UDC 10.36461/NP.2025.75.3.021
DOI 633.11+556.3.06;519.237.5;528.88

FORECASTING WINTER WHEAT YIELD BASED ON AGROCLIMATIC CONDITIONS AND NDVI IN THE PENZA REGION
A.S. Shcherbakov, Senior Lecturer, S.V. Bogomazov, Candidate of Agricultural Sciences, Associate Professor
“Penza State Agrarian University, Penza, Russia е–mail: This email address is being protected from spambots. You need JavaScript enabled to view it.

This article deals with statistical correlation between winter wheat yield, meteoconditions, and the Normalized Difference Vegetation Index (NDVI) to develop a methodological guide for forecasting yields and assessing winter wheat growing over time. The research is based on a nine–year data analysis (2015–2023) using statistical methods, including cluster and correlation–regression analysis. During the research 27 municipal districts of the Penza region were clustered, and then divided into four groups. Multiple linear regression equations with a determination coefficient (R²) ranging from 0.71 to 0.85 were done for each cluster. The model consists of five key variables, including precipitation at different growing seasons, thermal conditions, and max NDVI. Model verification showed the mean absolute error (MAE) of 2.7 c/ha and the mean absolute percentage error (MAPE) of 9.2%. The accuracy of the general yield forecasting model was 97.0% (error of 3.0%). The observed models made possible to forecast yields for 2024, with a range of 21.3 to 41.0 c/ha by the regional districts and the average yield of 31.9 c/ha for the region. The results of the current research can be used by agricultural authorities for forecasting and risk assessment in crop production.
Keywords: winter wheat, yield, forecasting, remote sensing, NDVI, regression model, cluster analysis, agroclimatic conditions.

Publication information View article