diversity of aquatic fauna found within the freshwater swamp forest, in particular more
rare taxa such as stoneflies which are generally not well supported in other catchments
in Singapore.
Mitigating erosion and sedimentation
a) Manage recreation: Restricting visitors to the catchment and repairing damage to
the tracks and bridges would help serve to reduce erosion.
b) Mitigation of stream-cha nn el erosion: fill in the re-sectioned channel to restore the
hydrological functioning of the swamp. However, this should be accompanied by an
assessment of potential effects on fish communities that have since established in the
channel.
c) Mitigation of stream-bank erosion: Investigation of sedimentation and potentially
sediment transport into and within the forest streams to formally examine the sources
of sediments, effects of erosion on stream fauna as well as trailing potential mitigation
techniques such as “soft engineering” of stream banks through planting of appropriate
plant species.
Conservation recommendations for Nee Soon
213
d) Mitigation of hillslope soil erosion: management of fire-associated loss of hillslope
vegetation, possible revegetation of hillslopes after considering the importance of
hillslope stability and ecological succession.
e) Monitor and manage tree falls and diebacks: i) investigate and potentially mitigate
against increased rates of tree falls and diebacks; and ii) manage loss of forest cover
caused by increased risk of fires due to drought / extended dry periods by improving
accessibility for fire fighters appropriately trained for operation in a nature reserve.
Maintaining ecological integrity
a) Forest recovery and replanting of endemic plant species to aid the ecological
resurgence of the reserve as well as the retention of water, sediments, and nutrients in
the system.
b) Propagation of rare plant species or those restricted to Nee Soon freshwater swamp
forest to facilitate replanting.
c) Continued exploration of flora and fauna alongside the further development of
barcoding techniques to yield more rediscoveries, new records, or rare species that can
be targeted for conservation action.
d) Long-term monitoring and standardised sampling of faunal communities as part of a
comprehensive, national ecological monitoring programme for inland waters to build
on current knowledge, capture long-term trends, and inform management decisions.
At least annual screening for surveillance monitoring should be conducted at multiple
stations (preferably representing a range of stream orders) alongside higher intensity
monitoring to investigate potential issues and to improve system understanding.
Establishment of minimum/maximum acceptable water-levels through refinement of
faunal response models can be applied as thresholds/targets for management.
e) Manage the spread of alien species into Nee Soon freshwater swamp forest from up-
and down-stream sources through the use of systems such as weirs, low head dams,
flap gates or electric fish barriers downstream from Nee Soon freshwater swamp forest
and/or upstream of reservoir input.
f) Mitigate against the high pH of reservoir water via chemical dosing or (preferably)
filtration through vegetated, “peaty” soil high in organic matter and tannins derived
from typical freshwater swamp forest plants (whereby the leaf litter provided by the
typical swamp forest flora provides acidity via humic acids typical of freshwater
swamp forest soils).
Mitigating effects of spillway
a) To reduce/prevent back-flooding into the swamp that flushes the system with
reservoir water (from the dam at the mouth of the Upper Seletar outlet, or “spillway”)
214
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
the feasibility and utility of a dam at the lower end of the lower channel to prevent
upstream encroachment should be investigated. This may be a flap gate designed to
allow outflow but prevent backflow of the reservoir water into the swamp.
b) Reduction of the influence of the spillway/discharge from reservoirs i) to mitigate
against changes in water quality; ii) to reduce input of, and local expansion in the
distribution of, less desirable (non-native) fish species within the swamp forest streams;
and iii) to maintain more “typical” forest stream communities of fish and invertebrates,
notably rarer taxa less prevalent elsewhere.
Mitigating impacts of construction and development
a) Identify, reduce, monitor and regulate impacts from new and existing construction
and development (e.g. water quality issues associated with heavy metals).
b) Ensure continued maintenance of the pipeline linking Upper Seletar Reservoir and
Lower Pierce Reservoirs to avoid leaks/input of reservoir water into the Nee Soon
freshwater swamp forest system.
