1996 Climate change scenarios

Comparison of this scenario with the WWF(1999) scenario

INTRODUCTION

Current scientific understanding predicts with confidence global warming in the next century. However, uncertainty surrounding future greenhouse gas and sulfate emissions, shortcomings in climate modelling, and difficulties in determining regional patterns of climate change from global estimates mean that predictions of future climate change at a regional level still cannot be made. Therefore, this is not a forecast.

This document is based on the best available science as of October 1996 and replaces the statement released by CSIRO in November 1992 (CIG, 1992). It is a highly simplified summary for general information only. More detailed climate change information is available from the Climate Impact Group for specific purposes including climate impact research.

In constructing climate change scenarios there are many sources of uncertainty. Where possible they have been taken into account. Thus the scenarios represent a multiplicity of plausible futures. However, changes outside the ranges given here cannot be ruled out, nor can confidence levels be reliably quantified.

The scenarios differ from those previously prepared by the Climate Impact Group in the following respects:

• The global warming scenarios of the Intergovernmental Panel on Climate Change (IPCC, 1996) are used;
• Some allowance has been made for the cooling effect of sulfate aerosol pollution;
• The results of fully coupled ocean-atmosphere climate models are now used, although results from models with a simplified slab ocean are retained for the present in relation to rainfall.

All changes to climate given in this summary are relative to 1990 levels.

GLOBAL GREENHOUSE GAS AND SULFATE AEROSOL EMISSIONS

The IPCC identifies a broad range of plausible future greenhouse gas and sulfate aerosol emissions in the absence of emission policies beyond those already adopted (IPCC, 1992). Figure 1 shows six IPCC scenarios for carbon dioxide (CO₂) emissions (IS92a-f). For other greenhouse gases and aerosols, the IS92a-f scenarios contain a similar range of future emissions.

Figure 1 IPCC scenarios for CO₂ emissions

The IPCC scenarios assume a strong link between CO₂ emissions and sulfate aerosol emissions because the burning of fossil fuel is a major source of both. Therefore most scenarios describe increasing emissions of both CO₂ and aerosols. However, because of the increasing use of sulfate emission reduction technology, the future strength of this link is uncertain.

GLOBAL WARMING

All global climate models (GCMs) show warming in response to increased greenhouse gas concentrations, but the sensitivity of models varies considerably. In particular, if the atmospheric CO₂ concentration is instantaneously doubled and the climate allowed to come to a new equilibrium, GCMs suggest that the resultant global mean surface air temperature increase would be between 1.5 and 4.5°C. This warming value is known as the climate sensitivity. However, this describes a highly idealised situation. In reality, CO₂ is increasing continuously, and is not the only greenhouse gas that needs to be considered. Furthermore, because of the large capacity of the oceans to absorb heat, changing greenhouse gas concentrations and global climate would not be in equilibrium.

New scenarios of global warming for 1990-2100 that take into account the range of estimates of climate sensitivity and the range of emission scenarios were presented in the latest IPCC assessment (IPCC, 1996). The simple model of Wigley and Raper (1992) which allowed for absorption of heat by the oceans was used. The calculations included an estimate of the cooling effect of increasing sulfate aerosol concentration. Figure 2 shows the resulting IPCC future global warming scenarios. High, mid and low scenarios are given.

The assumptions upon which each of these are based are as follows:

Figure 2 Low, mid and high case scenarios of global warming (IPCC, 1996).

The global warmings for each of these scenarios for years 2030 and 2070 are given in Table 1.

Table 1 Global warming scenarios (°C) for 2030 and 2070 (IPCC, 1996).

IPCC also presented global warming calculations for sulfate aerosol concentrations fixed at 1990 levels but with the greenhouse gas concentrations still varying according to the IS92 scenarios. By thus excluding the cooling effect of increasing aerosols, the warming scenarios are increased by twenty to thirty percent in the high and mid cases. This demonstrates the sensitivity of global warming to uncertainty in projecting sulfate aerosols. However, regional warming in the southern hemisphere is less affected by sulfate aerosol emissions because these are lower than in the northern hemisphere.

IPCC also calculated global mean sea-level rise scenarios. For example, the range of sea-level rise for 2070 is 9-59 cm assuming varying aerosols. A regional perspective on mean sea-level rise is the subject of further research.

