VOL. 118, PARTS 1 & 2

31 MAY, 1994.

Contents

Transactions of the

Royal Society of South

Australia

Incorporated

Williams, W. D. Climate change and its implications for South Australia — introductory

remarks - - - - - - - - - - -

Schwerdtfeger, P. A brief overview of climate research - - - - - -

Allan, R. J. Modelling climatic change and variability - = - - - -

Williams, M. A. J. Some implications of past climatic changes in Australia = - -

Fitzpatrick, R. W. & Wright, M. J. Climate change and its implications for South

Australian soils - > - - - - - =:

Bates, B. W., Charles, S. P., Sumner, N. R. & Fleming, P. M. Climate change and its

hydrological implications for South Australia - - - -

Harvey, N. & Belperio, A. P. Implications of climate change for the South Australian

coastline - - - - - - ~ - - -

Tyler, M. J. Climatic change and its implications for the amphibian fauna - -

Stott, P. Climate change and its implications for the terrestrial vertebrate fauna F

Boardman, R. Some possible effects of climate change on vegetation - - -

Burns, M. E. & Walsh, C. The economic implications of climate change - -

McMichael, A. J. & Beers, M. Y. Climate change and human Population health: tne

and South Australian perspectives - -

Thumlert, T. A. & Austin, A. D. Biology of Phylacteophaga Jroggatti Riek (Hymenoptera:

Pergidae) and its parasitoids in South Australia -

Bayliss, D. E. Description of three new barnacles of the genus Elminius (Cirripedia:

Thoracica) from South Australia, with a key to species of the Elminiinae

Lange, R. T., Lay, B. G. & Tynan, R. W. Evaluation of extensive arid Faneclance: the

land condition index (LCI) - - - - - -

Prideaux, G. J. A small sthenurine kangaroo from a Pleistocene cave deposit, Nullarbor

Plain, Western Australia = - - - - - - - -

Zbik, M. The Cook 007 meteorite: a new H4 chondrite from South Australia - -

Brief Communication:

Read, J. L. A major range extension and new ecological data on Oxyuranus microlepidotus

(Reptilia: Elapidae) - - - - - - - - -

Walker, S. J. Growth in the Australian ene. frog, Bink australis (Gray) Anns.

Leptodactylidae) - -

PUBLISHED AND SOLD AT THE SOCIETY’S ROOMS

SOUTH AUSTRALIAN MUSEUM, NORTH TERRACE, ADELAIDE, S.A. 5000

owe

115

133

139

143

147

TRANSACTIONS OF THE

ROYAL SOCIETY

OF SOUTH AUSTRALIA

INCORPORATED

VOL. H8, PART 1

Proceedings of a symposium entitled

“CLIMATE CHANGE AND ITS IMPLICATIONS FOR

SOUTH AUSTRALIA”

held on Li November, (993.

TRANSACTIONS OF THE

ROYAL SOCIETY OF SOUTH AUSTRALIA INC.

CONTENTS, VOL. 118, 1994

PARTS | & 2, 31 MAY, 1994

Williams, W. D. Climate change and its implications for South Australia — introductory remarks

Schwerdtfeger, P. A brief overview of climate research - - - - - -

Allan, R. J. Modelling climatic change and variability - - - - - -

Williams, M. A. J. Some implications of past climatic changes in Australia - - -

Fitzpatrick, R. W. & Wright, M. J. Climate wine and its herpligations for South Australian

soils - - - - - - - -

Bates, B. W., Charles, S. P., Sduiner, N.R. & Fleming, P.M. Climate cheng 2 and 1s

hydrological implications for South Australia - -

Harvey, N. & Belperio, A. P. Implications of climate change for the South Australian coastline

Tyler, M. J. Climatic change and its implications for the amphibian fauna — - - -

Stott, P. Climate change and its implications for the terrestrial vertebrate fauna —- -

Boardman, R. Some possible effects of climate change on vegetation - - - -

Burns, M. E. & Walsh, C. The economic implications of climate change = - - -

McMichael, A. J. & Beers, M. ¥. Climate change and human popalaeige health: global and

South Australian perspectives -

Thumlert, T. A. & Austin, A. D. Biology of Phylacteophaga froggatti Riek (ymenopter:

Pergidae) and its parasitoids in South Australia -

Bayliss, D. E. Description of three new barnacles of the genus Elminius (Cirripedia: Thorucica)

from South Australia, with a key to species of the Elminiinae - -

Lange, R. T., Lay, B. G. & Tynan, R. W. Evaluation of extensive arid rangelands: the land

condition index (LCI) - - - - - - - - -

Prideaux, G. J. A small sthenurine kangaroo from a Pleistocene cave deposit, Nullarbor Plain,

Western Australia - - : - - - - - -

Zbik, M. The Cook 007 meteorite: a new H4 chondrite from South Australia - -

Brief Communications:

Read, J. L. A major range extension and new ecological data on OxyaranDs microlepidotus

(Reptilia: Elapidae) - - - e zg . = -

Walker, S. J. Growth in the Australian burrowing frog, Cyclorana australis (Gray) (Anura:

Leptodactylidae) - - - - 7 : . - - =

45

143

147

PARTS 3 & 4, 30 NOVEMBER, 1994

Davies, M. & Watson, G. F. Morphology and reproductive biology of Limnodynastes salmini,

L. convexiusculus and Megistolotis lignarius (Anura: Leptodactylidae:

Limnodynastinae) - - - - - - - - -

Rounsevell, D, E., Ziegeler, D., Brown, P. B., Davies, M. & Littlejohn, M. J. A new genus

and species of frog (Anura: Leptodactylidae: Myobatrachinae) from

southern Tasmania - - 7 - - - - -

Pamment, D., Beveridge, I. & Gasser, R. B. The distribution of nematode parasites within

the stomach of the western grey kangaroo, Macropus fuliginosus -

Ainslie, R. C., Johnston, D. A. & Offler, E. W. Growth of the seagrass Posidonia sinuosa

Cambridge et Kuo at locations near to, and remote from, a power station

thermal outfall in northern Spencer Gulf, South Australia - - -

Molsher, R. L., Geddes, M. C. & Paton, D. C. Population and reproductive ecology of

the small-mouthed hardyhead Atherinosoma microstoma (Giinther)

(Pisces: Atherinidae) along a salinity gradient in the Coorong,

South Australia - ~ - - - = + - - -

Hutchinson, M. N., Milne, T. & Croft, T. Redescription and ecological notes on the pygmy

blubtonstie, Tiliqua adelaidensis (Squamata: Scincidae) —-

Fuller, M. K. & Jenkins, R. J. F. Moorowipora chamberensis, a coral from the early Cambrian

Moorowie Formation, Flinders Ranges, South Australia - - -

Peterson, M., Shea, G. M., Johnston, G. R. & Miller, B. Notes on the morphology and

biology of Ctenophorus mckenziei (Storr, 1981) (Squamata:

Agamidae) - - = ; + % 4 - - 7

McKillup, S. C. & McKillup, R. V. Reproduction and growth of the smooth pebble

crab Philyra laevis (Bell 1855) at two sites in South Australia during

1990-9] - - : E = = =: : - -

Olsen, A. M. The history of the development of the Pacific oyster, Crassostrea @ gigas CTanbee)

industry in South Australia = - “ = -

Bird, A. F. & Yeates, G. W. Studies on Aprutides guidettii (Nematoda: Seinuridae) isolated

from soil at Northfield, South Australia - - - - - -

Brief Communication:

Tyler, M. J., Davies, M., Watson, G. F. The hylid frog Litoria albapentare CGtimhet) in

the Northern Territory - - - 4

Insert to Transactions of the Roval Society of South Australia, Vol. H8, parts 3 & 4. 30 November, 1994

149

171

187

197

207

217

227

237

245

253

261

267

CLIMATE CHANGE AND ITS IMPLICATIONS

FOR SOUTH AUSTRALIA —

INTRODUCTORY REMARKS

By W. D. WILLIAMS*

Summary

It is my privilege, as President of the Royal Society of South Australia, to introduce

this important symposium on climate change and its implications for South Australia.

In doing so, I wish first to welcome all participants and I extend a particular welcome

to Sir Mark Oliphant, an Honorary Fellow of our Society. Our patron, Her Excellency

the Governor, Dame Roma Mitchell, has indicated that Vice-Regal commitments

preclude her attendance. | trust that all participants, students, distinguished scientists,

or mere mortals like me, will find the symposium interesting, useful and a

constructive addition to local debate on this most important matter.

Teiyacnany of the Royal Sacre ofS. Mase (994) TBR Le.

CLIMATE CHANGE AND ITS IMPLICATIONS FOR SOUTH AUSTRALIA —

INTRODUCTORY REMARKS

by W. D. WILLIAMS?

Tis any privilewe, as President af the Royal Society

of South Australia, 09 tntroduee this injportani

symposium on climate chimge and its implications for

South Australia.

Ii deity se, L Wish first iu welcome all participants

und | extend a particular welcome to Sir Mark

Oliphant, an Heaueary Fellow of our Society. Cur

putron. Her Excellency the Governor, Dame Roma

Mitchell, has indicated that Viee-Regal conunitinents

preclude her attendance. I trust that all participants,

dndents, distinguished scientists, or mere mortals like

me. will Fiod che sympesiuny interesting, useful and

¥ Constructive additron te local debate on this most

TMIpOrin Matter.

The Royal Society of South Australia held nis first

meoling i November IR80, shortly afer Queen

Vietoriacassemted ta the use of the word Royal The

Society, however, succeeded the Adelaide

Philosophies! Society tortaed much sarher, and indeed

not lon afler the foundation of the Stace of South

Austauia [rt was formed in 1853. Since those early

yours, the Society has met regularly. published

scientific papers, maintained an extensave Library,

awarded grants fir research oruwards for meritorious

CONTDUTIORS to schence, and its Fellows have docluded

most ot South Australia’s best known scientists,

The overall objective of the Society is the promotion

and ciffunon of saentitic Knowledge, und it is in

support of this objective that the present symposium

hus been sponsored, [need hardly remind this audience

that any change to the climate of South Australia will

have profound, sygarficant, and comprehensive effects

an the South Austrahan environment, economy. social

structure and public health — to name but the most

ohvious features that could be affected, It isto orig

this to the altention of the community in general asa

its decision makers in particular that the present

syinpesiuin us berg held, Such an sum falls squarely

wilhin the overall objective of the Society. To premate

farther the views expressed. the pitpers presented ire

heing published in this part of the Trqnsacrions of the

Riyal Socrerv of South Australia and are alse available

sepairulely ws a bowk,

OD} course, the symposium is only one of many thul

have heen and are plunined to he held to discuss chymnite

chinge At ihe international level. numerous mectings

have been held, several scientific journals specitically

angel rescurch emt climate change, and there 1 an

* Department of Zoulogy, University of Adelaide, South

Austrailia 3005-

iitermational commuter which reports revulurly (POC:

Intergavernmental Panel on Climate Change).

Likewise, at the national level. considerable activity

prevails. Indeed. the Australian Academy of Science

is Currently sponsoring # meciing in Canberra oa the

subject of climate change, The National Greenhouse

Respoase Strategy was finalised in 1992. and we the

Stale level, Various povernment ugencies maintain ut

least a watching brief on the subject, a number of

oatnral and medical scicnlsts, economists and others

ure actively interested in the subject. there is a Climuce

Change Committee (serviced by the Department of

Environment and Natural Resources and including

representatives from # wide variety of government

departments) which reports regularly to Cabinet and

has published several imporiant documents (e.g. South

Australian Climate Change Conimittee 1990, 1991) and,

just a few years ago (1988), a Wide ranging conference

was held in Adelaide on the subject of climate change:

1 refer 10 Greenhouse “88: Planning for Climate

Change, Adelaide Conference (Dendy 1989), Finally,

T note the publication in November 1993 of South

Australian Greenhouse News, Vol. 4, No 1, the first

of what is intended 1 bea regular series of newsletters

with particular eniphasis on South Australia. [tis

published by the South Australian Departament of

Environment and Natural Resources with the support

of the Office of Enersy.

Aguiost this flucry of recent and ongomig activity one

might ask, why should the Royal Society sponsor yet

one move meeting We address the subject? What

possible good can anse from another “ralkfesr”?

The answer 4s simple, First, reseanch and views on

this subject proceeds apace and there isa constant need

Io provide an opportunity for such research ty he

reculasly aired for the benefit af the wider Commutnity,

His indeed halla decade since the last major meeting

took place in Adelaide to discuss this, mater.

Seonnd. the Reval Society is quite independent af

government depanments, research instirudons and

universities, and is therefore in a unique position —

indeed, has # speci responsibility ~ to provide an

opportunity for views to be aired which may not

necessarily conform to the party line, curfent weology

or generally accepted scientific views. As will become

obvious i some of the pupers to be given, there are

views held by some which da nor agree with

widespread views. [Sir Mark Oliphant, ina comment

atthe conclusion of the Symposiuin potuted to az mceat

arhicle in Nerare, hand, which pointed ty the luck at

any evidende af change in the presen climate. |

i]

It is important for me to add that the views of

speakers are not necessarily those of the Royal Society

of South Australia or the Institution to which the

speaker belongs.

And a third reason behind this symposium is the

constant need for the South Australian community, and

through it, government departments and politicians,

to be reminded of the need fo plan ahead tor possible

change, which may, in any event, be more rapid than

predicted. A recent article in the New Sefentist (Nielsen

1993) documents some early results of the analysis of

an ice core from Greenland which records detailed

changes in the earth’s climate over the past 250,000

years. According to scientists who investigated the core.

present models used in predicting climate change may

be too simplistic. In general, such models predict

gradual change. Analyses of the core, however, suggest

that a climate just a few degrees warmer than now may

change very suddenly to become either significantly

colder or warmer. In other words, we may be forcing

W. D. WILLIAMS

the present climate into an unsteady state when large

natural changes in climate could be triggered by

relatively small events. Whether such climatic

instability has already started is a moot point. Bear

in mind, however, that severe storms have cost

insurance companies over $60 billion over the past six

years (Leggett 1993). Meteorologists have already

begun to point io the increasing likelihood of an

increased frequency of natural disasters following

climate change in the next century (Zillman 1993),

1 hope I have indicated sufficiently the need for

and the importance of this symposium. All that remains

for me to do is to thank the speakers for presenting

their papers and the work involved in preparation. to

thank my various colleagues, in particular Dr Margarct

Davies, for her background support for the symposium,

and to wish you all a good afternoon, bon appetit for

the evening meal, and, since this is the last meeting

of the Society for 1993, a Merry Christmas and Happy

New Year.

References

Lraarrr, J. (1993) Who will underwrite the Hurricane? New

Scientist. 7 August 1993, 29-33,

ZILLMAN, J. (1993) Memorial lecture. University of

Melbourne, 7 September 1993. (Unpublished)-

Nievsen. R. H. (1993) Chill warnings from Greenland. New

Scientist. 28 August 1993, 29-33.

Denny, T. (Ed.) (1989) “Greenhouse 88: Planning for

Climatic Change Adelaide Conference Proceedings” (South

Australian Department of Environment & Planning,

Adelaide),

SouTH AUSTRALIAN CLIMATE CHANGE COMMITTEE (1990)

“Implications of Climate Change for South Australia’, First

report of the Climate Change Committee, August 1990,

(Department of Environment é& Planning. Adelaide).

_—_____ (1991) “The Greenhouse Strategy for South Australia’

(Department of Environment & Planning, Adelaide).

A BRIEF OVERVIEW OF CLIMATE RESEARCH

By PETER SCHWERDTFEGER*

Summary

Schwerdtfeger, P. (1994) A brief overview of climate research. Trans. R. Soc. S.

Aust. 118(1), 3-7, 31 May, 1994.

The problems facing those attempting to generate reliable prognostic climate models

is formidable. Good estimates of trends in concentration of all radiating atmospheric

gases are necessary and these must be entered into a tested numerical model

incorporating all of the important feedback processes. Included in these are the

parallel processes occurring in the oceans. The simplest test of the prognostic prowess

of a model is to run it backwards in time — a simple test of veracity that has not

supported any modern long-term prediction model.

Key Words: climate change, models, radiating atmospheric gases, meteorology.

Tremsaeions af te Ravel Neeters af 8 Aust 1994), TEL), 3-7,

A BRIEF OVERVIEW OF CLIMATE RESEARCH

by PETER SCHWERDTFEUER*

Suinmary

SCHWERDTFEGER, P1994) 4 brief overview of cline rescareh, Trans. R, Soe 8. ast USC), 3-731 May, 1994.

The problems taeing those attempting ts generate relinble prognostic climute models is formidable Good estimates

af pends in-coneeatration of all radiating atmospheric gases ure necessary and these must be eniered into a (ested

numerical model incorporating all af the important leedback processes, Included itr these are the parallel processes

besurring in the oceans The siiplest rest of the prognostic prowess of a model is to run it backwards in lime

usimple test of veracity that fag not Supported any modern long-term climate predichan model,

KEY WORDS: climate change, models, rugiitting wimoxpherie pases, meleornlogy,

Introduction

Over a third of a century ago, | found myself as a

freezing. fledgling geo-scientist, upprehensively

prodding the frozen crust of Arctie sea ice beluw, in

order to render advice on our mutual safety to my Red

Indian compunion, who gingerly guided the hungry

dogs that, from meal to meal, dragged the sled which

held our food and yay bare-bones scientific equipment.

We knew thal we were venturing close lo the brink

of chaos when the supporting ice became less than six

inches thick. My yelling out “150 mm” would only have

generated dangerous confusion. The metcorologival

Parameter of greatest concern ta me then was the

profile of temperature from the surface downwards.

und forthe following ten years, [remained interested

in ige-bound temperatures: in the Antarctic as well, and

pondered on how these arose, At the Lime, these were

generally regarded as actvities ambecoming of

tumospheric svientists, but with the growing relevance

attached to the new science of climate, the perceived

significance of the extent of polar ice, and its thermal

as well as physical conten) has been dtamatrcally

changed.

During the last 25 years, I started to raise my head

and examine the way in which the radiant energy

incident on the Earth's surface was transferred not onty

to depths below burt also to the atmosphere above. In

Ihe 1960s, many “real” meteorologists referred

indulgently to the “boundary layer boys” (a phrase that

the well known Princeton University theoretical

Meteorologist Joseph Smagorinsky used ina talk in

Melbourne. when he unveiled the then excitingly new

results of numencal modelling with enormous

computers, leading to the promise of operational long-

range weather forecasts). Smagorinsky was not

sneering, he well knew that the most definitive

iransformalion of energy from the Sun into other

* Institute for Atmospheric and Marine Science, Flinders

University of South Austnilia, GPO Box 2100, Adelaide

SA SOU!

radiative, conducted, convective, and evaporative forms

wikes place at the Rarth’s surface. Apait from the

conducted heat which involves the material below the

surface only, the three remaining fluxes drive that all

important distnbution zone, the diurnal boundary layer,

the thickness of which cun vary from millimetres to

kilometres, depending on the season and Lime of day.

Without the critical knowledge of the flows of energy

into this. boundary layer, reselved for a sufficiently

fincly Spaced grid covering the atea being wodelled,

every numeneal meteorologist would be marooned,

Theretine for some time, | continued to feel happy with

the relevunce of my work, delving im the lowest 10m

of almosphere blanketing the surface of the Earth.

Only ten years ago,as a result of my colleague Jorg

Hacker bringing his combined knowledge of flying and

meteorology to the Flinders. University atmospheric

research vroup. 1 became persuaded that there was

inore to meteorology than could be achieved by

climbing with instruments to the top of a Ladder. The

experience of flying offers meteorologists definitive

perspectives of the almosphere, especially when the

possibility of both making and viewing the results of

actual physical measurements in real time ullows

important physical connections to be grasped, Mare

than ever now, my heart bleeds for those meleonMogists

who spend all of their windowless lives hypnotically

hunched in front of their ceimputes screens.

The science of the weather was, in earlier years,

often conveniently divided into wetearmloxy, which

involves stitements about aunospheric conditions it

vatious given times, including the fulure, when

concepts of forecasting are invulved, and climmelogy,

which is concerned with quintessential summaries of

meteorological conditions over specified places and

periods, Applied sciences ure strongly driven by public

perceptions and wishes, in which mutters, of course,

scientists are increasingly not averse, tf not forced, ta

offer their guiding influence, The last degade has seen

the more geographical science of climearolngy fall into

relative disfayour as the persuasive value of the study

4 FP SCHWERDTIFEGER

At /anan. in being linked with the drama of inexorable

crends aad woystery al extremely loag-ieria lrecasts.

hus been heralded as an-essential fowus oh serious

allention,

Newspapers and television have popularised the

wuiject oP melenrolosy (hrough the regular presenta

av weather forecasts, and Satellite inmages Of tie Barth

hive helped grewly to generuiing an appreciation pl

the global niture ot meteamlogtieal processes reflected

in Visite cloud structures, There is no daubr (hat an

advance koowledge of denils of metearologival

conditions iy rnporimt in many areas ef human

cndewout, Over telatively short terms and for specific

ures suchas landing grounds, aviation safety depends

ui, borceastng. also influenees many other

Commercuil operations, tneluding those connected with

vaviculnure, the manofaentre oF ieeereun ind Urinks,

the saleetion oF routes tor shipping as well as personal

clnices regarding clothing and recreational wenivities,

As iitesalt, the concepts of mereralogy and forecaying

fhe become alist synonymous in the public mind

However, more general climatological summatics

also belp to guide tie public. particularly travellers,

ever (hough a Le caution im required when too bitte

detail is offered. Forexample, the bald stitement thar

Adelkude’s mean January temperature i 23°C vould

easily. prove deceptive toa nuive visiter, when itis

known Ut (he inedn maximum fempenidture is 30°C

und rhe mean minimum 169CL with extreme values ot

these (wo parameters of almost 48°C and 79C hay ing

feed recorded, Equipped with this additional

Information, the intending traveller miht obtaji seine

dew as to what types of clothing te puck An even mure

useless piece ob “stand-alone climatological

(iteration, exgep! pechaps to wind makers needing

ju be usstired (hal (he teniperatures miintuined im leep

cellars aie benefienl to the maturation af their

prodives. ms that Adclades mean unmaal tempensnture

iy 179 Responsible aind useral chmitolegieal

comipilitiuns Usually ain te provide envelopes tor

monges ol vitlues within which porumelers of interest

might be expected ty lie, The choice of these

panimeters as well us the time scule wall depend an

the purpese ail the application. For example. tor

avrivolionil decisions, information about vasefell,

tenperulare, volar rudtton and winediness Gneludine

likely extreties) ts important. [nthe Hight of the present

Knowledge of the Earths regional clinuattes, ub is cisy

ib overlook the aehievernent of colunial planners such

a6 Colonel William Tight, the founderof Adelaide,

who with Sis Weeks of arrival. barely In advance af

the Hirst Puropedn setilers, was able to assure hunselt

as la the general hature of Adelaide's climate and te

realise (Nal ih was the only location near the oust of

what hid been desiznaled us South Austmilia where

a city based an 19th eenfiiny technol ooulh be

estublished.

The need for information on weather in general

terms Js readily accepted for seasonal forecasts for

agricultural and water resources planning, Inevitably,

because Of the: tnstorically observed hok between

protracted extremes in meteorological gonditions and

the economy, the possibility oF rebably uscertiininy

long-term trends. is an auractive propasivon, Therela

he the origins of (he clamour for, and the justification

of, modern cliniate research,

Meteorologioal processes vccur on many Uillergent

limie-seales, The Iwo most fimilar of these are ihe

varhitions in insolation whieh result in the annual cvele

al’ seasons and the daily msing and setting of the sun

Both ane associated With systematic insolation changes

which resull in every olher meieoroloyienl paramelyr

EXporiene ie coMecquent variations. Chanwes mpodher

lime-seales ire fat less predictable. At the breh

frequeney dnd, jurbulenve is best desembed by sing

stalishicul expressions, Low frequency meteurologieal

changes hive beea discovered in geological studies of

the history wl the Earth wath distinet climate events

being separated by frony thousands to millions of years

On this scale, goalogists und glactologists have lony

led meteorologists with the Intensity of their interest

Systematic periagie meteoralogical changes van he

sumimarised by the climatology of the perids under

consideration, Pooe to the development of modern

methods of reconling meteorological valucs, the

deseriptions of historians. including the authors.of the

Old Testment olter investigalurs of past clhimutey a

weulth of iMformaben, hocexwnple, during bis years

in [evpt, by warning of the seven years of famine te

follow years O) plenty, Joseph issued his Pharaoh with

the first recorded suvecesstul lone. range weather

forecast. Phe Lirle lee Age ot the Middle Ages is a

well recorded historical fact, casily verifiable through

s(udies ol the welvance and relreala Alpine glaciers.

Only in ihe 20Ur century, have mine-shulls. lust worked

some 600 vers avo. become visible aus with the

retreat ol these alaciers which begun oy the [9th

centuyy, Modem visitors to the “Belvedere Hotel"

whieh. a hundred years ooo Urinatieally ever looked

the Rhone Glacier in Switzerland. muy well ponder

on how Jong powerful binoculars will suthiee (a offer

a dimpse of the ghicier’s snout following its dramatte

recession, Tt wits this same “Middle Apes” ace ape

which contributed, substantially: to the elimination ot

the Viking colonies from Greenland at about the same

time that Christopher Columbus was busily diverting

Eivepean atention to warmer parts of the New World,

Bryson & Murrey (1977) offer accounts of climaly

changes und the cunsequenves thereel we various

vilisutions over several thutsund years ol documented

history.

There is #bsolutely me coubt that dre clamite oP (te

Fanh is amb has always been changing (Brooks 144)

Haawever whit his wlso been changing during (he lest

A BRIER (VERVIRW OF CLIMATE RESEARCH 4

Iwoilevades iy the pubhe perception of the nature of

climwte chanes and concern thal the contribu bo

tbese by human activines may greatly overshudow

nalural ramlom trends, Quick td respond to such

voncerns. many alinospheric screntists have xoghe te

qoanoly and offer physical explanations for both

primary wbservations dnd vssumed consequential

changes, Chanwes of climate parameters necureine

“quturally’. are un seemingly randori tine scales

because oF the dificrenc rates governing the physical

“teedback” systems which link most, if net sll.

processes in jhe atmosphere, Phe vast tinge of these

pracesves, uod theit intemetions, wher the perfect

regipe Tor chaos,

The subjection of past elimalie data to frequency

analysis in the hope of obtaining eycheally repetitive

iessaies has been in var. My Flinders Universily

colleagues, John Bye. Rolwnd Bynim-Scat and Adtian

Gordon have demonsiraled tis phenomenon of

randommcss in-an unusually ditecl way. They have

translated random levels ol met-insatation. by means

of a simple, physical computer model in whieh

atmospheric and oceanic momentum are plausibly

coupled. inte time series of air amd sea leniperature

which over a period of several simulated centuries

display an uticanny superticig) resemblunce to actual

observations.

Were the Kanth to be 4 simple, heat conducting

hollow shell, simalur perhaps to carly urpfietl

satellites. but not even containing asuffering dag, hen

meteorology would be simple: the surface tempeniture

would be the only reportable factor, and i could be

walculated front a KnoWledge of the sofar cerisreut

(which in fuct is pot quite constant) and the measurable

radiative properties al the satellite surface The resull

iscommonly feterred to as the planetary lemperaiure

and the (wo importint planetary radiative properties

qwhich this temperature is related are the albedo or

surface reflectivity toward the wavelengths of solar

radiation, Which wre predominantly sisihle and the

planet’, emissivity, a factor which determines the

efficiency of theexport of radiative energy, mainly in

the infra-red minge of wavelengths. out wo space

Otherwise, this loss of radianr energy is determined

only by the temperature, a feet discovered by: Newton

three centunes ago and quantified by Stefan und

Bollamann late last century, This relative

meteorological simplicay is further enhanced by rts

instaneinenty: a thin hallow planetary shell cannot store

uny heat enerey and the coupling of tempertare and

maididtional events ogcurs without any delay.

By virtue of its samosphere, not to mention its sulid,

heat sluring mass, out real Earth has made this simple

business far more complicated. For cxample. the

diurnal delay between maximuns insolation at che locstl

geoxraphical noon and the muximiun air lemiperature.

which may Occur 2,5 tours later is uligest entirely

ascribuble to the thermal auimittanee of Conductive-

cupueity of che ground. ‘The Earth's surface usell does

not have & unitorm albedo; this varies (nim 3-4 % over

large parts of the ucean tu 80-90% over freshly fallen

snow, The tevel af heat storage depends in the leat

espacity and thermal conductivity of the material

beneath the Earths surface, but these couductive

processes cannot compete wilh (he efficiency of both

horizontal ued vertical energy transler mechanisins

within the Earth's dynunsie aunmsphere. Viewed from

space, the clouds which result from the presence of

wiler vapour: are the most dominating and reflective

feature. Clouds are larzely responsible lor the Earls

ayerave albedo being about 34%- As in the case of the

hypothetical, hollow, attiosphereless planet discussed

carlier, he 66% of the incident solar enemy which

is accepted by the Earth can alse be usel as the basis

for the galeulanon of a steady state plimecary

cemperiture; in face this turns out to he jhout 229°C.

Although this figure has real meaning i Lerms of the

Earth's temperature when observed fram outer spite,

it js difficult, unless resting ona polar (ce floe, to

appreciate Uy signilicanee hack on Balth rselP where

the wreally integrated mean surfice ten iperalure js about

I6°C The corresponding figure in upper almospherie

levels hich still emit significant amounts of infra-reé

radiation 14 about -f0%C. ft is essentin) t realise whit

the planetary tempermure must be regarded as 4

radiatively weighted nian over the entire thivkness of

the ulmusphere.

Vhe aumosphere, becuusc it contains poly-atonnic

gases which partially absocb and re-emit their

characterising bands of infra-ted radiation which the

Earths surluce ranstorms from absorbed inculent solat

radiation, greatly modifies the simple, 2-dimensiunal

picture of planetary temperature, Sympson (1927, 1928,

1929) was one of the first tu quantity, ia tetms ot

spectrally selective absorption, the Greenhene Lfjecr

which is cssentially responsible lor the mean

temperature ur the surface of the Earth being

approaimately 379C warmer thao tie plinetary

temperature. The over-whelmingly important gas in

this radiative process is water vapour, H,O-

Understanding, how this process contributes tu the

distribution of tenspetuture within the aimosphere is

diffivult enuugh in itself because the concentration af

this vital vapour varies with beth location and ime

Is Cuncentration und temperature in tum decermine

the existence of clouds, Which increase the reflection

of in-cunnog solar radiation, thereby generating the

fundamentally important feedback process of the

Earth's. surface becom shaded below, thereby

reducing its emiperiture and consequently also the level

of evaporation. All of hese linked steps occur neither

instantaneously nor ar ralés which can be uniquely’

specified The physicul clarity of the entire progess

which as most really comprehended in a single,

fi P SCHWERDTFEGER

vertical, dimension, becomes obscured by large seule

horwontal wvective processes, of which (he winds near

the pround and jnovenent of clouds above are the most

obvious manifestations, The variability of HQ i our

altiosphere results from the fact that it readily changes

trom yapour to bquid and even sobd phases within the

range of temperatures encountered cin Earth, all forms

havent strikingly ditferent mobilines.

There are ober gases Which comribuie to the

ainaspherie Greenhouse Effert, which t should be

emphasised, % significantly diferent from the marke

warden (ype. the litter having enclosing impermeable

inembpanes (6 inhibit both convective jind wyter vapour

lusses fram the systens, The aduitioniel gases are alse

paly-atunite Whar is they have three or more atoms per

NOleeole and are present in concentrations which ate

Small compared ta (hat at water vapour, In the absence

of stranaly locilised sources of productionind of sinks,

veiich absonh them. these gases are considered to be

umtermty axed with the other dontimating gayes of

the almosphere. such us oxygen and nitrogen. All of

the early mihalion toodels ol the atmosphere in

Iovussing an the importance of water vapour. deigned

io consider anly carbon dioxide, CO, as being of even

secondary uportanee in their schemes. Other

members af the ulmospheric Emily ol poly -atomic

gases. (luding Methane. oxides of nimogen and ezone

were reparded as nepligihle in influence. The

importunce accorded these eusey churned dramatically

following the first lew years of ahservations, when

these hevume available during che 70s. drum se global

nenwork Ab menor sratiens which detected steady

mereases Th the Mean unnual coneentrarions ot

particularly CO, and CH, in the atmusphere. (ce.

M.LT. 197)) Those are readily explicable in terms oF

the ever inereasimg vombustion of firssil fucts. the

Hling and bumog of the world's remorselessly felled

remnant forests, and the Tatulence of the world’s

burgeoning hents oF tame ruminants. [ots reasonabic

to dewept that with significantly greater concentration

levels. these gases should be ieluded i the weqountings

of Gallelive energy iransfer aid heating. Phe success

wilh which this can be done alse depends on the ability

lo idemly the mature and magnitude of all ot the

lvedbavk processes

With the acknowledged dmercuse mM the “lesser

fHidiatively Thportint gases, urises the ynestion as ta

the Wieans Whereby Une atiiosphere is also able to

continuously shed part af (he crneentration, That this

ich always haye heen so ws clear since CO, wid CH,

huve both been released ti the wir ik leis) since lite

began on Earth. even in tines when their mean

concentrations Were relatively stuble. (da the basis ob

current estimates of the alobul rates of CO,

Tesssta; Moo1 1 f1962)) PHD, Thesis, Metecrolugy

Duparhovet University of Melbourne Unpubl

production and krewledge of tsuctual concentration

in the atmosphere, itis a simple matter to eateylate

the rite al which one af the Earths leedhack systems

is dewling with the “fising” of ever increasing lowds

of this gus by, for example producing carbonates as

uresull Of the dissolving of the CO, tp water, This

has lune been w fertile Geld of tnvestization lor

teulovists. However. it is now apparent thar this

process is nat coping with all af the CQ) currently

being produced. so thut is atmosphere concentration

is CONLINUINE to increase,

in suinmary, the problems facing those wiiiig Wo

generate a reliable prognostic chimate model are quite

lormulable. His necessary to have geod estimates uf

ihe trends jn coneentration of all mUditing atmospherte

guses and (ensure thal (hese ane entered! oto a tested

numerical model whieh incurporales all of the

importum feedback processes made even more comples

hy the upprectation in recent! yours that models of glohil

atmospheric circulation cannot function without

consideration of the parallel processes necurring pn the

ocedns, The simplest test ol the prognostic prowess

ot ainadel is to run it buckwards in time. When Uwe

Radok and Dick Jenssen deyeluped Australia’s lira

numerical short-term weather forecasting model il

Melbourne University in the late 19505. (Jenssen

192"), they used thiy simple and Honest procedure.

‘Yo the best of my knowledge, no modern long-term

climate prediction model his yet withstood this simple

(vst of veraeay:

In spite of the fact that one stmke of a US

presidential pencan huve Qrealer vunsequenees for the

Australian agricultural economy. thin even some of the

mare adverse predichons ollered by the proponents tt

climure models currently im vogue, | believe thal

climate resewreh is importunt. | alse believe that some

of the recommenditions af those whose faith iq curretit

Inodel predictions is clase to absolute, in which 4

radical reduction in the global sesle of combustion ts

urged, deserve substintial endorsement, particularly

where this may lead toa deceleration of the senseless

destruction of the world’s forests. Dihers may applaud

the development of niore vegetarian habus hy the

inereasinaly carnivorous human pace, whether or nor

aw reduction in vlobal catile numbers has any ulimate

significant impaeton climarcce wends. J do, however.

lind it regrettable, that the stranger proponents ol

climate research have dispropertionately triggered i

wide Tuoge of polineul imochurisinys by means of

potentially, alirming information. without proper

substantiation, apparently to generate and maintain high

levels of public attention. So fir the nwin result appears

Ww bave been to generale public confusion bemween thie

concepts underlying the Greenhouse kffect and Une

Ozone Mole T well recall two remarkably different

lectures held withia the space of a sitgle decade

id the seey same Universite of Adelaide. by ai

A BRIEF OVERVIEW OF CLIMATE RESEARCH 7

internationally acclaimed visiting atmospheric scientist.

The first of these chillingly fortold of an impending

Ice Age, the second unabashedly warned of Greenhouse

warming, Of course all scientists need public attention

in order to survive, but for credibility to be maintained.

science normally needs to disclose verifiable results.

even if these may require interpretation in a manner

conducive to greater public appreciation. It is

dangerous for any branch of scientists to test popular

indulgence too far and in the case of atmospheric

scientists, to allow the public to lose sight of the many

other useful purposes their scientific endeavours both

continue to and could be offering, particularly in non-

long-term prognostic ventures. With public and

political attention rivetted on such matters as burping

bulls in Brazil and the saturation of the ever expanding

seven seas with soda water, it is easy to overlook the

multiplicity of climate-related environmental problems

closer to home, where appreciation of diverse themes,

ranging from the security of vital water supply

catchments and the impact of large-scale land

management policies on regional climates (and vice

versa), to detailed investigations of the actual physical

and chemical processes constituting the links of the

complex chain of events loosely described as climate.

are vulnerable to chaotic diversionary forces which

thrive in confused societies.

References

Brooks, C. E. P. (1949) “Climate through the ages: a study

of climatic factors and their variations” (Dover facsimile

edition, New York 1970).

Bryson, R. A. & Murray, T. J. (1977) “Climates of hunger:

Mankind and the World’s changing weather” (University

of Wisconsin Press, Madison).

M.LT. (1971) “Inadvertent climate modification (Report of the

study of man’s impact on climate)”. (Massachusetts Inst.

of Technology Press, Cambridge).

Simpson, G, C, (1927) Mem. Rov. Meteorol, Soc. 2, 16.

(1928) Ibid. 3, 21.

_____ (1929) Ibid, 3, 23.

MODELLING CLIMATIC CHANGE AND VARIABILITY

By ROBERT J. ALLAN*

Summary

Allan, R. J. (1994) Modelling climate change and variability. Trans. R. Soc. S. Aust.

118(1), 9-15, 31 May, 1994.

Recent interest in climatic change has engendered considerable debate about the

enhanced greenhouse effect and its possible global impacts. Numerical computer

models of the climate system are important tools in the scientific assessment of the

enhanced greenhouse effect. This paper briefly reviews the model development and

approaches used to simulate the nature of anthropogenic changes and natural

variability in the climate system. Research in Australia is given particular emphasis.

Key Words: climate change, numerical computer models.

Trmmsaorions ih the Reval Sueiely ofS, dash (994), OBC), 91

MODELLING CLIMATIC CHANGE AND VARIABILITY

by RORERT J. ALLAN*

Abbas Rod (199d) Modelling climate change and variability, Vans, Ry See. So Aust, UBC. Y-15, AL May, 1994

Recent imerest 1 climatic change has engenderec. consideruble debate Whout whe enhanced greenhouse effect

anid ils possible global impacts. Numerical computer models of the chinate svsicu) are tnporuint fools in the

aulctititic wssessinent of the enhanced greenhouse efleet. ‘Thos paper briefly reviews the model development and

approaches used (o simulate the nature nf anthropogenic changes wad matupal variability an the chimate 4ystem.

Research w-Australia is given particulas eniphiois

KEY WORDS; Climate change, numeridal computer jeadels

Tntroduction

he wlobal climate system involves closely linked

interactions betyeen the atnosphere, oceans,

eryosphere and the biosphere, This system is doveu

hy energy derived trom sulur radiation, with the net

ilvoniing energy being balanced by that which ts fost

ta space, The nonlinear and highly dynamic climate

syste is at consequence of the redistribution of this

energy by thermodynamics and. the forces of motion

derived from the planet's rotation. Any Muctuahions or

changes in the climate system can be brought about

hy external foreing. or by provesses and readyustments

intrinsig to such a closely Coupled nuntinear syste.

Since the systein is finely balanced, what might appeir

whe minor changes. inany part of the system can-cause

large chunges in its character.

Until the latter part of dis century, utempts 10

understand the climate systent, and the variahons- and

changes in il, were puryued through observational

studies and the development of physical theories ol the

interactions and motions voverniny it, With the advent

of, und improvements in, computer technalogy, efforts

Jo produce computer models oF the climate syste have

expanded rapidly. Mueb initial modelling work focused

on the need to miprove day to day weather forecasts

usita models capable of resolving, synupuie scale

tealures in (he atmosphere. However, the longer term

envelope of weather events that constitutes the climate

has now become a significant areca in numerical

niodelinog. Concerns about the enhanced greenhiise

effect and a need to understand natural fuctuations in

the eliniuite system have seen u strong focus Ga The

development of models able to capture the progesses

governing interactions inherent ii the climate system.

Such interactions are shown schematically to Fig. 1.

This paper ptovides an overview of the important

characteristics of numerical models that have been used

to improve the scientific understanding of changes and

Nuctuations in the climate system, It foguses on both

Climate Impact Graup. CSIRO Division pf Atmospheric

Research, PMB No. |. Mociialloe 3195. Vietoria,

international efforts and the progress being made in

Australia 10 examine potential climate problems using

numerical computer models.

Low order models

Some models. have been developed lo improve

Understanding of particular features that dominate the

climate over regions of the globe und for various

periods oltime. Such models use what are often culled

“primitive” forms of the basic physical dynamics that

are necessary to encompass the processes underlying

climatic phenomena. These computer models contuin

sets af mathematical equations describing the most

important physical interactions at each point on

horizontal and vertical grids of pots that cover the

region in which the climatic feature oceurs. All the

equations governing, the physical properties of the

phenomenon are solved at time steps indicative of the

observed nature of the pracesses being examined, Ln

thts way, a model of the real system is built up.

However, such models are only as pood as the

mathematical resalution of the dynamics of the feature

being modelled, Some real processes are relauvely

casier to describe mathematically and ther mieractions

are known. Others are difficult to capture fully, or the

present physical understanding of then may be limited.

Tr general, Jow order models dave focused on resolving

the important atmospheric and/or ocewnie character of

a particular climatic phenomenon,

Perhups the mont concentrated emphases in this ares

has been in the construction of low order models aimed

at capturing the essence of EI Nifio-Southern

Oxcillation (ENSO) events, The ENSO phenomenon

is a large-scale ocean-atmosphere interaction. that

occurs uregularly. dnd is centred in the Indo-Pacific

basin. It involves a Cloye coupling of the important

features of the climate system across this regton aml

has significant impacts in the marine and terrestrial!

environments. Some of the maw variables and

feedbucks Known tu operate during ENSO events are

shown schematically in Fig. 2, and a review of the

phenomenon can be found in Allan (1991).

10 R, J. ALLAN

Cr

a

Changes of }=9=-—~ US

solar radiation us “f 4) -

H '\\

Terrestrial

radiation =

H2O, Ny, Os, COz, O4, etc, Clouds (

aerosol Ls sy Y

Atmosphere

Atmosphere-land coupling

Biomass

VW ’

Fey Sea-ice | wy *1 Wind stress

Atmosphere-ice coupling TA pT

: |

A

Ice-sheets, ,

snow Heat Precipitalion |

Fee = Evaporation

Nee ral MN) cl Otel fs exchange Pp

Ice—-ocean A h fi Ocean

Changes of coupling tmosphere-ocean coupling

atmospheric composition . LaF

Changes ol land fealures,

orography, vegetation,

albedo, etc.

Changes of ocean basin

shape, salinity, etc.

COUPLED OCEAN-ATMOSPHERE

ATMOSPHERIC CIRCULATION

aan

siomen Uy

(RAINFALL )

SURFACE WINDS

=— — > ee

CURRENTS => > => SEA-SURFACE

wr ates TEMPERATURE

NESS POSITIVE

7 pYER THICK FEEDBACK

NEGATIVE

FAST RESPONSE FEEDBACK

OCEAN CIRCULATION

SLOW RESPONSE

MODELLING CLIMATIC CHANGE ANI VARIABILITY Il

Most low order tnodels of ENSO have locused on

the Pacific Ocean region botnded by the suimrapics

in_bath Iyemispheres, Such models are very sunple,

with ninimal physical representation of the almesphere

(often only two layers in the vertical) and a constriction

designet! lo resolve (he basic dynamics of the ENSO

phenomenon alone. The building of models in this

context bas required substantial resources and

particular recourse to ahservational studies and

dynamical theories whieh provide a conceptual

lramework of the physical processes likely to underlie

the phenomenon. At present, there is au firm theory

describing all aspects of ENSO events, but rather a

dynamically consistent understunding of seme of the

IMporiant processes Governing its behaviour. Curren

models of ENSO have been built under this constraint.

