ABSTRACT
The simulations of ecosystems and their services require weather and climate data. There are two available sources for these data: (1) meteorological stations, and (2) general circulation models (GCMs). However, sparsely distributed meteorological stations are often unable to satisfy the data requirements of such studies. Most GCM simulations use coarse resolution with horizontal resolution being about 200-500 km (Xue et al., 2007). It is difficult to use GCMs to assess climate change impacts on various ecosystems on regional and local levels because of their coarse spatial resolution, although CGMs can provide a good overview of both current and future climates on a global level. Grotch and MacCracken (1991) found that the range of changes in temperature and precipitation predicted by different computer models is much broader at finer spatial scales; many shortcomings were apparent in the model simulations of the present climate. Von Storch et al. (1993) observed that simulations of GCMs were questionable on a regional level. Ciret and Sellers (1998) stated that the accuracy of regional climate predictions was linked to the spatial resolution at which GCMs operated; increasing the spatial resolution of GCMs would improve the simulation of climate, and hence, increase confidence in the use of GCM output for impact studies. Covey et al. (2003) argued that it was difficult to determine whether or not the models were good enough to be trusted when used to study climate either in the distant past or for future predictions. Raisanen (2007) concluded that many small-scale processes could not be simulated explicitly in current climate models. Prudhomme and Davies (2009) indicated that GCMs often resulted in different climate outputs from the same atmospheric and oceanic drivers, especially at regional scales.
