Investigating trends, variability in observed and simulated upper tropospheric humidity and outgoing longwave radiation

Upper tropospheric humidity (UTH) is a diagnostic of the atmospheric water cycle and strongly contributes to climate sensitivity. Therefore, it is important to understand UTH variability and how this is represented by global climate models. Here, infrared and microwave brightness temperature observations and satellite simulations based on ECMWF Reanalysis v5 (ERA5) and the Hadley Centre Global Environment Model version 3 (HadGEM3) Atmospheric Model Intercomparison Project (AMIP) data are used to evaluate and characterise UTH variability since 1979. UTH satellite observations have been simulated using a radiative transfer code (RTTOV) from ERA5 and HadGEM3 to provide a more direct comparison of the model and reanalysis to observations.

We present results on the sensitivities of water vapour brightness temperatures. There are competing influences of temperature and specific humidity on the brightness temperatures. The effect of these is such that fluctuations can be considered as a proxy for relative humidity. Despite this, a spurious increase in UTH of up to 1% is identified for a 1K increase in profile temperature when relative humidity remains constant.

We also investigate trends and variability of UTH. Using Principal Component Analysis, we explore the spatial and temporal impact of El Niño Southern Oscillation (ENSO) on UTH distribution and link this to changes in outgoing longwave radiation (OLR). Trends show increased UTH over the Indian Ocean and decreases over the western Pacific. This mirrors large-scale dynamic changes in the Walker Circulation, which shows a weakening of the circulation over the same period.