AUSTRALIA’S CLIMATE

Temperature change

Methodology

The scenarios of regional changes in temperature are based on:

• The global warming scenarios outlined above, which provide information on the magnitude of the global climate response over time.
• The regional pattern of the temperature response taken from five GCM experiments forced by changes in greenhouse gases only. The GCM results are used to produce ranges of local temperature change per degree of global warming. The ranges reflect the fact that for a given level of global warming some models show stronger regional warming than others.
• An adjustment to the regional response patterns to allow for how sulfate aerosols affect the Australian region relative to the global average effect. This is based on the results of a single GCM experiment which considers increasing sulfate aerosols as well as greenhouse gases.

The regional climate change information used here comes from coupled atmosphere-ocean GCMs. Coupled models employ general circulation models of the full ocean whereas earlier models, such as those used for earlier CSIRO scenarios (CIG, 1992 and Whetton et al., 1996a) represent only the surface layer of the ocean (and are known as ‘slab ocean’ models).

The pattern of cooling is quite sensitive to the regional and temporal pattern of assumed emissions and the way the climatic effect of aerosols is represented in models. Our estimate of the regional effect of aerosols has been based on the results from the only available GCM simulation. The uncertainty in projecting future global aerosols emissions was discussed earlier.

Ranges of temperature change

Table 2 gives scenarios of temperature change for Australia. Low and high scenarios for local warming per degree of global warming are given for three regions. These ranges mainly represent differences amongst GCMs and seasonal variation (current simulations produce slightly less warming in winter). Using these values and the global warming information in Table 1, we then calculate the low and high case warmings for 2030 and 2070. The scenarios for these years incorporate additional uncertainties associated with estimating global warming, namely the range of climate sensitivities, and the range of plausible future emission scenarios. Mid-case warmings may also be calculated but are not presented here.

Table 2 Scenarios of temperature change (°C) for locations in the Australian region.

Models suggest that increases in temperature are likely to be similar for daily maximum and minimum temperature except where there are also changes in rainfall and cloudiness. Wetter, cloudier conditions will lead to a greater increase in minimum temperature, whereas drier, clearer conditions will produce a greater increase in maximum temperature.

Temperature variability and extremes

There is little consistency between models regarding the direction of change in daily temperature variability. However, a significant increase in mean temperature will imply a marked decrease in the frequency of extremely low temperatures and a similar increase in the frequency of extremely high temperatures. For a given scenario and site, this effect may be quantified given certain assumptions (e.g., Hennessy and Pittock, 1995). For example, under a 2°C warming the number of days over 35°C in Canberra increases from 4 to 10 days per summer, and the number of days less than 0°C in Ballarat decreases from 9 to 2 days per winter.

Precipitation change

Methodology

As for temperature, the rainfall change scenarios are based on:

• Global warming scenarios as outlined above, which provide information on the magnitude of the climate response as it varies over time.
• The regional pattern of percentage precipitation response taken from a range of GCM experiments which are forced by changes in greenhouse gases only. The GCM data are used to produce ranges of local precipitation change per degree of global warming.

However, because of large uncertainty in its estimation, no adjustment is made for the effect on regional precipitation of sulfate aerosols.

A set of scenarios of precipitation change is shown using the five coupled experiments used in preparing the temperature scenarios. However, an additional set of scenarios are shown using five experiments with slab ocean GCMs (which represent only the surface layer of the ocean).

There is good reason in principle for preferring the results of coupled models. However, for the following reasons, we are not yet ready to dismiss scenarios based on slab models:

• The difference between the rainfall changes simulated by the two classes of GCMs is greater in Australia than it is elsewhere in the world. Over Australia, coupled models tend to simulate summer rainfall decreases and slab models summer rainfall increases.
• The rainfall change in coupled models differs from that in slab models due to a strong delay in warming in the higher latitudes of the southern hemisphere in coupled models. There is considerable uncertainty regarding the reliability of the processes in ocean models that lead to this result (e.g. Whetton et al., 1996b).
• Some aspects of the pattern of warming in the southern hemisphere simulated by coupled models conflict with observed 20th century trends (although the latter may not be due to the enhanced greenhouse effect; Pittock et al., 1996).

Future rainfall over Australia may also be affected by the following:

• Local changes in ocean circulation;
• Changes in large-scale atmospheric circulation due to increasing sulfate aerosols in Asia;
• Change in El Niño - Southern Oscillation (ENSO) behaviour.

Deficiencies in the current state of the science mean that the scenarios cannot make explicit allowance for these factors (see below for more on ENSO).

