Current atmosphere-ocean general circulation model (AOGCM) simulations have limited ability to suggest likely changes in tropical cyclone activity; until high-resolution GCM simulations are available, it is not possible to say whether the frequency, area of occurrence, time of occurrence, mean intensity, or maximum intensity of tropical cyclones will change in Tropical Asia. However, more recent studies with models of higher resolution appear to be able to simulate tropical cyclone climatology (Bengtsson et al., 1995). Bengtsson et al. (1995) found a decrease in the number of tropical cyclones under enhanced greenhouse scenarios, especially in the Southern Hemisphere, although their geographical distribution remained unchanged.
A study by Holland (1997) indicated that tropical cyclones are unlikely to be more intense than the worst storms experienced under present-day climate conditions-although some potential exists for changes in cyclone intensity in tropical oceanic regions, where sea-surface temperature (SST) is betwen 26�C and 29�C. According to Henderson-Sellars and Zhang (1997), recent studies indicate that the maximum potential intensities of cyclones will remain the same or undergo a modest increase of up to 10-20%; they add that these predicted changes are small compared with observed natural variations and fall within the uncertainty range in current studies. Lal et al. (1995a), in their ECHAm3-T106 experiment, found no significant change in the number and intensity of monsoon depressions in the Indian Ocean in a warmer climate. Similarly, likely changes in the ENSO phenomenon under enhanced greenhouse conditions and their possible impact on the interannual variability of the summer monsoon are not known.
The spatial patterns of increases in surface temperature during winter (DJF) and summer (JJA), simulated by the more recent coupled AOGCMs, tend to confirm, to a large extent, the projections of the Climate Impact Group (1992) for the monsoon Asian region. Area-averaged increases in temperature for doubled CO2 conditions are, however, lower by 15-20% in the transient experiments (using AOGCMs and based on a 1% increase in CO2) than in equilibrium experiments (using atmospheric GCMs and based on an instantaneous doubling of CO2 and a mixed-layer ocean). The area-averaged increase in summer monsoon rainfall and its spatial distribution are highly variable among GCMs, although all models produce an increase in monsoon rainfall.
The response of the monsoon climate to transient increases in GHGs and sulfate aerosols in the Earth's atmosphere has recently been examined by Lal et al. (1995b), using data generated by the MPI-ECHAm3 atmospheric model, coupled to a large-scale geostrophic ocean model (ECHAm3 + LSG). The authors identified the potential role of sulfate aerosols in obscuring GHG-induced warming over the Indian subcontinent during the past century and found that year-to-year variability in simulated monsoon rainfall for the past century is in fair agreement with observed climatology.
Lal et al. (1995b) then presented a scenario of climate change for the Indian subcontinent for the middle of the next century, taking into account projected emissions of GHGs and sulfate aerosols. They suggested an increase in annual mean maximum and minimum surface air temperatures of 0.7�C and 1.0�C over the land regions in the decade of the 2040s with respect to the 1980s. This warming would be less pronounced during the monsoon season than in the winter months. A significant decrease in the winter diurnal temperature range, with no appreciable change during the monsoon season, also is projected (Lal et al., 1996). Moreover, projected warming of the land region of the Indian subcontinent is likely to be relatively lower in magnitude than that of the adjoining ocean, resulting in a decline in the land-sea thermal contrast-the primary factor responsible for the onset of summer monsoon circulation.
As a consequence-and contrary to simulations that consider only CO2 forcing-Lal et al. (1995b) found a decline in mean summer monsoon rainfall of about 0.5 mm/day over the region, which is marginally above the range of interannual variability for the present-day atmosphere. This decline in summer monsoon rainfall resulting from combined GHG and aerosol forcing in the southeast Asian region also has been suggested in the UKMO GCM experiments (see Figure B-2).
Despite the great strides that have been made in objective modeling of climate and climate change through AOGCMs, there still are uncertainties associated with model projections, especially for climate change on regional scales. The problem is even more complicated when researchers attempt to project time-dependent regional climatic responses to future increases in radiative forcing from anthropogenic GHGs and aerosols. Because the anthropogenically induced sulfate aerosol burden has large spatial and temporal variations in the atmosphere, its regional-scale impacts could be in striking contrast to the impacts of GHGs-the concentrations of which are likely, in most cases, to change uniformly throughout the globe. Spatially localized radiative forcing resulting from anthropogenic aerosols is confined largely to the Northern Hemisphere and tends to yield a steepening of normalized meridional temperature gradient in that hemisphere. The effects of aerosols also yield distinct precipitation responses in the tropical region in general and over the Asian monsoon region in particular. The implications of enhanced aerosol loading on tropospheric clouds, which could strongly modulate the monsoon climate, still are not clear. Because anthropogenic sulfate aerosol loadings are projected to be substantial over the southeast Asian region, their impact on the Asian summer monsoon needs to be more carefully examined. Their precise magnitude, as well as the roles of other localized potential forcings (e.g., aerosols from biomass burning, increases in tropospheric ozone), also must be known before confident predictions of regional changes in the monsoon climate and its variability can be made.
Other reports in this collection