Climate Change Impacts on Biodiversity in Australia · Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002 1 The workshop A workshop held in

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002 i

Logo

Climate change impacts on biodiversity in Australia

Outcomes of a workshop sponsored by the Biological

Diversity Advisory Committee, 1–2 October 2002

Edited by Mark Howden1, Lesley Hughes2, Michael Dunlop1, Imogen Zethoven3, David Hilbert4 and Chris Chilcott5

1 CSIRO Sustainable Ecosystems, GPO Box 284, Canberra, ACT 2601

2 Department of Biological Sciences, Macquarie University, North Ryde, NSW 2109

3 World Wide Fund for Nature, PO Box 710, Spring Hill, Qld 4004

4 CSIRO Sustainable Ecosystems, Tropical Forest Research Centre, Atherton, Qld 4883

5 Climate Impacts and Natural Resources, Queensland Department of Natural Resources and Mines, 80 Meiers Rd, Indooroopilly, Qld 4068

June 2003 CSIRO Sustainable Ecosystems, Canberra

Supported by Environment AustraliaBIOLOGICALDIVERSITYADVISORYCOMMITTEE

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002ii

Climate change impacts on biodiversity in Australia: Outcomes of a workshopsponsored by the Biological Diversity Advisory Committee, 1–2 October 2002

Editors: Mark Howden, Lesley Hughes, Michael Dunlop, Imogen Zethoven,David Hilbert and Chris Chilcott

For bibliographic purposes this report may be cited as:

Mark Howden, Lesley Hughes, Michael Dunlop, Imogen Zethoven, David Hilbert and Chris Chilcott (2003):Climate Change Impacts On Biodiversity In Australia, Outcomes of a workshop sponsored by the BiologicalDiversity Advisory Committee, 1–2 October 2002, Commonwealth of Australia, Canberra.

© Commonwealth of Australia (2003)

This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may bereproduced by any process without prior written permission from the Commonwealth, available fromEnvironment Australia. Requests and inquiries concerning reproduction and rights should be addressed to:

Assistant SecretaryNatural Resource Management BranchEnvironment AustraliaGPO Box 787Canberra ACT 2601

The views and opinions expressed in this publication are those of the authors and do not necessarily reflectthose of the Commonwealth Government or the Minister for the Environment and Heritage. CSIRO hascollated and edited this publication under contract to the Commonwealth Government. While reasonableefforts have been made to ensure that the contents of this publication are factually correct, theCommonwealth and CSIRO do not accept responsibility for the accuracy or completeness of the contents,and shall not be liable for any loss or damage that may be occasioned directly or indirectly through the useof, or reliance on, the contents of this publication.

ISBN: 0 9580845 6 4

This report is available at:

Website of the Biological Diversity Advisory Committee (BDAC)http://www.ea.gov.au/biodiversity/science/bdac/index.html

and

Environment Australia, Community Information UnitJohn Gorton BuildingPO Box 787, CANBERRA ACT 2601Telephone 1800 803 772Facsimile (02) 6274 1970

Cover photographs: O. Hoegh-Guldberg, S.E. Williams, CSIRO Sustainable Ecosystems

Cover and title page: Design One Solutions, Hall, ACT

Science editing, indexing, layout: Science Text Processors Canberra

Printing and binding: Union Offset Co Pty Ltd, Fyshwick, ACT

Printed in Australia on recycled paper

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002 iii

Climate change is happening—the extensive andcritically accepted reports of the Inter-governmental Panel on Climate Change (IPCC)have removed any remaining doubt.

One particularly visible sight that brings climatechange to the attention of the public is the seriesof large tracts of bleached coral on the GreatBarrier Reef resulting from unusually warmsummers. The Great Barrier Reef is anAustralian icon, symbolic of the wealth ofAustralia’s biological diversity. It is the nurseryto fishing industries and is visited by thousandsof tourists each year, bringing hundreds ofmillions of dollars into our community. Whilethe high visibility of coral bleaching has madeus aware of this impact of climate change weshould be equally concerned about the manyother, less obvious but perhaps equally damagingimpacts of climate change on biological assetsand ecosystem processes. Are there more subtleresponses hidden in our Gondwanan rainforestsand extensive rangelands that we are failing tosee? Will these responses only become apparentwhen it is too late to act and we have alreadylost ecosystem services and biological assetsforever?

Although the Earth has gone through episodesof climate change in the past, this is the firsttime such major changes can be attributed tohuman-induced causes. The AustralianTerrestrial Biodiversity Assessment 2002 andState of Environment reports have shown thatas a direct consequence of continuing humanactivities, many of our natural systems arealready under severe stress. There is no doubtthat we are facing real and serious threats to ourbiodiversity. Our modification of the landscapehas reduced habitat. At the same time, introduced

species and diseases threaten many species inthe restricted and modified habitats that remain.These threats are likely to be even moredamaging to native biodiversity under theinfluence of changing climates. We must respondwith adequate and timely policy andmanagement action. The Biological DiversityAdvisory Committee (BDAC) is concerned thatinvestment in ameliorating known threats andrepairing current damage will be inadequate ifclimate change is not taken into considerationin future policies and management actions.

The BDAC has been established to advise theMinister on matters relating to the conservationand ecologically sustainable use of biodiversity.The Committee believes that these four questionsare important for governments:

� What are the current and future impacts ofclimate change on biodiversity?

� What can we do to buffer these climaticimpacts?

� Who needs to make the relevant policy andmanagement decisions?

� What information is out there to help withthese decisions?

To begin to answer these questions, we broughttogether climate change researchers, biodiversityresearchers and policy makers, covering a widerange of expertise.

We recognise that, while much valuable work onbiodiversity conservation is being undertaken,there are many research gaps in Australia. Thereis an enormous amount of information onbiodiversity, but very little has been put into thecontext of climate change—the required

��

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002iv

linkages have not yet been made. It is up to policymakers to describe their information needsclearly to researchers, and to help direct researchtowards filling these needs.

The resulting workshop report has collated up-to-date information on the impacts climatechange is having on a range of Australian

ecosystems and how we might go aboutmeasuring these impacts. It provides a first andvitally important contribution demonstratingwhere we should be directing our attention, bothin research and in policy. The report is a fineexample of researchers, decision-makers andtheir institutions working together.

Acknowledgements

1 Professor Possingham is also Director of the Ecology Centre andProfessor of Mathematics and Zoology at the University ofQueensland.

Prof Hugh Possingham 1

ChairBiological Diversity Advisory Committee

I would like to thank the Committee, and inparticular Imogen Zethoven, for raising andpursuing this issue. Thanks to Mark Howden,Michael Dunlop (CSIRO) and Lesley Hughes(Macquarie University) for developing thisproposal with BDAC and for their significantcontribution to the report. I would like toparticularly thank the editorial team, whichincluded the above contributors as well asDavid Hilbert and Chris Chilcott. The team putin significant hours compiling the report.Environment Australia and the Australian

Greenhouse Office have been supportivethroughout. Environment Australia also providedfunding for the workshop and publication of thereport.

Thanks to the workshop participants for theirenthusiastic involvement throughout the process.Their unpaid expert advice is acknowledged—it is a testament to the strength and breadth ofthe ecological sciences in Australia that suchworld-class experience can be brought togetheron this topic.

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002 v

��

Foreword .......................................................................................................................... iiiAcknowledgements ....................................................................................................... iv

Summary ............................................................................................................................ 1

Chapter 1. Biodiversity and climate change: workshop aims and background ............ 6(Lesley Hughes, Mark Howden and Imogen Zethoven)

Chapter 2. Climate trends and climate change scenarios ................................................ 8(Mark Howden)

Chapter 3. Summaries of workshop presentations ........................................................ 14

3.1 Recent impacts of climate change on species and ecosystems ........................ 14(Lesley Hughes)

3.2 Coral reefs, thermal limits and climate change ................................................. 17(Ove Hoegh-Guldberg)

3.3 Existing and potential impacts from changing climate: Australia’s coral reefs ...... 20(Terry Done)

3.4 Potential future impacts of climate change in Queensland rangelands ........... 22(Chris Chilcott, Steve Crimp and Greg McKeon)

3.5 The Western Australian Rangeland Monitoring Systems (WARMS) and itspotential for detecting impacts of climate change ........................................... 25(Ian Watson)

3.6 Rangeland biodiversity and climate change: key issues and challenges ......... 29(Anita Smyth)

3.7 Advancing statistical science for climate research: a perspective fromthe Indian Ocean Climate Initiative (IOCI) ....................................................... 31(Eddy Campbell)

3.8 Vegetation change related to climate change in Australian subalpineenvironments ..................................................................................................... 33(Kerry Bridle and Jamie Kirkpatrick)

3.9 Impacts of global warming on the Snowy Mountains ....................................... 35(Ken Green)

3.10 What’s the link, if any, between recent changes in distribution ofAustralian birds and greenhouse climate change? ............................................ 37(Julian Reid)

3.11 Birds and climate change.................................................................................. 40(Michael Weston)

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002vi

3.12 Palaeoecological perspectives on climate change and its impact on biodiversity .. 43(Simon Haberle)

3.13 Global change: the direct effects of the increasing concentration ofatmospheric carbon dioxide on the climate and the biota ................................. 46(Sandra Berry)

3.14 Potential global warming impacts on terrestrial ecosystems andbiodiversity of the Wet Tropics ......................................................................... 48(David W. Hilbert)

3.15 Impacts of global climate change on the rainforest vertebrates of theAustralian Wet Tropics ..................................................................................... 50(Stephen E. Williams)

3.16 Effects of greenhouse warming scenarios upon selected Victorianplants and vegetation communities ................................................................... 53(Graeme Newell, Peter Griffioen and David Cheal)

3.17 Impact of climate change on terrestrial antarctic and subantarcticbiodiversity ........................................................................................................ 55(Dana Bergstrom)

Chapter 4. Indicators of climate change .......................................................................... 58(Lesley Hughes)

Chapter 5. Modelling biodiversity and climate change ................................................. 63(Chris Chilcott, David Hilbert and Mark Howden)

Chapter 6. Policy discussion ............................................................................................. 67(Michael Dunlop and Mark Howden)

Appendix 1. Existing Commonwealth programs .............................................................. 80

Appendix 2. List of participants ............................................................................................ 83

Appendix 3. Glossary ............................................................................................................ 85

References .......................................................................................................................... 87

Index .......................................................................................................................... 97

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002 1

��

The workshop

A workshop held in Canberra on 1–2 October2002 brought together for the first timeresearchers and policymakers from all overAustralia to assess the issue of climate changein relation to biodiversity in Australia: agrowing policy concern.

The participants had expertise in the keyecosystems of concern: coral reefs, rainforests,alpine regions, rangelands, grasslands, aridzone, temperate forests, coastal marine systems,rivers, mammals and birds. Presentations weremade about these key ecosystems, outliningexisting trends in climate, as well as futureclimate change scenarios (Chapter 2), currentunderstanding of the impacts of climate changes,and likely sensitivities to future impacts (Chapters1 and 3). The workshop identified potential signsof climate change (Chapter 4), modellingapproaches that could improve both predictionsof impacts and assessments of possibleadaptation strategies (Chapter 5), and possiblepolicy responses (Chapter 6).

The workshop was part of an iterative processthat will help meet the objectives of the Councilof Australian Governments’ Impacts andAdaptation Work Plan and an action plan underthe National Objectives and Targets forBiodiversity Conservation (2001–2005). Theseplans will identify potential impacts of climatechange on Australia’s biodiversity, andmeasures to address those impacts.

Climate changes

Human activities appear to be affecting theglobal climate (see Chapter 1). Global meantemperatures have risen approximately 0.7oCsince the mid-1800s. Changes in rainfallpatterns, sea-levels, and rates of glacial retreatare being seen and found to be consistent withexpectations of ‘greenhouse’ climate change.

The eight warmest years measured globallyhave occurred in the 1990s and 2000s. Last year(2002) was the second warmest on record. The1990s was the warmest decade ever recordedwith measuring instruments, and the last 100years were the warmest of the millennium,according to instruments and other means ofestimating temperature.

The above trends also apply in Australia. In2002, Australia recorded its highest-everaverage March–November daytime maximumtemperature, with the temperature acrossAustralia 1.6°C higher than the long-termaverage and 0.8°C higher than the previousrecord. Evaporation rates were also the highestrecorded so far.

The most recent report of the IntergovernmentalPanel on Climate Change (IPCC) concludedthat there is now strong evidence for a humaninfluence on global climate, and that thesetrends will continue for the foreseeable futurebecause of continued emissions of fossil fuelsand other greenhouse gases.

Current projections predict an increase in globalaverage temperatures of between 1.5 oC and 6oCby the end of the present century (Chapter 2).To place these changes in perspective, a 1oCrise in average temperature will makeMelbourne’s climate like that currentlyexperienced at Wagga Wagga in southern NewSouth Wales (NSW), and with a 4oC or 6oC rise,Melbourne’s climate will resemble the climateat Moree in northern NSW or just north ofRoma in Queensland. Intuitively, given thedifferences in ecosystems across this transectof around 1400 km, the workshop participantsrecognised that such changes are likely to havelarge implications for Australia’s biodiversity.

The relatively modest warming experienced sofar has already had measurable impacts on thedistributions, physiology and life cycles of arange of plants and animals across the globe.

Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 20022

For example, the distributions of some speciesof birds, mammals and insects have apparentlymoved toward the poles or upwards in altitude,in response to shifting climatic zones. There isalso increasing evidence of earlier flowering andfruiting in plants, and earlier reproduction inamphibians and birds, in response to warmertemperatures. Unfortunately, few studies havebeen made in Australia because of a lack of long-term recording and little support for suchresearch to date.

Likely climate change impacts onecosystems

The workshop presentations identified potentialimpacts to a range of key Australian ecosystems(Chapter 3).

Coral reefs

The natural balance between reef-building(accretion) and reef-breakdown (dissolution)normally maintains and increases the bulkvolume and complex architecture of coral reefs.But currently the balance is under threat on twofronts: coral is dying apparently because of morefrequent and severe heat waves that stress andbleach it, and also there are detrimental changesin sea-water chemistry caused by increasingamounts of carbon dioxide (CO

2) in the

atmosphere. Evidence suggests that with evenmild warming (+2oC), tropical near-shorecommunities will change from coral-dominanceto algal-dominance. It is likely, given thisevidence, that global warming will have majorimpacts on biodiversity and will affect theecosystem services currently being supplied bycoral reefs.

Near coastal marine systems

It is possible that plankton productivity couldbecome significantly more variable in near coastalmarine systems, and that change could have flow-on effects to system ecology and productivity.

Rangelands

There is likely to be a decrease in rangelandproductivity, an increasing risk of degradation,increasing sensitivity to disturbance, a changein ecosystem function, and alteration to plant

and animal community composition. In aridzones particularly, the effects of climate changemay be hidden in the ‘noise’ caused by climatevariability, management actions and otherdisturbances, at least in the next few decades.

Alpine regions

Large reductions in snow cover are likely to leadto declines in alpine flora and fauna as a resultof changes to habitats, alterations in fire regimesand incursion of feral animals and weeds.

Temperate forests

As well as changes in vegetation compositionin temperate forests, it is likely that changes instructure, productivity and foliage quality willhave flow-on effects to other components ofbiodiversity. Additionally, likely increases in firefrequency and intensity will have also haveimpacts.

Tropical rainforests

The tropical forests of north Queensland arehighly sensitive to the range of climate changesexpected in the next two to three decades. Higherrainfall favours some rainforest types (lowland,for example) while reduced rainfall increases thearea suitable for woodlands and forestsdominated by Eucalyptus. Highland rainforestenvironments, the habitat for many of theregion’s endemic vertebrates, may decrease inarea by 50% even with only a 1oC warming. Alarger rise in average temperatures may resultin the fragmentation and then disappearance ofenvironments suitable for highland rainforest.For fauna, the evidence suggests that globalwarming will have severe effects on the long-term survival of many species. A 1°C increasein average temperatures is predicted to decreasethe bio-climatic range of endemic species byabout one-third (37%). A rise of 3.5oC willreduce bioclimatic range to an average of 11%of current area, and will completely eliminatethe current bioclimates occupied by 30 speciesof endemic vertebrates. Workshop presentationswarned that there is a real possibility of at least30 to 50 species becoming extinct this centuryand that most highland faunal species willdisappear if average temperatures increase by1–5°C.

Summary


Priorities for action

The workshop identified several areas for policydevelopment and action (Chapter 6).

Understanding and managing for climatevariability

As we have found with agriculture and othersectors, there are opportunities to manage theconservation estate effectively in spite of climatechange. Climate information, together with anunderstanding of ecosystem dynamics, isincreasingly being brought into managementdecisions (such as for fire management, cullingpermits, harvest licences, river flows, visitornumbers). The more we incorporate climateinformation into management tools now, thebetter we are likely to manage future climatechange. Experience in other sectors such asagriculture shows that the most effective way toincorporate the information is throughparticipatory research in which usefulapplications are developed in collaboration withon-ground managers, building their capacity foreffective decision-making.

Immediate preservation of components ofbiodiversity that are sensitive to climatechange

There is evidence suggesting that the rate ofclimate change will be faster than the rate atwhich most species can adapt, either bymigration or by changing their behaviour,physiology or form. Hence, one short-term goalfor management is to ensure the survival ofspecies in spite of additional threats resultingfrom climate change. A first step is to identifythreatening processes and threatened species orcommunities. If there are insufficient resourcesfor managing problem situations,straightforward guidelines and priorities will beneeded, which everyone can understand. Theguidelines should set out the risks to biodiversity,how the risks are linked to resourcing, and thereasons for particular courses of action.

Some existing programs designed to managethreatening processes may also enhance species’adaptability or resilience to impacts from climatechange, especially if implemented morevigorously. Examples include management

programs for pest animals, disease and weeds,and activities that set out to reduce the risks offire, or sediment and nutrient discharge to theGreat Barrier Reef.

Where activities such as these are inadequate toreduce risks, new activities will be needed thatexplicitly target climate change. For example,we should assess the possibilities fortranslocating species, or identify thecharacteristics of sites that can act as ‘refugia’(refuges) from climate change: perhaps specificsites, linked sites across regions, or moreintensively managed areas.

For some species, the only practicablepreservation options may be in aquaria, zoos andbotanical gardens — although issues of geneticdiversity, animal welfare and cost-effectivenessare likely to arise.

Facilitating long-term adaptation

Long-term adaptation may be achieved byspecies in environments where naturaladaptation processes (such as migration,selection, change in structure) can take placebecause there is sufficient connectivity in thelandscape and few other threatening processes.Two key components will be:

� appropriate management of both on- and off-reserve areas with high conservation value;

� a system of comprehensive, adequate andrepresentative (CAR) reserves that takes theeffects of climate change into consideration.The scale of potential climate change posessignificant challenges in achieving the CARgoal.

However, there may be some conflict betweenthese and other natural resource managementgoals. For example, well-connected landscapes(to facilitate natural adaptation) may also allowmovement of, and give refuge to, pest animalsand weeds, or provide avenues for fire (therebythreatening other species preservation goals).There may also be conflicts for revegetationteams that wish to use local provenances,because that objective may be at odds withproactive planting of ‘future climaticallyappropriate’ species or provenances to facilitatelong-term adaptation.

Summary


Monitoring and research, setting currentpriorities and adjusting them as climatechange occurs

There is very little information to guide prioritiesfor managing for climate change. Focusedresearch is needed, as well as predictions offuture impacts, if an active adaptive managementapproach is to be adopted. Key elements inecosystems of concern will need to be monitored.

� Some past approaches for predicting climatechange impacts have been based simply oncorrelations between existing speciesdistributions and climate variables (seeChapter 5). While these analyses can beuseful as a first ‘filter’, they can be a problemif used and interpreted naïvely. We needbetter prediction techniques that canincorporate the effects of increasedatmospheric CO

2 concentrations, and other

environmental constraints such as soilcharacteristics, as well as managementadaptations. Such systems have recentlybeen tested using simple models of keyecological process (for example,establishment, growth, reproduction, death)in response to climate factors, CO

2,

management and soils.

� Given the huge range of possible climatechange across Australia (from 1 to 7oC by2070) monitoring will be very important,enabling progressive adaptation to climatechanges as they occur. Monitoring activitiesshould be undertaken for clearly definedpurposes: either to make specific decisionsor to learn about the system or species ofconcern.

Several types of indicator organisms can be usedfor monitoring. Of these, ‘detectors’ and‘sentinels’ are most pertinent. Detectors arespecies occurring naturally in an area of interestthat may show measurable responses toenvironmental change, such as changes indistribution or behaviour. Sentinels are sensitiveorganisms introduced into the environment asan early-warning device. These organisms maydisplay changes in physiology, phenology anddistribution and abundance. In Chapter 4 a setof possible indicators is outlined and a set ofcriteria for evaluating the appropriateness ofthese and/or other indicators is developed.

Understanding the balance betweenmitigation and impacts or adaptation

Ideally, decisions on climate change responsesshould be based on five elements:

� costs and/or benefits of mitigation,

� costs and/or benefits of adaptation (and costsof any residual impacts if adaptation is lessthan complete), and

� the interaction between these.

Workshop participants discussed the trade-offsbetween requirements for short-term responses(mitigation) and longer-term responses(adaptation). There is likely to be a ‘see-saw’interaction between these, with large mitigationeffort likely to reduce the need for, and cost of,future adaptation and vice versa. It is not possibleat present to adequately inform policy-makerson any of the last three elements of the abovedot points. Furthermore, such an analysis shoulddeal with all sectors likely to be affected(including agriculture, forestry, transport, humanhealth, tourism), of which biodiversityconservation is only one.

Information gaps

A large number of information gaps wereidentified at the workshop. Some of the key gapswere:

� documentation of impacts that are alreadyoccurring in response to existing climatetrends;

� an understanding of the factors affecting thedistribution and abundance of species,particularly those affecting the establishmentand death phases of life-cycles, the likelymigration rates, and the identification ofmigration barriers and refugia;

� analyses of the species, ecosystems andregions most vulnerable to climate changes,including those likely to be negativelyaffected by species advantaged by thechanges (such as weeds and feral animals);

� a comprehensive assessment of adaptationoptions available, including the modificationsneeded to existing conservation planning

Summary


and practice, and the existing conservationestate;

� analyses of present and future social andeconomic costs of climate change impactson biodiversity with or without adaptations;

� an understanding of the factors determiningthe resilience and adaptive capacity ofecosystems, including the roles of habitatextent, connectivity and quality, flowregimes, disturbances and management ofmosaics;

� how to develop policy that is robust to theuncertainties and long response times ofclimate impacts and adaptations, includingassessments of trade-offs between costs ofmitigation and future costs of impacts andadaptation.

One of the recurring themes through this reportis uncertainty. We are uncertain about

1) scenarios for emissions;

2) the implications of emissions for global andregional climate change;

3) the impacts of increased CO2 concentrations

and changed climate on ecosystems and theirconstituent species; and

4) the adaptation options available, and theirassociated effectiveness and cost.

One approach to deal with such pervasiveuncertainty is to adopt a risk-management

approach, which includes assessing the climatechange impacts on biodiversity in relation to asuite of other pressures and factors (one of whichis mitigation).

However, there is a range of barriers to adoptingan integrated, risk-management approach.Adversarial institutional structures seem to be akey barrier; they stretch from the landscape levelthrough to the peak political processes (forinstance, approaches that foster either agricultureor conservation, but not ways to integrate thetwo). Lack of information is a barrier tounderstanding the nature of impacts of climatechange or elevated CO

2 on ecosystems, and lack

of information prevents full acknowledgementof the values of Australian ecosystems andspecies (even when abstracted down to only theecosystem service components).

A first step

This workshop report should be seen as only aninitial step in the iterative process needed toinform the Australian public and policymakers.It can be viewed as an input into further targetedassessment.

The workshop has identified the relatively poorand uncoordinated information base on the issueof climate change and its effects on biodiversity.Workshop participants agreed that in somerespects, this area of work is at the stage ofdevelopment that global climate science wassome 15 years ago.

Summary


Human activities appear to be affecting theglobal climate. Global mean temperatures haverisen approximately 0.7oC since the mid-1800s.Changes in rainfall patterns, sea-levels, and ratesof glacial retreat are being seen and found to beconsistent with expectations of ‘greenhouse’climate change. The 1990s was the warmestdecade ever recorded, and the last 100 years werethe warmest of the millennium, according tovarious means of measuring or estimatingtemperature. Last year (2002) was the secondwarmest on record. The most recent report ofthe Intergovernmental Panel on Climate Change(IPCC 2001a,b) concluded that there is nowstrong evidence for a human influence on globalclimate and that these trends will continue forthe foreseeable future because of continuedemissions of fossil fuels and other greenhousegases. The most up-to-date predictions indicatean increase in global average temperatures of1.5–6oC by the end of the present century. Toplace these changes in perspective, a 1oC rise inaverage temperature will make Melbourne’sclimate like that currently experienced at WaggaWagga in southern New South Wales (NSW),and with a 4oC or 6 oC rise Melbourne’s climatewill resemble the climates at Moree in northernNSW or just north of Roma in Queensland.Intuitively, it is hard to conceive that suchchanges will not have implications forAustralia’s biodiversity.

The IPCC Third Assessment Report (2001 a,b)concludes that Australia is significantlyvulnerable to the changes in temperature andrainfall predicted over the next decades to 100years. Natural resources and conservation aretwo of the key sectors identified as likely to beaffected strongly. Climate change will add to theexisting substantial pressures on these sectors.

Until recently, it was expected that the effectsof climate change on species and communitieswould not be detectable for at least anothercouple of decades. However, over the last fewyears, evidence has accumulated which indicatesthat the relatively modest warming experiencedso far has already had measurable impacts onthe distributions, physiology, and life cycles ofa range of plants and animals. For example, thedistributions of some species of birds, mammalsand insects have apparently moved toward thepoles or upwards in altitude, in response toshifting climatic zones. There is also increasingevidence of earlier flowering and fruiting inplants, and earlier reproduction in amphibiansand birds, in response to warmer temperatures.Consequently, there is growing recognition thatthese biological responses are very sensitiveindicators of climate changes, as they integrateclimate over long periods and often havethresholds over which abrupt changes occur.

The data collected so far on these changes havebeen almost exclusively from the NorthernHemisphere. The data have been made possibleby the tradition of long-term monitoring ofdistribution, abundance and life-cycles of severalgroups of organisms by the InternationalPhenological Gardens in Europe, the ButterflyMonitoring Scheme in Britain, and others. In theIPCC Third Assessment Report (IPCC 2001b),of the approximately 100 physical processes and450 species analysed for such signals, there wereno Australian studies.

��

��

1. Dept of Biological Sciences, Macquarie University,North Ryde, NSW 2109

2. CSIRO Sustainable Ecosystems, GPO Box 284,Canberra, ACT 2601

3. World Wide Fund for Nature, PO Box 710, Spring Hill,Qld 4004

Lesley Hughes1, Mark Howden2 and Imogen Zethoven3


How will species respond to futureclimate change?

The effects of future climate change on speciescould be profound, as indicated by the earlierexample of potential geographical shifts inclimate zones within Australia. Species that areunable to tolerate changed conditions withintheir current range, or that cannot migrate fastenough to keep up with moving climate zones,face eventual extinction. The most vulnerablespecies will be those with long generation times,low mobility, highly specific host relationships,small or isolated ranges, and low geneticvariation. Remnant populations within reservesand ecosystems such as alpine zones, coral reefs,polar regions and coastal wetlands are likely tobe particularly vulnerable.

Considering that temperature and other climatictrends in Australia appear to be consistent withthose happening globally, is there any evidencethat the early impacts on species in the NorthernHemisphere are also occurring in Australia?

We cannot yet answer this important questionwith any degree of confidence, because fewlong-term data-sets exist in Australia. No co-ordinated monitoring of sensitive species hasbeen undertaken, and analyses that demonstrateexisting change have not yet taken place forthose species for which we have the data. Weurgently need to identify vulnerable plants,animals and ecosystems that are sensitive toclimate change, in order to inform policy-makersabout climate change in relation to conservationmanagement. The need for work on this issuewas the first research priority listed in the IPCCThird Assessment report chapter on Australasia(page 629, IPCC 2001b). The work couldeventually tie into the development of effectiveindicators of climate change. Ideally, suchindicators would already have some historicaldata available to serve as a baseline againstwhich future change can be compared. Ongoingmonitoring of these indicators would allow us

to ‘calibrate’ the rate of ecological changeagainst climate trends, so we could implementadaptive management procedures. They wouldbe intended to supplement, not replace, themonitoring of climate itself.

Aims of the workshop

In response to the above issues, the BiologicalDiversity Advisory Committee supported aworkshop held on 1–2 October 2002 inCanberra. The aims of the workshop were to:

� assess whether there is evidence for existingimpacts from the climate change that hasalready occurred;

� compile knowledge about climate sensitivities(and thresholds if they exist) of species andcommunities;

� identify key characteristics of species,communities and ecosystems that will resultin them being highly sensitive to climatechanges;

� start to assess the state of data-sets that couldbe useful for analyses of climate impacts;

� assess the analysis and modelling approachesthat could be used to identify past and futureimpacts of climate change.

The workshop was part of an iterative processthat is helping meet the objectives of the Councilof Australian Governments’ Impacts andAdaptation Work Plan and an action plan underthe National Objectives and Targets forBiodiversity Conservation (2001–2005)(Environment Australia 2001). These plans willidentify the potential impacts of climate changeon Australia’s biodiversity, and initiate measuresto address the impacts. Hence, the workshop alsoaddressed the policy-making environment anddeveloped suggestions for attention in the policy-making processes.

1. Background and aims


Are changes happening now?

The IPCC Third Assessment Report (2001b)documents trends in a large number of aspectsof climate and related factors. Primary amongstthese is the continued increase in the concentrationof atmospheric carbon dioxide (CO

2). Levels

have increased from 280 ppm, during the pre-industrial era, to 371 ppm today. Post-1800s CO

2

concentrations grew almost exponentially(Figure 2.1), with the concentrations of othergreenhouse gases such as methane and nitrousoxide showing similar growth curves.

The increase in greenhouse gas concentrationshas contributed to the increases in temperatureexperienced over the past 50 years, althoughthere are other factors contributing totemperature rise, such as volcanic eruptions,variations in solar output and changes in aerosolconcentrations. The measurements of increasedsurface temperature come from thermometers,tree ring data, ocean temperature soundings,coral data, borehole temperatures and satellitemeasurements of tropospheric temperature.Factors such as the influence of ‘urban-heatislands’ have been removed from the signals.Previous apparent conflicts between the satellitedata and surface temperature records have beenreconciled, by re-analysis which has allowed fororbital decays (Wentz and Schabel 1998), as wellas by calibration factors on successive satellites(Prabhakara et al. 1998), and by separation ofthe temperature signals into those arising fromthe lower atmosphere (troposphere) and thosearising from the upper atmosphere(stratosphere). The satellite tropospherictemperature record is now consistent with thesurface record (IPCC 2001a).

Consistent with global trends, Australia’scontinental average temperature increased 0.7ºCfrom 1910 to 1999 as measured by high-qualitysurface temperature records (Plummer et al.1995, Torok and Nicholls 1996), with most ofthe increase recorded since 1950. These resultsare supported by data from dendrochronological(tree ring) studies (Cook et al. 1991). The year1998 was Australia’s warmest year on record,and the 1990s the warmest decade (Skirving etal. 2002). Annual minimum temperatures haveincreased by 0.85ºC per century and maximumtemperatures by 0.39ºC per century (Wright etal. 1996). These increases have not been spatiallyor temporally uniform, with minimumtemperatures increasing more in the north-eastpart of the continent and greater increasesoccurring in autumn, particularly in May(McKeon and Howden 1993, McKeon et al.1998). Part of this autumn warming in the north-east may be related to recorded increases in sea-surface temperatures in the Coral Sea (Salingeret al. 1996, Zheng et al. 1997). Upper oceantemperatures have also been increasing globally.Changes from 1955 to 1996 were consistent withthe increase in net radiative forcing arising fromincreased concentrations of both greenhousegases and aerosols (White et al. 1998), althoughlong-term natural variability may also beinfluencing temperatures (Latif et al. 1997).

The increased minimum temperatures recordedin Australia may be partly related to a 5%increase in cloud cover since 1910 (Jones 1991).This trend is similar to that recorded globally(McGuffie and Henderson-Sellers 1994). Thechange in cloudiness appears to have resulted

��

��

1 CSIRO Sustainable Ecosystems, Gungahlin,GPO Box 284, Canberra, ACT, 2601.

Mark Howden1


in significant decreases in solar radiation in bothcoastal and inland Queensland, in autumn andwinter, and increases in vapour pressure in May(McKeon et al. 1998).

The trend to warmer temperatures appears tohave already reduced frost frequency andduration (Stone et al. 1996) resulting in anincrease in Australian wheat yields (Nicholls1997) and the frequency of heat stress inlivestock (Howden and Turnpenny 1997).Higher temperatures have also reduced snowcover (Green and Pickering 2002).

Global and regional air temperature increasesare closely linked to reductions in global glacialmass (Dyurgerov and Meier 1997b, Haeberli andBeniston 1998), although in some regionsincreased precipitation has increased glacialmass (Dowdeswell et al. 1997). Increased airtemperatures during the period 1961–1990 havebeen associated with a marked reduction inglobal glacial mass, which has contributed anaverage of 0.25 mm/year to sea-level rise (14 to18% of the 100-year average sea-level rise). Therate of glacial shrinkage has increased significantlysince the mid-1980s (Dyurgerov and Meier1997a,b) with European glaciers losing 10–20%of their mass in this period (Haerberli andBeniston 1998). Glaciers in New Zealand are alsoshowing substantial retreat (de Freitas 1988),whilst in the Antarctic there have been changes

in ice flow rates (Bindschadler and Vornberger1998) and retreats and break-up of large ice-shelveswhich are close to the climatic limit (Doake andVaughan 1991, Skvarca 1993, Vaughan andLachlan-Cope 1996). These changes in ice shelfarea and dynamics have been linked (Doake andVaughan 1991) to the significant warming thathas occurred at the edge of Antarctica and inthe surrounding ocean (Jones 1995). There hasalso been a marked decline in the extent ofantarctic sea ice, with a southerly migration (2.8degrees of latitude) of the summer sea iceboundary between the mid-1950s and the 1970s.There is some suggestion that this has alreadyaffected antarctic marine productivity (de la Mare1997). Increasing sea-surface temperatures are alsoimplicated in a general increase in wave energyand wave heights (Grevemeyer et al. 2000).

