Detection of geologically contrasting structures of the soil cover of arable land using neural network filtering of big remote sensing data

Arable land can have different degrees of fertility contrast within the same agricultural field. One of the reasons for the formation of highly contrasting soil cover structures (HCSPS) is the different depths of the Permian deposits underlying the Quaternary deposits. HCSPS on alternating Quaternary and Permian deposits are common in the Republics of Tatarstan and Bashkortostan, as well as in the Orenburg, Samara, and Ulyanovsk regions. The development of methods for processing large amounts of remote sensing data (LSDD) using neural networks (construction of a multitime soil line) allows for the detection of the spread of VKSFP in large areas with the detail of precision farming systems. The distribution of different crop yields coincides spatially with the VKSFP and is determined by the contrasting properties of the soil cover. The greatest differences in crop yields were observed for sunflower, with a variation of more than 2.5 times from one fertility zone to another. The circular pattern of the VKSFP and the repeated alternation of rings can only be exploited to increase the productivity of the territory through the use of precision farming systems based on remote sensing data.