Soil Moisture Estimation by Assimilating In‐Situ and SMAP Surface Soil Moisture Using Unscented Weighted Ensemble Kalman Filter

Abstract

Highly accurate soil moisture information is necessary to understand land surface processes. However, observational techniques do not produce adequately accurate spatial‐temporal continuous regional soil moisture data. The data assimilation method can be used to improve the soil moisture estimations by merging multi‐source observed data, but its performance is affected by error covariance and the quality of assimilated data. We designed eight numerical experiments to analyze how to improve the filter performance through soil moisture assimilation using the unscented weighted ensemble Kalman filter (UWEnKF) and 1‐D Richards equation at Maqu and Erlinghu (ELH) observational sites in the source region of Yellow River (SRYR), China. The experimental results show that the filter performance improves as the quality of assimilated data increases in the soil moisture assimilation experiment when assimilating in‐situ surface soil moisture (SSM) observations, SMAP SSM data and downscaled SMAP SSM data. In other aspects, filter performance is readily affected by model and observation error covariances in soil moisture assimilation. If the SMAP SSM data are taken to be perfect (i.e., small bias), UWEnKF performs differently between different sites because of the underestimation or overestimation of SMAP SSM and model simulations compared to the in‐situ observations. Additionally, different soil moisture assimilation results can be obtained with different initial values at the beginning of the assimilation period. Overall, filter performance can be improved primarily by improving the quality of assimilated data (e.g., downscaling the remote sensing data), and by creating a reasonable and effective method for determining error covariance.This article is protected by copyright. All rights reserved.

A Review of Root Zone Soil Moisture Estimation Methods Based on Remote Sensing

“…In general, the estimation of RZSM has been achieved by assimilating multi-source remote sensing data into physical models, resulting in a series of operational remote sensing products. However, data assimilation is often associated with high computational costs and complexity, and its accuracy is largely dependent on the quality of numerous physical model parameters (such as soil properties, atmospheric forcing, and vegetation cover) [19,91]. Moreover, the mismatch of scale between remote sensing soil moisture data or land surface assimilation soil moisture simulation data and ground-based observation data can also result in significant errors in assessing RZSM [92,93].…”

The Impact of Satellite Soil Moisture Data Assimilation on the Hydrological Modeling of SWAT in a Highly Disturbed Catchment

“…DA techniques provide an integrated framework to analyze all sources of uncertainties from modeling and observation, including the errors of model forcing inputs, model parameters, and model structures, as well as the observation errors from in situ measurements and remote sensing [10]. The assimilation of high-quality in situ SM data in hydrological modeling has been proven to benefit the SM and streamflow estimation [11][12][13]. Nevertheless, the benefits are limited by sparse field measurement sites with lacking capacity to fully account for the spatial heterogeneity of SM.…”