On the Co-Variations of Temperature and Humidity Above the Tropical Convective Boundary Layer across Multiple Timescale

Understanding the moisture and temperature relations in the region immediately above the tropical convective boundary layer (~ 600-800 hPa) across multiple timescales has important implications, such as stability of the tropical middle troposphere, pre-moistening condition for MJO developments and associated precipitation events, water vapor and lapse-rate feedbacks in response to the global warmings. Two decades ago, radiosonde observations, reanalysis, and climate models tended to disagree with each other on the co-variability between tropical averaged temperature and humidity interannual anomalies in the region mentioned above, more so than in any other parts of the tropical troposphere. The most recent generations of satellite observations, reanalyses, and climate models have this discrepancy much reduced, but still with noticeable differences in the same region.

Using weak buoyancy gradient approximation, it can be argued that, at high temporal resolution, the correlation between temperature and humidity in the tropical free troposphere at a given tropical location should be negative. 6-hourly output from reanalysis and low-resolution and high-resolution climate model simulations all support this argument of the dominance of negative correlations. The correlation becomes bi-modal distribution when the seasonal cycle is examined, with large positive correlations in the subtropics and significant negative correlations primarily over the deep tropical oceans. At the interannual timescale, the spatial patterns of the correlation closely resemble the ENSO pattern. When tropical-averaged temperature and anomalies are considered, most climate models and most recent reanalyses exhibit a result closely following the “constant relative humidity” hypothesis at all vertical layers. This is not true for temperature and humidity retrieved from AIRS (Atmospheric Infrared Sounder) observations, especially above the tropical convective boundary layer. But AIRS results changed noticeably from version 6 to version 7 for this vertical layer, highlighting the challenge of confidently deriving temperature and humidity interannual anomalies from observations. Last, we review the current status of middle-tropospheric moisture and temperature observations and deliberate on possible improvement on such observations.