Application of the two-source energy balance model with microwave-derived soil moisture in a semi-arid agricultural region

Authors: XU, Y. - Southwest University; SONG, L. - Southwest University; Kustas, William - Bill; XUE, K. - Southwest University; LIU, S. - Beijing Normal University; MA, M - Southwest University; XU, T. - Beijing Normal University; JIANG, H. - Beijing Normal University

Submitted to: International Journal of Applied Earth Observation and Geoinformation
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/19/2022
Publication Date: 6/30/2022
Citation: Xu, Y., Song, L., Kustas, W.P., Xue, K., Liu, S., Ma, M., Xu, T., Jiang, H. 2022. Application of the two-source energy balance model with microwave-derived soil moisture in a semi-arid agricultural region. International Journal of Applied Earth Observation and Geoinformation. 112. Article e102879. https://doi.org/10.1016/j.jag.2022.102879.
DOI: https://doi.org/10.1016/j.jag.2022.102879

Interpretive Summary: The remote sensing-based Two-Source Energy Balance (TSEB) model uses the land surface temperature (LST) from satellites as the key boundary condition for estimating water use or evapotranspiration (ET), soil and plant stress, and has been applied from field to global scales. However, LST may not always provide a sufficient boundary condition to constrain both soil evaporation and plant transpiration, especially under water limited conditions. Since soil moisture plays a critical role in ET, soil moisture from microwave remote sensing is integrated along with a more sophisticated transpiration algorithm into the TSEB model (TSEB-SM). TSEB-SM was applied over an agricultural and natural landscape in a semi-arid region in China using the airborne-based microwave-derived soil moisture data and satellite based LST. The TSEB-SM model yielded better agreement with flux tower measurements of ET than the original TSEB, especially over the dry/sparsely vegetated natural ecosystem. This suggests that combining microwave soil moisture and LST from satellites with TSEB-SM can improve ET estimates over semi-arid areas resulting in better water management decisions in water limited regions.

Technical Abstract: Evapotranspiration (ET) is the crucial component that connects the land surface’s water, carbon and energy cycles. Remote sensing-based models provide the possibility of mapping ET from field to regional and even global scales. Here, a Two-Source Energy Balance modeling scheme in combination using microwave-derived near surface soil moisture and thermal-infrared surface temperature (TSEB-SM) as key boundary conditions were used over irrigated agricultural and surrounding semi-arid sparse natural vegetated area. The TSEB-SM yielded better agreement with tower measurement of the surface fluxes, especially over the dry/low vegetation covered sites. Model outputs from TSEB-SM reduced the values of mean absolute percent difference (MAPD) by 2% for net radiation (Rn) and soil heat flux and soil heat flux (G) by 2%, 4% for sensible heat flux (H), and 5% for latent heat flux (LE) when compared with the performance of the TSEB model only using radiometric surface temperature as the key boundary. Latent heat flux and its components of soil and canopy predictions from the TSEB-SM and TSEB models were compared for two days when the sky was mostly clear. The TSEB-SM model computed more reliable estimates of LE under dry soil surface/low vegetation cover conditions, reducing the overestimation of LE by the TSEB model, which was likely due to overestimation of the canopy LE. The TSEB-SM model yielded similar LE results as TSEB model over moist soil surface/density vegetation cover conditions, indicating that TSEB-SM can provide reliable flux estimation and LE partitioning over wide range of environmental conditions. With reliable microwave remote sensing of soil moisture, TSEB-SM has potential for ET monitoring under various landcovers and environmental conditions in natural and agricultural landscapes.