Considering the views of stakeholders, public perceptions and the views and interests of
other agencies should also be factors in deciding conservation measures for Nee Soon
freshwater swamp forest. All management options should be considered in concert,
in the form of a whole-catchment plan of research, monitoring and management to
address both hydrological (flood, drought) conditions as well as emerging pressures
such as erosion and reservoir discharge.
ACKNOWLEDGEMENTS. This paper is based on contributions from leaders of the
multidisciplinary team, listed here in alphabetical order: Cai Yixiong, Chong Kwek Yan, Esther
Clews, Liong Shie-Yui, Rudolf Meier, Hugh Tan Tiang Wah, Robert Wasson, Sin Tsai Min,
Darren Yeo Chong Jinn, and Alan D. Ziegler. We thank them for the opportunity to include their
findings and recommendations here. We would like to thank staff of the National Parks Board,
especially the Conservation Division and National Biodiversity Centre, who have provided
support, helped to acquire permits, and worked tirelessly from the beginning of the project. We
thank the Ministry of Defence for permitting research teams to access areas around the firing
ranges. Colleagues from the Public Utilities Board have been very helpful and cooperative in
providing essential hydrological data (e.g. rainfall, reservoirs’ water levels) required for the
numerical eco-hydrological model. PUB’s participation in this project is greatly appreciated.
Past and present staff and students at the Tropical Marine Sciences Institute, the Department
of Biological Sciences and the Department of Geography, National University of Singapore,
contributed to the body of work and knowledge-base on which this study was developed as
well as providing practical, logistical and technical support for the project team. Our gratitude
goes to the Ministry of Finance and the National Parks Board for their funding without which
it would have been impossible to carry through this project.
Conservation recommendations for Nee Soon
215
References
ANZECC (Australian and New Zealand Environment Conservation Council) (2000a).
Australian and New Zealand Guidelines for Fresh and Marine Water Quality (the
Water Quality Guidelines). National Water Quality Management Strategy (NWQMS)
Document 4. The Australian and New Zealand Environment and Conservation Council
and the Agriculture and Resource Management Council of Australia and New Zealand.
www.mincos.gov.au/publications/australian_and_new_zealand_guidelines_for_fresh_
and_marine_water_quality.
ANZECC (Australian and New Zealand Environment Conservation Council) (2000b).
Australian Guidelines for Water Quality Monitoring and Reporting 2000 (the Monitoring
& Reporting Guidelines). National Water Quality Management Strategy (NWQMS)
Document 7. The Australian and New Zealand Environment and Conservation Council
and the Agriculture and Resource Management Council of Australia and New Zealand.
www.mincos.gov.au/publications/australian_and_new_zealand_guidelines_for_fresh_
and_marine_water_quality.
Barbour, M.T., Gerritsen, J., Snyder, B .D. & Stribling, J.B. (1999). Rapid Bioassessment Protocols
for Use in Streams and Wadeable Rivers: Periphyton, Benthic Macroinvertebrates and
Fish , 2 nd ed, EPA 841-B-99-002. Washington, D.C.: U.S. Environmental Protection
Agency, Office of Water.
Beisner, B.E., Ives, A.R. & Carpenter, S.R. (2003). The effects of an exotic fish invasion on the
prey communities of two lakes. J. Anim. Ecol. 72(2): 331-342.
Blakely, T.J., Eikaas, H.S. & Harding, J.S. (2014). The Singscore: a macroinvertebrate biotic
index for assessing the health of Singapore’s streams and canals. Raffles B. Zool. 62:
540-548.
Cai, Y., Li, T., Lim, W., Tok, C., & Woo, C.M. (2016). Biodiversity assessment of freshwater
shrimps and crabs in the Nee Soon swamp forest, Singapore. In: Kawai, T. &
Cumberlidge, N. (eds) A Global Overview of the Conservation of Freshwater Decapod
Crustaceans , pp. 373-418. Heidelberg: Springer Nature.
Cai, Y., Ng, C.Y. & Ngiam, R.W.J. (2018). Diversity, distribution and habitat characteristics of
dragonflies in Nee Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore 70
(Suppl. 1): 123-153.