More recent developments in low order coupled ocean-

almosphere models have scen improvements in the

ibility of these tools ro predict ENSO behaviour The

mist comprehensive review of the current slalus of

ENSO jodelling as given in Neelin er al, (1992). In

Australi, such work ts perhaps best charucterised hy

ihe law order model developments at the Bureau of

Metcorology Research Centre (BMRC) (Kleeman 1991,

1993), Further propress cun be expected as the

underlying dynamics powering ENSO episodes

become better understood. wd the vapabilidies of

models to represent such Complex processes develop

turther,

Gencral Circulation Modelling

More detailed numerical models. for climatic

research are the General Circulation Models (GCMs),

which are three dimensional models of the global

climate system. These models represent interacuions

between features of the climate system through a series

of mathematical equations on a spherical grid. The

grid, representing a latitude by longitude army, is

repeated tn the vertical plane aver the desired number

of levels or layers throuyh the depth of the simusphere.

This structure is shown in Pig 3, and provides: the

Irimework within which the model simulates the

dominant processes of the climate system over cime.

Physical processes and interactions are calculated at

euch grid point simnultancously in the active maul.

In addition, in some models important boeanic links

to the GCM atmosphere are mimicked by including

a“slabyocean” in Lie model. This iy a 50 m deep layer

covering the Eaah’s oceans that performs like a well-

muxed oceame “near surface” layer. However, sea

surface temperatures simulated by this approach most

be continuously corrected to take uccount of the hear

transport by hortvonlal currents and deep ocean

processes. This is termed w Q-flux correction to the

GCM, Using this technique and given the constraints

mentioned, un evolving simulation of the Earth's

climate is produced by the model that captures many

observed aspects of the climate system, Detailed

specifications of the CSIRO Divison of Atmospheric

Research (DAR) and the BMRC GCMs that haye beet

developed in Australia, are-given in MeGireger rf al

(1993b) and Hart er e/, (1990).

Despite the relatively sophisticated nature of GCMs,

they are coarse in their spatial resolgtion with

endpoints spaced al arqund 500 kin apart and vertical

levels representing the atmospheric structure usually

limited to between 10-20 layers. As u result of these

Spatial restrictions nposal by limilalions of computing

resources, many smaller features of the climate system

such as tropical cyclones and deep cyclonic systems

cunnot be resolved by GCMs, Another importunt

problem that stents from ihe spatial limitations is that

some physical processes, such as those relating la

uspects of cloud physics, cannot be fully resolved and

are parameterised in these models. These factors limit

the current capabilities of GCMs to capture the full

inincacies of global climate. At present, for example,

GCMs are unable w produce realistic ENSO events.

Nevertheless, it must be realised that these are early

generation models and that many of the large-scale

aspects of the real climate ere seen in the current GOM

simulations

An important advance in the development of GCMs

is the Coupling of atmospheric and oceanic models

(fortnulated similarly to the-GCMs) in order th) capture

more of the essential elements responsible for observed

climatic patterns. Currently, such coupled madels

require correction for fluxes of energy, miinentum and

moisture exchange berween the atmosphere and the

aceans if they are to resolve full ocean-almosphere

imeractions. Mast simulauons of enhanced preenhuuse.

effect conditions with current coupled ocedn-

almosphere GCMs have used {ux correction,

However, as ned in the section on low order models.

coupled models provide the basis for a more reulixuc

Fiz, | TOP. A simple schematic view of the major interactions in the global climate sysiem.

fig 2. BOTTOM, A schematic diagram of ENSO interactions involving major oceanic and atmospheric yanables in the

Pacifie Basin. Whe lwo responses (fiat and slow) in the ocean refer in differen feedbacks rosulling fran che natire ol

the dynamical waves generated by tropical oceunaumesphere teractions inherent in the phenomenen

12 R J. ALLAN

a. sea

on the surface

Fig. 3. A schematic representation of a GCM as i divides the earth inty a regular three-dimensional grid.

7.0

= 0.8

i

3

® 0.6

=

oO

bd BMRC

s a4 CSIRO

= CSIRO4

o

we GFDLH

ccc

UKMOH

0.0

DJF MAM JJA SON

Fig. 4. Pattern correlation for seasonal yalues of model

simulated and observed mean $ea level pressure over the

Australian region. The abbreviations are for the GCMs and

BMRC, CSIRO (4 and 9-levels in the vertical), United

States. Geophysical Fluid Dynamics Laboratory (GFDL)

(High resolution), Canadian Climate Centre (CCC) and

the United Kingdom Meteorological Office (High

resoluhon).

representation of the real world. Oceanic GCMs have

a similar structure to those described for the

atmosphere, but must represent a more “sluggish” tluid

with spatial and temporal scales of processes and

interactions. that ate slower than in the atmosphere. It

must also be remembered that models of the

atmosphere owe much to a Jong history of scientific

endeavour to improve weather forecasts. Oceanic

model technology has lagged behind and these models

require considerable further development. Much of this

is dué to sparse Observational data coverage for the

aceans, which limits the validation of model output,

and the need to resolve smaller scale processes

involving meso-scale eddies in order to achieve more

realistic simulations. The status of coupled model

development at the Australian Bureau of Meteorology

is given in Power ef al. (1993), Similar development

at CSIRO’s Division of Atmospheric Research is ina

rather more advanced state.

In using GCMs for studying variations or changes

in the global climate, one important prerequisite is that

the models can perform an acceptable simulation of

current climatic elements. A number of

intercomparisons between various GCM simulations

of climatic fields and observed climatic parameters

MODELLING CLIMATIC CHANGE ANT) VARIABILITY 3

have new been perlormed by reseurch groups around

the world (Houghton ef af 1990, 1992), Specific

intemumparisons for GCM simulations aver the

Australian region are detailed in Whetion & Pitoek

(1991, 1993) and Whetton er af. (1994) (Fig. 4). The

most encouraging aspect ot these checks. is (hil they

have objecuvely revealed 4a general improvement ub

GCM simulations as model resolution and physics have

heen refines!

The procedure employed to sei up a model run ta

produce “contra” Ar present climate conditions usually

volves an-experanent where the GCM as run forward

im line under ihe current wlmosphene CO,

concemiration with observed sea surface lemiperatures

und initial ptmospheri¢ input fields, The aim of this

lype of éxperinient is to obtain a sanyple of simulated

climaly lang enaugh for statistically valid analysis. In

practice. this normally requires between 10-30 years

of simulated chmate variables. The various elds

generated ean then be examined using cureful staustical

techniques thir check the pattern of cormlations

between exch of the GCM: control uutpur dite

distributions with the appropriate observed field

patterns. Sivulated fields can also be examined for grid

point and total spatial freld significance, If this proves

lo be satisfactory. then the GCM can be used to address

the possthle change ta climate that could result trom

the enhanced greenhouse effect

Two Lypes of experiments are commonty conducted

to-assess. possible climatic change duc to increasing

CU, The first, and most widely employed technique

mine in which the experiment is periomied by simply

running the GCM with doubled CO, These

equilittrium simulations require about a 10-year time

peciod for the GCM to adjust to the new CO, limit

alter itis instiniunedusly imposed, and then a sample

of the next 20-30 years of data is ken fo represent

the enhanced areenhouse climate stutistics. A doubling

oF COs Is a standard experiment used as a benchmark

ugains! which all (XCMs can be compared. It does nat

refer ly any particular dale in the future and implicitly

\heluides ‘the radiative effect of increases. in other

greenlmuse pases, which are expressed as

CO ,-equivalents, The more physically realistic

approach, which ts being undertaken as coupled necan-

atmosphere modets begome more widcly available and

computing capabrliaes improve. is tw perforant a

tunsient experiment. In this type of simulation the

CO, ithe model atmosphere is gradually increased

Tha manner which loosely appruxemates the proyected

increases in this gas, lo a time when the level has

doubled.

Intemarional evaluations of climiute change research

and GCM siniiitiens and intercomparisons arc

compiled in the reports of the hnfergovernmental Pancl

on Climate Change (POC) (Houghian er af 989(),

W921 Th Australi, GOM research with “on hose”

models 1s concemrated of BARC and CSIRO Division

of Atmospheric Research. Both groups have produced

doubled CO, cquillhniun culis using impreved versions

of their GC Ms, und are dewelopmg GCMs tor transient

experumenis (Power et a/. 1993) and te simulate ENSO

and patueral climate variability, At the CSIRO Division

of Atmospheric Research, resulis of enhanced

ercenluuse effect experiments have been used as part

of the input inte regional climate change scenarios for

Austealia. Tiaportanl contributions io IPCC evaluations

have been made by (SIRO'S Divisiun ul Almospheric

Researeh through studies examining the resoludon of

possible changes in rainfall amaunts ind intensities in

doubled CO, simulations (Gordon et a, 1992), This

anup alsa has tepaned un the nalure ul variahilily

within the climate system sunulatcd hy duuhled CO,

experiments (Gordon & Hunt 1994) Climate variahiliry

studies ai CSIRO include those using the recent

Atmospheric Model lotercompurisan Project (AMIP)

results Io highlight the ampact ol chaotic influences on

the atmosphere (Dix & Huint 1994) and an ongoing

experiment simulating 300 years of climate

Limited Area and nested modelling

The neal po focus on likely changes un regional

climates, together with the poor spatial resolution of

GCMs, trewns thal models that can resolve smaller scale

processes are required, Limited area models (LAMS)

have finer spatial resolutions than the GCMs and can

resolve Jeatures of orography and geography thal arc

smoothed .in coarser global models (Giorgi 1990; Giorgi

et al, 1990, 1992), Typical horizontal grid resolutians

that have been achieved by LAMS in Australia alt as

Jow as 60 km with IS levels in the atmosphere

(McGregor & Walsh 1993; MeGregor er al. 19934).

LAMs are constructed to cover limited regians af the

Earth's surface and to focus on specific feaures of the

chimale system while being computationally economical,

Such models can be used to examine how: a particular

synoptic system, such as a tropical cvelone. may hive

behaved 1 it were in have been wdlueneed hy different

environmental conditions (Evans 1993; Evans eral. in

press}. For enhanoal greenhouse experiments fis aisy

involve increasing, sea surlisce temperatures in mimic

doubled COj conditions snd then tesung ihe sensitivaly

of the tropical cyclone anil the amount of rain i) proaluces

ta uch a change in boundary conditions. These medels

are particularly useful in ussessmy the sensitivity of local

preciptation-producing systems ta @ changed

environment. Juint research by BMRC and DAR has

Concentrated on the simulation of “east-coast’ lows ov

what are often termed "cut-off" lows that can produce

copious rainfall aver southeastern Australia, This cu-

operaive study is examining the effect that increased

Sex Surface Pemperslujes May have on tle precipitalion

regime of these synoptic systerns (MeTrines et al, 1992)

it kod ALLAN

a —

7

ee TEPrr eben,

a ee

pee a)

=

Fig, 9 Schematic showing the nesting-ol the LAM gridowiltin

the CSIRO GCM grid over Australia,

An important new development using both GCMs

and LAMSs is the combination of the two tn what es

termed a “nested” model, The GCM generutes the

information trom a full globsl simulation and provides

the boundary conditions for the GAM that is embedded

or nested within the larger, coarser model. This ts

illustcated in Fig. 5. The nested modelling approach

is capable of resolying features as small as

60 * 60 km. Recent research at the CSIRO Division

of Aunospheric Research has focused on the use of

this approach ta produce finer scale regional resolution

of climatic features and their sensitivity te changed

eavivonmental conditions. ‘So far. LAMs bave anly

been ncstcd in equilibrium boundary conditions

produced by GCMs with a simple Q-flux corrected

ocean model, As such, these LAMs. ure limited in their

performance by the accuracy of the GCM boundary

cundilions,

Other types of models are currently being used or

developed internationally and in Australia. One

interesting, application is the use of tracer transport

tnodels to examine the atmospheric CO, budget,

Studies ut CSIRO DAR using such models are

described in Enting er al, (1993). Such work is

purticularly important in helping to estimate the global

carbon budyet.

Conclusions

Over the last five years, progress in developing

numerical computer models of the climate system has

heen substantial. mainly as a result of the stimulus

provided by concern about the enhanced greenhouse

effect and the need to investigate its possible globat

impacts, Today a suite of numerical models ts being

Weveloped i juckle various problems relating to the

climate system. Specialised low order models focusing

om particular features of the climate systent are best

exeniplitied hy models of the ENSO phenomenon.

Such research has already been responsible for

significant improvements io the torevasting uf ENSO

phases. Larger atmospheric and oceanic GCMs are

used to-simulate the global climate and hkely changes

and fluctuations in climatic pallerms due to the

enhunced greenhouse effecr and nariral variability,

Efforts are also progressing with the full coupling ol

the oceun-atinosphere system und the addition of

realisne und interactive cryospherie and biospheric

elements. Finer-seule resolution. in the numerical

modelling of climate has been attempted using LAMs

that have been nested in the coarser GCMs, or used

as “stind alone” models cupable of resolving specific

synoptic features and their sensitivity to: potentiul

changes in the climate system.

Considerable further research is needed to continue

such developments and to take the current generation

of models to a stage where fully interactive Earth

system representations are possible. Such

improvements then open up the possibility of

simulating the nature of fluctuations and changes in

the global climate system more realistically. However.

these developments must be linked with research to

improve observational data records and understanding

of chmate dynamics, so that model resulis can he

assessed in their proper context.

Acknowledgments

J would like to thank Dr-Chris Mitchelland Messrs

Barrie Hunt and Kevin Hennessy, CSIRO Divison of

Atmospheric Research and De Janette Lindesay.

Department of Geography, Australian National

University for discussions and their constructive

comments on the initial manuscript of this paper. This

work contributes to the CSIRO Climate Change

Research Program «nd is part funded through

Australia’s National Greenhouse Research Program.

References

Auian. RJ. (991) Austrtlasia pp. 73-120, Jn Glante, M.,

Katy, BR. & Nicholls, N, (ds) “ENSO: Teleconnections

Linking Worldwide Climate Anomaties Sujentific Basis

and Societal Impacts” (Cambridge University Press.

Cumbridge, UK).

Enting, 1 G., Tropincre, C M, FrRancoy, Ro I &

Graneék., H. (1993) Synthesis inversion of ulmospheric

CO, using the GISS tracer transport model, C§/RO

Division of Annnsypherie Rescarch, Wil, Map. Na, 29,

Melbourne +44

MODELLING CLIMATIC CHANGE AND VARIABILITY 15

Evans, J. L, (1993) Sensitivity of tropical cyclone intensity

to sea surface temperature. J. Climate 6, 1133-1140,

. Ryan, B. F & McGrecor. J, L. (In press) A

numerical exploration of the sensitivity of tropical cyclone

rainfall! intensity to sea surface temperature. [hid.

Giorct, F. (1990) Simulation of regional climate using a

limited urea model nested ina general virculation model.

Ibid, 3, 941-963,

, Marinucel, M. R, & Visconrt, G. (1990) Use of

a liniited-area model nested in a general circulation model

for regional climate simulation aver Europe. J. Geapitys,

Res. 95, 18413-18431.

. de (1992), A 2 x CO, chimuate

change scenario over Europe generated using a limited area

model nested in a general circulation mode! 2: Climate

change scenario, Jhid. 97. 1OOU-10028,

Gorpor, H. B, & Hunt, B, G. (1994) Climate variability

within an enhanced greenhouse simulation, Climate Dyn.

9, 195-212.

. WretTon, PH Prrtock, A. B., FowLer. A. M.

& Hayiock. M. R. (1992) Simulated changes in daily

rainfall intensity due to enhanced greenhouse effect:

Impheéations for extreme rainfall events. Zbid. 8, 83-102.

Harr, T. L., Bourke, W., McAvaney. B,. J., FORGAN,

B, W. & McGercer, J. L. (1990) Atmospheric general

circulation simulations with the BRMC global spectral

model: ‘The impact of revised physical parameterisations,

J. Climate 3, 436-459.

HoucHton, J. T.. Inwgins, G. J. & Eeurauss, J. J. (1990)

“Climate Change: The IPCC Scientific Assessment”

(Cambridge University: Press, Cambridge),

, CALLawpER, B. A. & WARNEFY, S. K. (1992)

“Climate Change 1992: The Supplementary Report to the

IPCC Scientific Assessment” (Cambridge University Press,

Cambridge).

Keeeman, Ry (1991) A simple model of the atmospheric

response lo ENSO sea surface temperature anomalies, ,/.

Anos. Sei. 48, 3-18.

(1993) On the dependence of hindeast skill on ocean

thermodynamics in a coupled ccean-atmosphere model.

J Climate 6, 2012-2033,

McGreoor, J, L. & Wats, K. (1993) Nested simulations

of perpetual January climate over the Austraiian region.

J. Geophys. Res, 98, 23283-23290.

, WALSH, K, & Katzrey, J. J. (19934) Nested

modelling. for regivnal climate studies. Jk pp. 367-386

Jakeman, A. J.,. Beck, M. B. & McAleer, M. J. (Eds)

“Mouelling Change in Environmental Systems” John Wiley

& Sons Lid., Chichester).

____, Gorpon, H. B., WaTTeRSON, 1. G,, Dri, ML RR. &

Rotstayn. L. D. (1993b) The CSIRO 9-Level atmosphene

general circulation model. CSIRO Division uf Atmospheric

Research, Tech. Pap. Na. 26, 1-89.

McInnes, K, L., Lesuié, 1.. M. & McBride, IL. (1992)

Numerical simulation of cut-off lows on the Australian east

coast! Sensilivily to sea-surface temperatures. Ini J

Climaol, 12, 783-795.

NeeLin, J, D,, Latir, M., ALLAART, M. A, R, Cane. M,

A., Cupascu, UL, Gates, W. L., Gent, PR, Gan, M.,

Gornon, C., Lau, N-C.. Mecnoso, C, R., Mecan, G,

A.. Onrrtuber, J, M,, PHiranner, S, G, H., ScHore,

P. S., SPeRBER, K, R., StrRI, A-, Toxrona, T., Tripp,

J. & Zesiak, §. FE. (1992) Tropical air-sea interaction in

general circulation madels. Clim: Dyn. 7, 73104,

Power, S. B., Couman, R. A., McAvaner, B, J,, DaHit,

R_R., Moorr, A. M, & Smiry. N, R, (1993) The BMRC

coupled atmosphere/oveun/sea-ice Model. BMRC Res, Rep.

No. 37 (Bureau of Meteorology, Melbourne).

WHETTON, PH. & Pittock, A. B. (1991) Australian region

intercomparison ol the resulls of some general circulation

models used in enhanced greenhouse experiments. CS/RO

Division of Atmospheric Research, Tech, Pap, No. 24, 1-73.

& (1993) The growing consensus in simulated

regional climate for enhanced greenhouse conditions: Is

there a palaenclimatic analogue? Quaternary Australasia

H. 1, 81-87.

____, Raywer, P. J., Prvtocn,, A, B. & HayLock, M_R.

(1994) An assessment of possible climate change in the

Australian region based on en intercomparison of general

circulation modelling results, J Climate, 7. 441-463.

SOME IMPLICATIONS OF PAST CLIMATIC CHANGES

IN AUSTRALIA

By M. A. J. WILLIAMS*

Summary

Williams, M. A. J. (1994) Some implications of past climatic changes in Australia.

Trans. R. Soc. S. Aust. 118(1), 17-25, 31 May, 1994.

An important question, in the context of possible human impact on global climate, is

whether past climatic history can offer insights into possible future climatic change.

Equally critical is the likely response of the physical environment to any future

climatic changes. Evidence from pollen analysis, lake fluctuations, desert dunes and

coastal plains from Holocene deposits in Queensland, Victoria, New South Wales and

Northern Territory, demonstrates that the response of different elements of the

Australian landscape to geologically-recent changes in temperature and precipitation

was often time-transgressive. Any attempt to use palaeoenvironmental data to predict

possible future change must therefore take due account of the varying response times

of different constituents of the Australian landscape to external disturbance. A

synchronous response to climate change is more likely with relatively simple

biophysical systems such as small closed lake basins or source-bordering dunes than

with more complex systems such as tropical rainforests and tropical coastal plains.

Key Words. Climatic change, Australian landscape, past climates, Queensland

rainforest, Victorian lakes, New South Wales dunes, Northern Territory coastal

plains, response times.

Vrveciins af the Reval Soviery of S. aust (99d). Le 17-25

SOME IMPLICATIONS OF PAST CLIMATIC CHANGES IN AUSTRALIA

by M.A. J. Wittiams*

Sunumary

Wituiams. M, A. J, 994) Some implications of past ulinuitic changes in Australia, Trans, R, See, 5. dusr LBC),

17-25, 31 May; 1994

An important question, in the context of possthle human impact on global climate, is whether past climatic

husiory can offer insigbtyinto possible future clinic change. Equally evitical 1 the likely response of the physical

environment to any furure climatic changes, Evidence from pollen analysis, lake Noctuarons, deseer dunes and

coustal plains from Holocene deposits in Queensland, Victoria, New South Wales and Northern ‘Territory,

demonstrates that the response of different elements of the Australian landscape to pedlogicully-recent changes

in lempecatuce aad precipitation way often Ume-transgressive. Any attempt to use palaeoeayironmental data to

predict possible future change must therefore take due account of the varying response times of different constituents

of the Austruliinn landscape to external disturbance. A synchronous response to climatic change is, more likely

with relatively simple biophysicalsystems sach as small closed luke basis or source-bocdering dunes than with

more complex systems kuch vs tropical ramforests and tropical coastal plums,

KEY WORDS. Climatic chunve, Australiin landscape, past climates. Queensland rainforest. Victorian lakes,

New South Wales dunes, Northern Territory. coasta: plains, response times.

Introduction

There is in¢reasing concern thut the aceclerating

impact of human activities upon our natural resources

of land, air. water, plants and animals is seriously

daimaging many of our more Tragile egosystems,

culminating 10 irreversible losses-of genetic as well as

cultural diversity. Hunan actions may alsa be

contributing. to possible changes in world vlimute

through a combination of burning of fossil fuels und

changing land use (most notably deforestation). both

of which reduced the terrestrial store of carbon by dbint

60 + 0.5 and 1.6 / LO PeC in 1990, respectively

(Houghton e af. 1992).

Although the increase in the atmospheric

concentration of carbon dioxide. methane, nitrous

oxide und other “preenhouse” gases since the start of

the industrial revolution is well documented, the Hkely

impact Upon global climate and ecosystems remains

unclear (Pearman 1988; Houghton ef al. 1990;

Dunnette & O Brien 1992), Notwithstanding the

unavoidable scientific uncertainties over the magnitude

and frequency of future climatic Auctuations, there

appears to be w significam measure of agreement that

the increase in anthropogenic aerosols and trace gases

will enhance the greenhouse effect, culminating it

global warming of the lower atmosphere, particularly

in middle to high Jutitudes (Houghton er al, 1990;

1992). However, there is far less agreement on how

the global patiern of precipitation might respand to

enhanged greenhouse warming, prompting a aumber

of researcher's to look tw the past as a sourec of possible

= Mawson israduate Centre for Environmental Studics,

University of Adelaide, South Australia 5005.

analogues for future global warming (De Deckker er

al, )988; Petit-Maire et.al. 1991; Street-Perrott 1994),

The .aim of this paper is to consider some of the ways

in which wstudy of geologically recent changes inthe

Australian environment can offer insights into how the

various clements Of Our nutoral environment are |ikely

to respond to [ulure climatic change.

Past climatic changes

Will climatic history repear irself!

The search tor past climatic analogues of a warmer

future planet earth is based on a number of explicit

ond implicit assumptions. A major but unproven

assumption is the notion of cyclic change: “as in the

Past, so, lou, in the future”. Linked to this assumption

of recurrent climatic changes is the equally unproven

assumption of similar recurrent boundary conditions.

Boundary conditions in this context means the global

distribution of land. sea and ice; the global (iistribuuion

al major vegetation zones; and the global albedo

pattern. tosofar as global fuctuauons in surtiee

insoladion and in ice volume are linked 16 eyelet

changes in the ocbitul geometry of the carh at time-

scales of 010° years, vertain climauc changes are

indeed cyclic, at least when set against the last two

mullion years of the Quaternary period (Williams v7

al. 1993), But what of time-scales of shorter duration,

more obviously relevant to present human pre-

uccupalious? Here it is important to distinguish

between climatic change ard climatic vartubility.

Climatic change and climatic: variability

Much contusion over climatic variability and

climatic change stems froma failure to specify the time

a) M.A. J, WILLIAMS

scale involved. Fig. | illustrates climatic variability at

different time scales during the last 0.9 million ycars

(0.9 Ma) of the Quaternary. From 20 000 to 10 000

years ago (20-10 ka), the trend in global climate was

from cold to warm, but relative to the preceding

0.9 Ma, this warming was but one peak in a series of

global climatic fluctuations from cold to warm to cold

again. repeated at least eight times in the last 0.9 Ma.

Tt is important to remember that the Little Ice Age

(Fig. lb) affected both hemispheres more or less

synchronously. World temperatures have increased in

the last hundred years, with a brief return to cooler

4 b

Oh 5 ow (ow

1960 1980 (A) Lune fea Age

1700 @

a si Last Glacial

1880

1100)

ho 0. 06 300 (Last Interglacial

are AD, res

(2)

Max. Min

©

02

04

© os

~ 10°C 08

my,

~ 5x107 km

Fig. |. Climatic variability at different scales in time during

the last 0.9 million years (from Williams et a/. 1993, adapted

from Australian Academy of Science 1976).

temperatures between about 1940 and 1960 (Fig. lat.

None of these global temperature changes seem fo be

linked in any obvious way to ihe steady exponential

increase in the concentration of atmospheric carbon

dioxide — # point conveniently forgotten by over

enthusiastic advocates of enhanced greenhouse

warming.

The reconstruction of past climatic events

The reconstruction of past climatic events is based

on many independent sources of proxy data. These

“nalural archives” include terrestrial and marine

deposits and fossils, as well as evidence from

archaeology. isotope geochemistry and archaeology.

The temporal resolution, temporal range and type of

information which may be gleaned from the proxy data

sources are illustrated in Table 1, and are discussed

in detail by Bradley (1985, 1990) and by Williams ef

al, (1993).

Limitations of palaeaclimatie enquiry

Certain elementary precautions are essential when

using any particular type of evidence to reconstruct

past environinents and climates. Each may be useful

for a particular purpose and for a particular spatial and

temporal time-scale. However, as Williams eral. (1993)

note: “Difticulties anse immediately when we use

Proerustean tactics to force the data to yield

palaeoenvironmental information at particular scales

in space or fime for which those data are totally

inappropriate. A related issue is the precision available

in dating the proxy data or samples used in

reconstructing past events” (ap cit., p. 9).

Reconstruction of past climates thus requires good

chronology as well as careful analysis of appropriate

natural archives. Nor must it be forgotten that climate

is a second-order concept. First-order interpretation

TABLE 1. Characteristics of natural archives used in palacuclimatic reconstruction. (After Bradley 1990, Table I.)

* Minimum sampling interval in most cases. T = temperature; H = humidiry or precipitation; C = chemical compasition

af air (C,), water (C,) ar soil (C,)) B = bronuiss and vegetation patterns; ¥ = valeunic eruptions; M = geomagnetic field

variations, L = sea levels; S — salar activity.

Best Temporal

Archive Resolution*

Historical records day/hr

Tree rings season/yr

Lake sedinients tr to 20 yr

Ice cores ye

Pollen 100 ver

Loess TO) yr

Qceon cores 1000 yr

Corals yr

Paléosals JOO yr

Gcomorphic features 1) yr

Sedimentary rocks yr

‘lemporal

Range (yr) Information derived

1 Ta, BV, M,L,S

107 T.H.C,, B, V, M, §

104408 T, H, Cy, B, V, M

10° T.H,C,, B, VM, 5

10° T.H.B

10° H, B.M

0" T, Cy, BM

10" Cy, L

Ww T,.H, C,, ¥

10" T.H.Y.L

in! B.C. V, M,L

SOME IMPLICATIONS OF PAST CLIMATIC CHANGES IN AUSTRALIA 1

of w natural archive usually provides direct information

about some component of the natural environment.

§uch a§ a river, lake or dune. To infer climate (or some

particular attribute of climate, such as temperature or

precipitation) from a particular element of the

landscape is a far more circuitous and difficult

procedure, with far greater scope for interpretative

error.

Consider, for instance a lake. The outstanding work

by Gasse, Fontes, Street-Perrot’ and co-workers on

reconstructing the climatic history of lakes in Africa

has demonstrated the need for calibration ysing presen(-

day chemical, physical und binlogical data, as well as

the need tor taking due account of local hydrological

factory and of exireme events when deducing

palaeohydrology from palaeolimnology (Fig, 2) (Street-

Perrott & Roberts 1983; Fontes et al. 1985; Gasse er

ul, 1987; Street-Perrott 1991). Only then is it possible

to attempt a reconstruction of paldéoclimate. From

reconstruction of local palacoclimates to regional or

even global palaeoclimatic modelling 1s yet another step

removed from the onginal field data, Provided that the

limitations of pulacoclimatic enquiry are clearly

recognised, and the appropriate steps are followed in

interpreting, past climate Irom proxy data. useful

insights are possible when using past environmental

analogues,

Past eavironmental analogues

A decade has clapsed since Pittock & Salinger (1982,

1983) supgested that the early Holocene climate of

Australia towards 9600 — 700U years ago may be a

suitable analogue for the continent in a CO -warmed

carth, Granted that many early Holocene lake and

LAKE AND AIVER FLUCTUATIONS

AS PALAEOCLIMATIC INDICATORS

PALAEOLJMNOLOGY

(lakas and ¢lvars)

calibratian using

yo oo Medern fotereices

PALAEOHYDROLOGY

(walo’, salls and sofules and stable

isotope falaeobudgets)

calrenie evotits —

affects of local

pdralagloat

lachors

PALAEOCLIMATE

(estimation oF palasa-precipilatlon,

-avaporalinn, -temperature, etc)

~_Y

Fins, 2. Lake and river fluctuations as PalacoClimale indicators

(Source: F, Gasse 1986, ynpublished seminar paper to

Department of Geography, Monash Uneversity),

pollen sites in Australia and Papua New Guinea do

seem. t0 indicate a warmer and/or wetter climate at this

time (Williams 1984), just how reliable is the evidence?

Can it be quantified? Is the "Climatic Optimum” or

“Hypsithermal” in Australia indeed a valid analogue

of future warming? Finally, can we glean anything else

of value to future climate prediction from the record

of early Holocene environmental change in Australia?

In an attempt to answer some of these questions. it

is appropriate to considet the evidence from a variety

of different Holocene sites across Australia. In the

ensuing discussion we consider four distinct types of

Holocene site in tour widely separated localities: the

rainforests of northeas! Queensland, “maar” lakes of

western Victoria, the source-bordering dunes of central

western New South Wales, and the coastal plains of

the Alligator Rivers area in Northern Territory.

The Holocene rainforests of northeast Queensland

Kershaw's (1983) palynological studies of the late

Pleistocene and Holocene vegetation history at four

pollen bearing lake or swamp sites in the Atherton

Tableland of northeast Queensland have tevealed

that the period of maximum rainforest expansion

lasted from 6000-8000 to 3600 years B.P. (Fig. 3).

Lake

Euramoo

Quincan| Bromfield Lynch's

Crater Swamp Tater

1

Maxlrrum

2 Rainforast

Expanslorr

3-

a4

m

Ss

a

a)

a &

Oo

=]

- 7

Time of arrival

B of rainforest

9 Period of

precipitation

increases

1490

Present mean annual tainiall (mm)

Fig. 3. Veretahon changes in northeast Queensland deduced

from pollen analysis of Holocene cater lake sediments,

iron De Deckker vr al, 1988, adapted from Kershaw

1983.)

7600. 1800 2000 2200 2400 2601)

20 M.A. WILLIAMS

However. as Figure 3 clearly shows. the response of

the rainforest to increasing warnith and imereasing

effective precipitation was strongly time-transgressive.

Rainforest appeats m the Bromfield Swamp pollen

record towards 8500 years BP, at Lake Quincan

Crater not until afer 7000 years B.P. The demise of

the ramfOrest at all dour sites was roughly synchronous

Go000 years B.P.), indicating that the rainforest

responded slowly to climatic amelioration (warmer and

wetter conditions) but rapidly to climatic deterioration

(cooler and drier conditions).

Application of the CSIRO binclimate prediction

system developed by H, A, Nix to Kershaw’s

Queensiand rainforest pollen dite enabled Kershaw &

Nix (1989) to derive more quantitative estimates af the

Holocene climates. Their analysis revealed that

Holocene temperature maxima were not achieved in

the Atherton Tableland until S000 years B.P., with

mean annual temperatures up to 3.5° higher than today

persisting until abyut 3500 years B.P) Men annual

Precipitation at that time appears to have been at lenst

300 mm higher than today, and most probably

500-800 mm higher (Kershaw & Nix 1989, De

Deckker et al. 1988).

One obvious conclusion to be drawn from the

Queensland pollen record is that the interval 7000-90)

years B.P, was not the ume of “Chmatic Optimum” for

northeast Queensland. The wettest and warmest period

vame considerably later, and on present evidence seems

to have lasted from 5000 to 3500 years B.P. If this is

true for the tropical northeast of Australia with its

monsoonal suramer rainfall regime is it true also. for

the temperate southeast of the continent, where the

rainfall comes mainly from the westerly airmasses

which pass across the southern margins of the continent

most persistently in winter, when the Antarctic

convergence ts Im ts most northerly postion?

The Holocene crater flakes of western Victoria

Lake Keilambete in Victona is perhaps the best

Studied Holocene lake in Australia (Dodson 1974;

Bowler 1981, De Deckker 1982, Chivas ef al. 1985,

1986), Like other volognic explosion-erater or “maar”

lakes in Victoria, 1 occupies a small closed basin anil

is highly sensitive to changes in precipitation and

evaporation over is catchment. Five “maar” lakes on

western Victoria, including Lake Keilambete. have

yielded usefu) information about Holocene changes.in

the balance between precipitation and evaporation (De

P/E Estimates for 5 Victorian Maar lakes for the Holocene

Kellambete Gnotuk

x1000 years before present

wm

19

Bullenmerri

re

ee

changing

[tow | aan [ttn [ cw ecn | ii

West Basin East Basin

siege 2) se

ee a

50 km

Lowest

Fig. 4. Changes in Holocene precipitation/evaporation (P/E) ratios lor five voleune “maar” lakes in western Victoria. (Froipy

De Deckker ef al, WR)

SOME IMPLICATIONS OF PAST CLIMALIC CHANGES IN AUSTRALIA pa]

Deckker 1982; De Deckker et al. 1988). Highest

precipilation/ evaporauon (P/E) ratios, peak lake levels

and lake salinity minima were all within the interval

7000 to SOO) years B.P. From 5000 to 3000 BLP (when

morthcast Queensland was warmer and wetter than

today) Jake levels were lower and salinity valucs wer

hivzher than in the preeeding 2000 years or sa (Fig, 4)-

One possible urference is that winter Tainfall may

have been lower towanis 501)-3000 years BoP. Eauislly

Plausibly, tolal annual raintall (including both summer

and winter precipitation) may have been reduced al that

timc. We cannot, as yel, choose between these vay

possibilities. What does appear certain is that the time

af greatest effective precipitation (7QQ0-SQKX) years

B.P.) was several thousand years earlier than the time

of maximum effective precipitation in northeast

Queensland, It is tempting to speculate thitt we are

secing the etlects of pwo distinctive climatic systems:

one (in northeast Queensland) controlled by the tropical

summer monsoon, the other (in western Victoria)

controlled hy the winter westerlies. Ifso, the response

of both systems io pusiglacial warming was not

synchronous along the eastern. third of Australia. Nor

may jthe synchronous in the face of any furure global

warning. We Cur tow to the semi-arid inland areas

of New South Wales

The Holocene source-hordering dunes of

central western New South Wales

Holocene palaeoclimaue data are-excecdingly scarce

for the semi-arid and arid areas of Australia which

together comprise 75% of our present land areg, ‘The

late Pleistocene fluctuations in the Willandra Lakes

region of western New South Wales so carefully dared

and clucidated by Bowler (1970', 1983) do aut yield

a Holdcene signal and in any event have far more to

do with runoff from the Eastern Highlands vis the

Pleistocene Lachlan-Willandra river system than with

local changes in rainfall and evaporation (Williams, ef

ul, 1986),

One climatically-sensitive area capable ol providing

useful information about /vcal hydrological events is

the desert margin system of central western New South

Wales studicd by Wasson (19757, 1976) und by

Williams er a/, (199}), The study area occupies about

80 000 km! in the semi-arid region bounded by the

Darling River to the north and west, and the Willandra

'BawLer. 7 M_ (i270) Late Quaternary environments: 4

study of Jakes and associ#ied sediments in southeastern

Austratia, Ph.D, thesis, Australian Natuonal University,

Unpubl,

“Wasson. R. J. 0975) Evolution of alluvial fans in two areas

of southeastern Australia Ph.D. thes, Maequane

University, Unpubl.

Creek distributary of the Lachlan River to the south

(Fig. 5}. There are po perennial rivers iq the entire

area, The ephemeral sircum channel Known docally as

Crewl Greek of Sandy Creek (Fig. 3) [lows

intermittently during very wel years and the West to

east trending linear dunes are toduy vegetated and

stable.

a

o— Lake Nulohars s

+L

~

atari Range

elvanhce «>

[ns

Fig, 5. Map of central western New South Wales showing

ocation of Sandy Creek. The shaded urcays are isolated

Tanpes over 200 10 in elevation (Saurce: Williams et al:

1991).

The Holvcene climatic history of this vast area ts

poorly understood, but (he pattern of local

environmental changes is now reasonably well

documented (Figs 6, 7). Source-bordering dunes were

actively forming from channel sands Jerried in by

Sandy Creek between about S$00 to roughly 600) years

B.P. Both before and after that mterval the dunes were

inactive, vegetated and stable (Williams et al, 1991),

Souree-bordering dunes are dunes which develop

immediately downwind vf a parent source of sand, such

as the sandy bed of w river, asandy lake beach or a

sandy alluvial fan. There are three prerequisiles tor

the formation of source-bordering dunes:

(1) A regular replenishment of the sand supply (for

instance, from a seasunally-fowing sand-bed

channel),

(2) Strong unidirectional winds for at least part of the

year, und

(3) A sparse or limited vegetation cover adjacent to

the sand source.

Any interpretation of the Holocene climatic histury

of this region Must take into account the requitemnents

for source-borering dune formation, the apparent

absence of carly to middle Holocene dune deposits in

22 M. A. J, WILLIAMS

Netrts csifFS csG Males CSiFS CSG Metres

t+ o7-

1,840 + 50 B.P,

Loamy.

fine-medium

red sands

Soft red

sands

Palaeosol

Low angle planar ay.

cross-bedding

dipping downstream

Crude

horizontal

bedding

Cobbles

Sandy clay

Red clayey sand

+ green mottles

§5B: Bronzewing Valley

Fig. 6, Representative alluvial fan stratigraphic sections exposed in the gullied western piedmont of Belarabon Range (see

Fig. 5). The block diagram shows Sandy Creek, the source-bordering dunes and the wind direction. (Source; Williams

etal, 1991).

Years Before Present

1 o ~~

400-3,600 Aboriginal fires. ()

Alluvial and aeclian ganas

3)

ates BBPP 7,400-7,500 a. Hight lake levels (2)

Pedogenic carbonate (4) ~

13,500-15,500 _ Calcareous dust

Alluvial and aeolian sands F

14,000—1 8.000 <O Gypseous jurette 12)

Fig. 7. Generalised late Quaternary piedmont stratigraphy in

the central western New South Wales study area. The

numbers in brackets denote numbers of samples dated, and

includes both radiocarbon and thermoluminescence dates

(Source: Williams eral, 91, revised to show additional

dates, }

Padagenic tarbonale (6) ~S

24,000-32,000 aie Caleareaus dust

Alluvial anc gealiar sands

a)

SOME IMPLICATIONS OF PAST CLIMATIC CHANGES IN AUSTRALIA 23

this area (Fig. 7). and the fact that the present-day * 600 years BLP. to present:

dunes are vegetated and stable, A tentative Hot summers, cold winters

palaeoclimatic interpretation. based mainly upon the Raintall more uniform

geomorphic evidence and needing to be tested by future Wind vyelocitics low and/or

independent work, is 4 follows; Denser vegetation

¢ 10,000-5500 years BP: If this interpretation is correct, any change to a wetter

Rainfall more uniform climate or to. a more seasonal distribution of rainfall,

Wind velocities low with more runolf during the winter months, could lead

Surface well vegetated to a renewed phase of source-bordering dune formation

Erosion and deposition very slow provided the summers remained dry and windy and

® 5500-600 vears BP: the riparian vegetation relatively sparse. Should the

Higher winter rainfall future climate in this area become both warmer and

Seasonal channel flow wetter, the somewhat paradoxical outcome could be

Summers yery hot, dry and windy a replenished sand supply and reactivation of the linear

Summer deflation suurce-bordering dunes.

Arnhem Land Plateau

(Middle Proterozoic

Sandstones & Dolerites)

Sandstone Outlier / 4

e Je fi aa

Wooded Lowlands

Paperbark Swamps

Shelly Sand Ridge

(Holocene Cheniers)

Uranium bearing

Lower Proterozoic

Metasediments

Strike Ridge

Tidal River Sy 1 2 Ag ae

Late Catnozojc

Coastal Plains 5 Sands & Gravels

Late Quaternary

Marine , Estuarine

& Freshwater Clays

Fig. 8. Block diagram showing the Arnhem Land plateau of Northern Territory and the geologically youthful coastal plains

wo the north. (Source: Williams. 1991.)

24 M.A

The Hivecene coastal plains of the

Alligator Rivers area, Northern ‘Territory

The tropicul eoastal plains of Northern Territory

sti in alinos: diameirie contrast w the ephemeral

streaings and desert dimes of semi-arid western New

South Wales. Their significance stems {ron the Tact

that the Holocene coustal plains which extead tor some

thousinds of kilometres from the far narthwese of

Western Australia to the fac northease of Queenstane

are very dynarme landforms, and highly sensitive ti

even minor chasges in sea devel, sediment tnad, salinity

and wave climate (Williams (991). They ure also hese

toa umque and “abundant fauna and Nora (Haynes e¢

al, W991), In addition, the coastal plains west of Darwin

dbut Ono the Arnhem Land plateau (Bias. 8). with its

uhundant galleries of Aboriginal rock art..and its record

ul over SQ 000 years. of prehistoric Aburipiril

senilement (Roberts er al, 1990),

With the melfiing of the ale Pleistocene ice Sheets

of North America and Europe, world sea levels rose

from their last glaeial maximunt level oF about -[35 m

(at 18 QUU years BOP to their present levels towarils

BOWU-7000 years BOP. (Williams-er al’. 193d

The initial rise was rapid, and vast areas of the

contuvental shell off northern Australia were

submerged at @ rate of over 20 metres.a year or roughly

40 emia week. Quce this rise had slowed down, which

it did in the last few thousand yours, coastal nungroves

hes to colonise the intertidal muds. allowing muddy

sediments to accumulate on the old late Pleistowene

land surface inland of the coastal mangrove fringe. The

process accelerated once the sea altained its present

level. and widespread mangrove swamps developed

across the present area of the coastal plains in mid-

1 WILLTAMS

Holocene times, from about 5000 to 2000 years B.P.

(Woodcotle et al, 1985). ‘These mangrove swamps

proved to be highly efficient sediment tnsps, and Were

eventually buried by estuarine and, ultimately, by

freshwater couds (‘Woodrotle etal, 986; Williams 199);

Wasson 1992), The present-day coastal plains are thus

4 relatively youthtul feature, ind are generally less thin

2000 years old,

How these plains might respond to sca leved change

will depend upon 4 vainet of factors. including the

rate of sca level rise, the impact of any climatic change

upon uno and sediment yield in the litlal rivers; Lhe

magnitude and frequeney of future cyclones. und the

relative duration of the wet und dry seasons. Since none

of these variables is. accuracly predictable, speculation

seems unwarrinted beyond noting that the coustal

plains are likely to remain @ geomorphically dynamic

and ucuively developing feature of (he landscape, just

as in the past SOG) years.

