An assessment of the quality of surface sensible heat flux derived from reanalysis data through comparison with station observations in Northwest China

Abstract

The present study compares seasonal and interdecadal variations in surface sensible heat flux over Northwest China between station observations and ERA-40 and NCEP-NCAR reanalysis data for the period 1960–2000. While the seasonal variation in sensible heat flux is found to be consistent between station observations and the two reanalysis datasets, both land-air temperatures difference and surface wind speed show remarkable systematic differences. The sensible heat flux displays obvious interdecadal variability that is season-dependent. In the ERA-40 data, the sensible heat flux in spring, fall, and winter shows interdecadal variations that are similar to observations. In the NCEP-NCAR reanalysis data, sensible heat flux variations are inconsistent with and sometimes even opposite to observations. While surface wind speeds from the NCEP-NCAR reanalysis data show interdecadal changes consistent with station observations, variations in land-air temperature difference differ greatly from the observed dataset. In terms of land-air temperature difference and surface wind speed, almost no consistency with observations can be identified in the ERA-40 data, apart from the land-air temperature difference in fall and winter. These inconsistencies pose a major obstacle to the application in climate studies of surface sensible heat flux derived from reanalysis data.