CEC (Commission of the European Communities) (2000). Directive 2000/60/EC of the
European parliament and of the council of 23 October 2000 establishing a framework
for community action in the field of water policy. Off. J. Eur. Communities L327: 1-72.
Chia, L.S. & Foong, S.F. (1991). Climate and weather. In: Chia, L.S., Rahman, A. & Tay,
D.B.H. (eds) The Biophysical Environment of Singapore, pp. 13-49. Singapore:
Singapore University Press for the Geography Teachers’ Association Singapore.
Chong, K.Y., Neo, L., Tan, S.Y., Koh, C.Y., Lim, R.C.J., Loh, J.W., Ng, W.Q., Seah, W.W.,
Yee, A.T.K. and Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon
Swamp Forest (I): Lauraceae. Nat. Singapore 9: 1-28.
Chong, K.Y., Lim, R.C.J., Loh, J.W., Neo, L., Seah, W.W., Tan, S.Y. & Tan, H.T.W. (2018).
Rediscoveries, new records, and the floristic value of the Nee Soon freshwater swamp
forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 49-69.
Clews, E., Lin, Y. & Eikaas, H. (2012). Biomonitoring of Singapore’s Freshwater Ecosystems.
City Green 5: 28-33.
Clews, E., Low, E., Belle, C.C., Todd, P.A., Eikaas, H.S. & Ng, P.K.L. (2014). A pilot biotic
index of water quality for Singapore’s reservoirs. Ecol. Indie. 38: 90-103.
216
Ganl. Bull. Singapore 70 (Suppl. 1) 2018
Clews, E., Corlett, R.T., Ho, J.K.I., Koh, C.Y., Liong, S.Y., Memory, A., Ramchunder, S.J.,
Siow, H.J.M.R, Sun, Y., Tan, H.H., Tan, S.Y., Tan, H.T.W., Theng, M.T.Y. & Yeo, D.C.J.
(2018) The biological, ecological and conservation significance of freshwater swamp
forest in Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 9-31.
Davison, G.W.H. (2008). Appendix 1. Criteria for determining category of threat. In: Davison,
G.W.H., Ng, RK.L. and Ho, H.C. (eds) The Singapore Red Data Book: threatened plants
and animals of Singapore, 2 nd ed., pp. 268-271. Singapore: Nature Society (Singapore).
Davison, G.W.H., Cai, Y., Li, T. & Lim, W. (2018). Integrated research, conservation and
management of Nee Soon freshwater swamp forest, Singapore: hydrology and
biodiversity. Gard. Bull. Singapore 70 (Suppl. 1): 1-7.
Dobbs, C., Nitschke, C.R. & Kendal, D. (2014). Global drivers and tradeoffs of three urban
vegetation ecosystem services. PLoS ONE 9(11): el 13000.
Friess, D.A., Richards, D.R. & Phang, V.X. (2015). Mangrove forests store high densities of
carbon across the tropical urban landscape of Singapore. Urban Ecosyst .: 1-16.
Ho, J.K.I., Ramchunder, S.J., Memory, A., Theng, M., Li, T., Clews, E., Cai, Y., Tan, H.H.
& Yeo, D.C.J. (2016). A Guide to the Freshwater Fauna of Nee Soon Swamp Forest.
Singapore: Lee Kong Chian Nature History Museum & Tropical Marine Science
Institute, National University of Singapore, Singapore.
Ho, J.K.I., Quek, R.F., Ramchunder, S.J., Memory, A., Theng, M.T.Y., Yeo, D.C.J. & Clews,
E. (2018). Aquatic macroinvertebrate richness, abundance and distribution in the Nee
Soon freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 71-108.
Kutty, S.N., Wang, W., Ang, Y., Tay, Y.C., Ho, J.K.I. & Meier, R. (2018). Next-generation
identification tools for Nee Soon freshwater swamp forest, Singapore. Gard. Bull.
Singapore 70 (Suppl. 1): 155-173.
Larson, H.K., & Lim, K.K.P. (2005). A guide to gobies of Singapore. Singapore: Singapore
Science Centre.