As well as temperature changes, there have beenconsistent trends in precipitation over manyregions of the world, with some increasing andsome decreasing. There have also been trendstowards increased intensity of precipitation(IPCC 2001a).

These trends are also present in Australia, withstudies documenting an increase in both heavyrain events and average rainfall over large areasof Australia from 1910 to 1990 (Suppiah andHennessy 1996, 1997, Lavery et al. 1997). Thelargest increases were along the east coast,

Figure 2.1. Atmospheric CO2 levels derived frommeasurements in ice cores (Etheridge and Wookey1989, Etheridge et al. 1996, 1998, Morgan et al. 1997)and direct measurement from Mauna Loa (Keeling andWhorf 2002)

250

270

290

310

330

350

370

390

1000 1200 1400 1600 1800 2000

Year

CO

2concentr

ation (

ppm

)

Atmospheric CO2 concentrations (ppm)

1006 to 2002 AD

2. Trends and scenarios


particularly in NSW, but decreases occurred inthe south-west of Western Australia and inlandQueensland over this period. In the summer half-year, the all-Australian average rainfall (basedon area-weighted station data) increased by 14%,heavy rainfall increased by 10–20%, and thenumber of dry days decreased by 4% (Suppiahand Hennessy 1996, 1997). In the winter half-year, the changes were about half these figures.The trends in heavy rain events are partially butnot totally explained by ENSO (El NiñoSouthern Oscillation) fluctuations over recentdecades (Suppiah and Hennessy 1996, 1997).The changes in rainfall are also reflected in thenumber of wet days. There are up to 20% morewet days now in parts of NSW and the NorthernTerritory, but fewer by about 10% in WesternAustralia (WA) (Hennessy et al. 1999).However, a significant part of the trends (bothpositive and negative) consists of an almost step-wise change around the mid-1970s. A similar‘state-change’ in climate and oceanic systemsoccurred almost globally at this time.

The possible relationship between ENSO andrainfall amount has altered since the 1970s, withhigher rainfalls for any value of the SOI(Southern Oscillation Index) than would havepreviously been expected (Nicholls et al. 1996).Furthermore, it appears that the ENSO systemitself is changing (Nicholls et al. 1997) with anapparent functional change in the mid-1970s(Graham 1994) to increased frequency of ElNiño events and fewer La Niña events. Thesehave been assessed in the context of the historicalrecord as being very unusual and unlikely to beaccounted for solely by natural variability(Trenberth and Hoar 1996, 1997). Others, suchas Mantua et al. (1997) and Allan and D’Arrigo(1998), have a conflicting view that the recentpersistent El Niño events could be part of long-term natural variability. Ongoing research ondecadal climatic fluctuations (Power et al. 1999)should eventually allow us to more effectivelydifferentiate any climate change signal fromother, natural influences.

Tropical cyclones contribute a large part of thesummer rainfall in many tropical areas, but thenumber of tropical cyclones observed in theAustralian region has declined since the start ofreliable satellite observation (1969–1970),

reflecting the increased incidence of El Niñoevents (Nicholls et al. 1998).

Sea-level

Global sea-level has risen by 10–25 cm over thepast 100 years (Peltier and Tushingham 1989,Barnett 1988 cited in Quayle and Karl 1996) withtrends of 1.7 to 2.7 mm/year for Australia and NewZealand (Salinger et al. 1996), similar to the globalestimated rate of 1.8 mm/year (Douglas 1991).However, more recent analyses suggest thatAustralian sea-level rises, and those of theSouthern Hemisphere in general, are less than theglobal average (Mitchell et al. 2000). This rise insea-level is a combination of thermal expansiondue to observed temperature increases of the oceans(Church et al. 1991), glacial melting (Dyurgerovand Meier 1997a,b), groundwater extraction(Sahagian et al. 1994) and an uncertaincomponent from changes in the Antarctic andGreenland ice caps (Zuo and Oerlemans 1997).Preliminary analyses of data from new satellite-based sensors indicate sea-level rises of about 4mm/year in the last few years, perhaps linked toENSO behaviour (Nerem 1995).

UV-B

The short wavelength portion of the ultravioletrange (UV-B; 290–320 nm) is damaging to bothplants and animals (Rozema et al. 1997) and isstrongly absorbed by ozone. Emission ofchlorinated fluorocarbons and similarcompounds has led to reductions in stratosphericozone levels particularly around the poles inwinter (Farman et al. 1985, Rex et al. 1997).One result has been an increased intensity of UV-B radiation at the Earth’s surface, with theAntarctic ‘ozone hole’ being of record size inthe year 2000 (Watkins 2001). The increases inUV-B have been significant at latitudes greaterthan 35ºS, and range from about 7% per decadefor Tasmanian latitudes to 3% per decade forlatitudes in northern Australia (Herman et al.1996). Similar trends of increasing UV-B arefound in data from New Zealand (Basher et al.1994) and other regions (Blumthaler andAmbach 1990). In polluted areas, aerosols andhigh concentrations of tropospheric ozone tendto absorb UV-B radiation and reduce its intensityat the ground surface (Bruhl and Crutzen 1989).



Shindell et al. (1998) suggest a link between theobserved increasing concentrations ofgreenhouse gases and the decreases in polarstratospheric ozone. The mechanism they exploredthrough global climate models is that higherconcentrations of greenhouse gases increase thestability of the stratosphere and the polarvortices, resulting in a significantly colder lowerstratosphere and thus greater ozone loss. Furtherresearch is required to explore the relationshipsbetween these two trends (Salawitch 1998).

Climate change scenarios

The IPCC has released a report describingscenarios for greenhouse gas emissions (IPCC2000). The scenarios are based on internally-consistent and plausible projections of futurepopulation, economic growth, energy technologyand political situations. They span a range ofpossibilities, from rapid growth in total energy useto smaller increases in emissions early on, whichplateau then fall later in the century (Figure 2.2).In all cases, however, CO

2 concentrations in the

atmosphere are likely to increase over the next100 years, and hence some level of climatechange seems inevitable. A further point to noteis the broad span of possible outcomes—thisinherently expands the uncertainty in anyclimate-change scenario.

Even the most conservative of the scenarios inFigure 2.2b will result in concentrations farexceeding any experienced over the past 420 000years and possibly the past 20 million years.Implicitly, this means that nothing in ourexperience as a species provides a precedent forthis sort of change (Homo sapiens sapienshaving been in existence for 90 000 to 130 000years). This increase in CO

2 concentrations is

the most certain of potential changes (that is, it isthe primary cause of the climate changes).Increased CO

2 concentrations will have significant

direct impacts because they increase thephotosynthetic rate and water use efficiency ofplants. The outcome tends to be increasedbiomass but decreased nitrogen content inplants—with a variety of flow-on effects onherbivores. Another consequence can be achange in competitive interactions, and alteredspecies composition within communities.

CSIRO has recently updated their climate changeprojections for Australia (http://www.dar.csiro.au/publications/projections2001.pdf) andtheir likely impacts (http://www.marine.csiro.au/iawg/impacts2001.pdf). These projectionsindicate that by 2030, annual averagetemperatures will be 0.4–2.0ºC higher thanpresent. By 2070, annual average temperaturesmay increase by 1.0–6.0ºC (Figure 2.3).

Figure 2.2. Scenarios of a) CO2 emissions andb) atmospheric CO2 concentrations over the 21stcentury (IPCC 2000)

A1BA1TA1F1A2B1B2IS92a

Scenarios

(a) CO2 emissions

2000 2020 2040 2060 2080 2100

Year

25

20

15

10

5

CO

2 e

mis

sio

ns (

Gt C

/yr)

A1BA1TA1F1A2B1B2IS92a

Scenarios

(b) CO2 concentrations

2000 2020 2040 2060 2080 2100

Year

CO

2 c

oncentr

ation (

ppm

)

1300

1200

1100

1000

900

800

700

600

500

400

300



Projections for rainfall vary much more, bothspatially and seasonally (Figure 2.4). In general,the projections indicate substantial reductionsin autumn, winter and spring rainfall (the mainrainfall seasons) for much of southern Australiaand particularly the south-west of WesternAustralia. In the north of Australia, increases insummer rainfall—the main rainfall season—aresuggested in some regions. In most areas, anincrease in rainfall intensity is indicated—evenin projections where mean rainfall decreases.The projections also suggest increases inevaporation, provided that the increases intemperature remain roughly equivalent betweenday and night. If minimum temperaturescontinue to increase faster than maximumtemperatures, then the results in terms ofevaporative demand are likely to be equivocal.

Overall, the scenarios emphasise a more variableand unpredictable climate in Australia, withincreased incidence of extreme events such asfires, floods, droughts and tropical storms. River

Figure 2.3. Possible average seasonal warming ranges(oC) for years 2030 and 2070 relative to 1990 (CSIRO2001)

flows in particular are likely to be significantlyaffected, with a rule of thumb being that in humidregions there is a doubling of reduction in riverflow for a given reduction in rainfall (that is, a10% decrease in rainfall would translate to a20% decrease in river flows). In drier zones themultiplier is greater than two.

Climate is a key determinant of the location,structure and function of natural ecosystems. Wesee this every day in the landscapes in whichwe live: for example, there are progressivechanges in vegetation up a mountainside; andforest types are related to rainfall. There are alsomany examples showing that climate change, ongeological timescales, has affected thedistribution of flora and fauna and the survivaland ranges of whole ecosystems. Theseobservations lead us to predict that the climatechanges projected above will have profoundimpacts on natural ecosystems and theircomponent species.

Summer Winter

Autumn Spring

2030

2070

0 1 2 3 4 5 6 7 8

0 1 2 3 4 5 6 7 8Temperature change (°C)

Temperature change (°C)



Figure 2.4. Possible ranges of average seasonal rainfallchange (%) for around 2030 and 2070 relative to 1990(CSIRO 2001)

Summer Winter

Autumn Spring

Drier 2030 Wetter

–80 –40 0 40 80

–80 –40 0 40 80

Drier 2070 Wetter

Gidyea (Acacia georginae) in Central Australia.Climate changes are expected to have profoundimpacts on natural ecosystems and theircomponent species. (Photo: Peter Canty, courtesyof CSIRO Sustainable Ecosystems)



Increasing greenhouse gas concentrations areexpected to have significant impacts on theworld’s climate on a timescale of decades tocenturies. There is accumulating evidence fromlong-term studies that the climate trends of thepast few decades are anomalous, and that thesetrends are already affecting the physiology,geographic distributions and phenology (life-cycles) of species. Predictions about the waysin which species respond to climate change can bebroadly summarised into the following categories:

Effects on physiology: Changes in atmosphericCO

2 concentration, temperature or precipitation

will directly affect rates of metabolism anddevelopment in many animals, and processessuch as photosynthesis, respiration, growth andtissue composition in plants.

Effects on distributions: A 3oC change in meanannual temperature corresponds to a shift inisotherms of approximately 300–400 km inlatitude (in the temperate zone) or 500 m inaltitude. Species’ geographic ranges aretherefore expected to move upwards in altitudeor towards the poles in response to shiftingclimate zones, in those species capable ofmoving range relatively rapidly.

Effects on phenology: Life cycle eventstriggered by environmental cues such as degree-days may be altered, and the result may break thecoupling of life-cycle interactions between species.

Changes in the physiology, phenology anddistributions of individual species will alter theircompetitive relationships and other interactionswith other species. This will lead to changes inthe local abundance of species and to changesin the composition of communities. Inevitably,at least some species will become extinct, either

as a direct result of physiological stress, or viainteractions with other species.

Recent analyses of long-term datasets show thatsome species are already responding to theanomalous atmosphere and climate of the 20thcentury (Hughes 2000, McCarty 2001, Waltheret al. 2002, Parmesan and Yohe 2003, Root etal. 2003). Here are some examples.

(i) Changes in plant physiology, productivityand growth

� Tree-ring records in both the Northern andSouthern Hemispheres show clear, century-scale increases in the amount of wood laiddown each year. It is possible that theincrease has been caused by highertemperatures and/or CO

2 supply and

nitrogen deposition.

� Since the early 1960s, the difference betweenannual maximum and annual minimumconcentrations of atmospheric CO

2 has

increased by 20% in Hawaii and by 40% inthe Arctic. A likely explanation for this trendis that more CO

2 is being absorbed by land

plants during the growing season.

� Basal areas of trees and lianas (climbers)have been increasing in the Amazon, withlianas increasing their dominance faster.

(ii) Changes in phenology (life-cycle timing)

� Data from the International PhenologicalGarden (1959–1993) show that spring eventsnow happen 6 days earlier and autumn events

��

��

3.1 Recent impacts of climate change on species and ecosystems

Lesley Hughes1

1 Dept of Biological Sciences, Macquarie University,NSW 2109


happen 4.8 days later, with the result thatthe growing season is now 10.8 days longerthan in the late 1950s–early 1960s.

� Monitoring of life cycles at Cardedeu, Spain,since 1952 finds that leaves now unfold 16days earlier and fall 13 days later, plants fruit9 days earlier, butterflies emerge 11 daysearlier and spring migratory birds arrive 15days later.

� Twenty of 65 British bird species surveyedfrom 1971 to 1995 show statisticallysignificant tendencies to lay eggs earlier(about 9 days earlier on average). Only onespecies lays significantly later. Paralleltrends of earlier reproduction have beenreported for several British amphibians(during 1978–1994), with every 1oC increasein maximum temperatures corresponding toan advance in spawning date by 9–10 days.

� Phenological mismatches have started toaffect predator–prey and plant–herbivorerelationships, such as the interactionsbetween the oak, the winter moth and thegreat tit in Europe.

(iii) Changes in species distribution andabundance

� Ten of the 16 new heat-loving (thermophilic)plankton species that established in the NorthSea during the 20th century have beenrecorded for the first time in the 1990s.

� On many mountain ranges in western NorthAmerica, the treeline has moved upwardssince 1890, reaching maximum extentbetween 1920 and 1950. By 1992–1993,there was greater plant-species-richness(number of plant species) than is noted inhistorical records, at almost three-quarters(70%) of the sites inspected on 30 peaks inthe European Alps, possibly as a result ofupward colonisation.

� Both species of antarctic vascular plantshave dramatically increased in numbers atmany locations in maritime Antarctica as aresult of better seed germination andseedling establishment.

� Surveys of 35 non-migratory Europeanbutterflies find that the ranges of 22 species

(63%) have shifted northwards by 35–240km during the 20th century. Only two species(3%) have shifted south. For two-thirds ofthe species, the northern boundaries movednorth while the southern boundaries remainedstable, so their geographic ranges effectivelyexpanded. Northward shifts of the ranges ofEuropean birds have also been recorded.

� Malaria and dengue fever have been recentlyreported at increasingly higher altitudes inPapua New Guinea, South America, Asiaand Africa.

� Warming of the California Current (during1951–1993) has been associated withsignificant declines in populations ofzooplankton, macroalgae, reef fish andpredatory seabirds.

� Increasing sea-surface temperatures havebeen associated with a lifting cloud base inCosta Rican cloud forests and a decline inprecipitation. The changes have beenaccompanied by upward colonisation oflowland birds and sharp declines inamphibian and lizard populations.

Trends and impacts in Australia

Climate trends in Australia have been consistentwith global patterns, but few biological impactshave been recorded, largely because we lacklong-term datasets and active monitoringprograms. Nonetheless, there is evidence ofchange (see Hughes, in press), such as:

� thickening of vegetation in eucalyptwoodlands as a result of CO

2 supply (in

addition to the impacts of grazing and alteredfire regimes);

� increased establishment of snow gums insub-alpine meadows;

� intrusion of mangroves into swamps thatwere formerly freshwater, especially in theNorthern Territory;

� higher wheat yields associated with lessfrequent frosts;

� rainforests invading eucalypt woodlands;

� behavioural changes in the sleepy lizard(Tiliqua rugosa) and changes in the position

3.1 Recent impacts


of the parapatric boundary between lizardticks;

� new seabird colonies to the south of theirhistorical range;

� more frequent and intense coral bleaching as aresult of increasing sea surface temperatures.

Future monitoring of carefully selected speciesand communities in Australia will be vital if weare to understand the impacts of global change.A national monitoring scheme such as thatdeveloped in the United Kingdom by theDepartment of Environment and Transport(DETR) would be highly desirable (see page 60).

Wanderer butterfly on apoached egg daisy. Thegeographic ranges ofAustralian butterflies may beaffected by climate change.(Photo: CSIRO SustainableEcosystems)

3.1 Recent impacts

Bleached corals near GreatKeppel Island on the GreatBarrier Reef in early 2002.Large sectors of associatedreefs bleached due tounusually warm seatemperatures. Some corals,like the one shown in theforeground can turn pink orblue due to the stimulationof host pigments. Thesepigments appear to beinvolved in the stressresponses of corals. (Photo:O. Hoegh-Guldberg)

Massive bleaching occurred on the GreatBarrier Reef during early 2002 due tounusually warm sea temperatures. Inshoresites like the Keppel Islands nearRockhampton experienced almost 100%bleaching. More than 50% of the GreatBarrier Reef was damaged by bleaching in2002. (Photo: O. Hoegh-Guldberg)


3.2 Coral reefs, thermal limits andclimate change

Ove Hoegh-Guldberg1

1 Centre for Marine Studies, University of Queensland,St Lucia, Qld 4067

Coral reefs are the most spectacular and diversemarine ecosystems on the planet, and providemajor resources for at least 100 million peopleworldwide. At the heart of these highly productiveecosystems are Scleractinian corals that build theframework of coral reefs and provide a largeportion of the primary productivity associatedwith reefs. Despite their importance, coral reefsare highly threatened by human activities (waterquality, over-fishing); it is expected that at least30% of today’s coral reefs will be destroyed by2050 (Wilkinson and Buddemeier 1994). Climatechange has recently been added to these concerns,and it is now considered to be the primary threatto coral reefs over the next 50–100 years. Thispaper reviews the evidence for the impact ofclimate change on coral reefs such as Australia’sGreat Barrier Reef, and explores projections forthe future if current rates of warming continue.

Corals are a mutualistic symbiosis betweenScleractinian corals and single-celleddinoflagellate algae belonging to a number ofspecies from the genus Symbiodinium. Thesealgae give Scleractinian corals an overall browncolour. Symbiodinium photosynthesises while inthe tissues of the coral host, providing most ofthe coral’s energy requirements. The high ratesof calcification that are a key feature of coralreefs are directly due to the abundantphotosynthetic energy provided bySymbiodinium (Trench 1979).

Several conditions can lead to failure of thesymbiosis between corals and Symbiodinium, andas a result the corals turn from brown to white—an occurrence referred to as ‘coral bleaching’. Coralbleaching is a response to environmental stress,and can be triggered by a range of stressfulconditions including abrupt changes to salinity,

light and temperature (Hoegh-Guldberg 1999).While it may occur at a local scale (affectingperhaps 100 m2 on a coral reef), episodes of‘mass’ coral bleaching have begun to affectthousands of square kilometres of coral reefs.These events appear to be increasing in severityand frequency, and are unknown in the scientificliterature prior to 1979. Coral bleaching events havegreat significance for Australia, which harboursover a quarter of the world’s coral reefs within itswaters and some of the most spectacular examplesof coral reef biodiversity (for instance, in the GreatBarrier Reef and on the North West Shelf). Tworecent mass bleaching events (in 1998 and 2002)

Concern for biodiversity in the faceof increased coral bleaching andmortality has increased as studiesshow the disappearance of coral-dependent organisms like thebeautiful Orange-spotted Filefish,Oxymonacanthus longirostris.(Photo: O. Hoegh-Guldberg)


affected over 50% of all coral reefs within theGreat Barrier Reef Marine Park.

Mass coral bleaching events are caused byincreased water temperature, which results fromdisturbances to the pattern of water movementand the development of hot and still conditions.Satellite mapping of sea-surface temperatureshas revealed that temperatures as little as 0.8oCabove the long-term summer maxima will leadto mass bleaching. In these cases, coral reefs mayrecover. If sea temperatures climb to as much as2–3oC above long-term summer maxima, or seatemperatures remain higher for longer, coralsmay not recover from bleaching. In cases, massmortality events have occurred. In the extremelywarm conditions of the 1998 El Niño SouthernOscillation (ENSO) event, it is estimated that 16%of the world’s coral reefs died. In some areas, suchas the Western Indian Ocean, as much as 48% ofliving coral cover was removed (GCRMN 2000).

Current trends in sea temperature have scientistsand managers concerned about how thefrequency and severity of coral bleaching willchange. Tropical sea temperatures are nowapproximately 0.6oC higher than they were atthe beginning of the last century (Lough 2000).How conditions for coral reefs will change hasbeen modelled using the output of generalcirculation models from three of the world’smajor climate centres. These analyses all tell thesame story. The thermal limits currently seen oncoral reefs will be exceeded on an annual basisby 2030–2050, even under best-case climateprojections (Hoegh-Guldberg 1999). Perhaps ofgreatest concern is that levels of thermal stressare set to rise to more than 4–6 times the levelsof stress known to have caused major mortalityevents in the past (Hoegh-Guldberg 2001). Theseconditions suggest that Scleractinian corals arelikely to become remnant organisms on reefs thatare currently dominated by these importantorganisms today.

How will reef systems like Australia’s GreatBarrier Reef respond to increased rates of coralmortality? Several people have suggested thatcorals and their symbionts will adapt despite therapid changes to environmental conditionsaround them. There is some concern about thesesimplistic scenarios. Corals are slow-growingasexual organisms that are unlikely to respond

fast enough relative to rates of climate change—reasoning supported by recent increases in theimpact of mass bleaching events. Spatial(cellular automaton) models built usingparameters measured from coral populations onthe central Great Barrier Reef show that evenrelatively mild changes to the mortality of coralpopulations, such as those that occurred duringthe 1998 mass bleaching event on the GreatBarrier Reef, lead to rapid declines in coralcover. Even with the assumption that oceanwarming continues at the rate of 0.1oC perdecade, with warm anomalies occurring every10 years (note: they currently are occurring every3–4 years) coral cover declines steadily to lessthan 15% after 100 years (Johnson et al. 2002).

Attempts to project the future of coral reefsunder climate change are based on the keyassumption that the thermal tolerance of reefsdoes not change over the short periods involved.Analysis of the thermal tolerance of corals andtheir symbionts reveals that tolerance varies withlatitude. This clear evidence of geneticadaptation to sea temperature has been taken asa sign that populations of corals will thereforeevolve in response to climate change. Proponentsof this idea neglect to consider three majorissues. First, corals have evolved to toleratedifferent sea temperatures but over relativelylong periods (probably hundreds and thousandsof years but not years or decades). Second,current rates of increase in sea temperature arevery high—they are increasing almostmonotonically. Third, the observation ofincreasingly frequent and severe bleachingevents over the past 20 years contrasts with thesituation expected if coral reefs were rapidlyevolving tolerance to thermal stress.

Other workers have suggested that re-mixing ofof the host and symbiont within symbiosis mayproduce additional mechanisms for rapidevolutionary change. The Adaptive BleachingHypothesis (Buddemeier and Fautin 1993), forexample, suggests that corals may bleach to takeon other strains of dinoflagellate symbionts (withhigher thermal tolerances). While a tantalisingpossibility, the Adaptive Bleaching Hypothesisremains unproven. Creation of new symbioticcombination is likely to be a complex co-evolutionary step that is unlikely to occur at

3.2 Coral reefs, thermal limits


ecological time scales. The ratio of thecomponents in multi-strain symbioses may beable to change, but that is not the same as asymbiosis that involves a new partnership.Again, the fact that coral populations areshowing signs of increasing death over the lasttwo decades suggests that bleaching does notlead to increased tolerance in coral populations.

Given these scenarios, we are left trying toproject what coral reefs will look like as coralcover is reduced. Impacts on overall biodiversitywill depend on the type of organism beingconsidered. For example, in fish populationsstudied after mass bleaching events, only thosespecies that are dependent on corals for foodshow major declines. In the Seychelles, Spaldingand Jarvis (2002) found that the overall structureof fish communities had changed very littledespite very large decreases (3–20 fold) in livingcoral cover after the 1997–1998 bleaching event.The authors noted that a few individual fishspecies did show change, as was also discoveredin a number of other studies. The Orange-spottedFilefish (Oxymonacanthus longirostris), whichdepends on coral, disappeared from Okinawanreefs rapidly after the 1998 bleaching event (Kokitaand Nakazona 2001). Similar observations havebeen made for coral-eating Chaetodon fishes.Other coral-dependent species are expected toshow major impacts. For example, over 55species of decapod crustacean are associatedwith living colonies of a single coral species,Pocillopora damicornis (Abele and Patton 1976,Black and Prince 1983). Nine of these are knownto directly depend on living pocilloporid coralcolonies. Similar relationships exist for othercoral genera. If reefs lose the coral genera andspecies that support the numerous coral-dependent organisms, many other organisms willalmost certainly become extinct.

In summary, even with warming at the ratespredicted in lower-edge scenarios put forwardby IPCC (+2oC by 2100; IPCC 2001b), it is likelythat many tropical near-shore communities willchange from coral-dominance to algal-dominance. This will almost certainly havemajor impacts on biodiversity, and change theecosystem services currently being supplied bycoral reefs. It is estimated that coral reefs containthe greatest number of marine species of anyecosystem (estimates range from 1 million to 9million species). There is likely to be anenormous loss of species as climate change setsabout transforming the basic structure of reefslike the Great Barrier Reef.

It is clear that Australia must meet thesechallenges by taking action. Existing studiesneed to be refocused so that the changes can bedescribed in greater detail. Only then can theimportant changes to the industries and socio-economic systems that depend on coral reefs bedesigned. Science must drive policy changes thatwill ensure that the devastating impacts areminimised. Reef resilience (the ability to surviveand ‘bounce back’ from stress) must become aprimary goal of reef management—only if otherstresses on reefs (such as over-exploitation andcoastal run-off) are reduced is there a chancethat reefs will have the resilience to survive theenormous increases in thermal stress beingprojected. Most importantly, major decreases ingreenhouse gas emissions are required ifecosystems like coral reefs are eventually torecover. If we do not act strongly on greenhousegas emissions over the next decade, coral-dominated ecosystems are almost certainly tobe condemned to the past.

3.2 Coral reefs, thermal limits


3.3 Existing and potential impacts from changing climate:Australia’s coral reefs

Terry Done1

In the last two decades, summers with extremetemperatures and rainfall have made significantvisual and ecological impacts on Australian coralreefs. Disturbance and population turnover incorals and associated communities are normalaspects of ecological dynamics in coral reefs,but there can be little doubt that global climatechange has made the previously infrequent andlocalised ‘coral bleaching’ disturbancecommonplace and widespread (Hoegh-Guldberg1999). ‘Coral bleaching’ is a reduction of coralpigmentation and of the symbiotic algae(zooxanthellae) that are a primary source ofnutrition to the coral host.

A spectrum of bleaching impacts can be defined.‘Sub-lethal’ coral bleaching reduces thephysiological and reproductive fitness ofaffected individuals; ‘sub-lethal’ bleaching atone reef thus has the potential to significantlyweaken that reef’s supply of coral larvae to otherreef areas. A ‘minor’ coral bleaching event killsor injures a small proportion of corals at a place,but has little immediate impact on thecomposition and diversity of the localcommunity. ‘Catastrophic’ coral bleaching cankill virtually all corals over an entire reefscape,sometimes including ‘old-growth’ individualcorals or stands of corals. Catastrophic bleachinghas poorly known but potentially serious flow-on effects on local reef architecture, biodiversityand fisheries production, and significant impacton reproductive output.

Data-sets relevant to bleaching impacts arelimited up to the 1980s, but more targeted datahave become available since the 1990s. Seatemperature data collected in situ (Lough inpress) and remotely by satellites (Skirving et al.2002) show extremely high sea-surfacetemperatures during the summers of 1997–1998

and 2001–2002. These were major potentialbleaching years on Australia’s north-west andnorth-east coasts, respectively, as has beenconfirmed by assessments of the coral reefs(www.gbrmpa.gov.au; Done et al. 2003a). Theoceano-graphic pattern of the heating anomalywas patchy in relation to the distribution of coralreefs in both years. In 1998, most of the coralsat Scott Reef (under a persistent ‘hotspot’ offNW Australia) died to a depth of 40 m (Heywardet al. 1999), whereas Ningaloo Reef and RowleyShoals, some hundreds of kilometres to thesouth, were in cooler waters and showed nosymptoms. On the Great Barrier Reef, the 1998impact was mainly coastal—possibly acombined effect of the stresses of heat and therelatively fresh waters associated with floodingrivers (Berkelmans and Oliver 1999; Sweatmanet al. 2001). In 2002, Flinders Reef in the CoralSea (under a persistent ‘hotspot’) had damagesimilar to that at Scott Reef (Done et al. 2003a).The Great Barrier Reef was subjected to acomplex mosaic of relatively hotter and coolerareas in summer 2001–2002. In the coolest areas(hundreds of square kilometres in extent), noneof the reefs surveyed (each several squarekilometres in area) had damage that could beattributed to coral bleaching. In the hottestpatches of water, there was significant coraldeath, attributed to the bleaching event. Therewere complex patterns of coral mortality andsurvival within individual coral reefs that wererelated to habitat, water depth and speciescomposition (Done et al. 2003a).

At the Great Barrier Reef, extreme conditionsof heat and rainfall are likely to become muchmore commonplace as climate changes. More

1 Australian Institute of Marine Science, PMB No. 3,Townsville MC, Queensland


extremely hot days are expected,threatening to increase the frequency andintensity of coral bleaching leading to coraldeath (Hoegh-Guldberg 2001) and causepotentially catastrophic transformations incoral communities by mid-century (Done1999; Done et al. 2003b). Cyclone frequencyseems likely to remain the same, but as eventsbecome more extreme, there may beincreasingly severe destruction of reefcommunities, and the most destructiveswathes may be broader (Pittock 1999).More extreme floods may cause coral to diefrom hypo-osmotic stress (Coles and Jokiel1992) at ever increasing distances into thetract of coral reefs (King et al. 2002). Ahigher sea level will stimulate some reef-topcommunities of organisms that are currentlylimited by sea level, but others will be smotheredas a result of redistribution of reef-top sediments(Wilkinson 1996). Increased atmospheric CO

2

concentrations will marginally lower thealkalinity of reef waters, raising the rate ofchemical dissolution of existing reef limestone,and decreasing the rate of deposition and/or thestrength of new limestone deposited by reeforganisms (Kleypas et al. 1999).

To conclude, there appears to be a two-frontedthreat to the natural balance between reef-building (coral growth, reef accretion) and reef-breakdown (coral death, reef erosion), which,over recent millennia, has produced the bulkvolume and beautiful, complex architecture oftoday’s coral reefs. The balance is being tippedboth by more frequent and severe heat wavesthat stress, bleach and kill coral reef builders,and also by detrimental changes in seawaterchemistry caused by increasing amounts of CO

2

in the atmosphere. The key uncertainties in thisscenario are:

� the patchiness in both the environmentalstresses and the ecosystem’s vulnerability tothose stresses; and

� the capacity of coral reefs to adjust to achanged environment without permanentloss of their essential attributes of beauty,biodiversity and productivity so valued bypeople.

In other words:

� Are there less vulnerable areas, and whereare they?

� How fast can coral reefs adjust, relative toclimate change, and how will it be manifestat all levels throughout the coral reefecosystem, from individuals to localcommunities and regional reef systems?

These questions are all areas of active research.

An area of bleached corals. Both the white and thepastel coloured corals are ‘bleached’ as aconsequence of loss of most of the microscopicsymbiotic algae that normally live by the millions inevery square centimetre of corals, and give themmuch darker tonings. This loss of algae is aresponse to heat stress and causes physiologicalstress in the corals. (Photo: T.J. Done)

3.3 Impacts on coral reefs


Climate variability has been identified as themost important driver of change in Queensland’snative and naturalised pastures (collectivelyknown as the ‘rangelands’). The failure toadequately manage for climate variability in therangelands has been a major source of reducedresource condition and degradation. Recentreports by McKeon and Hall (2002) demonstratehow processes of degradation and recoveryinteract with climate variability. In particular, acombination of heavy grazing pressure anddrought during the normal growing season hasbeen shown to accelerate loss of desirableperennial species, leading to increased soil loss,reduced burning opportunities and woody weedinvasion.

Climate change simulations indicate thatenhanced seasonal climate variability andextremes can be expected. They will interactwith current maladapted management practicesto amplify the risks of resource degradation, andmost likely will lead to long-term reductions inresource condition. There are several key factorsadding to the risks for biodiversity from climatechange: for example, it is likely that conditionswill be much warmer and drier and that therewill be increased disturbance from tropicalcyclones.

By 2030, annual mean temperatures overQueensland are expected to have increased by0.3 to 2.0ºC compared to 1990 values. Muchlarger ranges are projected for 2070, withincreases from 0.8 to 6.0 degrees. For example,in Brisbane, the annual mean temperature was20ºC in 1990, but it may increase to 20.3–22ºCby 2030 or 20.8–26ºC by 2070. Thesetemperature changes, in combination with theconsensus scenario predictions of little changein annual rainfall, imply a general decrease in

soil moisture over Queensland. The decrease isprojected to be most pronounced in the farinterior.

There is an emerging consensus that maximumtropical cyclone wind speeds are likely to haveincreased by 5–10% by some time after 2050.This will be accompanied by increases of 20–30% in peak precipitation rates during tropicalcyclones. Cyclonic disturbances are importantin the dynamics of a variety of ecosystems:rainforests, inland streams, and the rangelands,where recruitment of both desirable andundesirable species can occur.

Climate change will result in impacts to manyfacets of natural resource functioning.

� To recover after over-grazing, woodland andherbaceous vegetation generally requiresequences of above-average rainfall so thatthe plant populations and perennial rootsystems can build up. In a highly variableclimate, there are an infinite number ofcombinations of climate and managementthat might trigger improvements in landcondition, species composition or landscapefunctions. Improvements may not appearuntil well after the change in managementor climate sequence has occurred.

� Ecosystem function, litter decompositionand soil fauna communities are highlysensitive to changes in environmentalconditions (especially soil water, soiltemperature, nutrient and carbon status).These organisms drive important ecosystemfunctions such as decomposition and nutrient

3.4 Potential future impacts of climate change inQueensland rangelands

Chris Chilcott1, Steve Crimp1 and Greg McKeon1

1 Climate Impacts and Natural Resources, QueenslandDept of Natural Resources and Mines, 80 Meiers Road,Indooroopilly, Qld 4068


mineralisation, and they maintain soil structurethrough burrowing and forming water stableaggregates. Climatic pulses are importanttriggers to the function of these organisms, andwill determine how populations will changeunder climate change.