Conclusion

Provided certain commun-sense preciutwins are

observed, an appreciation of Holocene climatic und

olher environmental changes in Austrilia can be i

useful guide to pusmble future landscape responses to

global warming whatever its ultimate causes, The

Austnilian landscape is a palimpsest of landforms, soils

and plant associations, all of which will respond to

lulure climate change in a variety of ways, The way

im which our rainforests, lakes. dunes and coastal plains

have responded to Holocene climatic changes offers

us some guide to their possible future responses, It is

highly unlikely that these will be either simple or

synchronous,

References

AustkauWAN AcAbEMY Ob Scrence (1976) “Repo of a

Committee on Climate vlunge” (Australian Academy pl

Science, Canberra).

Bowler, J. M-. (1981) Austtalion wall lakes: a

pidacohydoblogical approach Hydrabiologia 82, 431-44,

(1983) Luneites-as mdives ol hydrological change:

A review of Australian evidence. Pree. Ro Sec Men 95,

147 hk

Brantey, RS. (1943) “Quaternary Paleoo nnatelogy.” (Allen

& Unwin. London).

(Hed) (990) “Global Changes. of the Past” (EiCA RY

Office far Interdisciplinary Earth Studies, Boulder.

Colorado),

Chamece. IM. A. & GrinpRow, A. (1983) “Proceedings

of the First CLOIMAN YZ, Conference. Howmans fap. 1941.

Voluines Joand 2° (Dept of Biowcogruphy & Geotnor-

plislogy, Austrahian National University, Cunberral

Caras A RL. De Deckeer P & Sueriry. 1 MOG Yas)

Simmnuint cantent of ostraceds indicates lacustrine

pilacosalinuy, Mature Sip, 251-253.

; J & (1986) Magnesii conient of

pon marine ostracod shells: a new poleasalinameter and

palueothermometer, Falueayeage Palueorlimatal.

Palaeoecul, 54, 43.0).

De DerkkER. FP. (1983) Holocene ostracods. other

Invertebmaites and fish remains from cores ot four maar lakes

in southeastern Australia, Pro, Re Soo, Pied, 94, 193-270.

> Kprsiaw, A, Po & Wiiiams, M, A. J. (1988) Past

enyironmentil analogues. pp, 473-488. Ja G. I. Pearman

(bd.) “Greenhouse: Planning tor climare chanwe” (Bod

Brill. Leiden & CSTRO, Metbourne),

Bonsor. J. (1974) Vegetation und climutic history near Lake

Keilambete, western Victoria. Ames. A Bor 22, 704-717.

Duspeire, DO A, & O'BRim, RK. od (Rds) (1992) “lhe

Seience-at Global Change; The linpact of Huraan Actrvities

on the environment” (American Chemical Society.

Washingtin).

Fontrs, 1-C, Gasse, Fo. CAtuor, Y. Praviat, I.-C.

Cyrsoner. P. Depevece, PAL & KacaManska. L-

(1985) Lreshwater and marine-like environmenis trom

Ifalocene lakes in Northern Sahara. Marnie 317, 608-6100

SOME IMPLICATIONS OF PAST CLIMATIC CHANGES IN AUSTRALIA

Gassr. F. Fortes, J. C.. Puaziat, 3-C., Carsonel, P,,

KACZMARSKA, 1, De DeckkKER, P.. SouLi-MAKSCHE, L,,

Cartyr, Y. & DupruPle, P. A. (1987) Biological remains.

veochemistey and Stable isotopes for the reconstruction of

environmental and hydrological changes in the Holocene

lakes from North Sahara. Palagogeoyr: Palaecelimetol.,

Palaeoeco!, 60, \-46.

Hayses, C, D., RioparH. M. G. & Wariiams, M.A, J,

(Eds) (1991) "Monsoonal Australia” (Balkema, Rotterdam)

HoucHton, J. T., Carranper, Bo A. & Varnny. 5. K,

(Eds) (1992) “Climate Change 1992, The Supplementary

Report to the IPCC Scientific Assessment.” (Cambridge

University Press. Cambridge).

, Jenkins, G. J, & Eprraums. J. J. (Eds) (1990)

“Climate Change, The IPCC Suientific Assessment’.

(Cambridge University Press. Cambridge),

Kersuaw, A, P (1983) The Vegetation reeotd from

northeastern Australia; 7 + 2 KA. pp. 100-101, Vol. | and

Fig. G7, Vol.2, Jn Chappell 1. M, A & Grindrod, J, (Eds)

“Proceedings of the First CLIMANZ Conference.

Hewmuns Gap. 1981. Volumes 1 and 27” (Dept. of

Biogeography & Geomorphology, Australian National

University, Canberra),

— & Nix, H. A. (1989) The use of bieclimutic

envelopes for estimation of quantitative palaeoclimutic

yvalués, Proc. CLIMANZ 3 Symp., 78-85. (CSIRO. Division

of Water Resources, Canberra).

Perit MARIE, N,, ForTeGNr, M, & Rouianp, €. (1991)

Atmospheric methane ratin and environmental changes in

the Sahara and Sahel during the last 130 Kyrs. Ailaeoveour.

Palaeoelimatel. Palaeaecal. 86. 197-204.

Pirtock, AB. & SALINGER, M. J. (1982) Toward regional

scenarios for a COj-warmed vanth. Climate Change 4,

23-40,

. & (1983) The climatic optimum and a

CO ;-warmed earth: the Australasian Region. pp. 122-125.

Jn Chappeil J. M, A & Grindyod, J. (Eds) “Proceedings

of the First CLIMANZ Conference, Howmans Gap, 1981.

Volumes ] and 2.” (Dept. of Biogeography &

Geomorphology, Australian National University, Canberra),

Rowerrs. R, G., Jones, R. & Soir. M. A, (1990)

Thermoluminescence dating of a 50 GO00-year-ald human

occupation site m northern Australia. Nature 345, 153-156,

SrReet-Prrrotr, FA. (1991) General circulation (GCM)

Md

S

modelling éf palacoclimates: a critique. The Holocene 4,

74-40,

(1994) Palgeo-perspectives! Changes in terrestrial

ecosystems. Ambio 23(1). 37-43,

& Roperrs, WN, (1983) Fluctuations in closed-basin

lakes as an indicator of past atmospheric circulation models.

pp. 331-345. Jy Street-Perrun, F A., Beran. M, & Ratclifle,

R. (Rds) “Variations in the Global Water Budget” (D.

Reidel, Dorilrecht).

Wasson. R, J, (1976) Holocene veolian landforms in the

Belarabon area, SW. of Cobar. N.S.W, Journal Proc. R.

Sac. N.S ME 109, 91-101.

_. __— (1992) "Modern Sedimentation and Late Quaternary

Evolution of the Magela Creek Plain” (Research Repoit

6. Supervising Scientists lor the Alligator Rivers Region,

Australian Government Publishing Service, Canberra),

Wittiams., M. A, J. (1984) Palaeoclimates and

palacocnvironments: (a) Quaternary environments, pp:

42-47 Th Veevers, J. J. (Ed.) “Phanerozoic Farth History

of Australia.” (Clarendon Press, Oxford),

______ (1991) Evelution of the landseape. pp. 207-221 In

Haynes, C, D,. Ridpath. MG, & Williams, M.A. J. (Eds)

“Monsoonal Austnilia® (Balkema. Rotterdam).

, Apsmson, D. A. & BAXTER, J. T. (986) Late

Quaternary environments in the Nile and Darling basins

Ausiralian Geographical Stadies 24. (28-144,

. Dr Deckker. P., ADAMSON, D. A. & TaLror,

M. R. (1991F) Episodic fluviatile, lacustrine and avolian

sedimentation in a late Quaternary desert margin system,

comral western New South Wales, pp, 258-287 Jn Williams,

M.A. J, De Deckker. P. & Kershaw, A, P, (Rds) “The

Caimoame in Australia: A Re-Appruisal of the Evidence”

(Geological Society of Australia Special Publication No.

18. Sydney).

, DuUmKERLEY., D, L., Dr DEeckKER, P., KERSHAW.

A. B & Stokes. T, (1993) “Quaternary: Environments”,

(Edward Arnold, London).

Wooprorre, C. D., Tom, B. G, & CHaperit. J. (1985)

Development of widespread mangrove swamps in ntid-

Holocene times. in Northern Australia. Nature 397, 71-713.

_____.. CHAppFII. J. M. A., THoM, B. G. & WALLENSKY,

E. (1986) “Geomarphological dynamics and evolution of

the South Alligator tidal river and platns, Northern

Territory’. (Australian National University, North Australia

Research Unit. Mangrove Monograph No. 3, Darwin)

CLIMATE CHANGE AND ITS IMPLICATIONS FOR

SOUTH AUSTRALIAN SOILS

By R. W. FITZPATRICK* & M. J. WRIGHT*

Summary

Fitzpatrick, R. W. & Wright, M. J. (1994) Climate change and its implications for

South Australian soils. Trans. R. Soc. 8. Aust. 118(1), 27-34, 31 May, 1994.

The nature and distribution of dominant soils in South Australia is briefly reviewed

with particular reference to major issues relating to climate change. For this purpose

the State has been divided into two regions: (i) the southern regions or agriculturally

developed area which lies south of latitude 32°S and is mainly used for dryland

cereal/sheep production and (11) the northern region or semi-arid and arid areas which

mostly lie north of latitude 32°S and are mainly used for low intensity grazing of

natural rangeland. A large proportion of South Australia, including many texture

contrast soils of the high rainfall areas, has dispersive soils subject to sodicity and

which are highly prone to waterlogging and salinity throughout a significant

proportion of the profile.

Key Words: climate change, soils, South Australia, soil moisture, soil temperature,

soil organic content, salinity, sodicity, soil acidification.

Transactions of the Royal Sociery of 3. Aust. (994), TES), 27-34

CLIMATE CHANGE AND ITS IMPLICATIONS FOR SOUTH AUSTRALIAN SOILS

by R. W. Fitzpatrick*t & M. J, Wriout*

Summary

birveatrick, RK. W.& Wear, MJ. (1994) Climate change und its implications for South Austrahan sulle eins

R. Sac. S. Aves, WSO), 27-34, 31 May, 1994

The nature and distribution of dontinant soils in South Ausitalia is briefly reviewed with particular reference

to major issues relacing 1 climate change, For this purpose the Stute hay been divided ints two regions; (1) the

southern eeion or agriculturally developed arca which lies south of latitude 32°S and is mainly used for dryland

ceéreal/sheep production und (i) the northern region or semi-arid and arid areas Which mostly le north of latitude

32°S and are mainly used lor low intensity grazing of naturel sangeland. A large proportion of South Australia.

including, many texture contmst soils of the high rainfall areas, has dispersive soils subject lo sudicily aad whieh

are highly prone to walerlogging and salinity throughout a significant proportion of the profile.

Chis paper attempts to forecast the most likely conscquences of plobal climate change on the doniinant soils

of South Australian, The direet iflugnee of imereasing winter temperatures and decreasing winter rainfall on lhe

wide range of soil types (hat o¢cur in South Austrahy is as yet unclear, However, ihe magnitude and extent of

supposed degradation or enhancement of panicular soil morphological properties, and decline or increase of

auil site properties (e.g. salinity or Crosion) are matters of speculation, In fact, these changes can affect the processes

od global climate change by affecling production of greenhouse gases or causing changes in vegetation, How

swils modify under the influence of changing climate will depend principally on seul type. roposraphy and changes

of vegetation.

South Australia’s soils will respond ot global climate change through changes In soil motsbare, soil temperature

and soil organic maiter content, Decline in winter rainfall in the high raintall regions occurring to (he south of

latitude 32°S will have a beneficial effect by substantially reducing seasonal waterlogging and formation of aquic

soils (including non-dal acid sullate sits), In contrast, dryland salinity may continue to expand. with a corresponding

decline in stream water quality.

Tlis nol yet poxsible to accurately quantify regional sort elagges resulting from climate change given the present

Uncertainty about the amounts und rates of global climate change, and particularly concerning regionul patveras

of lensperature, precipitation und coastal geomorphic changes. The cuprentlysinereasing tate of land and water

exploitation in Sonth Austealia will likely. hayes greater inmpact on scils, adverse or benvticial, than the effects

of limale change.

Key Worws: climate change. setts, South Australia. sed! mersiure, sol temperate, soil organic coment,

salinity. sodicity, seil acidification.

Introduction

The greenhouse effect is the warming of the carth

resulting from increases in the cancentralion of carbon

dioxide (CO,) and other radiatively active gases

including methane (CH;), nitrous oxide (NO), ozone

(O,) and the chlorofluorocarbons (CFC's) (hat reduce

the Joss of outyoing infrared radiation (i.e. limit heat

fosses frony earth into space; Table [). A global

increase in CO, has been reliably documented during

the last 100) years and although this is duc mainly to

the burning of fossil Fuels there isan appreciable Mux

of CO, trom the oxidation of soil organic matter und

the burning of forests (Table 1). CO, levels of twice

the present are forecast by the year 2050 (Pearman

1988; Houghton 1990: van Breemen & Feijtel 1990).

Soil is, not only a source of CO, but also of CH, and

N,O (Table 1), Atmospheric warming of about 0.5°C

* CSIRO Division of Soils, Private Bag No. 2. Glen Osmond,

South Australia. 5064,

+ Cusuperative Research Centre fe Sol and Land

Management

has been estimated over the last 100 years, estirnates

of warming up to 5°C over the next 100 years have

been made (Bouwmun 1990), A corresponding

warming of the oveans-and rise of sea level of up to

0.5 mare also forecast, Changes of this magnitude and

rale represent 4 Significant change in our environment

that would have a profound effect on the Earth's

ecosystems and human activities (Bolin et al. 1986:

Wild 1993).

Much of the emphasis in recent discussions of the

impact of greenhouse induced climate change in

Australia has focussed on the atmosphere and on

agricultural and forestry production (Pearman 1988),

However, the effeets of climate change om Australian

soils is also a fundamental issue. Given the present

state of severe soil degradation in Australia (e.g.

Chartres ef a/, 1992) the pessibility of further

degradation should be of national concern,

South Australia is a region which 1s currently

influenced mainly by winter rainfall but where the

tairrent rainfall varies greatly from north to south

Hence, for discussion purposes, we have divided South

rs

oe

house gases, and there contrthidion to global warming (adapted from Wild

for major green

ie international society af seil science, 1990),

eranimles of th

f

‘centrations, increase, residence time, sources and sinks fe

1993; Bounwaun 1990); Standing committee on international pra

TABLE 1, Atraspheric cone

R. W, FITZPATRICK & M. J. WRIGHT

fa Australia into the following two regions: (i) the

gs southern region — the agriculturally developed urea

z a5 which lies south of latitude 32°S and is mainly used

[=e as e fay

ial 4 ai for dryland -cereal/sheep production and (ii) the

a, _ * ’ - - ‘

Pa as 3 as, northern region — the semi-arid and arid areas which

is

wif 7S I Zé ' H atl Jon a}

ul+.2= & Se mostly lie north of latitude 32°S and are mainly used

SeSAL1 1 a for Jow intensity grazing of natural rangeland. The

> greenhouse effect is likely to influence the soils in each

az region differently, The southern region is considered

3B hkely to suffer a drying climate duc to a reduction in

uv ¥ . 2 < : .

= sit winter rainfall whereas the northern region is more

= aE likely to reveive enhanced summer rainfall, The

ait = BRA ™ . “fee » * a

SIF. § & ZE objectives of this paper are to: (i) summarise existing

Scifaiol | a RS information on soils in South Australia, (i) highlight

the most likely consequences of global climate change.

q . ‘ whe . .

nn on the range of South Australian soils and (ili) identify

z= future priorities for research on global climate change

= cp on South Australian soils.

or af

eS 3 :

Bun a8 Soils

Zz 4S ao

A Ce so F F

ol Ex =. aoe = Ber Soils are complex systems which are strongly

a 25s e¢ ; Sane oy 3

z 2 ae eso Fres z 22 < influenced by processes occurring in the atmosphere,

SSRAG+S3R 7922 2d biosphere, hydrosphere. and lithosphere (Fig. 1).

In general, soils develop slowly over thousands of

S years, the absolute rates depending to a large extent

is on past climates. Their component parts also differ in

= Boe *. 8 their rates of development. For example, soil biotic

; S foi gD tl a f :

a ey = Sra eS gece processes resulting from the interaction of the

< zoe er Em 358 atmosphere and biosphere are much more rapid than

SS = u ee A 7) ' . . 4 ‘

| —- © SUstsety ae 2 = soil weathering processes resulting from the interaction

= 384745323 7 TT i

il 5 = of = a g 228° C2 58 of the hydrosphere and lithosphere. Thus, in response

Die Pongo ee = Eg SEZES=ES to greenhouse imduced climate change. the more

D . = Cc os r a6 r “Tr

eo oma FR ase eS oe Sagat environmentally sensitive soil biotic pracesses would

y lolic p

+4 be expected to respond more rapidly (Bouwman 1990;

al Varallyay 1990; Wild 1993). In coastal landscapes,

os me many sails would also suffer relauvely rapid water

i FS : , Bee :

= = erosion, waterlogging and salinisation as sea levels rise

u Ds vers with global warming.

é G 4&8 ww BS

i Eos me an ‘a oS Zz 2 a A h

G nese 25688 imosphere

Clone sH228 ER FS

Sones Boao

y

a

i

£2

BE Biosphere ———» («ais +— Hydrosphere

a

“3

58

me? Osh

SERFS =

nc eee Bs wz ot

gerbe Er

ue eae a 4 i= } n oO

E BES = 8 z c g Lithosphere

ose D5 a S Et ae . " . ° . . .

dEfuss 3 5 ea = “a| Fig. 1. Simplified diggrammatic representation illustrating (hat

Sake 2 z “4 Lt = € soils are complex systents which are influenced by processes

Bee 2 ze ae 2 2 fram the uttiosphere (climate), biosphere (vegetation and

2258 6fa $ G & 2 fauna), hydrosphere (hydrology) and lithosphere (geology

and topography).

CLIMATE CHANGE AND ITS [MFLICATIONS FOR SOUTH AUSTRALIAN SOTLS 24

Influence of lund clearing

An important factor im assessing greenhouse etfect

is the change already brought about to Australian soils

as @ result of Buropean settlement, Land clearing,

praying un) cullivalion generally bave resolled in a

rapid decline of soil organic niatter and organisms and

consequently a deterioration of soil structure and

increasing suscepuibility ky erosion. A marked

reduction in the diversity of fauna and flora of most

ecosystems bas also accompanied European settlement

and this his made our soils and lundseapes less resilient

ty change

The effects of gecenhouse induced climate change,

therefore, must be assessed for soils and landscapes

already degraded: pristine conditions are virtually non-

existent, Given this scenario, there is a signiticant

environmental question t be answered: If there is

greenhouse induced climate change of the magnitude

forecast, what will be its effects on South Australian

sos and landscapes?

Sotl Carbon and the COs cycle

The amount of carbon stored in the world’s soils as

fresh organic matter, stable humus und charcoal, & two

to three times highee than carbon stored in natural

vegetation and crops (Table |, Bouwman 1990:

Standing committee on international programy of the

International Society af Soil Sctence 1990; Wild 1993).

Distribution and major properties of soils in

South Australia

Ciencially speaking, Sour Australian soil landscapes

are extremely variable and complex due. i part, 10

the great age of much of the continent. A signiticant

proportion (>60%) of soils io South Australia are

saline, or sodie and/or texture contrast (ic. duplex —

have bleached sandy to sandy clay loamy topsoil

hurmens overlying clayey subsoil horizons that may,

or may mol be mutiled) (Table 2), Sodic duplex soils

are falierently subject t waterlogeing because of water

perching on the more impermeable clayey sobsoil

horizon and consequently saturating the topsoil

hocizons, especially where there is.an Increased inpul

of water following removal of native vegetation, A

similar effect cam be caused. in the short term, by

excessive irrigation. Therefore, to proceed from u mure

general understanding of the processes of waterlogging

and dryland salinisation to 4 more definite

understanding. of these processes in an individual

catchment, in order to predict the effects of

management or climale change. 1s extremely difficult.

The centri ynd southern Mt Lofty Ranges area is

further compheated by hwhly variahle geology and

weathering patterns of the soils in those Jandscapes.

Tt 1s apparent that in adjacent landscape positions one

is confronted with deeply weathered soils which

contain ancient stored salt, juxtaposed with very

youthful soils on partly weathered rocks which are

generating sallas a result of contemporary weathering

processes.

The higher rain(all areas (i.e. greater (han 500 mm

average annual rainfall) which he south of littitude 32%

ure restricted to coastal and sub coastal plains and

ranges. A high proportion of both regions: has soils

which are sodic throughout or through a significant

proportion of the profile. Table 3 lists the estimated

areas ol’sodic soils in South Australia, south of latitude

32°S. The names used represent general profile

characteristics that may each include several great soil

groups. ‘This has been done to reduce the table to

manageable proportions suitable for reproduction here.

The semi- and arid regions north of this latitude contain

even more significant areas of saline and sodic soils.

particularly the desert lounts and associated red clays.

The table was prepared by te-interpreting Maps

published by Northcote (1960) und Northcote ef al,

(1948), Sheets | und 10, respectively, of the Adus of

Australian Suis; ihe original scale of which was

12 100 000,

TABLE 2. Uiviribulion of saline ane Sedic sails in relation we rainfall in South Austrilia (after Northern & Skene 19721.

Percentage dred of each map don within anowal rainlall zones

Annual Alkaline strongly sodic Now-alkuline sodic

Rainfall Area Saline Uniform Gradationsl Duplex Neutcal Avid Talal

(mm tke Sails texture texture prafile duplex duplex (%

<)50* 450,215 25 9 6¥ 18.7 - _ 31.35

190-254)** 351.510 | 1.8 470 — — 39.9

250-350"=* 82.570 23 11 54.8 V5 - TOT

3SU-SSD6F* 72,745 - is 224 a4 8 42 _ 62.4

> 5a0t** 25,560 12 | 24.2 a4 28 43.2

Total 983,000

* All within northern region of South Australia.

“+ Largely within northern région hut a significant area of southern region inelodial,

rH AL WITHIN sotiTheen regia cecepl Jor small acca at nortberm region in 250-350 zane:

uw KW. FITZPATRICK & M | WRIGHT

On an area basis two broad groups of sodic sails

are prominent in the southern region, the grey-brown

calcurcous louis (36% of the area) and the duplex suils

with alkaline soil reaction trend (18%), thal is, soils

with pH venerully increasing with depth and alkaline

in the subsoil, These two soil groups inclide the

majority of the cereal and sheep producing, land in the

State, Although of much smaller area, the sadic duplex

soils of southern Eyre Peninsula, Yorke Peninsula, the

Mid-North, eastern Mt Lofty Ranges, Kangaroo Island

and the South-Bust ure critically important beeauge they

have high rainfall and a high production potential. They

are alsey important because Unecy indicate the regions

where waterlogging and secondary salinisatien is 4

current or potential problem,

TABLE 3. Estimated vreas af soits in South Australie seats

af latitinde 32°S (madified ufier Nai et al, 1293),

Area

kunt a

Soils that wre sogie throughyut:

al Cliayéey soils. uniform texture

profiies 3450 if

b) Grey-brown culcareous loans,

unitarm and gradational texture

profiles. 75660 30,1

¢) Alkithine duplex soils 3335 lq

Sub-tutul Rd § 4u4

Soils that fave sodie Bubsolls

i) Alkaline duplex. soils 38685 IBS

ec! Neutral duplex soils W185 24

tf) Acidic duplex soils 2930 14

Sub-total 47800 23.4

Tolal sodi¢ soils 2245 64.2

Now sagdic soils;

) Alkuline duplex soils 947 14

nh) Neutral duplex suils WIG ne

t) Acidic duplex swrls 5682 2.7

i) Clayey soils. uniform textire

profiles 1683 0.8

Kk) Girey-brown caleareous foams,

uniform and gradational bexture

profiles 44212 2uu

lh Sandy soily incudinge podzots WSR 1)

Total not sodic soil 7IATS 36.8

Land area South of latitude 32°8 209720 Tw

Prajections for climate change

Forécasis of greenhouse induced climate change ure

based on global models that incorporate the circulation

of the Earth's almosphere and oceans and the alhedo

effect of its polar ice caps. Present models vary in their

forecasts of global CQ, and temperature changes

although a commonly accepted view is that a doubling

oF CO, will result in a temperature rise of 4°C

(Pearman 1988; Wild 1993), These models also predict.

io varying degrees, a Gonsequent merease In

evaporation and previpitation Le. an intensification of

the hydrological cycle with increased rauntall erosiv ity.

Greater uncertainties exist in the prediction of regtonal

climate change und ils.consequences within continents.

In Australia, a “best guess” scenario tas been

developed by Pittock & Nix (986) based on historic

records and a global model from which a rise at

helween 2 and 4,7°C in meun annual temperature by

the year 2030 is predicted (see also Pearman (988),

[i sumimury, lemperslure increases are predicicd to be

ereatest in southern Australia, particularly in winter;

summer raintall areas will receive um increase in

precipitation of up to 50%, and the monsoonal

influence will extend farther south; the winter rainfall

areay Of southern Australia will receive less

precipitation, particularly in winter,

The climatic models vary even more widely in their

prediction of the moisture status of Australian soils

under doubled atmospheric CO; Only uw broad

generalisahon of increased soul moisture in the north

and a decrease in the south can be made.

The uncertainties in predictions of climate chunge

ints the next century are consideruble, Nevertheless,

it 18 clear thal even a modest greenhouse effect will

significantly change the regional climatic regiines of

Australia, In the southern parts of the continent,

particularly, the climate change forecast will create

Urier soil moisture conditions and impose further

limitations fora range of land uses. When this scenario

is superimposed on the present pattern of soil

Gegradalon in South Australia, it is apparent that large

areas of land could be at risk of further degradation,

Not all the predicted! consequences of the greenhouse

effect, however, are negative. Forecast increased

precipitation in the northern part of South Australia

would have (he potential fo increase soil moisture stores

with 2 positive outcome for soil stability and for the

pustoral and agricultural mdustries. In contrast,

Jecreased precipitation in the higher rainfall areas of

southern Sourh Austraha would reduce seasonal

waterlugging (sce beluw),

Impact of past climate change

Because soils and landscapes develop over long

periods, miny are found to contain information about

past climate change, Among the most significant

examples are the soils and associated sediments of

coastal flood plains wbout Australia (Division of Suils,

CSIRO 1983), At the peak of the workd?’s last glaciation

20,000 years ago, sea levels were about 140. m below

the present and the Australian coastline was

consequently much more cxtensive. AL the cnd of

glaciation, plobal seas rose rapidly ta reach their

present level about 6000 years ago. During this rise,

ocean waters invaded large ureas of coustal land aod

embayed all codstul rivers, The rivers depnsitel

CLIMATE CHANGE AND ITS IMPLICATIONS FOR SOUTH AUSTRALIAN SOILS i]

sediment ia mace and estuarine-enviroamems so thir.

today, their coastal fload plains contain a continuous

record of sedimentation. Around Australia. high water

tables. high salinity, aod sulphates characterise the soils

of these flood. plains.

Vastu 4, Possible negative modifications vf lane use due to

Manes in temperature, rainfall antl windiness in South

Aastralia,

Coastline

Rising sea levely.

= wave vrgsion

* flooding of coastal fond plains

*salinindtion of coastal Hood plains

Northern Region (Arid and semi aril zones which

mostly lie warthoof ketimde 32°98)

Increased rainfall and erasiviiye slighty lugher

LOC LANETEN

* mcreased risk of lindstides

® incrensed pink of water erosian

* fouding of lowlands

Southern Region (High raintall zones which lic south of

latitude 32°S)

Changed rainfall invulence. iicivased vrasnvisy

higher lemperalures

* lower sol moisture

* decreased vegeitive cover

* loss of sbil biotic companenty and otganic matic

® dererioration of soil strucnire

* increased risk ol water crosiay

* increased salinisation

© ditenaranon of strean: water quality

* incredsed wind erosion

® (listicr atmosphere

Tn such coastal landscapes, the ale geological pas;

provides an uccurate guide for predicting the

cumsequences of sea level cise nesulting from the

greenhouse effect, These are increased flooding and

in elevation of water tables, and increased salimisation

und encroachment of marine and estuarine

enV Ironments, accompanied by a build up of sulphidic

sediments in expanded mangrove swampy (‘Tables 4,

S\, This will lead to the development of greater areas

ef avid sulphate soils. The area of coastal Mood plan

actually affected, particularly io the gulf regions oF

Sourh Australia, would depend on the magnitude oF

the sew level rise: a special study would be required

tor relivble estimates.

Rises in sca level also cause widespread wave erosion

of coastal svils. near shorelines, as well as

destabilisation of coastal dune Jandscapes (Table 4)

Much of the evidence for these effects in the late

veolagical past have been. covered by the ocean or

obliterated. Nevertheless, a significant risk ty couxdines

around South Australia is apparent (Harvey & Relperio

Unis volume),

‘Tame 5. Possible positive modificarions in kinduse dup to

changes in tenperature, rainfall and wiadiness in South

Australie.

Cnastline

Ristnse seq levels

® flooding und formation of coastal mangrove swaps

Northern Region (Arid and semi-arid zones whieh

mostly lie north of lautude 32°S)

Increased rainfall: slightly higher temperatures

* increased soil moisture

@ increased vepelalive caver

@ reduced. Wafer erosion

* jncreased soil orwinte matter

Southern Region (High rainfall zones whieh he soutlr ot

fatitude 32°)

Decreased rainfall: higher temperarnres

* lower water lables — ground und perched treduced

incidence of seasonal walerlogping)

@ decreased soll acidification

The usefulness of the geological record for predicting

the effect of greenhouse mduced cliniate change on

other Australian soils and landscapes is much less

certain, This is because much of the stratigraphig and

seomorphie evidence for climaic change during the laxt

20,000 years relates to world-wide (Pleistocene) glacial

climates, In Australia, a large part of Tasmania and

a small arca of the mainland were glaciated. inland

lake levels were high, river discharges were higher than

the present, lurge areas of hill slope land were

crusivnally unstable and eroded sediment filled local

alley floors. In central Australia, the dry phases of

the Jate Pleistocene were characterised by extensive

movement of sand dunes and the mobilisation of large

quantities of dust which wus deposited as a clayey

mantle over Jandscapes in eastern Australia These

unstuble soil and landscape conditions changed to

relative stability as climatic warmed during the fast

{0,000 years (Holocene),

If we use the past ax an analogy, the predicted

greenhouse warning should be seen as leading to soil

formation and. therefore, not expected to result in

widespread instability and degradation of soils and

landscapes um non-coastal Australia, However, two

factors muke the present different fron the late

geological past, Furst, the rate of climate change due

to greenhouse warming is predicted to be much quicker

than climate change of the same magnitude in the

geological past. Second, most of our sotls and

landscapes haye already been modified follawing

European settlement an such a Way as lo make them

more vulnerable to rapid environmental change. In

many parts of South Australia, thresholds of stability

have been exceeded, resulting in accelerated

walctlogging, saliosauon and erosiog (Fiepauick

al. 1992, 1993). If present land use ts not responsive

Ls RW. FITAZPAPRICK & M_ J. WRIGHT

ti ereenhnuse induced climate change, these and other

fanny ol soil degradation are likely to increase,

particularly ip the southern region where decreascd

soil mowiure wuld make present land use marginal

ir sume Areas

In this connovnon, Pitock & Nix (1986) have

modelled changes in plant productivity a5 4 result of

projected greenhouse induced climate change in

Australia. In northern Australia, the positive outeeme

of the greenhouse effect in inereased precipitation and

vegetative cover could mean i shill inwards an anercuse

in the stability of Sauls ly water erosion. As elsewhere,

however. much would depend on Land use practices,

A decrease in plant production is anticipated in the

southem region of South Australia, signalling a

decreuse in soil vegetative cover and soil biotic activity

ft is Suggested that these effects would be relatively

rapid with 4 cotresponding decrease in sei) structural

stability and au) inereased susceptibility 6 erosion Li

southem Australia, lherefure, the year 2050 dould see

w return towards conditions that prevailed m the Tale

Pleistocene given the Opper Qvirst) greenhimse

scenarin,

Projections for Soils and Landscapes

From the foregoing, ts possible 10 oudline some

uf ie faain Consequences of greenhouse induced

climate change in South Australan soils and

landscapes. Although these can only be froadly slated.

they have a potential value in idenefying like areas oF

Positive und negative outcomes. These are summiarisect

in Tables 4 and 5S. Lt may be argued that some of these

projections are alarmist, However, waterloysiig.

sulinisation, acidification, wind and water erosian are

already major problems in South Australia anu there

is no evidence that they are under control (Fitzpatrick

etal. 1092. 1993. 1944: Naidu ep al, 1993). Even the

conservative “niddle scenarios” of greenhouse induced

climate change would exacerbate these problems in

many areas.

Sail erasian

South Australia has a high proportion of sodic and

saline soils (Northecte & Skene 1972: Naidu er al.

(993), The xudie:saline rao of approximately 3-1

mages hetween 4 anil WO times that reported tor other

continents (Szaboles (989) and is conststent with the

high proportion of sodium present in soil solutions and

groundwaters. [nm South Australia, most sodium

affected sails are the result of past inundations by

brackish water supplemented possibly by cyclic sult

Thus, im subsoils, chloride 4s the dominant anion and

exchunveable Mg/Ca ratios are high. The imcidence

of sodicity often coincides with the spatial distribution

iW duplex soils. This associuun nibcates a patern of

environtnemal hazard which appears to at beast broadly

coincide with many areas under agriculture in the

southern region, and substantially limits its

productivity, Saline, sodic and sodic duplex soils are

often predispused te land degradation (e.g. Naidu ez

al, 1993. Isbell e¢ 7/1983) and, signifieunily, cover

> AN ul the South Australian land mass. The effects

of adsorbed sodium on clay dispersien are most

pronounced in denus alkaline subsails which comprise

over 86% of Australian sodic soils (Northcote & Skene

1972),

The impart of sail sodicuy on the eaviranment is

# inost important land degradation issue in Sauth

Australia. Both primary and secondary sodification of

soily cuuses undesirable changes in soil structure,

severe hillslope crosion, waterlogging and erosion of

downstream watercourses. The assucialed increase in

colloid and nutrient loading of streams alsa conte ibutes

ty sedimentution and the consequent loss al reserverr

storuge capacity, Iagether with serous water quality

problems. Heavever. itis the contact of sodium-attected

soils with water in the form of rain. splash, surface

runoff, thoaugh Now, natural ground water flaw, or that

pumped for irmgation frony other catehments, which

provides He expanded scale on Which environmental

problems and harards become must conspicuous

(Fitépatrick ef al. 1994).

Tn South Australia generally, climate change is likely

to increase the tncidence of water erosion (Table 4),

Although lower preetpilauion an the south wall generally

reduce the rate of water erosion, this 1s counterhulunced

by the less intensive sail conservation influence of poor

vegetative growth due tran inadequate moisture supply

for plants (cxucerbited by an increase in temperature,

as well), Lower precipitation and higher temperatures

will alsu lead to increased wind erosion, especially

where sandy soils are dominant (ive i the semi anid

northern portions -of the southern region). In the

northern région, increased summer rainfall is Itkely

to be of higher intensity than at present and i, thus,

expected to present o higher erosion risk here tou,

particularly to the widespread saline and sodic desert

luans and red clays of the stony tablelands,

Waierlogeing und dryland saliniry

Extensive deforestation which followed Europes

settlement im South Australia has conrnbuted to a

detrimental shift in the hydroligic balance of muny

¢utchments, This has resulted in 4 rise of perched and

saline groundwater tables causing waterlogging anil

dryland salinity to be a major land degradation problem

in South Austnilia, Over 220.000 ha oF land are

estunated. to be affected by dryland salinity throughout

the major agricultural districts at an annual cost to this

state of approximately $25 nullion in lost agricultural

production. The problem is worsening. In particulir.

there is # prowimg Concern by pruperty hulders in the

MI JLofiy Ranges over fhe rapal merease in saline,

CLIMATE CHANGE AND ITS IMPLICATIONS FOR SOUTIL AUSTRALIAN SOILS 4s

structureless acid sulphate sails that are waterlogged

and highly prime te water erosion (Firzpauick er al,

1992, 1993)

Bespne an obvious reduction in seasonal

Waterlopgzing due ta déeline in winter raintall in high

rainfall regions south of latitude 32°S, dryland salinity

may continue to expand with x corresponding decline

instream water quality (Tables.4 and 5). According

to Sadler et al, (1988), the processes involved in such

“saline” sodie soul cavironments, due to redueed winter

precipitation, ave further complicated by the balance

hetweern rates of leaching and recharge. Gverall, the

consequent reduction of avater quality and land

capability are, therefore, only products of 4 counples

set of interactions which can occur between. the

following factors in Mediterrancan climates: sadium-

affected soils, ambient levels of soluble salts and the

teniporal Now patterns, spatial distribution. relative

comributions and quality of surface water, and through

Naw and ground water In particular, the leche of

saline sodic soils with good quality water te.

Tatnwalet) also poses a threat iy structural stability in

huildinus, bridves, earth dams and embankments

(Fitzpatrick er ef 1994)

Seil acidification

In the Southern region, a. reduction in winter

Precipijation may decrease downward infiltration and

leaching of water and hence lower the rale at soil

acidification. Similarly, a decrease in winter

preipitation and increase in winter temperatures will

discourage heath vegetation which produces acidifying,

litter (Table 4), In contrast. in those presently high

raintall regivns where climate change will cause an

iNerease Mn Sumer precipitauion, soil acidification may

accelerale due to stronger leaching snd chemical

reactivity (eg carbonate dissolution may occur)

Bidlegieal charges including the rutrient regime

Graetz ef al. (1988) in discussing the implications

of climate change: tor arid zone vegelaton stresses the

importance af the spatial patterns of soil types.

Furthermore, given the importance of this spatial

dimension, there 164 requirement, as yet not met, Lor

models of vegelation response to plant-available

myistire und avaiable nutrients thal.ure specitic to the

major soil types. Three reqhiremients Were identified

tor forecasting vegetation change. The first was that.

becuuse the spatial patterns of suil type are still

expected to determine the distribution of vegetation,

it is essential that the forceasts of future climatic

conditions have a spatial resolution equivalent to that

of the mupped soil landscapes.

We consider that much of the carbon stored in South

Aumralian sous is in the torn of charcoal due to a Jong

history of burning, With climate change in the

directions discussed, it is reasonable to expect that ii

the northern regions, biomass, and hence charcoul will

increase, whereas in the suutbern region reduced

canfall will lead toa generally reduced biomass and

consequent reduction in the amount of charcoal

produced, even though line trequency. could wierease

because of higher temperatures and lowered moisture

Carneclusiins

Changing agricultural practices ({hat lead to erosion,

salinisuhen, sodihealion, acidification and

waterlogging) make jt difficult to monitor the effects

of climate change in individual cutehments, Special

rescarch that takes account of these complications 15

needed on specific South Austritlian soil types lo

quantify regional effects of climate change and ty plan

yinilegies fo cope with te changes because of

implications for the well being of our communities and

the environment. Th would appear thal the diornnmently

pastoral activities of the northern region stand to benefit

by reason of incrensed Summer rainksll ala time when

temperatures are suitable for major growth of herbage:

Salinity of surtiace sails, at least, should alse be reluced

by the increased precipitation (Brinkman & Brammer

1990). In the southem region, on the other hand,

agricultural pursuits could be adversely affected by

reduced rainfall, although increased (emperusures will

improve winter growth of cereal ctops. Agricultural

activities may also be able to expand somewhat in the

northern marginal fringe to take advantage of increased

rainfall there, Finally, a] a time when the demand lor

good quality water js esculating, the limited

untdlersignding of the factors and interrelatinuships

involved in soil and water management may well be

further compheated hy » changing, global climate

induced by the “greenhouse effect”

Acknowledgement

The authors Wish to acknowledge Dr P, H- Walker.

formally. of CSIRO Division of Soils, who contributed

significantly tu dhis paper by providing expert advice

in the form ofa craft review chat formed an original

framework for it

References

BainkMan, R. & BraniMer, H. (1990) The Influence of a

changing climale on Soil properties, Trans, (4h tn, Caren

Soil Sei, Kyate, 1990, 283-289.

Boin, B.. Doos, BR. & De JAGR. J. (Eds,) (1986) “The

Greenhouse effect, climate change, and ecosystems” (Wiley.

Chichester),

Bouwaan, A, PF) (Rd. (890) “Soils and the Greenhouve

Fftect’ (Wiley, Chichester).

34 R. W. FITZPATRICK & M. J. WRIGHT

Chartres, C. J.. Hetyar. K. W., Firzeaprick, R. W. &

Wiutiams, J, (1992) Land Degradation as a result. of

European Settlement of Australia and its Influence on Soil

Properties. pp. 3-33, dn Gifford, RM. & Barson, M.M.

(Eds) “Australialy Renewable Resources Sustainability and

Global Change’, International Geosphere-Biosphere

Programme Australia Planning Workshop, Oct 2-4 1990,

Burcau of Rural Resources Proceedings Nu. 14, (AGPS,

Canberra}.

Divisios OF Soils, CSIRO (1983) “Soils; an Australian

viewpoint’ (CS£RO: Melbourne’ Academic Press,

London).

Firzpatrick, R, W., NAipu, R- & Seip, PG. (1992) Tron

deposits and micre organisms occurring in saline sulfidic

soils with altered soil water regime in the Mt. Lofty Ranges,

South Australia, pp. 263-286 In Skimner, H. C, W. &

Fitzpatrick, R, W: (Eds) “Biomineralization Processes of

Iron and Manganese — Modern and Ancient

Environments”, Catena Supplement No, 21.

» Hupnacr, W. H., SELF, P. G, & Narpu, R, (1993)

Origin and properties of inland and tidal saline acid sulfate

soils m South Australia. pp. 71-80. /n Dent, D. L. & van

Mensvoort, M, E. F. (Eds) “Selected papers of the Ho Chi

Minh City Symposium on Acid Sulfate Soils.” International

Inst. Land Reclamation and Development Publication 53.

___ , Boucuer, S., Naipu, R. & Fritscy, E- (1994)

Environmental consequences of soil sodicity. Aust. J. Sail

Res. 32 (In Press},

Graerz. K. G., WALKER, B. H., & Wacker. P. A, (1988)

The consequences of climatic change for seventy percent

of Ausualia. Jn Pearman, G. 1. (Ed.) “Greenhouse: planning

for climate change” (CSIRO, Australia).

HouGuron, J. T., JENKins, G.J. & EpHraums, J. J. (Eds)

(1990) “Inter governmental panel on Climate Change”

(Cambridge University Press, Cambridge),

IsheLL, R. F, Reeve, R., & Hurron, J. T. (1983) Salt and

sodicity, pp. 1O7-IL. Jn “Soils: an. Australian viewpoint",

(Division of Soils, CSIRO. Melbourne/Academic Press,

London).

Nau. R,, MERRY, R. H., CHurcHMAN, G. JL, Wricit, M.

TL, Murray, R. §., Fitvpatrick, R, W, & ZaArcinas, B.

A. (1993) Sodicity in South Australia; — a review. Aust,

J, Soil Res., 31, 911-929,

Nortucote, K, H. & Skene, J. K. (1972) Australian soils

with saline and sodie properties. CSIRO Soil Pub. No. 27-

Prarman, G. IL. (Fd.) (1988) “Greenhouse: planning for

climate change”. (CSIRQ Ausiralia).

Prrrock, A. B. & Nix, H. A. (1986) The effect of changing

climate on Australian biomass production — a.preliminary

study. Climatic Change 8, 243-255.

SADLER, B. §., MauGer, G, W. & Stokes, R, A, (1988) The

waler resource implications of a drying climate in south-

west Wesiern Australia, pp. 296-311. /n Pearman, G, |.

(Ed.) “Greenhouse — planning for climate change” (CSTRO,

Australia),

SzaBocs, [, (1989) “Salt affected soils” (CRC Press, Boca

Raton, Florida).

