Validation of River Flows in HadGEM1 and HadCM3 with the TRIP River Flow Model

Abstract

Abstract The Total Runoff Integrating Pathways (TRIP) global river-routing scheme in the third climate configuration of the Met Office Unified Model (HadCM3) and the newer Hadley Centre Global Environmental Model version 1 (HadGEM1) general circulation models (GCMs) have been validated against long-term average measured river discharge data from 40 stations on 24 major river basins from the Global Runoff Data Centre (GRDC). TRIP was driven by runoff produced directly by the two GCMs in order to assess both the skill of river flows produced within GCMs in general and to test this as a method for validating large-scale hydrology in GCMs. TRIP predictions of long-term-averaged annual discharge were improved at 28 out of 40 gauging stations on 24 of the world’s major rivers in HadGEM1 compared to HadCM3, particularly for low- and high-latitude basins, with predictions ranging from “good” (within 20% of observed values) to “poor” (biases exceeding 50%). For most regions, the modeled annual average river flows tended to be exaggerated in both models, largely reflecting inflated estimates of precipitation, although lack of human interventions in this modeling setup may have been an additional source of error. Within individual river basins, there were no clear trends in the accuracy of HadGEM1 versus HadCM3 predictions at up- or downstream gauging stations. Relative root-mean-square error (RRMSE) scores for the annual cycle of river flow ranged from poor (>50%) to “fair” (20%–50%) with an overall range of 20.7%–1023.5%, comparable to that found in similar global-scale studies. In both models, simulations of the annual cycle of river flow were generally better for high-latitude basins than in low or midlatitudes. There was a relatively small improvement in the annual cycle of river flow in HadGEM1 compared to HadCM3, mostly in the low-latitude rivers. The findings suggest that there is still substantial work to be done to enable GCMs to simulate monthly discharge consistently well over the majority of basins, including improvements to both (i) GCM simulation of basin-scale precipitation and evaporation and (ii) hydrological processes (e.g., representation of dry land hydrology, floodplain inundation, lakes, snowmelt, and human intervention).