� Many vegetation types in southernQueensland regenerate successfullyfollowing the disturbance caused bybroadscale clearing. While regrowth bywoody communities is considered amanagement problem in a productionlandscape, it is critical to the sustainabilityof many ecosystem functions, and may havesome positive biodiversity consequences.This ability to regenerate is a function ofnative vegetation retention levels andvegetation condition. As with other keyecosystem processes, regeneration capacityis principally driven by climatic factors andmay be affected by climate change withinthe confines of varied land managementregimes (such as fire and grazing).

� Recruitment and death in grazed woodlandsare, for the most part, driven by climaticpulses. A series of above average rainfallevents is required for recruitment, althoughit can also drive recruitment of weed species,an example being Acacia nilotica invasioninto Mitchell grass plains. Incidences of treedieback in mature canopies of grassywoodlands have been observed followingextended periods of drought. In the DalrympleShire, between 1991 and 1994, there were largereductions in tree canopy cover, with tree deathin some cases, probably because of thedepletion of sub-surface soil water.

� Although a majority of the farming andgrazing systems in Queensland are limitedby water, periods of above-average andexcessive rainfall lead to pulses of water lossvia runoff or deep drainage. Removal ofnative vegetation tends to increase deepdrainage (rainfall percolating to thegroundwater) if the replacement pastures andcrops have shallower roots or transpire forfewer days of the year than the nativevegetation (as happens with winter dormantor frosted pastures). The degree to whichremoval of trees changes these two factors

(rooting depth and ‘green days’) is currentlypoorly understood for pasture-woodlands inQueensland. Where groundwater systemscannot accept the increased recharge, thereis likely to be salinisation of the soil orincreased export of salt into rivers.

� Current research into thresholds for nativevegetation retention does not consider theimplications of climate change on landscapedesign or habitat condition. Most currentresearch assumes complex native ecosystemscan be modified down to some threshold,where most species and functions areretained. Below that threshold, themodification will result in significantdeclines in either species or function.However, the level of disturbance (be it fromland management or from climate change)within a remnant could have a greater impactupon the functionality of remnant vegetationthan the extent of fragmentation, thuschanging thresholds.

If rangelands are to avoid permanent loss ofresource condition and worsening regionaldegradation, some change in current managementpractices is necessary. One key change needed forthe management of both grazing systems andparks is an appreciation of safe carrying capacity.This innovation to improve grazing landmanagement is gaining momentum through StateGovernment initiatives (such as the South WestStrategy and the desert uplands initiative) andindustry developments (such as the grazing landmanagement education package funded by Meatand Livestock Australia). Safe carryingcapacities are derived directly from

(1) graziers’ knowledge and experience of safecarrying capacity, and

(2) calculation of pasture growth, tree densityand pasture condition for a given land type(using the GRASP model).

These calculations are limited in their capacityto deal with either climate variability or climatechange because they are based on averageexperiences and past climatic records (generallyprior to climate change).

As well as the largely off-reserve grazingmanagement approach above, other strategic

3.4 Queensland rangelands


planning and tactical management activities areneeded. The protection of biodiversity (both on-and off-reserve) depends almost entirely on theeffectiveness with which in situ protectionmeasures can be located, designed andmaintained, particularly in the light of additionalstresses introduced through climate change.Pressy and Taff (2001) believe that to measure

the effectiveness of protected areas there shouldbe an assessment of vulnerability bias (the extentto which reserves have been dedicated in partsof the region with most risk of vegetation loss)and of representativeness and efficiency. If suchelements are included in reserve planning theycould enhance the resilience of biodiversity toclimate change.

3.4 Queensland rangelands


What is WARMS?

The Western Australian Rangeland MonitoringSystem (WARMS) is a ground-based systemdesigned for primary production rather thanbiodiversity purposes. It indicates changes in thepastoral rangelands at broad scale, based onattributes of the soil and perennial vegetation ata set of representative sites (Watson et al. 2001).The emphasis is on detecting change rather thanassessing condition.

The current WARMS program began in 1993,although it emerged from an earlier system andso some sites have records (of varying quality)back to the mid-1970s.

There are about 1600 WARMS sites locatedwithin about 98,000,000 ha of pastoralrangelands in Western Australia. The averagedensity therefore is about one site per 61,000 haor just over three sites per lease (Figure 3.5.1).

WARMS is designed to provide information atthe scale of vegetation type, IBRA (InterimBiogeographic Regionalisation of Australia)region or other regional scale, for use bygovernment rather than individual managers.

There are two types of WARMS sites. Grasslandsites are used in the Kimberley, Pilbara andnorthern Gascoyne. Shrubland sites are used inthe shrublands from the Pilbara through to theNullarbor. Vegetation type at the site determineswhether a Grassland or Shrubland site is installed.

Grassland sites are reassessed on a three-yearschedule and Shrubland sites on a five- to six-year schedule.

What is recorded at WARMS sites?

Site photo

Both Grassland and Shrubland sites use a standardphoto, taken obliquely from about 2.5 m height fora defined trapezoid area of 132.5 m2 that fits withinthe viewfinder of a camera. The photocomposition is about 1/3 sky so that the generallocation as well as the specific photo area canbe seen.

3.5 The Western Australian Rangeland Monitoring System (WARMS) andits potential for detecting impacts of climate change

Ian Watson1

1 Centre for Management of Arid Environments, Dept ofAgriculture Western Australia, PO Box 483, Northam,WA 6401

Figure 3.5.1. Locations of WARMS sites: greyed areais pastoral tenure; IBRA regions are shown with khakilines; different site colours represent different years ofreassessment


Vegetation on Grassland sites

The frequency of all perennial species (mostlygrasses) is assessed in 100 quadrats, each 0.5 m2.Frequency is the number of quadrats (out of 100)in which the species are present. The crowncover of all woody species (>1 m height) isassessed using a Bitterlich stick. Site dimensionsare about 50 m by 30 m. While the ends of thetransects are permanently located, the individualquadrats are not; rather, they are ‘thrown’ evenly,20 per transect.

Vegetation on Shrubland sites

A direct census technique is used to locate allperennial woody species (with at least a 5 yearlifespan) individually. This technique allows thepopulation dynamics to be determined (thesurvivors, deaths and recruits) because the sameindividual is located at each assessment. Themaximum width and height (to 205 cm) of theliving canopy is also recorded for eachindividual. Three permanent belt transects areused, with total area dependent on shrub density(range is 30 m2 to 1200 m2).

Landscape Function Analysis

Standard CSIRO Landscape Function Analysis(Tongway 1994; Tongway and Hindley 1995) isdone on all sites. It assesses permanent and semi-permanent attributes such as position in thelandscape, mode of erosion and the spatialarrangement of the patch and inter-patch zones,as well as assessments of the soil surface from asample of ‘query points’ for the most prevalentpatches.

Permanent and semi-permanent attributes

A number of permanent and semi-permanentattributes are also recorded for each site, suchas location (with Global Positioning System),land-system, distance from water, paddockname, and so on.

Stratification and location criteria and theirimplications

Stratification at the regional scale

The sites have been installed on land held underpastoral tenure. Sites have been allocated tocoarse vegetation groups based on an index of

pastoral productivity, fragility and areal extent.This means that fragile productive systems (suchas alluvial areas) are over-represented at theexpense of robust and/or low productivity areas(such as hard spinifex).

Stratification at the lease scale

The sites have been allocated to leases on a prorata basis, based on the regional stratification.

Location at the local scale

The sites have been installed so as to representthe full range of vegetation states across the localarea, although the majority of sites are on themost common state. They were located onrepresentative areas of land, using the largestgrazed area within each paddock. They werelocated towards the centre, rather than margins,of each area of vegetation but not on extremelydynamic areas such as gully heads. They aretypically at least 1.5 km from permanent water,and in the Gascoyne–Murchison region, forexample, 75% of sites are between 1.5 km and3.5 km from water (Watson and Thomas 2003).This means that small areas of vegetation or‘special’ vegetation types (such as at the tops ofbreakaways, or in riparian zones) are notsampled, nor are there many sites beyond normalgrazing radius of permanent water.

Potential outputs from WARMS fordetecting impacts of climate change

The major factors driving change on each siteare (in no particular order): demographic inertia,domestic livestock and other grazing (principallygoats and kangaroos), fire, seasonal effects, othersignificant perturbations (such as insectdamage), and probably climate change.

Possible effects of climate change that might becaptured through WARMS include woodythickening (or decline) both in terms of plantdensity and cover, change in distribution ofspecies across sites, shifts in species compositionwithin sites, changes in species richness at thesite scale and changed demographic rates (seeFigure 3.5.2 for example, and Watson andThomas 2003 for more detail).

Because the sites are point based, summaries canbe prepared for any spatial aggregation, as

3.5 WA rangelands monitoring


Figure 3.5.2. Changes on 223 reassessed WARMSShrubland sites from the Gascoyne–Murchison Strategyarea. Date 1 was between 1993 and 1997 and Date 2was between 1999 and 2001. The average periodbetween Date 1 and Date 2 was 5.3 years.

Results on a site by site basis

Population growth rate (density)

Canopy area (cover)

Species richness (by site) or occurrence (by species)

Results on a species by species basis

Date 1: number of shrub species per site

1 10 50

Da

te 2

: n

um

be

r o

f sh

rub

sp

ecie

s p

er

site

1

10

50

Date 1: number of sites each species

was found on

1 5 10 50 100 150

Da

te 2

: n

um

be

r o

f site

s e

ach

sp

ecie

s w

as f

ou

nd

on

1

5

10

50

100

150

Date 1: total canopy area at each site (m2/m2)

0.01 0.1 1

Da

te 2

: to

tal ca

no

py a

rea

at

ea

ch

site

(m

2/m

2)

0.01

0.1

1

Date 1: average canopy size (m2/plant)

of each species

0.001 0.01 0.1 1 10

0.001

0.01

0.1

1

10

Date 1: number of plants on each site

10 50 100 500

Da

te 2

: n

um

be

r o

f p

lan

ts o

n e

ach

site

10

50

100

500

Date 1: number of plants of each species

1 5 10 500 1000 5000

Da

te 2

: n

um

be

r o

f p

lan

ts o

f e

ach

sp

ecie

s

1

5

10

500

1000

5000

Da

te 2

: a

ve

rag

e c

an

op

y s

ize

(m

2/p

lan

t)

of

ea

ch

sp

ecie

s



defined by the user. Because data are recordedto the species level, summaries can be preparedby species, functional group, turnover ratecategory or other categorisation determined bythe question.

There are some privacy issues associated withthe use of WARMS data. The greatest concernis that summaries of change should not beprepared for individual leases, since the systemis not designed to report at the lease scale. Ingeneral, WARMS data are made available forbona fide scientific pursuits.

Other potential sites

There are a large number of other sites, installedfor various purposes, which are now largely

inactive but may still be useful to specific studiesof climate change impacts (Watson et al. 2001).These include almost 4000 ‘old-WARMS’ andother miscellaneous sites (data from mid-1970sto early 1990s), old flightlines (1970s and 1980s)and about 30 or so exclosures (late 1960s to mid1990s). The Rangeland Survey Program has alsoleft behind about 4000 inventory-and-conditionsites and over 80,000 traverse assessments ofrange condition at 1 km intervals from the late1960s to the present. Ongoing pastoral leaseinspections also provide traverse assessments foreach station on a one- to six-year cycle.



Is there evidence of climate change impacton rangeland biodiversity?

Global climate change caused by increased ratesof greenhouse gas emissions is a recentphenomenon that is now predicted todramatically change the global thermal budget.We know that the Australian rangelands havebeen exposed to global warming and coolingevents over tens of thousands of years (Frakeset al. 1987). As Australia drifted towards theequator, it passed through different climaticzones with changing wind systems andcirculation of ocean currents that causedfluctuating thermal budgets. This geologicalevolution has mediated our present-daybiodiversity leading to some global extinctionsof less adapted species, the evolution of existingenvironments and species, and the isolation ofothers in refugia (Heatwole 1987). The way biotaresponded to these fluctuating global thermalbudgets in the past is a good indication of theway they respond naturally to broad-scaleclimate fluctuations (Swetnam et al. 1999). Allthe same, we need to be cautious with thiscomparison, ensuring that present-day thermalbudgets are matching similar budgets of thepast—a significant challenge in historicalreconstruction studies (Millar and Woolfenden1999).

Can we detect present-day impacts?

Climate extremes

The present-day climate of the rangelands varieswidely and unpredictably from place to placeand through time (Pickup and Stafford Smith1993; McKeon et al. 1998). For example, rainfalland sequences of events in the Alice Springsregion based on 113 years of records show that

3.6 Rangeland biodiversity and climate change: key issuesand challenges

Anita Smyth1

several events of 50 mm rainfall (42 in total)can occur in a year, but many years can also passwithout any (Stafford Smith and Morton 1990).This rainfall variability is reflected in thevegetation growth patterns. Substantial rainfallevents produce a significant vegetation responsethat resembles a series of pulse and decayfunctions of growth and vegetation senescencerespectively, over time. However, another keyissue associated with detecting the impact ofclimate change on vegetation growth in therangelands is that these pulse and decayvegetation responses may or may not form trends(Griffin and Friedel 1985; Friedel 1990; Pickupand Stafford Smith 1993). There is otherevidence of vegetation responses to climate, asshown by a study on the Mitchell grasslandswhere primary and secondary floristic gradientswere correlated with gradients in mean annualrainfall and temperature (Fensham et al. 2000).We also know that changing climate affects thefrequency, intensity and extent of fire in therangelands through variable fuel loads (Griffinand Friedel 1984).

Climate changes in both vegetation andhydrological cycles are also predicted to affectfaunal biodiversity (Woinarski 2000), althoughthe evidence appears weak for the rangelandbiota (Whitehead 2000). Predictions for futurebiotic responses as described by climateenvelopes have been attempted using variousclimate-change scenarios, but these scenarioshave been shown to be very imprecise at leastfor northern Australia (Howden et al. 1998).Another important issue with assessing theimpact of climate change on biodiversity using

1 Centre for Arid Zone Research, CSIRO SustainableEcosystems, PO Box 2111, Alice Springs, NT 0871


bioclimatic predictive approaches is that present-day distributions only reflect present-day climateand other environmental driving factors (H.A.Nix, pers. comm. 2002). Unless compared tohistorical predictions, the signals can bemisleading because then it may be assumed thatpresent-day distributions do not change. In the caseof animals that are broad-scale seasonal migrants(such as wetland birds), that is not correct.

Megadiverse biome

The rich diversity of the rangelands is anotherfeature that makes detecting impacts to climatechange a real challenge. Over half of Australia’sbioregions are in the rangelands. The landscapesare extremely fragile but support nine of 12major vegetation types, and the land systemsdiffer in their ability to recover once substantialchange has occurred. Most of Australia’s birdsand reptiles can be found in the rangelands, butso can the world’s highest mammalian extinctionrates. Not only is this richness mediated byextremely variable climates but human activitiesand land uses also mediate present-day biodiversity.

Other key human environmental drivers

Grazing, artificial water sources, tree clearing,introduced predators and plants, and changes infire regimes are highlighted, with goodsupportive evidence, in the National Land andWater Resources Audit as key drivers of impactson rangeland biodiversity (Whitehead 2000;Woinarski 2000). These drivers operate intandem to produce land degradation, habitat loss,habitat degradation and fragmentation, andoverall homogenisation of ecological systems.The processes lead to a loss of biodiversity andshifts in its composition at all scales, from plotto regional and continental.

Climate change, which is a more incipientprocess, adds to this impact overall, especiallythrough broad-scale habitat change. However,because of the very long, delayed responses bybiota to climate change, less attention is givento studying the impacts of climate change onrangeland biodiversity, particularly when theother human environmental drivers are of greaterimportance to natural resources management inthe shorter term.

Conclusions

Biodiversity of the rangelands has survivedprolonged fluctuations in the global thermalbudget in the past, but has never experiencedthe accelerated pace of present-day globalclimate change. Such rapid change is likely tospeed up the extinction rates of climate-sensitivebiota, although the time scales in which this mayhappen in the rangelands are unknown. Littleresearch has been conducted on the impacts ofclimate change on rangeland biodiversity,especially fauna, and this is understandable asthe evidence is much stronger for the moreimmediate impacts of other human activities andland-use practices in the rangelands. To fullyunderstand the separate impact of climate changeon biodiversity, we must seek innovativeapproaches for teasing apart the signal of climateimpact on biotic response variables, and the‘noise’ explained by natural diversity and otherhuman environmental drivers. By identifyingsynergies with other biodiversity initiatives andadopting historical and present-day climatestudies, opportunities for innovative approachesare likely to be forthcoming.

3.6 Key rangeland issues


Climate research has traditionally advancedalong two major lines of inquiry: observationand modelling. Statistical methods have beenused widely in observational work, but to only alimited degree in modelling studies. It is commonin modelling work, for example, to parameterisesome relationships using available data, andstatistical methods are often used to do this.However, the application of these methods tendsto be ad hoc, and the resulting uncertainties notcaptured. In observational studies the statisticalmethods used are typically well established ratherthan contemporary, and require the application ofoften physically unreasonable assumptions. Chiefamongst these assumptions would be stationarityand independence, and the climate system isunlikely to comply with either of them.

The Indian Ocean Climate Initiative (IOCI) is a5-year project funded by the Government ofWestern Australia (WA). It is somewhat uniquein that it has a statistical research capability aswell as scientists from the more conventionalfields of climatology, hydrology and climatemodelling. The project has led to thedevelopment of new statistical methods for thestudy of nonlinear phenomena and for modellingclimate extremes.

The IOCI has accumulated a substantial bodyof knowledge over the past 5 years. Quoting fromthe executive summary of the just publishedIOCI report Climate Variability and Change inSouthwest Western Australia:

� Winter rainfall in the southwest of WesternAustralia has decreased substantially sincethe mid-20th century. This has altered theperceptions of the climate of the region eventhough a similar, albeit less severe, dry

3.7 Advancing statistical science for climate research: a perspectivefrom the Indian Ocean Climate Initiative (IOCI)

Eddy Campbell1

sequence was experienced around thebeginning of the 20th century.

� Temperatures, both day-time and night-time,have increased gradually but substantiallyover the last 50 years, particularly in winterand autumn.

� The rainfall decrease was only observed inearly winter (May–July) rainfall; late winter(August–October) rainfall has actuallyincreased, although by a smaller amount.

� The winter rainfall sharply and suddenlydecreased in the mid-1970s by about 10–20%. It was not a gradual decline but moreof a switching into an alternative rainfallregime.

� The reduction in winter rainfall resulted inan even sharper fall in streamflow in thesouthwest.

There appears to be sufficient evidence toattribute the gradual warming to the enhancedgreenhouse effect. The rainfall decline appearsto be associated with a change in atmosphericcirculation at about the same time. The reducedrainfall and circulation changes bear someresemblance to greenhouse changes projectedby most climate models. The changes are notsufficiently similar to attribute them beyondreasonable doubt to greenhouse at this time. Thecurrent climate regime may simply be a modeof natural variability, which happens to be inline with greenhouse projections for the region.Southwest WA may therefore be experiencing ataste of its future climate.

1 CSIRO Mathematical and Information Sciences,Private Bag 5, PO Wembley, WA 6913


In terms of water resources, there is an urgentneed to adapt to a climate of less rainfall and soreduced streamflow, with increased evaporation.Historically the water supply system has reliedon occasional extreme winter rains to rechargethe system. The change in the likelihood ofextreme rainfall events is quite startling. UsingManjimup in southwest WA as an example,Table 3.7.1 shows the drop in probability ofextreme rainfall, since 1930.

These estimates in the table were derivedthrough some developments in the statisticalmodelling of extreme events. Our work innonlinear methods has indicated that mid-IndianOcean sea surface temperature may beimplicated in the switch between the dry andwet seasons in the southwest. This has thepotential to open up new lines of inquiry intofactors influencing climate variability andchange in the region.

The key question we should perhaps start within climate research is: how can we best forecastthe future? The next logical step is to apply thebest available science to integrate physicalunderstanding encapsulated in models andobservations of the climate system. Contemporary

Table 3.7.1. A comparison of exceedance probabilitiesfor extreme daily rainfall in 1930–1965 and 1966–2001, at Manjimup, WA

Daily rainfall Probability Probability (mm) (1930–1965) (1966–2001)

30 0.0132 0.00472

34 0.00837 0.00281

38 0.00543 0.00167

40 0.00448 0.00129

50 0.00181 0.000363

60 0.000835 0.000363

statistical science offers some excitingdevelopments for such integration, while alsoproviding the means to capture associateduncertainties.

The climate system is very hard to understandand even harder to predict, so we should expectto undertake statistical research to make best useof the observations we make. It is not enough toapply textbook methods and then hope that ourphysical knowledge and intuition will makegood. With crafted statistical tools we can askmore insightful questions of our data anddevelop new physical understanding.

3.7 Statistical input: the Indian Ocean Climate Initiative


Global warming has been a reality since the latterhalf of the twentieth century. Many models havebeen created to predict the impact of climatechange. These models suggest that arctic andalpine regions will be greatly reduced in extentbecause of increases in temperature, and thatplant species adapted to these cold climates arelikely to become extinct as a result ofcompetition from other species (Busby 1988).Much research has been carried out on thepossible loss of biodiversity in these rare andspecies-rich biological environments (see Green1998, Korner 1999, Theurillat and Guisan 2001).However, recent evidence suggests that whilethe general trend is for temperatures to increaseacross the planet, this may not be the case at alocal scale (Salinger and Griffiths 2001, Doranet al. 2002).

A recent study of long-term climate records inNew Zealand shows that there has been nosignificant change in the number of frost daysat higher altitudes (Salinger and Griffiths 2001),despite there being a general decrease in thenumber of frost days across the country as awhole. This departure from the perceived normof a trend of increased temperatures due to globalwarming has also been shown for a mountainenvironment in Tasmania (Kirkpatrick et al. 2002).

In Australia, climate predictions have beenmodelled at a national and a state scale.However, there has been no analysis of long-term climate records surrounding high altitudesites. Such an analysis would indicate whether thegeneral climate change predictions are accurate andapplicable at a regional level. These data wouldalso determine whether the environmentallysensitive and nationally rare Australian alpineand subalpine belt is under threat from globalwarming as predicted in the literature (Scherrer

and Pickering 2001). Furthermore, if linksbetween changes in climate and changes insubalpine vegetation over the same time periodwere to be evaluated, the result would providestrong evidence for rejecting or accepting thepredicted impacts of climate change onAustralian alpine and subalpine vegetation.

Using the long-term data available from sites(Kosciuszko National Park for 1950s to present,Tasmania for 1973 to present and 1991 topresent) we can explore ecological hypothesesabout the sustainability of subalpine plantspecies, communities, lifeforms and functionalgroups. These existing long-term data-sets willprovide an ecological context in which we canexamine the resilience of alpine and subalpineplant communities to invasion by exotic or lowlandspecies, driven by climate change or otherdisturbance factors (time since burning or grazing).

On a regional scale, the impact of globalwarming on alpine and subalpine plantcommunities has been speculated upon (seeScherrer and Pickering 2001, Theurillat andGuisan 2001), but little evidence exists of anyimpacts. Analysis, both of vegetation changes(especially lifeform groups or functional groups)in long-term monitoring sites in sensitive alpineand subalpine environments, and of changes inregional climate patterns, will show whetherecological processes are consistent across theAustralian alpine zone. The Kosciuszko data-set provides an opportunity to test whetherAustralian alpine vegetation is respondingconsistently to global warming or whether themaritime environment of Tasmania differs fromcontinental Australian alpine regions.

3.8 Vegetation change related to climate change in Australian subalpineenvironments

Kerry Bridle1 and Jamie Kirkpatrick1

1 School of Geography & Environmental Studies, Universityof Tasmania, GPO Box 252-78, Hobart, Tasmania 7001


Creating a Tasmanian peatland inventory,and assessment of the possible impacts ofclimate change on the carbon content ofTasmanian organic soils

Organic soils cover over 1,000,000 ha ofTasmania (Jarman et al. 1988). Organic soildeposits (bogs, peats, peatlands), especiallyraised bogs and blanket bogs, are extremelysensitive to environmental conditions. Raisedbogs (dominated by Sphagnum) and blanket bogs(dominated by rushes and sedges) form inpoorly-drained environments where there ismore precipitation than evaporation. Tasmanianblanket bogs are of national and internationalsignificance. Tasmania contains the largest areaof organic soil deposits in Australia, but there iscurrently no peatland inventory for Tasmania.No data are available on organic soil type, depth,or organic content. Therefore there is no estimateof carbon storage in these systems.

Limited climate data from western Tasmaniaindicate that evaporation exceeds precipitationduring the summer months, especially February.These climatic conditions pose a threat toexisting organic soil deposits, because alowering of the watertable may result inoxidation and decay of the peat deposits. Thewatertable in several western Tasmanian blanketbogs has been 15–20 cm below the soil surfacefor much of summer 2000–2001 and autumn2002, causing a probable increase in the releaseof carbon dioxide into the atmosphere.

Although peatlands are regarded as ‘carbonsinks’, it appears that they are likely to become(or may have already changed to) a carbonsource, creating a positive feedback into thephenomenon of global warming. Therefore, dataon the role of organic soil deposits as carbonsources or sinks are crucial input into any modelsof atmospheric carbon and climate change.

An inventory of Tasmanian organic soil depositsis necessary, with subsequent estimation of carbonstorage within peatlands. The data collected in theproject will offer valuable national insights intothe role of peatlands in the carbon balance. Oncethese data have been collected, the project canattempt to quantify the possible impact ofclimate change on these organic soil deposits. Itwill also identify the potential magnitude of thecarbon source that these deposits may form inthe light of climate change.

The inventory will provide baseline surveys ofa range of organic soil deposits across the state.This inventory may then be used to determinethe impact of fire on organic soil deposits, if thefire history is known for a particular site. Theremoval of peats up to 1 m deep during a swampfire on King Island in 2000 would have releasedsignificant amounts of carbon into theatmosphere. It has had a severe impact on thehydrology of the swamp system.

3.8 Subalpine vegetation


The Snowy Mountains contain the largest areaof contiguous subalpine and alpine habitats inAustralia. Other subalpine and alpine habitatsoccur as relatively isolated areas in the mainlandVictorian Alps, and in the Central Highlands andhigher peaks of island Tasmania. The largeSnowy Mountains’ area presents a uniqueopportunity to monitor regional changesattributable to global warming. Even a modestwarming (‘best case scenario’ of only +0.6°Cby 2070) will result in a 39% reduction in thearea that receives 30 days of snow per year inthe Snowy Mountains (Whetton 1998). Seasonalcover of snow is a major determinant of theanimals living in the subalpine and alpine areasof the Snowy Mountains (Green and Osborne1994, 1998).

There has been a significant decline in meansnow cover at Spencers Creek, as measured atthe Snowy Mountains Hydro-electric Authoritysnow course (Osborne et al. 1998). The snowcourse has been visited weekly through the snowseason since 1954. Examination of the data bydecade shows a total of 2283 metre-days of snowin the 1960s, with a 20% reduction to the 1970s(1843 metre-days) and a further 10% reductionto the 1980s and 1990s (1655 and 1706 metre-days respectively). The last five years, occurringin the warmest decade of the century (AustralianBureau of Meteorology), had the lowest five-year average metre-days of snow of the series:7.5% less than the previous lowest five yearsand 53% less than the highest five years (Greenand Pickering 2002). For fauna sensitive to depthand extent of snowcover, this decline might beexpected to be reflected in changes in itsdistribution.

Mammals and birds have responded to this 30%reduction in snow cover over the last 45 years.

There is an increased penetration of feralmammals into alpine and high subalpine areas,and there is a prolonged winter presence ofbrowsing macropods. The only three species ofnative mammal to increase in abundance withaltitude are affected adversely by years ofshallow snow. Birds are less constrained byaltitude, but among migratory species there hasbeen an observable change in timing of arrivalin the mountains, with earlier arrival in the 1980sand/or 1990s compared to the 1970s. Of 11species for which good data exist, seven arrivedin the mountains at least a month earlier in the1980s/1990s than in the 1970s. The implicationof these trends is that predicted impacts of globalwarming on snow cover will result in a significantchange in distribution of animal communities bothgeographically and through time.

To test this hypothesis, ongoing research islooking at:

� weekly recording of snow depth at theWhites River snow course (it has beenrecorded monthly since 1954). Whites Riveris close to long-term fauna-monitoring sitesand is in marginal snow conditions at1680 m. This area is expected to respondmore quickly to changes than the snowcourse at Spencers Creek.

� date of ice breakup and September ice depthat Blue Lake. In the past thirty years therehas been a two-month difference in the dateof ice breakup. It has ranged between lateOctober and mid December.

� snow depth, percentage snow cover,percentage of open flowers, and bird species

Ken Green1

3.9 Impacts of global warming onthe Snowy Mountains

1 NSW National Parks and Wildlife Service,PO Box 2228, Jindabyne, NSW 2627


present at seven 1 ha stations, 50 m apartaltitudinally, from 1350 to 1650 m, onDisappointment Spur, and at a further twostations at 1600 m and 1650 m. This studyis done four times a month from mid-Augustto mid-October and aims to document the

relationship between snow duration, plantphenology and bird migration. Tracks ofmacropods are also recorded and are relatedto snow depth and extent and proportion ofuncovered shrubs.

3.9 Impacts in the Snowy Mountains


Introduction

The mobility of birds allows them to respondrapidly to changes in the environment, generallyas a distinctive biotic group, over a range ofspatial scales.

Therefore, birds may be a good and earlyindicator of climate change. However, birds havethe same capacity to respond to anyenvironmental change, whether human-inducedclimate change, natural climatic variability,disturbance events or a host of other human-induced landscape changes. How can we tell thedifference?

First, we have to demonstrate beyond reasonabledoubt that a change in distribution has occurredor is occurring over the relevant time and spatialscales. Geographically, range changes ought tobe fairly steady and pervasive. I examine in moredetail the specific requirements and sufficientand necessary evidence to meet the conditionsfor change attributable to global warming andother ‘greenhouse’ effects.

Range changes that meet greenhouseclimate change criteria within Australia

� Range expansion to the south (or a move tohigher altitude)

� Range contraction from the north (or a moveto higher altitude)

� Pervasive range shift across a range front,not isolated occurrences

� Steady range shift over the past few decades,as opposed to earlier responses to massiveland-use change coincident with Europeanoccupation

For a discussion of these criteria see Hughes(2000).

Range changes in Australian birds

The breadth of range changes is quickly sketchedto eliminate those types of movements and rangeshifts that cannot be construed as evidence for‘greenhouse’-induced changes.

Reversible range shifts

Irregular, short-term range shifts (periods of afew years or less) are a striking characteristic ofmany native Australian bird species. Two examplessuffice. Nomadism allows many species to exploitgeographic shifts in food resources, and manyspecies have been described as ‘locally tocontinentally nomadic’, depending on the scaleat which these range changes typically occur.Related phenomena, and topical currently(September 2002), are drought-induced,extraordinary movements that take species welloutside their customary range, usually towardsthe southern, eastern and presumably northerncoastal regions. There are many reports inAustralian ornithological circles currently oflarge influxes of a wide range of species, butparticularly waterbirds, to coastal areas wherethe majority of ornithologists reside. These areepisodic and temporary distributional changesand so cannot be ascribed to greenhouse effects.

Range expansions

Many native species historically considered to betypical of Australian arid and semi-arid regions

1 Visiting Fellow, CSIRO Sustainable Ecosystems, GPOBox 284, Canberra, ACT 2601

3.10 What’s the link, if any, between recent changes in distribution ofAustralian birds and greenhouse climate change?

Julian Reid1


have invaded temperate parts of Australia overthe past 100 years. In some cases, these speciesare still steadily expanding to the south and intohigher altitudes. The Galah Cacatua roseicapilla,Little Corella C. sanguinea and Crested PigeonOcyphaps lophotes are the best examples(Blakers et al. 1984) and the latter species’colonisation of the Australian Capital Territoryin the last 20 years has been closely monitoredand documented. It is widely agreed that, ratherthan climate change, it is land-cover and land-use changes (primarily vegetation clearance andprovision of artificial water sources) that havecaused these southern range expansions, andsimilar expansions along the eastern seaboard.

Many non-native bird species have colonisedAustralia and substantially expanded their rangesfrom initial entry and release points. Their rangeexpansions have proceeded in all directions and insome cases aggressively continue; for example, theIndian Mynah Acridotheres tristis. The mynah’smain direction of expansion has been north inaccordance with its climatic provenance. Againclimate change is not implicated, thoughurbanisation and land-use change is.

Range contractions

Many native species are in serious decline acrossall parts of the continent. Range contractionsoften occur at the periphery of a decliningspecies’ range first or most clearly, though thisis by no means general (Brown 1984).

Are there any species with northern range limitsinside the northern Australian coast exhibitinga steady pervasive range contractionsouthwards? This would provide firmer evidenceof greenhouse impacts, particularly if the rangewere not also contracting from the south. Evenstronger evidence would be provided by anaccompanying range extension southwards.Exploratory research is required to see if thesepatterns emerge.

Torresian species expanding south

(1) Landbird occurrence: Figbird Specotheresviridis, Channel-billed Cuckoo Scythropsnovaehollandiae, Common Koel Eudynamysscolopacea, Australian Brush-turkeyAlectura lathami.

(2) Waterbird occurrence and breeding: Black-necked Stork Ephippiorhynchus asiaticus,Pied Heron Ardea picata.

We have to decide whether these rangeexpansions are more parsimoniously attributableto land-use and landscape transformation thanclimate change. There does appear to be a strongtrend for Torresian migratory landbirds to beextending south. The first three species listed aremigratory landbirds, but they are all fruit-eaters,so other factors need to be investigated.

Conclusions

Are birds necessarily good indicators of pervasiveclimate change? Might they not be too responsive(vagile)? For a start, birds are endothermic and soare substantially buffered against the directimpacts of temperature change. Birds tend to uselarge amounts of space, from the scale of theindividual home range to that of the populationand species. Physiologically then, they seem tobe tolerant of a broad climatic range and, beinglong-lived, also tolerant of weather and short-term climatic variability.

Multiple causes of shifts in distribution can bedifficult or impossible to disentangle,particularly if there are direct and indirect effectsas well as interactions between causes, operatingin tandem. ENSO episodes and their impacts ondistribution patterns exemplify this problem(Reed et al. 2000). Climate change is likely toalter the timing and magnitude of ENSOreversals and so have an indirect role in theseeffects in addition to its direct effects. Teasingapart indirect effects and interactions will beproblematic (but see Martin 2001).

