SPATIAL RELATIONSHIPS BETWEEN SOIL PROPERTIES AND MAIZE YIELD

Abstract

Soil properties play an important role in the spatial variability of maize yield. The aim of the present in-vestigation was to assess the role of soil factors in the patterns of spatial-temporal variation of maize yield in the Polissia and Forest-Steppe zones of Ukraine. To analyze the impact of soil factors on maize yield, we used the following indicators: humus reserves, pH, and density of soil, sand, clay or silt content for different soil layers. Principal Component Analysis identified 6 principal components, which explain 98.5 % of the total soil dispersion indices. Each of the six components demonstrates the highest correlation with one or more soil variables. By establishing these correlations, we can identify the nature of each principal compo-nent, as well as mark the areas of hypersensitivity to each of the principal components on the map. It was established that maize yield dynamics on the research territory was represented as logistic model. The fol-lowing indicators were used as parameters of the yield pattern: yield lower limit indicates the lowest yield level during the research period which was observed in the early and mid-1990s; slope of the trend curve, shows the speed of changes in yield over time; time from experiment’s beginning (ED50) – the time, which is required to achieve half of the maximum level of yield growth and at the same time the moment of the highest rate of productivity growth; upper yield level – the highest level of yield, at which under this level of agricul-tural technology, the yield is determined precisely by the biotic potential of the territory. Using regression analysis the relationship between maize yield parameters and soil properties was established. It is found that the slope of the logarithmic model, which determines the potential rate of yield growth, mostly depends on humus reserves in the soil. The lower and upper limits of maize yield are negatively correlated with sand content of the soil, and the time of sharp increase in yield is also mostly dependent on the particle size distri-bution of the soil and is positively correlated with sand content.