Li, T.J., Chay, C.K., Lim, W.H., & Cai, Y. (2016). The fish fauna of Nee Soon swamp forest,
Singapore. Raffles B. Zool. Suppl. 32: 56-84.
Lim, K.K.P. (2008). Fishes. In: Davison, G.W.H., Ng, P.K.L. & Ho, H.C. (eds) The Singapore
Red Data Book: Threatened Plants & Animals of Singapore, 2 nd ed. Singapore: Nature
Society (Singapore).
Lim, R.C.J., Chong, K.Y., Neo, L., Tan, S.Y., Koh, C.Y., Loh, J.W., Ng, W.Q., Seah, W.W.,
Yeee, A.T.K. & Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon
Swamp Forest (III): Myristicaceae. Nat. Singapore 9: 29-39.
Lim, W.H., Li, T.J. & Cai, Y. (2018). Terrestrial snails and slugs diversity in Nee Soon freshwater
swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 109-121.
Murphy, D.H. (1997). Odonata biodiversity in the Nature Reserves of Singapore. Gard. Bull.
Singapore 49: 333-352.
Neo, L., Chong, K.Y., Tan, S.Y., Koh, C.Y., Lim, R.C.J., Loh, J.W., Ng, W.Q., Seah, W.W.,
Yee, A.T.K. and Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon
Swamp Forest (II): Cratoxylum (Hypericaceae). Nat. Singapore 9: 29-39.
Neo, L., Chong, K.Y., Koh, C.Y., Tan, S.Y., Loh, J.W., Lim, R.C.J., Seah, W.W. & Tan, H.T.W.
(2017). Short-term external effects of increment coring on some tropical trees. J. Trop.
For. Sci. 29: 519-529.
Ng, P.K.L. & Lim, K.K.P. (1992). The conservation status of the Nee Soon freshwater swamp
forest of Singapore. Aquat. Conserv. 2(3): 255-266.
Ng, P.K.L. & Lim, K.K.P (1997). The diversity and conservation status of fishes in the nature
reserves of Singapore. Gard. Bull Singapore 49(2): 245-265.
Conservation recommendations for Nee Soon
217
Nguyen, C.T.T., Wasson, R.J. & Ziegler, A.D. (2018). The hydro-geomorphic status of the
Nee Soon freshwater swamp forest catchment of Singapore. Gard. Bull. Singapore 70
(Suppl. 1): 33-48.
O’Dempsey, T. & Chew, P.T. (2013). The freshwater swamp forests of Sungei Seletar
Catchment: a status report. In: Leong, T.M. & Ho, H.C. (eds) Proceedings of Nature
Society, Singapore’s Conference on “Nature Conservation for a Sustainable Singapore”
- 16 October 2011, pp. 121-166. Singapore: Nature Society (Singapore).
Ogutu-Ohwayo, R. (1990). The decline of the native fishes of Lakes Victoria and Kyoga (East
Africa) and the impact of introduced species, especially the Nile perch, Lates niloticus,
and the Nile tilapia, Oreochromis niloticus. Environ. Biol. Fish. 27(2): 81-96.
Rahman A. (1991). Soils. In: Chia, L.S., Rahman, A. & Tay, D.B.H. (eds) The Biophysical
Environment of Singapore, pp. 89-133. Singapore: Singapore University Press.
Schumm, S.A. (1973). Geomorphic thresholds and complex response of drainage systems.
Fluv. Geomorphol. 6: 69-85.
Sun, Y., Wendi, D., Kim, D.E. & Liong, S.Y. (2015). Technical Note: Application of artificial
neural networks in groundwater table forecasting - a case study in Singapore swamp
forest. Hydrol. Earth Syst. Sci. Discuss. 12: 9317-9336.
Sun, Y., Wendi, D., Kim, D.E. & Liong, S.Y. (2016). Development and application of an
integrated hydrological model for Singapore freshwater swamp forest. Procedia
Engineer. 154: 1002-1009.
