Climate Change Under Enhanced Greenhouse Conditions In Northern Australia
Final Report (1994-1997)
Contents
Executive Summary
- Introduction
- Present Climate Variations and Interactions
- Temperature and Rainfall Changes Under Enhanced Greenhouse Conditions
- Simulations of Northern Australian Climatic Variability During the Past Century
- El Nino - Southern Oscillation (ENSO) Phenomenon
- Extreme Events Including Tropical Cyclones
- Potential Climate Change Impacts for Northern Australia
- Conclusions and Recommendations
Contents
Consultancy Details
Report on research undertaken as part of consultancy for the:
- Department of Lands, Planning and Environment, Northern Territory
- Department of Environmental Protection, Western Australia and in Queensland,
- Department of Primary Industries, Department of Natural Resources, Department of Business Industry and Regional Development, Department of the Premier, Economic Development and Trade, Department of Family Services, Aboriginal and Islander Affairs, Department of Environment and Heritage, Department of Mines and Energy, Department of Lands, Department of Housing, Local Government and Planning, Department of Transport and Tourist and Travel Corporation
CSIRO Division of Atmospheric Research Ramasamy Suppiah, Kevin Hennessy, Tony Hirst, Roger Jones, Jack Katzfey, Barrie Pittock, Kevin Walsh, Peter Whetton and Steve Wilson
January 1998
Important Disclaimer
This report relates to climate simulations based on computer modelling. Models involve simplifications of real physical processes that are not fully understood. Accordingly, no responsibility will be accepted by CSIRO or the clients (the Northern Territory Department of Lands, Planning and Environment; Queensland Department of Primary Industries; Department of Natural Resources; and the Western Australian Department of Environmental Protection) for the accuracy of forecasts or predictions inferred from this report or for any person's interpretations, deductions, conclusions or actions in reliance of this report.
Contact Details
Address for correspondence:
CSIRO Division of Atmospheric Research
PMB 1 Aspendale, VIC 3195
Telephone: (03) 9239 4400
Fax: (03) 9239 4444
e-mail: rms@dar.csiro.au
January 1998
ISBN 0 643 06017 0
For further information please contact:
Environment Protection Division
Department of Lands, Planning and Environment, Northern Territory
GPO Box 1680
Darwin NT 0801
Ph: (08) 89 244004
Department of Natural Resources, Queensland
80 Meiers Road
Indooroopilly Qld 4068
Ph: (07) 3896 9379
Department of Environmental Protection, Western Australia
Westralia Square
141 St George's Terrace
Perth WA 6000
Ph: (08) 9476 7406
Acknowledgements
Cher Page performed data processing, data manipulations and plotting.
Observational data have been provided by the National Climate Centre of the Bureau of Meteorology, Melbourne. The European Centre for Medium Range Weather Forecasts supplied global atmospheric data. Greg Holland and Noel Davidson from the Bureau of Meteorology Research Centre supplied the code for Holland's maximum potential intensity (MPI).
Liaison between CSIRO Division of Atmospheric Research (DAR) and the Governments has been facilitated by Roger Stone from the Department of Primary Industries; and Greg McKeon, Ken Brook and Terry Loos from the Department of Natural Resources; Queensland, Michael Ward from the Department of Lands, Planning and Environment, Northern Territory; and Ruth Clark from the Department of Environmental Protection, Western Australia.
Editorial support was provided by Paul Holper and the layout was designed by Julie Penn.
This work was produced by CSIRO DAR as part of its Climate Change Research Program with the support of funding from the Department of Lands, Planning and Environment, Northern Territory, Department of Environmental Protection, Western Australia, and in Queensland, Department of Primary Industries, Department of Natural Resources, Department of Business Industry and Regional Development, Department of the Premier, Economic Development and Trade, Department of Family Services, Aboriginal and Islander Affairs, Department of Environment and Heritage, Department of Mines and Energy, Department of Lands, Department of Housing, Local Government and Planning, Department of Transport and Tourist and Travel Corporation and also from the Commonwealth via the National Greenhouse Research Program.
Executive Summary
Introduction
This report describes simulations of climate change and variability under enhanced greenhouse conditions in northern Australia. This The work of the authors draws upon research findings of many colleagues within the Division of Atmospheric Research. CSIRO global climate models (GCMs) were developed largely by the members of the Climate Modelling Group of this Division. John McGregor, Jack Katzfey and Kim Nguyen carried out the limited-area modelling experiments. Barrie Hunt and Ian Smith have done the GCM experiment with observed SST. Their continued cooperation and support is greatly appreciated. is the final report of a three year consultancy (1994-97) with the Governments of the Northern Territory, Western Australia and Queensland.
