ANNUAL SYMPOSIUM

Australia's science future 3-4 May 2000
Full listing of papers

Dr Leon Rotstayn graduated with honours in applied mathematics from Monash University in 1987, and then worked for a time in the private sector. In 1990, he joined CSIRO Atmospheric Research, where his early work focused on computational aspects of climate modelling, such as parallel processing of climate models. He received a PhD from Monash University in 1998 for studies on the treatment of clouds in climate models. His recent work has focused on aerosol-cloud interactions, as well as the treatment of clouds in climate models and the role of clouds in climate change.

Symposium themes - Changes in the global environment
Climate change and greenhouse
by Leon Rotstayn
leon.rotstayn@dar.csiro.au

Abstract
Is the Earth's climate changing? There is observational evidence that climate has changed; several factors may be contributing, including radiative forcing by anthropogenic greenhouse gases and aerosols, changes in solar output, and natural variability. Dr Rotstayn will discuss our present understanding of the relative importance of these different factors. Climate models are the most powerful tools that are available to integrate our understanding of the different physical processes that act in the climate system. He will describe these complex tools, their strengths and limitations, present some projections of future climate from climate models and discuss the uncertainties. He will conclude with some thoughts about where the science is going.

Recently the Melbourne Age had an article headed, 'Greenhouse effect leads to disasters'. Is it true? Or is it part of a plot by greenies and scientists who want more research funding?

If we look at the changes in globally averaged surface temperature from 1856 to 1999, it seems that something is happening to the climate (Figure 1). The ten warmest years are all in the 1980s and 1990s, and eight of the ten are in the 1990s. The warming trend from 1856 to 1999 is statistically significant at the 0.1 per cent level. (This means there is only a 0.1 per cent chance of getting this result due to ‘chance’.) So, most climatologists are now confident that the Earth is warming. There is also some evidence of more heavy rainfall events and changes in El Niño, but these are less clear. Scientists refer to this issue as the 'detection and attribution' problem. While we are now quite confident that climate change has been detected, it is more difficult to attribute the observed changes to particular causes.

Figure 1: Globally and annually averaged near-surface temperatures, shown as differences from the 1961-90 average (based on Parker, D. E., C. K. Folland and M. Jackson, 1995: Marine surface temperature: observed variations and data requirements, Climatic Change, 31, 559-600, and Jones, P. D., M. New, D. E. Parker, S. Martin and I. G. Rigor, 1999, Surface air temperature and its changes over the past 150 years, Rev. Geophys.
(37, 173-199.)

Possiblae causes of the observed changes are:

• the 'well-mixed' greenhouse gases;
• ozone (also a greenhouse gas);
• aerosols;
• solar variations (which are entirely natural).

Aerosols are small particles in the atmosphere, produced by both natural and anthropogenic (human) sources. Some important types of aerosols are:

• sulfates (from the burning of fossil fuels, volcanoes and marine biota);
• carbonaceous particles (from the burning of fossil fuels and biomass, as well as natural sources);
• sea salt particles;
• mineral dust.

Aerosols' main direct radiative effect is to create a haze that reflects solar radiation back to space, thus exerting a cooling effect on climate. (Some aerosols, such as soot aerosols, are highly absorbing of solar radiation, and are thought to exert a warming effect.) Less well understood is the indirect effect of aerosols, through their modification of cloud properties. This occurs because cloud droplets condense on these particles. In a polluted atmosphere, clouds therefore contain lots of small droplets, producing more persistent, brighter clouds than in an unpolluted atmosphere. These clouds reflect more solar radiation, making the ground cooler. This cooling effect of aerosols may (partially) offset the global warming effects of greenhouse gases. The indirect aerosol effect is currently a big uncertainty in climate change research.