STANDING COMMITTEE ON INTERNATIONAL PROGRAMMES OF

THt INTERNATIONAL Society OF Sot ScieENCcE (ISSS)

(1990) “At global change . . . Do soils matter?” (c/o ISRIC,

P.O, Box 353, 6700 AJ Wageningen, Netherlands),

Van Breemen, N, & Feurnn, T. C. J. (1990) Soil processes

and properties involved in the production of greenhouse

gases, with special relevance to Soil Taxonomic systems,

In Bouwman, A_ F. (Ed_) “Soils and the Greenhouse Effect”

(Wiley, Chichester).

VARALLYaY, G. Y. (1990) Consequences of climate changes

induced in soil degradation processes. Trans. 14th Int,

Congr. Soil Sci.. Kyoto, 1990. 265-270.

WiLp, A. (1993) Soils and the environment: an introduction.

Chapter 1, (Cambridge University Press, Cambridge).

CLIMATE CHANGE AND ITS HYDROLOGICAL

IMPLICATIONS

FOR SOUTH AUSTRALIA

By B. C. BATES*, S. P. CHARLES*, N. R. SUMNER*, & P.M. FLEMING?

Summary

Bates, B.C., Charles, S. P., Sumner, N. R., & Fleming, P. M. (1994) Climate change

and its hydrologic implications for South Australia. Trans. R. Soc. S. Aust. 118(1),

35-43, 31 May, 1994.

The possible effects of doubled atmospheric concentrations of carbon dioxide on the

North Para River at Penrice catchment are evaluated using results from a general

circulation model, a stochastic weather generator and a conceptual water balance

model. Although results show only a marginal decrease in median monthly runoff

during winter months, large increases in monthly runoff maxima are indicated for

August and September due to large increases in extreme monthly rainfall. Modest

increases in evapotranspiration were also indicated for these months.

Key Words: Climate change, hydrology, South Australia, water yield.

Treva tins vf the Koval Secen of & Ame, US) LET Ly

35-43

au

CLIMATE CHANGE AND f'l'S HYDROLOGICAL IMPLICATIONS

FOR SOUTH AUSTRALIA

by B.C, BatTes* S. P Coaries* N, R, SumNrr*® & PM. FLEMING=

Sumeary

Hates, BL C., Cuaknes, 5. PB, Sumner, N. R., & Fleming. PM (1994) Climate change and its hydrologic implications

for Soot Australia Trans, R. Soe. S. Aust, 80), 35-43, 3) May, 1994,

The possible effects of doubled atmospheric concentrations of carbon diceide othe North Pura River al Penrice calchinent

ure evalualed asm results from) a general circulation model. a stochastic weather generator ynd a conceptual water halance

model, Although results show only a mitrginal decrease in tiedian monthly funoff during winter mondhs, large increases

in Monthly unolt machina are mdicated tor August and Seprember due to large increases in extreme mounttily ruintall, Modest

Increases in evapotranspiration were also indteated for these Maoaths

Kev Woros: Cliniite chunge, hydrology, South Australia. water viel

Introduction

Global climate change caused by rising atmospheric

concentrations ol carbon dioxide (CQ) and other

trace gpuses may have a significant unpuct on regional

Willer resuurecs, Reecot research suggests that plausible

climatic changes will affect the timing and magnitude

of cunoff and soi moisture, change lake levels. and

increase evapotranspiration (Cohen 1986; Lettenmaier

& Gan 1990; Allene al. 1991, Lettenmaier & Sheer

1991. Mimikou er afl 1991; Nash & Gleick 199);

Panagoulia 1992), Such scenarios have important

implications (or future water resources planning and

Management, the cnvironment and the national

economy.

Mest hydrologic scenarios are based on the climatic

predictions of numerical models of the generil

circulation of the atmosphere. General Circulation

Models (GCMs) can produce long-term simulations

of the energy and water Muxes in the atmosphere and

land and ocean surfaces on a global computational grid

uf cells and a number of vertical layers, Predictions

af changes in climatic variables such as precipitation

und teinperature ave considered to be more reliable than

those for runoff and soil moisture (Gleick 1989). The

predictions are provided as sputial averages over areas

of the order of (0% to 10° km? due te ihe limrtations

of present-day: computers. Current GCMs perform

reasonably well in simulating ihe present climate with

respect to annual or seasonal averages at this spatial

scule. However, the direct use of GCM outputs to drive

hydrelugic models is considered to be impnyper due

to the coarse (relative to river basin scale} resolutiog

* Division of Water Resources, CSIRO, Private Bag P.O,

Wembley. Wester Australia 601+

+ Division of Water Resourees. CSIRO, GPO Bow [ber

Canberra, Australian Capital Territory 2601

of the spatial grids used by GCMs and the simplified

GCM represemations of land surface processes, energy

transfer wilhin oceans, wid subgrid-scale atmospheric

processes such as convective storms, Moreover, GCM-

based assessments uf climate change are based on

steady-state simulations of current climate and the

climate associvted with a doubling of current

atinospheri¢ concentrations of CQ,, whereas in

reality the concentration of CO, is. increasing

continuously. Despite these limitations. GCMs offer

the most detailed quantitative information on potential

large-scale climatic changes due to increasing

utrnospherie concentrations of trace gases

Consequently, assessments of the impact of climate

change on hydrologic systems frequently use doubled

CO, scenarios consisting of a spectrum of unitorm

shifts to historical temperature series and scalings of

histurical precipitation scenes based on GCM trends.

The size of the adjustment may vary from month tw

month to reflect seasonality in the assumed changes,

Estimates of potential cvapotranspiration for the

changed climate scenarios are usually oblained by using

simple sculing factors, sometimes varying seasonally,

and by laking qualitative account of precipitation and

temperature trends, The historical series and changed

climate scenurios are usedl as input to mathematical

models of hydrologic processes anu the model outputs

vare evaluated to discern pussible changes in soil

moisture and water yield for a given catchinent.

It is frequently argued that working with hypothetical

scenarios suits the purpose of a sensitivity analysis of

water resources and that results are not intended to be

4 predicuion of changes. However, this approach suffers

from three major limitations. First. ic Ignores any

changes in the distribution and frequency of

precipitation events and any changes in the nature and

variability of temperature series, This may be

regardless of whether such changes wre indieated by

3

GCM predittions Second, the use of hypathetical

soominos based on an arbitrary: number of temperature

wid precipitation perturbations and stationary wind and

relative hunndity series, say, 18 not realistic and may

lack the imerdal consistency of GCM simulations of

fulure climates, Third, concurrent historical climatic

series are relatively: short in length which may

compromise the evaluation of the response of a

hydrologic system to elimaté variability as well vs

climate change,

At alternative to the conventional approach is to vse

4 stochastic model representing daily weather

vuriauons aca iocation, The parameters of such a model

characterise the behaviour of the present day climate,

Changed chute sequenees can be produced by

adjusting the model parameters ina maine consistent

with GCM trends. The model is then used to generate

long-lerm sequences of synthetic dally weather records

which in urn are used to drive a hydrologic model

(Wilks. 1992; Bates et uf, 1993; Charles eral 1993)

Such an approach allows consideration of changes in

the distribution and [requency of precipitation events

und chines in the variability of other efimaue

variahles. It also preserves the mtcrnal coosmtency of

GCM simulations of future climates

In this paper, an attempt made to assess The impact

of a doubling of current atmospheric concentrations

of CO, on the North Puri River ut Penriee catchmenc

in Saoth Austrglia, Historical climatic and hydratogic

series und results from a sinule GCM, a Stochusuc

wauther geucrator, and a conceptual water balance

model have been used. Conclusions are drawn on

possible changes im Water yield ay a cesult of climate

change.

Methads

Creneral circulation nuutel

The GCM used in this study, CS7ROG, has been

develaped by the CSIRO Division ol Atmospheric

Research, The model operates with nine vertical levels

in the atmosphere and a horizontal resolution uf about

UH km ~ 600 km. A computauonal ine step ol St

minutes is used, The simulated climate dala come from

equilibrium (constant CO, concentration) tuus tor

present day (control) and future (dovbled CO.)

climates. A complete descrippon ol the mudel ws given

by McGregor et al (1993).

Charles ev af, (1993) reported distinct differences

between the monthly raintall patterns found in

Australian meteorological station records and those in

the CS/ROY control nuo. The differences were with

respect Ky muinfall amount and its peographicil

distribuuon, Maximuns daily grid cell rainfalls for

CSIRO control runs were also found lo be between

‘wand '4 ofthe maxinam daily rainfalls recorded on

nesoscale walersheds, Nevertheless, the comparison

36 & C BATES. S. P. CILARLES. N. R. SUMNER & PM. FLEMING

of GCM rainfalls. by Charles ef u/.. revealed a general

(rend towards increasing rainfall amounts and changes

in the rainfall occurrence process under doubled CO,

conditions. The latter is consistent with the work ol

Gordon ef al, (1992) who found marked changes jy

(he magnitude and frequency of extreme raintall events

when comparing results from equilibria experiments

with the CS/RO4 GCM,

SMechastic weather generalyr

The stochastic weather generator used in this study

is based on the WGEN generator deseribed by

Richardson & Wright (1984), The daily climatic

variables simulated by WGEN are precipitation

occurrence and amount. maximum and minimum

lemperature, and globitl solar radiation (R,)-

Precipitation occurrence is described by a two-state

(wel or dry day), first-order Markoy chain, the

transition probubililies for a given location are allowed

to vary (hteugh an anaual cycle by defining separate

probabilities for each of the [2 calendar months:

py = Prid, =i | day = 7

where py, = probability that a day Oy state ¢ will be

followed by a day in state 7 (7,7 = 0 denotes u dry day;

i, J =t detwtes a wet day): und /, =indicator

variable denoting the presence or absenee 4

precipitation on day a, The variation of precipitation

ndumly on wel days is characterised using cither a

gamma or mixed exponential distribution,

‘The temperature and solar radiation components are

represented as.

x*(7) = ANF) + Beit)

where x = (3 % 1) veetor oF climatic variables:

© =(3 * 1) random forcing vector consisting of three

independent standard normal variates; A and

B — (3 ™% 3) matrices obtained from the lag-( and

lag-l correlation matrices for the components OF s

(Matthias 1967), and the asterisk Genotes

standardisation:

SAE = ALY — fg) Foyt)

in which kK— 1. 2. 3 and y—O. be yt) = mean

for climatic variable A and state jo and

a0 = corresponding standard deviation. The annual

cycles Of jy, and ay, dire modelled by simgle Fourier

harmonies with fixed phase angles

HYDROLOGICAL IMPLICATIONS OF CLIMATE CRANGE FORK STH. AUSTRALTA x7

Jnvestigation of historical Australian daily max anu

and minimum temperature series conditioned on wet

and dry days has shown thit the use of 4 single

hurmanic to describe annual cycles of means and

standard deviations is often unjustified, Higher

harmontes. ure present if the mean series and the

sland deviation series ure more tealistnatlly

represented by fixed monthly values tsee alsvv

MeCuskill 1992), Thus a mean temperature series (7)

tay be writer as

TA) = Hy) | E Ry coy Jarre (b+ a) / 365] -F F,

where K=t. 2) =O. te + =Jolian date.

4 = constant: ay=3; RK, = amplitude of {he rth

harmonic: w = 3,14 in (his case; = phase anple:

und 5, = tth residual,

Similarly. investigation of Australian solar radiation

series has indicated that the fitting of single harmonics,

Ihe use Of a fixed phase angle and the use of the

generation scheme defined above ts an inadequate

approwch (sve Bates eral. 1993), Our approach tavolves

the calculation of the upper envelope for clear day

conditions, R,. (A documented FORTRAN-T7

compuler program tor calculating the upper envelope

can be ubtained from the authors). This thearetcal

maximum is based on geographical location and

uverage clear day atmospheric conditions Por each

calendar month, a generulised beta distribution is fined

to the daily resuluals (the difference between the

theoretical maximum and recorded data) for wet .and

dry days,

Climate sequences for doubled CO, conditions

Wilks (1992) presented. a method for the adaption

of stochastic daily weather models fitted to current

elinnitic series to the generation of synthetic series for

future climates. “The adjustments to the model

parameters were made in a manher consistent with the

changes in monthly: statistics derived from comparisons

al GCM runs for contrat and doubled CO,

conditions. This approach was based-on the oouon thin

GCM results ure often available in terms of monthly

rather thin daily values or even monthly means and

varianees, The CSTRO9 GCM runs provide daily values

lor 30 climatic variables for 30 year periods, The

variables include; precipitation; humidity at level t;

maximum and minimum screen temperature:

temperature at level |; and.nct solar radiarioo at ground

level (&,,). Consequently, our method for making

parameter adjustments to stochastic duily weather

models uses this information,

Let P = monthly precipitation for a month

comprised of Nuduys, The mean af P and is varinee

ane defined by (Wilks 1992)

wy) — Noreep

wtP) = Naa? ita = abe !(l-d)

where a = uncenditional probability of a wet day

le =poll laa pk a 8 = shape and scale

parameters of the gamma distribution: and

d= pPy-Py WS # Measure OF persistence. For

simulating doubled CO, conditrons, we construct {he

rats (Wilks 1992):

BPD) Cote

wP) rap

(PY) — ae'B [1 bec! (Lay +d)! (ed)

o (PF) pais? [1 + a(l-ryl + d) / ed

where the prime deootes doubled CO, conditions.

The solution of the above equations requires twa

additional constraints, For catchments in the arid and

semi-arid regions of Australia where x < 05, we set

wv! =m, and d' =day/ dy) in which the

subseripts | and 2 denote GCM values fora nearby

GCM wrid cell tor conwol and doubled CO,

conditions, respectively,

For mean doubled CO, daily maximum and

mninimum temperatures, we assume:

Tylt) = TAO + Ty — They

where the prime denotes doubled CO, conditions;

7 = harmonics fited to the observed mean series; and

T, P= harmoniés fitted 1 screen temperatures

from the control and doubled CO, GCM runs,

respectively. Standard deviations are assumed to be

unchanged.

Clear day solar radiation is affected by precipitable

water content and CO, concentration, Here we derive

daily tevel 1 vapour pressures using the temperatures

and relative humidity. at level | and hence upper

envelopes for the control and doubled CO, GCM

runs. The GCM solar radiation is R, = R,, / (a)

where a is jhe albedo assigned to the GCM grid cell.

Generalised beta distributions are fitted to the

corresponding residuals (R,-R,) for each calendar

month and rainfall state. Parameter estimates for

doubled CO, residual distributions are obtained by

the method of moments, Here the historical residual

means and variances tor cactr monthare scaled by the

ratios of the corresponding doubled CO, and control

means and variances, respectively. The upper envelope

for doubled CO, conditions is phtained by scaling

historical daily vapour pressures by the ratio of derived

doubled CO, and contro) GCM vapour pressures.

4S RC BATES, & P. CHARLES, N. R, SUMNER & P.M. FLEMING

Hitter balance model

A modified version of Boughton’s (1984) SFB model

was used for the study (Fig. 1), The madel uses daily

rainfall and potential evapotranspiration data us input

to estimate monthly streamflow. The original model

has three parameters requiring calibration: the surface

slurage capacity (S$). the daily infiltration capacity (F)

cantrolling the movement of water from the surface

store to lower store; and the basellaw factor (8)

determining the portion of the daily depletion of water

in the lower store that appeurs as basefluw (O<B<1),

The faur remaining parameters are fixed: the fraction

of the surtace storage capacity that does not drain to

the lower store (VDC = 0.5): the maximum limiting

rate of evapotranspiration (£,,,, = %.9 mm d!); the

wetee

lower store depletion factor (DPF = 0,005); and a

baseflow threshold for the lower store

(SDR,,,, = 25 mm) defining the depth of water in the

lower stoce al which baseflow will cease.

roll

t Eg

PX 0,

Surface Storage

S

ie

Lower

t

Storage Ob

rt. =8-DPF'SS

| SDRmas

Dp

= (1 B}DPF-SS

Hip. 1 Structure of the SFB model.

Drainage

= (1-NOC}§

Non-drainage

=NDO'S

‘The model operations may he summarised as

follows. Rainfall begins to fill the surlace store and

any water in excess of the non-drainage component of

that store infiltrates into the Jower store at 4 maximum

daily rate of F mm d'. Surface runoff (Q.) oceurs

when the drainage component of the surfave store 1s

full and may be written as

QO, = P — F tanh (P/F)

where P = rainfull excess remaining after the surface

store is filled. The surfiice store contents are depleted

by evapotranspiration which oceurs at the potentlal rate

(E,,,) When the non-drainage component is full,

Otherwise. the actual evapotranspiration rate (&,,) is

given by

E, = min f&,,.5 / (NDCS): E,,,,|

where » = depth of water im the non-drainage

component of the surtace store. The lower store is

depleted by deep percolation (D,,) and baseflow (Q,,)

which are defined by

D, = (!-B). DPF SS

Q;, = BDPE(SS-SDR

wee )

where SS = depth of water in the lower store.

In this study, estimates of daily potential

evapotranspiralion in mm d! were obtained using the

Priestley-Taylor equation (Priestiey & ‘Taylor 1972):

L Ey = aR, ANA + ¥)

where L = latent heat of vaporisation in MJ kg", rz

= Lain this study, A = slope of the vapour pressure

curve in kPaeC', RX, = net radiation in MJ m? d'

and + = psychrometric constant in kKPa?C'.

Although the SFB model parameters have some

physical basis. they cannot be readily determined by

physicul measurement. Thus parameter estimates must

be obtained by fitting computed to observed monthly

streamflow hydtographs, Formal optimisation

techniques can be used to facilitate the estimation

process. These lechniques use a subjectively chosen

colterion (the ‘objective function’) to quantity

discrepancies between the computed and observed

hydrographs for a given set of parametér values. The

estimates of the inodel parameters are those parameter

values which result in the minimum possible value of

the objective function, The accuracy of these estimates

affects the accuracy of streamflow predictions.

In this study, the estimates for the SFB model

parameters were obtained using a simulated annealing

algorithm (Kirkpatrick ef a/, 1983; Press eral, 1992),

Simulated anneulme 15 a stochastic, multivariate

optimisation technique which seeks the global or near

global minimum of 4 user-defined objective function

without geting Wapped in a local minimum. The

function need not be smooth or even continuous. in its

domain. The method can he considered as a biased

random walk that samples the full parameter space and

provides @ solution that is independent of the starting

HY PROLOGICAL, IMPLICATIONS OF CLIMATE. CHANGE FOR SOUTH AUSIRALIA a9

point. I can aecept i ‘move’ thal increases the value

ab the objective function a5 a part of @ full series af

moves for which the general trend is to decrease the

lunctiom value, Dewils of tbe algorithm used will be

published elsewhere,

The objective functron used inthis study may be

Whitten as

a

mus Ls,

fw fF tel

where the model parameters vector 8 = (8. FB, NDC,

Ena DPF. SDR,,,.) in which the prime denotes the

transpose OF a Vector, #4 = number of tonths iy the

observed streamflow hydrograph record; & = number

ol months inthe ‘warm-up’ period which fs excluded

from the calculation of the objective funetion value;

und the disturbance z, ts defined by

MW « Ay

.) (QD, + Ad) > (Q, + Aa) yy &, #0, Ay=U

mowhich @ = observed streamflow. @ = computed

streamflow (Q = QO, + G\; and Ay A =

franstormation constants which can be estimated by

(nial and error (Bates & Walls I988), Generally, the

best ftw are obtarned with Ay < 1 und Ay = 0

Case Study

Deseription of catchment anet cletea

The above methodology was applied tn the Narth

Para River at Penrice catchment in South Australis.

The catchment comprises an weea of (8 kor and 1s

located within the Gawler River basin some 50 kn

north cast of Adelaide. It rises from 300 i to S00 m

AHD and has duplex scils and a mined ground

cover!, Land use includes horticuliure, viticullure,

sawing. and arable farming.

Daily streamflow at the Penrice sauging Station

(34° 28'S, 139° 4B) and daily rainfalls recorded at

Angustun Post Office (34° 30'S, 139° 3’B) and

Keynetun (34° 33S. 139° &'F) for the period January

1978 ta December 1989 were used in the study, Over

this period, the mean annual runoff was 6.200 MI

(30 mm) and the ean annual rainfall (65% of which

fylls mn the period May to September) was 550 inm.

The average daily maximum winter and summer

temperatures were 13°C and 29°C, respectively (Chiew

& McMahan 1993!).

| Crew, FH, S. & McMauon, T, A, (1993) Complete set

of daily minfall, potential evapotanspiration und streamflow

data for 28 unregulittud Australian catchments. Centre fir

Eoviranmenal Applied Hydrology. University of

Melbourne. 34 pp. (unpubl >

Estimates of daily potential evapotrunspiration were

Obtained wusing the guidelines described by Smith

(1991), clear day radiation and day length estimates,

und temperature and sunshine hours data from the

Nurivotpa Vittcuhural climate station (34° 29'S, 139°

O'R) which is located some 10 kim from the eutchment

cuntrotd,

Model calthration

Preliminary investigations reveyled that the

performance of the 3-parameter SFB model was

inadequate. However, satistagtory model fits were

obtained by selfing A, = Of E,.= 89 mm dt,

NBC = 0.5, SDR,,,, = Omm, und allowing the

parumeters $, 68, and DPF to vary dung ealibrarion

runs. The finul parameter estimates obtained wene:

S=l65 mm, F=0.9mm dl; #=HAld, and

DPF = 012, The B and DPF estimates indicate (hat

there is a significant loss of water from tie catchment

lo the regional groundwater,

An important assumption in this study is thal the S&B

model is able to simulate runoff from the North Para

cutchment under climati¢ conditions thit are different

to those for which the model has been calibrated, For

example, the eflect of possible chinges in plint

Lranspinition ralcs und vegeiative cover due tw CO,

doubling on the model parameters is ignored. This was

considered to be a réasoouble assumption given che

current level of uncertainty regarding the nature and

magnitude of these changes.

Results

To simplify the analysis of the results, the behaviours

of three variables under control (Ll « CO.) and

doubled CO, (2 x COs) conditions were examined:

tl) monthly raintall, (2) monthly evapoternspiration:

and (3) monthly total runoff. Synthetic ¢limane

sequences of 1,000 years duration were generated for

control and doubled CO, conditions and used io drive

the S&B model. Exploratory analyses of the data

revealed that the variable distributions were highly

skewed, Thus the median rather than the mean was

adopted as the measure of central tendency

Fig, 2 compares the distributions ol simulated

monthly caintall for control and doubled CO,

conditions, "Three features are worthy of note under

doubled CO, conditions: (1) there is an inerease in

minimum monthly rainfall for the period from June

io August and a decrease for the months of Mary,

September and October: (2) median rainfalls. for the

period from March 10 November ure greater than or

équil to those for present day conditions: and (3) there

ate marked increases in extreme (high) rainfalls for

the month of January. and the period from August lo

October.

Fig, 3. compures the distributions of modelled

monthly evapotranspiation for coairel and doubled

40 B. C, BATES, 8. P. CHARLES, N. R. SUMNER & P. M. FLEMING

TaBLe 1. Seasonality and distribution of monthly runoff for control and doubled CO z conditions.

Percentile Monthly Runoff (mm)*

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Noy Dec

Minimum 0 0 {) 0 0 0 0 QO 0 0 0 0

(QO) (0) (0) (0) (0) (0) (O)} (0) (0) (0) (O) (0)

10th 0 0 0 Q 0 0 0 17 10 0 0 0

(0) (0) (0) (0) (0) (Q) (0) (1.4) (0 4) (0) (0) (0)

30th 0 0 0 0 v0 0 1,3 7.3 5.0 03 0 )

(0) (0) (0) (0) (QO) (0) (1.5) (7.2) (4.3) (0,3) (0) (0)

501h 0 0 0 0 0 0 5.0 10.8 8.7 14 0 0

(0) (0) (0) (QO) (QO) (0) (3.0) (10.6) (8.4) (L,7) (0) (D)

70th 0 0 0 9] 0 Lo 87 7 1.0 4.2 0.1 0

(05 (0) (0) (0) (0) (h4) (8.8) (IL7) (11,0) (4,3) (OU) (0)

90th 0 0 0 0 0 58 IL.6 7.4 N14 8.5 0.5 0

(O) (0) (Q) (0) (()) (6.1) (H.6) (16.0) (13.9) (8,45) (0.6) (0)

99th 1) 0 0 0 3.5 10.9 367 61,5 A().4 1.8 3.9 0.1

(Q) (0) (0) (0.1) (4.7) (108) (44.7) (63.1) (89.5) (16.9) (37) (0,2)

Maximum 0) 0 0.6 4.9 1.0 412 69.5 99,2 53.2 51.9 68 0.4

(20) (2.5) 02) 27) 5) Ol) (966) (145) (103) (82.3) (76) (1.0)

* Values in parentheses are for doubled CO; cOnditions.

1 x CO2 2 x CO2

=) oO

R R

_ Oo oOo

—E & a

£ - _

= oa ~ — fo)

€¢ 2 22280 2 : :

= = | =

oO _ tr 3! : _ = - 5 =

2 =": = ole =: s+

=e Ff). El: z =~) 2) 7 =

= =E-s & = -- =

5 a i AEE: ,

2 °e uy 8 af

a a [. RA

F JF

| core ---

MAMJJASONOD MAMJJASOND

Month Month

Fig. 2, Box plots of modelled monthly rainfall for control (L x CO ) and doubled CO, (2 x COg) conditions. [Edges of

boxes mark upper and lower quartiles, and horizontal blank line within each box depicts the median. Distance between

ihe quartiles is the interquartile range (IQR). End points of whiskers attached to boxes denote either data extremes (no

adjacent horizontal lines) or adjacent values defined by upper quartile plus 1.5 < IQR and lower quartile minus L5 x IQR.

Horizontal lines mark data points that lie beyond adjacent values. |

HYDROLOGICAL. IMPLICATIONS OF CLIMATE CHANGE FOR STH. AUSTRALIA Al

1 x CO2 2 x CO2

150

150

100

I

100

i

+ TM me

-- OR

2 |

mT CMR - - - -

11 EMINC- - 208 --a

50

50

ee)

7m Ces: -- - Sail

Monthly evapotranspiration (mm)

MAMJJASOND J

MAMJJASOND

Month Month

Fig. 3. Box plots of modelled monthly evapotranspiration for control (1 x CO ) and doubled CO, (2 x CO3) conditions.

1 x CO2 2x CO2

Oo oO

sf =

oO Oo

nu ~

E 8 8

t 3 ° ~

°o = ie) ==

Cc — =

2 * ==

© = i=] = => =

~ ive) = _ © ia

xo > = =.

= 9 eas g sg:

Ss ga 5 - zs

2 AaR= ° 7M

m4 eS Ss Se

pag? es

S | a oS pecan.

JFMAMJJASOND JFMAMJJASOND

Month Month

Fig. 4. Box plots of modelled monthly runoff for control (1 x CO ) and doubled CQ, (2 x CO) conditions.

42 GB C, BATES. 8. P CHARLES, N. R. SUMNER & P.M. FLEMING

CO, conditions, There appears us be a veneral

increase in evapotranspiration for the period from

March to November under doubled CO, condifions.

Furthermore, a review of evapotranspitation values

greater than the 75th percentile point revealed un

increase in high evapotranspivation Values oVer the

entire year, The size of the increase is quite large for

the months of January, Februsry and November

Table | lists various percentile points for simulated

monthly runoetf tor control and doubled CO,

conditions, The distributions of the runoff data are also

Ulustrated in Fig. 4. Overall. there is lithe evidence

of changes in the median and upper and lower quartiles

ot monthly runotf. However, a ts evident thai the

catremes (high runotf values) tor the doubled CO,

climate are greater than those for present day conditions

for the period trom July t October. There are also

indications of reductions im the 10th. 30th and 50th

percemile pomts for Seprember.

Fig, 5 shows an empirical quantile-quintile plot of

the annual maxima of monthly runoff. The annual

maxima for both conditions are very similar lar

maxiina below their medians (about 1) mmy. Above

the medians, the maximuy are higher tor the doubled

CO, climate than for the present day and the

difference between the series appears to grow with

invreasing percentile rank, Thus there is a suggestion

of an increase in flood risk A review of the maxima

reveuled that the extreme events were due to the marked

increases in extreme rainfalls and modest inercases in

evapotranspiration for August and September under

doubled CO, conditions.

Percentile ranks. fot 1x CO2

55050

5

ta

99

Percentile ranks tor 2% GO2

Annual maximurn runotl tor 2x CO2 (mm)

t zo 40 60 20 00

Annual maximum runot tor 1x GO2 (mm)

Fiz 5, Empirical quamtile-quintile plot of annual maxima of

monthly runoff tor conrral (| = 05) and sloubled CO,

(2 ¥ COs) conditions

Conclusions

The most prominent effeer of doubled atmospheric

concentrations. of carbon dioxide on the water yield

of the North Para River catchment, South Australia,

is a Sizeuble increase in the annual maxima of monthly

runoff. “This muy indicat an increase in flood risk.

However, there appears to be litthe impacr on the

seasonality and magnitude of the median and upper

and lower quartiles: of monthly runoff.

The wbove results highlight the benefits: of using

stochastic weather generators in assessments of the

impucl ol climate change on hydrologic systems, The

estimation of extreme hydrologieal events based oni

original and perturbed historival records are not

Teliable when the return periods of these events are

of the order of (he original record length, An impasse

arises when the return periods of interest are preuter

than the period af record. Thus the length of the

historical revord Jimits what can be suid ubout the

changes tn the tails of the distributions of hydrologic

variables that may be caused by CO, doubling, In

contrast, our approach is capable of producing climatic

and hydrologic series of arbitrary Jength. This enables

amore detniled investigation of changes in interannval

variability,

The results presented herein were obtained using one

GCM (CS/ROY), a stochastic weather generator and

a water balance model (S78). Thas the quality of the

results rely heavily on the accuracy and relevance of

these models and the implicit assumption that the long-

lerm changes in vegetation that would be associated

with CO, doubling represent a second-order effect.

Nevertheless, it can be argued that the present

generation of GCMs are fundatientally similar and that

they differ principally with respect to their

patameterisuvion of certain processes. such as cloud

formation and rainfall, Thus the use of one rather thar

a suite of GCMs should not seriously: compromise the

results of this stucly.

Acknowledgments

This Work contributes to the CSIRO Cliniate Change

Research Program and is part funded through

Australia’s National Greenhouse Research Prograin.

We are indebted to FHS. Chiew and T.A. McMahon

(Centre for Environmental Applicd Hydrology,

University of Melbourne), the Bureau of Meteorology,

and the Engineering and Water Supply Department of

South Australia for permission to use their data sets,

Special thanks are due ta the CSIRO Division of

Almosphene Research for supplying CSIRO9 GCM

dats.

HYDROLOGICAL IMPLICATIONS OF CLIMATE CHANGE FOR STH. AUSTRALIA 43

References

Alpen, R. G., Gicuuxi, FN. & Rosenzwec., C. (1991)

CO,-induced climatic changes and irrigution-waler

requitements, <0 Hater Resour, Plug. Mem. Amer: Soc,

Civ, Eng. LI7, 157-178.

Bares, B. C., Crarves. S. Po & Fueminc, P. M. (1993)

Simulation of daily climatic series for the assessment of

climate change impacts on water resources. pp 67-72. In

C, Y. Kua (Ed.) “Engineering Hydrology,” (Amer. Soc.

Civ, Eng., New York),

BATES, D. M, & Warts, D, G. (1988) “Nonlinear Regression

Analysis and Its Applications.” (J. Wiley, New York).

RouGHTON, W. C. (1984) A simple model for estimating the

water yield of ungauged catchments. Civ. Eng. Trans. frist.

Eny, Aust. CE26, 83-88.

Carers, §, P, FLEMING, P.M. & Barns, B. C. (1993)

Problems of'simulation of daily precipitation and other input

time series for hydrological climate change models. Prac.

Hydral, & Water Resour, Symp. Inst. Eng Ausi. Nat. Con:

Publ. No. 93/14. 469-477.

Cowen. §. J. (1986) Climatic change, population growth, and

their effects on Great Lakes water supplies. Prof, Gevg.

38, 317-325,

Gietcn, P. H. (1989) Climate change, hydrology, and water

resources, Rev. Geophys. 27, 329-344,

Gorpos, H. B.. WHETTON, PH, Prrmock, A. B.. FowLnr,

A. M. & Haytock, M. R. (1992) Simulated changes in

daily rainfall intensity due to the enhanced greenhouse

elfect: implications for extreme rainfall events. Cliniate

Dynamics 8, 83-102.

Kirkpatrick, §,, GELATT, C.D. Jr, & Veccri, M. P. (1983)

Optimization by. simulated annealing. Science 220, 671-680,

LETTENMAIER, 1. P, & GAN, TI. Y, (1990) Hydrologic

sensitivities of the Sacramento-San Joaquin River basin,

Po ie to glahal warming. Water Resour. Res. 26,

69-86,

_& SHeer, D. P. (1991) Climatic sensitivity of

Calitornia water resources, J, Mater Resour. Png. Mygmi.

Amer. Soc, Ciy, Eng, U7, 108-125,

Matatas, N. C. (1967) Mathematical assessment of synthetic

hydrology. Water Resour. Res, 3, 937.945,

McCaskice, M. R. (1992) Generation of long-term weather

tlata sets for Charters Towers. Tropical Agronomy Tech.

Mem. No, 72, CSIRO Diy. Tropical Crops and Pastures.

Brisbane, Australia, 24pp.

McGreaor. J. L.. Gorpon, H. B.. WATTERSON, | G., Dix,

M. R. & Rovstayn. L. D. (1993) The CSIRO 9-level

atmospheric circulation model. Tech. Paper 26, CSIRO Div,

Aunos. Res. 89pp.

Mimixou, M., Kouvoroutos, Y.. CavapiAs, G. &

VaytANas, N, (1991) Regional hydrological effects of

climate change. J. Hydrol, 123, 119-146,

Nasu, L, Lo & Gieick, PH. (1991) Sensitivity of

streamflow in the Colorado Basin fo climaue changes. [hid

125. 221-241,

PanaGOouLiA. D. (1992) Impacts of GISS-modelled climate

changes on catchment-hydrology, Aydrot. Sei. J, 37, 141-163.

Pkess, W. H., Teuxousky, $.4., VerrerLinc, W. T. &

FLANNERY, B, P. (1992) “Numerical Recipes in FORTRAN:

The Art of Scientific Computing” 2nd ed., (Cambridge

Univ. Press. Cambridge, )

Proestiey, C. H. B. & Tayuor, R. LE (1972) On the

assessment of surface heat flux and evaporation using large

scale parameters. Mon. Weather Rev. 100, 81-92.

RICHARDSON, C. W. & WRIGHT, D, A, (1984) WGEN: 4

model for generating daily weather variables. Rept No.

ARS-8. U.S, Dept. of Agriculture, Agricultural Res.

Service, 83pp.

Smith, M. (1991) “Report on the Expert Consultation on

Procedures far Revision of FAO Guidelines for Prediction

of Crop Water Requirements.” (Land & Water Development

Div. Food & Agricultural Organization of the United

Nations, Rome),

Wiiks, D.S. (1992) Adapting stochastic weather generation

algorithms for climate change studies. Climate Change 22.

67-84,

IMPLICATIONS OF CLIMATE CHANGE FOR THE

SOUTH AUSTRALIAN COASTLINE

By NICK HARVEY* & TONY BELPERIOF

Summary

Harvey, N. & Belperio, T. (1993) Implications of climate change for the South

Australian coastline. Trans. R. Soc. S. Aust. (1994) 118(1), 45-52, 31 May, 1994.

Recent attention has been focused on the effects of climatic change on sea level and

the effects of a rising sea level on coastal environments. However, the variation in

physical and geological processes which are responsible for sea level change is often

overlooked or underestimated. This paper presents recent geological and tide gauge

data from South Australia to demonstrate that neotectonic and anthropogenic

influences have resulted in a general overestimation of the current rate of sea level

rise, apart from two sites where the reverse is true. The paper concludes that all tide

gauge sites must be corrected for vertical crustal movements before any conclusions

are drawn regarding local or global sea level change. The implications of this for

South Australian coasts are that adjustments to sea level trends should be made before

any vulnerability assessments are conducted.

Key Words: sea level change, coastal environments, climate change, vertical crustal

movements, South Australia.

Tronsaciions af the Novel Sevvery af S. dead. (1994), ARG), 13°52,

IMPLICATIONS OF CLIMATE CHANGE FOR THE

SOUTLL AUSTRALIAN COASTLINE

by Nick Harvey? & Tony BELPERIOT

Suminacy

Hanvry, N, & Boiehkin, T. (1993) Implications of climate change tor the South Australian coastline, Trans,

KX See. 8. Aas, (1994) LEIB(). 45-52. 31 May, 1994,

Recent attention his been focused on the effecrs of climatic change on sea level and the elfeets of a ral sca

lew! an coastal environments. However, (he variation in physical atid geological processes which ure responsible

Tor sed level change 19 often overlooked or underestimated. This paper presents recent geologic’ vial tide gauge

dati fron South Australis to demonstrate that avotectonic and ynthropapenic influences have resulted in i general

overestimation of the current rate of sea level rise, apart from swo Sites where the reverse is (rie. The paper

voneludes that ull tide gauge sites must be corrected Jor verte crustal movements before any conclusians are

drawn regarding focal or global sca level change, The implications of this tor South Australian cousts are That

adjustinents td sea level trends should he made hetore any vulnerability tasessments are conducted

Key Worps: sea devel change, coastal environments. climate change, vertipal crustal movements, South

Australia.

Introduction

‘The recent Greenhouse debate has focused attention

on the effects of climune change on sea level (Warrick

ef al. 993) and also the effects of a rising seu level

on coastal environments (Bird 1993, Tooley &

Jelversma 1992). Tn addition, the work of the

Intergovernmental Panel on Climate Change IPCC)

has produced scientific assessments of climatic change

(Houghton ef a/. Wl. Houghton er @/, 1992), weether

with the IPCC Common Methodology for assessing

the vwinerability of coastal areas to Sea level mse (IPCC

19M). However, there has been some critigism of the

applicability of this IPCC Common Methodology to

the Australian region (Kay er ul, 1992; Woodrotle &

Mcbeuty 1993),

Belore discussing the implications of this recent

research forthe South Australian coast, itis important

to note that there isa great variation in physical and

geological provesses which are responsible for sex level

change, ‘These variations, which have been categorised

by Pugh (1993) in terms of their spatial and temporal

influence. can vary Considerably from short term wind

waves with periods of about 10 seconds and an extent

af tens Of metres, up to global changes iit sea level

relaicd tw sea floor spreading: with time periods of

hundreds of millions of years. In addition, the pricess

of redistribotion of mass over the earth resulting From

devlaciauon, addition of meltwater to the beeans and

transgression and regression over the continental

shelves, wsclf results in a variety of isostatic fesponses

af the crust to the changing louds, The resultant lack

of uniformity in global sea level change is often

overlooked or underestimated.

* Mawson Graduate Centre tor Environmental Studies,

University of Adelaide, Seuth Australia SOOS_

P South Australian Departmental Mines and Enerey, PC) Baik

15. Enstwood, S$. Aust, 5063

Perhaps the most studied geological period in terms

of sea level Change is the Quaternary (Williams er af.

1993) where the effect of climatic change has resulted

in numerous Muctuations of sea level in response to

the waxing and waning of continental ice shects.

Consequently there has been alternate flooding and

exposure ol-continental margins together willl periods

of erosion and sedimentation

The most recent time when climatic conditions and

sea level were similar to the present was during the

lase interglacial period a around 125,000 years before

present (BP) when sea level in the Australian region

was between 2 and & m higher than today (Chappell

1987). Since then, sea level Nuctuations have always

been lower chan present with evidence from the

Australian region of a law sea level of hetween 130

and 165 m lower (Chappell 1987) at 18,000 years BP,

afer which if rose at.a rate of between 6 and 12 mi

vr! prior to reaching its present level between 6000

and 7000 years BP.

ft is in the record of the last 6000 years that scientists

have focused attention on finding evidence for any long

term trends in sea level cither by direct sea level

indicators in the geological record or by analogous

paleoclimatic change evidence. Careful field studies

from many coastal localities have supported

geophysical models which indicale that subtle

differences in sea level behaviour are the norm even

for the South Australian coast, long regarded as stable

and uniform (Lambeck & Nakada 1990),

There has also been direct measurement of sea level

and detailed analysis of recent tide gauge records te

extripalate relative trends (Gornitz, 1993; Pirazzoli

1991), As noted by Bird (1993) there are humerous

lactors ulfecting rélative sea level change. Apart from

eustatic sea [evel change there iy the tectonic response

of the Jand and the isosialic response of the continental

margins relating to changing volumes of ice. water or

a6 N. HARVEY & T. BELPERIC)

sediment. In addition, human aeniyities such 4s

groundwater br hydrocarbon extructron, land

reclamation, arihicwl coastal sructires, dredging and

pumping of sediment can atteet local sea level change.

Pirazzoli (O89) sugeested that local seculur lide pauge

data are dominated by neotectonic and anthropogenic

effects, resulting in an overestimation of glubal sea

level ise hy 2 lu 3 dames when these factors are ignored

South Australian policy of coast protection

and new develupment

The current South Australian policy on const

protection amd new cuustil development was prepared

by the Coast Protection Board and endorsed by the

South Australian Government in May 91. The policy.

Which is deseribed in detail elsewhere (Coast

Prection Boaid (992). relics in part oo loca records

of coastal erosion. flooding and sea level rise bul mone

importantly has incorporated the IPCC estumiutes ul

wrcenhouse inmluoxd eustatic sca level rise. These

estimates predict a seu level rise to the year 2100 ul

approximately (65 m (range 0.33 m to 1.10 m) for a

“business as usual” scenamo (Houghton eral. WY9l).

Given these extimates the Coast Protection Board used

the “preciuinnary principle” .in preparing ity policy,

The precautionary principle which was adopted by wil

Australian governments states that “where there arc

(reals oF serious or ureversible environmental damage,

lack of full scientific eertainty should not be used as

a reuson lor postponing measures to prevent

emjronmental degradation” (ntergoverninenial

Agreement on the Efivironment 1992, para 3.5.1).

In acgordance with this principle the Coast

Protection Board has adopted the policy that any new

coastal development should be capable al being

reasonahly protected from a bm sea level rise by the

year 2100, The policy estblishes the 100 year average

return interval (ARI) water level as a standard for

coastal development in South Australia. It reconimends

thal sile and building levels should be determined by

adding {4 m to the }00 year ARID water level and

(where appropriate) making an adjustment for locahsed

subsidence or uplift, Floor levels of buildings should

hewn additional 0.25 m above this level. and buildings

should not be approved unless they are capable of being

proweted of raised (o withstand a further 0.7 m of sea

level rise (c.g. by means of a bund wall). In the case

af Mood protected sites. the calculation of the 100 year

ART design flood level must incorporule the extreme

ude (plus surge) and stormwater everts, together with

wave effceis within te development.

The policy also makes a general recommendation

for an crosiog setback distance, This: is to he based

wron 100 years OL erosion at a site, allowing for local

coaslal processes and g sea level rise OF 03 11 to the

yeur 2050, und taking account of slurm erpsion from

a series of severe storms. For majar coastal

development tt ts suggested that calculuejons are birsext

Upon 20) years of erusion.

The policy is less spevitic about the protection of

existing property although a reaffirms an earlier

Government policy not to protect private property,

Although part of the Coast Protection Bouru's duties

are lo protect the coast. most coast protection works

wre carried by local councils. Acvording tothe (991

policy statement, the Coust Protection Bowrd provides

councils with grants of up to 8U% Ob the cust of

approved coast protection works and up to the same

winount tor storm damage repairs. The issue af uns

sharing bewween State and local covernmer os currenny

an issue of debate bue the underlying question for any

protection Works is essentially a devision whether to

protect or reloeate. This is conyplicuted by the level

af public of private involvement and the relative

responsibilities i State and lucal government. bor this

reason the Coast Protection Board and local

2overnment determine these sues an & case hy case

situation,

Global sea fevel rise estiniates

A major problem in identifying the corrent rate ol

eustutic sea level change Irom tide gauge data as the

influence of neotectonie, isostatic and anthropogenic

effects, compounded by a gengraphicul biey in the

dhsteibuvon of reliable tide gauge data (Warrick 1993;

Gornitz 1993; Aubrey & Emery 1999: Woodworth

1993). These problems create uneertuinry 1

extrapolating the custatic companent ofsea level change

and bas Jed authors such as Aubrey & Emery (1995)

ro-expressed caution in attempting lo exlrapolate actual

sea level changes ftom the data. They supecst thar the

apparent post 1930 accelerated sea level cise muy be

relaled to factors other than human induced fichucs such

as a Uclayed response to climatic warming following

the Litle Ice Age, oceanographic factors, or perhaps

my nat even be statstieally significant.

