Changes in relative humidity profiles over Earth’s oceans in a warming climate: a satellite data based inference

Abstract

Recently we presented a classification of ‘primitive’ relative humidity (RH) profiles into 8 distinct clusters over Earth’s oceans, based on about 18 years (2003-2020) of observations from the AIRS on NASA’s Aqua satellite. Here we investigate the seasonal variability and decadal trends, both in the vertical structure of these profiles, and in their associated area of occurrence. Since vertical structures (except in the marine boundary layer) of each RH-class are generally robust across all seasons and change only weakly in a warming climate, seasonal or decadal changes to their occurrence areas shift patterns of global moisture distribution. Globally, the marine boundary layer exhibits non-linear moistening effects after about 2010, the end of thewarming hiatus. Annual timeseries of ocean areas dominated by RH-classes have linear trends, which are positive only for the most moist and driest RH-classes (in terms of the free troposphere) associated with deep convection and large-scale subsidence favouring conditions for low-level stratocumulus clouds, respectively. Based on estimated linear trends of RH-class occurrences and sea-surface temperatures, we infer projected linear responses of RH in a warming climate. Ocean areas dominated by most moist and driest RH-classes (in terms of the free atmosphere) are estimated to increase by about 1 and 2 %, respectively (corresponding to about 2.5%K−1 and 4.5%K−1, respectively). The averaged global and tropical RH-structure remain almost constant in a warming climate. While this is consistent with other studies, our results show how increases in most moist and dry areas compensate each other, indicating possible increases in the frequency or persistence of future extreme events.