ERA5-Land overestimates runoff coefficient but underestimates runoff recession rate in the central Tibetan permafrost region

• Published in: Journal of Hydrology: Regional Studies
• Main Authors: Li Liu, Yonghong Yi, Huiru Jiang, Youhua Ran, Deliang Chen
• Format: Article in Journal/Newspaper
• Language: English
• Published: Elsevier 2024
• Subjects: ERA5-Land; Runoff recession; Runoff coefficient; Freeze/thaw dynamics; Tibetan permafrost region; Physical geography; GB3-5030; Geology; QE1-996.5; permafrost
• Online Access: https://doi.org/10.1016/j.ejrh.2024.101792; https://doaj.org/article/b0fc92037b8949a2a1e3f5403047d0ab

Study region: The source region of Yangtze River (SRYR) Study focus: Better understanding the influence of soil freeze/thaw (F/T) activity on runoff production is vital for cold regions like the Qinghai-Tibet Plateau. While global reanalysis products offer full spatial coverage and fine temporal resolution for runoff estimates, their accuracy in depicting runoff dynamics within permafrost-dominated regions remains uncertain. To address this gap, a thorough assessment of the ERA5-Land runoff products within the SRYR was undertaken to determine their effectiveness in representing specific permafrost hydrological mechanisms. New hydrological insights for the region: Our analysis reveals that ERA5-Land adequately captures the intra-annual variability of runoff, but tends to overestimate runoff coefficients while underestimate runoff recession rates. The monthly runoff coefficient fluctuates during the thaw season, showing decreasing trends from May to increasing trends after July. The runoff recession rate generally shifts from acceleration to retardation during the freezing period (October-March), influenced significantly by snow cover and soil F/T dynamics. There's a discernible delay in the observed transition time (0.3 days/a) of runoff recession from 1980 to 2020, closely linked (R>0.35, p<0.05) to soil F/T indices. Notably, the simulated transition time lacks significant trends, and displays contrasting correlations with soil F/T indices. This highlights a critical need for the ERA5-Land model to refine its representation of soil F/T dynamics, particularly snow physics and soil thermal schemes.