Журнал

About Editorial Board Information for authors and reviewers Publication ethics Contacts User Account: send / review a manuscript
Archive
Volume 23, 2026
Number 1 Number 2 Number 3 Number 4
Volume 22, 2025
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 21, 2024
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 20, 2023
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 19, 2022
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 18, 2021
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 17, 2020
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 16, 2019
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 15, 2018
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 14, 2017
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 13, 2016
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 12, 2015
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 11, 2014
Number 1 Number 2 Number 3 Number 4
Volume 10, 2013
Number 1 Number 2 Number 3 Number 4
Volume 9, 2012
Number 1 Number 2 Number 3 Number 4 Number 5
Volume 8, 2011
Number 1 Number 2 Number 3 Number 4
Volume 7, 2010
Number 1 Number 2 Number 3 Number 4
Issue 6, 2009
Volume 1 Volume 2
Issue 5, 2008
Volume 1 Volume 2
Issue 4, 2007
Volume 1 Volume 2
Issue 3, 2006
Volume 1 Volume 2
Issue 2, 2005
Volume 1 Volume 2
Issue 1, 2004
Volume 1
Search
Find:
русский
ISSN 2070-7401 (Print), ISSN 2411-0280 (Online)
Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa
CURRENT PROBLEMS IN REMOTE SENSING OF THE EARTH FROM SPACE

  

Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2026, V. 23, No. 4, pp. 282-292

Predicting wheat crops nutrient status using multispectral and hyperspectral remote sensing data, machine learning and spatial cross-validation

A.A. Struev 1 , E.P. Mitrofanov 1 , V.P. Yakushev 1 , V.M. Bure 1, 2 , O.A. Mitrofanova 1 
1 Saint Petersburg State University, Saint Petersburg, Russia
2 Agrophysical Research Institute, Saint Petersburg, Россия
Accepted: 16.07.2026
DOI: 10.21046/2070-7401-2026-23-4-282-292
Traditional methods for diagnosing the supply of plants with macro- and micronutrients are labor-intensive, expensive, and do not allow obtaining continuous spatial maps. The development of remote sensing technologies opens new opportunities for operational crop monitoring; however, for many elements that lack characteristic absorption bands in the visible and near-infrared ranges, the problem remains unsolved. Moreover, many existing studies use random cross-validation, ignoring spatial autocorrelation, which leads to overestimation of model accuracy. In this work, a reproducible machine learning approach is developed and validated for predicting the concentrations of 12 macro- and micronutrients in the aboveground biomass of spring wheat using multispectral (5 bands) and hyperspectral (300 bands, 390–1031 nm) data. A feature set was constructed comprising 44 broadband vegetation indices, window-based statistics, and textural descriptors characterizing the spatial heterogeneity of the canopy. For model evaluation, buffered leave-one-out cross-validation was applied, accounting for individual spatial autocorrelation radii. The approach employed mRMR feature selection, Bayesian hyperparameter optimization, and ensemble machine learning methods. It was found that under correct spatial validation all 12 nutrients in the multispectral pipeline yield positive coefficients of determination R2 in the range of 0.11–0.46. A physical separation of signal sources was revealed: for macronutrients (N, S, Ca), vegetation indices played the leading role (50–55 % importance), whereas for micronutrients and cobalt, textural features dominated (62–100 %). Comparison with the hyperspectral data showed that, despite 53–80 % of predictive information lying outside the multispectral range, at equal sample size the multispectral pipeline with texture channel outperforms the hyperspectral one without textures for 11 out of 12 elements, especially for cobalt and nitrates. Random cross-validation overestimates R2 by 0.13 on average, confirming the necessity of spatially-correct evaluation. The developed approach is promising for constructing nutrient distribution maps (especially for nitrogen, phosphorus, and potassium) suitable for variable rate fertilization, as well as for future development of hyperspectral orthomosaics, integration of textures with hyperspectral spectra, cross field validation, and use of the SWIR range for micronutrients.
Keywords: remote sensing, multispectral imagery, hyperspectral imagery, vegetation indices, textural features, machine learning, spatial cross-validation, micronutrients, spring wheat, precision agriculture
Full text

References:

