Multiproduct Characterization of Surface Soil Moisture Drydowns in the United Kingdom

• Published in: Journal of hydrometeorology Vol. 24; no. 12; pp. 2299 - 2319
• Main Authors: Tso, Chak-Hau Michael, Blyth, Eleanor, Tanguy, Maliko, Levy, Peter E., Robinson, Emma L., Bell, Victoria, Zha, Yuanyuan, Fry, Matthew
• Format: Journal Article
• Language: English
• Published: Boston American Meteorological Society 01.12.2023
• Subjects: Atmosphere; Atmospheric boundary layer; Climate; Climate and hydrology; Climate models; Cosmic rays; Decay; Dielectric properties; Drought; Ecological research; Environmental impact; Evaporation; Hydrologic models; Hydrology; Land cover; Land surface models; Machine learning; Moisture content; Neutrons; Rain; Rainfall; River flow; Satellite data; Satellites; Soil moisture; Soil properties; Soil surfaces; Soil types; Statistical models; Support systems; Variables
• ISSN: 1525-755X; 1525-7541
• DOI: 10.1175/JHM-D-23-0018.1

The persistence or memory of soil moisture ( θ ) after rainfall has substantial environmental implications. Much work has been done to study soil moisture drydown for in situ and satellite data separately. In this work, we present a comparison of drydown characteristics across multiple U.K. soil moisture products, including satellite-merged (i.e., TCM), in situ (i.e., COSMOS-UK), hydrological model [i.e., Grid-to-Grid (G2G)], statistical model [i.e., Soil Moisture U.K. (SMUK)], and land surface model (LSM) [i.e., Climate Hydrology and Ecology research Support System (CHESS)] data. The drydown decay time scale ( τ ) for all gridded products is computed at an unprecedented resolution of 1–2 km, a scale relevant to weather and climate models. While their range of τ differs (except SMUK and CHESS are similar) due to differences such as sensing depths, their spatial patterns are correlated to land cover and soil types. We further analyze the occurrence of drydown events at COSMOS-UK sites. We show that soil moisture drydown regimes exhibit strong seasonal dependencies, whereby the soil dries out quicker in summer than winter. These seasonal dependencies are important to consider during model benchmarking and evaluation. We show that fitted τ based on COSMOS and LSM are well correlated, with a bias of lower τ for COSMOS. Our findings contribute to a growing body of literature to characterize τ , with the aim of developing a method to systematically validate model soil moisture products at a range of scales.