Residual Diagnosis of Diabatic Heating from ERA-40 and NCEP Reanalyses: Intercomparisons with TRMM

Diabatic heating is diagnosed from the 40-yr ECMWFRe-Analysis (ERA-40) circulation as a residue in the thermodynamic equation. The heating distribution is compared with the heating structure diagnosed from NCEP and 15-yr ECMWF Re-Analysis (ERA- 15) circulation and latent heating generated from Tropical Rainfall Measuring Mission (TRMM) observations using the convective-stratiform heating (CSH) algorithm. The ERA-40 residual heating in the tropics is found to be stronger than NCEP's (and ERA-15), especially in July when its zonal-vertical average is twice as large. The bias is strongest over the Maritime Continent in January and over the eastern basins and Africa in July. Comparisons with precipitation indicate ERA-40 heating to be much more realistic over the eastern Pacific but excessive over the Maritime Continent, by at least 20% in January. Intercomparison of precipitation estimates from heating-profile integrals and station and satellite analyses reveals the TRMM CSH latent heating to be chronically weak by as much as a factor of 2! It is the low-side outlier among nine precipitation estimates in three of the four analyzed regions. No less worrisome is the inconsistency between the integral of the CSH latent heating profile in the tropics and the TRMM precipitation retrievals constraining the CSH algorithm (e.g., the 3A25 analysis). Confronting TRMM's diagnosis of latent heating from local rainfall retrievals and local cumulus-model heating profiles with heating based on the large-scale assimilated circulation is a defining attribute of this study.