Sun, Y., Kim, D.Y., Wendi, D., Doan, D.C., Raghavan, S.V., Jiang, Z. & Liong, S.Y. (2018).
Projected impacts of climate change on stream flow and groundwater of Nee Soon
freshwater swamp forest, Singapore. Gard. Bull. Singapore 70 (Suppl. 1): 175-190.
Tan, H.H., Li, T.J. & Lim, W.H. (2014). A food item of the blackwater mud snake. Singapore
Biodivers. Rec. 2014: 311-312.
Tan, S.Y., Koh, C.Y., Siow, H.J.M., Li, T., Wong, H.L., Heyzer, A., & Tan, H.T.W. (2013).
100 Common Vascular Plants of the Nee Soon freshwater swamp forest, Singapore.
Singapore: Raffles Museum of Biodiversity Research, National University of Singapore,
Singapore.
Tan, S.Y., Chong, K.Y., Neo, L., Koh, C.Y., Lim, R.C.J., Loh, J.W., Ng, W.Q., Seah, W.W., Yee,
A.T.K. & Tan, H.T.W. (2016). Towards a field guide to the trees of the Nee Soon Swamp
Lorest (IV): Xanthophyllum (Polygalaceae). Nat. Singapore 9: 29-39.
Thomas, N., Li, T.J., Lim, W.H. & Cai, Y. (2014). New record of the blackwater mud snake in
Singapore. Singapore Biodivers. Rec. 2014: 309-310.
Turner, I.M., Boo, C.M., Wong, Y.K., Chew, P.T. & Ibrahim, A. (1996). Lreshwater swamp
forest in Singapore, with particular reference to that found around the Nee Soon Tiring
Ranges. Gard. Bull. Singapore 48: 129-157.
Wendi, D., Liong, S.Y. Sun, Y. & Doan, C. D. (2016). An innovative approach to improve
SRTM DEM using multispectral imagery and artificial neural network. J. Adv. Model.
Earth Syst. 8: 1-12.
Yeo, D.C.J. & Lim, K.K.P. (2011). Lreshwater ecosystems. In: Ng, P.K.L., Corlett, R.T. &
Tan, H.T. (eds) Singapore Biodiversity - An Encyclopedia of the Natural Environment
and Sustainable Development, pp 52-63. Singapore: Raffles Museum of Biodiversity
Research, Department of Biological Sciences, National University of Singapore,
Singapore.
Zuzao, V.H.D. & Pleguezuelo, C.R.R. (2008). Soil-erosion and runoff prevention by plant
covers: A review. Agron. Sustain. Dev. 28(1): 65-86.
Instructions for contributing authors (continued)
Title and authorship. The title should concisely describe the contents. If a scientific name is used, its
authority is normally excluded, but the family name would be provided. Authors’ names, affiliations
and postal/ e-mail addresses are stated below the title. If more than one author, indicate “corresponding
author”. Avoid footnotes. A short running title (up to six words) should also be provided.
Abstract. The abstract should be less than 200 words. It should concisely indicate the article’s contents
without summarising it; mentioning novelties and name changes. Keywords. Suggest at most eight
keywords, in alphabetical order.
Scientific names and author abbreviations. Genus and species names of organisms must be italicised and
followed by the authority (with family name in parentheses) when first mentioned in the text or diagnoses.
Standards for author abbreviations should follow:
1PNI <http: //www. lpm. org/ipni/authorsearchpage. do>
Authors of Fungal Names <http://www.indexfungorum.org/authorsoffungalnames.htm>
Herbarium abbreviations follow
Thiers, B. (2013). [continously updated] Index Herbariorum: A global directory of public herbaria and
associated staff. New York Botanical Garden's Virtual Herbarium, <http://sweetgum.nybg.org/ih/>
Journal and book title abbreviations. For journals:
Bridson, G.D.R., Townsend, S.T., Polen, E.A. & Smith, E.R. (eds) (2004). BPH-2: periodicals with
botanical content: comprising a second edition of Botanico-Periodicum-Hunlianum. Pittsburg: Hunt
Institute for Botanical Documentation.
