Reconciling modeled and observed trends in the contrast between precipitation in tropical wet and dry regions*

Climate models have long been known to predict a strengthening contrast between wet and dry regions in the tropics and subtropics (30 °S–30 °N), due primarily to anthropogenic forcing. This behaviour has also been observed in blended in situ and satellite products and this hydrological change has been robustly attributed to forcing using detection and attribution approaches, leading to a figure in the latest IPCC report. Intriguingly, such analyses have shown that the observed strengthening contrast is significantly greater than what CMIP6 models simulate, in particular the dry regions display dying trends greater than in any individual model simulation. This result has important implications as it provides powerful evidence that the expected signal of an intensification of the low-latitude water cycle is already underway, and may be larger than expected. Here we will discuss these previous studies but also show new analysis using an updated instrumental dataset (the GPCP-3.2 monthly product) that still displays a detectable strengthening but also, crucially, no discrepancy with the models. We will explore the reasons behind this finding and discuss the implications.