Machine learning models for predicting spring wheat yield using vegetation indices

The study explores the application of vegetation indices derived from remote sensing for forecasting spring wheat yield using machine learning methods. The test site was a field experiment on intensive spring wheat cultivation conducted from 2019 to 2023 at the plot located in the Novosibirsk region. The soil type is leached chernozem. Annual Sentinel-2 satellite images with a spatial resolution of 10 meters per pixel for the June-July period were used. A preliminary analysis of the relationship between vegetation indices and wheat yield using four methods revealed that the most significant indices, with the highest integral importance values, were NDVI (0.63), GCI (0.47), RECI, NGRDI, and NDRE (0.40). To develop yield forecasting models, Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Support Vector Regression (SVR) were used. The models used three datasets: the full set (21 vegetation indices), a selected subset (5 vegetation indices), and a single-index set (NDVI only). The best results were achieved with the XGBoost model, which demonstrated high efficiency (R² = 0.75) for both the five-index subset and the NDVI-only dataset, with the latter achieving the lowest mean absolute error (MAE = 0.24 t/ha). The RF model also performed well on reduced datasets (R² = 0.75 with five indices and R² = 0.70 with NDVI. The SVR model showed a significant decline in performance as the number of features decreased (from R² = 0.74 to R² = 0.64). The study results have practical significance for optimizing remote crop monitoring, demonstrating the feasibility of effective yield forecasting using a minimal set of spectral data.

1. Khanal S., Kc K., Fulton J.P., Shearer S., Ozkan E. Remote Sensing in Agriculture – Accomplishments, Limitations, and Opportunities. Remote Sensing, 2020, vol. 12, no. 22, р. 3783. DOI: 10.3390/rs12223783.

2. Weiss M., Jacob F., Duveiller G. Remote sensing for agricultural applications: A meta-review. Remote Sensing of Environment, 2020, vol. 236, р. 111402. DOI: 10.1016/j.rse.2019.111402.

3. Denisov P.V., Sereda I.I., Troshko K.A., Lupyan E.A., Plotnikov D.E., Tolpin V.A. The state of grain crops in the European part of Russia and Siberia in June 2021 based on remote sensing data. Sovremennyye problemy distantsionnogo zondirovaniya Zemli iz kosmosa = Current problems in remote sensing of the Earth from space, 2021, vol. 18, no. 2, р. 171. (In Russian). DOI: 10.21046/2070-7401-2021-18-2-171-185.

4. Yakushev V.P., Yakushev V.V., Blokhina S.Yu., Blokhin Yu.I., Matveenko D.A. Prospects for revealing identification indicators of crop canopy based on aerospace imagery and field precision experimentation. Sovremennyye problemy distantsionnogo zondirovaniya Zemli iz kosmosa = Current problems in remote sensing of the Earth from space, 2022. vol. 19, no. 4, рр. 113–127. (In Russian). DOI: 10.21046/2070-7401-2022-19-4-113-127.

5. Kurbanov R.K., Zakharova N.I. Application of vegetation indexes to assess the condition of crops. Sel'skokhozyaystvennyye mashiny i tekhnologii = Agricultural Machinery and Technologies, 2020, vol. 14, no. 4, рр. 4–11. (In Russian). DOI: 10.22314/2073-7599-2020-14-4-4-11.

6. Hatfield J.L., Prueger J.H., Sauer T.J., Dold C., O’brien P., Wacha K. Applications of Vegetative Indices from Remote Sensing to Agriculture: Past and Future. Inventions, 2019, vol. 4, р. 71. DOI: 10.3390/inventions4040071.

7. Stepanov A.S., Aseyeva T.A., Dubrovin K.N. The influence of climatic characteristics and values of NDVI at soybean yield (on the example of the districts of the Primorskiy region). Agrarnyy vestnik Urala = Agrarian Bulletin of the Urals, 2020, no. 1 (192), рр. 10–19. (In Russian). DOI: 10.32417/1997-4868-2020-192-1-10-19.

