A METHOD FOR ASSESSING VINEYARD PRODUCTIVITY BASED ON SPECTRAL INDICES OF SOIL PATTERNS
DOI 10.12737/2782-490X-2025-4-4-30-39 УДК 634.8:631.4
Keywords:
remote sensing, vineyard, soil patterns, spectral indices, productivity, Sentinel, machine learning, precision farming, terroir, monitoringAbstract
This article presents a method for remotely assessing of vineyard agroecosystem productivity based on the
analysis of stable spectral patterns of the soil and vegetation cover. The study was conducted in vineyards of Anapa district
of Krasnodar Krai using time series of multispectral data from Sentinel-2 satellites. The key objective was to develop an
automated approach to identifying soil patterns reflecting terroir heterogeneity and establishing their relationship with
actual yield. The methodology includes a comprehensive analysis of vegetation indices (NDVI, SAVI, NDWI, ELI) during
key phenological phases, their statistical correlation with soil agrochemical properties (soil organic matter content, mobile
phosphorus and potassium, pH, carbonate levels) and meteorological factors. Machine learning algorithms were used to
classify image pixels, identifying three stable types of spectral patterns corresponding to zones of high, medium and low
productive potential. The results showed a strong dependence of crop yield on soil conditions and moisture availability,
expressed through spectral indices. The highest productivity is associated with a pattern characterized by elevated NDVI
(>0.55) and SAVI (>0.8) values, as well as moderate water stress NDWI (>-0.5). A regression equation for crop yield
forecasting based on these indices was developed (R² = 0.86). The accuracy of automated soil pattern extraction was
85-90% (kappa 0.78-0.84). This method enables precision vineyard management by optimizing agronomic measures differentially for each identified microzone, which is the basis for increasing the efficiency and sustainability of viticulture.
For citation: A method for assessing vineyard productivity based on spectral indices of soil patterns / Orlov V.A.,
Lukyanov A.A., Mikhaylovskaya O.I., Mikhaylovskiy S.S. // Agrobiotechnologies and digital farming. 2025; 4 (16).
30-39 p. http://dx.doi.org/10.12737/2782-490X-2025-4-4-30-39
References
Литература
Орлов В.А., Лукьянов А.А., Михайловская О.И. Определение морфометрических показателей почвенной поверхности виноградного насаждения по спектральным каналам спутниковых изображений // Аграрная наука. 2024. № 10. С. 159-164. doi: 10.32634/0869-8155-2024-387-10-159-164.
Орлов В.А., Лукьянов А.А. Оценочные признаки виноградопригодных земель по спектральным паттернам // Вестник Казанского государственного аграрного университета. 2023. Т. 1. С. 69.
Савин И.Ю., Савенкова Е.В., Кучер Д.Е. Оценка контрастности почвенного покрова пахотных угодий
по спутниковым данным Sentinel-2 // Почвоведение. 2021. № 11. С. 1295-1305. doi: 10.31857/
S0032180X21110125.
Прудникова Е.Ю., Савин И.Ю. Спутниковая оценка дегумификации пахотных почв в Саратовском
Поволжье // Почвоведение. 2015. № 5. С. 597. doi: 10.7868/S0032180X1505007X.
Abdi A., Khormali F., Emamian V. Soil properties estimation using Sentinel-2 satellite data and machine
learning algorithms: A case study in vineyard terrains // Geoderma. 2020. Vol. 365. P. 114178. doi: 10.1016/
j.geoderma.2020.114178.
Тимиргалеева Р.Р., Гришин И.Ю., Рыбалко Е.А. Анализ методов дистанционной диагностики плодородия виноградных агроценозов // Дистанционные образовательные технологии: Материалы IV Всероссийской научно-практической конференции (с международным участием), Посвящается 75-летию ГПА, Ялта,
–21 сентября 2019 года / Под ред. В.Н. Тарана. Ялта: ООО «Издательство Типография «Ариал», 2019. С.
-338.
Becker S. J. et al. Bare ground classification using a spectral index ensemble and machine learning models
optimized across 12 international study sites //Geocarto International. 2025. Т. 40. №. 1. С. 2465452.
Simón Sánchez A.M. et al. Convolutional neural networks for agricultural land use classification from Sentinel-2 image time series // Remote Sensing. 2022. Vol. 14. № 21. P. 5373. doi: 10.3390/rs14215373.