Five climatic aspects are assessed: (1) the interaction between the monsoon, tropical cyclones, the El Niño - Southern Oscillation (ENSO) phenomenon and variations within the monsoon season, (2) the ability of regional and global climate models to simulate present-day and future climates, (3) the behaviour of the ENSO phenomenon under enhanced greenhouse conditions, (4) changes in extreme events including tropical cyclones, and (5) potential impacts of climate change and identification of future research projects of an applied nature.
Climate change due to the likely increase in greenhouse gas concentrations is now inevitable. Climate change will continue to increase until a stabilisation of greenhouse gas concentrations in the atmosphere is reached. We therefore need the best possible estimates of regional climate changes and of the sensitivity of various sectors to such changes.
During the consultancy period, we have used several approaches to determine possible future changes in north Australian climate, including temperature and rainfall patterns, and the behaviour of the ENSO phenomenon and tropical cyclones. These approaches include results from global climate models (GCMs), theoretical considerations, and the application of higher resolution regional climate models "nested" in the GCMs.
The results bring out the importance of the progression from the simple "slab-ocean" GCMs available at the start of the 3-year project, to the more detailed "coupled-ocean" GCMs which have now become available. The former were not capable of simulating gradual changes in the oceans, and gave results appropriate to a changed "equilibrium" climate which will not occur in the foreseeable future. The latter can simulate a gradually changing or "transient" climate more like what we should expect in the coming decades.
We must therefore move rapidly to climate change assessments based on coupled-ocean GCM simulations, while bearing in mind that even these may not yet be fully realistic. In particular, the recent generation of coupled-ocean GCMs, with their coarse spatial resolution, do not realistically simulate ENSO. A major task over the next few years is to refine coupled-ocean GCMs to provide ENSO simulations in which we can have confidence, and to use the models also to examine other climatic changes, including possible changes in the behaviour of tropical cyclones.
Model results indicate a range of plausible changes in rainfall, higher temperatures and increased rainfall intensities, and possible changes in ENSO and cyclone behaviour. The practical implications of these scenarios of climate change suggest possible widespread impacts across northern Australia. Such impacts may affect coastal zones, ecosystems, agriculture, water resources, the mining industry, tourism, human health, fisheries, pests, disease, mangroves, coral reefs and low lying islands.
The importance of developing better methods of forecasting the possible climate changes resulting from the enhanced greenhouse effect has now been widely recognised and the results of the consultancy suggest promising future lines of investigation. The need to translate this information into possible scenarios (and responses) for different sectors of the economy has also been widely recognised. The Queensland Government has moved in this regard to establish the Queensland Centre for Climate Applications (QCCA). With the advent of QCCA, we look forward to a growing interaction and collaboration with research and applications agencies in the region. Even in the absence of confident local scenarios of climate change, developing the ability to assess the impacts of climate change on particular activities and sectors is a valuable undertaking. It will enable rapid assessments to be made of potential impacts and adaptation strategies once the uncertainties in the climate scenarios are reduced.
Present climate variations and interactions
Globally, surface temperature measured in standard meteorological screens at 1.5 m above the ground has increased by 0.3 to 0.6°C since the late 19th century. In Australia, the average temperature has increased by 0.76°C from 1910 to 1990. This estimate accounts types of instruments and their changes and urban heat island effects, mainly by excluding data from stations affected by these problems. There are however major differences between measured temperatures near the surface and those obtained from satellites which measure temperature over a considerable depth of the lower atmosphere. These differences are not surprising as the measurements are of different quantities. Moreover, the satellite-measured 19-year temperature record is too short to determine long term trends (see the CSIRO Division of Atmospheric Research World Wide Web site at: http://www.dar.csiro.au/Christy.htm.
Observed records of wind and rainfall have been analysed to understand within-season and year-to-year variations of the Australian monsoon. The primary purpose is to learn about the relationships between monsoon behaviour and the cross-equatorial flow of air from south-east Asia.
A strong relationship has been found between the Australian monsoon onset, rainfall variability over northern Australia and wind surges from the South China Sea and west coast of Australia, but the relationship is weak once the monsoon is well established.
This work provides information useful for short-term (five to ten days) forecasting of monsoon activity, particularly the active and break periods of the monsoon in tropical Australia based on wind variations over the South China Sea and West Australian coast. The results were also used as a basis for assessing the realism of simulations of the present-day climate from CSIRO global and regional climate models.
