Agroecological modeling of water conditions in fallow fields using remote sensing data
Abstract
Global warming has intensified the ecological challenge of moisture deficit in agriculture, making the dynamic monitoring of water resources essential for sustainable crop production. Traditional on-land surveys are costly and time-in-tensive, highlighting the need for cost-effective remote sensing solutions.
Purpose. This study aimed to determine the suitability of remotely sensed indices – the Normalized Difference Water Index (NDWI) and the Soil Moisture Index (SMI) – for the dynamic control of Moisture Accumulation (MA) in fallow fields within a semi-arid climate zone.
Methods. Trials were conducted in 2025 at the “Vostok” experimental farm in the Kherson region. The soil of the experimental fields was represented by typical dark-chestnut slightly saline soil. Climatically, the area of the study conduction belongs to the semi-arid Steppe zone. Satellite imagery from the OneSoil platform provided the required indices with corresponding uniform spatial resolution of 10 m from Sentinel-2 satellite, while a meteorological gauge measured MA in the studied fields. Regression analysis was performed on 500 data pairs (250 for “NDWI-MA” and 250 for “SMI-MA”) and 103 complex data pairs (“NDWI-SMI-MA”) by the ordinary least square (OLS) algorithm. The regression performance was assessed with Pearson’s correlation coefficient (r), coefficient of determination (R²), mean square error (MSE), and mean absolute percentage error (MAPE). Besides, clustering was performed by the K-means algorithm.
Results. The results indicate that NDWI exhibits a significantly higher correlation (0.9160 vs. 0.6884) and stronger regression relationship with MA than SMI. Consequently, NDWI is preferred for dynamic monitoring in these environments. The combined “NDWI-SMI-MA” model provided the best overall performance for estimating MA with the least MAPE of 26.02%. Cluster analysis successfully distinguished three major moisture groupings, revealing that most fallow fields belonged to a “dry cluster,” indicating a severe deficit of soil moisture content.
Conclusions. Based on the results, the NDWI is a better spatial index for soil moisture accumulation assessment. The SMI, while showing a positive relationship with soil moisture, is less suitable for accurate quantitative modeling. The best performance is achieved using combined NDWI-SMI model for soil moisture prediction. Though convincing, it is necessary to emphasize that these models should be used cautiously in practice, as they still lack the robustness built on the longevity of observations and bigger data analysis. Further research will be aimed to enlarge the dataset including more edge cases in the data, collecting data for multiple years so that more robust and reliable models of soil moisture accumulation could be developed.
References
2. Ingrao C., Strippoli R., Lagioia G., Huisingh D. Water scarcity in agriculture: An overview of causes, impacts and approaches for reducing the risks. Heliyon. 2023. Vol. 9 (8). P. e18507. DOI: 10.1016/j.heliyon.2023.e18507
3. Sehler R., Li J., Reager J., Ye H. Investigating relationship between soil moisture and precipitation globally using remote sensing observations. Journal of Contemporary Water Research & Education. 2019. Vol. 168. P. 106–118. DOI: 10.1111/j.1936-704X.2019.03324.x
4. Gaona J., Quintana-Seguí P., Escorihuela M., Boone A., Llasat M. Interactions between precipitation, evapotranspiration and soil-moisture-based indices to characterize drought with high-resolution remote sensing and land-surface model data. Natural Hazards and Earth System Sciences. 2022. Vol. 22. P. 3461–3478. DOI: 10.5194/nhess-22-3461-2022
5. Chiaravalloti F., Brocca L., Procopio A., Massari C., Gabriele S. Assessment of GPM and SM2RAIN-ASCAT rainfall products over complex terrain in southern Italy. Atmospheric Research. 2018. Vol. 205. P. 151–162. DOI: 10.1016/J.ATMOSRES.2018.02.019
6. Akbar R., Gianotti D., Salvucci G., Entekhabi D. Mapped hydroclimatology of evapotranspiration and drainage runoff using SMAP brightness temperature observations and precipitation information. Water Resources Research. 2019. Vol. 55. P. 3391–3413. DOI: 10.1029/2018WR024459
7. Iamampai S., Talaluxmana Y., Kanasut J., Rangsiwanichpong P. Enhancing rainfall–runoff model accuracy with machine learning models by using soil water index to reflect runoff characteristics. Water Science and Technology. 2024. Vol. 89 (2). P. 368–381. DOI: 10.2166/wst.2023.424