For books:
Stafleu, F.A. & Cowan, R.S. (eds) (1976-88). Taxonomic Literature. 2nd edition. 7 vols [Regnum Veg.
vols 94, 98, 105, 110, 112, 115, 116],
Stafleu, F.A. & Mennega, E.A. (eds) (1992-). Taxonomic Literature. Supplements [Regnum Veg. vols
125, 130, 132],
A useful source of verifying names of publications is
<http://asaweb.huh.harvard.edu:8080/databases/publication_index.html>. If in doubt, list full titles.
Other abbreviations and units of measurement. If using standard abbreviations and acronyms, give the
full term on first mention. Dates are cited as: 1 Jan 2000. SI (metric) units of measurement are used and
spelled out except when preceded by a numeral; they are abbreviated in standard form: g, ml, km, etc.
Coordinates should be given in the form 11°46'44.8 ,, N 109°10 , 44.5"E.
Tables. Tables are numbered in arabic numerals in the order they are first mentioned in the text and carry
an indicative legend at the head. Tables are given at the end of the manuscript.
Illustrations. All drawings, maps, graphs and photographic images (individually or collected in a plate)
are to be numbered in arabic numerals in the order they are first mentioned in the text, as Fig. 1, Fig. 2,
etc. (plate components would be referred to in the text as Fig. 1A, IB, Fig. 1A-D, etc.). If relevant, scale
bars should be used to indicate magnification.
When grouping photographs, the maximum page area 19.5 x 13 cm must be heeded. High resolution digital
images may be submitted as separate files (line drawings in black and white at 600 dpi, photographs at
300 dpi) sent electronically or in a CD. Captions on figures should be added as a layer so that they can be
further edited or send both captioned and uncaptioned versions. Do not embed images into the main text
file and provide the legends for all figures at the end of the manuscript.
References in the text. Citation in the text should take the form: King & Gamble (1886) or (King &
Gamble, 1886), or King et al. (1886) if more than two authors to a work. Use 2000a, 2000b, etc. if several
papers by the same author(s) in one year are cited.
References listed at the end. There, works mentioned in the text are listed alphabetically as follows:
Dallwitz, M.J., Paine, T.A. & Zurcher, E.J. (1999). User’s Guide to the DELTA Editor, http://
biodiversity.uno.edu/delta/ (accessed on 2 Aug. 2010).
Persson, C. (2000). Phylogeny of Gardenieae (Rubiaceae) based on chloroplast DNA sequences from the
rps 16 intron and trn L (UAA)-F( GA A) intergenic spacer. Nordic J. Bot. 20: 257-269.
Ridley, H.N. (1930). The Dispersal of Plants Throughout the World. Ashford, U.K.: L. Reeve.
Smith, A.C. & Darwin, S.P (1988). Rubiaceae. In: Smith, A.C. (ed) Flora Viliensis Nova, A New Flora
of Fiji 4: 143-193.
References to web-based resources should include either a doi (digital object identifier) specification
or full URL mentioning also the date it was accessed. Use of DNA sequences from GenBank should be
acknowledged and the studies for which the sequences were generated should be cited.
Style of nomenclatural summaries. The following style is required:
Ornithoboea arachnoidea (Diels) Craib, Notes Roy. Bot. Gard. Edinburgh 11: 251 (1920); Burtt, Notes
Roy. Bot. Gard. Edinburgh 22: 294 (1958).
Ornithoboea panishii C.B.Clarke in ADC. & C.DC., Monogr. Phan. 5(1): 148 (1883).
If authors include full bibliographic data for these works in the list of references at the end of the paper,
they should also be mentioned in the text briefly, e.g., “Nomenclatural references researched include
Blume (1849) and Roxburgh (1824).”
Homotypic synonyms should be provided in a block, stating the type at the end.
Front cover picture: Female of Libellago aurantiaca (Photo by Y. Cai)
national'
PARKS
LET'S MAKE SINGAPORE OUR GARDEN
Singapore
Botanic Gardens
The Gardens’ Bulletin Singapore VOL. 70 (Supplement 1) 2018