Climate Change 2001:
Working Group I: The Scientific Basis
Other reports in this collection

10.3.2.2 Climate variability and extreme events

Gregory and Mitchell (1995) identified in an equilibrium 2xCO2 simulation with the Hadley Centre model a tendency for daily temperature variability over Europe to increase in JJA and to decrease in DJF. Subsequent work on temperature variability at daily to monthly and seasonal time-scales has tended to confirm this pattern, as found by Buishand and Beersma (1996) over Europe, Beersma and Buishand (1999) over southern Europe, northern Europe and central North America and Boer et al. (2000b) throughout the northern mid-latitudes. This tendency can also be seen in the results of Giorgi and Francisco (2000a) for a set of transient HadCM2 simulations over different regions of the globe.

Daily high temperature extremes are likely to increase in frequency as a function of the increase in mean temperature, but this increase is modified by changes in daily variability of temperature. There is a corresponding decrease in the frequency of daily low temperature extremes. Kharin and Zwiers (2000) and Zwiers and Kharin (1997) found that in all regions of the globe the CGCM1 model simulated substantial increases in the magnitude of extreme daily maximum and minimum temperatures, with an average frequency of occurrence of once per twenty years. Delworth et al. (1999) considered simulated changes of a ‘heat index' (a measure which combines the effect temperature and moisture) in the GFDL R15a model. Their results indicated that seasonally warm and humid areas such as the south-eastern United States, India, Southeast Asia and northern Australia can experience increases in the heat index substantially greater than that expected due to warming alone.

There is a strong correlation between precipitation inter-annual variability and mean precipitation. Increases in mean precipitation are likely to be associated with increases in variability, and precipitation variability is likely to decrease in areas of reduced mean precipitation. In general, where simulated changes in regional precipitation variability have been examined, increases are more commonly noted. Giorgi and Francisco (2000a) found a tendency for regional interannual variability of seasonal mean precipitation to increase in HadCM2 simulations in many of the regions they considered. Increases in interannual variability also predominated in the CGCM1 simulation (Boer et al., 2000b) although there were areas of decrease, particularly in areas where mean rainfall decreased. Beersma and Buishand (1999) mostly found increases in monthly precipitation variance over southern Europe, northern Europe and central North America. A number of studies have reported a tendency for interannual rainfall variability to increase over South Asia (SAR; Lal et al., 2000). McGuffie et al. (1999) identified a tendency for increased daily rainfall variability in two models over the Sahel, North America, South Asia, southern Europe and Australia. It should also be noted that in many regions interannual climatic variability is strongly related to ENSO, and thus will be affected by changes in ENSO behaviour (see Chapter 9).

The tendency for increased rainfall variability in enhanced GHG simulations is reflected in a tendency for increases in the intensity and frequency of extreme heavy rainfall events. Such increases have been documented in regionally focused studies for Europe, North America, South Asia, the Sahel, southern Africa, Australia and the South Pacific (Hennessy et al., 1997; Bhaskaran and Mitchell, 1998; McGuffie et al 1999; Jones, R.N. et al., 2000) as well as in the global studies of Kharin and Zwiers (2000) and Zwiers and Kharin (1998). For example, Hennessy et al. (1997) found that under 2xCO2 conditions the one-year return period events in Europe, Australia, India and the USA increased in intensity by 10 to 25% in two models.

Changes in the occurrence of dry spells or droughts have been assessed for some regions using recent model results. Joubert et al. (1996) examined drought occurrence over southern Africa in an equilibrium 2xCO2 CSIRO simulation and noted areas of both substantial increase and decrease. Gregory et al. (1997) looked at drought occurrence over Europe and North America in a transient simulation using both rainfall-based and soil moisture-based measures of drought. In all cases, marked increases were obtained. This was attributed primarily to a reduction in the number of rainfall events rather than a reduction in mean rainfall. Marked increases in the frequency and intensity of drought were found also by Kothavala (1997) over Australia using the Palmer drought severity index.

Fewer studies have considered changes in variability and extremes of synoptic circulation under enhanced GHG conditions. Huth (1997) noted little change in synoptic circulation variability under equilibrium 2xCO2 conditions over North America and Europe. Katzfey and McInnes (1996) found that the intense cut-off lows off the Australian east coast became less common under equilibrium 2xCO2 conditions in the CSIRO model, although they had limited confidence in this result.

10.3.3 Summary and Recommendations

Analysis of transient simulations with AOGCMs indicates that average climatic features are generally well simulated at the planetary and continental scale. At the regional scale, area-average biases in the simulation of present day climate are highly variable from region to region and across models. Seasonal temperature biases are typically within the range of ±4°C but exceed ±5°C in some regions, particularly in DJF. Precipitation biases are mostly between -40 and +80%, but exceed 100% in some regions. These regional biases are, in general terms, smaller than those of a similar analysis presented in the SAR. When it has been assessed, many aspects of model variability have compared well against observations, although significant model-dependent biases have been noted. Model performance was poorer at the finer scales, particularly in areas of strong topographical variation. This highlights the need for finer resolution regionalisation techniques.

Simulated changes in mean climatic conditions for the last decades of the 21st century (compared to present day climate) vary substantially among models and among regions. All land regions undergo warming in all seasons, with the warming being generally more pronounced over cold climate regions and seasons. Average precipitation increases over most regions, especially in the cold season, due to an intensified hydrological cycle. However, some exceptions occur in which most models concur in simulating decreases in precipitation. The magnitude of regional precipitation change varies considerably among models with the typical range being around 0 to 50%, where the direction of change is strongly indicated, and around -30 to +30% where it is not. There is strong tendency for models to simulate regional increases in precipitation variability with associated increases in the frequency of extreme rainfall events. Increased interannual precipitation variability is also commonly simulated and, in some regions, increases in drought or dry-spell occurrence have been noted. Daily to inter-annual variability of temperature is simulated to decrease in winter and increase in summer in mid-latitude Northern Hemisphere land areas.



Other reports in this collection