To quantify the importance of the different possible causes of climate change, atmospheric scientists use the concept of 'radiative forcing'. Radiative forcing is a measure of the change in net downward radiation at the top of the atmosphere induced by the particular factor (such as aerosols or greenhouse gases). Radiative forcing gives a guide to the likely effect of each of these factors on the Earth's surface temperature. A positive radiative forcing represents an increase of radiation and hence a warming tendency. Greenhouse gases are thought to be the largest factor increasing the amount of radiation since pre-industrial times (Figure 2). Other smaller 'warming' factors are increases in tropospheric ozone, soot aerosols from fossil fuel burning, and solar variations. The largest factor decreasing the amount of radiation may be the indirect effect of aerosols, but this is very uncertain. Other smaller factors are stratospheric ozone depletion, and increases in sulfate aerosols and aerosols from biomass burning. In particular because of the large uncertainty in the indirect aerosol effect, we do not have a good sense of the overall change in radiative forcing since pre-industrial times. But the overall picture that scientists have tentatively arrived at is one of climate change that is driven by the warming effect of greenhouse gases, modulated by the other smaller warming or cooling effects.

Figure 2: Estimates of the globally and annually averaged radiative forcings in watts per square metre from 1850 to the present day (from Intergovernmental Panel on Climate Change (IPCC), 1996: Climate Change 1995: The Science of Climate Change, J.T. Houghton, L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg, and K. Maskell, Eds., Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 572 pp.)

Climate models can help unravel some of the complexities of atmospheric interactions. A climate model is a set of equations that simulates the atmosphere, ocean, sea-ice and land surface. In most models each grid square is at least 300 kilometres across, and processes that occur at a scale smaller than the grid may be important. More accurate representation of these 'sub-grid' processes is an ongoing aim of climate model development.

The strength of models is that they enable us to integrate our understanding of many complex processes into a whole that is more powerful than the individual parts. They allow us to do experiments that we might not want to try with the real Earth (such as to cut down the Amazon rainforest and see what will happen to the climate). They simulate the present climate surprisingly well. But models also have their limitations. Sub-grid physics is very simplified in some cases, due to a lack of understanding of some processes and limited computational power. There are also errors in climate simulations. Perhaps more importantly, different models give different responses to climate change. When eight models forced by changes in greenhouse gases and sulfate aerosols simulated temperatures from 1850 to 2100, their predicted surface warmings for 2100 varied from 2.0ºC to 4.7ºC. A better understanding of the processes and more computer power will produce better models in the future. These will have higher resolution and more realistic sub-grid physics, especially regarding the treatment of clouds. They will incorporate greenhouse gas emissions (rather than simply prescribing the atmospheric levels of these gases), vegetation that interacts with climate, and take account of the direct and indirect effects of aerosols in a more physically consistent way.

For another example of climate modelling, we can look at the results of a snow model that was driven by a range of climate scenarios given by climate models for Australia in the year 2070 (Figure 3). In the worst case scenario, the temperature is 3.4ºC warmer and there is 20 per cent less precipitation. The snow cover in south-eastern Australia then almost disappears (there is 96 per cent less snow). Even in the best case considered, there is 39 per cent less snow cover.

Figure 3: Simulated duration of snow cover in days per annum over the Australian Alps, for the present climate and for the worst case considered for 2070. (The method is explained by Whetton, P. H. and M. R. Haylock and R. Galloway, 1996: Climate Change and Snow-Cover Duration in the Australian Alps, Climatic Change, 32, 447-479.)

CSIRO has used results from a range of global climate models to assess likely changes over Australia in the coming decades. Here is CSIRO's projection for Australia's climate in 2030:

• the temperature will be between 0.3 and 1.4ºC warmer;
• there will be 5 to 30 per cent more days over 35ºC;
• there will be 10 to 60 per cent fewer frosts;
• it will be 0 to 10 per cent drier in winter;
• it will be 0 to 10 per cent drier in summer in the east and south-west;
• there will be more storms and heavy rain;
• sea level will rise by 5 to 25 centimetres.