Applications of a thermal-based two-source energy balance model coupled to surface soil moisture

Authors: SONG, L. - Southwest University; DING, L. - Chinese Academy Of Agricultural Sciences; Kustas, William - Bill; XU, Y. - Southwest University; LIU, S. - Beijing Normal University; MA, M. - Southwest University; XUE, K. - Southwest University; BAI, Y. - Chinese Academy Of Sciences; XU, Z. - Beijing Normal University

Submitted to: Remote Sensing of Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/24/2022
Publication Date: 2/3/2021
Citation: Song, L., Ding, L., Kustas, W.P., Xu, Y., Liu, S., Ma, M., Xue, K., Bai, Y., Xu, Z. 2021. Applications of a thermal-based two-source energy balance model coupled to surface soil moisture. Remote Sensing of Environment. 271(2022):112923. https://doi.org/10.1016/j.rse.2022.112923.
DOI: https://doi.org/10.1016/j.rse.2022.112923

Interpretive Summary: Evapotranspiration (ET) and its components of soil evaporation (E) and plant transpiration (T) are a nexus of the water, energy, and carbon cycles, which has been applied in climate, hydrology, drought monitoring, and crop yield predictions. The two-source energy balance (TSEB) model using remotely sensed land surface temperature has been shown to address key soil and plant exchange processes for many complex canopies. A new transpiration algorithm was developed and evaluated under a wide range of surface soil water content values and vegetation cover conditions and compared with the performance of the original TSEB model. Results showed significant improvement in ET and T and E partitioning, especially for a semi-arid shrub-forest site with sparse vegetation cover. With satellite land surface temperature data and surface soil moisture retrievals from microwave satellite observations, the refined TSEB model potentially will provide more reliable monitoring of ET partitioning in semi-arid regions under sparse canopy cover conditions.

Technical Abstract: The two-source energy balance (TSEB) model using the land surface temperature (LST) as a key boundary has been used to estimate land surface evapotranspiration (ET) over various landcovers and environmental conditions. However, LST may not always provide an adequate boundary condition to simultaneously constrain the soil evaporation and plant transpiration especially under water limited conditions. A refinement to TSEB model by coupling surface soil moisture information to derive the soil and vegetation component temperatures and a new transpiration algorithm was developed (TSEB-SM). The TSEB-SM model was evaluated under a wide range of surface soil water content values and vegetation cover conditions and compared with the performance with the original TSEB model using only LST. While the results showed that the TSEB-SM model produced similar agreement in the fluxes and ET for the cropland, grassland as the original TSEB, TSEB-SM model performance was notably improved at the shrub-forest site with a significant reduction in mean absolute percent difference in daily ET from nearly 65% to 24%. It also appears to be more reliable in partitioning ET into soil evaporation and plant transpiration when compared to the partitioning using the water use efficiency (uWUE) approach in combination with the eddy covariance measurements. With satellite data such as MODIS LST and leaf area index, and surface soil moisture retrievals from microwave satellite observations, the TSEB SM model may potentially be a more reliable tool for monitoring regional ET partitioning under sparse canopy cover conditions.