8. Storchak I.G., Yeroshenko F.V., Oganyan L.R, Shestakova E.O., Kalashnikova A.A. Assessment of winter wheat plant development during the seeding and tillering stages according to the Earth remote sensing data. Inzhenernyye tekhnologii i sistemy = Engineering Technologies and Systems, 2021, vol. 31, no. 1, рр. 21–36. (In Russian). DOI: 10.15507/2658-4123.031.202101.021-036.

9. Rodimtsev S.A., Pavlovskaya N.Ye., Vershi nen S.V., Gor'kova I.V., Gagarina I.N. The use of the vegetative index NDVI to predict grain crop yields. Vestnik NGAU = Bulletin of NSAU, 2023, no. 4, рр. 56–67. (In Russian). DOI: 10.31677/2072-6724-2022-65-4-56-67.

10. Radočaj D., Šiljeg A., Marinović R., Jurišić M. State of major vegetation indices in precision agriculture studies indexed in web of science: A review. Agriculture, 2023, vol. 13, no. 3, p. 707. DOI: 10.3390/agriculture13030707.

11. Xue J., Su B. Significant Remote Sensing Vege tation Indices: A Review of Developments and Applications. Sensors and Sensor, 2017, article ID: 1353691. DOI: 10.1155/2017/1353691.

12. Ali A., Martelli R., Lupia F., Barbanti L. Assessing multiple years’ spatial variability of crop yields using satellite vegetation indices. Remote sensing, 2019, vol. 11, no. 20, р. 2384. DOI: 10.3390/rs11202384.

13. Yang M., Hassan M.A., Xu K., Zheng C., Rasheed A., Zhang Y., Jin X., Xia X., Xiao Y., He Z. Assessment of water and nitrogen use efficiencies through UAV-based multispectral phenotyping in winter wheat. Frontiers in plant science, 2020, vol. 11, р. 927. DOI: 10.3389/fpls.2020.00927.

14. Liu Y., Sun L., Liu B., Wu Y., Ma J., Zhang W., Wang B., Chen Z. Estimation of Winter Wheat Yield using multiple temporal vegetation indices derived from UAV-Based multispectral and hyperspectral imagery. Remote Sensing, 2023, vol. 15, no. 19, р. 4800. DOI: 10.3390/rs15194800.

15. Vallentin C., Harfenmeister K., Itzerott S., Kleinschmit B., Conrad C., Spengler D. Sui tability of satellite remote sensing data for yield estimation in northeast Germany. Precision Agriculture, 2022, vol. 23, no. 1, pp. 52–82. DOI: 10.1007/s11119-021-09827-6.

16. Vidican R., Mălinaș A., Ranta O., Moldovan C., Marian O., Ghețe A., Ghișe C.R., Popovici F., Cătunescu G.M. Using remote sensing vegetation indices for the discrimination and mo nitoring of agricultural crops: a critical review. Agronomy, 2023, vol. 13, no. 12, р. 3040. DOI: 10.3390/agronomy13123040.

17. Haque M.A., Reza M.N., Ali M., Karim M.R., Ahmed S., Lee K.D., Khang Y.H., Chung S.O. Effects of environmental conditions on vegetation indices from multispectral images: A review. Korean Journal of Remote Sensing, 2024, vol. 40, no. 4, pp. 319–341. DOI: 10.7780/kjrs.2024.40.4.1.

18. Rose M., Rose T., Kage H. Spectral reflection and crop parameters: can the disentanglement of primary and secondary traits lead to more robust and extensible prediction models? Precision Agriculture, 2023, vol. 24, pp. 607–626. DOI: 10.1007/s11119-022-09961-9.

19. Cheng E., Zhang B., Peng D. Zhong L., Yu L., Liu Y., Xiao C., Li C., Li X., Chen Y., Ye H., Wang H., Yu R., Hu J., Yang S. Wheat yield estimation using remote sensing data based on machine learning approaches. Frontiers in Plant Science, 2022, vol. 13, Р. 1090970. DOI: 10.3389/fpls.2022.1090970.

20. Uribeetxebarria A., Castellón A., Aizpurua A. Optimizing wheat yield prediction integrating data from Sentinel-1 and Sentinel-2 with Cat-Boost algorithm. Remote Sensing, 2023, vol. 15, no. 6, p. 1640. DOI: 10.3390/rs15061640.