Interannual and decadal scale climatic variability will continue in the future and will remain a source of uncertainty in projecting future climate.

Ranges of change

Regional low and high case scenarios of rainfall change in summer and winter are given in Table 3 for the coupled models. The corresponding information for slab models is given in Table 4. Changes in percent per degree of global warming are given as well as the percentage changes in 2030 and 2070. In winter, regions that have insignificant rainfall will remain dry. In general, rainfall changes in spring and autumn are transitional between the summer and winter patterns of change.

Table 3 Scenarios of precipitation change for locations in the Australian region based on coupled models

Winter

Summer

Table 4 Scenarios of precipitation change for locations in the Australian region based on slab models

Winter

Summer

Changes given in the tables apply to broad areas. Significantly larger or smaller changes would apply at the local scale, particularly in locations where topography strongly controls rainfall patterns.

In winter, the scenarios for rainfall change are broadly similar in the coupled and slab models. Both show a pattern of rainfall decreases over most of Australia, increases over the oceans to the south, and a region of uncertain change in between. In the coupled models the area of decrease is stronger and more extensive and the other regimes are located a little further south.

In summer, rainfall increases predominate in slab models whereas decreases predominate in coupled models. This reflects a major systematic difference in regional response of coupled and slab models, and demonstrates the sensitivity of simulated rainfall change over Australia to oceanic processes. This large uncertainty in summer rainfall change highlights the need to consider more than one climate change scenario.

In general, the regions used for a given season and model type were chosen so as to highlight those areas where most models (at least 4 out of 5) show the same direction of rainfall change. Areas where the direction of change differs amongst models are thus also highlighted (such as the transitional region B in the winter scenarios). Region boundaries are indicative only and do not reflect topographical barriers unresolved by the global models. Notably, in a high resolution regional climate modelling experiment for southeastern Australia, the southern boundary of the zone of winter rainfall decrease lies along the Great Dividing Range in Victoria.

Rainfall intensity

Where models simulate an increase in average rainfall, this is associated with an increase in daily rainfall intensity leading to more frequent or heavier rainfall events. This tendency is less marked or absent where reduced rainfall is simulated.

Tropical cyclones

These important but small-scale events are difficult to model and continue to be a major research priority. Present indications are:

• Region of origin is likely to remain unchanged;
  Possible modest increase in intensities in some parts of the globe - poleward extent still uncertain;
  Preferred paths may alter;
  Location and frequency is affected by ENSO.

ENSO

• New observational studies have given us a fuller picture of the variation in ENSO behaviour in the past;
• Coupled models can now crudely simulate ENSO-related climatic variability;
• These advances have not provided compelling evidence that the frequency or intensity of ENSO phases will change beyond that seen in the past.

Soil moisture and runoff

Levels of soil moisture and runoff can be very sensitive to changes in precipitation and evapotranspiration. Increased temperature will increase evaporation, which will reduce soil moisture and runoff except where there is a compensating increase in precipitation. However important uncertainties remain in quantifying the hydrological response for a given climate change scenario.

IMPLICATIONS

This document is not intended to canvas in any detail the potential implications of these scenarios. However, it should be appreciated that climatic changes of the magnitude envisaged here, even by 2030, could have practical implications. For example, any significant change in soil moisture and runoff would be important for water resources, agriculture and biodiversity. Other important impact areas may include health, pests and diseases, and fire occurrence (Hennessy et al., 1995; Bouma et al., 1996). The direct fertilising effect of CO₂ will also need to be considered in assessing impacts in some areas.

FURTHER INFORMATION

This document is intended to provide a brief summary of the latest science behind regional climate scenarios for Australia. Those who wish to use regional scenarios or climate model output for research on climate impacts and adaptations are encouraged to contact the climate impacts liaison officer at CSIRO Division of Atmospheric Research who can provide more detailed data.

The Climate Impacts Group has several other initiatives supporting impacts research. The first is OzClim, a regional scenario generator and impacts software package for Australia. Data from a number of global and regional climate models are also available. Included are a wider range of variables than summarised in this document as well as high-resolution data on a 125 km grid for Australia.

Contact information is given at the end of this document.

REFERENCES AND FURTHER READING

Bates, B.C., Jakeman, A.J., Charles, S.P., Sumner, S.R., and Fleming, P.M., 1996: Impact of climate change on Australia’s surface water resources. In Greenhouse: Coping with Climate Change, Bouma, W.J., Pearman, G.I., and Manning, M.R. (eds), CSIRO Publishing, Melbourne, pp. 248-262.

