Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Hydrology and Remote Sensing Laboratory » Research » Publications at this Location » Publication #377116
Research Project: Integrating Remote Sensing, Measurements and Modeling for Multi-Scale Assessment of Water Availability, Use, and Quality in Agroecosystems

Location: Hydrology and Remote Sensing Laboratory

Title: The added value of brightness temperature assimiilation for the the SMAP Level-4 surface and root-zone soil moisture analysis over mainland China

Author
item QUI, J. - Sun Yat-Sen University
item DONG, J. - US Department Of Agriculture (USDA)
item Crow, Wade
item ZHANG, X. - Nanjing Agricultural University
item REICHIE, R. - National Aeronautics And Space Administration (NASA) - Johnson Space Center
item DE LANNOY, G. - Catholic University Of Leuven

Submitted to: Hydrology and Earth System Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 2/14/2021
Publication Date: 3/29/2021
Citation: Qui, J., Dong, J., Crow, W.T., Zhang, X., Reichie, R., De Lannoy, G. 2021. The added value of brightness temperature assimiilation for the the SMAP Level-4 surface and root-zone soil moisture analysis over mainland China. Hydrology and Earth System Sciences. 25(3):1569–1586. https://doi.org/10.5194/hess-25-1569-2021.
DOI: https://doi.org/10.5194/hess-25-1569-2021

Interpretive Summary: Soil moisture is an important climate variable because of its impact on the land surface water, energy, and nutrient cycles. For example, soil moisture controls how much water from a given rainfall event is stored in the soil (and becomes available for natural and agricultural plant growth) versus runs off into streams, lakes, and reservoirs. New global estimates of soil moisture are now available from the NASA Soil Moisture Active Passive (SMAP) satellite mission. Using an extremely dense network of ground-based soil moisture observations in China, this paper evaluates the added value of SMAP soil moisture estimates relative to existing soil moisture estimates available from numerical land surface models. In addition, it identifies key ways in which SMAP soil moisture estimates can be improved and provide even more value. Once applied to SMAP soil moisture estimates acquired over the United States, these insights will improve our ability to monitor, and thus mitigate, the impact of agricultural drought on domestic agriculture production.

Technical Abstract: The Soil Moisture Active Passive (SMAP) Level-4 Surface Soil Moisture and Root-Zone Soil Moisture (L4) product provides global estimates of surface soil moisture (SSM) and root-zone soil moisture (RZSM) via the assimilation of SMAP brightness temperature (Tb) observations into the Catchment Land Surface Model (CLSM). Here, using in-situ measurements from 2474 sites in mainland China, we evaluate the performance of soil moisture estimates from L4 and from a baseline "open- loop" (OL) simulation of CLSM without Tb assimilation. Using random forest regression, the efficiency of the L4 data assimilation (DA) system (i.e., the performance improvement in L4 relative to OL) is attributed to 8 control factors related to the land surface modelling (LSM) and radiative transfer modeling (RTM) components of the L4 system. Results show that 77% of the 2287 9-km EASE grid cells in mainland China that contain at least one ground station exhibit an increase in the Spearman rank correlation skill (R) with in-situ measurements for L4 SSM compared to that of OL, with an average R increase of approximately 14% ('R = 0.056). RZSM skill is improved for about the same percentage of 9-km EASE grid cells, but the average R increase for RZSM is only 7% ('R = 0.034). Results further show that the SSM DA efficiency is most strongly related to the error in Tb observation space, followed by the error in precipitation forcing and microwave soil roughness. For RZSM DA efficiency, the three dominant control factors remain the same, although the importance of soil roughness exceeds that of the Tb error. For the skill of the L4 and OL estimates themselves, the top control factors are the precipitation error and the SSM-RZSM coupling strength error (in descending order of factor importance for ROL), both of which are related to the LSM component of the L4 system. Finally, we find that the L4 system can effectively filter out errors in precipitation. Therefore, future development of the L4 system should focus on improving the characterization of the SSM-RZSM coupling strength.