Increased intensity of precipitation events in a future climate with increased greenhouse gases was one of the earliest model results regarding precipitation extremes, and remains a consistent result in a number of regions with improved, more detailed models (Hennessy et al., 1997; Kothavala, 1997; Durman et al., 2001; Yonetani and Gordon, 2001). There have been questions regarding the relatively coarse spatial scale resolution in climate models being able to represent essentially mesoscale and smaller precipitation processes. However, the increase in the ability of the atmosphere to hold more moisture, as well as associated increased radiative cooling of the upper troposphere that contributes to destabilisation of the atmosphere in some models, is physically consistent with increases in precipitation and, potentially, with increases in precipitation rate.
As with other changes, it is recognised that changes in precipitation intensity have a geographical dependence. For example, Bhaskharan and Mitchell (1998) note that the range of precipitation intensity over the south Asian monsoon region broadens in a future climate experiment with increased greenhouse gases, with decreases prevalent in the west and increases more widespread in the east (see further discussion in Chapter 10). Another model experiment (Brinkop, 2001) shows that extreme values of the convective rain rate and the maximum convective height occur more frequently during the 2071 to 2080 period than during the 1981 to 1990 period. The frequency of highest-reaching convective events increases, and the same holds for events with low cloud-top heights. In contrast, the frequency of events with moderate-top heights decreases. On days when it rains, the frequency of the daily rates of convective rainfall larger than 40 mm/day in JJA and greater than 50 mm/day for DJF, increases. Generally, one finds a strong increase in the rain rate per convective event over most of the land areas on the summer hemispheres and in the inter-tropical convergence zone (ITCZ). Between 10 and 30°S there are decreases in rain rate per event over the ocean and parts of the continents.
In global simulations for future climate, the percentage increase in extreme (high) rainfall is greater than the percentage increase in mean rainfall (Kharin and Zwiers, 2000). The return period of extreme precipitation events is shortened almost everywhere (Zwiers and Kharin, 1998). For example, they show that over North America the 20-year return periods are reduced by a factor of 2 indicating that extreme precipitation of that order occurs twice as often.
Another long-standing model result related to drought (a reduction in soil moisture and general drying of the mid-continental areas during summer with increasing CO2) has been reproduced with the latest generation of global coupled climate models (Gregory et al., 1997; Haywood et al., 1997; Kothavala, 1999; Wetherald and Manabe, 1999). This summer drying is generally ascribed to a combination of increased temperature and potential evaporation not being balanced by precipitation. To address this problem more quantitatively, a global climate model with increased CO2 was analysed to show large increases in frequency of low summer precipitation, the probability of dry soil, and the occurrence of long dry spells (Gregory et al., 1997). The latter was ascribed to the reduction of rainfall events in the model rather than to decreases in mean precipitation. However, the magnitude of this summer drying response may be related to the model's simulation of net solar radiation at the surface, and more accurate simulation of surface fluxes over land will increase confidence in the GCM climate changes.
Alhough of great importance to society for their potential for causing destruction, as well as their human and economic impacts, there is little guidance from AOGCMs concerning the future behaviour of tornadoes, hail or lightning. This is because these phenomena are not explicitly resolved in AOGCMs, and any studies that have been done have had to rely on empirical relationships between model features and the phenomenon of interest. For example, Price and Rind (1994a) derive a relationship between lightning activity and convective cloud-top height to infer an increase of lightning with increasing CO2. They take that relationship one step further to suggest a future increase in lightning-caused fires due to the increased lightning activity and decreased effective precipitation (Price and Rind, 1994b). Using another empirical relationship between daily minimum temperature and severe convective storm frequency for France, Dessens (1995) connects an increase in daily minimum temperature with greater convective storm frequency and more hail damage in a future climate with increased CO2. However, there have been no recent studies examining this problem with the current generation of global climate models. Due to the fact that these severe weather phenomena are sub-grid scale (even more so than discussed below for tropical cyclones), and that second and third order linkages between model output and empirical relationships for limited regions must be used to derive results, we cannot reach any definitive conclusions concerning possible future increases in hail and lightning, and there is no information from AOGCMs concerning future changes in tornado activity.
Other reports in this collection