Surface Water and Energy Budgets for Sub-Saharan Africa in GFDL Coupled Climate Model

Abstract:

This study compare surface water and energy budget variables from the Geophysical Fluid Dynamics Laboratory (GFDL) FLOR models with the National Centers for Environmental Prediction (NCEP) Climate Forecast System Reanalysis (CFSR), Princeton University Global Meteorological Forcing Dataset (PGF), and PGF-driven Variable Infiltration Capacity (VIC) model outputs, as well as available observations over the sub-Saharan Africa. The comparison was made for four configurations of the FLOR models that included FLOR phase 1 (FLOR-p1) and phase 2 (FLOR-p2) and two phases of flux adjusted versions (FLOR-FA-p1 and FLOR-FA-p2). Compared to p1, simulated atmospheric states in p2 were nudged to the Modern-Era Retrospective Analysis for Research and Applications (MERRA) reanalysis. The seasonal cycle and annual mean of major surface water (precipitation, evapotranspiration, runoff, and change of storage) and energy variables (sensible heat, ground heat, latent heat, net solar radiation, net longwave radiation, and skin temperature) over a 34-yr period during 1981-2014 were compared in different regions in sub-Saharan Africa (West Africa, East Africa, and Southern Africa). In addition to evaluating the means in three sub-regions, empirical orthogonal functions (EOFs) analyses were conducted to compare both spatial and temporal characteristics of water and energy budget variables from four versions of GFDL FLOR, NCEP CFSR, PGF, and VIC outputs. This presentation will show how well each coupled climate model represented land surface physics and reproduced spatiotemporal characteristics of surface water and energy budget variables. We discuss what caused differences in surface water and energy budgets in land surface components of coupled climate model, climate reanalysis, and reanalysis driven land surface model. The comparisons will reveal whether flux adjustment and nudging would improve depiction of the surface water and energy budgets in coupled climate models.