On February 1, 2022, Sofia Drozd, a student of the Institute of Physics and Technology of NTUU «KPI», an active participant in projects of the National Research Fund, World Bank and e-shape, made a report «Forecasting the yield of major crops in Ukraine» at the international conference Virtual Assembly.
As you know, the agricultural sector has a strong impact on Ukraine’s financial situation, and crop failures are hitting the economy, causing protracted financial crises. Fortunately, modern mathematical methods can predict yields. In particular, satellite sources play an important role in forecasting, from which information on soil properties can be obtained. For example, the NDVI vegetation index.
In her work, Sofia presented the results of correlation and regression analyzes to predict land productivity for the main spring crops of Ukraine in Kyiv region based on the vegetation index obtained by remote sensing and weather data for spring and summer vegetation.
In addition to presenting the results of the analysis, the program implementation of the study was also demonstrated and the work of the written program on real data in real time was shown. NDVI vegetation index and weather conditions were provided at the entrance, and annual projected land productivity based on two models, based on the NDVI vegetation index and a combination of vegetation index, appeared in a few minutes. The most influential monthly temperatures and precipitations were determined for each crop. Overall, the results were good. The error of the model based only on NDVI for each culture did not exceed 32%, and the combined model – 10%.
In general, Sofia noted that this program can be used not only for Kyiv region, but also for any region. In addition, in the future it is planned to conduct forecasts not only in large areas, but also in individual fields. Thus, using the developed program, any farmer will be able to predict the yield in his field and, accordingly, choose the best place for sowing crops. As a result, the program has the potential and can be used in practice to reduce crop risk.
We wish Sofia Drozd success at future international conferences!