Peculiarities of remote sensing diagnostics of agrochemical properties of arable soils

Agrochemical surveys of arable soils are laborious, costly and time-consuming. The introduction of modern remote sensing and digital technologies has great potential to overcome these disadvantages, but requires additional scientific research. The article presents the results of a comparison of the efficiency of using different types of remote sensing data for spatial modeling the main agrochemical properties of soils by the example of a test site in the Tver region. The spectral reflectance of the soil surface was determined in the field, its simultaneous measurement by an unmanned aerial vehicle (UAV) with a standard camera, and additionally the test plot was analyzed on Sentinel-2 satellite images. Regression analysis showed that the most accurate predictive models of soil agrochemical properties can be obtained from field spectrometry, lower quality models are obtained from UAV data, and the lowest quality models are obtained from satellite data. The main reason for this seems to be the spatial variation of soil agrochemical parameters and the generalized representation of their open surface on UAV data and satellite images.