Observational constraints on the uncertainties of the future precipitation change projections

Future projections of global mean precipitation change (ΔP) based on Earth system models have larger uncertainties than those of global mean temperature changes (ΔT). While many observational constraints on ΔT have been proposed, constraints on ΔP have not been well studied and are often complicated by the large influence of aerosols on precipitation. By analyzing the Coupled Model Intercomparison Project Phase 5 and 6 ensembles, we show that the upper bound (95^th percentile) of ΔP (2051-2100 minus 1851-1900, % of the 1980-2014 mean) can be lowered from 6.2% to 5.2-5.7% (min-max range of sensitivity analyses) under a medium greenhouse gas concentration scenario. ΔP for 2051-2100 is well correlated with the global mean temperature trends during recent decades after 1980 when global anthropogenic aerosol emissions were nearly constant. ΔP is also significantly correlated with the recent past trends of precipitation when we exclude some tropical land areas with few rain gauge observations. Based on these significant correlations and observed trends, the variance of ΔP can be reduced by 8-30%. The observationally constrained ranges of ΔP should provide further reliable information for impact assessments.