RESULTS OF USING SPECTRORADIOMETERS FOR IN SOIL MOISTURE OF MONGOLIAN STEPPE ECOSYSTEM

Abstract

Abstract. Determining soil moisture remains an important issue in environmental, agricultural, and crop fields. The use of remote sensing data in the study of soil surface moisture has many benefits, such as: saving time, resources, and money, as well as monitoring soil condition (surface soil moisture). Based on point data from field measurements, can be used to obtain results of soil moisture. The importance of this study is highlighted by the useof spectral measurements and laboratory-determined moisture content of a variety of soils with different geographical conditions spread over three different natural zones, combined with satellite data. The purpose of this research is to determine and analyse surface soil moisture using Sentinel-2 satellite data and Spectro radiometric measurements. For this research area, four areas representative of forest steppe, steppe, and Gobi eco-regions of Mongolia were selected, field research was conducted, data were collected, and Sentinel-2 satellite data was processed and analysed. According to the results of the spectral reflectance measured during the field research, the high reflectance of the wet soil spectrum in the forest field, and the low reflectance of the spectrum representing the Gobi region, respectively, this create the possibility of studying soil moisture by spectral measurement and analysis. The relationship between the spectral reflectance of the red (665 nm) and near-violet (842 nm) light regions shows that the soil of the forest-steppe region is relatively moist compared to the soil of the Gobi region and that the Gobi soil is drier, and the field soil is drier, and these parameters are confirmed by laboratory analysis. A model developed from a linear relationship between field-measured spectrometer reflectance and laboratory-determined moisture provides a formula that can be used for regular monitoring of moisture content using the Normalized Difference Water Index (NDWI) estimated from satellite data. Surface soil moisture (%) = −44.133*NDWI − 11.553. The NDWI calculated from the spectroradiometer compared to the humidity determined in the laboratory had a linear correlation of R2=0.74 or 74%, and when comparing the humidity calculated from Sentinel-2, the agreement was high. In terms of space, the results calculated from satellite data show that the forest and steppe areas have relatively high humidity compared to the Gobi region, and the total area representing the Gobi region has a lower moisture content. Based on these results, it is considered that it is possible to continuously determine the amount of moisture from satellite data using NDWI (surface soil moisture) in a certain space-time range and use it for soil control and monitoring.