Study of the dynamics of herbs productivity based on long-term monitoring data

Abstract

The paper shows methods for studying the data bank for monitoring the yield of grasses within the agrolandscape. Monitoring of the yield of different-aged (1, 2 and 3 years of use) grass legumes was carried out in 2001-2019 at an agroecological test site located within a finite moraine hill at permanent sampling points on a transect that crosses all the main micropositions of the agricultural landscape. The influence of the features of various landscape structures (slopes of different exposure, landscape plots within their limits and the variegation of the soil cover) on the dynamics of grass yield was studied. The data array was processed using the methods of Variable Components Analysis, principal components, and correlation analysis. It is shown that the productivity of grasses is reliably influenced by the features of the landscape, agroclimatic conditions of the year and the age of the grass stand. The formation of the spatial variegation of the herb yield is most influenced by the nature of the soil cover (≈ 21 %), to a lesser extent by the features of individual parts of the slopes (≈ 17 %). It was found that with the aging of the herbage, there is a noticeable increase in the influence of exposure (from 3.4 to 8.0 %) and microlandscape (from 16 to 22 %) factors on the spatial variability of their productivity, and a decrease in the influence of soil cover characteristics (from 24 to 18 %). It was revealed that the dynamics of agroclimatic parameters noticeably affects only the dependence of the variegated yield on the exposure factor. An attempt has been made to divide the years of research by the nature of the dynamics of the yield of grasses into agro climatically homogeneous groups. It has been determined that different groups of observation years differ in productivity and in the nature of its spatio-temporal variability, as well as in the factors that determine them and in the conditions that affect these factors. This makes, when predicting the yield of grasses of different ages, to create mathematical models of its dependence on landscape conditions for different time clusters.