Organisms which have particular life-historytraits tightly controlled by temperature may bethe best candidates to screen for climate changeimpacts already underway (see McGeoch 1998).Examples include certain phenological (lifecycle) phenomena in plants, and ectothermicanimals (such as insects). Montane areas mayprovide the best arenas for study here because steepgradients are conducive to sharp zonal patterning.

Conversely, extensive flat regions where climategradients are gentle (such as the broad openplains of inland Australia, like the tundra of the

3.10 Birds and climate change


Northern Hemisphere) provide a potential arenafor large shifts in geographic range over tens tohundreds of kilometres for those organisms orecological indices capable of rapid adjustment.Birds certainly satisfy the criterion of rapidadjustment potential.

The recent Australian ornithological literature,including the regional journals and newsletters,reports the pervasive southward shift along theeastern seaboard of a range of Torresian bird

species. The selection of species involvedincludes many more than the four species citedand warrants close study. It may also be fruitfulto examine the stability of northern range limitsof Bassian bird species, separating the effectsof climate changes from those of land-coverchange and comparing the northern limits to thestability of their range limits on other fronts.

The answer to the question in the title of thiscontribution is … we don’t know!

3.10 Birds and climate change


3.11 Birds andclimate change

Existing impacts

As identified in the paper by Hughes, a range ofstudies suggests that breeding seasons arealtering in response to climate changes. InAustralia, as discussed by Reid, there are alsosome shifts in species distributions towards thesouth or along changing rainfall contours.Further analysis is needed to determine if thesechanges are associated with climate change.

Potential impacts

On the basis of existing understanding, there isa large range of likely potential impacts ofclimate change on birds in Australia.

Being relatively mobile, most birds will probablyfare better than many other life forms. However,the ability to move and colonise varies amongbirds, and those least able to disperse or colonisemay face extinction, particularly where newhabitats are distant or where intermediate habitatis unsuitable. There is also likely to be varyingability among bird species to adapt to changinghabitats and food sources, and this will tend toput more pressure on species that need specifichabitats (habitat specialists) than on generalists.In addition, some other processes (such asfragmentation of habitat) may reduce the abilityof species to adapt.

One of the solutions in such cases may be toestablish habitats that can act as ‘stepping stones’.However, some efforts to minimise social oreconomic impacts may further disadvantage somehabitats and the birds that use them. For example,efforts to prevent coastal erosion associated withrises in sea levels will also have an impact on theHooded Plover because they breed in the upperbeach and adjacent dunes. Lastly, these impactsmay vary geographically, as critical parts of the

range may be affected while the bulk of the rangemay remain suitable; for the Swift Parrot forinstance. As there is little information knownabout many of the above factors, predictions ofthe impact of climate change are difficult.

Birds Australia databases

Birds Australia owns a number of databases onbirds that may be useful for assessing the impactof climate change on birds (and other elementsof biodiversity, depending on one’s faith in birdsas indicators). The following table lists thecurrent Birds Australia databases. Of these, twoare of particular note as substantial long-termdata-sets.

� The Atlas of Australian Birds is an ongoingdatabase that contains information aboutdistribution and relative abundance(frequency of occurrence) at thecontininental scale.

� The Nest Record Scheme database containsdata on bird breeding (timing, etc.). It ismostly computerised, but needs someverification.

Birds Australia wants to be a partner in researchabout climate change and birds. There are somelimiting factors—we are limited in resources.Also, our organisation has made an enormousinvestment in these databases and so we wish toprotect our intellectual property. Thus, access tothese data involves:

1) formal data-sharing agreements (large datarequests have to be referred to our Researchand Conservation Committee), and

1 Research & Conservation Department, Birds Australia,415 Riversdale Road, Hawthorn East, Victoria 3123

Michael Weston1


Table 3.11.1. Birds Australia data-sets, and their scope in timeframe, geography and species. Asteriskedrows are probably of greatest interest.

2) some funding to cover extraction time(covering costs only).

3) Additionally, the analysis and interpretation ofAtlas data are not straightforward activities,and we will need to negotiate any staffinvolvement. Ideally, we would like to be

involved in the analysis of these data, andencourage any grant applications or proposalsto factor in some funding for our involvement.

If you are interested in using Birds Australia data,please contact Michael Weston, email:[email protected] .

3.11 Birds Australia databases

NAME SCOPE SCOPE SCOPE TYPE(time scale) (geographical) (birds)

*Atlas of Australian Birds 1997–1981 (1) Australia, including All wild birds Location with some1998–2001 (2) territories and count data of

territorial waters wetland birds

Australian Bird Count 1989–1995 Australia Mainly bush birds Location and(ABC) (passerines) numbers

AWSG Banding Database From 1981 Australia Waders Morphometrics andongoing (shorebirds) movements

AWSG Colour-flag From c.1995 Australia Waders MovementsDatabase ongoing (shorebirds)

Beach Patrol Australia Seabirds

Bird Monitoring at 1997–2002 Engine Ponds, Mostly waterbirds Total countsEngine Ponds Sydney

Bird of Prey Watch 1986–1990 (1) Australia Raptors Transect counts(BOP Watch) 1996–2000 (2) within regions

Birds in Backyards 2000–2002 Sydney and Terrestrial urban Location, counts,surrounds birds nests, vegetation

Birds in Tree Hollows 1997 Victoria Hollow-nesters Breedinginformation

Birds on Bore Drains 1998 Roxby Downs, SA All Presence/absence

Birds on Farms 1995–1998 E and SW Australia All Point surveys(1080 sites)

Birds on Golf Courses 1990s All Counts by habitat

Corner Inlet Oyster Farm 1989–1990 Corner Inlet, Mostly waterbirds CountsVictoria

Freckled Duck Count 1987–1988 Victoria Waterbirds Site counts

Gulf Plains ConservationFramework

HANZAB (Handbook of 1980–2004 Australia, New All ComprehensiveAustralian, New Zealand Zealand, Antarctica literature reviewand Antarctic Birds)

HEC (Hydroelectrical 1999 NW Tasmania Moving birds Moving birdsCommission) windfarmcounts

Homebush Bay 2000–ongoing Homebush Bay, All at site Site countsSydney, NSW

Kooragang (Rehabilitation 2001–2002 Ash Island, Mostly waterbirds Total counts/of Waterbird Habitat on Newcastle behaviourAsh Island)

Little Terns 1989 E and N Australia Little Terns with Counts andnotes on other information onspecies breeding

Melbourne Airport 1996–2000 Melbourne Airport Airport birds Counts in grids

(continued next page)


Table 3.11.1 continued. Birds Australia data-sets. Asterisked rows are probably of greatest interest.

3.11 Birds Australia databases

NAME SCOPE SCOPE SCOPE TYPE(time scale) (geographical) (birds)

Murray-Darling Basin 1993–1996 Murray-Darling Wetland birds Location/count/someWaterbirds Basin catchment: breeding records

scattered across theBasin in four statesand the ACT

National Hooded Plover 1980–2001 Australia Waders Location andCounts (shorebirds) numbers

National Wader Counts From 1981 Australia Waders Location andongoing (shorebirds) numbers

*Nest Record Scheme (NRS) From 1964 Australia All breeding wild Breeding detailsongoing birds

Orange-bellied Parrot counts From 1984 to SE Australia Neophema and Counts by sitepresent coastal seed eaters

Pt Lillias and Avalon Saltworks 1994–1995 Pt Lillias and Avalon Waders and Location andSaltworks, Victoria waterbirds numbers

Pt Wilson Counts (mid 1990s) Pt Wilson, Victoria Waders (shorebirds)

Question of Survival Bird 1993 Australia All Site countsSurvey

Recent Ornithological 1998 onwards Worldwide All Bibliography includingLiterature abstract

Regular Wader Counts 1986–1990 Australia Waders (shorebirds) Location and numbers

Rolling Bird Survey (RBS) (mid 1980s) Australia Mainly bush birds Abundance(passerines)

Red-tailed Black Cockatoo W Victoria and RTBCs 1. Sightings Sightings, nests and(RTBC) SE South Australia 2. Nest site flock counts

3. Flock count

Seabird Atlas SE Australia Seabirds

Summer Waterfowl Count 1987–1992 Victoria Waterbirds Site counts(Victoria)

Swift Parrot Counts at 1999 Melbourne Airport Mostly bush birds Transects andMelbourne Airport flowering phenology

Sydney Airport Birdstrike 1988–2002 Sydney Airport All birds Bird strikesInformation RetrievalDatabase (BIRD)

Sydney Airport Foraging Data 1998–2002 Sydney Airport All birds Location andabundance

Sydney Olympic Park Authority 1996–2002 Sydney Olympic All birds Location counts/OCA Parklands

Tasmanian Shorebirds 1998–2002 Tasmania Shorebirds

Vicgroup Rushworth banding 1996–present Rushworth State Mainly bush birds Captures, recaptures,data (planned till 2016) Forest, Victoria (passerines) morphometrics

Victorian Wetlands Survey 1989–1992 Victoria Wetland birds Location and numbers

WA Hooded Plovers 1995–ongoing SW Western Australia Hooded Plovers Location and numbers,breeding information

WA Waterbirds 1981–1985 SW Western Australia Wetland birds Location and numbers


Introduction

Palaeoecological data provide valuable insightsinto past climate variability, the process of climatechange under the influence of greenhouse-gasforcing and the nature of ecosystem response, andsurvival under rapidly changing conditions.Consideration of long-term (1000–100,000 years)as well as short-term (10–100 years) timescalesshould be incorporated into biodiversity planning.

Long-term climate change

The northwards progression of Australia sincethe Tertiary has led to progressive aridification,the emergence of a biota adapted to complexcycles of climate change, and the appearance ofshort growing seasons in the south.

The last 2 million years have been dominated byglacial climates in which mean land temperatureswere as much as 9°C cooler than present in theinterior of Australia (Barrows et al. 2000).Therefore, many species are maladapted to present-day (interglacial) conditions. The continent is largeenough to cover a wide range of climates, and sohabitats have been preserved for species capableof migration. Currently, rare species are often thosewith limited capacity to disperse or habitats thathave diminished during interglacials; for instance,Wollemia nobilis (Wollemi Pine, Blue Mountains),Swainsona recta (Purple Pea, springtime plantof grassland steppe).

Palaeoceanographic records of past sea-surfacetemperatures from the Coral Sea have shown thatthere was an increase in temperature of around4°C around 245,000 years ago (Figure 3.12.1).This temperature increase may have been relatedto the development of the Western Pacific WarmPool leading to the formation of the Great BarrierReef, an increased temperature gradient across

the Pacific Ocean, and the onset of El Niño–Southern Oscillation activity (ENSO) (Isern etal. 1996). A resultant increase in climaticvariability could then explain the trends towardsincreased burning and the development of moreopen vegetation (Kershaw et al. 2002). Thecombination of higher precipitation resultingfrom increased sea-surface temperatures andclimate variability might also have allowed theexpansion of both complex rainforest andsclerophyll vegetation into adjacent areas.

Abrupt climate change

Recent palaeoclimate research has revealed thatthe Australian landscape has been influenced byabrupt climate change events during the last10,000 years. The onset of modern ENSOperiodicities around 5000 years before present(BP) followed by an abrupt increase in ENSOmagnitude around 3000 years BP are key timeperiods of rapid change in climate variability andtherefore ecosystem disturbance. Disturbance isseen as a major determinant of forest structure,and more frequent and intense storm events ordrought–fire events have the potential todrastically shorten the turnover period for treesand thus alter forest composition. This can beseen in high-resolution (ten year samplingintervals) pollen records spanning the last 10,000years from the Atherton Tablelands (LakeEuramoo) and Southern New South Wales (BegaSwamp). There, increased disturbance relatedto altered fire regimes around 5000 years BPresulted in long-lived shade-tolerant treesbecoming rare, and a collapse of dominant treespecies (Figure 3.12.2).

1 Research School of Pacific and Asian Studies,The Australian National University, Canberra, ACT 0200

3.12 Palaeoecological perspectives on climate change and its impact onbiodiversity

Simon Haberle1


Figure 3.13.2. High-resolution pollen diagramfrom Bega Swamp, southern NSW, showsperiods of major species collapse during theHolocene possibly related to changingdisturbance regimes (burning) at a time of gradualclimate change (Green et al. 1988)

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

11000

12000

Podocarp

us

Pom

aderr

isE

ucaly

ptu

s type 1

Eucaly

ptu

s c

f. p

auciflo

ra

Eucaly

ptu

s type 2

Acacia

Calli

tris

Casuarina

Epacris type

Kunzea type

Lepto

sperm

um

Monole

te F

ern

spore

s

Ble

chnum

Poaceae

Tubulif

lora

e

Charc

oal (>

5m

)

Age (

years

BP

)

Bega Swamp 1025 m, Southern New South Wales

Figure 3.12.1. Summary pollen and charcoal records

from terrestrial and marine records in north-east

Australia (Kershaw et al. 2002) compared to the

extended δ18O record of ODP 820 (Peerderman et al.

1993) showing a rise of approx. 4°C in sea-surface

temperatures at the end of Isotope Stage 8 (IS-8)

3.12 Palaeoecological perspective


Human impacts

Aboriginal land-use practices in existence for over50,000 years have had a major impact on vegetationdistribution and the survival of large marsupials inAustralia (Bowman 1998). However, by far thegreatest impact on biodiversity, and this is reflectedin many palaeoecological records, has been thearrival of Europeans around 220 years ago. Alteredfire regimes, forest clearance, and the introductionof herbivores and weeds have led to fragmentedbiomes that are likely to exacerbate the impact ofrapid climate change events on future regenerationand the migration potential of plants and animals.Unprecedented rates of vegetation change(Dodson and Mooney 2002) combined with theintroduction of exotic species can fundamentallychange ecosystem structure through competitiveexclusion of available resources. Climate changesthat increase the aggressiveness of exotic species,or reduce the competitiveness of indigenousspecies, will be significant. Understanding the rateof spread of weed species and their impact onbiodiversity in the Australian landscape requiresa multi-disciplinary approach—the developmentof an historical ecology.

Research areas

1. Comparison of climate models andpalaeoecological data to refine ourunderstanding of the interaction betweenclimate and organisms through time.

2. Further research aimed at extending existingpalaeoecological records and understandinghow the Western Pacific Warm Poolinfluenced Australia’s climate, given that theregion has already experienced rapid rises intemperature equivalent to those possibleunder the influence of greenhouse-gasforcing.

3. High-resolution studies of biotic responsesduring key time periods of abrupt climatechange within a range of biomes (forexample, 5000 years BP, 3000 years BP).

4. Multi-disciplinary studies examining therelative influence of human land-use changeversus climate change, developing newtechniques in palaeo-biodiversity andhistorical ecology.

3.12 Palaeoecological perspective


Relationships between vegetation, CO2 andclimate

Ecologists and biogeographers routinelycorrelate distributions of organisms withvariables that can be easily measured. There isa long history of relating distributions toprecipitation and air temperature. The reason forthis is that precipitation and temperature werethe things that were measured a century ago. Thisapproach does not make sense for at least tworeasons. Firstly, vegetation modifies airtemperatures through radiation and heat budgets.The optical properties of vegetation coverdetermine its albedo (α), and thus the amountof global solar radiation (R

s, in J/m2/s) that it

absorbs according to the energy budget

Rn = R

s(1–α) + R

Li – R

Lo ,

where Rn (J/m2/s) is the net all-wave radiation

at the surface, and RLi

and RLo

are the downwardand upward fluxes respectively of longwaveradiation. The partitioning of R

n is also affected

by vegetation cover, as indicated by thesimplified form of the heat budget equation

Rn = LE + H + ∆G,

where L (J/kg) is the latent heat of evaporation,E (kg/m/s) is the rate of evaporation, H (J/m2/s)is the heat convected to the atmosphere (whichwe measure as air temperature) and ∆G (J/m2/s)is the change in the heat in the subsurface layers.When water is readily available at the surface,such as after rain or from transpiring leaves, LEwill dominate the heat budget and ∆G and Hwill be small.

The second illogicality of the precipitation andtemperature approach arises because plantsrequire other resources, such as CO

2, O

2 and

nutrients, in addition to water and energy, andthe availability of these other resources(especially CO

2) varies through time (Barnola

1987). Variations in the concentration of CO2 in

the atmosphere through time are linked to theenergy and heat budgets through spatial andtemporal changes in vegetation cover, and wateruse efficiency of photosynthesis.

Changes in vegetation structure (but notdistributions of individual species) in responseto resource availability can be predicted usingthe TMS scheme (Berry and Roderick 2002a,b).In the TMS scheme the vegetation at a place isdescribed according to the proportionalrepresentation of turgor (T), mesic (M) andsclerophyll (S) leaf functional types. Theproportions of each leaf type can be plotted on atriangular diagram. The TMS scheme is a linkbetween leaf properties, vegetation structure andenvironmental conditions, and can be used topredict responses of the vegetation to changesin both the seasonal and long-term availabilityof moisture, nutrients and atmospheric CO

2

concentrations. The underlying theory can alsobe applied to predict changes in vegetationstructure in response to fire, livestock grazingand land clearance.

Direct effects of atmospheric CO2

enrichment on the vegetation

Responses of plants to increasing concentrationsof CO

2 can occur at all levels from leaves to

ecosystems. There is evidence for change in leafmorphology, by decreases in specific leaf area

3.13 Global change: the direct effects of the increasing concentration ofatmospheric carbon dioxide on the climate and the biota

Sandra Berry1

1 Ecosystem Dynamics Group and CRC for GreenhouseAccounting, Research School of Biological Sciences, TheAustralian National University, Canberra, ACT 0200


and stomatal frequency (Peñuelas and Matamala1990, Peñuelas and Azcon-Bieto 1992), and inthe ratio of surface area to volume (Roderick etal. 1999), with increasing CO

2 concentration. The

increase in ratio of leaf surface area to volume intrees and wild non-woody species is commonlyassociated with an increase in relative allocationto structural tissues (Pritchard et al. 1999) andincreasing leaf longevity (Li et al. 2000). Increasedflower production and reproductive output (fruitsand seeds) have also been observed (Lawlor andMitchell 1991, Osborne et al. 1997, Deng andWoodward 1998, Ziska and Caulfield 2000). Atthe plant level, there is evidence of increasedwater use efficiency (Drake and Gonzalez-Meler1997) and increased tolerance of salinity (Rogerset al. 1994). At the ecosystem scale, whentranspiration is limited by water availability, theincreased plant water use efficiency resultingfrom higher CO

2 concentration would usually

result in the available soil water being able to

support more leaves (that is, either support moreleaves for some of the time, or support the sameleaves for more time) and hence morephotosynthesis (Farquhar 1997, Owensby et al.1999, Li et al. 2000). As the dynamics ofmoisture availability have been shown to largelydetermine the vegetation height and cover(Specht 1972, 1981), a change in ecosystemwater use efficiency can also be expected tomodify competition dynamics at a site. Inenvironments enriched in CO

2, the mass

concentration of dry matter in woody stems islikely to increase rather than decrease (Roderickand Berry 2001), and this would enable woodystems to escape from the flame zone ingrasslands (Higgins et al. 2000), and wouldpresumably assist woody vegetation to establishon some grassland sites. All these factors couldpotentially affect the distribution and the relativeabundance of plant species and consequentlyanimal species as CO

2 concentration increases.

3.13 Effects of increasing CO2


Introduction

Globally, forest clearing is thought to be thegreatest threat to biodiversity in the tropics, andrates of clearing are certainly highest there,particularly in tropical Asia. Climate change issometimes discounted and has been less studiedin tropical regions than in temperate, boreal orarctic ecosystems. However, preliminary studiesindicate that climate change is a particularlysignificant threat to the long-term preservation ofthe animals and plants in the Wet Tropics WorldHeritage Area.

The Wet Tropics is dominated by mountainranges with altitudes varying sharply between sealevel and over 1600 m. Environmental gradientsassociated with this complex topography largelydetermine species composition and general patternsof biodiversity. Partly because of extensive clearingin the lowlands, most rainforest is above 300 mand most of the regionally endemic species arecool-adapted upland species. The regionallyendemic rainforest vertebrates generally aredistributed over areas with a very narrow range inannual mean temperatures. Consequently, thebiodiversity and regionally endemic species thatare keystone elements in the Wet Tropics WorldHeritage Area may be under severe threat over thenext few decades. An enormous loss of habitatleading to significant loss of biodiversity ispossible. Ecosystem processes and the provisionof ecosystem services (such as clean and reliablewater) also could be severely affected by climatechange.

Warming can have a particularly strong impacton mountainous regions like the Wet Tropics.The mountain tops and higher tablelands can bethought of as cool islands in a sea of warmerclimates. These islands are separated from each

other by the warmer valleys and form a scatteredarchipelago of habitat for organisms that areunable to survive and reproduce in warmerclimates. Many of the Wet Tropics endemic specieslive only in these cooler regions. An additionallikely effect of warming is a significant lifting ofthe base of the cloud-bank on tropical mountains.There is significant trapping of cloud-waterdroplets within cloud forests—so-called ‘cloudstripping’—which can considerably augmentrainfall, especially in the dry season. Globalclimate simulations with a scenario of doubledCO

2 concentration show that the relative

humidity surface is shifted upwards on tropicalmountains by hundreds of metres during thewinter dry season. This is the period when theseforests are most reliant on the moisture fromcloud contact. Because of their great sensitivityto climate, cloud forests are likely to displayclimate change effects in the very near future.

Have ecological and biological effects ofwarming already been observed ?

Palaeoecological studies in the Wet Tropics havedemonstrated the extensive biogeographicchanges that have occurred as a result of pastclimate change (Hopkins et al. 1993, 1996,Hilbert and Ostendorf 2001). However, there hasnot been any climate-change monitoring in theWet Tropics, making it impossible to statewhether warming in the 20th century has had animpact on the Wet Tropics flora or fauna orecosystem processes. Monitoring in other partsof the world has identified a large number ofecological and biological changes due to recentclimate change. For example, a lifting cloud-base

1 CSIRO Sustainable Ecosystems, Tropical ForestResearch Centre, PO Box 780, Atherton, Queensland 4883

3.14 Potential global warming impacts on terrestrial ecosystems andbiodiversity of the Wet Tropics

David W. Hilbert1


associated with increased ocean temperatureshas been implicated in the disappearance of 20species of frogs in highland rainforests ofMonteverde, Costa Rica.

Preliminary assessments of impacts in theWet Tropics

Forest type, defined by structure andphysiognomy, is an important habitat variable,although many trees and other organisms are notrestricted to a single forest type. Therefore, theextent and distribution of forest types in the futureis an important indicator of potential changes inbiodiversity. Recent work (Hilbert et al. 2001a,b,Ostendorf et al. 2001) has found that the tropicalforests of north Queensland are highly sensitive toclimate change within the range expected earlyin the 21st century. Overall, the location andextent of rainforests are determined by rainfalland its seasonality, with some influence of soilfertility and water-holding capacity. But the typeof rainforest and the kinds of organisms foundthere depend more on temperature.

In the Wet Tropics, one degree of warming canincrease the potential area of rainforests as awhole, as long as rainfall does not decrease.However, large changes in the distribution offorest environments are likely with even minorclimate change, and the relative abundance ofsome types could decrease significantly.Increased rainfall favours some rainforest typeswhile decreased precipitation increases the areasuitable for woodlands and forests dominatedby sclerophyllous genera such as Eucalyptus andAllocasuarina. The area of lowland MesophyllVine Forest environments increases withwarming, while upland Complex Notophyll VineForest environments respond either positivelyor negatively to temperature, depending on

precipitation. Highland rainforest environments(Simple Notophyll and Simple Microphyll VineFern Forests & Thickets), the habitat for manyof the region’s endemic vertebrates, decrease by50% with only a 1oC warming. Using this model,we have mapped the current and potential futuredistributions (1oC warming and –10% rainfall)of upland and highland rainforest types.Environments suitable for these forest typesdecline greatly and become very fragmented inthis climate-change scenario. Obviously, if theupper range of predicted warming occurs (5.8oC),no appropriate environments would remain withinthe Wet Tropics. Whether and where appropriateclimates might come to exist further to the south,say in the Border Ranges, is unknown. However,regional rainfall patterns and topographicconstraints imply that such new habitat wouldbe very far removed from the Wet Tropics.

There are two published studies of climatechange impacts on a vertebrate of the WetTropics focusing on the endangered NorthernBettong (Hilbert et al. 2001a) and the GoldenBowerbird (Hilbert et al. in press). Estimates ofthe potential habitat for the Northern Bettong inthe future depend on whether they require tallopen forests and whether rainfall increases ordecreases. Assuming that tall open forests arenot essential, habitat would decline if warmingis accompanied by greater precipitation, andwould increase if rainfall decreases. Ourmodelling of habitat for the Golden Bowerbird,a charismatic upland and highland endemic,predicts that the current habitat (1199 km2 inseveral distinct patches) will shrink by 63% withone degree of warming and a 10% decrease inrainfall. Three degrees of warming and a 10%reduction in rainfall reduces potential habitat toonly 28 km2.

3.14 Impacts in the Wet Tropics


Climate change is a particularly significant threatto the long-term preservation of the biota in thetropical rainforests of Australia’s Wet TropicsWorld Heritage Area. The Wet Tropics isdominated by mountain ranges with altitudesvarying sharply between sea level and over1600 m. Environmental gradients associatedwith a complex topography dominate thebiogeography of the region (Nix and Switzer1991, Williams et al. 1996). The gradient inaltitude is the most significant environmentalgradient determining species composition andgeneral patterns of biodiversity (Williams et al.1996, Williams and Pearson 1997). Mostrainforest is above 300 m and almost all of theregionally endemic species are cool-adaptedupland species (Nix and Switzer 1991, Williamset al. 1996).

Most of the regionally-endemic rainforestvertebrates are distributed over areas with a very

narrow range in annual mean temperatures. Therange of annual mean temperatures encounteredwithin the distribution of most species is verysmall (between 1.8°C for Cophixalus bombiensand 9°C for the Leaf-tailed Gecko (Saltuariuscornutus)) with an average range of annual meantemperature across regionally-endemic speciesof just 5.5°C (Williams unpublished data). Thesedata reveal just how significant an increase ofseveral degrees would be. The implication is thateven minor changes in temperature will havesignificant impact on available environments.Hilbert et al. (2001b) predict significant changesin the geographic distributions of vegetationtypes in the region, including a 50% decline inthe extent of upland rainforest types if there isonly a 1°C increase in temperature. The results,using neural network models, produce conclusionssimilar to those of the BIOCLIM models based onthe distributions of endemic species discussed here.

Understanding patterns of biodiversity,endemism and species abundance in theWet Tropics is a study of limiting factorsat various geographic and time scales. Ona regional scale, biodiversity has beenlimited by contractions in the extent ofrainforest area during the Pleistocene, andsubsequent episodes of expansion, bothwithin the region (Williams and Pearson1997) and between the regions (Winter1988). Local assemblage structure andlocal species diversity are limited by theway vegetation structure varies from placeto place within an area (Williams et al.2002). Recent results (Williams

3.15 Impacts of global climate change on the rainforest vertebratesof the Australian Wet Tropics

Stephen E. Williams1

1 Rainforest CRC, School of Tropical Biology, JamesCook University, Townsville, Queensland 4811

Rainfall seasonality (CV mean monthly)

10090807060

Me

an

no

in

div

idu

als

pe

r ce

nsu

s

60

50

40

30

20

Subregion

Paluma

Kirrama

Carbine

Atherton

Total Population

Rsq = 0.3439

Figure 3.15.1. Rainfall seasonality and bird abundance


unpublished data) show that the abundanceof a species at local and at landscape scalesis severely limited by short-term bottlenecksin the supplies of resources. Variability ofrainfall from month to month is the mostsignificant climatic variable explainingpatterns of bird abundance (Figure 3.15.1).This is despite the fact that bird speciesrichness and species composition do notvary across the same gradient. Similarly,variability of temperature from month tomonth limits the species richness andabundance of reptiles (Figure 3.15.2). Shortperiods of dry weather limit insect biomass(Frith and Frith 1985) and probably fruitbiomass. A plausible hypothesis toexplain these patterns is that shortbottlenecks in available resources limit thepopulation sizes of insect-eaters and fruit-eaters.

Basically, the population sizes and diversity ofspecies in a Wet Tropics habitat both declinewhen climate differs strongly from season toseason. Variation between seasons is predictedto increase under greenhouse conditions.

Microhylid frogs are another group whereregional patterns of diversity are stronglyrelated to consistent moisture levelsthroughout the year, and limited by low rainfallin the dry season (Williams and Hero 2001).Increased rainfall variation between seasons anda lifting cloud base both have seriousimplications for the long-term survival of mostspecies of microhylid frogs.

Preliminary BIOCLIM modelling of thegeographic distributions of a representativecollection of vertebrates unique to the WetTropics suggests that global warming will havesevere effects on the long-term survival ofmany species. A 1°C increase in meantemperature is considered a certainty and eventhis small increase is predicted to decrease therange of these rainforest species to a mean of63% of their current range. A rise of 3.5°C(average scenario predicted) will reduce rangesizes to a mean of 11% of current area andwill completely remove the bioclimatescurrently occupied by 30 species of endemicvertebrates. There is a real possibility ofbetween 30 and 50 species becoming extinct

this century. Most upland species will disappearif average temperatures increase by somewherebetween 1°C and 5°C (Figure 3.15.3 showsexamples of predicted declines in range size).This would be calamitous in a region that hasbeen preserved as a World Heritage Area mostlyon the basis of the biodiversity valuesrepresented by these regionally endemic species.

In the models discussed above, only the effectsof temperature have been considered. The effectsof climate change will conceivably be muchmore severe. Increased CO

2 concentrations will

Temperature seasonality

6420–2–4–6–8

Re

ptile

ab

un

da

nce

200

100

0

–100 Rsq = 0.4134

Figure 3.15.2. Relationship between reptile abundanceand seasonality of temperature

Figure 3.15.3. Rate of decline in distribution area offour species as mean temperatures increase

0

10

20

30

40

50

60

70

80

90

100

110

current 1 1.5 2 2.5 3 3.5 4 4.5 5

Temperature increase

Core

are

a (

% o

f curr

ent are

a)

A. katherina

A. adustus

C. laevis

C. ornatus

3.15 Impacts on rainforest vertebrates


reduce the nutritional value and increase thetoughness of most foliage (Kanowski 2001).This will have significant detrimental effects onthe abundance of leaf-eating species (ringtailpossums and many insects, for example).Furthermore, predicted changes in geographicdistribution will move species off nutrient-rich,basaltic soils and onto increasingly poorgranitic soils at higher altitudes. It has alreadybeen shown that rainforests on these poorer soilssupport sparser population densities of ringtailpossums and tree kangaroos (Kanowski 2001).Increasingly unpredictable rainfall may havesignificant effects too. Insect biomass hasalready been shown to decline during dry months(Frith and Frith 1985). These falls in insectbiomass will probably have large impacts on

insect-eaters and many other ecologicalprocesses at all levels of the food web. If thecloud base rises higher above the ground, speciesthat need consistently wet conditions will beaffected (microhylids, litter skinks, soilinvertebrate faunas, microbes, etc.). Then alllitter-feeding insect-eaters (many species ofbirds, skinks, microhylid frogs, dasyuridmammals, bandicoots, etc.) will be affected.Longer dry seasons will probably affect both theyield and the life-cycles of fruiting plants. Thesechanges will undoubtedly cause many flow-oneffects across the trophic levels (for instance, adecline in fruit biomass might result in declinein fruit-eaters that in turn will affect seeddispersal and plant recruitment processes).

The Green-eyed Treefrog (Litoria genimaculata) (males3–4 cm long and females 4–7 cm long), is endemic tothe Wet Tropics and is relatively common throughout therainforest. Males generally congregate along slow flowingstreams. However females are often spotted throughoutthe forest, even at the tops of the canopy. (Photo: StephenWilliams)

Lampropholis robertsi (4 cm snout-vent) is a highaltitude, regionally-endemic skink. It is generallyrestricted to above 900 m and has a very limitedand fragmented distribution on mountain tops in thenorthern mountain ranges of the region. (Photo:Stephen Williams)

3.15 Impacts on rainforest vertebrates


Variability in climate is a natural phenomenonthat has occurred throughout the Earth’s history.However, the warming of the Earth’s atmosphereover the last decade or so appears to beoverriding these natural climatic fluctuations,and has led to serious alarm. The causes of theserapid changes in the atmosphere are not entirelyclear, but they are most plausibly explained bya range of human-induced influences, such asincreased concentrations of atmospheric carbondioxide and other ‘greenhouse’ gases.

There has been considerable speculation aboutthe effects of global warming on activitiesranging from the future of tourism (for instance,Australia’s ski resorts), agriculture, forestry andhuman health, to the long-term survival of low-lying and coastal towns and nations. The effectsof these putative changes in climate across theseareas and activities vary from being beneficialto benign to potentially catastrophic.Considerable concern has also been raised aboutthe potential effects of climate change uponbiodiversity, with commentators suggesting thatglobal warming is likely to have severeconsequences for many species and biologicalcommunities (Newell et al. 2001). This paperdescribes a pilot project undertaken by theFlora Ecology Research Section at the ArthurRylah Institute to examine the potential effectof enhanced greenhouse gases (that is, globalwarming) upon selected Victorian vegetationcommunities. This work builds upon similarstudies conducted at the Institute in the past.

The bioclimatic profiles of 12 different plantspecies were modelled using ANUCLIM 5.1software. A major innovation of this study wasthe integration of up-to-date regional climatemodels (courtesy of CSIRO Division ofAtmospheric Research) into the process. This

has allowed, for the first time, an examinationof different combinations of both proposedchanges to the global climate (that is, globalclimate models) and regional climate models thatapply the global climate at the local level in arealistic manner. Using this new approach it hasalso been possible to model the bioclimaticprofiles of species at five-yearly increments fromthe years 1990 to 2100.

Twelve plant species were selected either asrepresentative of specific plant communities (forinstance, swamp heathlands or alpinecommunities), or as individual species that werelikely to decline or increase their range in a warmerenvironment. Predictions (to the year 2100) weremade about the responses of these species tomodelling processes. Results indicate that thebioclimatic profiles (that is, the amount of landsurface in Victoria that is likely to provide theappropriate climate for a species) decline foreach species over the time period and withineach group of scenarios. These results suggestthat the models may be overspecifying oroverconstraining the bioclimatic profiles ofspecies.

