DOUBLED CO2 PRECIPITATION CHANGES FOR THE SUSQUEHANNA BASIN - DOWN-SCALING FROM THE GENESIS GENERAL-CIRCULATION MODEL

Artificial neural nets are used in an empirical down-scaling procedure to derive daily subgrid-scale precipitation from general circulation model (GCM) geopotential height and specific humidity data. The neural net-based transfer functions are developed using a 2 degrees x 2.5 de grees gridded data assimilation product from the Goddard Space Flight Center, applied to a 4 x 4 matrix of grid-cells centred on the Susqueh anna river basin. The down-scaled precipitation is a close match to th e observed data (temporal correlations at individual grid-points range from 0.6 to 0.84). Doubled CO2 climate change scenarios are produced by applying the same transfer functions to the geopotential height and specific humidity fields from 1 x CO2 and 2 x CO2 simulations of vers ion II of the GENESIS climate model. The analysis indicates a 32 per c ent increase in spring and summer rainfall over the basin, resulting f rom changes in both moisture availability and the orientation of the s torm track over the region. The down-scaled precipitation increases, d erived from the change in the GCM's circulation and humidity fields, a re considerably larger than the change in the model's actual computed precipitation. (C) 1998 Royal Meteorological Society.