Thaler E.A., Larsen I.J., Yu Q. A new index for remote sensing of soil organic carbon based solely on visible
wavelengths // Soil Science Society of America Journal. 2019. Vol. 83. № 5. P. 1443-1450. doi: 10.2136/
sssaj2018.09.0318.
Eslava-Lecumberri F.J., Jiménez-Ballesta R. Delineating vineyard management zones: Intrafield spatial
variability of soil properties of carbonate vineyard soils // European Journal of Soil Science. 2024. Vol. 75. № 6. P.
e70029. doi: 10.1111/ejss.70029.
Павлова А.И. Применение вегетационных индексов для цифрового почвенного картографирования
на основе космических снимков Sentinel-2 // Siberian Journal of Life Sciences and Agriculture. 2021. Т. 13, № 6.
С. 119-131. doi: 10.12731/2658-6649-2021-13-6-119-131.
de Campos Assunção J.M.S. Terrain classification using machine learning algorithms in a multi-temporal
approach A QGIS plug-in implementation: Universidade NOVA de Lisboa, 2021. doi: 10.3390/rs14153802.
Ramos T.B. et al. Soil salinity assessment using vegetation indices derived from Sentinel-2 multispectral
data. application to Lezíria Grande, Portugal // Agricultural Water Management. 2020. Vol. 241. P. 106387. doi:
1016/j.agwat.2020.106387.
Abdi A., Khormali F., Emamian V. Soil properties estimation using Sentinel-2 satellite data and machine
learning algorithms: A case study in vineyard terrains // Geoderma. 2020. Vol. 365. P. 114178. doi: 10.1016/
j.geoderma.2020.114178.
Виноградарство России. 2022 Электронный ресурс. https://geonovosti.terratech.ru/upload/
iblock/653/20o2jr9s6ofi590on9tbcua3rl4w1h09/Vinogradarstvo-Rossii.pdf /дата обращения 18.03.2025/
Eslava-Lecumberri F.J., Jiménez-Ballesta R. Delineating vineyard management zones: Intrafield spatial
variability of soil properties of carbonate vineyard soils // European Journal of Soil Science. 2024. Vol. 75. № 6. P.
e70029. doi: 10.1111/ejss.70029.
Zeyada A.M. et al. Sentinel-2 Satellite Imagery Application to Monitor Soil Salinity and Calcium Carbonate
Contents in Agricultural Fields // Phyton. 2023. Vol. 92. № 5. doi: 10.32604/phyton.2023.027267.
Gomez C. et al. Topsoil clay content mapping in croplands from Sentinel-2 data: Influence of atmospheric
correction methods across a season time series // Geoderma. 2022. Vol. 423. P. 115959. doi: 10.1016/
j.geoderma.2022.115959.
Crespo N. et al. Satellite remote sensing tools for drought assessment in vineyards and olive orchards: A
systematic review // Remote Sensing. 2024. Vol. 16. № 11. P. 2040. doi: 10.3390/rs16112040.
Lasko K., O’Neill F. D., Sava E. Automated mapping of land cover type within international heterogenous
landscapes using sentinel-2 imagery with ancillary geospatial data //Sensors. 2024. Т. 24. №. 5. С. 1587.
doi: 10.3390/s24051587
Ammoniaci M. et al. State of the art of monitoring technologies and data processing for precision viticulture // Agriculture. 2021. Vol. 11. № 3. P. 201. doi: 10.3390/agriculture11030201.
References
Orlov VA, Lukyanov AA, Mikhaylovskaya OI. [Determination of morphometric indicators of the soil surface of
vineyard plantations using spectral channels of satellite images]. Agrarnaya nauka. 2024; 10. 159-164 p. doi:
32634/0869-8155-2024-387-10-159-164.
Orlov VA, Lukyanov AA. [Evaluative characteristics of vine-suitable lands based on spectral patterns]. Vestnik
Kazanskogo gosudarstvennogo agrarnogo universiteta. 2023; Vol.1. 69 p.
Savin IYu, Savenkova EV, Kucher DE. [Assessment of soil cover contrast of arable lands using Sentinel-2 satellite
data]. Pochvovedenie. 2021; 11. 1295-1305 p. doi: 10.31857/S0032180X21110125.
Prudnikova EYu, Savin IYu. [Satellite assessment of arable soil dehumification in Saratov Volga region].
Pochvovedenie. 2015; 5. 597 p. doi: 10.7868/S0032180X1505007X.
