Publications

Tian, JX; Lu, H; Yang, K; Qin, J; Zhao, L; Jiang, YZ; Shi, PF; Ma, XG; Zhou, JH (2023). Improving Surface Soil Moisture Estimation Through Assimilating Satellite Land Surface Temperature With a Linear SM-LST Relationship. IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING, 16, 7777-7790.

Abstract
Soil moisture (SM) plays a vital role in linking the global terrestrial water, energy, and carbon cycles. Land data assimilation (DA) is typically applied for acquiring more accurate SM estimates by incorporating remote sensing (RS) retrievals to constraint model parameters and system states. In addition to RS SM retrievals that are frequently used in land DA, the land surface temperature (LST), which can be directly and independently retrieved from thermal infrared measurements, is closely linked with SM, and can also be used in the land DA to improve SM estimation. However, it is hampered by the lack of a general observation operator that links LST retrievals with SM in the land surface model (LSM). Therefore, a novel LST DA scheme is developed in this article, in which RS LST is linked to SM by a linear relationship between the simulated ensembles of SM and LST generated by the Ensemble Kalman filter. Subsequently, MODIS LST is assimilated into the LSM to estimate SM and calibrate the soil parameters (soil porosity, soil texture) with the dual-cycle assimilation algorithm. The DA SM results are validated using an in situ SM network deployed in the central Tibetan Plateau. The results demonstrate that our DA scheme improved the SM accuracy-leading to a decrease in the root-mean-square error from 0.11 m(3)m(-3) to 0.026 m(3)m(-3). This article provides a novel scheme of LST assimilation, which is beneficial for the assimilation of high-resolution RS data.

DOI:
10.1109/JSTARS.2023.3305888

ISSN:
2151-1535