Effects of water-salt interaction on soil spectral characteristics in Hetao Irrigation Areas of Inner Mongolia, China

Abstract: Hetao irrigation area is characterized by less precipitation and more evaporation, which makes the sever soil salinization. Satellite based on remote sensing technology can quickly and accurately monitor the dynamic changes of soil salinization, and provide effective information and methods for land management and ecological restoration. Soil moisture and salinity have similar effects on soil spectrum. In order to explore the influence of soil water-salt interaction on the spectral characteristics of Sentinel-2 satellite, a total of 280 surface soil samples were collected from April to May in 2018 and 2019 in Shahao canal irrigation area of Hetao Irrigation Area in Inner Mongolia, China. Soil moisture content and electrical conductivity were measured, and Sentinel-2 satellite remote sensing data were obtained to construct the soil water-salt-reflectance model based on theory of soil spectral characteristics. Combined with soil water-salt interaction, the water-salt interaction model was constructed, and the simulation effects of the two models on soil spectrum were compared, and the influence of soil water-salt interaction on soil spectral estimation was analyzed. The results showed that: 1) the water-salt interaction was affected by soil moisture and salinity. In the visible light range, the effect range was from -0.11 to 0.29. In the near infrared and short wave infrared ranges, the effect was relatively strong with the range from -0.35 to 0.61. When the soil moisture or electrical conductivity were high, the influence was weak. When the soil moisture or electrical conductivity was similar, the effect was strong. 2) Compared with the theoretical model of soil spectral characteristics, the water-salt interaction model significantly improved the simulation accuracy of soil spectrum, increased the simulation correlation coefficient from 0.14-0.44 to 0.29-0.59, and reduced the root mean square error from 0.032-0.082 to 0.029-0.068. At the same time, compared with the theoretical model, the range of relative error of soil water-salt interaction model was smaller, and the area occupied by low error was larger, indicating that the relative error distribution estimated by reference spectrum can provide valuable information for the retrieval of soil moisture or salt content by Sentinal-2 satellite, and the inversion can be further improved by considering water- salt interaction. 3) The characteristic quantities based on the brine ratio could reflect the water-salt interaction of soil, but the correlation between them changed with the change of brine ratio. When the brine ratio was greater than 1, the correlation between water-salt interaction and characteristic quantity was mainly 0.26-0.49; when the brine ratio was less than 1, the correlation between water-salt interaction and characteristic quantity was mainly 0.35-0.54. This study reveals the interference process of salt and water on spectral characteristics, provides strategies and methods for soil salt estimation, and has certain significance for realizing accurate monitoring of soil salinity at regional scale. However, vegetation coverage is not involved in the study. Therefore, it is necessary to study the spectral characteristics of water-salt interaction from vegetation growth and development and water-salt response mechanism in the future.