Climate Change 2001: Working Group I: The Scientific Basis

	Other reports in this collection

10.5.1.2 Mean climate: Simulations using GCM boundary conditions

Since the SAR, evaluation of RCMs driven by GCM simulations of current climate has gained much attention (Appendix 10.2), as this is the context in which many RCMs are used (e.g., for climate change experiments). Errors introduced by the GCM representation of large-scale circulations are transmitted to the RCM as, for example, clearly shown by Noguer et al. (1998). However, since the SAR, regional biases of seasonal surface air temperature and precipitation have been reduced and are mostly within 2°C, and 50 to 60% of observations (with exceptions in all seasons), respectively (Giorgi and Marinucci, 1996b; Noguer et al., 1998; Jones et al., 1999 for Europe; Giorgi et al., 1998 for the continental USA; McGregor et al., 1998 for Southeast Asia; Kato et al., 2001 for East Asia). The reduction of biases is due to both better large-scale boundary condition fields and improved aspects of internal physics and dynamics in the RCMs.
The regionally averaged biases in the nested RCMs are not necessarily smaller than those in the driving GCMs. However, all the experiments mentioned above, along with those of Leung et al. (1999a,b), Laprise et al. (1998), Christensen et al. (1998) and Machenhauer et al. (1998) clearly show that the spatial patterns produced by the nested RCMs are in better agreement with observations because of the better representation of high-resolution topographical forcings and improved land/sea contrasts. For example, in simulations over Europe and central USA, Giorgi and Marinucci (1996a) and Giorgi et al. (1998) find correlation coefficients between simulated and observed seasonally averaged precipitation in the range of +0.53 to +0.87 in a nested RCM and –0.69 to +0.85 in the corresponding driving GCM.

The role of the high-resolution forcing was clearly demonstrated in the study of Noguer et al. (1998), which showed that the skill in simulating the mesoscale component of the climate signal (Giorgi et al., 1994; Jones et al., 1995) was little sensitive to the quality of the driving data (Noguer et al., 1998). On the other hand, interactions between the large-scale driving data and high resolution RCM forcings can have negative effects. In simulations over the European region of Machenauer et al. (1998), the increased shelter due to the better-resolved mountains in the RCMs caused an intensification of the GCM-simulated excessively dry and warm summer conditions over south-eastern Europe.

Horizontal resolution is especially important for the simulation of the hydrologic cycle. Christensen et al. (1998) showed that only at a very high resolution do the mountain chains in Norway and Sweden become sufficiently well resolved to yield a realistic simulation of the surface hydrology (Figure 10.10). An alternative strategy is to utilise a sub-grid scale scheme capable of resolving complex topographical features (Leung et al., 1999a).