A suite of potential refinements has beensuggested that essentially lead to a tighterselection of the bioclimatic variables that wouldbe used for any particular species. This selectionwould be based upon both relevant ecologicalinformation and the biophysical informationderived from ANUCLIM. Also, the modeloutputs could be incorporated into a geographicinformation system (such as ARCView), which

1 Arthur Rylah Institute for Environmental Research,Department of Sustainability and Environment,PO Box 137, Heidelberg, Victoria 3084

3.16 Effects of ‘greenhouse’ warming scenarios upon selected Victorianplants and vegetation communities

Graeme Newell1, Peter Griffioen1 and David Cheal1


allows for the development of more realisticscenarios for individual plant species andcommunities. These scenarios would be basedupon our current knowledge of soils suitable forparticular species, the distribution of remnantnative vegetation across Victoria (that is, the reallikelihood that a species can alter its distributionover the next 100 years), potential areas of

salinity discharge, and so on. The incorporationof these refinements should lead to greenhousemodelling procedures producing outputs thathave a biological reality. Then Victoria’s landmanagers should be able to plan for the futurein a more strategic and informed way. Thefurther refinements and methodologicalinvestigations are being undertaken currently.

3.16 Effects on Victorian plant communities


Patterns of climate change

There is firm evidence of recent climate changein antarctic and subantarctic regions. Trendsshow significant increases in annual surface airtemperatures on all subantarctic islands wheredata have been collected. Changes in meanannual temperatures have been as high as +1.5°Cover the last 30–50 years. These are among themost rapid rates of increase observed in the world.

On Australia’s subantarctic Heard Island, anincrease in mean annual surface air temperatureof around 1.3°C over a 50-year time period(1948–2001) has been calculated (using linearregression models). Associated with thistemperature increase there has been extensiveglacial retreat. Between 1947 and 2000,approximately 35 km2 of new terrain, includingseveral large lagoons, has been exposed by iceretreat. This represents nearly 10% of the total areaof the island. Using similar analysis methods, asimilar rate of increase of annual surfacetemperature was reported for Macquarie Island,although with a more conservative analysistechnique (ARIMA time series) this rate wasscaled back to 0.3°C over 50 years. It still representsa total increase of around 16% and is within therange of current global change models.

On continental Antarctica, many areas areexperiencing warming, including Australia’sCasey station where the calculated trend inannual surface temperatures for the period 1959–1996 is +3.1°C/100 years. In other parts ofAntarctica, very rapid warming has beendocumented on the Antarctic Peninsula, but localcooling has occurred in the Dry Valleys region,with increased surface ice on lakes and associatedreduced primary productivity.

Signals in ice-sheet mass balance (significant toglobal sea-levels) are mixed. There is thickeningin some areas and thinning in others, but the mainmessage is that changes to the ice sheet can occurquite rapidly and the processes that cause thesechanges are poorly understood.

Apart from changes in temperature regimes,there are changes in other biotically-relevantenvironmental parameters such as precipitation,wind speed, evapotranspiration and UV. OnMacquarie Island mean monthly precipitationhas increased, while on other subantarcticislands, such as South Africa’s Marion Islandand France’s Iles Kerguelen, precipitation hasdecreased dramatically. In Antarctica, theincidence of rain has increased in somelocations. The significance of this is that rainincreases the availability of free water to life.Conversely, temperature increase can reducelocal supplies of snow or ice, lessening springand summer melt run-off.

Inter-annual variability is very pronounced in theregion and impacts of climate change may be moresignificant when considering the extreme eventsin contrast to the trends. The summer of 2002–2003 was significantly anomalous, with droughtson subantarctic islands, extreme summer sea-iceextents in some areas, and heat waves and flashflooding in the Dry Valleys and at the South Pole.At McMurdo a new maximum record of 10.5°Cwas recorded. Temperatures in January exceeded10°C for more than five days—a temperatureapproximately 11 standard deviations above the

3.17 Impact of climate change on terrestrial antarctic andsubantarctic biodiversity

Dana Bergstrom1

1 Australian Antarctic Division, Channel Highway, Kingston,Tasmania 7150


mean for the month (–3.2 ± 1.48°C). On theeastern side of the Antarctic Peninsula,anomalous warm conditions contributed to therapid and unparalleled break-up of an ice shelf(Larsen B) leaving newly revealed benthic areasavailable for colonisation. As unusual events likethese become more frequent and intense, thereis potential for ecosystems to undergo rapidchange in structure and composition.

Impacts of climate change on biodiversity

(i) Impacts on current taxa: population level

Increased ambient temperatures (and for plants,increased CO

2) will allow many species to

increase performance, improve reproductivecapacity and increase their distribution. Manyspecies are not currently living in optimalconditions. On the Antarctic Peninsula, thepopulations and distributions of two floweringplant species have shown dramatic increasesover recent decades. On Macquarie Island, thereis evidence of a bank of viable propagules oflowland species in upland areas, ready to growwhen climatic conditions permit. On HeardIsland there has been a trebling in the numberof Black Browed Albatross in approximately

50 years. This appears to be related to increasedbreeding success because of a milder climate andmore food being available as a result of newfishing practices around the island.

(ii) Impacts on current taxa: communitylevel

The enlarging areas that have been freed ofsurface ice by glacial and snow-field retreat areallowing, and will continue to allowcommunities to expand their distributions. OnHeard Island, land newly exposed during the lastcentury has been rapidly colonised by plants andassociated animals, increasing the biodiversityin these areas. However, areas where water willbecome less available will lose biodiversity.

Interactions between species, particularlyspecies competition, pathogen–host interactionsand predator–prey relationships, also respond toaltered climatic conditions. Ultimately theseinteractions will alter community dynamics. Forexample, on Heard Island, a native creeping herbappears to be out-competing and overgrowingthe dominant plant species (a cushion plant) andis found in many more plant communities nowthan in the mid-1980s. The warmer conditions

Heard Island—one of Australia’s World Heritage-listedsubantarctic islands (Photo: Dana Bergstrom)

3.17 Impacts on terrestrial antarctic and subantarctic biodiversity


are probably reducing the ratio of growth torespiration in the cushion plant, and it will retreatto higher altitudes (Heard Island is just under3000 m in altitude) as it has done on warmerislands. On Macquarie Island, a related butendemic cushion plant species may not havespace to retreat when lowland communitiesmigrate to higher altitudes, because the island islow in profile (maximum altitude approx 400 m).As a result, the species may be lost in the future.Little is known about the temperatureperformance of a newly recorded plant virus.This virus does not appear to kill the host plantspecies, but the balance could be disrupted inthe future.

(iii) Impacts on current taxa: alien impact

Better performance and wider distributions arealso expected in many alien species that havebeen introduced to the region. It is predicted thatthe status of many of these species will alter frompersistent to invasive, ultimately leading to areduction in indigenous biodiversity. OnMacquarie Island, rats are being found in uplandherbfields where they were previouslyunrecorded, and there is anecdotal evidence ofrats now having two litters per year comparedto only one in the past. Rats are now having anegative impact on the reproductive success ofthe regional endemic megaherb Pleurophyllumhookerii.

(iv) Impacts on ecosystem processes

On subantarctic islands currently, nutrientturnover depends on the concentrations ofdetritivores (usually earthworms and insects)because conditions are too cold and wet forsubstantial bacterial decomposition. Peataccumulation is the norm. Increases in surfaceair temperatures may alter this balance. If so,the change would have flow-on effects intocommunity structure and dynamics.

(v) Impacts on migration

It can be expected that as the climate warms,there will be successful migrations of species.However, the difficulty for scientists is todistinguish between alien arrival and naturalarrival. The region’s biodiversity is recognisedas missing functional groups. Elements of thesefunctional groups can be expected to coloniseand establish in the future—woody uprightshrubs on subantarctic islands, for example,which would lead to major change in communitybiodiversity. The threat of introduction of morealien species is increased with climate warmingbecause the number of species that cansuccessfully establish is increased.

Evidence is emerging that the formation of theantarctic ozone hole is having an impact. Duringspring, approximately two-thirds of theprotective ozone layer dissipates over thesubantarctic and antarctic region, leading togreater penetration of UV-B radiation to theEarth’s surface. Maximum levels of UVradiation are occurring much earlier in the seasonthan they did in the past, and lower and moredamaging wavelengths are penetrating to theEarth’s surface. Among impacts identified arealterations in the production of protectivepigments in some species, and reductions in thepopulation sizes of insects.

Conclusion

Understanding the impact of climate change onsubantarctic and antarctic biodiversity willremain one of the Federal Government’santarctic research priorities for the near future.Models of these ecosystems will be simpler thanmodels for continental Australia. They are partof ongoing research within the internationalRiSCC program (Regional Sensitivity to ClimateChange in Antarctic Terrestrial and LimneticEcosystems: http://www.riscc.aq).

3.17 Impacts on terrestrial antarctic and subantarctic biodiversity


Introduction

There is now clear evidence that the relativelymodest climatic changes over the past centuryhave already had significant impacts on thedistribution, abundance, phenology andphysiology of a wide range of species. Recentreviews have documented many instances ofshifts in species distributions towards the polesor upwards in altitude, and progressively earliermigrations and breeding (Hughes 2000, McCarty2001, Walther et al. 2002, Parmesan and Yohe2003, Root et al. 2003). The studies documentedin these compilations are almost exclusivelyfrom the Northern Hemisphere, and an extensivereview of species responses to recent climaticchanges by the IPCC includes no Australianexamples at all. As climate changes in Australiaare consistent with global trends (see Howden,this report), it is likely that the lack ofdocumented impacts is not because Australianspecies have been unaffected, but rather becauselong-term data-sets in which such trends couldbe detected are scarce (Westoby 1991). Our lackof systematic, long-term monitoring data ongeographic distribution, phenology and othercritical responses of Australian species willseriously hamper our ability to predict, adapt to,and mitigate the impacts of climate change onAustralian biodiversity.

We need to devise a coordinated nationalindicator program to monitor the impacts ofclimate change in Australia. Therefore, one ofthe main aims of this workshop was to assembleexperts from a wide range of disciplines to discussspecies and systems that could form the basis ofsuch a program.

What is an indicator?

An environmental indicator is a species or groupof species that responds predictably andsensitively, in ways that are readily observed andquantified, to an environmental disturbance.Indicators are generally selected to act assurrogates of at least a subset of other organismspresent in the same habitat. Indicators have beenclassified into a number of categories (Jenkins1971, Spellerberg 1991):

� sentinels—sensitive organisms introducedinto the environment as early-warningdevices,

� detectors—species occurring naturally in anarea of interest that may show measurableresponses to environmental change such aschanges in distribution or behaviour,

� exploiters—species whose presenceindicates the probability of disturbance orpollution,

� accumulators—organisms that take up andaccumulate chemicals in measurablequantities,

� bioassay organisms—organisms selected foruse as laboratory reagents to detect thepresence of particular chemicals.

Of these five categories, it is anticipated thatdetectors will initially be the most useful typeof indicator for the purposes of monitoringclimate change impacts. However, monitoringprograms using sentinels may also be an option;they would provide an element of experimentalcontrol over the variation between individualsand environments, and therefore increase thesignal-to-noise ratio.

�� !��

Lesley Hughes1

1 Dept of Biological Sciences, Macquarie University,North Ryde, NSW 2109


Indicators of climate change impacts

Species may respond to climate changes in anumber of ways, including physiologically,phenologically (by adjusting the timing of lifecycles) and in geographic distribution.

(i) Physiological indicators

Many physiological processes in plants andanimals are sensitive to changes in greenhousegas concentration (notably CO

2) and climate. The

processes of photosynthesis in green plants andrespiration in both plants and animals respond toCO

2 concentrations and temperature. Other

physiological characteristics that may respondto climate change include diapause (dormancy)in insects and the breaking of dormancy in plants.

(ii) Phenological indicators

The life cycles of many species are stronglydetermined by climatic factors such astemperature and precipitation. Warmerconditions in spring as a result of climate changemay lead to earlier onset of growth and floweringin plants. Species with a wide geographicdistribution may not show the same phenologicalresponses over the whole of their range;responses may show a gradation with latitudeor altitude. Earlier emergence and a longergrowing season may enhance the production andviability of seed in some plant species. Species thatreproduce several times in a year (some weeds,many insects) may be able to produce moregenerations per year. Spawning in amphibians andegg-laying in birds are also expected to occurearlier in the season.

(iii) Distribution and abundance

Physiological and phenological responses toclimate change, in combination with alteredinteractions between species, will influence therelative abundance and geographic distributionsof most species. It is expected that the ranges ofmany species will gradually shift to the south(in the Southern Hemisphere) or upwards in altitudeas climate zones change. These shifts will occurthrough the progressive extinction of localpopulations in areas that become unsuitable, andthe progressive colonisation of new areas.

Criteria for selecting indicators

Once a list of potential indicators is produced,the following criteria can be used as a series offilters (as suggested by de Groot et al. 1995) torefine the selection.

Climate sensitivity

Appropriate indicators will be those specieswhose distributions, physiology or life cyclesare predominantly sensitive to climate(especially temperature and precipitation) andless vulnerable to other changes inenvironmental conditions (such as land-usechange, acidification, pollution). Species thatrespond to accumulated changes may beparticularly suitable.

Availability of historical data

Where possible, indicators should be chosenwhere previous data have been collected. Thedata will serve as a baseline against which tomonitor future change. Unfortunately, thiscriterion will rarely be met for Australian speciesbecause of the lack of long-term monitoringstudies available.

Position within geographic range

Climate-related changes in performance will bedetected earlier at the boundary of a species’geographic range than at its centre. Shifts in theboundaries between vegetation types (ecotones)may be particularly sensitive indicators.

Dispersal capacity

Species with the greatest mobility (flying animalspecies, plants with small wind-dispersed seeds)will be the most capable of shifting with climatezones and should thus indicate changes soonerthan species with limited dispersal capacity.

Functional position in the ecosystem

Ideally, an indicator should be representative ofa whole functional group (such as pollinators,or top predators). Changes in the response ofthe indicator could then be more readilygeneralised to other species.

4. Indicators


Suitability for monitoring

Ideally, the indicator species should be easy toobserve, recognise and identify. The cost ofmonitoring it should be minimal, andmeasurements should be readily repeatable.

Recommendations

1. Further evaluation of data-sets identifiedat the workshop

A systematic survey of long-term biologicaldata-sets available in Australia should beundertaken to investigate whether usefulbaseline data exist, against which future, climate-induced change can be evaluated. This processwas begun at the workshop. Resources areneeded so that the usefulness of the suggesteddatabases can be explored further.

It is suggested that experts in the fields coveredby the databases be surveyed to provideinformation on the following issues:

� extent (number of years covered, number ofspecies or regions)

� reliability (consistency of data collection)

� format (electronic vs hardcopy)

� sensitivity of species to climate (especiallytemperature and precipitation) in relation toother environmental variables (such aspollution, harvesting and landcover change)

� availability (whether privately or publiclyheld, and cost of data retrieval).

2. Survey of additional data-sets

The workshop participants came from a broadrange of fields, but some important areas such

as freshwater and estuarine ecosystems were notrepresented. It is recommended that experts fromthose areas be consulted, so that data-sets forthose fields can be evaluated as described above.

3. Evaluation of potential indicators withouthistorical data

Potential indicator species were identified at theworkshop (Table 4.1) and in subsequentdiscussions. For several of these, there is littleor no historical data available about theirdistribution or phenology; that is, they fit all thecriteria outlined above except the second. Forthese species, it is recommended that furtherinformation be collected from both the publishedliterature and relevant experts, so their potentialusefulness for future monitoring programs canbe evaluated, given that baseline data are still tobe collected.

4. National monitoring program for climatechange indicators

Once the previous three recommendations havebeen acted on, we will need to consider selectinga subset of indicators to form the basis of anational monitoring program. It is anticipatedthat some of these indicators will be thosealready being monitored (coral bleaching, alpinevertebrate distribution, for example). For others,there will need to be a commitment to begincollecting the baseline data.

An excellent model for a national monitoringprogram is the set of 34 climatic, environmentaland socio-economic indicators of climateimpacts selected in the UK under the auspicesof the Department of Environment, Transportand Regions (DETR) (Cannell et al. 1999, http://www.nbu.ac.uk/iccuk/).

4. Indicators


(continued next page)

4. Indicators

Table 4.1 Potential indicators identified at the workshop and in subsequent discussions

Indicators Sensitivity Comments, references and contacts

Physiological indicators

Coral bleaching Cumulative degree-days of Ongoing research program by Australian Institute of Marinesea-surface temperatures Sciences (AIMS) (e.g. Jones et al. 1997, Berkelmans & Oliverabove various thresholds 1999, Hoegh-Guldberg 1999, Lough 2000). Acropora spp. and

other inshore coral species noted as being particularly sensitive(Ove Hoegh-Guldberg, pers. comm.).

Calcification in Ocean temperatures Annual variation in the density of calcium carbonate (CaCO3)

long-lived corals skeletons in some massive coral species such as Porites (Lough& Barnes 1997, Lough 2000) can be used retrospectively tomonitor coral growth in a way analogous to the study of tree-rings.For each 1oC of temperature increase, calcification increases byapprox. 0.3 g/cm2/year and linear extension of the coral increasesby approx. 3 mm/year.

Vegetation thickening CO2 and climate impacts Vegetation thickening over the last 50–100 years has beenrecorded in environments such as semi-arid woodlands (Fensham& Fairfax 2002), and eucalypt savannas (Bowman et al. 2001)using historical collections of aerial photos, direct measurementand other techniques. Further measurement of this indicator hasparticular relevance to studies of carbon sequestration.

Distribution as an indicator

Incidence and Temperature and Arthropod vectors (mainly insects) expected to move south, andgeographic location precipitation impacts on previously-eradicated diseases such as malaria may re-emergeof arthropod-vectored mosquito vectors (Walker 1998)diseases, e.g. malaria,Ross River fever

Boundary of C3/C

4CO

2, temperature and Recorded changes in the distribution and abundance of grass

plant species in rainfall change species reflect changes in temperature, rainfall and CO2

subtropics concentration (Howden et al. 1999) due to changes in competitionbetween C

3 (cool season) plants and C

4(warm season) grasses

Boundary between CO2, temperature and Expansion of rainforest into eucalypt forest and grasslands is wellrainforest and precipitation change documented in Queensland over the past two centuries (e.g.eucalypt woodland Harrington and Sanderson 1994, Fensham and Fairfax 1996,

Hopkins et al. 1996). Recent invasions of warm temperate rain-forest species to higher altitudes in northern NSW, and expansionof Nothofagus into eucalypt woodland on plateaus in the BarringtonTops region have also been documented (e.g. Read and Hill1985). The relative roles of recent atmospheric and climaticchange, relative to changes in fire regimes and long-term warmingtrends after the last glacial maximum, are unclear at present.

Tree establishment Temperature, duration and Encroachment by Eucalyptus pauciflora into subalpine grasslandsin sub-alpine zones depth of snow cover near Mt Hotham, Victoria, has been documented by Wearne and

Morgan (2001). Contact: Dr John Morgan, La Trobe University.

Composition and Temperature, duration and Trends in shrub senescence reported for shrublands in thestructure of alpine depth of snow cover Victorian Alps in plots fenced to exclude grazers since 1945plant communities in (Wahren et al. 1994). These trends have been interpreted mainlylong-term grazing as slow recovery from grazing disturbance and post-fireexclosures regeneration, but may also reflect recent climatic changes.

Estuarine/freshwater Sea-level rise and Dramatic expansion of some tidal creek systems in NT hasvegetation boundary saltwater intrusion occurred since the 1940s. In the Lower Mary River system, two

creeks have extended more than 4 km inland, invading freshwaterwetlands (Woodroffe and Mulrennan 1993; Bayliss et al. 1997,Mulrennan and Woodroffe 1998). Rates of extension of saltwaterecosystems inland in excess of 0.5 km/yr have been measured(Knighton et al. 1992). Over 17,000 ha of freshwater wetlands havebeen adversely affected and a further 35–40% of the plains isimmediately threatened (Mulrennan and Woodrofffe 1998).Landwards transgression of mangroves into saltmarsh environmentshas also been documented in the estuaries of Queensland, NSW,Victoria and SA over the past five decades (Saintilan and Williams1999). Multiple causes are likely, but include sea-level rise.


Table 4.1 Potential indicators (continued)

Birds Temperature and Various databases, including the Australian Bird Census, held andprecipitation, sea-level rise collated by Birds Australia, may be analysed for climate-related(for sea bird nests) trends. The most suitable species will be relatively sedentary with

restricted ranges and/or specialised habitat requirements.Contact: Dr Mike Weston, Birds Australia.

Upland epiphytes Temperature and Epiphytes are dependent on occult (mist) precipitation.in wet tropics precipitation, seasonality Contact: Dr Dave Hilbert, CSIRO Sustainable Ecosystems,

of rainfall, length of dry Atherton.season, lifting of cloud base

Alpine vertebrates Temperature, duration and Monitoring by Ken Green, NPWS*, indicates shifts in vertebrate(macropods and depth of snowcover geographic ranges to higher altitudes over the thirty-year period toferal animals, 1999. Wildlife Atlas records indicate a higher maximum altitudinale.g. horses) distribution for all three macropod species and for four species of

feral mammals (Green and Pickering 2002).

Range boundaries Temperature, precipitation, The range of Pteropus poliocephalus, the grey-headed flying fox,of flying foxes in frost frequency has contracted south from its northern boundary by about 750 kmeastern Australia since the 1930s (Tidemann 1999). The black-heading flying fox,

P. alecto has apparently extended its range south by a similardistance. Contact: Dr Chris Tidemann, Australian National Univ.

Colony formation Temperature, El Niño Colonisation of new breeding sites to the south of the historic rangein seabirds frequency and intensity has been documented for at least 8 seabird species in the region

between the Houtman Abrolhos and the Naturaliste and LeeuwinCapes, off the coast of WA (Dunlop 2001). Some of the shiftsbegan as early as the 1920s, others in the 1950s and 1960s, whileothers did not begin until the 1990s. Impacts of El Niño via itseffect on the Leeuwin Current may be responsible.

Phenological indicators

Alpine plant phenology Temperature increases and Some data are available in herbarium records (collated in electronicsnow decreases, format by Dr L. Hughes, Macquarie University, from records at theadvancement of spring National Herbarium). Some areas have been identified as beinggrowing season suitable for ongoing monitoring of flowering onset.

Arrival of migratory Temperature increases and A trend towards earlier arrival of migratory bird species in the alpinealpine birds snow decreases, zone in the 1980s and/or 1990s, compared with the 1970s has also

advancement of spring been documented (Green and Pickering 2002). For the 11 birdgrowing season species for which there are sufficient data, the earliest record was

in the 1990s for five species and in the 1980s for four.Contact: Dr Ken Green, NPWS.

Egg-laying dates Temperature and precipitation Nest Record Scheme collated by Birds Australia (see above)of birds

Australian plague Temperature and precipitation Records held by Australian Plague Locust Commissionlocust emergence

Behaviour in the Temperature and precipitation A long-term study on the sleepy lizard Tiliqua rugosa by Michael Bullsleepy lizard, and colleagues (Flinders University) since 1983 has documentedTiliqua rugosa changes in mating behaviour correlated with climatic changes (Bull

and Burzacott 2002). Changes in the abrupt parapatric boundary oftwo reptile ticks for which the sleepy lizard is a host have also beenmonitored (Bull and Burzacott 2001).Contact: Dr Mike Bull, Flinders University

Economic/Agricultural indicators

Trends in honey Honey yields reflect changesyields by region in nectar and pollen production

of flowers, which in turn reflectboth climate and CO2-inducedchanges to plant physiologyand phenology

*NPWS = National Parks and Wildlife Service

4. Indicators


Modelling and analytical tools have beendeveloped to assess the effects of climate changeon biodiversity and biological function. Acapacity to make predictions can give policymakers and regional planning groups someconfidence to move forward, implement orchange current policy instruments, identifycritical thresholds of climate for importantbiomes, develop and test key indicators ofchange, and highlight iconic species orcommunities under immediate threat. In thefuture, modelling developments, particularly inprocess-based simulation models, will help usto better understand the effects of climatechanges on biodiversity and assess theeffectiveness of adaptation strategies.

Simulation models already exist that can be usedto assess the impact of climate change onecosystem function. The workshop focused onmodelling tools that estimate the geographicdistributions or ‘climatic envelopes’ of speciesor vegetation classes and how climate changewould alter these, thus affecting biodiversity.However, they are just one type of severalpossible modelling approaches.

Models of species distributions or habitats canbe dynamic, predicting time-dependentresponses to a changing environment, or static,predicting equilibrium responses to a fixed setof independent variables (Prentice and Solomon1991). Each of these types can be correlative(correlating observed presence and absence withobserved environmental variables), ormechanistic (incorporating the ecological andphysiological and/or other mechanismsresponsible for an organism’s responses to theenvironment) (Beerling et al. 1995).

Dynamic, mechanistic models have theadvantage of being explanatory. They can becapable of some extrapolation beyond theenvironmental range used to develop the model.However, they require the modeller to haveextensive knowledge about how the environmentconstrains a species’ distribution (usually notavailable), or else they generalise from what isknown about similar species. Dynamic,mechanistic models also require an accurateestimate of the time-course of climate change.However, the timing of climate change is quiteuncertain and, for most species, the best or onlyinformation about environmental constraints isgeographic presence data or, less frequently, bothpresence and absence. While explanation, in thesense of ‘bottom-up’ simulation, is anintellectually satisfying goal, we must also bepragmatic when faced with immediate needs, andrecognise that explanations, in a reductionist,mechanistic form, are not always available or evennecessary. Consequently, static, correlative modelsare an important and widely used tool fordetermining the responses of organisms to climatechange (Guisan and Zimmermann 2000).

There are options (Howden et al. 1999) thatprovide intermediate pathways between the twoextreme forms of models. These intermediateapproaches combine existing knowledge (bothquantitative and qualitative) about ecological

��"� #��

��

Chris Chilcott1, David Hilbert2 and Mark Howden3

1. Climate Impacts and Natural Resources, QueenslandDepartment of Natural Resources and Mines,80 Meiers Rd, Indooroopilly, Queensland 4068

2. CSIRO Sustainable Ecosystems, Tropical ForestResearch Centre, PO Box 780, Atherton,Queensland 4883

3. CSIRO Sustainable Ecosystems, GPO Box 284,Canberra, ACT 2601


responses into simple, but process-orientedmodels. A number of approaches were identifiedand discussed at the workshop. Their advantagesand disadvantages are listed in Table 5.1

An important assumption of correlative modelsis that the species’ distribution is currently inequilibrium with climate and is limited primarilyby climate. In many cases these assumptions arenot valid or cannot be tested. When applying acorrelative model to altered climates, the modellerassumes that the climate variables continue to becorrelated with whatever unknown variablesactually control the distribution.

Correlative models are also highly sensitive tothe accuracy of the data about distributions, onwhich they are based. This, plus the possibilityof ‘over-fitting’ the model to too many climatevariables, can lead to substantial underestimatesof the potential range of species. Whencorrelative models model vegetation types(rather than species), the issue of change incommunity composition in the face ofenvironmental change is seen as important insome circumstances. Vegetation types definedby structure (Hilbert et al. 2001b) or plantfunctional types may be a more robust way to

Table 5.1. Types of modelling approaches for assessing climate change implications for biodiversity, examplesof specific models and approaches, and some of the disadvantages and advantages of each

Modelling approaches Examples Points for and against (pros and cons)

Correlative approaches BIOCLIM As descriptive approaches, these models are unable to account forCLIMEX functional relationships. Instead they develop relationships from a species’DOMAIN existing environmental domain and presence. They predict other potentialHabitat environmental domains where the species may occur, and changes in its

distribution under altered climatic conditions. In most cases, the models arebased on presence of species and cannot account for other determinants ofspecies distribution: soils, species interactions and competition, ormanagement and disturbance. Further, these approaches cannot accountfor the potential effects of faster change in CO2 concentration. The modelsare user-friendly and only require data on climate and distributions (ascomprehensive as is feasible). However, if applied naïvely, the models canresult in major over-estimation of likely impacts.

Statistical approaches General linear Similar to correlative approaches except they use presence, absence andmodels. General abundance data. They can also include categorical variables. They can beadditive models. used to develop some analytical understanding of ecological limitations.Artificial neural These approaches are not ‘packaged’ and require a good understanding ofnetworks. statistical approaches. They are thus not particularly user-friendly. As with

correlative approaches, it is assumed that the sample represents the entirehabitat, that the species’ distribution is principally determined by climaticconditions, and that the distribution is at equilibrium. More sophisticatedapproaches such as Artificial Neural Networks are considered something ofa ‘black box’ and require some experience in their application.

Simple simulation models GRASP These models have been developed to gain an understanding ofecosystems and interactions, based on basic ecological and/or physiologicaldata. They simulate the performance of an organism within an environment,based on factors such as climate, CO2 concentration, soil and management.Their users require an understanding of ecological interactions. The modelsmay be limited by available data, but will highlight limitations in theunderstanding of climate–species distribution interactions. They have thecapacity to generate hypotheses and to test uncertainty with different climatechange scenarios. Models of this type are important because theyincorporate at least some management options, can explore someadaptations, and can deal with a large range of interacting factors such asCO2 concentration, soil type, etc.

Bayesian approaches Bayesian approaches are used in conjunction with other models. For: theyare suitable for use when there is high uncertainty, and can build informationfor policy and science because they force the users to think about the natureof the data and the questions. Bayesian thinking can be thought of asparallel with processes of learning. Against: these approaches are notgenerally well understood, can involve highly complex statistics, and are notconsidered useful by everyone. Computer time is also an issue for someapplications.

5. Modelling


model future vegetation, but this approach willnot provide information on composition.

When properly implemented and interpreted byusers knowing the limitations, statistical andcorrelative models can be a useful ‘first filter’for identifying locations and taxonomic groupsthat are most (and least) threatened by climatechange. However, they cannot deal with 1) CO

2

concentration, which is the most certain aspectof climate change and likely to be very importantwith both positive and negative implications, 2)combinations of climate, CO

2 concentration, and

soils that are outside the range already existing,3) dispersal abilities of a species, and 4)management adaptations.

Where ‘hot spots’ or ‘hot species orcommunities’ are identified and where a widerrange of ecological information is available, furtheranalyses using simple simulation models (Howdenet al. 1999) can more robustly identify impactslikely to result from multiple changes that havenot yet been seen (in CO

2 concentration, rainfall,

temperature, frost, for example). They can assesswhether associated adaptation actions will beuseful or applicable. Where such analyses canbe made, the resulting knowledge could be used

to plan monitoring programs and identifypriorities for adaptation and options.

The following issues should be considered whenchoosing the most appropriate modellingapproaches to use:

� How will the different types of modelsaddress hitherto unexperiencedcombinations of climate variables, such ashigh rainfall, high temperature andinteractions with soil factors, fire andmanagement?

� How do models address response toincreased CO

2 concentration for which there

is no precedent in the last 400,000 years?

� Any modelling program requires themonitoring of change as way of refiningoutputs and predictive capacity.

Previous climate change impactassessments for Australia

Several of these modelling approaches have beenapplied previously to Australian ecosystems andbiota. They have been reviewed by Hughes (inpress) and are summarised in Table 5.2 overleaf.

5. Modelling


Table 5.2. Expected impacts of climate change and changes in CO2 concentrations on various ecosystemsand groups of organisms drawn from Australian studies (reviewed by Hughes in press)

Ecosystem/taxa Expected impacts Reference

Plant diversity in Little change in plant diversity as the benefits of higher CO2 levels offset Rochefort &

eastern Australia the deleterious effects of higher evapotranspiration under a +3oC and Woodward 1992+10% precipitation scenario

Tropical rainforest Most of the forest types examined will experience climates in the future that Hilbert et al. 2001aof north Queensland are currently more appropriate to some other structural forest type. Highland

rainforest environments (which are the habitat of many of the region’sendemic vertebrates) are predicted to shrink by 50% with only a 1oC warming.

Grasslands Some limited changes in the distributions of C3 and C4 grasses. Higher Howden et al. 1999temperature favours more southerly distributions, and higher CO2 favoursmore northerly distributions.

Northern Australia, The low relief of these areas means that even small rises in sea level could Woodroffe &extensive freshwater result in relatively large areas being affected by saltwater intrusion. The Mulrennan 1993,swamps and result would be expansion of the estuarine wetland system at the expense Bayliss et al. 1997,floodplains of present-day freshwater wetlands. Eliot et al. 1999

Riverine ecosystems Assuming that water allocation practices do not change, the projected Johnson 1998,and inland wetlands reduced river flows will reduce both semi-permanent and ephemeral wetland Hassall & Associates

vegetation substantially. They will reduce breeding events for colonial nesting 1998, Roshier et al.bird species and have negative impacts on many other taxonomic groups. 2001, Herron et al.

2002

Coral reefs Bleaching from raised sea-surface temperatures, hypo-osmotic stress if Coles & Jokiel 1992,the size of extreme floods increases, physical damage from increased Hoegh-Guldbergcyclone intensity. If higher CO2 concentrations in air lead to reduced alkalinity 1999, Leclerq et al.in sea water, substantial degradation of reef structure and composition is likely. 2002

Alpine areas Raised summer temperatures may not only increase the growth rates of Williams & Costinexisting shrubs but may also stimulate expansion of woody vegetation into 1994, Walter &areas now dominated by herbaceous species. If there is less snow cover, Broome 1998under-snow activity is likely to decline, but predation may increase andinterfere with hibernation of mammals.

Eucalypt species 53% of 819 eucalypt species have current ranges spanning less than Hughes et al. 19963oC of mean annual temperature. 41% have a range spanning less than2oC and 25% have ranges spanning less than 1oC. In addition, 23% ofspecies have ranges of mean annual rainfall that span less than 20%variation. If even a modest proportion of present day boundaries reflectthermal or rainfall tolerances, substantial changes in the Australian treeflora may be expected in the future.

Dryandra and The bioclimates of 28% of Dryandra species are predicted to disappear Pouliquen-Young &Acacia spp. in completely with a 0.5oC warming; this estimate increases to 66% with a Newman 2000Western Australia 2oC warming. The bioclimates of 59% of Acacia species would disappear

with a 1oC increase and all would disappear with a 2oC warming. In general,species are not predicted to track moving climate zones across the landscape,due to soil constraints, but instead to shrink to a smaller range within theircurrent distribution.

Mitchell grass Suitable and marginal bioclimates are predicted to decrease in northern Chapman & Milne(Astrebla lappacea) areas by 23% and 44% respectively. However, only about 50% of these 1998

areas occur on suitable soils. Note that inclusion of CO2 effects maysignificantly change these outcomes.

