Possibilities of remote sensing monitoring of soil fertility indicators of arable soils

On the example of the test field located in the Tula region (Russia), the possibility of using remote sensing data for mapping the fertility of arable soils is considered. Analysis of ground-based spectrometry data showed that the correlation between the spectral reflectance of the surface of arable soils and fertility indicators increases with an increase in wavelength. The maximum value of the correlation coefficient for mobile phosphorus was 0.84, for humus content - -0.83, for mobile potassium - -0.7, for pH - -0.86, for exchangeable calcium - -0.88, for exchangeable magnesium - 0.67. When analysing the Sentinel-2 satellite data, the correlation coefficients turned out to be higher than in the case of ground-based spectrometry data. Band 5 of Sentinel-2 was found to be the most informative for the detection of most of the analysed parameters of soil fertility. Depending on the property, the R2cv of the models varied from 0.47 to 0.93. The best models were obtained for pH, humus, mobile phosphorus, and exchangeable calcium. Based on regression models, maps of fertility indicators of arable soils for the test field were produced. The obtained models can be used to organize operational monitoring of the main parameters of soil fertility in the field. © Published under licence by IOP Publishing Ltd.