8. Latif S., Hazrin N., Koo C., Ng J., Chaplot B., Huang Y., El-Shafie A., Ahmed A. Assessing rainfall prediction models: Exploring the advantages of machine learning and remote sensing approaches. Alexandria Engineering Journal. 2023. Vol. 85. P. 387–400. DOI: 10.1016/j.aej.2023.09.060
9. Peng J., Loew A., Merlin O., Verhoest N. A review of spatial downscaling of satellite remotely sensed soil moisture. Reviews of Geophysics. 2017. Vol. 55. P. 341–366. DOI: 10.1002/2016RG000543
10. Lykhovyd P. V. The use of spatial normalized difference vegetation index for determination of humus content in the soils of southern Ukraine. Ecological Engineering & Environmental Technology. 2023. Vol. 24 (4). P. 223–228. DOI: 10.12912/27197050/162698
11. Yang X., Zhao S., Qin X., Zhao N., Liang L. Mapping of urban surface water bodies from Sentinel-2 MSI imagery at 10 m resolution via NDWI-based image sharpening. Remote Sensing. 2017. Vol. 9 (6). P. 596. DOI: 10.3390/rs9060596
12. Saha A., Patil M., Goyal V. C., Rathore D. S. Assessment and impact of soil moisture index in agricultural drought estimation using remote sensing and GIS techniques. Proceedings. 2019. Vol. 7 (1). P. 2. DOI: 10.3390/ECWS-3-05802
13. Burton A. L. OLS (linear) regression. In: The Encyclopedia of Research Methods in Criminology and Criminal Justice. Wiley-Blackwell, 2021. P. 509–514. DOI: 10.1002/9781119111931.ch104
14. Tatachar A. V. Comparative assessment of regression models based on model evaluation metrics. International Research Journal of Engineering and Technology. 2021. Vol. 8 (9). P. 853–860.
15. Sinaga K. P., Yang M. S. Unsupervised K-means clustering algorithm. IEEE Access. 2020. Vol. 8. P. 80716–80727. DOI: 10.1109/ACCESS.2020.2988796
16. Moreno J. J. M., Pol A. P., Abad A. S., Blasco B. C. Using the R-MAPE index as a resistant measure of forecast accuracy. Psicothema. 2013. Vol. 25 (4). P. 500–506. DOI: 10.7334/psicothema2013.23
17. McFeeters S. K. Using the normalized difference water index (NDWI) within a geographic information system to detect swimming pools for mosquito abatement: A practical approach. Remote Sensing. 2013. Vol. 5 (7). P. 3544–3561. DOI: 10.3390/rs5073544
18. Lykhovyd P. V. Irrigation needs in Ukraine according to current aridity level. Journal of Ecological Engineering. 2021. Vol. 22 (8). P. 11–18. DOI: 10.12911/22998993/140478
19. Ainiwaer M., Ding J., Kasim N., Wang J., Wang J. Regional scale soil moisture content estimation based on multi-source remote sensing parameters. International Journal of Remote Sensing. 2020. Vol. 41. P. 3346–3367. DOI: 10.1080/01431161.2019.1701723
20. Koohikeradeh E., Gumiere S. J., Bonakdari H. NDMI-derived field-scale soil moisture prediction using ERA5 and LSTM for precision agriculture. Sustainability. 2025. Vol. 17 (6). P. 2399. DOI: 10.3390/su17062399
21. Adab H., Morbidelli R., Saltalippi C., Moradian M., Ghalhari G. A. F. Machine learning to estimate surface soil moisture from remote sensing data. Water. 2020. Vol. 12 (11). P. 3223. DOI: 10.3390/w12113223
22. Das B., Rathore P., Roy D., Chakraborty D., Jatav R. S., Sethi D., Kumar P. Comparison of bagging, boosting and stacking algorithms for surface soil moisture mapping using optical–thermal–microwave remote sensing synergies. CATENA. 2022. Vol. 217. P. 106485. DOI: 10.1016/j.catena.2022.106485
23. John J., Jaganathan R., Shylesh D. D. Mapping of soil moisture index using optical and thermal remote sensing. In: International Conference on Structural Engineering and Construction Management. Springer International Publishing, 2021. P. 759–767. DOI: 10.1007/978-3-030-80312-4_65
24. Liu Z., Xia Z., Chen F., Hu Y., Wen Y., Liu J., Liu H., Liu L. Soil moisture index model for retrieving soil moisture in semiarid regions of China. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2020. Vol. 13. P. 5929–5937. DOI: 10.1109/JSTARS.2020.3025596
25. Babaeian E., Sadeghi M., Jones S., Montzka C., Vereecken H., Tuller M. Ground, proximal, and satellite remote sensing of soil moisture. Reviews of Geophysics. 2019. Vol. 57. P. 530–616. DOI: 10.1029/2018RG000618
26. Celik M., Isik M., Yuzugullu O., Fajraoui N., Erten E. Soil moisture prediction from remote sensing images coupled with climate, soil texture and topography via deep learning. Remote Sensing. 2022. Vol. 14. P. 5584. DOI: 10.3390/rs14215584
This work is licensed under a Creative Commons Attribution 4.0 International License.
