Preview

Soil Science and Agrichemistry

Soil Science and Agrichemistry

Advanced search

Optimization of cotton nutrition on saline soils of the Turkestan region using machine learning methods

https://doi.org/10.51886/1999-740X_2026_4_60

Abstract

The article presents the results of a three-year (2021–2023) field experiments and mathematical modeling of cotton (Gossypium hirsutum L.) productivity under irrigated agricultural conditions in the Maktaaral District of the Turkestan Region (Kazakhstan) at the "Sabyr" farm. The experimental plots were located on light medium-loamy sierozem soils characterized by 0.44–0.78% organic matter, 31.4–61.6 mg kg⁻¹ hydrolyzable nitrogen, 22.8–78.0 mg kg⁻¹ available phosphorus, and 270–407 mg kg⁻¹ exchangeable potassium. The study was conducted under two contrasting salinity levels: low (0.11–0.15% total soluble salts) and moderate-to-high (0.39–0.59%). Mineral fertilizer rates varied within the following ranges: N, 0–150; P₂O₅, 0–200; and K₂O, 0–120 kg ha⁻¹. The objective was to compare the predictive performance of conventional regression models, semi-empirical polynomial equations, and machine learning algorithms for forecasting seed cotton yield under the combined effects of soil salinity and macroelement availability. The predictor variables included total soluble salts (x₁), total nitrogen reserves (x₂), phosphorus reserves (x₃), and potassium reserves (x₄). Modeling was performed using an analytical radical regression equation and several Python-based algorithms (OLS, Ridge, Lasso, Random Forest, Gradient Boosting, SVR, and MLP). Model performance was evaluated using ANOVA and the statistical metrics R², RMSE, MAE, and MAPE. Gradient Boosting demonstrated the highest predictive accuracy (R²=0.994; RMSE=0.104 t ha⁻¹). The analytical radical regression model (R²=0.838) provided an interpretable description of soil–fertilizer–plant interactions, revealing synergistic effects of nitrogen and phosphorus nutrition that mitigate salinity stress, as well as antagonistic interactions between soil salts and potassium. The developed nonlinear models provide a scientific basis for precision nutrient management and digital decision-support systems for cotton production on irrigated saline soils. 

About the Authors

B. M. Amirov
Kazakh Research Institute of Soil Science and Agrochemistry named after U.U. Uspanov
Kazakhstan

050060, Almaty, Al-Farabi ave., 75B



A. T. Seitmenbetova
Kazakh Research Institute of Soil Science and Agrochemistry named after U.U. Uspanov
Kazakhstan

050060, Almaty, Al-Farabi ave., 75B



G. A. Saparov
Kazakh Research Institute of Soil Science and Agrochemistry named after U.U. Uspanov
Kazakhstan

050060, Almaty, Al-Farabi ave., 75B



S. O. Bazarbayev
Kazakh Research Institute of Soil Science and Agrochemistry named after U.U. Uspanov
Kazakhstan

050060, Almaty, Al-Farabi ave., 75B



O. S. Zhandybayev
Kazakh Research Institute of Soil Science and Agrochemistry named after U.U. Uspanov
Kazakhstan

050060, Almaty, Al-Farabi ave., 75B



References

1. Umbetaev I. Technology of cultivating new domestic cotton varieties in the South of Kazakhstan: monograph. - Almaty, 2005. - 144 p.

2. Umbetaev I., Dzhalankuzov T., Tagaev A., Suleimenov B. Influence of nutritional and water regimes on cotton yield // Soil Science and Agrochemistry. - 2008. - No. 1. - P. 90-94.

3. Ashirbekov M.Zh., Dridiger V.K., Batkaev Zh.Ya. Cotton yield depending on the timing and rates of phosphorus fertilizer application on irrigated sierozems of Southern Kazakhstan // Niva Povolzhya. - 2018. - No. 2. - P. 73-80.

4. Dospekhov B.A. Methodology of field experiments. - M.: Alyans, 2014. - 352 p.

5. Poluektov R.A., Smolyar E.I., Terleev V.V., Topazh A.G. Modeling the production process of agricultural crops. - St. Petersburg: St. Petersburg State University Publishing House, 2002. - 392 p.

6. Mannanova R.N. Transformation of nitrogen in the soil, its utilization by cotton, and the quantity and quality of the yield under various nutritional conditions: dissertation for the degree of Candidate of Agricultural Sciences. – Tashkent, 1984. – 130 p.