Mrker authors such as Gornitz. (1993) suggest that

after extraction of long-ternt trends and data averaging,

thalatas possible to obtain w true picture of sea level

rise, Gornitz presents evidence based on 16 tide piauge

Ua studies Wy suggest that estimates of global sea level

tise Over the last 100 years has been between 0.5 and

3mm yr’, wath most estimales in the range of 1 fo

2mm yr! (Gernitz 1995),

In addition lo studies alterpting to idenuty eustatic

sea level chunges based on woulysis of tide pauwe dala,

there are alse a timber af studies on sea level rise

projections related co climate change. The kev study

his been the [PCC seu level rise predictions with uw

best estimate of a 165m rise to the year 2100

(Houghton eral. 90) upon which the South

Australian coastal policy has been based, ‘The IPCC

IMPLICATIONS OF CHANGE FOR THE §.A. COASTLINE 47

report provided a significant downwards revision of

earlier sea level rise predichons but more recent

calculations have continued to produce similar best

estimate figures cither by more qualitative expert

analysis. (0.61 m by the year 2087, Woodworth 1993),

or by detailed re-calculation (0,46 m by the year 2100,

Wigley & Raper 1993).

The South Australian coast: Secular sea level rise

Records of sea level in Australia, from the period

1897 to present, have been monitored and analysed by

the National Tidal Facility (NTF) at Flinders

University in South Australia. As for the global

situation, there is significant spatial inhomogeneity in

the secular sea level trends resulting from the myriad

435°

Thevenard

EYRE

PENINSULA

Port

Lincoln

§@ Tide gauge sites

KANGAROO |S,

Victor Harbor

SIR RICHARD

PENINSULA

SOUTH

AUSTRALIA

Fig. 1. Location of tide gauges in South Australia.

Onkaparinga R.

of factors affecting relative sea level behaviour at each

lide gauge site. Although there is a variation in the

quality of tidal records available, the NTF analysis of

lidal data from gauges with an acceptable datum

stability indicate am Australian average of

151 +018 mm ye! at the 95% confidence level

(Mitchell (991). This figure 1s in dgreemtent with global

analyses of 10-20 mm yr! (Gornitz 1993),

notwithstanding the caution expressed about the

validity of these figures (Aubrey & Emery 1993).

South Australian sea level wends based on tide gauge

data (see Fig. 1 tor location of tide gauges) have been

presented by Mitchell (1991). although there are

significant tidal records such as Port Augusta, which

have yet to be analysed by the NTF (Table 1).

139°

Port Augusta

Port Pirie

Port Adelaide A.

@ ADELAIDE

$e

>

4

<)

Lake

Alexandtina

36°

Creek

YOUNGHUSBAND

PENINSULA

|

|

|

|

|

|

|

|

|

|

Mt

e Gambier

Port Macdonell 38°

48 N HARVEY & T. BELPERIO

South Australia — Neotectemic contribution

ta sea level change

Australia is frequently. and incorrectly, quoted as

a stable continent from which absolute sca levels can

be measured. Neovtectonic inovements caused by

structural geoidal. and isostatic processes. together

with factors such as sediment compaction, affect

Australia at various spatial and temporal seales. This

complexity of underlying factors that control relative

sea level change has been demonstrated by the

reconstruction of palaeu-sea level histories from

numerous sites around Australia and South Australia.

‘The sea reached its present level around the South

Australian coast beeween 7000 and 6000 yr BP. South

Australia is in the “far field” in its response to global

devlacttion, and is affected by subtle, onguiny isostatic

adjustment of shelf and coust. This is manifested us

an apparent highstand of the 6000 yr BP shoreline (Fig.

2), the height of which varies systematically and

predictably afound the coast. In particular. the height

Upmar Spencer Guit ¢4_fim)

Por! Wakeltlald (9.0re)

Port Gueter (2 1m

Cottin Bay €O.8m)

aes

Freseal_ Sea.

PALAEO SEA LEVEL RELATIVE TO PRESENT LOW WATER DATUM (mt

4

te] | e 4 4 5 5 7 2 q In

RADIOCARBON YEARS - BP (thousand years)

Tig. 2, Paluco seu levels relative to present sew levels at Various

sites in South Australia

of the highstand mereases up the two galts. with

increasing distance from the contimental margin, Sea

level change over the past few thousand years is

dominated by this regression, which increases in

magnitude from tm or less along Eyre Peninsula to

3.0 m at the head of Gulf St Vincent and 4.5 m in

Upper Spencer Gulf, Isostatic adjustments thus visry

from O.l mm yr? to 0.8 pam yr! averaged over these

ume scules. This has caused slow bul abvinus coastal

regression, particularly at the heads of both gulfs,

Superimposed on this geographically variuble

Holocene isostatic warping are longer term teclome

inevements, Tectonic effects are most noticeable along

the South East coastal plain, between Lake Alexandrina

and Mt Gambier, where Quaternary voleanisin has

resulted mm ongoing uplift and tilting of the coastal plain,

The scale and variability of this upwarp can be

iustrated by the changing elevation of the last

imerglacial shoreline (Fig. 3). This 125,000 year olel

shoreline rises progressively southwards, from 3 i

above present sea level at Salt Creck. fo in excess ol

IK m near Port Macdonnell. Uplift rates in the Port

Macdonnell region are a minimum of 0.2 mo ye! if

averaged our aver this entire time period.

Another, under-rated effect associated with cies 15

sediment compaction and land subsidence associated

with coastal reclamation and withdrawal of

underground fluids (Bird. 1993). Such effects are local.

but are sufficiently frequently associated with harbours

and tide gauge sites as to seriously question the validity.

of global averages obtained from their secular trends

(Davis 1987; Pirazzoli 1989).

The Jocal record from Port Adelaide clearly

illustrates these effects and the inherent danger of using

lide gauge data without adequate neatectonic correction

(Belperio 1989. 1993).. Daw from tide gauges,

muingrove migration patterns and trom dated subsurfice,

strata all indicate a contemporary relutive rise in sti

level within the Port Adelaide estuary. The geographic

restricdion of these effects to the Port Adelaide regio,

together with preliminary geodetic evidence, indicute

that the apparent rise in sea tevel is a local

phenomenon, resulting chiefly. from subsidence of the

lund. Belperto (1993) concluded that up ty 1.0 m ot

surficial compaction and land lowering had occurred

in association with wetland reclamation, acid sulphate

sou development, in¢reasing urban and industrial

development and groundwater withdrawal. Highly

variable rates of land subsidence, between 18 and

10mm yr'. were estimated to be occurring over

different parts of this regiou. More significanily, some

three quarters of the secular rise of sea Jevel indicated

by the Port Adelaide and Quter Harbor tide gauge dita

could be attributed to land subsidence over the last 50

years at this location.

Corrections to tide gauge derived sca level trend

data should be made for these various neotectoniv

IMPLICATIONS GF CHANGE FOR THE S.A, COASTLINE w

contributions. Preliminary corrections have been made

for South Australian tide gauge sites incorporating

known neotectonic variations (Table 1), The tide gauge

data from Port Adelaide and Outer Hatbor have been

used in global and Australian sea level rise averages

without adequate local negtectonic correction. Many

of the world’s tide gauges are similarly biased by land

subsidence effects, indicating the importance of making,

such local neatectonic corrections to all tide gauge data

before inferring local or global sea level changes.

TABLE t. See level trends caleulated from South Australian

tide vuuve data.

preliminary

years of trend* adjusted

record mit yr trend’

min yr!

Port Adelaide (Outer Hb) 48.2 2.82 06

Port Adelaide (inner Hb) 35.1 2.2% 0.5

Port Lincoln 25.4 0.90 07

Port Macdonell 217 0.49— 0.6

Port Pirie 5I- -0,20° 0.3

Thevenard 24 114 10

Victor Harbor 238 117 Vd

*Source: (Mitchell 199), p: 355) except ? Mitchell 1993. pers,

comm.) *

this paper,

Kielnerabilicry of the South Australian coast

io seu level rise

A number of papers presented at the Australian

“Greenhouse 87 Conference” discussed the gencral

coastal impacts of a greenhouse sea level rise around

the Australian coast (sec Pearman 1985). However,

very little work has been conducted on the vulnerability

of the South Australian coast to current erosion

processes, ar on the effects of an accelerated sea level

rise. Harvey: (1993) provides data on the sensitivity of

selected South Austrahan coastal environments to

development, Fothcringham & Caton (1989) give a

broad overview of potential impucts of a greenhouse

sea level rise othe South Australian coast, and Wynne

(1989) examines implications of a sea level rise for

coastal erosion and flooding.

The South Australian coastline is approximately

4000 km long tncluding a variety of coastal landforms,

including clifted coasts, rocky outcrops, mangroyes.

mudflats, extensive sandy beaches, coastal dunes, and

a number of off-shore reefs and islands. These arc

associated with a range of high energy exposed open

ocean coasts through to the protected low energy

shorelines of the upper gulfs.

The immediate impact of any sea level rise will be

to increase the magnitude and frequency of extreme

tides and levels of storm erosion, although the effects

of this will vary greatly around the coast. The least

vulnerable areas will be the resistant rocky coasts of

the Fleurieu Peninsula, Kangaroo Island and Eyre

Peninsula but there is likely to be greater erosion on

the more predominant softer aeolianite and Teruiary

limestone rocky cousts. The actual rate of chill or blutt

retreat on these coasts will vary with factors such as

Tock resistance, structure, the presence or absence of

shore platforms or nearshore reefs, exposure to wave

action, and tidal range (Bird 1993).

On the sandy coasts which represent about half

(1900 km) of the South Australian coast are likely to

have a variable response because of differing rates of

littoral drift, onshore-offshore sediment movernent, and

sediment size variability, The high energy beaches of

.

PORT MACDONNELL

ROBE

ry, + BEACHPORT

1 12 =——* ELEVATION OF THE 125,000 YR BP SHORELINE

Ww

1 10

<

uu

ie] B .

5

7 ir

O65 iva

re .

; :

wg a

6 : ‘

a —*

ii 2 a ee _— X. /

rd 7 -

im 0

=

-3 EYRE PENINSULA SPENCER GULF FLEURIEU SOUTH EAST

GULF ST. PENINSULA

VINCENT

hig. 3, Changing elevation of the last interglacial shoreline (125000 yr BP) m South Australia.

50 N

the Younghusband and Sir Richard peninsulas for

example are barked by an extensive dune coastal barrier

system which would be yulnerable to increiwsed staqn

utlack wilh elevated sea levels. Commline retreat and the

development of dane blowouts is likely to cause a

migration of the barrier towards the Coorg, As this

eccurs underlying calercte ant back harrier muds would

hecome exposed causing variable rates of retreat, In

addition raised water levels will impact on the Coorong.

In contrast. the metropolitan sandy coast lacks the

exrensive backing dune barners of the south-east, Urban

emeraachment across the frontal dune together with

extensive protective works have necessitated a sand

replenishment program to maintain the beaches, Elevated

sea levels are likely to have greatest financial impact

Wi this ared where starm protection will need to be

upgraded tigether with an increased sand replemshnical

programme, if the metropolitan beaches are to be

maintained, To the north of metrupolilan Adelaide, tand

subsilence has already been noted for che Port Adelaide

area, Pleveted sea levels will exuverbite the rate of

relalive sea level rise culsiny inangeoves wo advanes

further inland and changes to ecological zonations of

the intér-tidal wid supra-tidal biota. Jn sume plucus

mangrove advanvé may be restricted hy wrtificial

embankments resulting in mangrove dic back-

Th the gulf retions, similar displacement of ceological

commuities such as seagrasses, mangoes and

sampires would) be pronuiunced along the low ereahent

coasts. This rapid coastal retreat would be ussogiated

with reactivation of ual swamps, localised flooding of

the coastal plains and erosion of beach ridge systents.

In-other parts of the coast. an clevated sea level is

likely to cause flooding of low lying land, enlargement

a) coastal lakes and/or connection of some lakes to the

sea, mised groundwater levels ind alteratim (o estuarine

environments. South Australia has few estuunes

ulthough there could be major implications for urban

development adjacent to estuaries such as the

Onkaparings and the Port River. In the case of the Port

River estuary. upproaimeately 25% of the nearby urban

Ucevelopmenc is currently below high water, Elsewhere,

potential inipacts an the Murny River residual esiuury

(artificially consinéined by the Construction of bartages)

may be less significant for urban areas but could have

major implications for thé operating levels of the

hurrages und iiffect (he management of the Murray

Mouth and lower Murray Lakes region (Harvey 1988),

Assessing oulnerabilirys The IPCC common merhidology

In late IS91 the Coastal Zone Management Sub-group

of the [PCC released its “Common Methouology” for

the asscssmientof vulnerability of coastal areas tn sea

level vise (IPCC W491). An advisory group compresing

the United Nations Environment Progran (UNEP) and

12 nations Initiated a series of case studies to examine

HARVEY & ‘l

HELPERICY

appropriate national response strategies, and

linplementation requirements for cousts vulnerable tu

sea level rise. The Cotmnon Methodology lor assessing

vulnerability Comprises the following basic steps:

1) Delineation of case study ured and specification of

accelerated sea level rise and climate change

boundary condilions,

2 Lnventory of study area characteristics.

3 Projection of relevant development factors,

4 Assessment of physical chinges und naturil

TESPOMscs,

5 Formulation of response strategies and assessment

of their costs und effects.

6 The assessment of the vulnerability profile and

interpretation ol results.

7 Edentifcstion of actions to develop a long tem

coastal Zane management plan_

The assessment process, using !hese steps, 1s

explained in detail together wil tables and checklists

to assist un the compilation of data in 4 consistent. manner

GPEC 1991).

The Common Methodology approwch was fnund bo

be. deficient by Kay ef ail. (1992) in thew Western

Australian case study, They sugges} thar there jire

problems in the biophysical structure of the pssessment

and also in the engineering. dommated approweh bo cost

benelit response mechanisms. They also: indicate: that

there is an essential slep between the assessinent of

impacts and the formulation of a policy response. to

reduce those impacts, In Wester Australia, Ihe policy

formulation includes inter- and jntra-yovernment liaison,

public consultation, and political consultation, eventually

leading to strategic coastal zone management, However,

the lact |hal strategic coastal engineering decysions are

part of wider State und regional planning issues, means

that preference may be given to reactive coast protection

strategies fur short tenn eroston problems rather than

long-term coustal strategies related to sea level rise. This

type of problem highlights some ol the difficulties in

developing a common approach strategy encompassing

biophysical. wdininsteative and legislative factors (Kay

chal, WAR),

Tn South Australia, Ihe PCC Common Methodology

has yet to be tested but tt is likely that butemueratic nnd

politcal problems are curtently compomided by

uncertainty with the review of the 20 year old Cpasr

Protection Act, debale twer cost sharing for coastal

management between Stile and local government,

Discussion

Appropriateness. of current South Australian poitew

Tnitial estimates of the current Tate of sea level reste

obtained by averaging tide gauge dara from around the

world have been malucing., The principle reasui is che

recognition that lind level changes need to he pemenved

IMPLICATIONS OF CHANGE FOR THE S.A. COASTLINE S|

Troum (he tide gauge data before they can be used for

this purpose. Underlying, unrceognised neuieoimic

effects renal the mails reason for the geographic

variubility in secular tide gauge trends A global seo

level cise eannut be cxpecied to be detected untel

udequute Correctivins are ade for these effects at euch

lide gauge ste. With consensus estimates of predicted

sea level rise now down to 4,65 m over the next 100

years, these neotectonic effects will be determining

(actors an overall liscu! sea level behaviour, In the gulfs,

iiny rise of sea level will be mitigated by ongoing

iwostalic upwarp of up ti O83 mm yr'. In the

South-Rash, tectonic uplift will be ua mnitigating

circumstance. In major lowes, particularly where

wetland reclamation has occurred, or excessive

groundwater withdrawal is taking place, any

greenhouse sea level rise will be exacerbated by land

subsidence effects.

In the absence of accurate geadctic or wltimetric data,

the geolpgic record can be used ta obtain fipst order

estimates of land Jevel changes. In South Australis. the

chastal record has produced some useful results that

go some way towards explaining the variability ol sea

level change documented hy tide gauges. However.

there remains & pressing need for locul and global

crustal scule geodetic control of all tide gauge sites

In addition, a geographically suitable spread of tide

fue sites is required to account for the scale and

variability of neotectunic processes. In South Australia.

these fide gauge sites yre required within the gulfs as

well as along the oceanic coastline and offstore islands

The implications of this for South Australia’s coasts!

policy are twofold, First, it bighlights the need for

accurate local sea level data which are then corrected

for neolectonic and anthropogenic factors. The

preliminary corrected sea level trend figures presented

inthis paper indicate that the current rate of sea level

ns¢ in South Australia has generally been overestimated

apart lrom two sites where the reverse is true, Second.

ihis South Australian case study provides support fer

Pirazzoh’s concerns that (he global averages lor current

custatic sea level rise may also be overestiomated. The

compounding effect of uncorrected global sea level rise

trends added to uncorrected local trends will produce

inaccuracies in the sew level trend data upon which the

policies are based.

In addition the uncertainties. Surrounding elimate

change predictions und the associated sea level response

necessitate the adoption of the precautionary principle

allowing safety margins for building levels and erosion

set-backs, Although this: may huve major cost

implications lor copstal development, it is unlikely that

a grealet precisian for climate change models and sea

Jevel response will be reached in the near furure.

However, IL ts possible to reduce stme af the

uncertainties of current sea level measurements by

correcting sea level data for neotectonic and

anthropogenic influences.

Conclusions

The dala presenred in this paper indicate the need

for taking noutectunic und anthrupovenie factors into

atcount for calculating currence rates of seu level rise.

The secular trend obtained from tide gauge data 1s only

a relative sea level trend since jhe tide gauge cannot

distinguish between a real rise of sea level and crustal

subsidence al (he site. ‘This paper indicates thal the

current rate of sea level rise based om South Australia

fide gauge dita has genecally been overestimated apart

From two sites Where the reverse is true. [11s apparent

from this paper that all tide gauge sites must be

corrected for vertical crustal movements betore any

conclusions wre drawn regarding local ur global sea

level change. The implications of tis for South

Australian coasts are that adjustments to) sea level trends

sfiould be made before any Vulnerability assessinents

are conducted. It may also be possible to revise policy:

guidelines baxed on the revised sea Jewel trend data.

References

AuARFY. BD. G & Emery. K. ©. (1993) Recent global sed

levels and land levels pp. 45-56 /a Warrick, Ro A., Barrow,

EM. & Wigley, T M_ 1, (fds) “Climate and sea evel

chunwe, observations, projections and implications”

(Cambridge University Press, Cambridge).

BeLrerto. A. P (1989) The greenhouse debate and rising s¢s

levels: geological and other fictors controlling sea ‘evel

change pp. 77-00 It Dendy, T. (Ed.) “Greenhouse "BR:

planning for climate change” (Department of Pnvironmem

and Planning, South Australia),

__ (1993) Lund subsidence und seu level rise in ihe Port

Adelaide estuary: implicubons Jor moniporing she

greenhouse effect. Australian Journal of arth Scien ic AQ

499-4AK.

Bik. B.C. F 11993) “Submerging cousts: the eects oF a

rising sea level on coastal environments” Hohn Wiley &

Sons. CRichester).

Cuastat, PROTECTION Boaniz (1992) Comsial erssivn, Hooding

and sea level rise standards and protection policy Cyrtst/irte

26.

Tavis. G. H. (1987) Land subsidence and sea level rise on

the ALAC Coust phi Ardeermental Gevliey anil Water

Science 1, 67-80,

POTHERINGHAM, D, & Caros. B (1989) Soulh Australian

coastal landforms response te a greenhnuse sea level rise

pp. 77-80 Jn Dendy. T, (Ed.| “Greenhouse ‘88> planning

tor climate change" (Deparment of Environment and

Planning, South Australias,

"ty

Gorwitz, V. (1993) Mean sea level changes in the récent past

pp. 25-44 In Warrick, R. A,, Barrow, E. M. & Wigley,

T. M. L. (Eds) “Climate and sea level change: observations,

prajections and implications” (Cambridge University Press,

Cambridge).

Haavry, N_ (1988) Coastal management issues for the mouth

of the River Murray South Australia Coastal Managemenr

16(2), 139-149,

(1993) Sensitivity of South Australian environments

la marina construction pp. 327-342, /n Thomas, D. S. G,

& Allison, R, J. (Eds) “Landscape sensitivity” John Wiley

& Sons, Chichester).

Houcuron, J. T.. CALtenper, B. A. & Varney, &. K,

(1992) “Climatic Change 1992. The supplementary report

to the [PCC assessment” (Cambridge University Press,

Cambridge) -

Houcuton, J. T., Jnvnins, G: J. & Epriraums, J, J, (991)

“Chmaic Change. The IPCC scientific assessment”

(Cambridge University Press, Cambridge)

IPCC (1991) “Common methodology for assessing

vulnerability to sea-level mse” prepared by the IPCC Coastal

Zone Management Subgroup (Ministry of Transport &

Public Works. The Hague, Netherlands),

Kay, R. C., Eitor. L & Kiem, G. (1992) “Analysis of the

IPCC sea-level rise vulnerability assessment methodology

using Geographe Bay SW Western Australia as a case study:

im coastal risk management” (Report to the Department of

Arts. Sports, Environment and Territories)

MiTcHete, B. (1991) Sea fevel and climate change, pp:

327-342, Proceedings of the Second Australasian

Hydrographic Symposium Special Publicaiton 27

(Australian Hydrographic Sociely, Sydney).

PEARMAN, G. 1. (1988) “Greenhouse "88: planning for climate

change” (CSIRO Division of Atmospherit Research.

Mordialloc)

Pirazzoit, P. A. (1986) Secular Trends of relative sea level

(rs!) changes indicated by tide-gauge records Journal of

Coastal Research 1, 1-26,

(1989) Present and near-future global sea level

change, Paleogeagraphy, Paleaclimatology, Paleovcolagy

(Global and Planetary Change) 75, 241-258.

2 N. HARVEY & T. BELPERIO

PuGH, D. T. (1993) Improving sea level data pp. S7-71 In

Warrick, R, A.. Barrow, E. M. & Wigley, T, M. L. (Eds)

“Climate and sea level change: observations, projections

and implications” {Cambridge University Press,

Cambridge).

‘Tootry, M. J. & JecGersma, §, (1992) “Impacts of sea-level

rise on European coastal lowlands”, (Blackwell Publishers,

Oxford, }

Warrick, R, A. 1993) Climate and sed level change: a

synthesis pp-3-21 Ja Warrick, R. A.. Barrow, E. M. &

Wigley. T. M. L. (Eds) “Climate and sea level change:

observations, projections and implications” (Cambridge

University Press, Cambndge)

. Barrow, E. M. & WicLer, T, M. L.. (Eds) (1993)

“Climate and sea level change: observations, projections

and implications” (Cambridge University Press,

Cambridge).

Wiciry, TM. L. & Rarer, S.C, B. (1993) Future changes

in global mean ternperature and sea level pp. 1135 In

Warrick. R, A.. Barrow, E, M. & Wigley, T, M, L. (Eds)

“Climate and sea level change: observations, projections

and implications” (Cambridge University Press,

Cambridge).

WiiiamMs, M.A. J., DUNKERLEY, D. L,. Dr Drexker, P.,

Kersiaw. A. Po & Stokes, T. (1993) “Quaternary

environments” (Edward Arnold, London),

Wooororre, C. D. & McLean. R. F (1993) “Cocos

(Keeling) Islands: vulnerability to seaJevel rise”

(Depariment of Environment, Sport and Territories,

Canberra),

Woopworri. P-L. (1993) Sea level changes pp. 379-391 Jn

Warrick, R. A., Barrow, E. M. & Wigley, T. M. L, (Eds)

“Climate and sea level change: observations, projections

and implications” (Cambridge University Press,

Cambridge),

Wane. AR. (1989) tnplications for coastal erosion and

flooding in South Australia pp. 77-80 Jn Dendy, T. (Rd.)

“Greenhouse ‘88: planning for climate change” (Department

of Environment and Planning, South Australia).

CLIMATIC CHANGE AND ITS IMPLICATIONS FOR THE

AMPHIBIAN FAUNA

By MICHAEL J. TYLER*

Summary

Tyler, M. J. (1994) Climatic change and its implications for the amphibian fauna.

Trans. R. Soc. S. Aust. 118(1), 53-57, 31 May, 1994.

The dependence of frogs upon moisture makes them highly sensitive to environmental

conditions. The South Australian frog fauna includes 28 species which collectively

and individually experiences a wide temporal range of temperature and available

moisture. Any climatic change involving warmer and moister conditions is likely to

enhance their distribution and abundance.

Key Words: Frogs, moisture, distribution, South Australia.

‘Transactions at te Reval Spetery of 8, Aust, (1994), WI8(1), 53-57.

CLIMATIC CHANGE AND ITS IMPLICATIONS FOR THE AMPHIBIAN FAUNA

by MICHAEL J. TYLER*

Summary

‘Tyuer, M. J. (1994) Climatic change and its implicationy for the amphibian fauna, Trans. R. Soc. 3 Aisy, HIR(L),

53-57, 31 May, 1994

The dependence of frogs upon moisture makes them highly sensitive to envirenmental conditions. The South

Australian frog fauna includes 28 species which collectively and indivieually experiences a wide temporal range

of (emperature and available moisture: Any climatic change involving warmer and moister conditions is likely

tw-enbance their distribution and abundajice

Kry Wore; Frogs, moisture, distribution, South Australia.

Introduction

‘The relevance of the study of frogs in.any cvaluation

of the impacts of climatic change hinges upon

recognition of the dependence of these animals upon

tnoisture, Nevertheless the survival of frogs throughout

the period that witnessed the entire evolution of all

other terrestrial vertebrates, demonstrates the capacity

ot frogs to survive massive environmental changes.

The complete extent of the diversity of the modern

frog fauna is unknown: Duellman (1993) estimates a

total of 3967 species ut 31.xi),91, and numerous species

have heen described subsequently. In Australia the

current total is 203, but many more await description.

Despite the success and Jongevity of frogs as an

evolutionary lineage they remain dependent upon

moisture because of the relative permeubility of the

skin and (in most species) the need to deposit eggs in

tree bodies Of water.

It follows that frogs are highly sensitive indicators

of environmental pollution and, equally, that they

eomribute an carly warning system in terms of

detecting environmental changes.

Any climatic change in South Australia at a regional

or tobi level is likely fo impact upon the distribution

of species, and upon the number within (his State,

South Australian frog fauna

Currently 28 species of frogs have been reported

from South Austria, The largest number of species

in any area (1) is to be found in the lower southeast

whith is also the area of highest rainfall. However

comparable numbers are found in the arid northwest

(9) and northeast (10) (Table 1).

The northwest and the northeast are also the portions

of South Australia where numerically and

proportionately there are the most species not shared

wilh other areas (Table 1). In each instance these

“unique” species are known within South Australia

“ Department of Zoology, University of Adelaide, South

Austraha S0QS,

from fewer than six Jocalities, and each specics is more

widely distriboted outside the State. Numerous frog

species in South Australia are at the geographic limit

of their distribution, and climatic change may have a

dramatic influence upon their persistence or

abundance.

Tanti ty Geographic characte risies af South Australian frog

fauna.

Spenes = Known from

Nomner of = confined fo ess than six

Gengraphic Areu spevigs That urea localities

Northwest 9 aa 5

Northeast 10 5 5

Flinders. Ranwes 5 | 0

Fyre Peninsula 3 0 0

Yorke Peninsula 2 0 ()

Mt Lofty Ranges 7 0] 0

Murray Valley 9 3 iF

Kangaroo Island 6 i) 0

Lower southeast iy 4 )

Nullarbor Plain 2 0 a

Impact of Rainfall

Although the amount of rainfall. and its reliability

are not the only factors influencing frog distribution,

the generality can be made that the persistence of

moisture at or near ground level is most important.

To predict faunal impact in South Australia it is

pertinent to examine a geographic area in which there

is a progressive increase or decrease in rainfall along

a Jatitudinal gradient. The one area in Australia in

which this requirement is met is the Northern Territory.

where there is 2 progressive reduction in annual raintall

from north to south (Fig. 1),

The Northern Territory model

Based upon the distribution patterns of the 42 species

then known from the Northern Territory plotted by

rytetet ate!

I2°S

“

=, o

t

Newcastle Waters

1Ige—

Tennant Creek

400

e Barrow ade

Creek

Alice 200

prings

i)

2a°— | 50

LS Se es

|

|

|

| |

| |

30°— 455

Fig. | Northern Terrttory showing latitudes, isohyets und major centres,

CLIMATIC CHANGE AND ITS IMPLICATIONS FOR THE AMPHIBIAN FAUNA

Number of Species

19)

1200 1000

Rainfall (mm)

Fig. 2. Number of species plotted against isohyets (r = 2750).

Degrees Latitude

40 42 44 46

TOO = 600 SpQ 400 agg 200 jaQioO

48

55

Tyler & Davies (1986) there is a distinct association

between rainfall and the number of species (Fig. 2)

within rainfall zones where there is 100-500 mm per

annum. At rainfalls higher than 500 mm the trend is

not apparent, but the geographic area invalves Arnhem

Land which has been largely inaccessible to collectors.

I anticipate that the totals for the more northern

latitudes will increase when zoological exploration

becomes possible.

Because of their dependence upon having access to

sources of moisture, the evolutionary capacity of body

size and mass of frogs are closely linked to the

reliability of sources of moisture. Species of small size

have a surface area to mass relationship that is fitted

to reliable sources of moisture. Conversely the capacity

to withstand xeric conditions requires that the surface

area from which water is lost is, by some means,

reduced. Effectively large, bulky frogs are best

equipped for xeric conditions.

Although the length of the body bears no fixed

relationship to body mass, it is the standard expression

of size (S-V = snout to vent length), In Australian

species there is marked sexual dimorphism : males are

always smaller than females (except for the eastern

Australian species Adelotus brevis), To examine any

geographic trends in size | have used a median

eft5

OE dS o =.» anttéi‘“COON,CO~C«CssS;”:C~‘C<‘®SSOSOOSSSSOOOM

50 52 54 56 58 60

Mean of Medians (S-V}

Fig. 3, Number of species occurring within particular latitudes (boundaries at single degree positions) plotted against mean

of medians of snout to vent length (S-V). Two linear regression lines have been plotted: one between 12° and 20°S (r = .948),

the second between 12° and 26°S (r = .975).

Si Moh

measurement, berg the midpoint between the size oF

the smallest adult male and Jatgest adult lemale, The

first of these analyses followed the standard 1°

latitudinal divisions (Fig. 3). The seeond involved

sunilar Iattudinal divisions but separated ut the

intermediate 0.59 divisions (Fig. 4). Essentially the

fosults dre siintlars sinall species are assoeiated with

high levels of quoisture, and in arid areas larger species

predaminate.

\ ar

t

* +h

4

as

“| Lt

Pa

—>—— — — + ——- or

\e Aa ae aa SE ah ~ a aS we

“Wns ot Mudiis CSV)

lig. 4, Nurober of species occurring within particular latitudes

(boundaries at half degree positions) plotled against medians

of snout to yer fength (S-V). Regression line r = -921

The overall lifestyle of frogs also is reflected along

these north-south transects of decreasing available

moisture. Io tecms of the size of the individuals there

is a shift in the predominance of small species to larger

ones associated With rainfall. Equally there ts a distinct

change in the overall Jifestyle of these same species,

such chat the tecrestrial/aquatie mode shifts from more

than S0% in the extreme north of the Nonhern

Terrilory lo zere at 21° latitude south

Exiting geoyraphic trends in South Australia

Within South Australia there is a lineet assocratinn

between the number (and diversity) of species and

geographic area, but without the latitudinal decrease

from the arid north to the moist south, Instead there

are pockets of species, cach of which has ity own

unique features.

For example any gradient ix destroyed by the

presence of a wet refuge like that of the Coongie Lukes

system in the northeast of this State — an intrusion

of anarea wholly atypical within a broad zone of low

and irregular rainfall, The Coongie Lakes frog fiuina

js particularly rich, but not permanent becuuse wt the

period of drought. lt has been suggested that the

northeast af the State presents a dynamic sinmation in

which there is contraction and expansion uf populations

accuTdiog to the availability of moisture and repeated

PYLER

transportation of species. from southern Queensland

Myler 990; Fig, 3),

The River Murray presents a further dimension

because at Constitutes a route for the introduction of

species from the east. Litoria peroni and Crinta

parinsivnifera emer the State by this micans, and are

confined to its vicinily,

The toadlet Preudophryne vectdenalis bos an

extensive distribution in Western Australia. whereas

in South Australia at has been found only at Vietory

Well in the Everard Ranges. in the northwest, The

existence of species within such restricted areas is ot

significance uy palacovlimatic interpretation, Because

of the small gcogruphic area to which it is confined,

the dependance of the species upou morsture. and the

existence of more arid conditions around the area,

can be inferred that since the (ime of dispersal to thut

site, it has not been more arid there than at present.

With the excepuion of Crinta riparia which is wholly

confined to the Flinders Ranges, every species mm South

Australia has a more extensive distribution beyond the

Ste boundaries. In fact the percentage of endemism

in South Australia is less than tn any other Australtin

State.

The implications of climatic change

The climate variables that influence the yougraphic

distribution of Australian species of frogs are very

poorly known. Brattstrom (1970) was {he first tu

demonstrate an dssociaion in muntane species berween

alitudinal location and toleran¢e to temperature

fluctuations. In particular he indicated that the species

accupying the cooler habitats. had the least capacily

to adapt to a change in thermal regime.

Untortunutely the climatic variables that have

parucular influence upon ithe spatial distribution of

frogs aré, at best, inferred rather than demonstrated.

For example, what is the limiting factor of the

distribution of the (ree Frog Literta rubella whieh

extends as far south as Wilpena Creck in the Flinders

Ranges?

This species. ranges across the entire northern hall

ol the continent, principally within the area of summer

rains (Fig. 5), A commensal species. it is clearly

capable of adapting to changing environments, Itcan

he assumed that a climatic change producing warmer

and moister conditions will enhance the geographic

ranye of L. rubella.

Perhaps the most significant influence of climatic

chunge will be the creation of more aquatic breeding

sites and the persistence of these sites for longer

periods. Together these factors will modify habitats in

a manner (hal is advantageous lo frogs. Hence the

Coongie Lakes ure likely lo become more persistent,

and the fauna there nore stable in a temprral sense

(han at present.

CLIMATIC CHANGE AND ITS IMPLICATIONS FOR THE AMPHIBIAN FAUNA 57

Fig. 5. Geographic distribution of Litorta rubella

Conclusion

Any climatic change that results in a moister and

warmer climate in South Australia is likely to enhance

the geographic distribution of the constituent species.

This can be assumed because of the wide range of

temperatures experienced already by South Australian

species, and the fact that a warmer and moist regime

will ameliorate the existing seasonally harsh

environmental conditions.

Acknowledgments

| would like to express my deep gratitude to the

University of Adelaide and the South Australian

Museum, far creating a research environment

supportive of my studies of the South Australian frog

fauna over the past 35 years. I am also indebted to Kelly

Fennel and Veronica Ward for preparing the figures.

References

Bratrstrom, B. H.. (1970) Thermal acclimation in Australian

amphibians. Comp. Biochem. Physiol. 35, 69-103.

DueLLMAN, W. E. (1993) Amphibian species of the world:

additions and corrections. Univ, Kansas Mus. Nat. Hist.

Spec. Publ. (21), 1-372.

Tyzer, M. J. (1990) Biogeography. Chap. 20, pp, 223-226.

In Tyler, M. J., Twidale, C, R., Davies, M. & Wells, C.

B. (Eds) “Natural History of the North East Deserts” (Royal

Society of South Australia, Adelaide).

& DAVIES, M. (1986) “Frogs of the Northern

Territory”, (Conservation Commission of the Northern

Territory, Alice Springs).

CLIMATE CHANGE AND ITS IMPLICATIONS FOR THE

TERRESTRIAL VERTEBRATE FAUNA

By PHILip STOTT

Summary

Stott, P. (1994) Climate change and its implications for the terrestrial vertebrate fauna

(1994). Trans. R. Soc. S. Aust. 118(1), 59-68, 31 May, 1994.

A limited number of studies of the biology of a variety of species of terrestrial

vertebrates is used to speculate about their responses to climate change as predicted

by global circulation models. Dramatic changes in the distribution and abundance of

animals in Australia that has already occurred consequent to European settlement 1s

noted. Speculation about the impact of climate change on the relative abundance of

mammals and reptiles, range changes of kangaroos, rabbits in arid areas, food security

of the Spinifex Hopping Mouse, competition between two species of skinks, and

disease transmission is included. Nest-site selection by tortoises and social structure

of foxes are given as two examples where behavioural plasticity demonstrates some

capacity to cope in situ with the effects of climate change, but the ability of most

terrestrial vertebrates to track rapid climate change across different substrates is

questioned. Reservations are expressed about the knowledge base upon which the

speculations are based. For very few species is a suite of studies available to compare

detailed data on distribution and abundance with climate over a wide geographical

range, backed up with biological information sufficient to explain the mechanisms by

which the species interact with their environments.

Key Words: Climate change, mammals, reptiles, Testudines, abundance, distribution,

rainfall, drought, disease, Macropus giganteus, Macropus fuliginosus, Oryctolagus

cuniculus.

Transquliods af the: Raval Saevery af S. Aust, (19941. UBT S9-t

CLIMATE CHANGE AND ITS [IMPLICATIONS FOR

THE TERRESTRIAL VERTEBRATE FAUNA

by PHILIP S'ror'y

Summary

Sore, P. (1994) Climate change and its implications for the terrestrial vertebrate: tama (1494), Trans. Ro Sac.

So Anst, WBC), 59-68, 31 May, 1994,

A limited number of studies of the biology of a variety of species of lerrestridl vertebrates is. used to speculate

AlbQut the responses.to climate change ws predicted by ylobal circulation madels, Qramatic changes in the distribution

und abundance of animals in Australia that has already oecurred consequent to European settlement ts nord

Speculation about the impact of climite change on the relative abundance of mammals and reptiles, range chames

of kangaroos, ribbits i arid areas, food security of the Spinifex Hopping Mouse, compention between wo species

of skinks, and disease transmission is includvd, Nest-site selection by tortoises and social siructure of loxes ure

piven as (Wr exatnples where behavioural plasticity demonsiraies some cupacity to cope it yily with the effects

of climate change, but the ability of thost terrestrial vertebrates to track rapid climate change uctosy different

substrates 18 questinned. Reservations arc expressed abvut the knowledge base upon which the speculations are

based. Por wry few spewies is asune of studies available to compare detailed data on distrihuden sind abundance

with climate over a wide geographical Tange, backcd up with biological information sufficient w capkan the

mechanisms by which the species imteract with their environments,

KEY WORDS: Climate change, manutiais repoles, Testudines, abundince, distribulion, rainfall, drowght, discise,

Macropus efyanteus, Macropas filiginesus, Orvetolagus cuniculis

(ntroduction

Some 6000 years ago, within the span of 100 years,

ihe vegetation at Cold Water Cave, Towa, changed from

forest to prairie (Dorule ef al. 1992). The changes were

assmcsated with atemperature rise of e 39°C, and with

such a complete change in yeyetauion structure must

ligve come.a profound change in the vertebrate faunis

of the locality. ILimay be, during the next century, that

similar dramatic changes. in biota will result from an

anthropogenically-cohanced “greenhouse” effect, The

terrestrial vertebrate fauna of Austritlit has alrcady

undergone profound changes as a consequence of

European settlement (Recher & Lim 1990), climate

change would compound the impacts of introduced

competitors and predators. destruction and

fragmentation of habitat, altered fire regimes, hunting.

ond diseases.

Realise predictions about the impact of climate

change on the distribution and abundance of terrestrial

verfebrate species require a foundation of information

from several sources. Essential are delinled forecasts

from climatologists; predictions about changes in the

distribution, composition, and productivity of plant

communities and other elements of food webs: and

Uetailed information about the present distribution.

ubundance, physiology, and ecology of vertebrates, ‘The

temporal scale of possible changes should also be

considered.

* Dep of Environmental Scicnee and Ruapeland

Munagement, Roseworthy Campus of the University of

Adelaide, South Australia 5005

The sentiments expressed in this paper are

complementary to those expressed in essays by Arnold

(1988), Busby (1988), Graetz (98%), Muin (1988), and

Possingham (1993). ‘The paper augments. previous

contributions by exploring, some examples of the

mechanisms by which climate change might affect

vertebrates, Tt isa speculative paper, the scenarios are

presented with. the intent of illustrating the level of

complexity, rather than making confident predictions

about the outcome. The predictions are weskened

because a single study of a local population over a short

period of time is not necessarily representative of the

biology of a species during all seasons and over the

whole of its range (see Kemper er al. 1987), und the

range and habitat of a species as observed to date da

hot necessarily represent accurately the reilised

efivironimental niche of the species, nur this in turn its

fundamental niche *Possinghar 1993), Ln addition, the

predictions of global climate models are not considered

10 be reliable at the regional level (Gordon ef al. 1992).

As the number of logical steps increases, sy the crrors

are summed,

The paper reviews those climatic predictions of

particular Importance to animals. uses some exumples

w explore the mechanisms by which climate change

might atfect terrestrial vertebrates, then examines the

capacity of animals to cope wilh the predicted changes,

Predicted climate changes

Evidence from chmute modely jodicates that

emissions of “greenhouse” gases into the atmosphere

60 P. STOTT

will cause global warming of 0.3 (range 0.2-0.5) °C

per decade, and ussociated changes in patterns of

Precipitation (Muughivn eal, 1992), Whilst the general

predictions of the models are broadly aceepted. the

issuc 1s comiplicuted by the lick wl clear observational

evidence of changes attributable with certainty. tdi

enhanced “greentoase” elect, On examining records

of meteorological observations, Nicholls & Livery

(1992) were hot able to identify aay clear treads in

raintall yt reliable meteorological stations in South

Australia up until W988. Although Burruws & Staples

991) note a warming trend in South Australia sinee

WS), they cautioned that the trend was “close te the

bounds of past experience’, Models are not consistent

in their predictions of seasonal changes in temperature

patierns for South Australia (see Boer e7 af, 1992; Gites

ef el, 1992)5 at the present state of refinement of the

general circulation models such uncertainties in timing,

magnitude, and regional palterns are well recoynisert

Mfoughum ev al 1992).

Rainfall has a major influence oo food supply for

serivbrate species, partioularly in arid arcas. Maarsina

et al, (1993) predict a global inerease in tropical

disiurbances, Ab present, some of these disturbances

extend into northern South Australia as tropical-

extratropical cloudbinds (TECBs). und us a result

heavy tain falls on an averuge of nine days per year

(Kulinel 1990), However, the distribution and catent

of the rainfall is erratic. Gordon eral. (4992) cautionsly

report the results of simulations which indicate changes

in daily rainfall intensity across Australia, particularly

in summer. in the form of inereases in the frequency

of heavy (12.8-25.6mm, 7, 31%; > 25.6 min. «, 95%)

nuifall days. ane. 15% decrease in the frequency of

ht (<4.4 nim) rpintall days, and a decrease in the

number of rain days. Whetton eral. (in press) review

the predictions of five glohal circulation models; Hour

predict inereasing summer rainfall over the whole of

South Australia (all five predict invreasing summer

rainfall over much of the far narth), and four predict

a decrease yt winter rainfall over must wt Seuth

Australis.