  1. Mitrofanova O. A., Mitrofanov E. P., Blekanov I. S., Molin A. E., Specialized dataset of multispectral aerophotos for solving precision farming problems using artificial intelligence methods, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2023, V. 20, No. 6, pp. 67–79 (in Russian), https://doi.org/10.21046/2070-7401-2023-20-6-67-79.
  2. Mitrofanova O. A., Mitrofanov E. P., Bure V. M., Analysis of hyperspectral remote sensing data and wheat yield for the forecasting task, Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2025, V. 22, No. 1, pp. 93–105 (in Russian), https://doi.org/10.21046/2070-7401-2025-22-1-93-105.
  3. Adão T., Hruška J., Pádua L. et al., Hyperspectral imaging: A review on UAV-based sensors, data processing and applications for agriculture and forestry, Remote Sensing, 2017, V. 9, No. 11, Article 1110, https://doi.org/10.3390/rs9111110.
  4. Berger K., Verrelst J., Féret J.-B. et al., Crop nitrogen monitoring: Recent progress and principal developments in the context of imaging spectroscopy missions, Remote Sensing of Environment, 2020, V. 242, Article 111758, https://doi.org/10.1016/j.rse.2020.111758.
  5. Chancia R., Bates T., Heuvel J. V., Aardt J., Assessing grapevine nutrient status from unmanned aerial system (UAS) hyperspectral imagery, Remote Sensing, 2021, V. 13, No. 21, Article 4489, https://doi.org/10.3390/rs13214489.
  6. Friedman J., Hastie T., Tibshirani R., Sparse inverse covariance estimation with the graphical lasso, Biostatistics, 2008, V. 9, No. 3, pp. 432–441, https://doi.org/10.1093/biostatistics/kxm045.
  7. Gill A. K., Gaur S., Sneller C., Drewry D. T., Utilizing VSWIR spectroscopy for macronutrient and micronutrient profiling in winter wheat, Frontiers in Plant Science, 2024, V. 15, Article 1426077, https://doi.org/10.3389/fpls.2024.1426077.
  8. Gitelson A., Merzlyak M. N., Quantitative estimation of chlorophyll-a using reflectance spectra, J. Photochemistry and Photobiology B: Biology, 1994, V. 22, No. 3, pp. 247–252, https://doi.org/10.1016/1011-1344(93)06963-4.
  9. Hengl T., Nussbaum M., Wright M. N. et al., Random Forest as a generic framework for predictive modeling of spatial and spatio-temporal variables, PeerJ, 2018, V. 6, Article e26693v3, https://doi.org/10.7287/peerj.preprints.26693v3.
  10. Li Z., Zhou X., Cheng Q. et al., A machine-learning model based on the fusion of spectral and textural features from UAV multi-sensors to analyse the total nitrogen content in winter wheat, Remote Sensing, 2023, V. 15, No. 8, Article 2152, https://doi.org/10.3390/rs15082152.
  11. Maes W. H., Steppe K., Perspectives for remote sensing with unmanned aerial vehicles in precision agriculture, Trends in Plant Science, 2019, V. 24, No. 2, pp. 152–164, https://doi.org/10.1016/j.tplants.2018.11.007.
  12. Mashori A. S., Li F., Aman M. et al., Remote Sensing through UAVs for precision agriculture: Applications, technical foundations, current barriers, and future opportunities, Smart Agricultural Technology, 2026, V. 14, Article 102074, https://doi.org/10.1016/j.atech.2026.102074.
  13. Ploton P., Mortier F., Réjou-Méchain M. et al., Spatial validation reveals poor predictive performance of large-scale ecological mapping models, Nature Communications, 2020, V. 11, Article 4540, https://doi.org/10.1038/s41467-020-18321-y.
  14. Roberts D. R., Bahn V., Ciuti S. et al., Cross-validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure, Ecography, 2017, V. 40, No. 8, pp. 913–929, https://doi.org/10.1111/ecog.02881.
  15. Savitzky A., Golay M. J. E., Smoothing and differentiation of data by simplified least squares procedures, Analytical Chemistry, 1964, V. 36, No. 8, pp. 1627–1639, https://doi.org/10.1021/ac60214a047.
  16. Wold S., Sjöström M., Eriksson L., PLS-regression: a basic tool of chemometrics, Chemometrics and Intelligent Laboratory Systems, 2001, V. 58, No. 2, pp. 109–130, https://doi.org/10.1016/S0169-7439(01)00155-1.
  17. Xu L., Chen L., Luo Q. et al., Leveraging UAV hyperspectral imaging for crop physiology and biochemistry: A comprehensive review of feature extraction and selection methods, Plant Phenomics, 2026, V. 8, Iss. 1, Article 100141, https://doi.org/10.1016/j.plaphe.2025.100141.
  18. Xue J., Su B., Significant remote sensing vegetation indices: A review of developments and applications, J. Sensors, 2017, V. 2017, Article 1353691, https://doi.org/10.1155/2017/1353691.
  19. Yu J., Wang J., Leblon B., Song Y., Nitrogen estimation for wheat using UAV-based and satellite multispectral imagery, topographic metrics, leaf area index, plant height, soil moisture, and machine learning methods, Nitrogen, 2022, V. 3, No. 1, pp. 1–25, https://doi.org/10.3390/nitrogen3010001.