Abdi A, Khormali F, Emamian V. Soil properties estimation using Sentinel-2 satellite data and machine learning
algorithms: A case study in vineyard terrains. Geoderma. 2020; Vol.365. 114178 p. doi: 10.1016/j.geoderma.2020.114178.
Timirgaleeva RR, Grishin IYu, Rybalko EA. [Analysis of remote diagnostics methods of fertility grape agrocenoses.
Distance educational technologies]. Materialy IV Vserossiyskoy nauchno-prakticheskoy konferentsii (s mezhdunarodnym
uchastiem), posvyashchaetsya 75-letiyu GPA, Yalta, 16-21 sentyabrya 2019 goda. Editorship V.N. Taran. Yalta: OOO
Izdatelstvo Tipografiya “Arial”. 2019; 331-338 p.
Becker SJ. Bare ground classification using a spectral index ensemble and machine learning models optimized
across 12 international study sites. Geocarto International. 2025; Vol.40. 1. 2465452 p.
Simon Sanchez AM. Convolutional neural networks for agricultural land use classification from Sentinel-2 image
time series. Remote Sensing. 2022; Vol.14. 21. 5373 p. doi: 10.3390/rs14215373.
Thaler EA, Larsen IJ, Yu Q. A new index for remote sensing of soil organic carbon based solely on visible wavelengths. Soil Science Society of America Journal. 2019; Vol.83. 5. 1443-1450 p. doi: 10.2136/sssaj2018.09.0318.
Eslava-Lecumberri FJ, Jimenez-Ballesta R. Delineating vineyard management zones: Intrafield spatial variability
of soil properties of carbonate vineyard soils. European Journal of Soil Science. 2024; Vol.75. 6. e70029 p. doi: 10.1111/
ejss.70029.
Pavlova AI. [Application of vegetation indices for digital soil mapping based on Sentinel-2 satellite images]. Siberian Journal of Life Sciences and Agriculture. 2021; Vol.13. 6. 119-131 p. doi: 10.12731/2658-6649-2021-13-6-119-131.
de Campos Assuncao JMS. Terrain classification using machine learning algorithms in a multi-temporal approach
A QGIS plug-in implementation: Universidade NOVA de Lisboa. 2021; doi: 10.3390/rs14153802.
Ramos TB. Soil salinity assessment using vegetation indices derived from Sentinel-2 multispectral data. Application to Leziria Grande, Portugal. Agricultural Water Management. 2020; Vol.241. 106387 p. doi: 10.1016/
j.agwat.2020.106387.
Abdi A, Khormali F, Emamian V. Soil properties estimation using Sentinel-2 satellite data and machine learning
algorithms: A case study in vineyard terrains. Geoderma. 2020; Vol.365. 114178 p. doi: 10.1016/j.geoderma.2020.114178.
Viticulture in Russia. 2022. [Internet]. Geonovosti. [cited 2025, March 18]. Available from: https://
geonovosti.terratech.ru/upload/iblock/653/20o2jr9s6ofi590on9tbcua3rl4w1h09/Vinogradarstvo-Rossii.pdf
Eslava-Lecumberri FJ, Jimenez-Ballesta R. Delineating vineyard management zones: Intrafield spatial variability
of soil properties of carbonate vineyard soils. European Journal of Soil Science. 2024; Vol.75. 6. e70029 p. doi: 10.1111/
ejss.70029.
Zeyada AM. Sentinel-2 satellite imagery application to monitor soil salinity and calcium carbonate contents in
agricultural fields. Phyton. 2023; Vol.92. 5. doi: 10.32604/phyton.2023.027267.
Gomez C. Topsoil clay content mapping in croplands from Sentinel-2 data: Influence of atmospheric correction
methods across a season time series. Geoderma. 2022; Vol.423. 115959 p. doi: 10.1016/j.geoderma.2022.115959.
Crespo N. Satellite remote sensing tools for drought assessment in vineyards and olive orchards: A systematic
review. Remote Sensing. 2024; Vol.16. 11. 2040 p. doi: 10.3390/rs16112040.
Lasko K, O’Neill FD, Sava E. Automated mapping of land cover type within international heterogenous landscapes using Sentinel-2 imagery with ancillary geospatial data. Sensors. 2024; Vol.24. 5. 1587 p. doi: 10.3390/s24051587
Ammoniaci M. State of the art of monitoring technologies and data processing for precision viticulture. Agriculture. 2021; Vol.11. 3. 201 p. doi: 10.3390/agriculture11030201.
