Yield Dynamics in Long-Term Field Experience (Lyubertsy Experimental Field of the NIUIF). Message 3. Long-Term Yield and Soil Properties

On the territory of one of the fields of a long-term field experiment with a 4-pole crop rotation (the Lyubertsy experimental field of the NIUIF), the set of properties of soil samples obtained from wells in each plot was analyzed. The undisturbed structure samples were taken 5 cm deep to a depth of 150 cm. In this work, arable and sub-arable layers up to 50 cm were analyzed. In the arable layer, for dry dispersion aggregates >10 mm and for the equilibrium soil density (in spring before tillage), most of the plots were far from optimal, with the best plots being in the first 2 out of 4 field repetitions, and in density in the third. Also, in the arable layer, half of the plots had a low content of exchangeable potassium by Oil content, and the other half (like all plots) – the content of Ca, Mg and the amount of absorbed bases – was very low. Thus. The connection with the long-term application of K-fertilizers for the content of exchangeable potassium in the soil turned out to be not very strict. To obtain a spatial representation of the heterogeneity of soils in the field, the plots were grouped separately for physical, chemical and morphological soil parameters. Spatial ranges for each of the groups have been identified. For physical indicators, the groups of plots were divided mainly according to the indicators of the under–arable layer (the same for morphological ones), and for chemical indicators – according to the indicators of both arable and under-arable layers. A variance analysis of the dependence of average long–term yields on plots of land on experimental options and on grouped soil indicators showed the following decreasing number of shares of influence of factors: experimental options – 54–82%, groups of chemical indicators – 22–31%, groups of physical indicators – 12–14%, groups of morphological indicators – 5.5–5.8%. When the control and NP options were removed, leaving only 8 options with K-fertilizers in order to more clearly identify the influence of soil factors, we got the following number of shares of influence: groups of physical indicators – 37–43%, groups of chemical indicators – 27–35%, experimental options – 19–29%. Variance and regression analyses made it possible to clarify the specific soil indicators most related to the average long-term yield; these turned out to be indicators of the under-arable layer, mainly physical (fractional composition) and the content of exchangeable Ca. Significant regression equations are obtained.