A semi-empirical model for soil moisture content retrieval based on the Kubelka-Munk model

• Propose a semi-empirical soil moisture model based on the Kubelka-Munk model. • Explore the relationship between soil moisture and soil optical parameters. • Develop a new physically-based index soil moisture and valid it with imagery data. Soil moisture is a critical factor influencing crop phenological development, climate patterns, and environmental changes. Accurate monitoring of soil moisture is therefore crucial for the natural ecological environment. The traditional methods of soil moisture monitoring are time-consuming and labor-intensive, so it is essential to establish a soil moisture retrieval model using remote sensing technology, meeting the needs for rapid monitoring and large-scale areas. Among the many methods, a radiative transfer model can explain the relationship between the optical properties of the soil media and soil moisture from the spectral mechanism. A semi-empirical radiative transfer model—the soil moisture Kubelka-Munk (SM-KM) model—is presented here to describe the relationship between soil moisture and soil spectra through the optical parameters, based on the Kubelka-Munk (KM) model. Four types of soil were utilized to validate the ability of the model, which closely simulate the reflectance of different soils in various moisture contents. Since the model requires soil moisture gradient data as boundary conditions, simplification of this model should be conducted in practical application. The soil moisture index (SMI KM ) was constructed with the SM-KM model, which enables the retrieval of soil moisture based solely on the spectral information. The index was validated using Sentinel-2 Multi-Spectral Instrument (MSI) data from Yitong County, Jilin province, China, where it obtained good results with R C 2 of 0.640 and R p 2 of 0.633. A soil moisture content map was also generated using the SMI KM index. The proposed approach represents a convenient and rapid method for soil moisture estimation.