Vertebrates Range reductions are suggested for the majority of species although a Brereton et al. 1995,few might increase their range. For example in south-east Australia, of Dexter et al. 1995,42 species studied, 15 may have no suitable bioclimate if there is a 3oC rise Chapman & Milnein temperature. For some species (such as Mountain Pygmy Possum 1998, Pouliquen-Burramys parvus, and some frogs) their bioclimate may disappear Young & Newmancompletely if mean temperatures rise 1oC or less. Higher CO2 concentrations 2000, Hilbert et al.will tend to reduce foliage quality below critical levels. 2001b, Kanowski

2001

Invertebrates The bioclimates of 92% of butterfly species are predicted to decrease, Beaumont &with 83% declining by at least 50% if mean temperatures increase by 2.1 Hughes 2002, Johnsto 3.9°C. Large changes in range are projected. About 10% of species & Hughes 2002studied are vulnerable due to particular life history characteristics. Foliagequality could be affected, as for vertebrates.

5. Modelling


Background

There are many good reasons, relevant tocommunity, science and policy, for developinga national policy program to address the likelyfuture impacts of climate change on biodiversity.This section identifies a range of factors thatcould usefully be considered in the developmentof any such program. It also identifies existingpolicy programs that may include opportunitiesfor both monitoring the impacts of climatechange and initiating adaptations.

Definitions

To avoid any confusion in terminology, thefollowing definitions are offered in the contextof climate change and biodiversity policy.

Abatement and mitigation—managementactions to decrease emissions of greenhousegases or increase their sequestration by theenvironment. Most climate change action todate has focused on this aspect of climatechange policy.

Impacts—the effect of increased greenhouse gasemissions and atmospheric CO

2 concentration

(and other global changes) on physical andbiological processes in both natural andmanaged systems.

Adaptations—responses that decrease thenegative effects and capitalise on positiveopportunities associated with impacts.Adaptations can be split into ‘autonomous’(internal, automatic system adjustments suchas evolutionary responses in natural systems)and ‘planned’ (where a deliberate interventionis made in an attempt to achieve a specific goal,recognising the change in environment).

Three information needs

To develop an effective policy response to thepotential impacts on biodiversity of climatechange, information will be needed, to:

� establish the need for a policy intervention,

� identify and compare alternative policyobjectives, and

� develop and evaluate various possibledelivery mechanisms.

It is likely that different types of informationwill be required for each of these needs.

� Information about the impacts of climatechange on biodiversity will provide the weightof evidence required to justify investment inadaptation. Indicators of climate change andanalysis of impacts could contributesubstantially to this weight of evidence.

� To identify and compare adaptationobjectives, there will need to be:

(a) some knowledge of the scope of climateimpacts; that is, just how large a potentialimpact is likely to be (areas, species, ecosystemchanges). Scientific research can help identifythe scope of changes.

(b) knowledge and assessment of biodiversity‘values’; that is, species, communities,ecosystems, ecosystem function, ecosystemservices. The Biological Diversity AdvisoryCommittee may choose to offer some guidanceon the values of alternative biodiversityoutcomes.

1 CSIRO Sustainable Ecosystems, GPO Box 284,Canberra, ACT 2601

��$� %��

Michael Dunlop1 and Mark Howden1


� To develop and evaluate alternative policymechanisms, policymakers will needinformation about the adaptation options thatare feasible for different situations, and theircosts and benefits. A range of generaladaptation actions, and some factors toconsider when assessing possible actions,were discussed at the workshop.

What is the scope of the problem?Establishing the need for action tomanage the impacts of climate changeon Australia’s biodiversity

A clear message needs to be sent to both thecommunity and governments, by policymakers,interested groups and researchers, about theimpacts of climate change on biodiversity. Themessage should be based on clear, and preferablyquantitative, information about the likelyimpacts. We need to be able to identify whichaspects of biodiversity (that is, which species,communities and ecosystems services) are mostlikely to be affected, the scale of the severity ofthe impacts (for instance, an alteration of thegeographic range of a few species, replacementof ecosystems, or the extinction of a largefraction of species) and the rate at which theseimpacts may occur.

Many in the community and government, andindeed many biodiversity researchers, do nothave a good understanding of the potentialimpacts of climate change on biodiversity. Forexample, there is broad community awarenessof coral bleaching, but there is confusionbetween coral bleaching and coral mortality. Ifwe collect information about current impacts,and monitor future impacts, we will be able toeducate the community about the nature andscope of likely future impacts. If there iswidespread understanding of the issues, there ismore likely to be support for public investmentin adaptation action. Information about currentand expected impacts is also needed to guidepublic and private biodiversity-conservationprograms and to direct biodiversity research.Some of these issues are covered below.

Stressed and pristine ecosystems

Ecosystems that are already stressed as a resultof human-induced or other disturbance are likely

to be particularly vulnerable to climate changeimpacts. Any added stresses of climate change arelikely to further decrease resilience. However,species in relatively pristine ecosystems may alsobe particularly vulnerable if barriers to upwards orpolewards migration are present (for example, ifthe pristine ecosystems are habitat islands, adjacentto ranges, deserts, oceans, cities, or clearedareas), or if they are affected by invasive speciesor other changes in species interactions.

Winners and losers

Many species are likely to be negatively affectedby changes to mean temperature, rainfall, CO

2

concentration and disturbance regimes, and wemay naturally tend to focus impact (andadaptation) assessments on those species.However, the greatest community and ecosystemimpacts may come from those species that arefavoured by changed conditions or disturbanceand interact with other species (for instance,competitors, predators, invasive weeds, etc.).Such species could be native or exotic.

Uncertainty in impacts

The issue of what actions need to be undertakenis clouded in uncertainty. There are similarities withmany other environmental issues in that there is atrade-off between requirements for short-termresponses (mitigation) and longer term responses(adaptation). Both have tangible costs andlargely uncertain, intangible benefits. In the caseof climate change and biodiversity, we can thinkof structuring a rational decision in the form ofa simple, commonsense equation that has fivecomponents.

The decision is a function of 1) the costs ofmitigation, 2) the benefits of mitigation, 3) thecosts of adaptation, 4) the benefits of adaptation,and 5) the interaction between these previousterms, because we would expect that a largemitigation effort is likely to reduce futureadaptation costs and vice versa.

Unfortunately, it is not currently possible toproduce adequate information to guide policy onany of the five components of the equation. Eventhe most studied components (immediate costsand benefits of mitigation) are hotly debated andunresolved. Economic analysts are highly

6. Policy


polarised in their views. Some say that mitigationwill benefit the economy while others suggest itwill precipitate major economic losses.Nevertheless, the mitigation components arelikely to bias the analysis due to factors such asthe amount of prior research, and concepts such asapplication of discount rates to future (dis)benefitswhich favour short-term components at theexpense of longer-term ones. Effective long-termpolicy decisions are likely to be assisted byinformation and uncertainty that is roughlybalanced between the elements.

While this uncertainty is a difficulty to bemanaged in developing adaptation policy,dealing with such irreducible uncertainty is not anentirely new problem for public policy. Forexample, defence policy has to be developed inthe context of considerable uncertainty and shiftingthreats and priorities. Similarly, many

agricultural policies and farm managementstrategies are geared towards coping with themarked climatic variability in Australia. Inaddition, governments around the world are nowbeginning to take significant action to reducegreenhouse gas emissions and prepare forclimate change, despite a very wide range in themodelled climate scenarios. In the face of thisuncertainty about precise climate outcomes, theyhave the confidence to act as a result of the massivevolume of good science coordinated by the IPCC.

Even so, the impacts of climate changes onAustralia’s biodiversity are uncertain. This is dueto uncertainties in 1) the climate science (ocean–land–climate interactions), 2) projections offuture emissions, and 3) knowledge of ecologicalresponses. While improvements in climate andecological science may reduce some of thisuncertainty, there will always remain uncertainty

Figure 6.1. Synthesis of risk assessment approach to global warming

The left part of the figure shows global warming based on the six SRESgreenhouse gas emission marker scenarios (Jones 2000) with the zonesof maximum benefit for adaptation and mitigation. The right side showslikelihood based on threshold exceedance as a function of global warmingand the consequences of global warming reaching that particular levelbased on the conclusions of IPCC Working Group II (IPCC 2001b). Riskis a function of probability and consequence. SRES = Special Report forEmission Scenarios.

Source: Roger Jones, CSIRO, unpublished, © CSIRO.

Almost certain

Highly likely

Least likely

Low probability, extreme outcomes

Damage to the most sen-

sitive, many positives

Increased damage to many

systems, fewer positives

Considerable damage

to most systems

Moderately likely

Probability Consequence

Core benefits of adaptation and mitigation

Probability — the likelihood of reaching or exceeding a given level of global warming

Consequence — the effect of reaching or exceeding a given level of global warming

Risk = Probability × Consequence

Vulnerable to current climateHappening now

6. Policy


in future emissions. At least half of the uncertaintyin global warming is due to uncertainty in emissionsscenarios. These are a function of economic, social,cultural, political and technological change: theyare not amenable to scientific analysis. This shouldwarn us against expecting too much frominvesting substantial resources aimed at reducingthe uncertainty in the climate science. Theinvestment can only reduce overall uncertaintyto a limited extent. Hence, impact and adaptationpolicy will always have to deal with considerableuncertainty.

One approach to dealing with this is to focus, atleast initially and where possible, on the lowerend of the range of likely future climate changes,and develop strategies to cope with its impacts.This has the added advantage of enhancing ourmanagement of existing climate variability. Atthe same time, concerted actions to reduceemissions could avoid the need for adaptationto changes in climate at the more extreme endof the spectrum (see Figure 6.1).

This approach would focus attention on speciesand ecosystems that are most likely to beaffected—that is, they are most sensitive toclimate change. We suggest that these specieswill also be more seriously affected shouldclimate change be greater than the minimum.However, if it becomes necessary to prioritisethe use of limited resources for adaptation,focusing on species that are most likely torespond to adaptation actions rather than thosemost threatened by climate change, then adifferent approach might be required.

Much of the uncertainty in ecological responsesto climate change results from interactionsbetween different aspects of climate change andthe complexity of interactions between species.For example, decreases in rainfall may imposestress on some plant species, but increased CO

2

concentrations may increase their ability to copewith less water. These responses may differbetween species, leading to changes in competitiveabilities and hence community composition. TheCO

2 concentration may also alter the quality of

some foliage, thereby affecting herbivores thatdepend on those plants. The direct effects of

climate change on one species may be amplified,mitigated or transmitted to other species as aresult of changed species interactions. Forexample, there may be alterations to habitat andto predator–prey interactions, competitiveabilities and host–pathogen dynamics.

Information on climate impacts

The nature of the impacts of climate change onbiodiversity can be determined by a combinationof monitoring of specifically chosen indicators(climate variables and biological variables),modelling (shifting bioclimatic envelopes,physiological and ecosystem responses) andexperimentation (climate manipulation andtranslocation). Future activity could includecollating and mining existing time-series data todetect any climate-change signals, and setting clearagendas for future monitoring and research basedon the information requirements identified forassessing impacts and adaptation.

Societal values

Impacts on biodiversity will affect the valuesociety gains from biodiversity, quite apart fromany desire to conserve it for other reasons. The‘look and feel’ of ecosystems will change asspecies composition changes. Dominant species,familiar species, and less common species may allchange; for example, farmers are already noticingand lamenting the loss of some woodland species.The attractiveness for tourists may change (forinstance, by coral bleaching, rising cloud forests,decreasing snow depths and extent, dryingwaterways). Wild harvest may be affected (theproductivity of fisheries may be affected, honeyyield may drop as it has in past droughts); theresilience of agro-ecological ecosystems and theprovision of other ecosystem services may alsobe affected. As the new science of identifyingand quantifying the benefits that accrue frombiodiversity develops, and the understanding ofthe biophysical impacts of climate changeimproves, a more detailed picture will emergeof the likely scope of the social and economicconsequences of climate change impacts onbiodiversity.

6. Policy


Adaptation objectives: what are we tryingto achieve?

The United Nations Framework Convention onClimate Change sets out as an objective thestabilisation of greenhouse gas emissions withina time-frame sufficient to allow ecosystems toadapt naturally to climate change. However,there is considerable evidence to suggest thatthe likely rate of climate change may be fasterthan the rate of natural adaptation (autonomouscoping through migration or evolution) for manyspecies and ecosystems. This could result insubstantial losses of biodiversity.

We need to decide on an appropriate objectivefor Australia in responding to the threat ofclimate-change impacts on biodiversity. To makethe decision, we need to explore variouscomponents of both the threats posed by climatechange and the responses of natural systems andtheir components. We suggest a three-foldobjective is required. In the short-term theobjective is to preserve sensitive elements ofbiodiversity against the immediate threat ofclimate change. In the longer term the objectivesare both to facilitate adaptation to climate changeand to mitigate against large changes.

This distinction is analogous to the differencebetween first aid and healing (see Table 6.1).These objectives recognise that changes tobiodiversity are inevitable, but that humanintervention could have a significant impact inreducing the negative consequences of thatchange. A separate, but equally criticalobjective, is to reduce the need for furtherpreservation and adaptation by reducing futureclimate change itself through emission control

and carbon sequestration. This is analogous topreventative health care.

Individuals within species are likely to respondto climate change in different ways. Hence, thereare likely to be changes in species’ genetics andpopulation dynamics, assemblages of species,communities, ecosystem structure andfunctioning, and the goods and services providedby ecosystems. Some of these entities may beeasier to preserve than others, and some may bejudged by society as being of greater value (seeTable 6.2). For example, even if all speciessurvive climate change, differing environmentalrequirements and dispersal rates may makespecific communities impossible to preserve innature. Some species may be lost and gainedfrom an ecosystem without substantiallychanging its structure and function. Similarly,one ecosystem may be replaced by another butwith similar provision of water and carbon-processing services.

Pragmatically, the various objectives need to beassessed in the context of limited resources beingavailable for investment on biodiversityconservation. There are interactions (bothcompetitive and synergistic) betweeninvestments on biodiversity and other naturalresource management issues. This raises theprospect of prioritisation and trade-offs inallocating resources. Is society prepared todecide that the conservation of some ecosystemsmay be too expensive to manage; or that thereshould be no effort to conserve a given species,so that a greater investment can be committedto another species that might have a greater long-term chance of adapting to a changed climate?

Table 6.1. Analogies between managing climate change impacts on biodiversity and healthcare management

Climate change action Healthcare action

Managing existing climate variability better and Primary prevention, reducing exposure to risk factors (e.g.reducing emissions giving up smoking)

Reducing other pressures on biodiversity; Health promotion (e.g. getting fit)increasing ecosystem resilience

Preservation of species suffering from the First aid and life supportimpact of climate change

Facilitating long-term adaptation Treatment of a condition; healing

Prioritising investment Triage

6. Policy


Table 6.2. Examples of different levels of biodiversity and comments on the possible practicality and value of preserving them

Entity Comment

Intra-specific, genetic diversity Important for future evolution and perhaps relatively easy to preserve for mostspecies which can maintain large populations, but difficult for those with small orrestricted populations.

Individual species The best recognised unit of biodiversity. Possible to preserve a given individualspecies with generic biodiversity conservation actions and some specific climatechange adaptation actions, but would be expensive for some species and difficultto assess for others (e.g. invertebrates, the most biodiverse group).

Species diversity Similarly, it may be possible to preserve a diversity of species with combinations ofgeneric and specific actions.

Specific assemblages or Many assemblages will be impossible to preserve in the long term, as the constituent‘communities’ of species species are likely to have differing responses to climate change. Indeed the

popular concept of communities may not be robust or useful in the context of futureclimate change.

Ecosystem types in their Different types of ecosystems are strongly determined by climatic factors; hence itcurrent locations will be impossible to preserve some if not most ecosystems in their current

locations, depending on the magnitude of future climate change.

Types of ecosystems somewhere It is likely that many broad ecosystem types (e.g. temperate grasslands) will continueto exist (albeit not with the same assemblages of species) at some location.However, broad-scale management is unlikely to be able to preserve specific eco-systems with climatic and other requirements that cease to coexist in future climates.

Functioning and resilient ecosystems There are likely to be demands for future emerging ecosystems to be ‘healthy’; thatis, their primary productivity, functioning and resilience should be similar to existingnatural analogous ecosystems. Where this does not occur naturally in the future itmay be hard to create via management, although the study of how ecosystemscan be constructed is an emerging area of research.

Ecosystem services There is likely to be substantial demand for the maintenance of those attributes ofbiodiversity that support human well-being, including life-supporting services andnon-essential values. Where these do not occur naturally in the future they may bevery hard to create with management.

The comments in the table represent one initial assessment and should be subjected to further scrutiny and valuejudgement. Contrasting views might suggest:

� saving species and rejecting communities is artificial—species are also human concepts, genetic variation is aboutthe only fundamental and persistent biodiversity entity and covers all scales; and

� ecosystem function is not a good objective on its own—it may be possible to have good ecosystem function with lowbiological diversity, and zero ‘Australian’ species.

If such a triage approach were to be adopted,what would be the criteria for prioritisation:prospects for long-term adaptation? response toconservation investment? value to society? Dowe have the information necessary to make thosedecisions?

Decisions about how best to invest in managingthe various aspects of both climate change andbiodiversity conservation should be made in arisk management framework, with an integratedrather than adversarial, approach to differentobjectives. Are our current institutions andpolicy programs up to the task?

Many, but not all, existing conservationprograms are based, implicitly or explicitly, on

the notion that what we want to conserve is‘exactly what we have now, where it is now’—a static view of biodiversity. While this notionis not consistent with a far more dynamicbiogeographic or evolutionary perspective(10,000s of years), it is broadly consistent withthe shorter-term anthropomorphic perspective(1–10s of years). Hence it has maintainedcurrency. However, future climate change is verylikely to result in a pattern of shifting mosaicsof ecosystems in more human time scales (10–100 years), thereby undermining this staticnotion of biodiversity. To be effective, futureconservation programs will have to acknowledgeand accommodate this dynamic view ofbiodiversity. Indeed, in our highly modified

6. Policy


landscapes, active management of these shiftingmosaics will become increasingly important ifwe are to achieve conservation objectives.

Delivering on the objectives: adaptationoptions

A broad range of activities are likely to berequired to effectively deliver on the objectiveof conserving biodiversity under climate change.

� Preserving currently intact natural habitat isand will continue to be the cheapest and mosteffective biodiversity conservation action.

� The dynamics of many Australianecosystems, both natural and agricultural,are dominated by inter-annual climatevariability, although the details of climatethresholds and sensitivities are not alwayscurrently well understood. Understandingand effectively managing for this variationis likely to be an essential part of on-goingbiodiversity conservation and thedevelopment of future adaptation programs.

� Many existing biodiversity conservationpractices will be highly beneficial formeeting biodiversity conservation objectivesunder climate change, although there will bea need to do more of them and possibly todo them slightly differently.

� However, current conservation practice mayneed to be supplemented with activitiesspecific to dealing with climate change inthe longer-term (e.g. translocation).

� To guide adaptation actions for futurebiodiversity, there is an urgent need formonitoring of impacts and adaptations,understanding of climate sensitivities andspecies interactions and further developmentof both biodiversity and greenhouse policy.

Current biodiversity conservation effort is abroad mixture of activities, ranging from thegeneric to species-specific and site-specificactivities. Some examples:

� measures aimed at increasing the robustnessof ecosystems and preserving a large numberof species: for example, reducing landclearing, limiting sediment and nutrient flowinto waterways, establishing a connected,

comprehensive system of reserved areas,maintaining bio-security, communityeducation;

� targeted measures in situ, aimed at specificspecies but with multiple other benefits: forinstance, pest control, fencing, revegetation,flora and fauna reserves; and

� intensive single species programs, such asex situ conservation, captive breeding andreintroductions, perhaps even reconstruction(Tasmanian Tiger DNA, for example) butnoting the high cost and significant ethical,technical and other issues associated withthese types of activities.

Many of these activities could be beneficial foraddressing the impacts of climate change onbiodiversity. However, the nature and scale ofthe threats posed by climate change impacts mayindicate that we need a different balance ofinvestment in the various types of conservationactivity. Under climate change, there are likelyto be more species that are critically endangered,which will create added demand for expensivespecies-specific conservation actions. However,a more efficient long-term approach may be tomake greater relative investment in genericconservation actions—aimed at increasingresilience and preventing as many species aspossible from becoming endangered. Rapidlystabilising, then reducing atmosphericgreenhouse gas concentrations would probablybe most effective in minimising the long-termimpact on biodiversity, but this faces political,economic and cultural barriers.

In many situations, substantial increases in theability of species or ecosystems to cope withclimate change may be achieved by reducingother threatening processes (for instance,stopping land clearing), especially if climatechange is at lower end of the range of scenarios.In other situations, some climate-change-specificactions may be required. Broadly, those actionsmight include:

� increasing connectedness in the landscape,

� larger networks of protected areas,

� species translocations.

6. Policy


Given the potential scale of climate change andthe likely shifts in species’ climatic envelopes,substantial increases in connectivity acrosslatitudes and altitudes may be required, throughlarge-scale corridors and patches of habitat.Some practitioners suggest it may be more usefulto think about porosity in the landscape, ratherthan corridors per se (which tend to constrainthinking to specific geometrical layouts). Theessential feature that is required is enoughconnected habitat for large-scale ecologicalprocesses to continue. This could be achievedthrough coordination of vegetation managementacross land tenures and uses (for example,reserves, off-reserve protected areas and on-farmrevegetation) and across scales (connectingremnants, at one scale, or biogeographic regions,at another scale, for example).

The need for increased connectedness in thelandscape is not new: it has been recognised fordecades, and climate change just adds to the need.Despite this existing need, progress, if any, has beenvery slow. The scale and potential threat of climatechange might be useful in communicating the needfor more immediate action. The potential riskthat increased connectedness will lead to greaterhabitat for and movement of pests, weeds andfire also needs to be managed.

Translocation and ex situ conservation of speciesis likely to be a very cost-ineffectiveconservation mechanism, although it may be alast resort for some species; for example,threatened vertebrates and tree species that takea long time to become reproductively mature. Apriority should always be to protect species insitu to try and avoid the need for more radicalaction. If translocations are necessary, they arelikely to raise considerable ethical issues relatingto impacts on species in the recipient areas andeffective expenditure of funds. The expense ofsuch programs is unlikely to have broad supportif in situ conservation of other species orecosystems is being neglected. A possibly lesscontroversial form of translocation might be theproactive planting of ‘future climaticallyappropriate’ species, as opposed to localprovenances, in revegetation programs.

Investments in adaptations to climate change arelikely to be most effective if they are made inthe context of broader natural resource

management programs that, by their nature, havea broad range of objectives. Increasingly,multiple benefits are being sought from anysingle proposed investment. Hence adaptationswith other benefits (such as water-table control,forestry), and existing actions that can bemodified to provide adaptation benefits, aremore likely to be viable. Similarly, a riskmanagement approach to greenhouse policy canonly be implemented if assessments ofbiodiversity impacts and actions are integratedwith assessments of adaptation and impact inagriculture and other sectors, and with mitigationassessments. Thus, biodiversity adaptationpolicy should be coordinated with existingbiodiversity and greenhouse policy, especiallyAustralia’s Forward Climate Change Strategy.

A number of major existing programs havebiodiversity conservation as one of theirobjectives (see Appendix 1). These programs arelikely to have considerable potential formonitoring climate change impacts anddelivering adaptation actions, if they are giveninformation about possible impacts and aboutidentifying opportunities for adaptations.Information about regions and taxa that mightbe particularly sensitive would be useful formany existing programs. Regional-scaleinformation could be readily used in the regionalnatural resource management (NRM) programs.

There is significant opportunity for collaborationbetween the research community and the groupsdelivering NRM programs, for both impactassessment and management of adaptationprograms (especially in an adaptive managementframework). Staff from Environment Australiaand the relevant state agencies might play acritical role in facilitating information flows andbrokering collaborations.

While there are very few documentedopportunities for direct actions to promoteadaptation to climate change in marineenvironments, it may be possible to decreaseexisting pressures on such ecosystems andincrease their resilience by addressing land–sealinkages.

Given the uncertainties about future climatechanges and about biological responses to thesechanges and the effectiveness of particular

6. Policy


adaptation actions, it will be essential foradaptation programs to be managed in an activeadaptive management framework. That is, wherethere is uncertainty, management actions shouldbe designed specifically so that robust informationabout the outcomes of alternative actions isacquired and used to design further actions.

Information needs—filling the gaps

Investment and advice must be based on goodscience and information, particularly in the lightof the significant uncertainties in future climatechange and biological responses. Informationcan come from collating existing science, newresearch and adaptive management. This sectionlists many aspects of the information needed fordeveloping policy responses to the impacts onbiodiversity of climate change. Addressing thesequestions could help set research agendas andfoster partnerships between the science andpolicy communities. It is recognised that thesepartnerships can and should be based on far morethan application for and provision of researchfunding. The BDAC workshop provided a goodstart for the building of some science–policyrelationships; priority should be given to creatingand maintaining enduring linkages.

Impacts

How is biodiversity actually responding toclimate change?

Modelling studies can provide very usefulinformation about the potential scope ofbiological responses to climate change.However, it is essential that we also measureactual biological responses to past, present andfuture climate change, to complement, guide andvalidate modelling studies. A coordinatednational monitoring program would greatlyimprove our knowledge of impacts as they occurand improve our capability to assess future risksand adaptations.

What is the scale of the expected changes inbiodiversity?

At the regional scale, are we expecting changesin vegetation structural type, or are we simplyexpecting some species turnover withinecosystems resulting in a slightly differentassemblage but similar ecosystem functioning?

The answer to this question has consequencesfor flow-on effects and for options foradaptation; for example incarbon, nutrient andwater budgets and other ecosystem services,conservation status, prognosis and managementof threatened or restricted species. This questioncould be addressed with simulation analysesbased on ecosystem-level and species-levelbiological processes, and by careful analyses ofpast climatic changes.

How fast will biodiversity respond?

For many of the causes and pathways of change,we know the probable direction of change but wedon’t know the sizes of the effects, interactions andfeedback. For example, studies that overlie species’bioclimatic envelopes with future climate surfaces(and possible soil, topography, protected areas, etc.)may give an indication of the direction and possibleextent of species migrations. Studies based onecosystems’ physiological responses to CO

2

concentrations provide complementaryinformation. However, these studies give littleinformation about the rates of migration andinteractions that may actually happen. Predictedrates of climate change imply that migration rateswill have to be many times faster than the ratesof past migrations and faster than many speciesare likely to be able to migrate. Indeed, there isevidence that some species are still currentlylagging thousands of years behind past climatechange. What further analyses can be done toprovide information about possible dynamics ofthe compositions of ecosystems and speciesmigrations? This question could be addressed withtargeted experimentation, simulation analysesbased on ecosystem-level and species-levelbiological processes, and by careful analyses ofpast climatic changes.

How important is climate as a direct driver ofchange?

In what regions and for what species andecosystems will direct effects (CO

2 concentration,

temperature, rainfall change) or indirect effects(species interactions such as competition,predation, habitat, hosts, pathogens) be moreimportant? In situations where indirect effectsdominate, can studies of bioclimatic envelopesprovide useful estimates of the scope of futurechanges?

6. Policy


Which regions, species and ecosystems aremost vulnerable to the impacts of climatechange?

Various arguments (which are not alwaysconsistent) have been used to suggest particularsensitivities to climate change. For example:

� mountains, because they frequently havespecies stratified on steep climatic gradientsand are restricted in area;

� regions with low relief, where a small changein climate corresponds to a large geographicshift of favoured climate conditions;

� migratory species because they need avariety of habitats (winter migrating,breeding); and

� species with limited dispersal ability and/orthat reproduce slowly.

Ecosystems that appear particularly vulnerable,on the basis of existing studies, include coralreefs, tropical rainforests and alpine areas.

Which species or ecosystems will beadvantaged by climate change?

Some species that are otherwise restricted maybe favoured by climate change, become invasiveand become a threat to the persistence of otherspecies.

Which processes and phases of species lifecycles are most likely to be affected by climatechange?

It is likely that different ecological processesdetermine the warmer (northern or lower)boundaries and cooler (southern or upper)boundaries of species or ecosystem distributions.Hence the different boundaries may move atdifferent rates. The difference between theserates effectively determines whether a specieswill become extinct, invasive, or something in-between. Generally, establishment and death arethe key phases of the life cycle that influencethe abundance and distribution of species. Bothphases are highly sensitive to extreme conditionsor events—which are anticipated to increase infrequency and/or intensity with climate change.This provides us with a clear target for analysisof likely impacts and adaptations.

How will the impact of climate change onbiodiversity affect economic and societalvalues?

From a human perspective, not all aspects ofbiodiversity may be of equal value. Some aspectsmay have high value in social or economic terms.If those aspects are lost, the impact may notcorrespond directly to any objective measure ofthe loss of biodiversity. It is important that weunderstand the impacts of these societal andeconomic values, as they are likely to be majordrivers of any future action, and they shouldhighlight priorities for action. Currently, thereis no process to identify these values and howthey might change.

Adaptations

How should existing biodiversity conservationactions be improved to account for climatechange?

Many existing conservation actions will need tobe maintained or enhanced to preservebiodiversity under climate change. However,there is frequently uncertainty about howconservation should be practised now, and thisuncertainty will be amplified by climate change.More information is needed to understand thelinks between preserving biodiversity and

(a) flow regimes in waterways,

(b) habitat quantity and quality,

(c) landscape connectivity,

(d) invasive species, and

(e) the dynamics of small and isolatedpopulations.

In addition, more needs to be understood aboutthe characteristics that determine the resilienceof ecosystems.

What scale of connectivity is likely to be mosteffective for maintaining ecological processesand preserving biodiversity under climatechange?

Corridors and ‘stepping stones’ are frequentlypromoted as essential elements in landscapereconstruction and management, but there

6. Policy


remains little information about theireffectiveness. More information is required todetermine how such elements should bedesigned to achieve biodiversity outcomeswithout simply increasing habitat and refugesfor feral species, wildfire and new invasivespecies.

What are the requirements of a landscape thatis resilient and robust under climate change?

How can resilience to the additional pressuresand threats from climate change and the capacityto adapt naturally to future climates be increasedin different environments? Indeed, are resilienceand adaptability consistent or potentiallycontradictory characteristics? There is asubstantial body of theory on resilient systemsand a growing array of applications. There maybe lessons from this work that can guide bothconservation policy and conservation practice.

Is it practical to control the spread of nativeor naturalised species that may becomeinvasive under climate change?

Species that can survive changes in climate perse, may be threatened by other species, bothnative and exotic, that are benefited by climatechange. When and how might these threats bemanaged? Current quarantine and screeningpractices and weed and feral animal controlprograms may need to be modified to includeclimate change elements.

In which regions, and for which species, mightreductions in non-climate-change pressures onbiodiversity be sufficient to increase resilienceto climate change?

For many species, ecosystems and environmentsthere may be few direct actions that will reducethe impacts of climate change, but reducing otherpressures may increase their resilience andadaptive capacity. For example, some land-basedmanagement actions may be useful for reducingthe overall pressure on aquatic and marineecosystems.

What characteristics define locations thatcould become climate change refuges?

Are refuges from past climatic changes likely tobe important for natural adaptation to future climate

change? Possibly not, given that past climatevariations tended to be towards coldertemperatures, whilst future ones are expected tobe in the opposite direction. What role will activemanagement play in determining and protectingfuture refuges? Can such places be identifiedand protected from present and short-termthreats?

How should a shifting mosaic of ecosystemsand species be managed?

Can we identify the direction (both geographicaland ecological) in which ecosystems and speciesmight be moving in response to climate change,and then facilitate (or slow) that movement? Arethere general principles for such management,or does it have to be done on a case-by-casebasis?

Can potential barriers to migration andadaptation be identified and mitigated?

Some barriers to migration may be obvious: forexample, mountains, deserts, oceans andagricultural and urban land. However, otherbarriers (such as genetic paucity and smallpopulations) may be less obvious except in afew well-studied species.

What are the risks associated withtranslocating species?

Captive breeding and species translocation areexpensive, difficult and hazardous activities. Isit possible to predict and manage the risks offailure to establish (due to competition orinsufficient habitat, for example) and the risksthat some species may create a conservationthreat via competition or predation?

Delivery

What information about impacts andadaptations will be useful to regional NRMplanners?

The nature and scope of the impacts of climatechange on biodiversity suggest the issue needsto be addressed and managed at a national level.However, and increasingly, conservationprograms are being delivered through regional-scale processes. Can national-scale informationbe made useful for regional-scale management?

6. Policy


And is regional-scale delivery sufficient toachieve national biodiversity and climate changeoutcomes? What parameters would be useful forinclusion in the National Monitoring andEvaluation framework (a key part of the evolvingregional NRM process)?

Can policy cope with the level of uncertaintyabout climate change impacts and themanagement options?

The large uncertainty about climate change, itsimpacts and possible responses to adaptationssuggests that we need an adaptive managementapproach to guide and revise any adaptationprogram. However, responses to climate changeand to adaptation management are likely to takeplace over decades or more. Can adaptiveprograms be developed that will be robust to thetime lags and non-linearities in these biologicalresponses?

How might trade-offs in adaptation orconservation investment be managed?

If a triage or other some other trade-off approachis used to prioritise investment, what criteriashould be included (such as level of climateimpact, effectiveness of adaptation action andvalue of the entity be conserved)? And howshould the different criteria be assessed andcombined?

Summary: Main policy recommendations

We have identified high priority actions forbuilding a solid program for conservingbiodiversity under the impacts of climate change.These can be separated into the five categoriesdescribed below.

1. Understand and manage for climatevariability

Understanding the current impact of climaticvariability on natural and agriculturalecosystems will greatly help us to detect andmanage future climate change. Actions mightinclude management of:

� bushfire risk in relation to climatevariability;

� grazing in relation to climate variability inthe rangelands;

� water diversion and environmental flows inrelation to climate variability;

� estuaries and coastal areas in relation toclimate vaiability.

2. Preserve biodiversity that is sensitive toclimate change

Preservation of existing biodiversity will alwaysbe the most effective and cheapest means ofconserving biodiversity in the future. Many currentconservation activities will also help conservebiodiversity under climate change, especially ifgiven some enhancement. Actions include:

� preventing further land-clearing and loss ofhabitat;

� preventing introductions of new invasivespecies;

� reducing grazing in sensitive alpine areas;

� restoring environmental flows and waterquality;

� minimising land salinity and in-streamsalinity;

� reducing nutrient and sediment flows torivers, wetlands, estuaries and oceans(especially important for the Great BarrierReef and other reef systems).

Some actions specific to climate change mayalso be required, for example:

� suppression of species likely to becomeinvasive under climate change;

� ex situ conservation of species that cannotsurvive future climates;

� including climate change threats as amandatory consideration when draftingrecovery plans for endangered species andintegrated NRM plans.

