Divergent determinants on interannual variability of terrestrial water cycle across the globe

Jinyu Zhu,Dongqin Yin,Xiang Li,Ruirui Zhu,Hongxing Zheng
DOI: https://doi.org/10.1016/j.scitotenv.2024.174046
2024-10-01
Abstract:Intensifying variability in precipitation under a changing climate is projected to amplify fluctuation in terrestrial hydrological cycle, leading to more severe water-related disasters. The connections between interannual variability of hydrological components and factors influencing these connections have not been clearly defined yet. Based on terrestrial water budget from Climate Data Record, we identify dominant factors influencing partitioning interannual variability of precipitation (P) into that of evapotranspiration (E), runoff (Q), and water storage deviation (ΔS) across the globe by employing geographical detector model (GDM). Sensitivities of the variability partitioning to dominant factors are quantified for different hydroclimate regions by linear regression model and law of total differential. Results show that dominant factors influencing precipitation variability partitioning (VP) are different across distinct hydroclimate conditions. Comparing the statistical index (q value) of the GDM, it can be seen that surface air temperature (Ta), snow water equivalent (SWE) and water storage capacity (Smax) are dominant factors of VP in humid, semi-arid and arid regions, respectively. Changes in P variability largely can transfer into Q variability in humid region. The P variability partitioned into Q variability is dramatically reduced in semi-arid region with SWE decreasing, while P variability partitioned into ΔS variability increases with Smax increasing in arid region. Joint effects of Ta and coefficient of variation of precipitation (Pcv) are found to be the most important interaction in determining VP across the globe. Furthermore, warmer temperatures in humid region cause >90 % of the change in precipitation variability to be transferred to Q variability change. In semi-arid region with snowfall, decreased SWE has strong effect on changes in ΔS (30-40 %) and Q (20-40 %) variability. Our findings imply a changing VP and more severe impacts of hydrological extremes under future climate, where intensive changes in Ta, SWE and land cover are projected.
What problem does this paper attempt to address?