Bouma, W.J., Pearman, G.I., and Manning, M.R. (eds), 1996: Greenhouse: Coping with Climate Change. CSIRO Publishing, Melbourne, 682pp.

Chiew, F.H.S., Whetton, P.H., McMahon T.A. and Pittock A.B., 1994: Simulation of the impacts of climate change on runoff and soil moisture in Australian catchments, J. of Hydrology, 167, 121-147.

CIG, 1992: Climate Change Scenarios for the Australian Region. Climate Impact Group, CSIRO Division of Atmospheric Research, Melbourne, 6pp.

Hennessy, K.J. and Pittock, A.B., 1995: Greenhouse warming and threshold temperature events in Victoria, Australia, Int. J. Climatology, 15, 591-612.

Hennessy, K.J., Whetton, P.H. and Pittock, A.B. 1995: CSIRO Climate Change Research Program: Collaboration in Scenario Development and Impact Projects 1990-1995. CSIRO Division of Atmospheric Research Report, 51 pp.

Hennessy, K.J. and Pittock, A.B. 1996: Development and Application of Climate Change Scenarios: Climate Impacts Assessment Workshop Report. CSIRO report for the Australian Department of Environment, Sport and Territories, 47 pp.

IPCC, 1992: Climate Change 1992: The Supplementary Report to the IPCC Scientific Assessment, Houghton, J. T., Callander, B. A., and Varney, S. K. (eds), Working Group 1. Cambridge University Press, Cambridge.

IPCC, 1996: Climate Change 1995: The Science of Climate Change. Houghton, J.T., Meira Filho, L.G., Callander, B.A., Harris, N., Kattenberg, A. and Maskell, K. (eds), Contribution of Working Group 1 to the second assessment report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge.

Pittock, A.B., Dix, M.R., Hennessy, K. J., Jackett, D.R., Katzfey, J.J., McDougall, T.J., McInnes, K.L., O'Farrell, S.P., Smith, I.N., Suppiah, R., Walsh, K.J., Whetton, P.H. and Wilson, S.G., 1996: Progress towards climate change scenarios for the southwest Pacific, Weather and Climate, 15-2, 21-46.

Schreider, S.Yu., Jakeman, A.J., Pittock, A.B. and Whetton, P.H., in press: Estimation of possible climate change impacts on water availability, extreme flow events and soil moisture in the Goulburn and Ovens basins, Victoria, Climatic Change.

Whetton, P., Mullan, A.B., and Pittock, A.B., 1996a: Climate change scenarios for Australia and New Zealand. In Greenhouse: Coping with Climate Change, Bouma, W.J., Pearman, G.I., and Manning, M.R. (eds), CSIRO Publishing, Melbourne, pp. 145-168.

Whetton, P., England, M., O’Farrell, S., Watterson, I., and Pittock, A.B., 1996b: Global comparison of the regional rainfall results of enhanced greenhouse coupled and mixed layer ocean experiments: Implications for climate change scenario development, Climatic Change, 33, 497-519.

Wigley, T.M.L. and Raper, S.C.B., 1992: Implications of revised IPCC scenarios, Nature, 357, 293-300.

ACKNOWLEDGEMENTS

This background statement is based on research undertaken as part of the CSIRO Climate Change Research Program with funding by CSIRO, the Commonwealth Department of the Environment Sport and Territories and the governments of New South Wales, Northern Territory, Queensland, Victoria and Western Australia.

The GCM data used here were generously provided by climate modellers at CSIRO, Bureau of Meteorology Research Centre, Canadian Climate Centre, Deutsches Klimarechenzentrum, Geophysical Fluid Dynamics Laboratory and Hadley Centre. Some GCM data were supplied by the Climate Impacts LINK Project (UK Department of the Environment Contract EPG 1/1/16) on behalf of the Hadley Centre and the U.K. Meteorological Office. IPCC global warming and emissions data were provided by Prof. Tom Wigley.

Contact:
Climate Impact Liaison Officer
Roger Jones
CSIRO Division of Atmospheric Research
PB No. 1, Aspendale, Victoria, 3195
Ph: 03 9239 4555
Fax: 03 9239 4444
email: roger.jones@csiro.au

World Wide Web:
Model output
OzClim

Modified: April 3, 2008