Russian Federation
Russian Federation
Russian Federation
Russian Federation
UDC 632.111.51
UDC 632.111.52
UDC 004.852
The purpose of the study was to train and compare machine learning (ML) and deep learning (DL) models for long-term forecasting of minimum daily temperatures (frosts) - a key agrometeorological risk affecting crop productivity. The study was based on data from weather stations in St. Petersburg and the Leningrad region. To predict the year ahead, which aligns with the needs of operational agricultural production planning (sowing dates, harvest timing, protective measures), 8 methods were used: ForecasterAutoreg, Random Forest, Support Vector Regression (SVR), XGBoost, Convolutional Neural Network (CNN), SimpleRNN, Gated Recurrent Unit (GRU) and Long Short-Term Memory (LSTM). The analysis was carried out on two datasets: for the periods 1936-2024 and 1881-1995. The quality of the models was assessed using the metrics MAE, MSE, RMSE, R2 and adjusted R2. The LSTM model showed the most accurate results on the main dataset (1936-2024): MAE 2.9, MSE 14.661, R2 0.789. The SVR method (MAE 3.461, R2 0.775) demonstrated the best metrics based on archived data (1881-1995). It has been established that DL models (LSTM, GRU, CNN) generally outperform classical ML methods based on modern data. The LSTM method is recognized as the most effective for integration into precision farming systems and agricultural monitoring in the region for the advance planning of agrotechnological measures to protect crops from frost, optimization of crop rotation systems, and minimization of crop loss risks.
long-term agrometeorological forecasting, freezing, machine learning methods, deep learning methods, precision farming, agricultural management, decision-making in crop production
1. IPCC, 2012: Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change [Field, C.B., V. Barros, T.F. Stocker, D. Qin, D.J. Dokken, K.L. Ebi, M.D. Mastrandrea, K.J. Mach, G.-K. Plattner, S.K. Allen, M. Tignor, and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, UK, and New York, NY, USA, 582 pp.
2. Li X., Jiang D., Liu F. Winter soil warming exacerbates the impacts of Spring Low temperature stress on wheat // Journal of Agronomy and Crop Science. 2016. V. 202, no. 6. P. 554-563.
3. Mitrofanova O., Mitrofanov E., Blekanov I., Bure V., Molin A. Approach for long-term forecasting of frosts and droughts in smart agriculture // Agriculture Digitalization and Organic Production. ADOP 2024. Smart Innovation, Systems and Technologies, Springer, Singapore. 2024. V. 397. P. 35-46.
4. Hua W., Heinemann P., He L. Frost management in agriculture with advanced sensing, modeling, and artificial intelligent technologies: A review // Computers and Electronics in Agriculture. 2025. V. 231. Article 110027.
5. Eitzinger J., Daneu V., Kubu G., Thaler S., Trnka M., Schaumberger A., Schneider S., Tran T.M.A. Grid based monitoring and forecasting system of cropping conditions and risks by agrometeorological indicators in Austria – Agricultural Risk Information System ARIS // Climate Services. 2024. V. 34. Article 100478.
6. Harilal G.T., Dixit A., Quattrone G. Establishing hybrid deep learning models for regional daily rainfall time series forecasting in the United Kingdom // Engineering Applications of Artificial Intelligence. 2024. V. 133. Article 108581.
7. Guhan V., Raju A.D., Krishna R., Nagaratna K. Evaluating weather trends and forecasting with machine learning: Insights from maximum temperature, minimum temperature, and rainfall data in India // Dynamics of Atmospheres and Oceans. 2025. V. 110. Article 101562.
8. Hossain M.A., Rahman Md M., Hasan S.S., Mahmud A., Bai L. Analysis and forecasting of meteorogical drought using PROPHET and SARIMA models deploying machine learning technique for southwestern region of Bangladesh // Environment and Sustainability Indicators. 2025. V. 27. Article 100761.
9. Hsu C.-C., Lin Y.-P. Incorporating long-term numerical weather forecast to quantify dynamic vulnerability of irrigation supply system: A case study of Shihmen Reservoir in Taiwan // Agricultural Water Management. 2024. V. 306. Article 109178.
10. Chen L., Liu X., Zeng C., He X., Chen F., Zhu B. Temperature prediction of seasonal frozen subgrades based on CEEMDAN-LSTM hybrid model // Sensors. 2022. V. 22. Article 5742.
11. Zhang D., Ma Y., Jiang A., Ren Y., Lin J., Peng Q., Jin T. Long-term water temperature forecasting in fish spawning grounds downstream of hydropower stations using machine learning // Sustainability. 2025. V. 17. Article 4514.