A weakness Of the models 1s that they do not take

into account all mayor Known influences an climate. The

El Nine Southern Oscillation (ENSO) phenomenon is

nor coupled with the major global circulation models

{including the CSIROY model) despite ms influenee on

variability of rainfall in Australia (Whetton ef al. in

press}. ENSO as ol’ particular importance do the

fectuitment Of many arid zone species [Austin &

Williams 198%; Gractz er a/, 198%). The influecee of

ENSO needs tn be considered m addition la changes

predicted co accompany the cnhanced greenhouse efteet

(Walker ere. 1989). The issue is further complicated

hectiuse ENS# iisell may be affected. HW increased

(O. Works ta equalise temperawres in the gaters vik

the custero aud western Pacific Oeean, ENSO would

he weakened (Rind 1991), thus tending to reduce

Climatic vanability ty custom Australie and countering

ft some extent the predictions outlined above

Direct effects

Direet efteots of climate on vertebrates should be

more readily elucidated than woald be the case with

indirect effects. Two examples are presented al ulirect

cttects of cliimte on reptiles. The first is the process

of lemperature-dependent sea determinition, and the

second is the influence of temperature on che

distribution of testudines (lurtles arid tortoises) Hotls

examples demonstrate that the Knowledge base is

inudequare even inthis more straighrhorward caregory.

Temperature-dependent ser determination

Temperdture-dependent sex determination (TDSR)

has been demonstrated im miny species of testudines

jeg Mrosovsky eral. 1984), croeodilians (c.g. Webb

eral 1983) an aganid (Charnier 1966 vite Bull 1980).

und a gekkonid (Wagner 1980). Slight (<2°C)

departures Tram pivotal incubation temperalures may

result in. entirely male or entirely female sexual

phenntypes, ever in some of the speaes which have

heteromorphic sex chromosomes (Servan eral MR)

Some species have Wo thresholds, Wath mates resulting

from intermediate and females frony extreme incubation

lemiperatures (Yntema 1976: Webb & Smith [984).

TDSD has been demonstrited to occur in an Australian

erocodtlian (Webb er af 983). an Australian

freshwater/estuarine testudine (Webb eral, 1986), and

Australian marine testuclines (Limpus efal/, (985). In

contrast. it does not aceur in several Australian

freshwaner testudines, including species found in South

Austraha (Georges 1938; Thompson [988a),

Preliminary studies suggest that TDSD oceurs in at

least four Australian agamids (G. Johnston pers

cornet).

The influence of environmental temperature on sex

ratios is a mechanism by which climute change can

directly affect the distribution of reptiles. Species with

TDSD would be more susceptible to rapid climate

change than species without TDSD. Wehh & Smith

(1984) noted variauon in sex ratios of Crovvelylus

Jehaston’ hatchlings in the field. and speculated that

it was due to the interuction hetween TDSD jincl

geographic differences in mean air temperatuces. Ln

a-warmer climate. the effect might be ta skew the scx

rao completely. blocking reproduction, dnd thus

leading 1 the local extinclion of the species, Such i

scenario has been advanced by Pieau 1182) as a

pssible reason for the extinction of many Mesovoie

reptiles

IMPLICATIONS UF CLIMATE CHANGE ON TERRESTRIAL VERTEBRATE FAUNA hy

Distribution af wesacdines

Testudines, which require external sources uf fical

for merabalic wetivily, are limited in latitadinal

distribution by remperucure. However, ean anoual

lemiperuture ds far rn crude & meusure te predict limits

tochsunibution. Length of the growing season has been

generally weepted as a limiting factor, since ab the

latitudinal lims ol distnbution adults may aot have

enough tine &) aeeunulate sufficient energy reserves

to Survive the winter (MacCulloch & Seeay 1983).

Alternatively, distribution may be Timited by the abal ty

of hatchlings to survive overwintering in the nest

(Breitenbach ef al. 1984; Congdon eral 1987). Obban

& Brooks (1987) suggest that a critical factur ts

leniperature durlog the matunition of nwa, whieh

requires (he accumulation of heat units wer sponge

surficient for successful reproduction. Csugzest another

possibility, tht distribution is lirmied by the probability

of ambient temperatures being suffiviently high during

The nesting seaseo to permit the energy expenditure

necessary in dieging the nest chamber. un acuity

recouniseil as being eneryetically demanding (Congdon

& Gallen 1984),

To predict the effect |hat-climate change would have

Ona species reqitives detailed Knowledye both of the

nature of the change in climate and of the mechanisms

by which climate change would affect a species Tn

testudines, we need ia know whith of the above

hypotheses is acceptable befure we know whether w

examine number of duys hetween threshold

romperatures. padir winter ternperalures, number uf

spring days above i tinperature threshold, or the

probability of oeeurrence m spring of spot tenipentlares

abiwe a thteshuld.

Indirect effects

Hffewts mediated by other lactors are inherently more

comples than direct effects. Aliempts af predicting

changes in animal distributions based on detiiled

analyses. of complex mechanisms. which include

consideration of matters such as physiology, popalanon

dynamics. interspecific interachuns, behaviours!

changes. und microhabitat conditions are fraught with

potential for error. Ao alternative 15 to Wentify a

smaller suite of influences which drive the system and

determine the end result, Nix (982) saw chmute as

jhe major determinant of the distribution of terrestrial

urgumisms. and sever! authors have used various

climatic indices ti) explain the diversity and abundance

of particular Australian hota. OF particular value are

those fare studies which compare detailed data on

Uistribution and abundance with climate aver a wide

veographical raige Both mechanistic and deterministic

explorarions hallow,

Relative abundance of manmials wane reptiles

And Australia already has a more diverse and

abundant reptile fauna than arid areas in North

Americn and Africa (Pisnka TY85), In parm. the

diversity and ubundatice of the reptiles ts altribuled to

the high sariability of rainfall which is a feature of the

Australian inlanl (Morton & Juines 1988), Proposed

changes in clinmte inight therefore be expected to lead

To un increase in the diversity and abundance of reptiles

relative to mammals.

Not all mammals would necessarily be adyersely

affected, For the large avid arcas of Austrulia, nel

ynnual productivity and hence (he carrying capacity

expressed as Lotal biomass of vertebrites iy related 10

wnnudl ruinfall (Burbidge & MeKenzie 1989).

However, the species compasition of the total biomass

is largely determined by the predictability and

distribution of the tintall, rather than ily amount.

Paichy rainfall favours birds. buts. and mobile large

mammals. such. as kangaroos (Burbidge & MeKensie

1989) which are physically capable of moving long

distances. to environments made favourable by recent

heavy rains. Irregular rainfall favuurs repriles, which

huve very low field metabolic rates relutive tO maninals

(Nagy J9X7) and can survive for long periods without

food (Morton & James 1988). A change in rsinfull

patterns to fewer days with rain, o lower probability

of light rainfall, and an increase in the frequency of

heavy raintall would not have a great impact on mobile

animals, bot would favour tepliles over small

Thammals. Small mummalian cellulose-dependene

herbivores would be particularly disadvamaged; they

ate vulnerable becduse their energy intake is limited

by their gut sive. This represents a stinilar proportion

of the size of the individual as in larger animals, but

the energy eapenditure for mainiainanee of body

lemiperature Must be relatively higher than for larger

owmmals which have a lower surface area i body

mass rato (Mortun 190). ‘Chey ace also limited by

their restricted mobility in ther ability ta explo a

patchy environment; and they ure most vulnerable to

competition from rabbits (Burbulge & MeKenzie 1989,

Morton 1990).

Ib could be argued what the balance bewwcen the

diversity of reptihan and mammalian speeses as at the

tine of European settlement was determined id same

prion more severe period of aridity (such periods ane

known from prefistoric tines- see Sinh 1981), and

therefore the balance would not be affected by a turther

increase in temperature and in the variability of ruintall.

Further, any mammals which might have been affected

are diready extinct as a consequence of European

settlement, The counter argument is, that effective

aridity in the future may be more extreme than in recent

evolutionary time. Climahe aridity (wherein increased

evaponition is in excess of increused ramfally inay be

compounded by “emuluted aridity” becouse of the

2 P STOTT

consequences of the removal of primary productivity

from the arid system in the torm of livestock und

livestock products (Burbidye & McKenzie 1989), and

the sequestration within the arid system of primary

productivity and nutrients in the tissues of livestock

and rabbits. Thus the resources available to native

vertebrates would be significantly diminished,

particularly during the resource “bottlenecks” of

droughts, and in drought refuges (see Morton 1990),

Range changes af kanguroos

The responses of kangaroos can perhaps be predicted

with a little more confidence than those of other

vertebrates. Many studies of Kangaroos have been

undertaken, including thorough studies of their

distributions (Fig. la) as part of the basis for managing

populalions which are commercially harvested.

Cauehley eral. (1987) have demonstrated that the

distribution of three species of kangaroos. is, in the

major part. determined by climate. Whilsc the

distributions are directly determined by land use and

the availability of food, water, and shelter, these

attributes are in turn greatly influenced by climate, The

distributions of the two grey kangaroos, the Eastern

Grey Kangaroo (Macropus piganters) and the Western

Grey Kangaroo (Af. fidiginasys), are closely associated

with the seasonality of rainfall; they overlap in areas

of uniform seasonality of rainfall, but M. g/peeus

occurs inareds where summer ramfall predominates,

and M. fidiginosus oceurs in areas Where winter rainfall

predominates, The Eastern Grey tolentes higher

seasonal lemperalures than the Western Grey provided

(a) Present distributions of grey kangaroos

(Density > 0,1 per square km.)

that there is summer rainfall, Both require a

heterogenous habits! with shelter being an important

component (see Hill 981, Calmes eral. 99h. The

distributional data have been used by Walker (1990)

to develop an integrated modelling and mapping system

which could be used fo predict and map changes in

distribution consequent (o climahic change, Caughley

eral, (\987) sugeest that climate change in the past

has influenced the distribution of mucropods: it is

therefore reasonable to use their conclusions to predict

the distribubonal responses of these (hree species ty

future olimate change.

If. as predicted (ubove), the winter caintall zone

contracts to the south and temperatures rise, the

distribution of the Western Grey Kangaroo would ils

contrac. to the south (Fig. 1b). Perhaps more

remarkable might be changes to the distribution of the

Eastern Grey Kangaroo, At present the species occupies

two small and widely separated areas in Soult

Austatia, but these are minor projections. into this State

oF u distribution whose western boundary runs along

or just to the cast of the State's castern borders

(Caughley ef a/, 1984). M. gigentens could extend a

considerable distance to the west of its present

distribution. and hence across the north of South

Australia (a G¢eupy suitable habitats in the northern

part of the present range of M. fidtinosus, WP verain

conditions are met. They are that summer rain becomes

more common in northern South Australia, that rehable

water is provided by increased frequency of heavy

rauitall and/or divestock water supplies, and Uiat habitat

hetergenenly persias m the face af climate ubange.

MM. /ulquaasus

M gigaritaus <

(b) Possible fulure distribulions of grey kangaroos

Fie. | (4) Present distributions of grey kangaroos, (b) Possible future distributions of grey kKanugutoos. Adapted trom Cire.

efal HY99), Cuughley & Grigg (UXT), Cauehley eral (1983), Coughtey et al (984), Short eral (YK4)-

IMPLICATIONS OF CLIMATE UHANGE ON TERRESTRIAL VERTEHRATE FAL SA fa

Seven af rabies dn eavid arent

Inurid Australia. the Rurmpean Rabbit Oryverohaga.s

Hmioudasy is deercusing the probability af sumival of

small percanial plants during droughts, qd having

profound eflect un the reeruiment of some species,

sufficient th Uite to.eluninare then from the landscape

(Lange & Grahiim 1983: Cooke 1987). The response

af lhe rabbic to climate chinge 4s therefore of particulay

WMperhinee

Historical micords show that rabbits mand ancas have

buen severely reduced i Numbers during past droughts

(Griffin & Friedel 1985), and may, under drought

conditivns, Hecume eatinct over lunge areas (Myers &

Parker 1975). Reevuilment is most unlikely) under

dmauwghl vondilions (King -Cel) W983), ama! prolonged

drwghis such as the '%9a8-64 und 1925-38 droughts

at Alice Springs (Griffin & Friedel JY85) test the

longevity of the species txee e.g. Dunsimore 1974).

Been soca few cabbits s0rvive in refiges, The quitlity

of the refuges is determined by therr ability to harvest

and store water and tutrients flowing from kirger areas

of the landscape (Morton 1990) such that run-on from

Hight ramiilhis sufficient to scoate seme plant growth

during the drought period (Ludwig 1987), At Witehitie.

South Australia, Cooke (982) noted that a sharp Gall

of Hide more than Simm of rain might be sufficent to

yield cun-ofr which, if concenthicd along drainuge

lines, would ensure that succulent fred in the fort of

chenopod stiruby would be available ta rabbits living

U1 warrens along those drainage lines. Ones heavy rin

falls (>20 on near Carnarvon, Western Australi,

Kinw ey u/. 1983). rabbits begin to breed, and ean

expand from the retuges to rmecalonise the hulk of the

landlscupe.

Hoavy roinfalls sre rare in the Australianand zone.

whole years nay pass withour a rainiall event =

12.5 yyy (Stattard Smith & Morton (990). Light falls

ure more common, EVIL IT has been predicted (soe above)

thatthe heavy falls would became more commen, ant

light falls samewhal less cunmmon, Hence, droughts

fre ikely to be shorter in duration, but the refuges

which sustain the residual rabbit population during

droaughis woul! be uy hutle less reliable. With a

comeident mse fn lenaperature exacerbating the severe

physininvied! stress experienced by rabbits under

present summer conditions (Hayward 1961), Incul

extinetion becomes mire likely during drouchts, but.

with decreased return times lor heavy rin, plagues

tight he expected Move frequently in thase areas where

rabbits survive

Vegetation changes consequent lo the inercased

yaciabiliy of preeipitadion might not faviur rabbits.

Rabhiis are found ii chenopod shrublands, but the

majority of the teed is provided by the short grasses

und forbs between the shrubs; the chenopods are eaten

only during draughts (Hall er ad, 1964; Crrittiy &

Predel ¥5), Under condininns nf inereased vlimatic

yariahilily, perennial plants would be favoured iver

ephoment! plants (Sta/tord Smith & Nurtin 90), so

ribbit populations ciuld be expected ta became less

dense unless palatable penennid] grasses such as

Themeda sp repliceal antuul grasses and forbs. Further

expleration.of thin scenaiia weld need to take accent

of a potential southera extension al the dernnanee af

Cymer C. grasses (Henderson ef af, 1h press), the

relative importance of Cy and C, grasses fo rabbits.

und dhe impticutions for rahbity of a change on the

seasonal distribution of rain towards surmmer rainfall

in uric) arews of South Australi

Fond security for Spinifer Hopping Mouse

Nuramys alexis. he Spinifes Hoppin Mouse. ts

widely distributed through sandy aneis ab werthern

South Australia, mainly in association with spanilex

grasses (Walls & Astin 1981), A major component of

ty diet as seed (Pindavson (940). and trom the

carbohydrate in seed it is able to denve sufficient

metabolic waler to Survive indefinitely; one fernale i

known fo have reured wv singhe young without

Supplementary water (Baverstock & Watts 975), Henee

the regularity of seed produchin would iluence

survivability of N. alexis in northern press af the State.

Seed production may be untMuenced by soul olsture,

iemiperature. und CO, levels. and two mechunisnos by

which A. aleviy night be advantaged are explored.

Soil moisture is one of dhe most iiportanl

climatically determined variubles for gnrisshinus

(Pittock (993) and hence lor species such as N. alevis

which depend in large nieugure on grasses (MayeMilliin

& Loe 1969). Walker eral (M89] anticipate that mean

sol moisture 1 likely to diminish in northern South

Australia, wlthough at present the reliability of

prxlictions is questioned (Vinaikov 1991, Piituck 1993).

More important for seed production are episodes of

higher soil moisture following heavier rainfalls, which

are predieted (above) to become more frequent. Thus

pulses of seed supply may become more frequent, and

support denver populations ol N. clean.

The rate of growih and the speed at whieh secu

development occurs following rains inay be

accelerated. Imai ef al, (985) bseryed increased sced

yield per plant for ree grown under enhancesl

precuhouse conditions. Gifford (979, 198K) predicted

thal wheat yields in areas with more strongly seasonal

rainfirit would increase as a result ol the enhanced

“areenhouse” effect, and that some grain growth woukl

become possible under conditions of aridity. which

currently preclude any yield. ‘The factor influencing

yseld was stimulation to plant growth by both increased

CO, and warmer wemperanires, which would resull in

asharter growing season such that the grat was more

likely to he Alling under a favourable soil moisture

regime. Enhanced efficiengy of water tse would also

occur due bo partial stomata closure (Chaves & Percin

hd BO STON'T

1992). Thus dronghts as perevived by NV. ulet/s inay

he yinehorated by (he supply of some seed when pone

would) be mherwise be available.

Australion artd-vene soils are generally intertile

(Morton (990), and nuvlent limitation May counter

growth-stimulaling tiechanisms Atihough seed

production may be hinted by the availability at

phesphorus, nitrogen. is less likely to be lintiting for

Cy plums, bor & Bkely corollary is that the protein

content of their seeds would be lower [see Conroy

W921. NV. alexry wuntld not be disadvantaged. as at is

tore Ukely to sufvive drought on low protein diets

which obviate the need to expend water ty dispose ab

wise nitrogen (MacMillan & Lee 1969).

Competition between Clenotus species

Muny of the 70 speeies in the seincid genus Crenaries

are associated with spinifex (Cogger 1992), und tt is

nol unusual for several species fo be syntepie,

suggesting fine niche separation between them, IP

Climate influences the quiche separation, climate change

may altegt the hilance between the speyies

Creins helenae and C, pantherinus are wo species

pecurring sympatrivally in the far nomh-cast of South

Australia, Pianka (1969) noted that they shared similar

niches. and suggested thar C. panthers would be

excluded by C. helenae bul for its reproductive

capacity. James (1991a) tound a high degree of dietary

rvetlap between the twa species, and noted that dictary

overlap in Crenorus was highest during the driest period

of tis stady, There is evidence to suggest! that the

separatim between these wo species i4 based an theil

thermal responses; Pianka (Y86) found C parthertates

to hyve a lawer tnean body temperature than

helenve, and James (1991b) speculated thag C!

pantherinas (and C. breoksi) can be active at winter

temperatures: which preelude activity bv CL frelenae

(and gthee Credots species), This permits ©

pantherinus and CC breoksi to begin reproduction

carlicr than the nther species. If activiny at different

icmperalures |s critical either for maintaining stable

fiche separation Gr for sustaining a mechanism of

oscillating disequilibrium between. the species, an

increase In temperature during winter may result in

compeliuve exclusion ol C pantherinus by C. helenae

in those areax where they are sympatric

Epidemlology

In-stable ecosystems, there is generally a significant

level oF accommodation between host populating and

disease-uausing agents, particularly if they Tuve cp-

evolved, However, wansmission of infectious disease

16 a dynamic process, and in many cases is dependent

onthe capacity of the infections agent to.survive outside

of the host, Helmunth parasites often have obligatory

larval stages which may survive for long periods un

the ground, and thus be susveptible to elimatic

influences. Should the climate change, the

accomodation between the host and the parasite may:

be disturbed.

Amongst the purisites af livestock there are examples

of species whose Lransmisstbility is known [a be

alteered by climate. The larvae of Haewonchay

comoedns, a gaéstri-intestinal perasile of sheep, require

inean temperatures > 18°C for normal development.

whereas the development of Oytertagia circumcincia

lurvae is suppressed above 15.59C. As both require

moisture, Ihe former isin organism of summer rainfall

sereas, aad the latter of wanter rarnfall areas (Suutheort

et ul, \976),

There is less Knawn about the parasites of Australian

native vertebrates, and most of the published

investigations have been taxnnomie (e.g Beveridge &

Durette-Desset 1992). Arundel e7 al, (1990) undertook

one of the few cpidentulogieal studies, which

demonstrated that helminth parasites can cause

considerable mortality in Eastern Grey Kangaroas, and

concluded that development of free-living larvae os

influenced by climate. In North America, the Morse

Alces alees can exist sympatrically with White-tailed

Deer Qdocoilens virginianus only in Chose areas where

circumstances de not favour persistence of infective

larvac of the meningeal worm Fétre/aphastrongvlus

tenuis (Gilbert 1992). Hence. climate change pay

indirectly influence the distributions of terrestrial

vertebrates through its effect on the probability of

Uisedse transmissien,

Mechanisms for coping with climate change

Possingham (1993) recugrised that there are three

Tiewns by which a species mikhl survive climate change

range change to trick shifting climate zones.

lulerance uf ihe change, and/or microevalutionury

change Examples are presented which demonstrate

thab toleranioe in flee form of behavioural plasticity may

counter climate change. but wacking appears

implausible for many smull terrestrial species

Tealercrive

The Red Fox (hilpes vulpes) has 4 complex sociul

structure which can be modified to cope with

environmental change, Zabel & Taggart (1989) have

demonstrated an effect by the E) Nino phenomenon

on the food supply of 2 population of foxcs.on Round

Island, Alaska, Increased water temperatures in the

Bering Sea were asseciated with widespread nesting

failure dn the seahird species which comprise most of

the summer diet of the foxes. Resorpiton und

preimplantation Joss are known to occur in pregnant

vixens (Ryan 1976), a common cause of which 4s

nutritional stress (see Mousiguard 1969). Hence, tf Ue

available foad was uniformly distrebuted amongst the:

foxes, total reproducrive failure in che fox population

IMPLICATIONS OF CLIMATE CHANGE ON TERRESTRIAL VERTEBRATE FAUNAS 6S

may well have oecurred, However, on Round Island,

diehiry Ghanges th smaller, less common. and tess

accessible seabird species were associated with changes

inthe social structure of the foxes, Polyeyny, the

reproductive mode priee to the dietary change. was

supplamed by monogamy. The male's help ts essential

lov cupliu dig and delivering prey tou lactating female

and ber tiller (Klenmiin 1977); thus the change in the

social SoucTUre meant that assistance provided by the

mile fox was focussed on fewer cubs yl a time when

i would haye beet more difficult for the miles tn

procure food, Individual reproductive success (in terms

of cubs reared to Sexual maturity) of che reduced

number a! breeding females was nim significantly

affected by the Ef Nifo phenomenon. Herve, a

emmporury) climate change which lead wm total

reproductive. failure of the two seabird species mnst

prominent inthe diet al the faxes did notin tum lead

10 fatal reproductive tajlure in the foxes.

The Easter Long-necked tormuse (Clelexdina

longicollis) has simpler behaviount patterns thant the

Rea Fox, but still has seme plasticity, It appears to

idjust Ts selection of nesting sites to lake account of

metcormdlogical pérameiers hkely to affect incubauon

temperatore. At Armidale, New South Wales. the

species digs nesting chambers an unshaded areas. whieh

invreases insolation and hence egg temperature, and

stiriens incubation (Parmenter 1976). The sume

species ut Rosewurthy, South Australia, digs about two-

(hinds of ity nesis in sites shaded for more than half

of the day (Stott 1987, 1988). Nest temperatures were

not recorded at either site, but mean daily temperatures

duting the incubation period are higher at Roseworthy

(39°C higher in January) and cloud vover js less

frequent (07 oktas less in January). Thompson (988b)

has demonstrated that unshaded arests of Benydare

macquarit at Barmera, Seuth Australia cart be 26°C

warmer than Shad mests, and attributed deaths in

some unshaded nests ta excessive heal. Thus it is

reasonable co speculuce that €” /orgicvllis. Tike the

spévies of testudines considered by Bull et al. (1982)

and Schwarzkopf & Brooks (1987), positions its pests

relative fo shade to ohtain optinmint suhsuilace

temperatures (or meubation.

tracking

Toirale ef af, (1992) dated late Hulocene vexctrlion

changes.at two sites in Towa whieh correspond (9 a

rate oF retredt Of prairie of 3041-600m per annum.

However, the anticipated fate of ambhropogenie ¢]i mate

chine is rourh ereater (han in the past (Possirtehaim

1993) Wrh the low relief of the inlind plains of

northern South Australia. mean thermal eradients ane

slight, aid a typical distance between isetberns

vorresponding to the predicted annual rare of

temperature change of 003°C is 2000 t Because

inany bird species of the Australian arid zone are

nomadic (Wiens 199], tricking climate change is

physically and behayiourally possible, However,

sedentaty species may have behavioural difficulties.

Few dati are uviilable on the dispersal capability of

small lerrestrial vertebrates, bul the longest dispersal

movement recorded by James (1991) for any individual

of tive species of Creanras was OStn, indiwating thal

unassisted dispersal 18 most uolikely to be able to track

climate change al the predicted rate,

Possingham (1993) poms out that comparisons of

dispersal cupability wilh (he mile oF climube charge may

be sunplistic. The comparison is useful to identily

species which ure physically incapable of tracking

climate, but cannet by itself determine competence.

There must he subsequent stages in the process ul

identifying spevies at isk, such as consideration of

physical barniers and juter-relutionships berween

species. Even if Crenorus spp. were physically capable

of tracking climate change, there 1s a clase and

presumably obligatory associmtion berween many

species.of Crenatus and Trindia and Plecerechne spp.

(spinitex grasses) and their attendant termites. These

ure primarily distributed in inlertile, sumdy sutls

(Gruetz et al. 1988), which indicates. that spinifex-

dependent species of Creawias which ure less tolenainr

of uvereased temperatures would be unable to track

climate change acnss changes in soil fertiltly and type,

For vertebrates in the north of South Australia. nacking

femperiture changes means iw generally southern

extension in range (With or without @ northern

contraction, a separate issue which would depend on

the upper linus of tolerance}, bur for partherinus

Uiere would be constraints because long distance

dispersal ar even local spread of Uriedia and

Plectrachin: seem ty require considerable ime (acobs

1482). Also, these grasses would not extend ima heavy

clay soils and limestone plains, the lauer being

venetally south of the present distribotion of ©

pantherinus

The steepness uf chinatic gradicits [i MOUNaINOUS

areas ws much prester than on plains, cand thus

allitudinal tracking of climate change ts much more

feasible than latitudinal tracking tor smal! berrestrial

vertebrates. Generally, a shonm chmb in altitude

curresponds to a majur shilt in latijude (Peters &

Darling 1985), Over a distance of abour |S kin in the

Adelaide Hills, a 500i increase in altitude is associated

with a fallin January mean maximum teriperatire of

about 53°C. and arise in mean annual caintall of about

AOO.mm, However, whilst the climate as one

component of a species’ environment may track upa

Mountain, other comipuacnats of the enyiroameit May:

he fixed; for example, substrate structure and fertility

usually change with altitude

To allow tracking requires the linking of areas

managed primarily for conservation dleng lativudinal

and altitudinal gradients (Nertan 1990). The review

oa P STOTT

oF nature reserves in south-eastern New South Wales

undertiken by Marpules & Stem (1989) confirnied thal

u single. long. narrow. tectapgular reserve aligned

along an altitudinal gradient would be the configuration

which would most parsinoniously meet the dominant

environmental requirements of teniperalure, rainfall.

and substrate fiir 2h canopy tree species which o¢cur

inthe region. Mackey ev al. QY88) note the necessity

ind amue the validity of using vegetation dura as

surrugates tor data on fauna habitat in. the present

cireumstnces of paucity of the preferred primary data.

und ailvecate a Iicus of ecological gradients in onder

le provide @ nara of safety in assessing arcas for

conservation value, However becatise of the low tehel

of much of South Austmlia, only latitudinal gradients

wre possthle in Mest ates.

Conclusion

The paper has tocused on 9 number of studies which

have relevance to the issie of climate change On

reviewing the topre, it is apparent that the voologival

hase from which changes may be predicted 15

imperfect. Reliahle infermaten on the present

distribution, abundance, population dynamics, and

interspecitre relationships of Australian vertebrates 1

lume (Norton 1990), but there 15 sufficient

Information to indicate that climatic influences on the

distribution of many animals Operate through

Mechumsins which are subtle and’ us yet priorly

understood, and salfigient informution to wurrnint i

conclusion that climate change of any mugnituile ts

quite likely (4 affect the distributions of many species

of terrestrial vertebrates,

Conclusiins about the fate of individual spectes an:

ut presem speculative. Deterministic studies such 25

those undertaken on Kangaroo populations ure less

speculative than mechanistic studies because of the

complexity of the means by which climate influences

the biula, dul mevhanistic ckaminutions sre

complementary in thit they muy reves critical aspects

of detail not apparent to deterministic considerations,

Acknowledgments

Thanks are extendeal to Susana Carthew why) provided

constructive eviticisnt of the manuseripy ta Megan

Lewis who prepared lor presentation and publication

the fizures on the distriburion of kangaroos, and to

Kathleen Williams who cared tor the temple of the soul

References

Apsolt, GW (1988) Possible efteers of clinate change-on

wildlife in Western Australia, pp. 375-386 fy

Pearman, G. 1. (Bd.) "Greenhouse; Manning for climate

change” (CSIRO, Methoume)

Agunpnt, H., Dempsrex, Ke. Harnigan, K.E. & Bua,

RB. (1990) Epiemiologival observations on the helminth

parasites of Macrapuy givantens Shaw in Victoria. dust.

Wildl. Res. 17, 39-51,

Avstis, M oP & Wipiams O B (98%) [ifluenee ot

Climate and community composition on the population

UJemogruphy of pasture species in semiatid Austtalia

Koetatia 77, 43-49,

Raverstice, PR. & Waris, CHE (1975) Water balance of

small Jactsting rodents 1, dd fibinam water intikes and effects

of waler restriction on growth of young. Comp. Biochent.

Physiol, SOA, 819-825.

Beverinor, | & Dilrerre-Disser, M C (1992) A new

species of trichostronpyloid neraatade, Orlilia bulrae trom

native rodent, Rares fuseipes (Waterhouse). Trens. Re

Sov S Avant We, 124-128.

Bork, G. J... MePARLANE. No A, & Lazake. M. (1992)

Cireenkouse gas-induced climate change simulated with the

CCC second-generstion General Circulatiyn Model. J.

Climate §, 45-1077,

Berrringacn, G. L., Concpos, J.D, & Skis R. © (1484)

Winter temperatures of Chrvwempy pictd nests in Mictizun

Effects on tatehtine survival. Herpetalopica a0, 7681.

But). } J. 1980) Ses, determinunon in reptiles, 2 Ret Biol

55, 3-2)

—. Your, RB. ©. & MeCoy. C. J. (982) Sex

determining temperniures an rurtles: A geographic

companisen Fvalatian 36, 326-342.

Bike, AOA & MeKeere, NL (1989) Patterns if ihe

modern decline of Werlern Australias yertebrite lana:

Causes and conservation tmpheutiony, Mel. Gunn, 30.

194-19%

Birmkows. KOR. & Srarnns, Mt. CQU91) "Treads in Sour

Australia omuximum temperature regards S.A

Department of Environment and Planing, Adeliides

Busty, bh oR (WO) Brocline a btoelimanc analysis: ane

prediction system. Plant Protect. Quart. 6, 8-9.

Cares, & C, Popne, A, R.& Gricc, G, C, (1991) Densits

distributions and habilat associations oF Red Kangaroos.

Macrapus rufies, aid Western Grey Kangaroos, Macropies

filiginenus. 10 the South Australizn pastoral zone. Avst.

Wildl, Kes, WB, 377-402

CAUGHLEY..G., GriGG, GC. & SHoxr. J. (1983) How mary

kangaroos? Search (Sydney) 14, 151-152,

__. Brown, B,, Dosrine, Po & Grier, Be (1984) ‘Phe

grey kungarou overlap zone. Aust. Wildl, Rey. UL, 1-10

ss Strort, J. Gait, G. Co & Nex, A. (1987)

Kangaroos und climate: An analysis of distribution, J Anim.

Feul. 56, 751-761.

Crarming, M. (1966) Action de li temperature sur Ju sex

ralie chez Jémbryon dAyama age (Agamidiic,

Lacertilien). Sec. Biol. de Lavesr Afrieait 160, 620-622.

CHAVES, M, M, & Pureina, J, S. (1992) Water stress, COs.

and climate change, 4. exp, Bar, 43, 13) 139

CoceerR. Ho G, (1992) “Reptiles and amphitians af

Australia™ Ord Edin (Reed Books, Sydiiey).

Congpon, J DO & Garrex, ROB O89) Movements and

cuergetics of nesting Cirvemvy picte, Herpetulowca 45,

94-100,

___ Brerrenuacn, G, 1,., Sues, R,C, & Tinkrn, DB, We,

(987) Reprodugion and vesting eccolagy ol Snapping

Turtles: (Chelvdra serpentina) in Southeastern Michigan.

Mid. 43, 39-54,

Camroy, J, (1992) Influence of elevated atmexpheric COs

coneentrarions on plat nutrition, dust J Bor, Ah 445-456.

Cookr, BT) (1982) A shoriuge of water in naturel pastires

aso (hero Limiting at population of rabbis, Oryetclagur

cunteulas (Lod, inaritl, north-eastern South Australia, waz

Wildl, Kes, 9, 465 476

IMPLICATIONS OF CLIMATL CHANGE ON TERRESTRIAL VERTEBRATE FALNA 67

(1987) The etlects of rabbit grazing on regeneration

ul sheoaks, Alfocusuering verticiiald qos sullwater letrees,

Molaliiiva hulinatuaroram, io the Courong Naripnal Park.

Sout) Australia. aAnst J) Geol 23, 1-20.

Dorsip TA Gowanus tA Reacas, MK, Plone

D, A.. Murrent M, To & Baker. R.G. (1992) .4 hieh

resoldtion record of Holocene climate charge in speleothem

cqilode trom Dol) Warer Cave, norheast lowa, Senvies 258,

1626-1030,

BiMsMGrRE, § S. (974) Phe npbbil in subalpine soutl)-eastern

Australia L, Population structure and productreily, Awsr.

Wildl Rew We 16.

Fincaysuy. HH, (1940) On the central Ausrralian maramals

The Murilie Tae Ro San, S. da, 64, 125-136

Gares. W, L,. Mivenitt, J. F, Bork, G. J, Cupasen,

& Mitesukey VP 11992) Climate modelling, climace

revivetion antl muted validation, pp. 97-134 fa oughta,

JT, Callander, BOA. & Varney, §. K. (Eds) SClintate

change 497: The sipplementary report ty the [POC

Scientific Assessment!) (Cambridge University Press,

Cambridge)

Grokces, 4, WYRE) Sey determination is independent 4]

ineubation temperature tn another Chelid cunte, Qrelodina

longienlis. Capeia W988. 248-254.

Gieynep, R, M, QU78) Growth and yeld uf CO,-cnriched

wheat tuder weter limited conditions, dior J. Planar

Physniit, 6, 367-378

(1988) Divece effects of higher carbon dioxide

Lanverntrutions an vegetation pp. 506-519 fy Poarmin. th.

I. (Ed) “Greenhouse: Planning for climate chunge”

(OSTIROL Methane).

Giraven. FE 11992) Retroductive logic and the oflects af

meningeal wens. A comment J Wild! Munaye, St.

61-016,

Goroox, H. B Wretron, PH Prevocs.. A.B. Fowirs,

AM, & Hayteek, MR 942) Spmulsted vhanges in

daily mrnfall imensily due ta the enhanced erecnfovse

effect) Implications tor extreme rainfall events. Chumaze

Dynamos 8, 83-102.

Grammy. ROD. Wacker, ROH & Waker, PAL (IAS)

The vonstynences of cliniate change far seventy preteen

ef Australia. pp. 399-43) fi Peanian, Li l. (Ed.)

“Greenhouse, Planning for climate change’? (CSIRO,

Melbourne)

Ores, GO. & FRIEDEL, M. Hy (YS) Diseontinueus

change in central Australia: Some implications ef major

eeulyyieul events fir land manugement J And Evi

63-81),

Hasusya. Ro, Minieni. dF o& Sinton, © A, 1994)

Tropical disturbunves in 2 GCM. Climate Dyn. 8, 247-257,

Hau. EB AL. Seecar Rob & Barney, C M, 11964)

Repeneration of the vegetation on Kaunamore Vegctinn

Reserve, 1926-1962. Anat Bot, 12, 205-264.

Maywarn. dS. (961) The abiliey of the tablay at survive

condinons of water resiriction, CSTKO Wildl, Res. 6,

160) 179,

FpGERSON, 5, HATTERSLEY, PF YON -CatmMbker, 5. &

Osmonp, B, (ia press) Are C 4 pathway plants threatened

Ks wlobal chimate change? Jn Ecophystolepy of

raed 1) Ecological studies, Reielbers_ Springer

erlag. Cited in PITTOCK, AB. (1993). A cliinate change

Perspective on wrasshinds Prac ARM IA Croessl Colgr

W953

Hilt. G. J. (1981) Distribution of the grey kangaroos in

South, inland Queensland. Aine. Reowge. . 3, 3R-66.

Hovoirron, 1 7), CAntanper, B, A. & Varnxesy, 5, K,

(1992) “Climate change (992) The supplenentiry fepon

tothe IPCC Scientific Assessment’. (Cambridge University

Press. Cambridge),

IMAL. K.. CoLEMaAN. DF & Yanacisawa..T, (985) Inerense

in atmospheric paral pressure of carbo dioxide and

cowth arid Viel af ice (Arya yutive Ly, Jap. Crepe

cl, 54, 443-48.

Javoss, §. W. (982) Relutionships. distribution ane

evolution of THadia and Plecrachne (Graminsc). pp

287-290 Jy Barker, W. Ro d& Greenslade, PL cds)

“Evolution of the flora and fauna ofarid Australia’, (Peacock,

Publishers, Adelaide).

(amps, C1 (99Ia) Temparal varianidn in diets and trophic

partitioning by coexisting lizards (Crenortes: Scyneitae) i

central Australia, Gecofoete 85, 553-56]

—____ 11991b) Annual variation in reproductive cycles of

seincil lizards (Crenarus) mi ventral Australia Copeia 1991

7A4 760,

_____ (199ic) Population dynamics, demography, jn Hie

history of sympatric sctncd lizards (Crenoris) in central

Australiu. Merpetolegicn 47, 194-210

Kewerk, © M., Riveyespr, 1 J, Howey, Wo FE.

How, R. A., Beaorey, A oo & ScrMrry, L, H. (19R7)

The demography and physivlogy of Mefornys sp. (Rodentia:

Muridavt in the Mitchell Plateau urea. Kimberley. Wester

Australia, J. Zool. Land, 212, 533-362

King, D. R., Watnter, 5. H. & Scum, GL. 985)

Population Huctualions and reproduction of rabbits ina

pastoral oreaon the coast north of Carnaryon, WA, Aust.

Wildl. Res, 10, 97-104

Kieiman, D. G. (1977) Momogumy in manimals. @ Rev.

Biol, 52, 39-69,

Kunnen, § (990) Tropical-extratropical cluudbands mm the

Australian region. dar J) Climar 10, 341-364

Lance. R. T. & Guanam., C. RB. (1983) Rabbits and the

failure of regeneration in Australian arid zone Acacia, Aust.

J, Feol, 8, ¥77-381

limeus, C. 4, Rive, PoC, & Minuuk, | D, (19851

Temperature dependent sex determmation in Qucenstund

sed turtles: Intraspecific variition in Curetrd ccrene. pp

SMN-JAI In Grige, G. OC oShime. RO & Ehmoan, H- (Dds)

"Biology of Austrian Frogs and Reptiles’, [Surrey Beatty

& Sons, Sydney)

Lipwie, J. A. (987) Primary productivity in arid tunds

Myths and realities. deid Envir LR 17.

Mac€ritocn, R, D, & Seoov, DM, (1983) Demography,

growth, and food of Western Painted Turtles, Clirysevnys

prota belli’ (Gray), Trom southern Saskatchewan. Coat of

Zool 61, 1499-1509

Mackey, BoG., Nik, Ho A, HUTCHINSON, MF,

MacMaAuun, 2) & Puemiyu, PM, (1988) Assessing

representiliveness ef places Ter eonservanion reservar

und heritage listing. Environ. Manage 1, 501-514

MACMI ASR, EB & Len, & K. (989) Water metabolism

of Australian hopping mive, Comp. Biochem, Phesiol. 28,

493-514

Main, A. R- 11988) Climate efumps and its impact on nature

conservation in Austoilia pp. 361-374 dn Pear, GL

(Ed.) “Greenhouser Phinning for climate vhange” (CSTRO.

Melbourne),

Maraiies, C R & Srem, J. L. (89) Patrerns in the

distributions of spevies and the selection of nalire feserves:

An csample from Exevlypnas forests in soulh-cantern New

South Wales. Bint, Conver, 50, 2)/9-238

Moron, S. R. and Jases, C.D. (1988) The diversity ans

abundance of heards ingrid Australia: A reappraisal. Aer

Nat (32, 237-256.

(1990) The impact uf European setlerent oo te

sertebrate animals of aad Australia. A conceptual model

Proc. Ecol. See, Aust. 16, 201-213.

Muovsicaarp, J. (1969) Nutritive influences upon

reproduction, pp, 489 516 /n Cole, HH, & Cupps, FT,

(Bds) (1469) “Reproduction m odemesds anievals’,

(Aviademic Press. New York).

om POSIorr

Mrosevsky N | Derran. POH & Worse © PR Ys)

Sex natios of two species of sea turtle nesting in Suriname,

Can Jt Zewl #2, 2227 2239,

Mynrs, K. & PARseR, B.S, 11975) A study of the bielogy

of the wild rabbit in climaticully different regions in eustem

Australia VE Changes in numbers and distribution related

octiniate anc land systems in semiarid north-wesjem Now

South Wales, din Wildl, Rew 2, Uae

Naw. Ko A 987) Wield metabole rune und find

Teywirement scaling in entnals and binds. Leal, Meniase,

57, \\l-12K

Nreuona, No & Lavery, B99) Australian raurfall rrends

during (he bwontierh century. da. dc. Climatal, Bh. 183-163.

Sty. BM, A (1982) Ravinonmental Uererinivants of

Hinveugraphy amd evetution un Terre Austrilis, Ja Barker,

WR & Greenstile. Pod (Eds) “Evolution of the fora

and fauna of arid Australia” (Peacock Press, Aduluide,

Nurion. T. (1990) Mainline Biolagicdl diversity pp,

W753 fy Nocloo, Td Withatis, OF (bas) “Clinitte change

in Australias knvironmental, sograeconcmic and polticad

comsideratioas” (Centre for Resource wm Environmental

Studies, A.NUU., Canbern).

Ornnary, M. E. & Brooks, R, J, (IY87) Presticthan of the

unmet of the anual nesting season of the Camimon Snapping

Turtle. Chelvdra serpernina Herperilagitu AS, 324-328.

PaRMENTER, CJ (1976) The vatural Wrstary of the Ausra

freshwater turtle, Chelodina hamiculliy Shaw (Testydinata,

Chelidac) PhD. thesis. University of Now /nuland

Petes. RL. & Darcing, I BD. (985) ‘The greenhouse

effect and nuture reserves. Brasctence 335, 707-717.

Pranka, E.R. (1969) Sympatry of desert lizards (Chenetis!

in Westera Australia, Aeoloey $0, 1012-1030,

(1985) Some intercontinental comparisons al desert

lizards, Net, Geogr Res. 1, 490-504,

Pipat. C1982) Modalities of the action af temperarure on

sexual differentivtion in ficll-develuping embryos af the

Kuropeah Pond Turtle, Anis orbicuiaris (Emydide)./,

wap, foal, 220, 353-360,

Prrvoes, Av B. (1993) A climae chimge persective an

yetasadands, Prov NVM tne Grassl Corer Un press)

Pooir, WOR. (960) Breeding of the wild raobit,

Orvetolagus cuniculus (Lod, He rekition to (he ent ronment.

CSIRO Wiel. Res, 5 21-43,

Pessina, Ho P1993) Linpact of elevated aimuspheric

Cth Ui biodiversity: Mechanistic population-dynamne

perspeclivy dey J Bal ate Zt

Rrenek: WF a bam, L, (990) A review of current ideas

mothe extinetuen. gonservalion and ymunagement af

Australias terrestrial Acrtebrate fauna, Prov, Eenl Sen

Ait 16, 28730),

Ris. D. (99h) Chitnate viriability and climate change. pp.

6978 Je Schlesinger, ME (Bel), “Lireenhouse-sas-

imbiced climatic change: A entticul apprusal of sinulations

wel observations’ (Elsevier, New ‘ork.

