Validation of a new soil moisture product: Soil MERGE (SMERGE)

Authors: TOBIN, K. - Texas A&M University; Crow, Wade; DONG, J. - US Department Of Agriculture (USDA); BENNETT, M.E. - Texas A&M University

Submitted to: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/15/2019
Publication Date: 8/16/2019
Citation: Tobin, K., Crow, W.T., Dong, J., Bennett, M. 2019. Validation of a new soil moisture product: Soil MERGE (SMERGE). IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 12(9):3351-3365. https://doi.org/10.1109/JSTARS.2019.2930946.
DOI: https://doi.org/10.1109/JSTARS.2019.2930946

Interpretive Summary: Accurate soil moisture products are of value for important agricultural applications including: drought monitoring, irrigation scheduling and optimizing fertilizer usage. Within the past two decades, remotely-sensed surface soil moisture products have been produced from multiple satellite-based sensors. At the same time, significant advances have been made in the generation of soil moisture products by soil water balance models. However, to maximize the accuracy and length of available soil moisture data sets, strategies need to be developed to optimally merge concurrent soil moisture products acquired from different sources. This paper describes a new root-zone soil moisture product for the United States derived via the simple merging of independent soil moisture estimates obtained from both modelling and multiple satellite sensors. Validation results in the paper illustrate that the product represents a clear improvement over existing soil moisture data sets and motivate its potential application to important agricultural and hydrological decision-support problems.

Technical Abstract: SoilMERGE (SMERGE) is a 0.125-degree, root-zone soil moisture (RZSM) product (0 to 40 cm depth) within the contiguous United States (CONUS). This product is developed by merging RZSM output from the North American Land Data Assimilation System (NLDAS) with surface satellite retrievals from the European Space Agency (ESA) Climate Change Initiative (CCI). SMERGE, at present, spans four decades (1978 to 2016). Here, we introduce the SMERGE approach and describe the validation of SMERGE RZSM estimates using three geophysical observations. (1) Comparison with sparse in situ soil moisture data acquired from the Soil Climate Analysis Network (SCAN) and the U.S. Climate Reference Network (USCRN); (2) Ranked correlation analysis against Normalized Difference Vegetation Index (NDVI) datasets. (3) Ranked correlation analysis of antecedent RZSM with storm-event streamflow across a range of precipitation intensities (5 to 45 mm/day). Relative to in situ SCAN and USCRN observations, SMERGE has an average daily correlation of 0.7 to 0.8 and ubRMSE close to 0.04 m3/m3 - a level that is commonly applied as a validation target for large-scale soil moisture datasets. NDVI benchmarking allows us to indirectly evaluate SMERGE across CONUS and reveals it can predict near-term vegetation health anomalies with skill comparable to that of RZSM products generated by more complex data assimilation methods. In addition, streamflow-based evaluation results demonstrate that SMERGE antecedent RZSM can be used as a reliable predicator of storm-event runoff efficiency for rainfall events greater than 25 mm/day.