Contrasting Trends in Convective and Large-Scale Precipitation in the Intertropical Convergence Zone From Reanalysis Data Sets

We investigated long-term trends in convective and large-scale precipitation within the Intertropical Convergence Zone during the boreal summer (June-August) for the period 1980-2020 using five different reanalysis datasets. We found that there exists a large diversity in the trends. In particular, we focused on the European Center for Medium-Range Weather Forecasts reanalysis version 5 (ERA5) and the Japanese 55-year reanalysis (JRA55) datasets; because the convective precipitation amount statistically significantly decreases in the former, but increases in the later. In contrast, the large-scale precipitation amount increases in both of these reanalysis datasets. We examine the physical factors leading to contrasting features in the respective trends in convective and large-scale precipitation amount between two reanalysis datasets. A significant decrease of convective available potential energy, which is caused by enhanced atmospheric stability, leads to a decreasing trend of convective precipitation amount in the ERA5 dataset. In contrast, a significant increase of precipitable water is noted in JRA55 dataset. Additionally, the Clausius-Clapeyron relation is larger in JRA55 than ERA5, implying that the sensitivity of humidity to the temperature change is larger in JRA55 than that in ERA5 dataset. On the other hand, the increase of the large-scale precipitation is associated with the increase of medium cloud amount in both ERA5 and JRA55 datasets.Plain Language Summary Typically, the total amount of precipitation is divided into convective precipitation and large-scale precipitation in which the associated mechanisms are quite different. We investigated long-term trends in convective and large-scale precipitation within the Intertropical Convergence Zone, where tropical rainfall is concentrated in a narrow band that accounts for 32% of global precipitation, during the boreal summer for 1980-2020 using five different reanalysis datasets. We focused on the European Center for Medium-Range Weather Forecasts reanalysis version 5 and the Japanese 55-year reanalysis datasets; because the convective precipitation amount decreases significantly in the former, but increases in the later. In contrast, the large-scale precipitation amount increases in both of these reanalysis datasets. The interesting challenge in these results is that there is little agreement in the mechanisms for those trends and this disagreement is most notable in terms of convective precipitation. If our reanalysis products disagree in terms of mechanisms for the trends, understanding future projections will be even more difficult.