Temperature and rainfall changes under enhanced greenhouse conditions
Global climate models (GCMs) suggest a warming of 0.4 to 1.4°C inland and 0.3 to 1.0°C in northern coastal Australia by the year 2030. This range accounts for the uncertainties about future greenhouse gas emission scenarios (low case is Intergovernmental Panel on Climate Change (IPCC) scenario IS92c, and high case is IS92e) and the IPCC estimated range of global climate sensitivities (from 1.5 to 4.5°C equilibrium global warming for doubled CO2 conditions). See Appendix II for more details.
Simulated changes of winter rainfall by GCMs with simplified oceans (slab models) and models with full ocean circulation features (coupled-ocean models) broadly agree on decreases in northern Australia. Winter rainfall decreases by 0 to 8 per cent by 2030, although winter rainfall is low except in south-east Queensland.
Summer rainfall changes simulated under enhanced greenhouse conditions over northern Australia are model dependent. Changes simulated by five slab-ocean GCMs indicate increases from 2 to 12 per cent by the year 2030. As in other slab-ocean models, the CSIRO Mark 2 slab-ocean GCM also simulates an increase in summer rainfall in northern Australia.
Summer rainfall changes simulated by five coupled-ocean models show a decrease of 0 to 8 per cent by the year 2030. The latest CSIRO coupled-ocean model also simulates a decrease in summer rainfall in northern Australia.
Changes in both temperature and rainfall would be up to twice as large by 2070 compared with estimates for 2030.
Coupled-ocean model simulations are more realistic than slab-ocean experiments in that they can directly simulate climate change under transient conditions (steadily increasing greenhouse gas concentrations), as will be the case during the next century. Slab-ocean models simulate the results of an "equilibrium" climate change, that is, a stable climatic state after an increase in greenhouse gas concentrations. However, coupled-ocean model transient results are dependent on ocean behaviour which may not as yet be reliably simulated.
The regional climate model of CSIRO (DARLAM) simulates much more detailed and realistic rainfall patterns for present conditions when nested in the CSIRO Mark 2 slab-ocean GCM. DARLAM simulates an increase in summer rainfall in the western half and a decrease in the eastern half of tropical Australia under equilibrium doubled carbon dioxide (2xCO2) conditions. There is a need to nest DARLAM in the CSIRO coupled-ocean model to get more detailed information about rainfall change under simulated transient warming conditions.
Simulations of northern Australian climatic variability during the past century
The CSIRO atmospheric GCM has been used to simulate climatic variability during the past century. Observed sea surface temperatures since 1871 were used to represent the ocean. Results of five simulations of the atmospheric GCM using observed sea surface temperature indicate that there is good potential to predict ENSO impacts under present or enhanced greenhouse conditions once modelling of the oceanic processes is improved. However, problems remain regarding coarse spatial resolution and inadequate simulation of smaller-scale weather phenomena such as intense tropical depressions and tropical cyclones.
El Niño - Southern Oscillation (ENSO) phenomenon
Different models give different results about ENSO under enhanced greenhouse conditions. Both coarse resolution GCMs and fine resolution models indicate that the ENSO cycle continues to occur under simulated enhanced greenhouse conditions. The CSIRO fine resolution ENSO model (which simulates present ENSO events very realistically), run under near-equilibrium 2xCO2 conditions, simulates more frequent and weaker ENSO events, with increased average rainfall over northern Australia. However, the coarse resolution global coupled-ocean GCMs do not give such an increase in Australian rainfall. While there has been a significant improvement in the ability of models to simulate ENSO, at present there is no clear consensus on how ENSO might change under transient enhanced greenhouse conditions. It is possible that the response may change with time as the climate system moves from a transient warming to an equilibrium state (but this will not occur until well beyond 2100). Clearly, the nature of the transient response is a matter of high priority for further research.
Extreme events including tropical cyclones
Simulations using DARLAM nested in the CSIRO slab-ocean GCM indicate increases in heavy rainfall intensity during spring and summer under enhanced greenhouse conditions. Weather systems that produce heavy rainfall in north and central Australia become more intense, with heavier rainfall. Observations show that annual heavy rainfall intensity has increased by over 50 per cent (statistically significant) since 1910 in the north-west, with a 60 per cent increase in autumn (not significant). In the north-east, annual heavy rainfall has increased by 15 per cent and autumn heavy rainfall increased by 30 per cent, but these changes are not significant.
Sensitivity studies indicate an increase in the number of hot days and a decrease in the number of cold days. Again, similar tendencies have been observed since 1950 in Australia.