21. Zhang H., Zhang Y., Liu K., Lan S., Gao T., Li M. Winter wheat yield prediction using integrated Landsat 8 and Sentinel-2 vegetation index time-series data and machine learning algorithms. Computers and Electronics in Agriculture, 2023, vol. 213, р. 108250. DOI: 10.1016/j.compag.2023.108250.

22. Burdett H., Wellen C. Statistical and machine learning methods for crop yield prediction in the context of precision agriculture. Precision agriculture, 2022, vol. 23, no. 5, pp. 1553–1574. DOI: 10.1007/s11119-022-09897-0.

23. Bali N., Singla A. Emerging trends in machine learning to predict crop yield and study its influential factors: A survey. Archives of computational methods in engineering, 2022, vol. 29, no. 1, pp. 95–112. DOI: 10.1007/s11831-021-09569-8.

24. Han J., Zhang Z., Cao J., Luo Y., Zhang L., Li Z., Zhang J. Prediction of Winter Wheat Yield Based on Multi-Source Data and Machine Learning in China. Remote Sensing, 2020, vol. 12, no. 2, р. 236. DOI: 10.3390/rs12020236.

25. Bian C., Shi H., Wu S., Zhang K., Wei M., Zhao Y., Sun Y., Zhuang H., Zhang X., Chen S. Prediction of field-scale wheat yield using machine learning method and multi-spectral UAV data. Remote Sensing, 2022, vol. 14, no. 6, p. 1474. DOI: 10.3390/rs14061474.

26. Omia E., Bae H., Park E., Kim M.S., Baek I., Kabenge I., Cho B.K. Remote Sensing in Field Crop Monitoring: A Comprehensive Review of Sensor Systems, Data Analyses and Recent Advances. Remote Sensing, 2023, vol. 15, no. 2, р. 354. DOI: 10.3390/rs15020354.

27. Skobalski J., Sagan V., Alifu H., Akkad O.Al., Lopes F.A., Grignola F. Bridging the gap between crop breeding and GeoAI: Soybean yield prediction from multispectral UAV images with transfer learning. ISPRS Journal of Photogrammetry and Remote Sensing, 2024, vol. 210, pp. 260–281. DOI: 10.1016/j.isprsjprs.2024.03.015.

28. Xu H., Xu D., Chen S., Ma W., Shi Z. Rapid Determination of Soil Class Based on Visible-Near Infrared, Mid-Infrared Spectroscopy and Data Fusion. Remote Sens, 2020, vol. 12, р. 1512. DOI: 10.3390/rs12091512.

29. Liu S., Hu Z., Han J., Li Y., Zhou T. Predicting grain yield and protein content of winter wheat at different growth stages by hyperspect ral data integrated with growth monitor index. Computers and Electronics in Agriculture, 2022, vol. 200, р. 107235. DOI: 10.1016/j.compag.2022.107235.

30. Speiser J.L., Miller M.E., Tooze J., Ip E. Comparison of Random Forest Variable Selection Methods for Classification Prediction Modeling. Expert Syst Appl, 2019, vol. 15, no. 134, р. 93– 101. DOI: 10.1016/j.eswa.2019.05.028.

31. Liu H., Ong Y.S., Shen X., Cai J. When Gaussian process meets big data: a review of scalable GPs. IEEE Transactions on Neural Networks and Learning Systems, 2020, vol. 31, no. 11, pp. 4405–4423. DOI: 10.1109/TNNLS.2019.2957109.

32. Shi G., Du X., Du M., Li Q., Tian X., Ren Y., Zhang Y., Wang H. Cotton Yield Estimation Using the Remotely Sensed Cotton Boll Index from UAV Images. Drones, 2022, vol. 6, р. 254. DOI: 10.3390/drones6090254.

33. Fan J., Zheng J., Wu L., Zhang F. Estimation of daily maize transpiration using support vector machines, extreme gradient boosting, artificial and deep neural networks models. Agricultural Water Management, 2021, vol. 245, р. 106547. DOI: 10.1016/j.agwat.2020.106547.

34. Zhang Y., Liu J., Shen W. A review of ensemble learning algorithms used in remote sensing applications. Applied Sciences, 2022, vol. 12, no. 17, р. 8654. DOI: 10.3390/app12178654.