In this diagnostic study we analyze changes of rainfall seasonality and dry spells by the end of the twenty-first century under the most extreme IPCC5 emission scenario (RCP8.5) as projected by twenty-four coupled climate models contributing to Coupled Model Intercomparison Project 5 (CMIP5). We use estimates of the centroid of the monthly rainfall distribution as an index of the rainfall timing and a threshold-independent, information theory-based quantity such as relative entropy (RE) to quantify the concentration of annual rainfall and the number of dry months and to build a monsoon dimensionless seasonality index (DSI). The RE is projected to increase, with high inter-model agreement over Mediterranean-type regions—southern Europe, northern Africa and southern Australia—and areas of South and Central America, implying an increase in the number of dry days up to 1 month by the end of the twenty-first century. Positive RE changes are also projected over the monsoon regions of southern Africa and North America, South America. These trends are consistent with a shortening of the wet season associated with a more prolonged pre-monsoonal dry period. The extent of the global monsoon region, characterized by large DSI, is projected to remain substantially unaltered. Centroid analysis shows that most of CMIP5 projections suggest that the monsoonal annual rainfall distribution is expected to change from early to late in the course of the hydrological year by the end of the twenty-first century and particularly after year 2050. This trend is particularly evident over northern Africa, southern Africa and western Mexico, where more than 90% of the models project a delay of the rainfall centroid from a few days up to 2 weeks. Over the remaining monsoonal regions, there is little inter-model agreement in terms of centroid changes.
Bibliographical noteFunding Information:
Two anonymous reviewers provided constructive comments that helped to improve this manuscript. The authors acknowledge the World Climate Research Programmes Working Group on Coupled Modeling, which is responsible for CMIP, and the NOAA, for providing from their Web site the GPCC precipitation data. S.P., V.L. and S.H. wish to acknowledge the financial support provided by the ERC-Starting Investigator Grant NAMASTE (Grant No. 257106) and by the CliSAP/Cluster of excellence in the Integrated Climate System Analysis and Prediction. During this research S.P. was partially supported by the NOAA Climate and Global Change Postdoctoral Fellowship Program, administered by the University Corporation for Atmospheric Research. A.P. gratefully acknowledges NSF Grants: NSF FESD 1338694, CBET 1033467, EAR 1331846, EAR 1316258 as well as the US DOE through the Office of Biological and Environmental Research, Terrestrial Carbon Processes program (DE-SC0006967), the Agriculture and Food Research Initiative from the USDA National Institute of Food and Agriculture (2011-67003-30222). X.F. acknowledges funding from the NSF Graduate Research Fellowship Program.
- CMIP5 models
- Drought index
- Rainfall seasonality indicators
- Representative concentration pathways