Математичні моделі прогнозування врожайності біоенергетичних культур

Purpose. To determine mathematical models suitable for yield prediction of bioenergy crops and to establish the patterns of the crop health monitoring. Methods. Field studies were conducted in the zone of unstable soil moisture (Experimental Field of the Institute of Bioenergy Crops and Sugar Beet NAAS of Ukraine, Ksaverivka Druha, Kyiv region) and sufficient soil moisture (Yaltushkiv Experimental Breeding Station of the IBCSB NAAS of Ukraine, Chereshneve, Vinnytsia region) in the Right-Bank Forest Steppe of Ukraine in 2022−2024. Results. Frosts can destroy new giant miscanthus sprouts and reduce crop yield by 15.5% in the sufficient soil moisture zone and by 22.4% in the unstable soil moisture zone. Additionally, under the conditions of unstable soil moisture, yield formation of giant miscanthus is limited by insufficient soil moisture, as miscanthus plants can efficiently use only 20 kg/ha of nitrogen fertiliser. In this case, the yield of giant miscanthus can be predicted using the equation: y = 15.19 + 0.29X – 0.005X². Meanwhile, under the conditions of sufficient soil moisture, small doses of fertilisers do not limit the linear biomass growth function, and the equation for yield prediction is: y = 18.44 + 0.25X. Conclusions. For yield prediction in the specific region, it is advisable to use the regression model proposed by Vossen, which takes into account the average yield, linear time trend, and linear regression function. For a specific agroclimatic region, the potential yield of bioenergy crops could be determined by a formula that considers the total land area of the region, the share of bioenergy plantations, and the predicted yield for the specific conditions of the region. Yield prediction model for bioenergy crop consists of the following levels: (1) determining the areas of bioenergy plantations in the GIS map of the region, specifying the plantation age and the varieties used; (2) refining environmental conditions (soil type, moisture regime, and mineral nutrition); (3) weather observation; (4) vegetation index observation; and (5) yield modelling.