Partitioning of inter-annual variability in the terrestrial water cycle.

The variability of terrestrial water cycle, i.e., precipitation (P), evapotranspiration (E), runoff(Q) and water storage change (ΔS) is the key to understand the hydro-climateextremes. However, a comprehensive assessment for the partitioning of inter-annualvariability in P into E, Q and ΔS is still not available. In this study, we adopt therecently released global monthly hydrologic product known as the Climate DataRecord (CDR) to conduct an initial investigation of variance partitioning of global Pinto the variances (covariances) of (between) E, Q, ΔS. We also examined thelong-term mean and inter-annual variability in P, E, Q and ΔS, and further relatedthe variability to aridity and evaluated the empirical theory of Koster and Suarez.The empirical Koster and Suarez framework was found to overestimate σE/σPand underestimate σQ/σP. The biases of the empirical framework are shown to becaused by ignoring the variability in ΔS and the covariances between E, Q, ΔS,particularly in semi-arid/humid environments. We also find a great divergence of variancepartitioning under extremely hot and cold conditions. That divergence implies significantimportance of global warming in changing the variability of global water cycle. [ABSTRACT FROM AUTHOR]