Rass, OE 1976) Observations an the reproduction and age

structure of the fox, Males velpes Lo. in New South Wales,

Aust Wildl Rex, 3. 1-20,

SeuWwarekopr, 1. & Brooks, KJ (O87) Nest-sire selcenon

und offspring sex cativ in Painted Tartles, Chieyeernys pret

Crpeiu 1987, 53-6), ~

Srevan, b. Zanorser Po Woriazi, My & Pieau, CL (He)

Vermale biased sex rata in udutts cat phe nerthe Apeys

arhicilins al the nocthenn lirit of its distribution m Prince:

4“ probable Consequence of the interaction nf berporature

wih genotypic sex determination Can J Zon 67.

J2774. SF

Sina, G, (198), Late Quaternary pollen records and

seasonal palacoclimates Of Lake Frome, South Austratia.

Hdrohialopia 82, 49-430),

Starvorh ‘Smirn, DO M. & Moros, So oR. (1990) 4

tramework for the ecology ol arid Australia, J, Arid Kuvir

18, 255-278.

Srovi, PB (987) ‘Verresirial movements of the freshwater

lortose Cheluding longicollis Shaw us nioniioved with 4

spool tracking device. 4duxr Wildl, Res. $4, 559-567.

(988) Terrestrial movements of the freshwater

tortoise Chelodina longicollis, MSc, thesis, University of

Adelaide.

Souracor, WOH. Mafor G. W. & BARGER, 1A. (197A

Seasonal pisture contiminution and avanlability of

nemutodes lor grazing sheep, uss, J, aeste, Rex, 27,

277-286

THomeson, M. B (1988) titiuenee ol jueubation

empcralure and Wwaler potenti! on sex determintign tn

AKinydurd tee dain | Testudifes: Pleurodia)

Herpetatagicn 44, 66-9)

_____ (988h) Nest temperatures in the Plearodirun turtle.

Finydura macquearti: Copeia WS8, 996 1000.

Vinsikov, KY (1991) Soil nicosture Ganlent with global

warming. pp. 261-268 Ja Schlesinger, M, E. (Edu

“Greenhouse-gas-mdival climate chaiger A vyitical

appraisal of simulabous anc observations” CElsewer

Arnsterdum |

Wausrk. E, (i980) Temperature dependent sex

detecimiiionina gekke liza Appendis (p.2t) ta Bull,

1.3 (ISRO) Sex deverminstion in reptiles Q Rev, Brat, 55,

Wacker, Bo H., Yolwu, M.D, Parstow, J. $2. Cocks,

D,, Views. PM,, Mawspins, COR. & LaAnbsnikd, 3

J. 1989) Global climate change and Atotralias Bfleets an

renewable mitural resources. pp. 33-05 In Global climate

change assures tor Ausinibu. Papers presented al (he titst

Meeting of the Prime Ministers Science Council”

(Commonwealth af Austritia, Canherpa)

Watkiok PA (990) Madelling wildlife distributions using

4 geographic information system: Kungarnos in relation #4

climate. 2. Blogeoy, 17, 279-289,

Wars, C. (970) The foods eaten by same Australi

desert fodetts. 8. aust, Not, 44, 7-74

Jk Asti. HL (1981) “Phe rodents of Australia’

(Angus & Robertson, Sydney),

Wees, G, J, BuekWoRTH, Ro A Manoir, SC. (Ysa)

Craceetyhes jeinsront in the MeKonlay River, N-T, V1

Nesting hiolegy. dust Wildl Rey. 10, 607-637.

Satin, ALM. Y 84) Sey catio and survivorship

io the Austtulign (reshwaertevaxtile Crreod Wrs poleont

Syhyp. cael, Suc. Land. 52, 319-455,

CRoOOUENOT, DB. & WHITERBAn, PJ (19860) Nests,

cegs. und embryoni¢g development of Caretechelts

jnseutpta (Chelona: Curettochelidae) from northern

Australia, 2. Zoo/,, Lene. 1B) 1, S135

Wiierton. POHL, Fowirrr, A. Mo, Mayruck, My oR &

Pripork A. RL Cin press) Implications if vlinwte ehange

duc ta tte enhiinced greedhouse eect on Moods and

draughts in Australia, Climate Dwiinnies

Winns. 1A. (991) Eenlogionl similarity of sheub-desert

avilumes of Australia and Nonth Aimurica, Bealeey 72,

ATO-495,

Yrvems, © L (976) Elfeeu ot qenbanve tenipeniure on

zexual differentiation jn the jurtle. Chefwira seenenina.

J) Morph. i501, 453-462.

Magee. Co. & Vaoearr, S.J. (980) Shift im Red Fos,

Paipes vulpes. viiuing system assacialed with EN tine int

the Bering Sea. Anim. Behe, 38, 840 838.

SOME POSSIBLE EFFECTS OF CLIMATE CHANGE

ON VEGETATION

By ROBERT BOARDMAN*

Summary

Boardman, R. T. (1994) Some possible effects of climate change on vegetation.

Trans. R. Soc. S. Aust. 118(1), 69-81, 31 May, 1994.

One of the most difficult tasks for managers of land which grows long-lived plant

Species, (species that dominate ecosystems), is how to assess that change in

environmental conditions is actually of significance to the present ecosystems, and

what to do to ensure that the ecosystem, rather than the species themselves can be

maintained. Practical things to do which will make adaptation possible are limited by

the indefinite nature of time frames. Areas of the State north of Kangaroo Island

extending to the northern State boundary, are the parts most sensitive to changes in

vegetation. It is likely that the most strongly threatened species in the long-term are

the perennial species which dominate woodland formations. Low sclerophyll

woodland and mallee formations appear to be particularly at risk.

Key Words: arid zone, vegetation, climate change impact, Goyder’s line, Eucalyptus.

Transactions of the Royal Soviety if 8. Ast, (W994), UBC). 6Y-BL,

SOME POSSIBLE EFFECTS OF CLIMATE CHANGE ON VEGETATION

hy ROBER! BOARDMAN*

Summary

Busknuman, R. T. (984) Some possible effeets of cllivwte change On Vegetation, Prary. AL Se Suse DESC),

69-81, 3] Miiy, 194

One of the most difficult tasks for managers of land which geiws long-lived plant species, (species that dominate

ecosystems), iy how lo assess that charige in enviranimental conditions is actually af signifeinee 10 the presen

ecosystems, and What te do to ensure that the ceasystenn mither than the species themselves can be msintained,

Practical things to do which will make adaptation passible are limited by the indefinite nature of time frames.

Areas of the State north of Kangaroo Istand extending to the northenm Stite boundary. are the pxirts most sensitive

10 changes in vegetation, It is likely that the most stroagly threatened species im the long-term are the perennial

apovics which donate woodland formations. Low scletuphyll woodland and mallee formations appear to be

particalarly at risk. These formations occur m the zone with an average growing season, in the best possible

conditions. of fron 3,5 to 5.5 months, A large proportion of species in the seni arid and arid zones are trees

und shrubs which afe classed as “sensitive” by dendroclimatologists and dendrochronologists, The first visuil

changes to Vegetation will rosult From changes int physiological processes that are mediated (hrouph plait metabolisn.

Physiological processes of plants and microflora are more sensitive to tempeniture than ecological processes,

but this sensitivily to Wimperature change will be observed through the symptoms which will be ceslagical in

natue and reflect the impacts through interactions within and between species. It should be possible to use

dutecotogical information to help track the pathways ta new gynecological combinations of species: There is a

need and an opportunity to identify “indicator species” und particular features for Measurement. Alternative

approaches to the question of minimum water vequiccinent for a woodland without "gaps", und the extent af foliage

cover in the overnwood and whole system, are reported. The optimum cover in the overwood of 3245 % should

be udequate.to sustain low woodland on run-on sites in the 200-300 mim riinfall zone of 8.A. The evulogical

significance of Goyder’s line and its putential usefilness ave discussed. Features which permit adaplabon maliude

planting botjnigally closely-related land-races (pravenunces) und spovies, in five disnnct groups of Eucalypnes

Phat hive been matched ta three climatic mudels. The avn has been te provide a nest-egy of genes. allowine

adaptation lo n¢eur whilst retaining ceosystent structures ussociated wilh particular hind-user. Provision of gene

banks and corridors of fand able fo permil native species to niigtate, are still seen to be the best option in the iniecim

Key Worps: arid zone, vegetation. climate change impact, Goyder's line, Bucalypyits.

Introduction Climate change svenarios for the Australian region

(Climate Impact Group 1992) are now less definite than

South Austrilien vegetation is strongly Sssocisted

wilh climates. which are dominated by coo} wet winters

und waren to hot dry summers, Rainfall vod potential

evapramspiratin occur at levels which iinpart a range

from sub-humid to ard climate types, The vegeuytion

has heen classified by ils form (Specht 1975; Boonisma

& Lewis, 1976; Atlas of South Australia 1986). These.

forms shew marked correlation with isohyets and lines

of equal potential evaporation.

A large proportion of species in the semi-arid and

arid wanes ure trees and shrubs which are classed

as “sensitive” by dendroclimatologists and

dendrochronologists (Fritts 1976). Tree rings represent

the end point of the long-lived perennial plant's

ailocation of carbohydrate resources produced by

photosynthesis. This link emphasises the reliance of

semiarid and und zone vegetation on a tolerable

infrequency of favourable growmg seasons when

conditions enable species to miniain their place in

ecosystems of the region.

* Department of Primary Industries, SA, Forestry, 135

Waymonil) Strect, Adelaide, South Australia 3000

the original single emissions scenario (IPCC 1990)

fram which some of us worked in 1988 (Greenwood

& Boutdiman 1989; Boardman 1989). Global warming

models are less simplistic and now include a tange of

gases, responses in cloudiness. absorption of heat by

the oceans and a degree of climate “Sensitivity”, not

included previously. The CSIRO GCM madel has

provided plausible ranges of local temperature warming

and changes in rainlall per degree of global warming

civerall.

Sub-regions have heen added in the raintall change

scenarios which, so far as S.A. vegetation is

concemed, largely reflegt earlier findings. Arcas of the

State north of Kangaroo Island are placed in sub-region

A by Climate Change Group (1992) and this extends

most of the way to the northern heundary. The

reduction in Winter rain. despite a trend towands &

predter proportion of annual rainfall in summer

together with @ marked risé in potential

cvapotransprration has been modelled, and leads tu

greater droughliness and diminution of growing season

desenhed far this patt of (he Stale (Appendix 1)

(Boardrum 989, 1992),

™ K. BOARDMAN

Arcus of the State south of Fleuneu Peninsula are

placed in sub-region Bi this is a sub-region unlikely

la suffer murked change in gross ramtall and may not

even see a decrease in winter rainfall This fits in with

earlier progniises already deduced for this part al the

State, Linpacts on vegeuttion seem likely to be mininniil

and pissibly will see amore favourable sel of growing

conditions, Dendrometrical data on native forest m the

Lower Sruth-ease (Ruiter 1964) rndicate that the native

Iree species ire better adapted to a summer rainfall

climate than the Mediterranean type which his

persisted since the lust glacial cpoehs. That is, most

of their evolutionary adaplation luok place ina clinute

sunilar ta that now present in the castern part of NSW

(Boardinan 1986),

The impact of ENSO (El Nifto — Southern

Oscillation) Has not been mcluded in GEM models,

This uppears only to affect sub-region A of the Stabe

which Is bisected by it (Allen 1988), The boundary

between ENSO affected areas and those damunuted by

the Antarctic polar circulation systems lies along a NW-

SE direction which crosses the northern Flinders

Ranges. The “settled areas” of the State SW of the

boundary zone are not directly affected by ENSO, and

only indirectly influenced by the alternate system,

called La Nina,

[tis the time-scale of changes rather than their degree

whuch i different, compared with the early niodel. aad

it hay heen extended, A range of ternpenture change

is nOW given rather than an averuge. However the

median temperature change for the 2070 scenarios is

similar to that used in 188. and these will be fised

specifically here, Physiological processes of plants and

nucnoflor are more sensilive in temperature dhan

ecological processes, bul this Sensitivity th Len \perabure

change will be observed through the symptoms Which

will be ecological in nature and reflect ahe impacts

(through anteractions within and between species. They

will be the first changes to attract dtention We expect

to find features which can be meusured and which will

indicate the nature of changes taking place,

What is the nature of changes

that may be expected?

The sequence of reviews adopted by ihe organisers

ol this meeting imples that the most ebviows efieers

wf! climate change indiested tor South Australia will

he noticed in seals (which [think a unlikely) and in

arens We recognise as wetlands: namely, low coastlines

and coustal swamps, swales between sand-dunes, lakes

and lunettes, Most indicytions of the impact uf seasonal

changes in rainfall distribution, both in quantity and

intensity, and a Warmer elinnte incheate that it will be

topographically low-lying areas which will be affected

early during the process of change It seems reasonable

lo suggest (hat it will be the vegetation in areas (0st

sensitive to direct and intense climate change which

shows the visible signs first, rather than the animals

and (he soils. However, if may be lempting to sugeest

thot Wt will be the microflera and microfkunu which

will be must sensitive to climate change until is 1 1s

redlised that both dre more closely linked to the

dominant heher-plant specres uf an ecosystem anal the

weather, than to climate (Howard 1967: Lewin, 1985:

Simpson 1967).

It appears likely that the first yisual changes bs

vegetation will result from changes in physiological

pricesses that are mediated through plant metabolism.

In this respect, the tallest elements jf ecosystems.

especially Irees, are likely to shiw the effects {irst,

Trees are more likely to suffer climate change effects

than Shrubs, the result of thes ability wy pinduce a boke,

or trunk. As the bole lengthens and its diameter

becumes banger a grealer amount of usste feeds

sustenance which is unnecessary forlesser planis.

sheath of new Ussuc m added cach growing scasou

comprising extra phloem, or bust, bark and meristem

tissue, the cambiuin. Ibis a cast dor Keeping the tree

in a lofty posibea in the ceasystem. There ts no

difficulty with maintaining this structure in a stable

climate: allocation of resources is attuned 10 the

situation: wood erowih is laid down ita predictable

fashion including well into old age Waring (1988) and

Franklin er al. (988) have discussed the natute of

changes which may result in tree death, There ts 4

complex series of allernale pathways which lead ko a

single result. The fest of these is a reduction in food

substanee uvailable for new wood productin,

Different environmental stresses alleci the majur

components of the resource-hudgel allocation. There

is a hierarchy far normal allocation of carbohydrate

with a Ifge, Stem growih only occurs unee the resource

demands of new leaves, new roors and an intrinsically-

reguluted reserve ulloewtion into slurage (“reeavery

insunince”) has been agcommedated. Hence, the time

when physiological state (s most suitable for assessment

14 just before a Mush in new leaves becomes wppurent,

Flowers ure not produced annually antl when they de

depends on the amount of stored resources

aocunwulated. At this time carbohydrate and nutrient

reserves in (wigs, alder foliage, large dummeter routs

and the stem self are at their greatest, The urbiter

ofthis is the amount uf new woud produced per unit

of foliage

Fritts (976) bas prodiced model dixgrinis whiet

illustrate the factors which interact und lead by

formation of u narrow growth ring in the trunk. His

mudel has bye parts which can be related to the S.A,

situation, currently and as it may develop over the next

century. Pant A of his model contains relationships

associated with low peecipiration and high temperature

during the grawing seuscn that lead ta the foeniatin

Of a narrow ving in trees on Ury sues. Pin Bois a

SOME POSSIBLE EFFECTS OF CLIMATR CHANGE ON VEGETATIUN 7

possible precursory extension to plirt A. times when

low precipilation and high lemperiliire yecur rir hs

the growing seasert and auld wo the uinount al stress.

The complexity of these reactions is indicated by the

iwo parts which each have 24 possible stages. When

both paris nieer they aci additively and dimecthy on the

cumbiom, bul only combine tor the last three stages

Combination of parts A and Boas not an uncanimon

situution in SA, when dry winter presedes o draughty

growing season, The net result is reduced rates yf ect

division jn the cambium and so fewer wend cells

become differentiated, uy a conseyuence ol law

praluction of growth regulatory, substances (plant

hormones} and lood resource. Severa! other oomurwi

factors affect the cambiumn when a qarrow ring is

formed including years of intense flower and seed

production (mast years) and uttack by pests, mumly

deloliators bul also including bark beetles and sapwood

borers. Dendruchranologists have artempted ta idenmly

the range of causes when a narrow ring is formed and

some models exist for these sittations (Pritts 1976),

Palucoccolggical plan studi¢s have shown cycles of

change have alreauly been expenenced by the vetiera

und species stil] present in Australia. Species may be

adapted ts the seemirio io different ways or may be

tolerant in different degrees. A Sustained trend away

fron) normality will have an effect propertional te any

dormim evolutionary adaptation ulready incorporated

into the penes. We have rio ready way of knowing what

these mivhe be in specific cases, This position 15

agenwvatal by the appurent fag af amny centurles before

consequen| veewtatiin changes vecurred if ihe pust.

The current climate change seenurios preclude this

significant lag period apparent in the past. Ef 4 trend

in the mvc of change in climate becomes apparent ther

the siress impacts are likely ta accumulates tm witys uot

seen before. The impact iy then likely to be sunilar

to one of a range of possible impacts such ds dlustrated.

hy Barkers (1999) looking at different recurrent peyiods

herween fires in Aeacta woodland,

Periods of extended drought are no} particularly

hannful if trees have well-established mot system and

foliage canomes, partly because drought is reckoned

in relative derms 10 the normal climate. Where there

has bees natural selection by drought, trees mirely die

from draught, In the short term, adaptatvon may help

Telain reserves by temporury reduced demund, Trees,

such as those in S.A. are probably more peelimatised

to chrtmic drought than must high forest tree species.

They already have features which reduce the impact

of drought: Jow teal area indices (leaf area per unit

of ground area), smaller, thicker leaves. strap-like

petioles leading to panphotometric (drooping! lest

posture; leal mils and Wax coating, Some species

readily drop older foliage ta reduce leaf urea. The

adjustments can mean Wik woot preductton is

ewinparable with that on mare mellow sles because

respiration load is less A eonsiderable vegetation

inertia oevurs with existing, mature irec-vominated

ecusystems eeaise Irees iifluenee microclimate so

siwiuficantly, especially in the availability of sunlight

andl top-sor] moisture They can also alter the character

of spil chemistry ws they store and retranslocate scarce

NUCreNts in tissues away from recycling,

Consideration of microclimates emphasises thar che

regencration phase. when seedlings hecome

established, 15 passibly the mest eritical phase 1p the

life-cycle of higher plants The gume is one of numbers

and uf tesource reserves in the seed. Must species in

S.A. have adapted so that they produce large numbers

of seeds with small or nevlwible food reserves, 90

regeneration becomes a mater of numbers. Vege

falional change associated with climate change well

depend therefore, an the reproductive vigour of invasive

and endemic specees and whether Microelimates and

provision af soi) conditions are equally-sutted to: the

potential invader.

Persistent drought. however, falls photosynthesis.

Jeads to slepletion of varbohydrate reserves, and

depletion of defensive compounds, Individual species

wilhin a plant community will vary in the pate of

reaction, Air lurbulence and wind shear, especially of

strong. hot, dry winds. as well as chilling winds, will

agoravale drought and damuge loliages caller trees are

more at risk that shart anes. Shallew-rooted trees

growing over stony materials wllocate more resources

to rout erowth than co shoom. Even so, they are

susveptible to infrequedt intense drought anél are liable

( die off in wholé geuups. Drought will inhibit

Microbial activily amd teduee minerahsation of

nutrients from litter: lugher temperatures in tumes of

rain will accelerate mineralisation rates hut also

accelerate leaching losses — both conditions im 4

changed chmate will limit the quantity of nutrients

available. Nearly any stress. if it becomes sustained,

leads to forest und woodland decline: it reduces Icaly

canopy, photosyothelic activity, stored reserves and

delensive factors thoughout trees more so thar on

lesser vegetation. Consequently the balanve between

species und individuals will alter and. in euolopical

terms, a sefal change will be initiated,

Sciemific investigations since 1988 into impacts of

facturs in the climate change siluation on vegetation

mainly have bee conceme@ed upon the effect of elev-

ated atmosphere carbon dioxide concentrations on

physiological and metabolic processes. This hus been.

discussed frequently ds a “fertilizer effect” und evidence

inthe field as well as numerous laboratory studies have:

been made (o agluce is impact, The subject was re-

viewed in Australia by the Ecological Society and

Society of Plant Physielogists 12992, 1993; Gilfid

1993)

The impact on biodiversity has been given much less

allention, It has been reviewed by Possingham (1993)

rF RB. BOARDMAN

at the species level in terrestrial ecosystems. He took

a mechanistic approach to an implied rapidly-changing

physical environment and eoncluded that. in the short -

term, extinctions will oceur through direct interactions

of species with changes in the climate af their

environment. A second set of species face extinction

either beesuse vl uppartunities availahle to disease -

cuUsiNg species uF to loss of mutual support (ee. loxs

of shude) Or Unsustuinable symbiotie relavionships

Parucular “keystume” species ur Une associiqvon may

become extine: und so change the essence of that

system. Direct interspecific competition ts considered

ty be the beast likely cause of species loss frowe a

presenlday community, Passingham (1993) argues that

cach species Will be affected hy une. or any

combinarion ot four hasie pathways, change in

geographical position to follow favourable climatically-

controlled fealures uf (he current range; seleclion of

tolerance to vliniave change within the current 4onation.

Nicto-cvolutionary Ghinge, and extinction. He suggests

\hal ip should be possible ty use autecolumeal

information to help track the pathways to new

synecolimued! Combinations of species. This supports

the ides chat there Is 4 need.andl an oppertunity: te

identify “uidicator species” and particular fealures af

these for measirement

Suwth Anstralian vegelation

in Ube late 20th Century

The foregoing discussion has described! genenil

(rends and effects, Once une Wishes to review the local

Siluation, One has to observe its arregularites. ‘The

Flinders Ranges, for example. cause the seathern

climate patterns to extend northwards but any change

is (odified by 4 reduction in maritime cenditions

(Boardman 1992), This is reflecied, for exanyple, in

the direetion taken by Goyder’s Line between Eudundis

and Crystal Brook. When climate chunge is considered,

however, it hag to be noted that rainfall, temperatures

und particularly marine cunditioas are modified by

the topographically-high region:

Natural yewetalion in rural South Australis has beer

weected wi Jaw allitudes by the direct influence of

changes in sea levels associated with Pleismecne

glaciation. A significant pam has been affected by

marked climate changes associated with the Ice Ages.

Collectively there fuve been both eduphie and

geographic barrlers te plant immigration. Both of chese

barriers have mean| that a considerable, but

unknowable, mumber of native species has the capacity

ty wrow in S.A. 1o colomise sites and cumpete strongly

with locul species. hut these have been unuble to reach

at by Hatural means. On the other hand. the species

which are present haye suffered genetic depletion

through intense environmental selection pressures in

the last we million yeurs in their refigia_ and have

limited intrinsic adaptability io change in directions

not already encountered (Boardmun 1986).

Much of the vegetation outside the settled areas and

influenced by Aboriginal peoples. can be classed as

ancient secondary yegcwon, A une propartoen ol

the remnant vegetation, including conservution reserves

in the settled areas should be regarded as miodert

secondary ur tertiary vegetition (Barker |98Y:

Bourdmiin 1986). These last two categories were lirsi

affected by European migrauon purposefully, and

secondly, but without direction, from escapes of

introduced domestic aniinals and plants which have

become feral. Kloot (1985. 1985, 1987 ) has assessed

plant intraduetions in great detail wand has idenificd

these from plant migrants and sje pative species,

Klon| (1987) also evinced evidence (hal intense hunting,

especially on isolated land, like Kangarmo Island, very

cea Ly an the period uf sentlement by migranes, decunutd

hative grazing specics of kangaroo and Wallaby.

Iransforming the native plant ecosystems mpidly and

dramatically, Fire frequency and species life-span

interact and this affects the stability or vulnerability

of coosystems (Barker 1989), Such information is used

lo manage minge in the pastoral zone Frequent

Cevestating fires in the first half of this century, as the

population increased, and fire protection messunes in

the latter half, alse have had @ strong visible impact

on partially cleared und uncleared nalive ecosystems

io the areas south of the 32nd Paralle) (Cirandison

1983),

The studies of Kloot (1987) are a reminder that weeds

are likely to be the bane of any attempt by native species

al colomisation of new habitat, even ifthe habitat is

an amenable one, Many of the weed species iitretuced

from overseas are much more adaptable (notice Kloor's

method of separating, natural Migration from ganien

escape) and ofien they have a much more aggressive

behaviour, Many sre walural pioneer species, Two

examples of weed species likely to be affceted in the

climate change seenane and able In migrate southwards

and compete with native species needing in do the

same to survive, are Montpellier broom (Genlstr

monspessulana) and Salvation Sane (kchinnw

Plantaginium).

[tis likely that the mast strongly threatened species

in the long-term are the perennial species which

dominate woodland formations Low sclerophyll

woodland and mallee firmations appear ww be

particularly at risk. These farmatins occur in the zone

with an average growing season, in the best possible

conditions, of from 3.5 to 5.5 months. It is changes

in the distribution and remains of the physically-

dominant species in ecosystems which have signalled

past climate changes, whether ie is chrough differciat

levels of pollen deposits, whose abundance speaks fur

lisclh or (hrough macro-fossils, On the other hand,

many ofthe perennint herb and low shnih layer species

SOME POSSIBLE EFFECTS OF CLIMATE CHANGE ON VEGELATION 73

in Jocal ccosystems ane more ubiquitous than the

evosystem-dominants. and frequently less constrained

by particular site factors. This ts partly because of

similaricy i the microclimates they generale, and

punly, because Iree-duminated ecosystems provide

wiclioration of the weather From shelter and shide.

The huge demand for moisture by tees un lpcal

ecosystems means that tolerance of chron drought

hy understorey species is normal. ‘This is the case when

the understorey of the mallee woodlands is conipared

wish that Of forest ceosystens in the Mi Lofty Ranges

shuwing, that many species haye a continuity in

distribution across a wide range in rinkall from near

10250) (nintn the Murray Mallee to around 1000. mim

@ Cleland Conservation Park (McCann 1989; Dashirs

& Fessop 1990),

Investigations af woodland productivity

in the S.A. semi-arid zone in relation to climate

Trees. a3 long-lived perennial pliant spectes.

Frequently dominate all or major pars of the

ceosystems that contain ihem, They accumulate a major

part of ecosystem solid dry matter in their wood and

bark Hssues, Wood and hark ure that part of the

perrental plants which sequester CO, and which helps

to maintain stability of this gas in the aimusphere. The

Research Branch of the former Woods and Forests

Department has investigated the amounts und trends

in woud dicuntulation by wees growing in stands. li

has been rogerded on an unit area basis, either us

vile per hectare (preferred by foresters because i

is bulkiness which is of interest) or as dry weight per

hectare. The overall effect over a long time determines

productivity,

Studies in S.A, were stimulated by apprehension

ubout fossil fuel supplies, future costs of fuel, and the.

need {wv assess tenewable luel resources in the mid-70's

(Kiddle-er al, (985). This work was expanded upon

when the Flectrivity Trust of South Australia (TSA)

wixhed Iu ussess the opportunity to sequester CO,

produved frum their puwer stauons (Boardman e7 al.

1992), Productivity of twenty tree species growing in

pluntations over a range of sales. sails and climate. his

heen measured, mostly as single species, but also ay

mixtures of species. About one-third have been

remeusured Which means we have bec able to establish

trends with increasing age. The use of plantations as

useful as both the age and the density of trees in the

stand, or level of oocupancy of site are known factors,

It also means thal we Were able to extend mecthods trom

existing forestry science and use them fo extrapolate

findings over Lime and space, to other localities, We

have compared aur results with general models od foresc

productivity, to se¢ how local stands compare with their

coullerparts On a global-scale,

World plant life-zone classification (Holdridge 174)

indicates (hil sustainuble tree-duimnated ecosystems

are uilikely co existon land in S.A. receiving less than

250 mam rainfall. Consequently vur research set tui

to include all well-cstablished plantations in a zone with

average rainfall between 250 mm and 6) mim, bh the

ever, nene of the plantitions suitable for Une study

were tound in areas which received rainfall al less than

340 mm. Dryland cereal cropping, whiel) requires

220 mm of winter season taimtall la produce ai

econumic chop of obu/aere [§40L/ha|_ rarely extends

below the 280 mm isohyet (French IY8Y), However,

ETSA asked that petential for tree-growing lithe 200

— 300 mim rainfall zone should be given special

consideration. What we found is of interest in the

preseni coniteat-

Atan ecosystem level. soil moisture replenishment

and capacity lo store water ate critical factors which

determine whether trees will graw or nil, n'S.A, there

ix a surprisingly frequent inverse correlanan whiell

exists between Soil depth and fertility; a large number

of shalluw soils. which have low capacity to store

moisture and physically restrict roor development. anc

fertile and an even larger proportion of deep sulk, atle

to-absorb all the rainfall reaching the round and

provide unrestricted physical capacity for roat

development. are infertile in the topsqil where KS %

Or more of ruots are sonnally found in both cases

productivity is markedly below the level expected when

conditions ane the bese that ca be exposed,

Mallee woodland gives way to ql shrubland with

Acacia specres within the 201-300 min raintall Aue.

‘There are exceptions and an hiatus ts. present as in

western Eyre Peninsula, where it is associated soils

vonditions that are unsuited 10 either life-tormanan.

Average rainfalhas a poor delinuthay factor because it

is the critical periods and intensity of rainfall which

count the mest. The anoual rainfall in drought years

is often as Nitthe ws half that of median years. In Utis

yone, alin, infrequent long periods of drouglit persist.

These considerations have led to Lhe concep) and

definition oF “influential rainfall” The quantity uf

moisture available to prevent willing, expressed as its

equivalent in rainfall, is JOO mm und i has to be present

in the upper root zone, The factors which Jetermime

iiftuertial raanfall have been assessed for dryland cereal

cropping and sites with natural vegetation. potential ty

suitable for vereals. Influentitl rain. uo averawe is

needed for 3-4.5 months in 12. The mininwim amaune

of influential raintall ta produce 2.6 dev t/ha/y of wood

exoceds 200 iin. That os to say, in this zone sites

capable of producing wood have i receive waler im

exoess of thar received from natural rainfall directly.

In uther words, they have to be “run-on” sites whith

receive Water as run-off ar percolated as druinage trum

arcas of higher elevauion, The minimum amount of the

folal Water Which his to be available 1s hiund to be the

equivalent of 300 mint, da months when there is a tall

% RK, BOARDMAN

of tain suflicient to promote prowth (ie a surplus over

respiratory eeds!, u needs ky be supplemented by a

leas! 25% Seustinal vartation is eriti¢al for vegetation

survival if (his zone, lhe frequency distribulion al

sustained Tengths wf rainy scason for a tumber ul

representalrve weather stations has been jnvesiivsted

and is summarised in Table | (Boardiman 1980). It

emphasises that growth figs a Uees way oot be sin

imnval sequence of rings but of adequate rainievents.

Tani |, The feegrency per centiry of rain periods ef a given

rheraniant jor sites fn the 200 mm ta 300 un avertee rainfall

sone of South Australia,

Averige rinfall 300 mm 250 mini 200 mnt

Rainy scason

duration,

Al least Sinonths = (7 years 5 years | year

Atleast 4 months 44) years 17 years 5 years

Atleast 3 months 70 years 35 years 16 yeurs

Al least 2 months. 90 yours 43 years 33 years

Ac lewst J month 96 years 85 years = 45 years

‘Trumble (1937, 1939), considering dryland cereals

Tequirements, showed that 200 mm gross annual

randall will provide armual herbaceous Vegetation with

will-Free conditions for up to 3.6 months. Trees in

woodland, however, intercept-a proportion of the grass

rainfall with their wafy canopies, [f the trees are mature

and spaced so as to absorb all the capacity which is

available for them to lay dawn new wood, ie. the site

ix fully occupied by the smallest number uf trees able

to do so, then throuhfall, rain able to wet the soil.

is reduced to 160 min. This effective aceessian will

only provide for 14 months growing seasun, which

is Quite Insufficient to support woodland, At 300 mm

ratifall. fie throughfall rs likely to be sufficient for

4.5 months, An actual growing season of four months.

the minimum likely to curry a Sustamable stand ot

woodland, calls for a gross rainfall equivalera of

320 inm in Upper Fyre Peninsula and for 310 mm in

the Mutray Mallee. Following Trumble’s argument, the

“run-on" requirement fram these consideranons 1s tor

about 450 mm water, bur it muse be delivered in the

Winter ramfill season to be “influential”.

Altertative approaches to the question of minimum

water requirement fora woodland without “zaps” bas

heen provided by Specht (1973, 1975, 1983) and Walker

eral (J986, 1984}. Both have dealt with natural

*cgetation oF the kind being discussed here. Specht

worked from the direction of evaporative demand and

Uctined the tlerable lnnits to evapotzanspirslion siress.

Evergreen, perennial plant communities. under

conditions which are the best to be expected, need a

minimum net monthly supply of sail morsture

determined by rain that falls and any accession or loss

of! water in the Soil. He called this the “soil availible

wurer” requirement and it ranges fram 37.5 to

62.5 nm. Sites able to sustain woodland are

characterised by a need for additional Water Which 1s

equivalent to 27% of average rainfall for the month.

In turn, this iniplres that there is wdequale storage

available in the soil within reach of foots, Taking

canopy inlerceplinn into consideration indicales thal

Tun-on component needs to be nov less than 47% of

rainfall at a 200 mm site, again showing that abcul

300 mm af rainfall equivalent is needed to sustailp

woodland furmitions.

The stind structure of woodland which will optimise

net dry matter production from the site factors available

can he deduced from the research of Specht and

Walker. Specht has found strong relavonshipy between

the Inliage projected area ('PC) of the tallest plany

structural layer (overwood — FPCo), and of the whole

ecasysten (FPCL), to his evapotranspination coefficient,

and hasedl his findings on a very wide sample of mature

Australian nutive ecosystems. Mature low woodlanil

in the 204-300 mum rainfall zone has an overwood cover

@! 32 to 42% and total ecosystem foliuge cover of

=6%: ic. there will be about 20% bare ground.

Walkers graup took a different view of folie cover

which recognised its fragmentary structure

corresponding to about BOG of FPCo, The opumun

cover hased on their work, and adjusted to FPCo terms.

suggests that cuver uf 324+ 3% shoul be adequate tn

sustain Jaw woodland on pun-on sites in the

21H)-30) mim rainfall zone at S,A- Their models allow

one to assess stocking rates in youg-age and juvenile

Stands an well as mature ones.

Thus, the effectiveness: of water supply can be

associated with net primary production and the length

of the growing season, These have both been mapped

by for present-day climste and the median scenarios

by Greenwood & Boardman, (1989) and Boardaran.

(1992). Nee primary production is likely to change by

13% for cach incremental gain or lass of 10 mm in

average annual rainfall These preliminary findines,

nuw Itkely to be extended further abead in time, ure

sulficient to indicate the level of changes which can

be expected and where the changes are more likely to

uccur. [he maps show the “best that cun be expected”

and until the sub-optimal standing of existing vegetation

cun be assessed, they form a workiag basis Tor taking

adaptive action, Table 2 lists the tree and shrub species

whose survival and distribution Im the zones most

sensitive to change, are the ones likely 10 be influenced

by threats anid uppurtunities presented by the clinusle

change scenaris, This is based on Chippendale & Wolf

(198)): Boomsma (1981), Boomsma & Lewis (1976) axl

Costermatns (1983). Table 9 shows the cucalypt

spécies sorted inte their botanical affinities (Pryor &

Johnston, 1971) in relation to changes if ecological

slatus (hat are possible,

SOME POSSIBLE EFFECTS OF CLIMATE. CHANGE ON VEGETATION

a euolajdwa punoy piada piiejajdulay

ead Buapuend ‘ysng uinjg ‘poompepurg waiito.mu -y idaoxa ‘dds warping

a ysnquicaig pIpuloun nonanjayy

a yeuooyy DIDJOaIUD] vINALD|AaW

a “LAS Saa]]BUL pay sipid0s snjddjpangy

“a DJONNS XOQ 39]/FIAl vsasod snjd<jpany

a D:LIS aayew pai Assopy psoajo snjdaéjoonq

a“ VOVHVW yueqdsuins pay = pyoudysouopu dss ‘pysudysosoou “F

a avo1s Da]yPur paymy-aspry Dipsspiour snidsjomng

a 8-ZIS aa][eW ParRal-Jepuals npunraof smdsjoomy

a o:xXISs ad| [Bu papiniy-aurnbs puosardjna smds)panq

a :TOS XO AY suaqp snd\joanq

a PALO WOWTUOZY pxalfas VadOD

“1 ould ssaidAa AURA DSOINLEAA StII{)D

a BlBsang psoulds puosang

“A SNISIqIY Dep] ipasany audsorpy

= aque juedayq aDiaiia DIavIP

a wripwAds pasuiy wrdafipiaa umnpudds

e 4aMO[} BeuUIND naadsy DYIagquy

al ALaYyo IAB siudiofissaidna sndavx0xq

a VVHNVAN wind moug = poyfionnd dss ‘naoyfionod snidXjpony

a VHANAS xog quruaddag Mpsapa ~wA ‘tntepe snd<joonq

” aqa:ons xog Aairy pdinaodmnu snpdnpoong

vi WITS aaqTelu ajoosdury psodne snidsjnony

a ANIdS ung yaeqo|pue> apigns snidiypong

a aqayvian YsB 22]|BV pjoulad snidkjpoanq

a qaD'kns umd ang ‘W'S weydxoana) dss ‘uopAxoonay smiddppanq

a“ oyans OG IDArSE suaoysiny Fy

“ dAVANOS sare ujooury “1g vaandindagn dss. pupauaopsuny 7

“ YV4INNOS da(]RUW UOSWULID) DUDaUMOpSUD]

dss punaunopsuny smdXqpany

“ WiXIS Joror, Silopas snidsjpang

“a q4lds “OQ Panval-suoT xApDIOINOS SsmidApINT

al VWIEAIS Daye BuoD pinqo}suor snjdkjponz

“a WSIS una ining xAppoopyj? snjdajnany

“a ars eli XApooAyomug sniddjpong

“ OVO'TS psonsun snjidspoong

“ alas DaTPEUL UTA sdagup sniddjpang

“ arjdod aantn ymypofiuyoa sndansquepapy

1 yea ayg DIS DUMUDHSYD

“ aud ssaidAa Aan SUPISAUDI SPAIN IDD

“ PPA SAPOUNA DIIDIP

UOLESTO OD Apg[> 10U qwanbay uonnquisip ut USSPID awWe_Y UOLULOD soioads

ul wipe 10feyy SSO] Jo ureD ssay. — uoranpaz uoysuyor

SSO] ATS aUIOS ioe 2 IOAIg

SORIVNSDS SAONVHO ALVAITO ‘sdnioig,

HLIIM SN.IVLS NI SONVHO TVILLNALOd sHpds[Dony

4266] ‘dna woeduy amu) " uorSar-qus wu

Ajjodiiuidd 4naro paiaaffy suaisXsora ‘ajoN “pyp4isny YINnes 4of SolapUads aSunyo annul? Aq paivaffn Ajrunayiusis aq 01 Xpayty 81 SNIDIS asoyM Saizads qnays pun aap “7 FTAV I],

PAVNAY

VO:.LIS

W:LIS

avis

a@a4aNnns

Tas

R.. BOARDMAN

VRARNAYVADANLNALSLALALVAY

VHANAS

ScHs

OO°XNS

€D:e 1S

4O-XNs

WHNS

MHANS

OVEVA

HIS

IDeA TS

VOans

YVR3VYVYRVVVVVNVARY

16

ysnq yaoi

poomalpaan)

poomjoog

ysng 19h Jea|~JOATIS

OPIS YOON

tuniodsowid Suidaay,

Soysnqies

ysng aay

poomrsng

Ausyd sareu payuiog

aoqpeu RaRqUIn,

Sd][BWW EIRTBA

saqjew idoy

xog jurtuuaddag

SaTJEW J1BSAq, BUONIA

JOMOLXEAK

JOMOLXEAK

saysng nu

RISSRI LING {-FATIS

Ye}9g

auld ssaidaa wiayuoyy

YSU prac]

BQOOd JIATY

aVRM MOTTA, UOTYsNOIg

SVEN S.PleMsE_

ean |[ypurs

RAN EYaIqUIp,

vain

ysnq [pur

Da]]RW UIaity

2ayjeu Buudg evo,

tung anjq pueyuy

JOT[EU F9A]IS

und anjq “eq aAq

Yequooo yaeg-uINg

aayyeur Aan

BOT[BL AIIM ][PISPOD

aa][RW spur[ABLinyA)

daq[BUE SIY AA

xO Paavo[-peoig

au] Yasaq,

saysnqdopy

Wngy [Pq

aurd ssaidA> aayynog

[fRALU UaIsaAy,

ysng ioe

wnyofapajya unspuaposajayy

paaidomnay Day

DIDS ba} Wa)

pyxydoipuiau pappiaan

ppiydouad ppig

sapioaiXjijd uimnsodsoiig

‘dds pipospyy

‘dds puppy

‘dds vipodouay)

DLIIPAD]]IG DUDA

undaprdjojd umsodoxy

snqjdydo sndinzoxq

pupsipquin’ snidipang

Sistiaqwnjnd snidijpangy

XApDINDAIS snidKjnanz

pyofusnsuy ava ‘vinsiopo snidxppoanq

puunuos smdxjnong

SUpaul] UOWAasoUT

tunyofusnsup wousisouz

‘dds njrydowasy

‘dds vissvp

DIDISIA DULMDASDD

SLUBOUNJOI STUIYPOD

pysydouodnaral pLonoy

pypAydouais piano

MIIPS DIODIP

IP]DMSO DIDI

pypsydoul) viovop

DAYIDISAY. WAG DIDI

DANAUD DLIDIP

mopfirivd psafiarny

spi sniddjponq

SIpDAlal snidkjnang

psouinad “dss ‘uoj<xoonay snidipoong

pipapid snidkjpang

supjonad sniddjpany

pixayaiuy snidjoomngy

ussapuayl snidxjoonq

vyofissaaip snydxjoong

ppydydounka smidQnony

pupaspaypa snjdKyoanz

ounuyeg snidAjoony

pons snapaowAaALy

‘dds nauopog

syppuanitd sndiprouopavy

nsiaad Sino y

pdinsoiddod piovay

DSNIa4 DIUOIaTdWA]

‘WWOD 7 ITY L

T7

SOME POSSIBLE EFFECTS OF CLIMATE CHANGE ON VEGETATION

a MANS daypRur Aaicy uszapuny, smdkyoony

SANYOOILSAYTL

VIAVLIAESXd ‘uonsag

“f OIVOTS psojnsun snjdkjpang

e aVOTs aal[Rtu parinsy-adpry nyossasaul snpdapvINy

_ , aVLVSSVYONI

4 a-TTs ein xipoodyooig snidkypong

a VWs aay[eu ajoossuly nso8ns snidxjoony

AVIWNOYOL

“a aad 1S Ad]TBW OYA sdaaun snjdkjpong

“ VESTS aayeu Suod pipgqojsuod smiddjoony

“ Las JAT[BUI asa] BLO A puulouod snid<joong

“ H:A'1S DaypEw spurl AenyAy oyjAydouvad snidkjoong

al 99:4 1S SQT[EUL. ATU, pupainajpa snjdsjosngq

do:a 18 2ay[etw Satis pivapid smidjpang

wt HAS aojeur idoy xATRonesys shydAyeong

davsonnd

ce a‘ZIs 23][BLU PaAea-sapua|s ppunvraof snidkjnangy

“ a:xIs aa][RuL payimsy-arenbs puvsortjp2 snid<joang

a V'xIS ]]20K, sytonds snid(jpanq

id VO:LIS SaT[PlU BIRQUINA, pupsapgiund snidXjpang

rt WiLIS AaT[eUL BIETBA, sisuaynjod sniddjppong

“A TLIS saaqjeur pay sypios snidkjpony

a O-LIS aaj[eu par Assopy psoajo smidijpong

“ VeSIS iuns aesns axjpoopyp> smd(joony

cal US oaypeut duridg ROA Smyppaliy SnidkjpIny

WIAVLOASTA “vons9g

SN.LYAWOHdWAS

“SNNH9-ans

“1 PAV Yse aeq[RA piowad snidspoony

(@ wordas-qng)

“ VVHYVW wns moug = aiazfionnd dss ‘vaoyfionnd snidkjvang

: } AVNOI10

a“ VOVHVW yaeqAauins pay pijaudyonrmua “dss ‘pyoudysossow “J

AVIVTIFLId VD

ot OVEVIN SQTJEUL BIYM [BISeo pyafissaaip snjddjpanq

AVIIOSISHAAIG

VIASH.LNVNAS ‘YOlpag

SNLdATYDONOW

*SONA9-dNs

UOLJRSTUGTOI iraja yOu quanbay UOLINGLISIp ut uosuyor aueyy UOUTIUOD saioads

ul uied soley SSO] JO URE ssa], — uononpal ww IWAIG

SSOT als auiog Joley, ‘dnoin

SN.LIVLS NI JONVHO 'TIVILNALOd

WdNOld Salsas PUD Dsauad-qns ul pais] —

NOISAL-GNS Ul SOLADUAIS AsUDYD ated Aq pawaf{n Aporuaiod saizads smdAyeong “¢ ATVI

78 R. BOARDMAN

The ecological significance of Goyder’s Line

\ The practical limits to sustained primary production

for land-use historically have been contentious and are

likely to remain so, G, W, Goyder was put on his mettle

to define such « boundary. which, all things considered,

has been largely vindicated by time. He surveyed his

line of demarcation for sustainable Jand use options

LX oad in 1865 (McGowan 1990), He rode on horseback over

3000. miles (almost 5000 km) to define it. He

associated the boundary with repeatedly observable

facts related to the southernmost extension of drought

as indicated by vegetation and soil features. These were

the “chewed remains of saltbushes and other low

4 4 shrubs" and “light souls which were susceptible (o wind

erosion” when dry, indicating stlty-sandy textures and

low contents of organic matter in the top soi, It has

been alleged that he only surveyed the Upper- and Mid-

north Regions but extrepolated his criteria ro Eyre

Peninsula and the Murray Mallee regions without

Yu’ A VAY \ | Specific attention being given to them. However, the

boundary line in the central Section, (tom near Moonta

in the west to Eudunda in the east, is only 350 miles

long. His extensive distance travelled for the survey

demanded by. the Government suggests he ventured

. <men< wl hf nfie ius i

ra 5 a us <4 y eeaR vee much further afield to justify his boundary,

zee vu 28 £ ft 2 Be7ZzR 4Ax The Line, recognised as a broad limit, has come to

Aa 2 2P27 Rk F 55555 ses] , ¢ : ,

mA an 4 HF FRAADA AAW] be associated with three features, an average Winter

. ~ 7 ‘ 4 ' iB «

raintall of 220 mm, # rainfall/potential evaporation

ratio of 0.26, (French 1989). and the southern extent

of ecosystems dominated by saltbushes species: in

particular Maireana pyramidata and M. sedifolia.