7. Obidov K.A. Productivity of new cotton varieties depending on plant density and rates of mineral fertilizers: dissertation for the degree of Candidate of Agricultural Sciences. – Dushanbe, 2012. – 115 p.

8. Umbetaev I. Dynamics of biological activity of sierozems in relation to different depths of primary soil tillage // Soil Science and Agrochemistry. - 2010. - No. 2. - P. 40-41.

9. Maas E. V., Hoffman G. J. Crop salt tolerance // Journal of the Irrigation and Drainage Division. - 1977. - Vol. 103. - No. 2. - P. 115-134.

10. Marschner H. Marschner's Mineral Nutrition of Higher Plants. - 3rd ed. - London: Academic Press, 2012. - 651 p.

11. Breiman L. Random forests // Machine Learning. - 2001. - Vol. 45. -No. 1. -P. 5-32.

12. Friedman J.H. Greedy function approximation: A gradient boosting machine // The Annals of Statistics. - 2001. - Vol. 29. - No. 5. - P. 1189-1232.

13. Sychev V.G., Lunev M.I., Pavlikhina A.V. Current state and dynamics of soil fertility changes in the Russian Federation // Soil Fertility. - 2022. - No. 1. - P. 2-7.

14. Chinilin A. V., Savin I. Yu. Large-scale digital mapping of soil organic carbon content using machine learning methods // Bulletin of the V.V. Dokuchaev Soil Institute. - 2018. - No. 95. - P. 3-25.

15. Yakushev V.P., Yakushev V.V., Blokhina S.Yu., Blokhin Yu.I., Matveenko D.A. Precision agriculture: A comprehensive approach to managing crop production and analyzing spatial variability // Herald of the Russian Academy of Sciences. - 2023. - Vol. 93. - No. 10. - P. 955-969.

16. Munns R., Tester M. Mechanisms of salinity tolerance // Annual Review of Plant Biology. - 2008. - Vol. 59. - P. 651-681.

17. Hawkesford M., Horst W., Kichey T., Lambers H., Schjoerring J., Møller I. S., White P. Functions of macronutrients // Marschner's Mineral Nutrition of Higher Plants / Ed. by P. Marschner. - 3rd ed. - London: Academic Press, 2012. - P. 135-189.

18. Lauchli A., Epstein E. Mechanisms of salt tolerance in plants // California Agriculture. - 1984. - Vol. 38. - No. 10. - P. 20-22.

19. Prokoshev V.V., Deryugin I.P. Potassium and potassium fertilizers. - M.: Ledum, 2000. - 185 p.

20. Wang M., Zheng Q., Shen Q., Guo S. The critical role of potassium in plant stress response // International Journal of Molecular Sciences. -2013. -Vol. 14. -No. 4. -P. 7370-7390.

21. Flowers T.J., Colmer T.D. Salinity tolerance in halophytes // New Phytologist. - 2008. - Vol. 179. - No. 4. - P. 945-963.

22. Zhu J. K. Regulation of ion homeostasis under salt stress // Current Opinion in Plant Biology. - 2003. - Vol. 6. - No. 5. - P. 441-442.

23. Umbetov A.K., Balgabayev A.M., Shibikeeva A.M., Zhaksybaeva G.S. Dynamics of mineral and organic phosphates in various soil types of the foothill zone of the Zailiysky Alatau under natural and anthropogenic impacts // Soil Science and Agrochemistry. - 2019. - No. 2. - P. 64-74.

24. Novikova A.V., Tsoy Z.I. Drip irrigation of cotton // Collection of Scientific Works of SANIIRI named after V. D. Zhurin. - Tashkent, 2004. - P. 356-359.

25. Ganzhara N.F. Soil Science. - M.: Agroconsult, 2001. - 392 p.

26. Sheudzhen A.Kh. Agrochemistry. - Maykop: Adygeya, 2003. - 640 p.

27. Lynch J.P. Root phenes for enhanced soil exploration and phosphorus acquisition: Tools for future crops // Plant Physiology. - 2011. - Vol. 156. - No. 3. - P. 1041-1049.


Review

For citations:


Amirov B.M., Seitmenbetova A.T., Saparov G.A., Bazarbayev S.O., Zhandybayev O.S. Optimization of cotton nutrition on saline soils of the Turkestan region using machine learning methods. Soil Science and Agrichemistry. 2026;(4):60-73. (In Russ.) https://doi.org/10.51886/1999-740X_2026_4_60

Views: 0

JATS XML


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.


ISSN 1999-740X (Print)
ISSN 2959-3433 (Online)