3. Facilitate long-term adaptation

Some actions may be required to assist oraugment the long-term and natural adaptationof biodiversity to climate change, for example:

� increasing the connectivity in the landscape,including corridors and ‘stepping stones’,

6. Policy


especially along temperature- and othergradients;

� rehabilitating degraded habitat, includingrevegetation of cleared land and restorationof streams, rivers and wetlands;

� preserving locations that may become keyhabitat under future climates (for example,mountains and reefs to the south of thosewith potentially threatened species);

� ensuring that refuges are established andprotected, and that species can migrate toand from them;

� translocating species (noting the risksinvolved).

4. Monitor, research and develop policy

For adaptation programs to be successful in thelong term, substantial information is needed toaddress gaps in understanding. Someinstitutional issues may need to be managed.Actions should include:

� developing criteria for monitoring climatechange impacts

� monitoring and assessing climate change andbiodiversity impacts;

� designing adaptation actions to maximiseinformation for future actions (that is, activeadaptive management);

� ensuring that climate change impacts areintegrated into all existing and futurebiodiversity conservation programs. Forexample, biodiversity conservation shouldbe consistent with a shifting mosaic conceptof ecosystems;

� developing new policy mechanisms to allowintegrated risk management analysis of allgreenhouse issues (mitigation andadaptation, costs and benefits), as opposedto a segmented, potentially adversarial,approach;

� raising the priority of climate change inresearch funding processes such as those ofthe Australian Research Council.

5. Mitigate climate change and reduce otherpressures on biodiversity

There is no doubt that the tasks of preservingbiodiversity under climate change andfacilitating adaptation will be more achievableand less costly if the magnitude of future climatechange is reduced. The greater the magnitudeof future climate change the larger the numberof species that will become threatened and thegreater the likely need for investment to replacelost ecosystem services. When assessing thecosts and benefits of mitigating climate changewe must include the costs and benefits ofadaptation associated with different emissionscenarios.

6. Policy


A wide range of existing policy programs havebiodiversity conservation as one of theirobjectives. While few, if any, currently includeclimate change impacts in their assessments ofbiodiversity conservation needs, many of themcould do so if the appropriate information wereavailable. Similarly, adaptation to climate changecould feasibly be added as a specific objective tosome existing programs.

In some cases information about climate changeor potential adaptation benefits may be sufficientto increase, or decrease, the priority of a proposedaction; for example, in situations whererevegetation and water-table pumping optionswere being compared primarily for salinitycontrol.

There are several advantages of working withinexisting programs:

� the programs exist; therefore

(a) a delivery mechanism is alreadyestablished, and

(b) the transaction cost of getting a newprogram up is avoided;

� any actions will be developed in an integratedcontext; mechanisms will exist for addressingtrade-offs between different values or sectors,rather than having them emerge later as barriersto a specific adaptation action;

� there is generally a desire and role for climatechange information in existing programs;

and disadvantages:

� in some cases agendas may already be set;

� there may be limited additional scope forwork that does not conform with existingwork plans/actions/targets;

� the existing approach might not work for thepreferred climate change adaptations, e.g.mismatch of spatial or temporal scales;

� climate change impacts and the effectivenessof adaptations may have more uncertaintyand a longer timeframe than competingobjectives. Hence, without strong strategicinput, the comparisons and trade-offs mayfavour better understood and shorter-termNRM issues.

Regional NRM policy framework

Commonwealth natural resources management(NRM) and environmental investments aremainly delivered via ‘integrated regional plans’.These are essentially plans developed byregional bodies with some guidance and strategicinput from state and Commonwealthgovernments. The plans must set targets for anumber of specified issues and set out the actionsthat will be undertaken to achieve those targets.Funding of a plan is contingent on accreditationby the state and Commonwealth. The targetsfocus on natural resource condition issues andinclude biodiversity. This regional approach isthe primary delivery mechanism for both theNational Action Plan for Salinity and WaterQuality (NAP) and the second phase of theNatural Heritage Trust (NHT II).

The plans are expected to include detailed trade-off analyses of different objectives, includingobjectives specified by the region as well as thoseproscribed by the state and Commonwealth. Assuch, different objectives (possibly includingadaptations) essentially compete for priority witha regional body primarily determining the outcome,but there is a strong emphasis on actions that havemultiple benefits. The programs also have astrong component of monitoring which is mainlyconcerned with evaluating progress towards

&��'� � ('��

��


targets, but will also include some generalmonitoring of resource condition. Theinvestments can be used for capacity building(data and expertise) and facilitating land andwater-use changes as well as for activities suchas revegetation and fencing.

The NHT guidelines specify 10 areas of activityfor investment; of these biodiversity and climatechange issues could be covered directly in sixareas and indirectly in two areas. The remainingtwo areas essentially cover productivity ofagriculture and forestry (which will definitelyhave climate change issues). Hence, while theNHT does not explicitly include climate changeadaptations, it does not exclude them.

NHT national investments

The NHT II also includes investments to addressnational and state priorities that generally involveactivities at national or regional scale. Research,analysis, training and reserve acquisition areamong activities covered by this stream.Measuring impacts and large-scale adaptationprograms could fit into this category in the future,although it is highly competitive.

NHT local investments

The Envirofund provides small grants of up to$30,000 over a single year to community groupsto address local biodiversity or resource useproblems. The fund includes Bushcare, Landcare,Rivercare and Coastcare. If suitable adaptationactions could be developed at this scale, thismight be an effective delivery mechanism.

Greenhouse and agriculture framework

The developing national framework forgreenhouse and agriculture deals largely withreducing net emissions of greenhouse gases, butit also includes some opportunities forbiodiversity adaptations. There is an emphasison multiple benefits, such as planting for carbonsequestration and biodiversity benefit, althoughsequestration and economic uses are likely tohave priority.

The framework includes identifying regions thatare most likely to be affected and the monitoringof impacts. There may be opportunity toincorporate biodiversity into these activities.

Abatement programs

There are a number of abatement programs thatare investing in planting of woody vegetationfor carbon sequestration. Multiple benefitsinclude forestry and forest products, but plantingfor conservation, especially with regard toclimate change, could be possible. Theseprograms include:

� Greenhouse Gas Abatement Program(GGAP),

� Bush for Greenhouse—a program supplyinginformation about seeking maximum benefitfrom planting for carbon sequestration,

� National Carbon Accounting Scheme(NCAS):

(a) ‘developing tools to identify the carbonbenefits’ of revegetation for many purposesincluding biodiversity,

(b) seeking synergies between NRM andgreenhouse outcomes, such as via forestry,revegetation, vegetation management …greenhouse sinks … biodiversity,

� seeking to link greenhouse into regionalNRM delivery:

(a) NCAS, Bush for Greenhouse, adaptationsstrategies relevant to NRM.

Biodiversity toolbox

Biodiversity toolbox is an information kitdeveloped by the Commonwealth Governmentprimarily to help local governments withmanaging biodiversity. It is likely to be revisedfor the regional NRM framework, and there maythen be an opportunity for including informationon adaptation options.

National Reserve System

The National Reserve Scheme (NRS) aims toestablish a comprehensive, adequate andrepresentative (CAR) system of protected areas.Currently this includes minimal redundancy; thatis, criteria are achieved as long as an ecosystemis represented. Climate change may indicate aneed to reconsider the definition of adequate inthe NRS, because diversity, redundancy, andgeographic distribution will become increasingly

Appendix 1


important under climate change. Land acquisitionsare covered by the NHT national investments.

State of the Environment

State of the Environment reporting is not a directadaptation action. However, including specific

impact and adaptation information in futurereports may increase awareness of the issue.Those developing State of the Environmentreports have indicated they are keen for usableindicators of climate impacts and adaptations.

Appendix 1


&��'�� )��

Dana Bergstrom Australian Antarctic Division, Channel Highway, Kingston, TAS 7150(absent on the day)

Sandy Berry Greenhouse CRC, Research School of Biological Sciences,The Australian National University, Canberra, ACT 0200

Kerry Bridle School of Geography & Environmental Studies, University of Tasmania,GPO Box 252-78, Hobart, TAS 7001

Eddy Campbell Environmetrics, CSIRO Mathematical & Information Sciences,Private Bag 5, Wembley, WA 6913

Gerry Cassis Australian Museum

David Cheal Arthur Rylah Institute, Dept of Natural Resources & Environment,PO Box 137, Heidelberg, VIC 3084

Chris Chilcott QCCA, Climate Impacts & Natural Resources, Dept of Natural Resources& Mines, Gate 4, 80 Meiers Road, Indooroopilly, QLD 4068

Gerard Crutch Environment Australia

Terry J. Done Australian Institute of Marine Science, PMB No. 3, Townsville,QLD 4810

Ashley Fuller Australian Greenhouse Office

Michael Dunlop Biodiversity Policy Section, Environment Australia, GPO Box 787,Canberra, ACT 2601 (present address: CSIRO Sustainable Ecosystems,Canberra)

Heather Gibbs PhD Student, Deakin University

Alistair Graham Tasmanian Conservation Trust, 102 Bathurst Street, Hobart, TAS 7000

Ken Green National Parks and Wildlife Service NSW, Queanbeyan

Simon Haberle Research School of Pacific and Asian Studies, The Australian NationalUniversity, Canberra, ACT 0200

David Hilbert CSIRO Sustainable Ecosystems, Tropical Forest Research Centre,PO Box 780, Atherton, QLD 4883

Ove Hoegh-Guldberg Centre for Marine Studies, University of Queensland, St Lucia, QLD 4072

Mark Howden CSIRO Sustainable Ecosystems, GPO Box 284, Canberra, ACT 2601

Lesley Hughes Dept of Biological Sciences, Macquarie University, Sydney, NSW 2109


Roger Jones Climate Impact Group, CSIRO Atmospheric Research, Private Bag No.1,Aspendale, VIC 3195

Max Kitchell Environment Australia

Paul Marsh Environment Australia

Jo Mummery Australian Greenhouse Office

Jason Middleton School of Mathematics, The University of New South Wales, Sydney,NSW 2052

Jeanette Muirhead BDAC Secretariat

Julian Reid CSIRO Sustainable Ecosystems, GPO Box 284 ,Canberra, ACT 2601

Christine Schweizer Environment Australia

Joshua Smith Environment Australia

Anita Smyth CSIRO Centre for Arid Zone Research,PO Box 2111, Alice Springs, NT 0871

Ian W. Watson Centre for Management of Arid Environments,Dept of Agriculture WA, PO Box 483, Northam, WA 6401

Mike Weston Research & Conservation Dept, Birds Australia, 415 Riversdale Road,Hawthorn East, VIC 3123

Stephen E. Williams School of Tropical Biology, James Cook University,Townsville, QLD 4811

Imogen Zethoven World Wide Fund for Nature, PO Box 710, Spring Hill, Brisbane,QLD 4004


&��'��*��

adaptations responses that decrease the negative effects and capitalise on positiveopportunities associated with impacts

benthic associated with the bottom of a waterbody or the sea

bioclimatic range areas of land that are likely to provide a climate appropriate for a species

biodiversity the variety of life forms: the different plants, animals and micro-organisms,the genes they contain, and the ecosystems they form. It is usually consideredat three levels: genetic diversity, species diversity and ecosystem diversity.

biomes climatically controlled ecosystems with characteristic vegetation, occupying largeareas, typically on more than one continent, e.g. coral reefs, grasslands

biotic associated with living organisms

BP before present

breakaways land features in lateritic landscape, usually consisting of small escarpments

climate envelope the range of climate conditions within which a species or ecologicalcommunity can be found

decapod ten-footed creature (usually crustacean)

detectors species occurring naturally in an area of interest that may show measurableresponses to environmental change, such as changes in distribution or behaviour

detritivores consumers of detritus or dead biological materials

dinoflagellate algae mobile planktonic algae with more than one flagellum (for propulsion), veryimportant in marine and freshwater ecosystems

ecosystem service benefit provided by a particular ecosystem

ecotone transition zone between two or more vegetation communities, such as betweengrassland and forest

El Niño a combined ocean and atmosphere pattern which results in an increase insea-surface temperatures in the eastern Pacific off the east coast of NorthAmerica but reduced sea-surface temperatures around Australia, resulting inlower than average rainfall and a smaller likelihood of cyclones in Australia

endemic species species restricted to a particular geographic region or locality

endemism occurrence of endemic species

endothermic capable physiologically of maintaining a contant body temperature, as humansdo; ectothermic species have body temperatures that change with thetemperature of their environment

ENSO El Niño Southern Oscillation

exclosure a fence around a patch of ground, meant to exclude animals

genetic determined by or relating to genes

genetic paucity very little genetic variation

global warming anticipated increase in global temperatures resulting from human-inducedemissions of greenhouse gases


greenhouse climate changes in global and regional temperatures, rainfall and other climate-relatedchange factors as a result of human-induced net emissions of greenhouse gases

greenhouse gases gases such as carbon dioxide, methane and nitrous oxide, which act as a ‘blanket’in the atmosphere, keeping in more outgoing radiation than there wouldotherwise have been, maintaining higher temperatures in the lower atmosphereand at the surface of the land and oceans

habitat generalists species not confined to their range by a specific habitat

habitat specialists species specialised to suit a particular habitat or niche

hypo-osmotic a situation in which sea-water is less saline than usual, so that relatively freshwater diffuses into the cells of sea-living organisms by osmosis, damaging them

impact an effect of increased greenhouse gas emissions and atmospheric CO2

concentration (and other global changes) on physical and biological processesin both natural and managed systems

indicator an environmental indicator is a species or group of species that respondspredictably and sensitively, in ways that are readily observed and quantified, to anenvironmental disturbance.

IPCC Intergovernmental Panel on Climate Change

La Niña a combined ocean and atmosphere pattern which leads to an increase in sea-surface temperatures in the western Pacific and around Australia, resulting inhigher than average rainfall and a greater likelihood of cyclones in Australia

mining of data searching existing databases for patterns and relationships not part of theoriginal data-collection scheme

mitigation short-term relief; see abatement

monotonically steadily, in one direction

mutualistic symbiosis members of two different species living together, so that each benefits fromthe presence and metabolic activities of the other

net radiative forcing increase in energy exchange at the Earth’s surface caused by an imbalancebetween incoming and outgoing radiation

palaeoecology reconstruction of the relationships between past organisms and theirenvironments, usually studied via fossils, pollen and charcoal in layers of soil

parapatric having geographic or spatial distributions that abut but do not overlap

phenology life-cycles

physiognomy appearance

propagule seed, spore, tuber, egg—something from which new life germinates or hatches

riparian zone riverside zone of varying width, depending on context

sentinels sensitive organisms introduced into the environment as an early-warningdevice

signal-to-noise ratio ratio between the value that is being measured and interference from backgroundoccurrences

symbionts species that live together closely

symbiosis the situation of species living together closely

threatening processes situations, usually caused by human activities, that reduce the quality of ahabitat for a species, e.g. land-clearing, presence of feral animals

translocation moving a species into a new area of suitable habitat which the species couldnot have reached without human assistance


Abele, L. and Patton, W.K. (1976) The size of coralheads and the community biology of associateddecapod crustaceans. Journal of Biogeography3: 35–47.

Allan, R.J. and D’Arrigo, R.D. (1998) PersistentENSO sequences: how unusual was the 1990–1995 El Niño? Holocene 9: 101–118.

Barnola, M. (1987) Vostok ice core provides 160,000-year record of atmospheric CO

2. Nature 329:

408–414.

Barrows, T.T., Juggins, S., De Deckker, P., Thiede, J.and Martinez, J.I. (2000) Sea-surface tempera-tures of the southwest Pacific Ocean during theLast Glacial Maximum. Paleoceanography 15:95–109.

Basher, R.E., Zheng, X. and Nichol, S. (1994) Ozone-related trends in solar UV-B series. GeophysicalResearch Letters 21: 2713–2716.

Bayliss, B., Brennan, K., Eliot, I., Finlayson, M., Hall,R., House, T., Pidgeon, B., Walden, D. andWaterman, P. (1997) Vulnerability Assessment ofPredicted Climate Change and Sea Level Rise inthe Alligator Rivers Region, Northern TerritoryAustralia. Supervising Scientist Report 123, Su-pervising Scientist, Canberra.

Beaumont, L. and Hughes, L. (2002) Potential changesin the distributions of latitudinally restrictedAustralian butterflies in response to climatechange. Global Change Biology 8: 954–971.

Beerling, D.J., Huntley, B. and Bailey, J.P. (1995)Climate and the distribution of Fallopia japonica:use of an introduced species to test the predictivecapacity of response surfaces. Journal ofVegetation Science 6: 269–282.

Berkelmans, R. and Oliver, J.K. (1999) Large scalebleaching of corals on the Great Barrier Reef.Coral Reefs 18: 55–60.

Berry, S.L. and Roderick, M.L. (2002a) Estimatingmixtures of leaf functional types using continen-tal-scale satellite and climatic data. Global Ecol-ogy and Biogeography 11: 23–40.

Berry, S.L. and Roderick, M.L. (2002b) CO2 and land

use effects on Australian vegetation over the lasttwo centuries. Australian Journal of Botany 50:511–531.

Black, R. and Prince, J. (1983) Fauna associated withthe coral Pocillopora damicornis at the southernlimit of its distribution in Western Australia.Journal of Biogeography 10: 135–152.

Bindschadler, R. and Vornberger, P. (1998) Changesin the west Antarctic ice sheet since 1963 fromdeclassified satellite photography. Science 279:689–692.

Blakers, M., Davies, S.J.J.F. and Reilly, P.N. (1984)The Atlas of Australian Birds. Melbourne Uni-versity Press, Melbourne.

Blumthaler, M. and Ambach, W. (1990) Indication ofincreasing solar ultraviolet-B radiation flux in al-pine regions. Science 248: 206–208.

Bowman, D.M.J.S., Walsh, A. and Milne, D.J. (2001)Forest expansion and grassland contraction withina Eucalytpus savanna matrix between 1941 and1994 at Litchfield National Park in the Australianmonsoon tropics. Global Ecology andBiogeography 10: 535–548.

Bowman, D.M.J.S (1998) Tansley Review No. 101:The impact of Aboriginal landscape burning onthe Australian biota. New Phytologist 140:385–410.

Brereton, R., Bennett, S. and Mansergh, I. (1995)Enhanced greenhouse climate change and its po-tential effect on selected fauna of south-easternAustralia: a trend analysis. Biological Conservation72: 339–354.

Brown, J.H. (1984) On the relationship between abun-dance and distribution of species. AmericanNaturalist 124: 225–279.

Bruhl, C. and Crutzen, P.J. (1989) On the dispropor-tionate role of tropospheric ozone as a filteragainst solar UV-B radiation. Geophysical Re-search Letters 16: 703–706.

Buddemeier, R.W. and Fautin, D.G. (1993) Coralbleaching as an adaptive mechanism. BioScience43: 320–326.

Bull, C.M. and Burzacott, D. (2001) Temporal andspatial dynamics of a parapatric boundary be-tween two Australian reptile ticks. MolecularEcology 10: 639–648.

+��


Bull, C.M. and Burzacott, D. (2002) Changes in cli-mate and in the timing of pairing of the Austral-ian lizard, Tiliqua rugosa: a 15 year study. Jour-nal of Zoology 256: 383–387.

Busby, J.R. (1988) A biogeoclimatic analysis ofNothofagus cunninghamii (Hook.) Oerst. insoutheastern Australia. Australian Journal ofEcology 11: 1–7.

Cannell, M.G.R., Palutikoff, J.P. and Sparks, T.H.(1999) Indicators of Climate Change in the UK.Centre for Ecology and Hydrology, Departmentof Environment, Transport & Regions (DETR)Available at http://www.nbu.ac.uk/iccuk/

Chapman, A.D. and Milne, D.J. (1998) The Impactof Global Warming on the Distribution of SelectedAustralian Plant and Animal Species in Relationto Soils and Vegetation. ERIN Unit, EnvironmentAustralia, Canberra.

Church, J.A., Godfrey, J.S., Jackett, D.R. andMcDougall, T.J. (1991) A model of sea level risecaused by ocean thermal expansion. Journal ofClimate 4: 438–456.

Coles, S.L. and Jokiel, P.L. (1992) Effects of salinityon coral reefs. In: Pollution in Tropical AquaticSystems. Connell, D.W. and Hawker, D.W. (eds.),CRC Press, Florida, pp. 147–166.

Cook, E., Bird, T., Peterson, M., Barbetti, M., Buckley,B., D’Arrigo, R., Francey, R. and Tans, P. (1991)Climatic change in Tasmania inferred from a1089-year tree-ring chronology of Huon Pine.Science 253: 1266–1268.

CSIRO (2001) Climate change projections forAustralia. CSIRO Climate Impact Group,Aspendale, Vic, 8 p., http://www.dar.csiro.au/publications/projections2001.pdf

de Freitas, C.R. (1988) Perspectives on the impact ofshort-term climatic change in New Zealand. NewZealand Geographer 43: 169–198.

de Groot, R.S., Ketner, P. and Ovaa, A.H. (1995)Selection and use of bio-indicators to assess thepossible effects of climate change in Europe.Journal of Biogeography 22: 935–943.

de la Mare, W.K. (1997) Abrupt mid-twentieth-cen-tury decline in Antarctic sea-ice extent fromwhaling records. Nature 389: 57–60.

Deng, X. and Woodward, F.I. (1998) The growth andyield responses of Fragaria ananassa to elevatedCO

2 and N supply. Annals of Botany 81: 67–71.

Dexter, E.M., Chapman, A.D. and Busby, J.R. (1995)The Impact of Global Warming on the Distributionof Threatened Vertebrates (ANZECC 1991).ERIN, ANCA, Canberra.

Doake, C.S.M. and Vaughan, D.G. (1991) Rapid dis-integration of the Wordie Ice Shelf in response toatmospheric warming. Nature 350: 328–330.

Dodson, J.R., and Mooney, S.D. (2002) An assess-ment of historic human impact on south-easternAustralian environmental systems, using lateHolocene rates of environmental change. Austral-ian Journal of Botany 50: 455–464.

Done, T.J. (1999) Coral community adaptability toenvironmental changes at scales of regions, reefsand reef zones. American Zoologist 39: 66–79.

Done, T.J., Turak, E., Wakeford, M., De’ath, G.,Kininmonth, S., Wooldridge, S., Berkelmans, R.,Van Oppen, M. and Mahoney, M. (2003a) Test-ing Bleaching Resistance Hypotheses for the2002 Great Barrier Reef Bleaching Event. Aus-tralian Institute of Marine Science. Final Reportto The Nature Conservancy.

Done, T.J., Whetton, P., Jones, R., Berkelmans, R.,Lough, J., Skirving, W. and Wooldridge, S.(2003b) Global Climate Change and CoralBleaching on the Great Barrier Reef. AustralianInstitute of Marine Science. Final Report to theState of Queensland Greenhouse Taskforcethrough the Department of Natural Resources andMines.

Doran, P.T., Priscu, J.C., Lyons, W.B., Walsh, J.E.,Fountain, A.G., McKnight, D.M., Moorhead,D.L., Virginia, R.A., Wall, D.H., Clow, G.D.,Fritsen, C.H., McKay, C.P. and Parsons, A.N.(2002) Antarctic climate cooling and terrestrialecosystem response. Nature 415: 517–520.

Douglas, B.C. (1991) Global sea level rise. Journalof Geophysical Research 96 C4: 6981–6992.

Dowdeswell, J.A., Hagen, J.O., Bjornsson, H.,Glazovsky, A.F., Harrison, W.D., Holmlund, P.,Jania, J., Koerner, R.M., Lefauconnier, B.,Ommar, E. and Thomas, R.H. (1997) The massbalance of circum-arctic glaciers and recent cli-mate change. Quaternary Research 48: 1–14.

Drake, B.G. and Gonzalez-Meler, M.A. (1997) Moreefficient plants: a consequence of rising atmos-pheric CO

2. Annual Review of Plant Physiology

and Plant Molecular Biology 48: 609–639.

Dunlop, N. (2001) Sea-change and fisheries: a bird’s-eye view. Western Fisheries Magazine, Spring,11–4.

Dyurgerov, M.B. and Meier, M.F. (1997a) Mass bal-ance of mountain and subpolar glaciers: a newglobal assessment for 1961–1990. Arctic andAlpine Research 294: 379–391.


Dyurgerov, M.B. and Meier, M.F. (1997b) Year-to-year fluctuations of global mass balance of smallglaciers and their contribution to sea-level changes.Arctic and Alpine Research 29: 392–402.

Eliot, I., Waterman, P. and Finlayson, C.M. (1999)Predicted climate change, sea level rise andwetland management in the Australian wet–drytropics. Wetlands Ecology and Management 7:63–81.

Environment Australia (2001) National Objectivesand Targets for Biodiversity Conservation 2001–2005. http://ea.gov.au/biodiversity/publications/objectives/

Etheridge, D.M. and Wookey, C.W. (1989) Ice coredrilling at a high accumulation area of Law Dome,Antarctica. 1987. In: Ice Core Drilling. Rado, C.and Beaudoing, D. (eds), Proceedings of the ThirdInternational Workshop on Ice Core DrillingTechnology, 10–14 October, 1988, Grenoble,France, pp. 86–96.

Etheridge, D.M., Steele, L.P., Langenfelds, R.L.,Francey, R.J., Barnola, J.M and Morgan, V.I.(1996) Natural and anthropogenic changes inatmospheric CO

2 over the last 1000 years from

air in Antarctic ice and firn. Journal of Geophysi-cal Research 101: 4115–4128.

Etheridge, D.M., Steele, L.P., Langenfelds, R.L.,Francey, R.J., Barnola, J.M. and Morgan, V.I.(1998) Historical CO

2 records from the Law

Dome DE08, DE08-2, and DSS ice cores In:Trends: A Compendium of Data on GlobalChange. Carbon Dioxide Information AnalysisCenter, Oak Ridge National Laboratory, OakRidge, Tenn., USA.

Farman, J.C., Gardiner, B.G. and Shanklin, J.D.(1985) Large losses of total ozone in Antarcticareveal seasonal ClO

x/NO

x interaction. Nature 315:

207–210.

Farquhar, G.D. (1997) Carbon dioxide and vegeta-tion. Science 278: 1411.

Fensham, R.J. and Fairfax, R.J. (1996) The disappear-ing grassy balds of the Bunya Mountains, South-eastern Queensland. Australian Journal of Botany44: 543–558.

Fensham, R.J. and Fairfax, R.J. (2002) Detectinglandscape change in Australian savanna usingaerial photography. II. The influence of rainfalland management history on vegetation cover.Australian Journal of Botany 50: 415–429.

Fensham, R., Minchin, P.R., Fairfax, R.J., Kemp, J.E.,Purdie, R.W., McDonald, W.J.F., and Nelder, V.J.(2000) Broad-scale environmental relations offloristic gradients in the Mitchell grasslands ofQueensland. Australian Journal of Botany 48:27–38.

Frakes, L.A., McGowan, B. and Bowler, J.M. (1987)Evolution of Australian environments. In: Faunaof Australia. Dyne, G.R. and Walton, D.W. (eds.)Australian Government Publishing Service, Can-berra, Vol. 1A, pp. 1–16.

Friedel, M. (1990) Some key concepts for measuringAustralia’s arid and semi-arid rangelands.Australian Rangeland Journal 12: 21–24.

Frith, C.B. and Frith, D.W. (1985) Seasonality ofinsect abundance in an Australian upland tropi-cal rainforest. Australian Journal of Ecology 10:237–248.

GCRMN (2000) Status of Coral Reefs of the World:2000. Wilkerson, R. (ed.) Australian Institute ofMarine Science, Queensland.

Graham, N.E. (1994) Decadal-scale climate variabil-ity in the tropical and North Pacific during the1970s and the 1980s: observations and modelresults. Climate Dynamics 10: 135–162.

Green, K. (ed.) (1998) Snow: A Natural History; AnUncertain Future. Australian Alps LiaisonCommittee, Canberra.

Green, K. and Osborne, W.S. (1994) Wildlife of theAustralian Snow-Country. Reed, Sydney.

Green, K. and Osborne, W.S. (1998) Snow as aselecting force on the alpine fauna. In: Snow: ANatural History; an Uncertain Future. Green, K.(ed.) Australian Alps Liaison Committee,Canberra, pp. 141–164.

Green, K. and Pickering, C.M. (2002) A potentialscenario for mammal and bird diversity in theSnowy Mountains of Australia in relation to cli-mate change. In: Mountain Biodiversity: a Glo-bal Assessment. Korner, C.H. and Spehn, E.M.(eds.) Parthenon Publishing, London, pp. 241–249.

Green, D., Singh, G., Polach, H., Moss, D., Banks,D. and Geissler, E.A. (1988) A fine-resolutionpalaeoecology and palaeoclimatology from south-eastern Australia, Journal of Ecology 76: 790–806.

Grevemeyer, I., Herber, R. and Essen, H.H. (2000)Microseismological evidence for a changing waveclimate in the northeast Atlantic Ocean. Nature408: 349–352.

Griffin, G.F. and Friedel, M.H. (1984) Effects of fireon central Australian rangelands. II Changes intree and shrub populations. Australian Journalof Ecology 9: 395–403.

Griffin, G.F. and Friedel, M.H. (1985) Discontinuouschange in central Australia: some implications ofmajor ecological events for land management.Journal of Arid Environments 9: 63–80.


Guisan, A. and Zimmermann, N.E. (2000) Predictivehabitat models in ecology. Ecological modelling,135: 147-186.

Haeberli, W. and Beniston, M. (1998) Climate changeand its impacts on glaciers and permafrost in theAlps. Ambio 27: 258–265.

Harrington, G.N. and Sanderson, K.D. (1994) Recentcontraction of wet sclerophyll forest in the wettropics of Queensland due to invasion byrainforest. Pacific Conservation Biology 1: 319–327.

Hassall and Associates (1998) Climate Change andManaging the Scarce Water Resources of theMacquarie River. Australian Greenhouse Office,Canberra.

Heatwole, H. (1987) Major components and distri-butions of the terrestrial fauna. In: Fauna ofAustralia. Dyne, G.R. and Walton, D.W. (eds.),Australian Government Publishing Service, Can-berra, Vol. 1A, pp. 101–135.

Hennessy, K.J., Suppiah, R. and Page C.M. (1999)Australian rainfall changes, 1910–1995. Austral-ian Meteorological Magazine 48: 1–13.

Herman, J.R., Bhartia, P.K., Ziemke, J., Ahmad, Z.and Larko, D. (1996) UV-B increases (1979–1992) from decreases in total ozone. Geophysi-cal Research Letters 23: 2117–2120.

Herron, N., Davis, R. and Jones, R. (2002) The ef-fects of large-scale afforestation and climatechange on water allocation in the Macquarie Rivercatchment, NSW, Australia. Journal ofEnvironmental Management 65: 369–381.

Heyward, A., Smith, L.D., Halford, A.R., Rees, M.and Meekan, M.G. (1999) Natural Variability atScott Reef: Short Term Response of Coral andFish Assemblages to a Severe Coral BleachingEvent. Australian Institute of Marine Science,Queensland. 35 p.

Higgins, S.I., Bond, W.J., Winton, S. and Trollope,W. (2000) Fire, resprouting and variability: arecipe for grass-tree coexistence in savanna. Jour-nal of Ecology 88: 213–229.

Hilbert, D.W. and Ostendorf, B. (2001) The utility ofartificial neural network approaches for model-ling the distribution of vegetation in past, presentand future climates. Ecological Modelling, 146:311–327.

Hilbert, D.W., Graham, A. and Parker, T. (2001a) Tallopen forest and woodland habitats in the wettropics: responses to climate and implications forthe northern bettong (Bettongia tropica). TropicalForest Research Series No. 1. Available on theweb via http://www.TFRSeries.jcu.edu.au

Hilbert, D.W., Ostendorf, B. and Hopkins, M. (2001b)Sensitivity of tropical forests to climate changein the humid tropics of North Queensland. Aus-tral Ecology, 26: 590–603.

Hilbert, D.W., Bradford, M., Parker, T. and Westcott,D.A. (in press) Golden bowerbird (Prionoduranewtonia) habitat in past, present and futureclimates: predicted extinction of a vertebrate intropical highlands due to global warming.Biological Conservation.

Hoegh-Guldberg, O. (1999) Climate change, coralbleaching and the future of the world’s coral reefs.Marine and Freshwater Research 50: 839–866.

Hoegh-Guldberg, O. (2001) The future of coral reefs:integrating climate model projections and therecent behaviour of corals and theirdinoflagellates. Proceeding of the Ninth Interna-tional Coral Reef Symposium. October 23–27,2000. Bali, Indonesia.

Hopkins, M.S., Ash, J., Graham, A.W., Head, J. andHewett, R.K. (1993) Charcoal evidence of thespatial extent of the Eucalyptus woodland expan-sion and rain-forest contractions in NorthQueensland during the late Pleistocene. Journalof Biogeography 20: 357–372.

Hopkins, M.S., Head, J., Ash, J., Hewett, R.K. andGraham, A.W. (1996) Evidence of Holocene andcontinuing recent expansion of lowland rain for-est in humid, tropical North Queensland. Jour-nal of Biogeography 23: 737–745.

Howden, S.M., Walker, L., McKeon, G.M., Hall,W.B., Ghannoum, O., Conroy, J.P., Carter, J.O.and A.J. Ash. (1998) Simulation of Changes inCO

2 and Climate on Native Pasture Growth.

Evaluation of the Impact of Climate Change onNorthern Australia Grazing Industries. FinalReport to Rural Industries Research and Devel-opment Corporation.

Howden, S.M., McKeon, G.M., Carter, J.O. andBeswick, A. (1999) Potential global change im-pacts on C

3–C

4 grass distribution in eastern Aus-

tralian rangelands. Proceedings of the VI Inter-national Rangeland Congress, Townsville, Aus-tralia, pp. 41–43.

Howden, S.M. and Turnpenny, J. (1997) Modellingheat stress and water loss of beef cattle in sub-tropical Queensland under current climates andclimate change. In: International Congress onModelling and Simulation. McDonald, A.D. andMcAleer, M. (eds.) Modelling and SimulationSociety of Australia, Hobart, Tasmania. pp. 1103–1108.

Hughes, L. (2000) Biological consequences of glo-bal warming: is the signal already apparent?Trends in Ecology and Evolution 15: 56–61.


Hughes, L. (in press) Climate change and Australia:trends, projections and impacts. Austral Ecology

Hughes, L., Cawsey, E.M. and Westoby, M. (1996)Climatic range sizes of Eucalyptus species inrelation to future climate change. Global Ecol-ogy and Biogeography Letters 5: 23–29.