Simulated tropical cyclones in the Coral Sea and off north-west Australia tend to form and travel further south under 2xCO2 conditions. There are also indications that tropical cyclones could become more intense. Both results need further investigation.
Moreover, results to date have been obtained using DARLAM nested in a slab-ocean GCM which gives changes under equilibrium 2xCO2 conditions. It will be valuable to examine the possible changes to extreme events and tropical cyclone behaviour under transient (non-equilibrium) conditions as simulated by the newer CSIRO coupled-ocean GCM.
As yet there is no international consensus about changes to tropical cyclone behaviour due to the enhanced greenhouse effect, apart from a probable moderate increase in intensity. Changes found in the CSIRO simulations need to be studied carefully to establish their physical basis and statistical significance.
Potential climate change impacts for northern Australia
Initial results of our efforts to simulate likely climate change under enhanced greenhouse conditions suggest little room for complacency about potential impacts, despite large uncertainty. The practical implications of possible reductions in rainfall, higher temperatures and rainfall intensities, and possible changes in ENSO and cyclone behaviour, suggest potential for widespread impacts across northern Australia. Such impacts are likely on coastal zones, ecosystems, agriculture, water resources, the mining industry, tourism, human health, fisheries, pests, disease, mangroves, coral reefs and low lying islands. These impacts would generally increase in magnitude until well after greenhouse gas concentrations are stabilised.
Conclusions and recommendations
- Some significant degree of climate change across northern Australia now seems inevitable, and is likely to become increasingly apparent over the next 30-100 years as atmospheric equivalent CO2 concentrations exceed doubled pre-industrial concentrations. Changes are to be expected in both the average values and in the magnitude and frequency of extremes. This means that long-term planning should not be predicated on the assumption that future climate statistics and resources will be as they were over the last 100 years. Significant adaptation to a changing climate will be necessary.
- In particular, water supply, floods and drought frequency and severity, and the frequency of tropical cyclones and storm surges could change significantly. This has strong implications for the sustainable development of planned infrastructure including coastal developments, ports, bridges and urban centres.
- Higher temperatures and a changing frequency of drought and fire could have increasing impacts on agriculture and forestry. Significant changes in management may be required to minimise costs, maximise benefits, and ensure sustainability.
- Although there is a tendency for significant changes in the frequency of some extreme weather events in Australia over the last 100 years, it is not yet clear whether these changes are related to warming due to natural variability or anthropogenic influences.
- In seems likely that the relationship between rainfall in north-eastern Australia and the Southern Oscillation Index will continue under enhanced greenhouse conditions, although it may change. Thus seasonal rainfall predictions based on updated recent statistics may provide one means of adaptation to climate change.
- Decadal and century scale climate change is expected to affect the present savanna and tropical ecosystems in northern Australia. Some animal and plant species may come under increasing stress, causing long-term change in species composition.
- Coastal ecosystems will also be affected by sea-level rise and changes in runoff. Sea level is expected to rise due to thermal expansion of the oceans and melting of mid- and low-latitude glaciers. An increase in temperature in the Antarctic region would initially lead to an increased accumulation of ice, but to increased meltwater contributing to sea-level rise beyond the twenty-first century.
- Significant uncertainties remain in relation to the estimation of future climate. These can be reduced by: -
- moving to climate change scenario development based on improved transient coupled-ocean GCMs;
- improving the ability of coupled-ocean GCMs to simulate ENSO behaviour under present and enhanced greenhouse conditions;
- assessing the behaviour of tropical cyclones with DARLAM nested at fine resolution in a coupled-ocean GCM.
- Climate impact and adaptation assessment should be furthered through:-
- Development of versatile climate impact and adaptation models and methodologies for a number of key sectors and activities. These models should be developed and tested against observations as early as possible, so that these models can be used to assess the impacts of more reliable climate change scenarios as soon as these become available. Priority should be on the basis of potential sensitivity, impact model availability and stakeholder interest, and should include not only agriculture, water resources and coastal impacts, but also sectors such as health, transport, land planning and emergency services.
- This work should be done through close collaboration between climate modellers in CSIRO and relevant researchers and authorities in northern Australia across a wide range of sectors and activities.
- More integrated impact and adaptation assessments should be carried out, taking account of economic costs and benefits where possible.
- Climate change will have direct effects on rates of carbon-sequestration in soils, forests and regrowth vegetation which are vital to any greenhouse response strategy. Thus climate change influences should be factored in to carbon-sequestration modelling.