None of these features alone is adequate ty define the

sustainable land-use criteria for the 20th century. OF

particular concern here is whether its original joint-

ermeria will be assessable in time to serve in the latter

x a 2 E half of the 21st century.

ze 5 = C= 2 on ; ge . h

£3 oe: Zz Zve French (1989) showed that if the Winter (April-

= ss mes z

Bx a ° x Ge w= %! October) rainfall/evaporation ratio was to be

oo , eB we KOE FS. Bey) eo ;

$5 6 £53 3g BS Sf eg S 22] maintained, the boundary associated with Goyder's

ssp £ £7 86 5 Se mayer] : y ‘ hab

ei ee £ #§ # AS Ew .| Line would have to be located where the winter rainfall

Fe @ $2 56 £2 2 € See! ! . = E v9 !

33.65 £246 5 2 s o£ 2D; | wsohyer is currently 285 mm: ic. shift some 70 kin

_ | south of its present location on the plains of Eyre

= = =| Peninsula and the Murray Mallee, Greenwood &

Bes = =| Boardman (1989) assessed changes in growing season

a EE = =| and net primary production in native ecosystems. The

& E35 S «| northern limit of low woodland, defined as ecosystems

Ss = > el ;

4 = S$: * © | which increment new growth, under udequate

= & Sf ¢ 3 > ae is Ses contlitions, of less than 2.2 dry t/hit/y would move

PeSse § Boe 2 FES eu eS=| south by about (0Okm. These iwo independent

SEGSZ Fe tSsy, sesso lsezzt : ¥ ; 3 P

= S8Ss0etez Sees £5 SS5 x 35 =| estimates, considering the uncertainty, ave in reasonable

= =a EQ = CZSE*SESSEPRT SER LBES Ke oa aul ;

gut sa5S Ea 2G BE SS BSL S585 SS | agreement. There is likely to be a drastic impact on

ats. 25555 20 25 ges ES=f=S°0 25 =) vegemtion, To the east of the Flinders Ranges the

efsss SQSshsozosys z SSS5SSs| change indicates there would be a much more intense

Besos tert ce epgsagttss rit teywstt : : é ee . a

xR Ssgi< .= = ‘ ere = 5 H B, & fre 7 ‘ e plain: ;

<5 8§$8s5gS232acds VSsstagee gradient of change from the hills to the plains than is

presently the cage,

SOME POSSIBLE EFFECTS OF CLIMATE CHANGE ON VEGETATION cr]

Adlaptations to greenhouse-indsiced

Climate change

One of the mist difficult tisks for managers of land

which grows long-lived plant species, crees and shrubs.

espectally Where these species dommate ecosystems,

is how ty assess that change im environmental

conditions: is actually of significance to the present

ecosystems, and what do do to ensure that the

ecosystem, rather than the species themselves, can be

mamtained, Practical things to do which wilh make

adaplition possible are linmted hy the imlefinite nature

of the time frames. It is difficult to make a positive

contnbulion over such a protracted period without

efforts being conlounded hy normal variation in

weather and climate year by year. One positive

sugvesuion being widely Ginvassed is mainly applicable

to commercial tee species subject tu improvement

through hreeding. This has heen to progeny-test in

areas Which have 4 climate akin ls the changes expected

but currently outside the regions of economic viability:

The soit criteria adopted by Geyer can be

ascenaimed oi the basis Of textural classification and

degree of degradation associated with historic land-

use and the ramfall quanuily and evaporation trends

should be discerhible through use of runiing averages.

Consequently, a fulure equivalent of Goyder’s Line

could be used to highlight the progress of clinute

change. should it eventuate.

Better definition of the boundary could be achieved

with the selection of “indicator” plants, such as Mairued

pyramidifolia, indicative af change in favourable or

undavourable environmental condrions. arl gov

definition of the ecological factors which determine

their limitations or provide for stability is needed.

Ahility of indicator species to colonise sites south and

west of the current distribution should be priority

subjects for coology studies.

Ancther approach which has been adopted by

Primary Industees SA-Rorestry, that recognises forest

and woodland have multiple-uses, has been installed

(in 1992) near Gumeracha in a Demonstration Forest,

This investigation has been designed to integrate both

long term and short-term benefits. Uses included are

witer supply catchment protection, agro-forestry and

potential commercial foresury, Botanically closely-

related land-races (provenances) and species. in tive

disiinct groups of Eucalyptus similar. in fashion to Table

3. have been matched to three climatic models. Trees

are heing grown from seed collected (i) in good quality

lucal populations in $.A4. (in) fram central Victoria

where the rainfall distribution pattern ic like the

postulated change, bur the temperitures do not change,

and (iii) and populations in the central part of eastern

NSW where minfall disteibunen with a tendency to

surinee maxima and warmer temperatures occur. The

Nowering, seed set and natural regeneration phases of

ecosystems, us mentioned above, are likely to be the

MUST sensitive seral Stages to significant climale change

The Gumeracha investigation aims wa cover this

eventuality by providing genes in the dominant species

which contain adaptation to climate in the direcuons

in which it is very likely to change, The triaf is

providing a hank or nest-ege of penes to callow

adaptation to occur whalst sill likely to retain ecosystem

structure associated with land-use

A Forest Reserve al Whyte-Yarcowie. in the zone cast

of the Southern Flinders ranges. which our studies

incicate will suffer the most intense changes and

increases in stress, has been reserved tor scientitic

studies, One option we are considering is to plant a

suite of ‘indicator species, in range from those suited

well at present to others growing in regions akin to

the future climates of the scenario. Planting in a set

design at ten-year interval lor the next century would

be tuimensely valuable in an ccolugical sense. Planting

a bracket of three consecutive years at the decadal mark

would help to reduce the eflects of annual vanaton

in weather. Such a programme calls for dedication,

conscientious adherence to the plan and consistency

to be adopted by managers, and lasting, over several

generations of ecologists. [IL inay be too much toask.

A second question concerns conservation of the

current species composition of ecosystems which are

valued and preserved in Conservation and National

Parks. Greenwood d& Boyrdman (1989) reviewed the

impact of the median climate change scenarto on the

representativencss Of Parks in S.A, There has been no

significant addition to options suggested an 1988,

Provision of gene banks and corriders of land able to

permil species migration, mainly southwards is. still

seen to be the best option in the interim..In the ubsence

of physical evidence to justify the need or conservation

practices, sadly little has been added beyond revision

of the Native Vegetation Clearance Act, to suengthen

its preservation provisions, and to positively protect

roadside native vegetation. There is still a need to

include und emphasise whole-ecosystem conservation,

and to modify current preference and emphasis on rare

und endangered species.

80 R, BOARDMAN

APPENDIX L, Average annual temperature, annual rainfall and growing yeason esfimates Jar recent times and the late 2lst

Century adapting the IPCC 1991 median estimates of change in temperature and rainfall with latitude. Growing season length

has been estimated with the De Martonne Draught Index from mean monthly rainfall and teniperature data and estumates

far recent times and the late 2lst Cennry respectively {see Boardman, 1989),

Region Weathet Mean Average Growing, Season Change

Station Temperature Rainfall months in GS.

Recent Late 21C Recent Late 21C Recent Late 21C 4 of year

Eyre Pen. Cleve 16.5 19,9 404 410 5,75 25 -28

Ceduna 16.9 20,1 321 323 4.25 2.0 19

Polda 16.35 19,7 aq? 431 5.5 425 i

Kyancurla 17.0 20,3 330 340) 4.35 2.0 18

Whyalla 18.0 21.35 273 293 1,25 OS -6

Pt Lincoln 16,35 19,8 486 464 6.0 4.75 “1

Kimba 16.55 19,85 346 352 4.75 15 -28

Minnipa W2 20.45 325 323 4.) 1.0 26

Streaky Bay 17.45 20,75 378 36) 45 3.25 1

Yorke Pen. Price If 6 20,05 332 343 4.25 11 -26

Warooku 16,05 19.55 450) 433 6,0 4.5 -10

Maitland 15,95 19.35 509 50M) 6,75 §.25 -12

Kadina 16,65 20.05 346 495 5.25 2.75 2

Nowth:

Upper Hawker 17.65 20.85 301 $11 3A <0.5 27

Mid Yongala 14.4 {7.7 369. 376 5.5 1.5 -34

Bundaleer 14.8 17.1 554 544 7.0 7.0 +0

Georgetown 16,05 19.35 468 466 6.5 5.0 -12

Snawtows 16,4 19.7 407 405 6.0 3.0 -25

Lower Kapunda 15,75 19.2 496 493 6.75 6.0 -6

Roseworthy Coll, 16.35 19.8 439 443 6.5 4.5 -16

Murrayltands Milang 14.9 18.45 408 4n9 6.25 3.75 21

Nildottie 16.2 19.65 256 277 1.25 0 -10

Wanbi 15.7 19.2 307 325 3.0 <05 “2

Meningie 15.2 18.75 470 465 65 45 -lf

Wajkene 16.7 20.1 31] 342 1,75 0,75 =)

Lameroo 15.55 19.05 493 400 5.5 3.0 -21

Monarto 16.0 19.45 35] 371 4.5 0.5 -33

Mt Lolty Ranges Mt Crawford 13.15 16.6 784 755 R.0 10,0 +16

Myponga 13-6 17.05 763 734 1) 10:0 +20

Upper SE Bordertown 14.7 18,35 541 571 4.25 8.5 42

Keith 15.55 19.15 47] 474 6.75 5.0 -14

Lawer SE Konett: 13.45 17.35 713 716 20 11,75 +23

Naricoorte 14.45 18,05 586 580) 8.0 9.0 +8

Mt Burr 13.1 16.6 782 716 a5 11,0 +13

References

ALLen. R. J, (1988) El Nitto southern oscillation influences

in the Australasian region, Prog. in Phys. Gog. 12,

Va-344

AUSMAP (19894) Natural vegetatim about 1780, Map

CSU89/055 (Surveying and Land Information Group, Dept

of Administrative Services, Canberry).

____ {1989b) Present vegetation in Australia. Map

CSUS89/054. (Surveying and Land Information Group,

Dept of Administrative Services, Canberra),

BaRKER, §_ (1984) Natural ecosystems in the Greenhouse,

pp. 105-108 Jn Dendy, T. (Ed_)* Adelaide Greenhouse "88

Conference”, (Dept of Environment & Planning,

Adelaide),

BoarpMan, R. (1980) Perspectives on managed ecoxystems

for the semi-arid zone of South Australia pp, 1-16 Jn

Section 13. Agriculture and Forestry, Proceedings of the

50th Jubilee ANZAAS Congress, Adelaide (Woods &

Forests Department mimen).

_ (1986) The history and évolution of South Australia’s

forests and. woodlands. pp. 16-31 Jn Wallace, H. R. (Hd.)

"The ecology of the forests and woodlands of South

Australia” (S.A. Govt Printer, Adelaide),

_ (1989 The impact of the postulated enhanced

Greenhouse Effect Jn Dendy, T. (Ed_) “Adelaide

Greenhouse “88 Conference”. (Dept of Environment &

Planning, Adelaide).

__s {1 992) A study of the growth and yield of selected

plantations of native trees growing in the semi-arid cone

of South Australia and their capacity to sequester carbon

dioxide (Dept of Primary Industries, 8 A.; Woods and

Forests, Adelaide),

Boomsma, C, 0. (1981) Native trees of South Australia.

Ind edtn Bulletin 19, 1-228 (Woods & Forests Department,

Adelaide),

__ & ~ Lewis, N. B. (1976) The native forest and

woodland vegetation of South Australia. Bulletin 25, 1-313

(Woods and Forests Department, Adelaide),

CHirrenpAue, G M. & Worr, L, (1981) Natural

distribution of Exealyptus in Australia, Special Publication

No. 6 (Aust. Nat. Parks and Wildlife Serv.. Canberra)

Crmare Impact Group (1992) “Climate change scenanas

for the Australian Region”. (CSIRO Division ot

Atmospheric Research. Mordialloc).

SOME POSSIBLE EFFECTS OF CLIMATE CHANGE ON VEGETATION sl

CosrreMaAns, [.. (1983) “Native trees and shrubs of south

eastera Australia’ (Weldon Publishmg, Willoughby. NSW).

DasHokst, D. R. M, & Jessop, J. P, (1990) “Plants of the

Adelaide plains and hills’. (Kangaroo Press, Kenthurst,

N.SW.),

Frawkein, J.P, Saucart, AH, H, & HarMon, M, E. (1987)

Tree death as an ecological process. Bioscience, 37, 550-56.

Fresenu, R. J. (1989) The Greenhouse Effect and its

implications for primary production in South Australia,

pp. 137-142. Jn Dendy, T, (Ed,) “Adelaide Greenhouse "88

Conference” (Deprof Environmem & Planning, Adelaide),

Fritts, H. C. (1976) “Tree rings and climate’ (Academic

Press. New York),

Gitorb, R. M, (1993) Review! Responses of vegetation to

elevated atmospheric concentration of COs, Climare

Change Newsletter 5(2), 10-11.

GOVERNMENT CLIMATE CHANGE COMMITTEE (1990)

Implications of climate change for South Australia. [First

Report. (The Government ot South Australia. Department

of Environment & Planning, Adelaide),

OuANDISON, R. (1983) The relocation of fifteen sites. painted

hy S. 1G if whilst accompanying the Horrock’s expedition

of (846. Roy: Geesr, Soe Australasia, 8A, Branch, Prec,

83. 12-21,

Garenwoop, G. & BoarpMan, R, (1989) Climate change

and some possible effects upon the terrestrial ecology of

South Australia, pp. 112-132. Jn Dendy; T, (Bd.) “Adelaide

Greenhouse "84% Conterence’, (Dept of Environmenr &

Planning, Adclatde).

Grirtin, T. & MeCasnini, M. (Eds) (1986) “Atlas of South

Australia (S.A Govt Printing Divimon & Waketielll Press,

Adelaide).

Huppripoh, L. R (1974) “Life zone ecology”. (Tropical

Science Center, San Jose, Costa Rica),

Howarp. G. W. (1967) Arthropods inhabiting pine litter in

the South-east of South Australia. M.Ag.Sc. Thesis.

University of Adelaide Unpubl.

Kipoue, G,, BOARDMAN, R,, ARCHER, GR, & VAw Der

Sommer. f (1987) A study of the growth and

characteristics of woodlor and amenity tree plantings in

semi-and rural South Australia, (Woods & Forests

Department, Adelaide).

Kuoor, P.M. (1980) Dr. Richard Schomburgk’s “Naturalised

weeds” (1879), J, Adelaide Bot. Gard, 2, 195-224).

(1983) Barly records of alien plants naturalised in

South Australia. Joi, 6, 93-131.

____ (1985) Plant introductions to South Australia prioy

to 1840, Ibid, 7,, 217-231.

_ (1987) The naturalised Nora of South Australia. 4-

ts manner of introduction. hid. 10, 223-240:

Lewin, R, (1985) Plant communities resist climatic change.

Serence 228, 165-166.

McCann, I. R, (989) “The Malice in Tower” (The Vietoriant

National Parks Association, Melbourne).

McGowan, B. (1990) Guyder’s Line. Heritage Australia.

Spring, 90. 1-7.

Penripce, L, K. & WatkerR, J. (1986) Effect of

neighbouring trees on eucalypt growth in a semi-arid

woodland in Australia. J. Feel 74, 925-936.

PossivcHAM, H. P. (1993) Impact of elevated atmospheric

CO, on biodiversity: mechamstic population-dynamic

perspective. Aust. J. Bor. al, W-21.

Pryor, L, Di & Jonson. L, A, 8. (1971) A classification

of the eucalypts (ANU Press, Canberra),

Rurrer, Jo. (1964) Pinus radia D.Don in the lower south-

éast of South Austraha. M.Sc. thesis University of Adelaide.

unpubl,

Simpson, J, A, (1967) Mycological aspects of decomposition

of pine liter M Ag.Sc. thesis, University of Adclaide.

unpubl.

Sparrow, A (1989) Malice Vegetation in South Australia,

pp. 109-124 Jn Noble, J. C. & Bradstock, R. A. (Eds)

“Mediterranean landscapes in Australia; Mallee ecosystems

and their management” {CSIRO, Melbourne).

Soci#ry Or Plasy PHrsiglogisis (1993) dust) J Ber, 41.

SpecuT, R. L. (1973) Waler use by perennial evergreen plant

communities in Australia and New Guinea. dust, J Bot,

20, 273-299,

(1975) “The vegetation of South Australia” (Govt

Printer, Adelaide),

—— — (1983) Foliage projective cover of overstorey an

understorey strata of mature vegetation in Australia. Acsz.

J. Eeal. 8, 433-439,

Trumare, H.C. (1937) The climate control of agriculture

in South Australia, Trans, R. Soe. S. Aust, 61, 41 62.

——__ (1939) Climatic factors in relanon to the agricultural

regions of southern Australia, Jie. 63, 36-43,

WALKER. J, SHARPE, P. J., PENRIDGE, L, K. & Hspx-l Wu.

(1986) Competitive interaction between individuals of

different size; the concept of ecological fields, CSIRO;

Institute of Biological Resources, Division of Waler and

Land Resources. Tech. Mem, 86/11,

. CRAPPER, P. F, & Pewrince, L. K, (1988) The

crown gap Tatio (C) and crown cover: The Meld study. Azer

J. Eeal. 13, (O08.

WarinG, R. H. (1987) Chiracteristics of trees predisposed

to die, Bioyelenee 37, 569-574.

THE ECONOMIC IMPLICATIONS OF CLIMATE CHANGE

By MICHAEL E. BURNS* & CLIFF WALSH

Summary

Burns, M. E. & Walsh, C. (1994) The economic implications of climate change.

Trans, R. Soc. S. Aust. 118(1), 83-89, 31 May, 1994.

Feedback effects from the rest of the world on the economic structure and well being

of South Australia are likely to be more significant than direct economic impacts of

climate change in South Australia or even Australia as a whole. Economists can help

to identify climate sensitive sectors of the economy and analyse how change in them

will affect other sectors. Economists can also influence anthropogenic changes by

prescribing appropriate incentive structure to redirect human actions in the common

interest. Because of the nature of the data upon which economic models are built,

their predictive capacity is restricted to short periods of 1-5 years. It is postulated that

areas such as agriculture, forestry, fishing and hunting would be highly sensitive to

climate change whilst electricity, gas and water, construction, recreational and other

personal services together with ownership of dwellings would be only moderately

sensitive. Mining, manufacturing, wholesale and retail trade, transport, storage and

communications, public administration and community services, finance, property

and business services would be negligibly sensitive to predicted climate change. The

problems associated with control of greenhouse gas emissions in an economic

framework are controversial and difficult to resolve in a global context and the

significance of collaborative approaches to these and other problems cannot be

overemphasised.

Key Words: Climate change, economics, greenhouse gas emissions, anthropogenic

influence, economic models.

Transactions af the Raval Saeiety of §. Avy (999), TB), 84-89

THE ECONOMIC IMPLICATIONS OF CLIMATE CHANGE

by MICHAEL E. Burns* & CLIFF WALSHT

Summary

Hrkns, MB, & Watart, C. 1994) The ceonotine diplicatrons of inate change. Trans, R. Soe. S. dust, UB),

83-89, 31 May, 1994

Feedback etfeeta from the rest ef the world on the economic stracture and well being of South Australia ait

likely to be more significant than direct economic impacts of climate change in South Australia or even Australia

us a whole, Economists gun help tt identify climate sensitive sectors of the economy and andlyse how change

inthem will affect other seators. Economists cun also influence anthropogenic changes by prescribing appropriate

incentive structure to redirect hurnan actions in the common interest. Because of the nature of the data upon

which econamic models are built, their predictive capacity is restricted lo short periods of 1-5 years. Irs postilated

that areas such as agriculture, forestry, fishing and hunting would be highly sensitive to climate chunze whilst

electricity, 205 and water, construction, recreational and other personal services together with ownership of dwellings

would be only moderately sensitive. Mining, manufacturing, wholesale and retail trade, transport, storage and

communications. public administwtion and community services, finance, property und business servives would

be neghipibly sensitive 10 predicted climate change. ‘Vhe problems associated with control of greenhouse gas enissiuns

inan economic framework are controversial and diffieule (0 resolve inv global context and the significance of

vollaborative approaches to these and other problems cannot be overemphasised,

KFY WORDS ctimate change, ecnnomlcs, greenhouse gas emissions, anthropogenic miluence, economic models

Introduction

An important feature of the symposiusn tor which

this paper initially was prepared was the bringing

together of at least a litte move information abou the

likely regional and local effects of clitnale change

(beyond very tentative material available from the

Greenhouse "BS Adelaide workshop and elsewhere),

The contribution economists can make doesn't depend

totally on this information, however. Like other

sciences (social us well as physical), economics can

contribute to discussion and debate through increasing

understanding and application of its way of thinking

about issues.and problems. We are not here referring

ta the commion view that economists insist on reducing

everything 10 dollars and cents. Rather, we mean the

more general contribution that economics can make

through its development of frameworks for evaluating

public policy options which:

* emphasise costs as well as benefits of policy action;

* pay particular altentivin W incenlive structures 1

policy desien;

* jecugnise how the actions of one individual,

enterprise. region pr ration impact on others; and

= uotwirlistandine the usual presumption that

econumists think that markets work perfectly,

acknowledge causes Where “the invisible hand” of the

market 1s either arthriic or non-existent, such as

When, there are spilliver effects, poorly defined

property rights. Tigh transactions costs, or the need

*

School of Koonanvics, Flindets University of South

Austrulit, GPO Box 2100, Adeluide, South Austratia 5001

- South Australian Centre for Economic Studies, The

University of Adeliide, South Australia 5005,

to impate “existence values” (eg. of life on earth)

exists, and design policies to moderate their

consequences.

Moreover, like those physical sciemists who are

deeply engaged in uttempting to understand climate

change and the role of greenhouse gases in WW, the

cconomists’ framework is general equilibrium in neture

and, to the greatest extent possible and necessary for

the problem at hand, v1 is global in its recognition of

impacts, To be more specifie, the economists’

modelling includes feedback effects between markets.

including from. world markets.

This observation leads {o our Tirst, simple, and yet

protoundly significant poimt about the economic

implications of climate change, That is, for a small

region like $.A., with # not highly unusual economic

and physical structure, feedback effects on our

economic suructure and well being from the rest of the

world are likely to be more significant than direct

economic impacts of climate change within South

Australia, or even Australia as a whole.

On the basis of modelling of various scenarios

produced in late 1980's, estimates of the loss of World

output and income (often extrapolations from specific

country studies, especially the USA) ranged trom

1%-6% of GDP, for tempeniture increases in the range

2-10 degrees Celsius. These are very large oulput and

income changes in abyolute terms, and their effects on

growth of demand for South Australian products would

be very substantial and likely to ourweigh any direct

adverse effects of climate change in South Australia

oo South Australian production and incomes.

The estimates of climate change, however, remain

highly uncensin and have tended to be progressively

8 M. EF, BURNS & ©, WALSIL

reduced in more recent years, The uncertainties about

Whetlier, when and how climate change will occur.

mhoreover, are compounded by the fact that, they are

bused on a presuniprion that policies will net change

sOon enough, or sufficiently, fo ulier outcomes,

Predicting policy responses is as Important, in the end_

its predicting climate change isclF

Cneertainties in elimate predietian:

the policy feedback issue

Others are beter qualified thin ws la explain the

nalure Gl the models heme used to genenite cline

predictions based on observed jrends in greenhovsy

yas emissions, and (he uncertainties that sucround

them. What we might usefully focus some attention

on. Wowever, 18 the iitet-related iss,ue of patiey

feedbacks,

The ceonomic projections of the consequence of

vlimare vhange that currently exist are based on it

“worst case” policy scemirle of mo significant change

16 policy in response t emerging. or anticipated

climale changes, Thrs degree of pessumisii inty be

uuwartanted, bur climate ehange certainly involves

some characteristios which give pessimistic outcomes

i higher than asoal Weightin probability distributions

delined over Jikely responses.

Policy responses to climate change have

chanicterisnes siomlar to what cconanists, in ather

chmtexts, cull public goods, For example, utthe fanuliar

case Ob a xystemt of national defence, pnee a unit OF

“prowetion” is provided for one person il 1s

Simultancausly provided for all others, and those

unwilling lo contribute to the costs cannot be excluded

from benefits, In large numbers context, individuals

see themselves as having mevliible impact on

quantities produved whether they offerte contribare

te Costs or not, and therefore the rational respoose is

hecuterupt (ofree cide, Even theugh all others muy do

likewise. and nothing is produced ws a result. it sull

Will nat pay te offer fo venerbute because whether you

door aol senchally whl ootaffeet decisions of others

Thisis a classic case of market fullure, and a case

Where government provision, funded through

votipulsary taxes, is Inevitable und desirable.

Tt should he noted, nonetheless, that even mt the

relatively simple case of defence..giving governments

“the power lo coerce” does notensure that un optupul

quantity of defence will be produced. Voting. lobby ding

und other means by which people signal their

preferences in the political market have peculiar

chuructecities. und the inventive structures facing

political wad bureauwerutie decision-makers have a

temleney to encetirage oversupply of publie sect

pounds wld services.

Clearly, polictes to control climare change bave the

same characteristics as defence: effective netion wall

benefit all nations. whether or not they contribute to

the cost (financially and/or their own paliey actions),

They also face the added complennty. of course, that

&@ vovernment does not exist that can coerce mation

stiles Co pursue policy changes, We have to rely ob

IMernational cooperation, a nolariousty rare

phenomenon. pursued through cumplex procedures,

which ts bkely (o see countries and groups holding Out

(free riding) at others” expense — even when

perceptions of the Itkely costs of datlure ty obtain

ugreement ace browdly agreed to be high. The deferred

nature of the likely costs of global warming, even yt

agreed lo be inevitable, add durther difficultics ol

getting comperition

This as not intended to be totally pessimistic — just

to feinforee what common experience (if nOL common

sense) predicts ubout global pohey change. Given

scientific uncertainties, watehful caotion abour global

action ts desirable anyway. The claysic example ul

Thoms Malthus’ (8th Century prediction thatglobil

population would grow exponentially but fond

production only arcthimeveally, and the tailed Club of

Rome predictions tn the W60s, serve as cloquent

WAPNINBs aBATNSE precipitate palicy responses, On ib

more YpuMUsGe pote, ere 1s some evidence thal tree-

rider behaviour gives Way to greater conperation when

the stakes.are ugrecd tobe very high, if only tor tec

(hat failure to participate or contribute will resule im

retaliatory action,

Ta al hkehhood, the less developed world (LDC's)

will be Jess willing to hear the cost ol policy change.

and more likely 10 stay outside any agreements, even

(hough they also are more likely (o use technologies

which are ineffieet from the viewpoint of greenhouse

gas (GG) emissions One consequence of this. ts that

(he “price” the developed world, South Australia and

Australia included, has to pay tuight be not just the

costot reducing emissions oF thew own areas, but alse

substantial resource Trusters to LDCS to buy their

purtipation. Those eosts are likely te be high even

withoul Irinsfers to LDC's: it has been suyeested. fo

ckample. hit paducing areenhouse gas emissions by

the onder proposed under the Commonweulth’s O90

imerim planning. target would be cost 3% ob

Austria's GDP (IC 199).

Despite the lack of uceurate and detailed predictions

ubout Climate chanee, about which elements are due

lu human actions, and in what ways policy-makers

world-wide are likely to respond, there arc still a

number of wis in which economisis can usefully

contribute 1 the debate. For one thing, they can help

tu Wenuly climate sensitive sectors of the economy

und analyse hen clowes in them will alleen wiles

Sectors

THE ECONOMIC IMPLICATIONS OF CLIMATE CHANGT. %

Tanie | Breakdown of Sow Australian wross state product and emplovment by sensuivity to climaie change, 199U-Y!

GSP Employment

Value % of No.

Seclor $m total (000s) %

Stas ‘Toal 25,214 636

High Sensitivity 3 7

Agriculture, torestry, fishing and hunting S34 ' 43 7

Moderate sensitivity 24 i}

Elugtrienty, gas and water ASS A 5.9 !

Construction 208 & 37.1 if

Recreation. persanal and orher scrvives } 00S 4 22.8

Ownership of dwellings 2.546 9 nha

Nepligible sensitivity 73 83

Mining 755 a 3.6 |

Manufacturing 4.746 19 105.6 7

Wholesale and retail trade 3,571 14 135-8 |

Transport. storage and conmmunivation i875 s ASS 0

Public Administration and Cammiuhity Services 5.258 21 1S1.2 28

Finange, property and business services 1972 8 62.1 10

Identifying sensitive areas ancl

relevant Now-an effects

Economists have their own range of “global models”

wimed at capturing, the interactions between markets

within und across counties. As in other areas’ of

niedelling. there are well-understood limitations

associated with the use of models, For one thing, for

operaltonal reasons (iractability of solutions, etc.),

there are trade-offs between the Jevel of disaggregation

und detail, the richness of the dynamic specification

and feedback mechanisms, and the complexity of the

functional forms. use fo describe decision-making

behaviour, For another, not only are data limited, but

laboratory experimentation and repheation of

experiments are not yenerally possible, Moreover,

because of technological, climatic, political and

population chanyes amd so on, many of the parameters

of an ceonomic modg¢l alse will change over time:

depending On the yarjables in question, useful

predictions may only be made for the very short-run

(I-S years ahead).

Subject io the above reservations, however, uw is

possible both to identify qualitatively those parts of

the economy (and of other countries economies) which

ate most sensitive lo climatic change and to discuss

generally how changes in the affected sectors might

flow-o0 through the economy. as a whole, A siniple

approach, illustrated in. ‘Table 1 is lo vonsider a

breakdown of the economic activity of a region such

as South Australia inte the output and employment

contribution of different andustry sectors, noting the

sensiuivily of cach sector to clirnatie change. (We follow

the classification of sectors. adopted by TC 199),

With more detailed climate scenarios for South

Australia. beyond those yet able to be provided, we can

further refine this Classification as well as get ito more

specifics. For example, we could track the impact on

other sectors which supply inputs ta agriculture of a

climate-induced change in agricultural output through

the use of imput-output (or, preferably, computable

general equilibnum) models. It (5 important to note that

the relevant analysis must ke into account the global

economy. To illustrate this, while agriculture may be

hurt. (it may also become mere productive) in Australia,

becuuse it is hurt elsewhere such as China, India or the

USA, Australian exports may become more valuable and

Austrahians beeome better off, even though total

Australian production is reduced. By the same token,

however, if other countnes become poorer, the value

of our exports could diminish. Sorting out these issues

elearly is vital to understanding the consequences. of

climate change for us, and globally (since we assume

South Australians have the well-being of otters as part

of their concerns).

The need to take into account the global international

(ride interactions has already been the subject of analysis

and research by ctonomists in North America and

Europe. At the most recent mectings of the prestigious

American Economic Association, an entire session was

given aver to Greenhouse Warming considerations. At

this session. Reilly & Hohmann (1993), explored the

agricultural effect of climatic change and the worldwide

trade implications of a range of scenarios. Despite the

global nature of the approach adopted, the authors’ work

contributes significantly to the identifying of key impacts

and flow-on effects for South Australia.

The analysis by Reilly & Hohmann, through its

analysis of the interactions of global regions (North

86 M

America. European Conununity, former Soviet Union,

northern Europe. China, Japan, Australia, Argentina,

Brazil and “Rest of World"), gives powertul Ulustration

of the complexity of climatic change elfeets upon an

individual region. Thus. Argentina was predicted to

benefit overall. even though no increase in agricultural

production was expected in that country and their

domestic agricultural prices were expected to increase

(in line with international prices).

Evaluate particular scenarios

As was done by CSIRO (988 and revised 1992)

spewihe (and moderately plausible) chinate change

seenurios can be considered. The mujor characterisucs

of these scenarios for the Australian region aré set oul

in Table 2.

Using the most recent of these. a group of economics

graduates (Hutchins et al. 1993) ina project tor South

Austailin Department of Environment und Land

Management, estimated a loss of 20% in volume and

value of South Australia’s wheat production, The

conclusion regarding value, must be treated with a great

deal of caution, however, bearing in mind our earlier

discussion of the importance of taking into account

global trade interactions.

The results of Such exercises have led to two further

tentative explorations. First, there has been use of

general equilibrium economic models, such as

ORANI, to examine the consequences of teduced

wheat yields, increased forestry and fishery

productivity and so on. Second, there haye been

evaluations of some of the more extreme sources of

costs due to flooding, droughts and cyclones (an

eviluation of costs, of course, opens upon

considerations of costs of “abatement’).

B, BURNS. & C. WALSH

The first af these approaches, embedding the

predicted etfects of climatic change into supply shocks

in the agricultural sector of (national aod global)

economic models, has been widely used. Comparison

of the results GF such different research programs is

difficult. however, not least because the continuing

uncertainty’ about the direction and magnitude of

climatic change has resulted in quite different scenarios

heing considered froni ane study to another.

As an example, the previously cited work of Reilly

& Hohmann draws on the climate change forecasts

summarised in Houghton et al, 1990, The weneral

scenario. which involyes temperature increases of 2°C'

in the ipics and in the range 4°C to 12°C for the polar

regions. has agricultural productivity increasing iathe

North of the (former) Soviet Union, Canada anil

Europe, ‘buf reducing in the United States und most

of the rest of Europe due to drought (see Tobey e7 ul

1992).

In addition 10 incorporating economic feedback

mechanisms and imternational trade linkages, some

attempt has also been made to extend the interface

beiween the economy and the environment hy

embodying possible agricultural sector responses,

Questions of “idaptation potential” and “adaptation

capability” are considered to bear. respectively, upon

crop substitutions that ave poteutially available and

desirable due to climatic change, and upon the

constraints to these substinitions dae to poorly

developed tharkely lor crop inputs and a Tange of

Infrastructure considerations. including the skill level

of the agricultural labour force.

TaRLe 2, CSIRO climate change sconarins [987 anid 1992. Scenaries fiir xear 2030, Sources: Pearman (1988) une Climaiy

Chanue Group (1992).

JORT

1992

TEMPERATURE —210 4°

RAINFALL —50%% in north

-Z0% to south

RAINTALL INTENSITY 20-30% Meteuse

SEA LEVEL. +20 ty 40cm

TROPICAL CYCLONES

Extend 200-400 kia further south

+0).5 to 259C

Suminec O tar «209

Winter SA 0 to 104

St Alist -10 to 110%

General increase

—S ie 35 em

Uncertain

Frequeney increyxe 30-60%

SNOW LINE

WIND SPEEDS

Rise LOU m per L°C warming

Decrease 20% North of 3698

Increase south of 36°S

Rise 100 m per IC warming

Strengthening of monsoon westerlies

in north Aust. & the SE trade winds

in summer.

THE ECONOMIC IMPLICATIONS OF OLIMATE CRANGE Kr

Insights tote the occurrence of

snthrapopene cluanpes

None of the anilyses vonsidened so far distinguishes

between changes ue te human achons andl changes

which have other causes. Ib appears to be true that,

by vecivent Or otherwise, human impacts on climile,

svt ur Gind for some time ahead), may have made things

better tor humans than would otherwise huyve heen the

chse, There is also the face that, to some extent, the

living workd adapts lo new condilions.

‘These possibilities apart, ibis well understoral hy

ceoroviists Hhat malividiiels will sometimes (pack on

their environment in way thal reduces the common

good. through, for esanple:

* externalities (spillover efteces|;

*analeyualely defined’ property rights

* transaction casts

* Gailure lo reciunise “exisience values” (e.g. of ile

on earthy.

Torsuch cases, econdmists alsa ire able to presenhe

the appropriate juventive struwture tu redirect human

yetians in the Cormmon jnterest.

ixternalines occur where the sctiins ob one

decision-muker impose costs (ar bestow benefits) upon

other individuals. Problems aeeur when the relevant

decision-maker is not required to take into acemuint

these external costs fur benelity), sa thal an

inappropriate level of an activity 15 undertaken. As an

cxuniple, if nunulacrurer, in ihe course of generating

income, may Use & process that involves signifieant GG

cussions. Suppose that these emissions lead (oret

lime} to a climatic change which reduces globsl

abriculturel productivity and results in a range-of price

increases of basic commodities. Because the

munulacturer’s decision whether to produce ar not lows

not have to take into account the (external) costs

impased upon others. w situation can arise where the

valle of a tmuiufacturer’s oulpul, over and abowe uie

(inihernal) costs oF produchiod, is less than the oosis

imposed upon the rest of the community through the

impucts upon agricalture A tax on emissions would

eneyurage the manufacturer ty behave in a soceally

preferrecdl way,

Whether the presence ‘of externalities is a major

peoblent in practice Frequently depends upon how well

defined are the property tights of a particular resource,

Ownership of resources such as air, water and other

facets Of the natural enVironinent. for a varlety of

Tewsons, Often is poorly defined, TF itis nol defined

in law who-owns.a particular resouree. then consumers

af the resource are without legal recourse when the

quality and quantity of the resaurce is reduced due tn

the actions of others With regard to the earlier

caample, agricultural producers are Hot UL postion to

sue the Mmadufietuber(s) for a loss ih productieity

brought about by inereaset CMF ernissany.

It might seem that externality problems can be

handled by the sunple aet of cusuring tue che

ownership of resourees (such as air and water, ¢te,)

is tilly detined under the law, Leaving aside tor the

moment obvious Jegal complexities, if, for example,

the agricultural sector was given property rights over

the wimesphere. that sector could (im theery) sue fhe

manufacturer for the costs duc to GG emissions: Even

ifthe manufacturer was granted propercy rights aver

ihe atmosphere. then (in theory) a better wufoome could

be obtained as the aycicultural producers could decade

10 pery the manutacturer to decrease production orw

vest ina cleaner pricess. (This may scent “untyir’,

bur it Would result in more soctally efficient outeomes).

It is. perhaps. useful to remensber here that

economists da not prescribe zero emission of GG's or,

in general, zero levels of any other form of pollution’,

Mast unthropovuenre uclivitres ire “polluting” or have

externalilies (0 some degree and economists see the

problem as hiulancing (an the margin) the full costs

(including externalities: of an activity with the benefits

of thalLactivity, The assignment of property rights ean

fo sume Way by secing that decisions about activity

levels uke some qecount of externilitics,

This question of identifying and juking appropriate

wecuuOt all releviinl Costs ts central co Ure eeumomtist’s

trade. It alse leads directly to a funher complication

in the treatment of esternality producing activilics.

Even if property nghts could be cleurly defined, there

may he substantial transactions casts invelved an

abtaining the desired outcome. With regard to the

emissions of GG's, which manufacturers anc

responsible for which proportions of the emissions.

and how much are individual agneultiral producers

affected (and by whom)? At the very least, there could

be very substintial monitormye costs involved, and these

would be “compounded” by further legal and

administrative costs. Asa result. the costs of identifying

and achieving appropriate levels oMGG emissions tay

be prealer than any benelils oblained, Koonomists must

design policies that take into account the costs of

adnumistering the policy. Ibis for tls reason that a range

of other policy-measures are often considered, such

us “polluhion” taxes or deterrent lines. The proposed

hydro-carbon (ax, considered in more detail below. is

one such example.

Consideration of factors such us these give insight

as to why ordinary decent citizens of planet Earth may

tke decisions that are rational for thete individwal

stlves, but contrary to the global good, Understanding

these factors also is necessary for policy design, but

unlortunately may not he sutfivrent. Poley pardneters

can only be set on the husis of agreed valuation’s of

current and future states of the world. Such estimates

Must include Valaahonis of quite complex items, such

ws environmental resources orever "Ile an earth) The

ui fficullies fiere include the whsence of, on irrelevance

ne MoE

wy a simple market value and the question of how the

future should be valued eclutive to the present time (the

chaive of i discount rate).

The difficulties in evaluating assets such asx the Great

Barrier Reel or Anturedca are well understood The

orivate duction of a conmianal good will always Enere

the wilies held by tidividuals who wil be excluded

Ini use et the assets the imputtion OF value on the

basis ul vurrent use of a resource imores both ie

‘option values” of those who would value the

OPPOTLUTy 1G possibly tse a fesduree in future and

the “existence wilues” fell by those, Jor example. whe

Will never Use a resume Hany Way bur who simply

ville ils presenes und preservation

Analysis of alternative policy measures

Instatis Hibo (he peusors Why HOTA actions zine

likely to tead to undesinthle levels of greenhouse wats

emissary and.assocuted clintatie chune. such as those

discussed above. enable veanoimists to preseribe

inventive stroviures whieh Will redindet huinan detions

toward the common interest, There ure obvious

ivforminonal Obstacles io lhe determination oF such

iecntive strugtures, ws cun be illustrated by brief

consideration Of an ideal” data sinuatio),

Clearly. it would: be useful iit was Kitown exautly:

how che emissions associated with different production

Pricesses und GOMsUMpHON aelivites UAPHeted UPON

future production and consumption. the “tasie™

parameters of the population Thal desembe the

relationships hefween consumer choiwe and prices:

dnd. other faclurs, such as the structures of the diferent

Inarkels nthe ecosomy and the policy abjectives al

damestic und foreign governments, Ef all of this wats

Kaown, thes, 6.01 any other closed equilibrium system

whose descriplive parameters are Kiowa. a optinil

intervention strutegy could be concepluilised und

derived