IPCC (2000) Special Report on Emissions Scenarios.Nakicenovic, N. and Swart, R. (eds.). Intergov-ernmental Panel on Climate Change, CambridgeUniversity Press, Cambridge, p. 570

IPCC (2001a) Climate Change 2001: The ScientificBasis. The IPCC Third Assessment Report, Work-ing Group I report. Albritton, D.L. and MeiraFilho, L.G. (eds.) Cambridge University Press,Cambridge.

IPCC (2001b) Climate Change 2001: Impacts,Adaptation, and Vulnerability. The IPCC ThirdAssessment Report, Working Group II report.McCarthy, J.J., Canziani, O.F., Leary, N.A.,Dokken, D.J. and White, K.S. (eds.) CambridgeUniversity Press, Cambridge.

Isern, A.R., McKenzie, J.A. and Feary, D.A. (1996)The role of sea surface temperature as a controlon carbonate platform development in the westernCoral Sea. Palaeogeography, Palaeoclimatology,Palaeoecology 124: 247–272.

Jarman, S.J., Kantvilas, G. and Brown, M.J. (1988)Buttongrass Moorlands in Tasmania. TasmanianForestry Research Council Research Report No2, Hobart, Tasmania.

Jenkins, D.W. (1971) Global biological monitoring.In: Man’s Impact on Terrestrial and OceanicEcosystems. Matthews, W.H., Smith, F.E. andGoldberg, E.D. (eds.). The Colonial Press, USA,pp. 351–370.

Johns, C.V. and Hughes, L. (2002) Interactive effectsof elevated CO

2 and temperature on the leaf-miner

Dialectica scalariella Zeller (Lepidoptera:Gracillariidae) in Paterson’s Curse, Echiumplantagineum (Boraginaceae). Global ChangeBiology 8: 142–152.

Johnson, B. (1998) Consequences for the MacquarieMarshes. In: Climate Change Scenarios andManaging the Scarce Water Resources of theMacquarie River. Hassall & Associates (eds.)Australian Greenhouse Office, Canberra.

Johnson, C.R., Dunstan, P.K., Hoegh-Guldberg, O.(2002) Predicting the Long Term Effects of CoralBleaching and Climate Change on the Structureof Coral Communities. World Bank-UNESCOTargeted Working Group on modeling climatechange, Miami Florida, USA.

Jones, P.A. (1991) Historical records of cloud coverand climate for Australia. Australian Meteoro-logical Magazine 39: 181–189.

Jones, P.D. (1995) Recent variations in mean tem-perature and the diurnal temperature range in theAntarctic. Geophysical Research Letters 22:1345–1348.

Jones, R.J., Berkelmans, R. and Oliver, J.K. (1997)Recurrent bleaching of corals at Magnetic Island(Australia) relative to air and seawater tempera-ture. Marine Ecological Progress Series 158:289–292.

Jones, R.N. (2000) Managing uncertainty in climatechange projections: issues for impact analysis—an editorial comment. Climatic Change 45: 403–419.

Kanowski, J. (2001) Effects of elevated CO2 on the

foliar chemistry of seedlings of two rainforesttrees from north-east Australia: implications forfolivorous marsupials. Australian Ecology 26:165–172.

Keeling, C.D. and Whorf, T.P. (2002) AtmosphericCO

2 records from sites in the SIO air sampling

network. In Trends: A Compendium of Data onGlobal Change. Carbon Dioxide InformationAnalysis Center, Oak Ridge National Laboratory,US Dept of Energy, Oak Ridge, Tennessee, USA.

Kershaw, A.P., Van der Kaars, S., Moss, P. and Wong,S. (2002) Quaternary records of vegetation,biomass burning, climate, and possible humanimpact in the Indonesian-northern Australianregion. In: Bridging Wallace’s Line: The Envi-ronmental and Cultural History and Dynamicsof the SE-Asian-Australian Region. Kershaw,A.P., David, B., Tapper, N., Penny, D. and Brown,J. (eds.), Catena Verlag, Reiskirchen, pp. 97–118.

King, B., McAllister, F. and Done, T. (2002)Modelling the Impact of the Burdekin, Herbert,Tully and Johnstone River Plumes on the CentralGreat Barrier Reef. CRC Reef Research CentreTechnical Report 44. Applied Science Associatesand the Australian Institute of Marine Sciences.Available at http://www.reef.crc.org.au/publications/techreport/pdf/techrep_44_king.pdf.

Kirkpatrick, J.B., Bridle, K. and Lynch, A.J.J. (2002)Changes in vegetation and landforms at Hill One,Tasmania. Australian Journal of Botany 50:753–759.

Kleypas, J.A., Buddemeier, R.W., Archer, D., Gattuso,J.P., Langdon, C. and Opdyke, B.N. (1999)Geochemical consequences of increasedatmospheric carbon dioxide on coral reefs.Science 284: 118–120.


Knighton, A.D., Woodroffe, C.D. and Mills, K. (1992)The evolution of tidal creek networks, MaryRiver, Northern Australia. Earth SurfaceProcesses and Landforms 17: 167–190.

Kokita, T. and Nakazono, A. (2001) Rapid responseof an obligately corallivorous filefishOxymonacanthus longirostris (Monacanthidae)to a mass coral bleaching event. Coral Reefs 20:155–158.

Körner, C. (1999) Alpine Plant Life. Springer-Verlag,Berlin.

Latif, M., Kleeman, R. and Eckert, C. (1997) Green-house warming, decadal variability or El Niño—an attempt to understand the anomalous 1990s.Journal of Climate 10: 2221–2239.

Lavery, B.M., Young, G. and Nicholls, N. (1997) Anextended high quality historical rainfall data setfor Australia. Australian MeteorologicalMagazine 46: 27–38.

Lawlor, D.W. and Mitchell, R.A.C. (1991) The ef-fects of increasing CO

2 on crop photosynthesis

and productivity: a review of field studies. Plant,Cell and Environment 14: 807–818.

Leclerq, N., Gattuso, J-P. and Jaubert, J. (2002)Primary production, respiration and calcificationof a coral reef mesocosm under increased CO

2

partial pressure. Limnol. Oceanogr. 47: 558–564.

Li, J.H., Dijkstra, P., Hymus, G.J., Wheeler, R.M.,Piastuch, W.C., Hinkle, C.R. and Drake B.G.(2000) Leaf senescence of Quercus myrtifolia asaffected by long-term CO

2 enrichment in its na-

tive environment. Global Change Biology 6:727–733.

Lough, J.M. (1998) Coastal climate of northwestAustralia and comparisons with the Great BarrierReef: 1960 to 1992. Coral Reefs 17: 351–367.

Lough, J.M. (2000) 1997–98: Unprecedented ther-mal stress to coral reefs? Geophysical ResearchLetters 27: 3901–3904.

Lough, J.M. (in press) Weather on the Reef: AIMS’automatic weather station network. Bulletin of theAustralian Meteorological and OceanographicSociety.

Lough J.M. and Barnes D.J. (1997) Several centuriesof variation in skeletal extension, density andcalcification in massive Porites colonies from theGreat Barrier Reef: a proxy for seawatertemperature and a background of variabilityagainst which to identify unnatural change. Jour-nal of Experimental Marine Biology and Ecology211: 29–67.

Mann, M., Bradley, R. and Hughest, M.K. (1998)Global-scale temperature patterns and climateforcing over the past six centuries. Nature 392:779–787.

Mantua, N.J., Hare, S.R., Zhang, Y., Wallace, J.M.and Francis, R.C. (1997) A Pacific inter-decadalclimate oscillation with impacts on salmon pro-duction. Bulletin of the American Meteorologi-cal Society 78: 1069–1079.

Martin, T.E. (2001) Abiotic vs biotic influences onhabitat selection of coexisting species: climatechange impacts? Ecology 82: 175–188.

McCarty, J. (2001) Ecological consequences of re-cent climate change. Conservation Biology 15:320–331.

McGeoch, M.A. (1998) The selection, testing andapplication of terrestrial insects as bioindicators.Biological Reviews of the Cambridge Philosophi-cal Society 73: 181–201.

McGuffie, K. and Henderson-Sellers, A. (1994)Cloudiness trends this century from surface ob-servations. In: Proceedings of the InternationalMinimax Workshop, College Park, Maryland.Kukla, G.T. and Riches, M. (eds.), US Depart-ment of Energy, Washington DC, pp. 231–234.

McKeon, G.M. and Hall, W.B. (2002) Can seasonalclimate forecasting prevent land and pasturedegradation of Australia’s grazing lands? QNR14Technical Report for the Climate Variability inAgriculture Program. Queensland Department ofNatural Resources and Mines.

McKeon, G.M. and Howden, S.M. (1993) Adaptingthe management of Queensland’s grazing systemsto climate change. In: Climate Change:Implications for Natural Resource Conservation.Burgin, S. (ed.), University of Western Sydney,Hawkesbury, pp. 123–140.

McKeon, G.M., Hall, W.B., Crimp, S.J., HowdenS.M., Stone, R.C. and Jones, D.A. (1998) Cli-mate change in Queensland’s grazing lands: I.approaches and climatic trends. RangelandJournal 10: 151–176.

Millar, C.I. and Woolfenden, W.B. (1999) The roleof climate change in interpreting historical vari-ability. Ecological Applications 9: 1207–1216.

Mitchell, W., Chittleborough J., Ronai, B. and Lennon,G.W. (2000) Sea level rise in Australia and thePacific. Climate Change Newsletter 12: 7–10.

Morgan, V.I., Wookey, C.W., Li, J., van Ommen, T.D.,Skinner, W. and Fitzpatrick, M.F. (1997) Siteinformation and initial results from deep ice drill-ing on Law Dome. Journal of Glaciology 43:3–10.


Mulrennan, M.E. and Woodroffe, C.D. (1998) Salt-water intrusions into the coastal plains of theLower Mary River, Northern Territory, Australia.Journal of Environmental Management 54: 169–188.

Nerem, R.S. (1995) Global sea level rise variationsfrom TOPEX/POSEIDON altimeter data. Science268: 708–710.

Newell, G., Griffioen, P. and Cheal, D. (2001) Thepotential effects of ‘greenhouse’ climate-warm-ing scenarios upon selected Victorian plant spe-cies and vegetation communities. Internal Com-munication, Victoria Dept of Sustainability andEnvironment.

Nicholls, N. (1997) Increased Australian wheat yieldto recent climate trends. Nature 387: 484–485.

Nicholls, N., Drosdowsky, W. and Lavery, B. (1997)Australian rainfall variability and change. Weather52: 66–72.

Nicholls, N., Landsea, C. and Gill, J. (1998) Recenttrends in Australian region tropical cycloneactivity. Meterological Atmospheric Physics 65:197–205.

Nicholls, N., Lavery, B., Frederiksen, C. andDrosdowsky, W. (1996) Recent apparent changesin relationships between the El-Nino and Aus-tralian rainfall and temperature. GeophysicalResearch Letters 23: 3357–3360.

Nix, H.A. and Switzer, M.A. (1991) Rainforest Ani-mals: Atlas of Vertebrates Endemic to the WetTropics. Australian National Parks and WildlifeService, Canberra, Australia.

Osborne, W.S., Davis, M.S. and Green, K. (1998)Temporal and spatial variation in snow cover. In:Snow: A Natural History; An Uncertain Future.Green, K. (ed.) Australian Alps LiaisonCommittee, Canberra, pp. 56–68.

Osborne, J.L., Awmack, C.S., Clark, S.J., Williams,I.H. and Mills, V.C. (1997) Nectar and flowerproduction in Vicia faba L (field bean) at ambi-ent and elevated carbon dioxide. Apidologie 28:43–55.

Ostendorf, B., Hilbert, D.W. and Hopkins, M.S.(2001) The effect of climate change on tropicalrainforest vegetation pattern. EcologicalModelling 145: 211–224.

Owensby, C.E., Ham, J.M., Knapp, A.K. and Auen,L.M. (1999) Biomass production and speciescomposition change in a tall grass prairie ecosys-tem after long-term exposure to elevated atmos-pheric CO

2. Global Change Biology 5: 497–506.

Parmesan C.P. and Yohe G. (2003) A globally coher-ent fingerprint of climate change across naturalsystems. Nature 421: 37–42.

Peerderman, F.M., Davies, P.J. and Chivas, A.R.(1993) The stable oxygen isotope signal in shal-low-water, upper-slope sediments off the GreatBarrier Reef (Hole 820A). Proceedings of theOcean Drilling Program Scientific Results 133:163–173.

Peltier, W.R. and Tushingham, A.M. (1989) Globalsea level rise and the greenhouse effect: mightthey be connected? Science (USA) 244: 806–810.

Peñuelas, J. and Azcon-Bieto, J. (1992) Changes inleaf 13C of herbarium plant species during the lastthree centuries of CO

2 increase. Plant, Cell and

Environment 15: 485–489.

Peñuelas, J. and Matamala, R. (1990) Changes in Nand S leaf content, stomatal density and specificleaf area of 14 plant species during the last threecenturies of CO

2 increase. Journal of Experimen-

tal Botany 41: 1119–1124.

Pickup, G. and Stafford Smith, M. (1993) Problems,prospects and procedures for assessing thesustainability of pastoral land management in aridAustralia. Journal of Biogeography 20: 471–487.

Pittock, A.B. (1999) Coral reefs and environmentalchange: Adaptation to what? American Zoologist39: 10–29.

Plummer, N., Lin, Z. and Torok, S. (1995) Trends inthe diurnal temperature range over Australia since1951. Atmospheric Research 37: 79–86.

Pouliquen-Young, O. and Newman, P. (2000) TheImplications of Climate Change for Land-basedNature Conservation Strategies. Final Report 96/1306. Australian Greenhouse Office, Environ-ment Australia, Canberra, and Institute forSustainability and Technology Policy, MurdochUnviersity, Perth.

Power, S., Tseitkin, F., Vikram, M., Lavery, B., Torok,S., Holbrook, N. and Mehta, V. (1999) Decadalclimate variability in Australia during the twenti-eth century. International Journal of Climatology19: 169–184.

Prabhakara, C., Iacovazzi, R., Yoo, J.M. and Dalu,G. (1998) Global warming deduced from MSU.Geophysical Research Letters 25: 1927–1930.

Prentice, I.C. and Solomon, A.M. (1991) Vegetationmodels and global change. In: Global Changesof the Past. Bradley, R.S. (ed.), UCAR Office forInterdisciplinary Earth Studies, Boulder, CO, pp.365–383.

Pressey, R.L. and Taffs, K.H. (2001) Sampling of landtypes by protected areas; three measures of ef-fectiveness applied to western New South Wales.Biological Conservation 101: 105–117.


Pritchard, S.G., Rogers, H.H., Prior, S.A. andPeterson, C.M. (1999) Elevated CO

2 and plant

structure: a review. Global Change Biology 5:807–837.

Quayle, R.G. and Karl, T.R. (1996) Climate ChangeUpdate; The State of the Climate—1996; ASummary of the 1995 Policymakers Report of theIPCC. National Climatic Data Centre, Asheville,North Carolina.

Read, J. and Hill, R.S. (1985) Dynamics ofNothofagus dominated rainforest on mainlandAustralia and lowland Tasmania. Vegetation 63:67–78.

Reed, D.C., Raimondi, P.T., Carr, M.H. andGoldwasser, L. (2000) The role of dispersal anddisturbance in determining spatial heterogeneityin sedentary organisms. Ecology 81: 2011–2026.

Rex, M., Harris, N.R.P., Vondergthen, P., Lehmann,R., Braathen, G.O., Reimer, E., Beck, A.,Chipperfield, M.P., Alfier, R., Allaart, M.,Oconnor, F., Dier, H., Dorokhov, V., Fast, H., Gil,M., Kyro, E., Litynska, Z., Mikkelsen, I.S.,Molyneux, M.G., Nakane, H., Notholt, J.,Rummukaien, M., Viatte, P. and Wenger, J. (1997)Prolonged stratospheric ozone loss in the 1995–96 Arctic winter. Nature 389: 835–838.

Rochefort, L. and Woodward, F.I. (1992) Effect ofclimate change and a doubling of CO

2 on vegeta-

tion diversity. Journal of Experimental Botany43: 1169–1180.

Roderick, M.L. and Berry, S.L. (2001) Linking wooddensity, tree growth and environment: a theoreti-cal analysis based on the motion of water. NewPhytologist 149: 473–485.

Roderick, M.L., Berry, S.L. and Noble, I.R. (1999)The relationship between leaf composition andmorphology at elevated CO

2. New Phytologist

143: 63–72.

Rogers, H.H., Runion, G.B. and Krupa, S.V. (1994)Plant responses to atmospheric CO

2enrichment

with emphasis on roots and the rhizosphere.Environmental Pollution 83: 155–189.

Root, T.L, Price, J.T., Hall, K.R., Schneider, S.H.,Rosenzweig, C. and Pounds, J.A. (2003) Finger-prints of global warming on wild animals andplants. Nature 421: 57–60.

Roshier, D.A., Whetton, P.H., Allan, R.J. andRobertson, A.I. (2001) Distribution and persist-ence of temporary wetland habitats in arid Aus-tralia in relation to climate. Australian Ecology26: 371–384.

Rozema, J., Van de Staaij, J., Bjorn, L.O. andCaldwell, M. (1997) UV-B as an environmentalfactor in plant life: stress and regulation. Trendsin Ecology Evolution 12: 22–28.

Sahagian, D.L., Schwartz, F.W. and Jacobs, D.K.(1994) Direct anthropogenic contributions to sea-level rise in the 20th century. Nature 367: 54–57.

Saintilan, N. and Williams, R.J. (1999) Mangrovetransgression into saltmarsh environments. Glo-bal Ecology and Biogeography 8: 117–124.

Salawitch, R.J. (1998) A greenhouse warning con-nection. Nature 392: 551–552.

Salinger, M.J. and Griffiths, G.M. (2001) Trends inNew Zealand daily temperature and rainfall ex-tremes. International Journal of Climatology 21:1437–1452.

Salinger, M.J., Allan, R., Bindoff, N., Hannah, J.,Lavery, B., Lin, Z., Lindesay, J., Nicholls, N.,Plummer, N. and Torok, S. (1996) Observed vari-ability and change in climate and sea level inAustralia, New Zealand and the South Pacific.In: Greenhouse: Coping with Change. Bouma,W.J., Pearman, G.I. and Manning, M.R. (eds),CSIRO, Collingwood, Australia, pp. 100–126.

Scherrer, P. and Pickering, C. (2001) Effects of graz-ing, tourism and climate change on the alpinevegetation of Kosciuszko National Park. Victo-rian Naturalist 118: 93–99.

Shindell, D.T., Rind, D. and Lonergan, P. (1998)Increased polar stratospheric ozone losses anddelayed eventual recovery owing to increasinggreenhouse-gas concentrations. Nature 392:589–592.

Skirving, W., Mahoney, M. and Steinberg, C. (2002)Sea Surface Temperatures Atlas of the GreatBarrier Reef 1990–2000. Available on CD-ROMfrom the Australian Institute of Marine Scienceor by emailing [email protected].

Skvarca, P. (1993) Fast recession of the northernLarsen Ice Shelf monitored by satellite images.Annals of Glaciology 17: 317–321.

Spalding, M. and Jarvis, G.E. (2002) The impact ofthe 1998 coral mortality on reef fish communi-ties in the Seychelles. Marine Pollution Bulletin44: 309–321.

Specht, R.L. (1972) Water use by perennial evergreenplant communities in Australia and Papua NewGuinea. Australian Journal of Botany 20: 273–299.

Specht, R.L. (1981) Ecophysiological principles de-termining the biogeography of major vegetationformations in Australia. In: Ecological Biogeog-raphy of Australia. Keast, A. (ed.). W. Junk, TheHague, pp. 301–333.

Spellerberg, I.F. (1991) Monitoring EcologicalChange. Cambridge University Press, Cam-bridge.


Stafford Smith, M. and Morton, S. (1990) A frame-work for the ecology of arid Australia. Journalof Arid Environments 18: 255–278.

Stone, R., Nicholls, N. and Hammer, G. (1996) Frostin northeast Australia: trends and influences ofphases of the southern oscillation. Journal ofClimate 9: 1896–1909.

Suppiah, R. and Hennessy, K.J. (1996) Trends in theintensity and frequency of heavy rainfall in tropi-cal Australia and links with the Southern Oscilla-tion. Australian Meteorological Magazine 45:1–17.

Suppiah, R. and Hennessy, K.J. (1997) Trends in totalrainfall, heavy rainfall events and number of drydays in Australia, 1910–1990. International Jour-nal of Climatology 18: 1141–1164.

Sweatman, H., Cheal, A., Coleman, G., Delean, S.,Fitzpatrick, B., Miller, I., Ninio, R., Osborne, K.,Page, C. and Thompson, A. (2001) Long-termMonitoring of the Great Barrier Reef. StatusReport Number 5, 2001. Australian Institute ofMarine Science, 106 pp.

Swetnam, T.W., Allen, C.D. and Betancourt, J.L.(1999) Applied historical ecology: using the pastto manage for the future. Ecological Applications9: 1189–1206.

Theurillat, J.P. and Guisan, A. (2001) Potential im-pact of climate change on vegetation in the Euro-pean Alps: a review. Climate Change 50: 77–109.

Tidemann, C.R. (1999) Biology and management ofthe grey-headed flying fox, Pteropuspoliocephalus. Acta Chiropterol 1: 151–164.

Tongway, D. (1994) Rangeland Soil ConditionAssessment Manual. CSIRO Division of Wildlifeand Ecology, Australia. 69 pp.

Tongway, D. and Hindley, N. (1995). Manual forAssessment of Soil Condition for TropicalGrasslands. CSIRO Division of Wildlife andEcology, Australia. 60 pp.

Torok, S.J. and Nicholls, N. (1996) A historical an-nual temperature data set for Australia. Austral-ian Meteorological Magazine 45: 251–260.

Trench, R.K. (1979) The cell biology of plant–animalsymbiosis. Annual Review of Plant Physiology30: 485–531.

Trenberth, K.E. and Hoar, T.J. (1996) The 1990–1995El Nino Southern Oscillation event: longest onrecord. Geophysical Research Letters 23: 57–60.

Trenberth, K.E. and Hoar, T.J. (1997) El Nino andclimate change. Geophysical Research Letters 24:3057–3060.

Vaughan, D. and Lachlan-Cope T. (1996) Recentretreat of ice shelves on the Antarctic Peninsula.Weather 50: 374–376.

Wahren, C.H., Papst, W.A. and Williams, R.J. (1994)Long-term vegetation change in relation to cattlegrazing in subalpine grassland and heathland onthe Bogong High Plains: an analysis of vegeta-tion records from 1945 to 1994. AustralianJournal of Botany 42: 607–639.

Walker, J. (1998) Malaria in a changing world—anAustralian perspective. International Journal forParasitology 28: 947–953.

Walter, M. and Broome, L. (1998) Snow as a factorin animal hibernation and dormancy. In GlobalThreats to the Australian Snow Country. Abstractof a paper presented at the Global Threats to theAustralian Snow Country gathering, 17–19February 2001 at Jindabyne, NSW. AustralianInstitute of Alpine Studies newsletter 1, http://www.aias.org.au/newsletters/newsletter1.htm#snow.

Walther, G.R., Post, E., Convey, P., Menzel, A.,Parmesan, C., Beebee, T.J.C., Fromentin, J.R.,Hoegh-Guldberg, O. and Bairlein, O. (2002)Ecological responses to recent climate change.Nature 416: 389–395.

Watkins, A.B. (2001) Seasonal climate summarysouthern hemisphere (Spring 2000): a third suc-cessive positive phase of the Southern Oscilla-tion begins. Australian Meteorological Magazine50: 295–308.

Watson, I. and Thomas, P. (2003) Monitoring showsimprovement in Gascoyne–Murchisonrangelands. Range Management Newsletter 03/1, 11–14.

Watson, I., Blood, D., Novelly, P., Thomas, P. andVan Vreeswyk, S. (2001) Rangeland Monitoring,Resource Inventory, Condition Assessment andLease Inspection Activities in Western AustraliaConducted by the Department of Agriculture.Report prepared for the National Land and WaterResources Audit, Canberra. http://audit.ea.gov.au/anra/.

Wearne, L.J. and Morgan, J.W. (2001) Recent forestencroachment into subalpine grasslands nearMount Hotham, Victoria, Australia. Arctic Ant-arctic and Alpine Research 33: 369–377.

Wentz, F.J. and Schabel, M. (1998) Effects of orbitaldecay on satellite-derived lower-tropospherictemperature trends. Nature 394: 661–664.

Westoby, M. (1991) On long-term ecological researchin Australia. In: Long-term Ecological Research.Risser, P. (ed.) SCOPE, John Wiley and Sons,pp. 191–209.


Whetton, P. (1998) Climate change impacts on thespatial extent of snow-cover in the AustralianAlps. In Snow: A Natural History; an UncertainFuture. Green, K. (ed.) Australian Alps LiaisonCommittee, Canberra, pp. 195–206.

White, W.B., Cayan, D.R. and Lean, J. (1998) Globalupper ocean heat storage response to radiativeforcing from changing solar irradiance and in-creasing greenhouse gas/aerosol concentrations.Journal of Geophysical Research—Oceans 103:21355–21366.

Whitehead, P. (2000) A review of changes in the sta-tus of threatening processes. Report to the Na-tional Land and Water Resources Audit by theTropical Savannas CRC. Darwin, pp. B1-125 toB1-324.

Wilkinson, C.R. (1996) Global change and coral reefs:impacts on reefs, economies and human cultures.Global Change Biology 2: 547–558.

Wilkinson, C.R. and Buddemeier, R.W. (1994) Glo-bal Climate Change and Coral Reefs:Implications for People and Reefs. Report of theUNEP-IOC-ASPEI-IUCN Global Task Team onthe Implications of Climate Change on CoralReefs. IUCN, Gland, Switzerland, 124 pp.

Williams R.J. and Costin A.B. (1994) Alpine andsubalpine vegetation. In: Australian Vegetation.Groves, R.H. (ed.) Cambridge University Press,Cambridge.

Williams, S.E. and Hero, J.M. (2001) Multiple deter-minants of Australian tropical frog biodiversity.Biological Conservation 68: 1–10.

Williams, S.E. and Pearson, R.G. (1997) Historicalrainforest contractions, localised extinctions andpatterns of vertebrate endemism in the rainfor-ests of Australia’s Wet Tropics. Proceedings ofthe Royal Society of London B 264: 709–716.

Williams, S.E., Marsh, H. and Winter, J. (2002) Spatialscale, species diversity and habitat structure: smallmammals in Australian tropical rainforest.Ecology 83: 1317–1329.

Williams, S.E., Pearson, R.G. and Walsh, P.J. (1996)Distributions and biodiversity of the terrestrialvertebrates of Australia’s Wet Tropics: a reviewof current knowledge. Pacific ConservationBiology 4: 327–362.

Winter, J.W. (1988) Ecological specialization ofmammals in Australian tropical and sub-tropicalrainforest: refugial and ecological determinism.In: The Ecology of Australia’s Wet Tropics.Kitching, R. (ed.). Surrey Beatty, Sydney.

Woinarski, J. (2000) A review of changes in the sta-tus and threatening processes. Report to theNational Land and Water Resources Audit by theTropical Savannas CRC. Darwin, pp. B1-125 toB1-126.

Woodroffe, C.D. and Mulrennan, M.E. (1993)Geomorphology of the Lower Mary River Plains,Northern Territory. Australian National Univer-sity and the Conservation Commission of theNorthern Territory, Darwin.

Woodroffe, C.D., Chappell, J.M.A., Thom, B.G. andWallensky, E. (1986) Geomorphological Dynam-ics and Evolution of the South Alligator TidalRiver and Plains, Northern Territory. NorthAustralia Research Unit, Darwin.

Wright, W.J., de Hoedt, G., Plummer, N., Jones, D.A.and Chitty, S. (1996) Low frequency climatevariability over Australia. The Second Austral-ian Conference on Agricultural Meteorology,1–4 October 1996, University of Queensland.University of Queensland Printer, Brisbane.

Zheng, X., Basher, R.E. and Thompson, C.S. (1997)Trend detection in regional-mean temperature:maximum, minimum, mean temperature, diurnalrange and SST. Journal of Climate 10: 317–326.

Ziska, L.H. and Caulfield, F.A. (2000) Rising CO2

and pollen production of common ragweed(Ambrosia artemisiifolia), a known allergy-induc-ing species: implications for public health.Australian Journal of Plant Physiology 27:893–898.

Zuo, Z. and Oerlemans, J. (1997) Contribution ofglacier melt to sea-level rise since 1865—a regionally differentiated calculation. ClimateDynamics 13: 835–845.


!��'

adaptation 3, 4, 67, 72, 73–75, 76–77, 78–79in coral 18

aerosols 8, 10algae 15, 17, 18, 19, 20, 62alpine, subalpine 2, 33, 35, 61, 66, 78animal responses to climate 35, 61, 62, 66

birds 35, 37–39, 40, 59feral mammals 35indicators 59livestock 9subantarctic 56tropical 49, 50, 51, 52

antarctic 9, 15, 55–57Antarctica 9, 55–57arid (see rangelands)bats (flying foxes) 62, 66beneficiaries advantaged by climate change 4, 35,

68, 76, 7, 78bioclimates 66bioclimatic profiles 53biodiversity, levels of 74birds 15, 16, 35, 37–39, 40–42, 49, 50, 52, 56, 62bleaching, of coral 16, 17, 18, 20C

3, C

4plants 61, 66

carbon, sinks or sources 34carbon dioxide (CO

2) concentrations 8, 9, 11, 14,

21doubled 48variation in 46, 61, 66

climate change 1, 7, 8, 15, 31, 75impacts 2, 7, 22–24, 30, 33, 34, 40, 49, 66, 67,

75–76on animals 14–16, 50–52 (see also animal

responses, insect responses)on biodiversity 30, 56–57, 70, 71, 75

defined 67on plants 11, 14–16, 23, 51–52 (see also

plant responses)in past 43–44, 48social, economic, tourism 53, 62, 70, 71, 75

interaction with human impacts 22, 30, 45, 59,78

scenarios of 11, 12 (see also temperature,rainfall projections)

cloud 8, 48, 51, 52, 62, 66coastal marine 2coral 2, 17–21, 61, 66costs 4, 40–41, 68Council of Australian Governments 1, 7crustaceans 19cyclones 10, 21, 222data-sets, existing long-term 60

alpine 33, 35–36birds 40–42lack of 59rangelands (WARMS) 25–28various 60, 61–62

detecting climate change 29–30 (see alsoindicators)

diseases 15, 57, 61distributions, shifting 6, 15, 26, 37–39, 59, 66, 76,

77economic, agricultural indicators 62ecosystem services 19, 48, 72, 79El Niño Southern Oscillation (ENSO) 10, 18, 38,

43, 62emissions 10, 11, 70evaporation 1, 12, 34extinctions 2, 19, 51fire 2, 3, 29, 78fish 17, 19forests 43, 61

cloud 15, 48temperate 2, 15tropical 2, 15, 48–49, 66

freshwaters, estuaries 60, 61, 66river flows 12, 66, 76, 78

frogs 49, 51, 52, 59, 66frost 9, 33 (see also ice)glacial 9global warming 1, 33, 35, 48, 53, 69, 70 (see also

temperature, rising)grasslands 25–26, 66Great Barrier Reef 16, 17, 18, 19, 20, 43, 78


greenhouse gases 8, 11, 19Heard Island, 55–57ice 9, 35, 55impacts, see climate change impactsIndian Ocean Climate Initiative 31indicators 4, 38, 58–62

organisms 38, 40, 61–62detectors 4, 58sentinels 4, 58

processes 6, 59information gaps/needs 4, 67, 75–77insect responses 15, 59, 61, 66mammals 35, 49, 52, 57, 66managing for climate variability 3, 4, 68–78

risk 5, 69, 72methane 8migration 57, 76–77mitigation 4, 67, 68, 69, 79modelling 63–66, 70 (see also climate change

scenarios)statistical 31–32tropical habitats 49Victorian plant communities 53–54

models 64ANUCLIM 53–54BIOCLIM 50, 51, 64GRASP 23, 64neural networks 50, 64

monitoring, inventory 6, 60, 70, 74, 79 (see alsodata-sets, indicators)

alpine 35–36existing 60, 80in UK 16, 60lack of 7, 59need for 4, 16, 34, 60, 74Western Australian data 25–28

nitrous oxide 8ozone 10–11, 57palaeoecology 43–47, 48peat 34, 57phenological responses 14, 15, 59, 62physiological responses 14, 20, 59, 61plankton 2, 15plant responses to climate change 9, 14, 15, 26,

49, 53–54, 61, 62, 66, 74antarctic 56–57in rangelands 22, 23, 29past 43to CO

2 11, 46–47, 48

policy development 67–79information needs 67–68, 74–77investment 70, 73, 75

objectives 72–73recommendations 78–79risk 69, 72triage approach 71, 72, 78uncertainty 68–69

policy programs 80precipitation 9, 46, 61, 62 (see also rainfall)preservation of species 3, 71, 72, 78, 79radiation 46rainfall 9–10, 31, 50, 61, 66 (see also precipitation)

Antarctica 55projections 11, 13, 22variability 29, 51, 52

rangelands 2, 22–30, 78refugia, refuges 77, 79reptiles 15, 51, 52reserves 3, 24resilience 70, 76 ,77

biodiversity 24coral 19marine 74

revegetation, conflict possible 3, 74risk assessment 69sea level 10, 21, 61, 62sea-surface temperature 9, 15, 18, 20, 32, 61

past 43sea-water chemistry 21snow 9, 35, 36, 61, 66Snowy Mountains 35–36Southern Oscillation Index (SOI) 10soils 25, 34, 52, 56stress 16, 20, 24, 66, 68

hypo-osmotic 21, 66thermal 18, 19

subalpine (see alpine)subantarctic (see antarctic)Tasmania 33–34temperature 1, 6, 8, 51, 61, 62

effect of variability 1, 4, 6, 51factors raising 8, 17measures of 8, 17modified by vegetation 46past 8, 9, 43projections 1, 6, 11, 12, 18, 22, 35rising 31, 48–49, 50, 51, 55

translocation 3, 73, 74, 77, 79ultraviolet (UV) 10, 57vulnerability 24, 68weeds 3, 45 (see also beneficiaries)Western Australia 25–28, 31Wet Tropics 48–49, 50–52

Documents

Climate Change Impacts on Biodiversity in Australia · Climate change impacts on biodiversity in Australia: outcomes of a workshop, October 2002 1 The workshop A workshop held in