PubAg

Main content area

A climatology of tropospheric humidity inversions in five reanalyses

Author:
Brunke, Michael A., Stegall, Steve T., Zeng, Xubin
Source:
Atmospheric research 2015 v.153 pp. 165-187
ISSN:
0169-8095
Subject:
advection, autumn, climate, climatology, evaporation, humidity, melting, prediction, summer, temperature, troposphere, turbulent flow, winter, Polar Regions
Abstract:
Specific humidity is generally thought to decrease with height in the troposphere. However, here we document the existence of specific humidity inversions in five reanalyses: the National Centers for Environmental Prediction (NCEP) second reanalysis (NCEP-2), the European Centre for Medium-Range Forecasts (ECMWF) 40-year reanalysis (ERA-40), the Modern Era Retrospective Analysis for Research Applications (MERRA), NCEP's Climate Forecast System Reanalysis (CFSR), and the ECMWF interim reanalysis (ERA-Interim). These inversions are most frequent in the polar regions. Inversions do occur elsewhere, most notably over the subtropical stratus regions, but are less frequent and likely overproduced depending on the location. Polar inversions are the most persistent in winter and the strongest (as defined by the humidity difference divided by the pressure difference across the inversion) in summer or autumn with low bases (at pressures>900hPa). Winter humidity inversions are lower, being near-surface, due to the persistence of low-level temperature inversions associated with these humidity inversions, while summer humidity inversions tend to be located near cloud top providing moisture to prevent the melt season stratus from evaporating. The most important contributions to affect humidity inversions in MERRA are dynamics, turbulence, and moist physics. However, local advection may not play as much of a role as regional humidity convergence. The subtropical stratus inversions are as thick as polar humidity inversions but with higher bases generally at pressures <900hPa. These inversions are confirmed by rawinsonde data, but there are discrepancies between the observed annual and diurnal cycles in inversion frequency and those portrayed in the reanalyses.
Agid:
5459510