Climate change has altered not only the overall magnitude of rainfall but also its seasonal distribution and interannual variability worldwide. Such changes in the rainfall regimes will be most keenly felt in arid and semiarid regions, where water availability and timing are key factors controlling biogeochemical cycles, primary productivity, and the phenology of growth and reproduction, while also regulating agricultural production. Nevertheless, a comprehensive framework to understand the complex seasonal rainfall regimes across multiple timescales is still lacking. Here, we formulate a global measure of seasonality, which captures the effects of both magnitude and concentration of the rainy season, and use it to identify regions across the tropics with highly seasonal rainfall regimes. By further decomposing rainfall seasonality into its magnitude, timing and duration components, we find increases in the interannual variability of seasonality over many parts of the dry tropics, implying increasing uncertainty in the intensity, arrival and duration of seasonal rainfall over the past century. We show that such increases in rainfall variability were accompanied by shifts in its seasonal magnitude, timing and duration, thus underscoring the importance of analysing seasonal rainfall regimes in a context that is most relevant to local ecological and social processes.
Bibliographical noteFunding Information:
X.F. was supported by the US National Science Foundation (NSF) through the Graduate Research Fellowship Program. A.P. had support from the NSF through grant CBET-1033467, the US Department of Energy (DOE) through the Office of Biological and Environmental Research (BER) Terrestrial Carbon Processes (TCP) program (DE-SC0006967), the Agriculture and Food Research Initiative from the USDA National Institute of Food and Agriculture (2011-67003-30222), and the National Aeronautics and Space Administration (NASA) grant NNX09AN76G. I.R-I. acknowledges the support from NASA grant NNX08BA43A. We also thank P. Landolt and the Institute of Fazenda Tamandua in northeast Brazil for providing valuable insights during part of the study, and A. Nishimura for offering valuable suggestions for statistical analysis.