Agronomic traits of potato enabling yield prediction

Scientific research on potato breeding is aimed at increasing crop yields and the resistance of cultivars to diseases and pests, as well as improving product quality. Based on the results obtained, statistical models are developed that allow not only to create a system for forecasting the parameter under study, but also to explain the contribution of individual traits that form this parameter. The aim of the study is to determine the optimal model for forecasting potato yield based on the analysis of agronomic traits (average tuber weight per a potato plant; the number of tubers per a plant; content of starch, protein, total and reducing sugars, vitamin C and nitrates in tubers) of 100 potato cultivars harvested in 2024 (Sverdlovsk Region). In this work, a linear regression model of potato yield (t/ha) was developed with three predictors: average tuber weight (g), the number of tubers (units) and natural logarithm of nitrate content in tubers (mg/kg). The resulting model was able to predict 77.9 % of the data dispersion (R2 = 0.785, R2adj = 0.779, p<0.001), by this the degree of model fit was ideal, since the mean absolute error MAPE was less than 10%. The regression model assumptions were tested using the LINE algorithm: linearity, independence, normality, homogeneity. This algorithm proved the adequacy of the obtained model. Using the linear regression method, it was shown that with an increase in the average weight of marketable tubers by 1 g and the number of tubers by 1 unit, potato yield increased by 0.314 and 2.386 t/ha, respectively. However, with an increase in nitrate concentration per unit of natural logarithm, the yield decreased by 3.63 t/ha. While tuber weight and tuber number had a functional relationship with the yield, nitrate content had only an indirect relationship. According to the literature, the nitrate concentration in potato tubers at the end of the harvest represents a «final assessment» of the efficiency of nitrogen use by a particular cultivar throughout the growing season. The model obtained determines the selection criteria for breeding high-yielding potato cultivars: selection of genotypes with a large number and weight of tubers, low residual nitrate content in tubers under standard nitrogen nutrition conditions.

1. Devaux A., Goffart J. P., Kromann P., Andrade-Piedra J., Polar V., Hareau G. The potato of the future: opportunities and challenges in sustainable agri-food systems. Potato Research. 2021;64(4):681–720. DOI: https://doi.org/10.1007/s11540-021-09501-4

2. Amirova E. F., Gavrilyeva N. K., Grigoriev A. V., Sorgutov I. V. Digitalization in agriculture: problems of implementation. Siberian Journal of Life Sciences and Agriculture. 2021;13(6):144–155. (In Russ.). DOI: https://doi.org/10.12731/2658-6649-2021-13-6-144-155

3. Kamishenko G. A. Climatic component in the calculation potato yields. Prirodnie resursi = Natural Resources. 2021;(2):14–21. (In Belarus). URL: https://elibrary.ru/item.asp?id=47503785

4. Kuradusenge M., Hitimana E., Hanyurwimfura D., Rukundo P., Mtonga K., Mukasine A. et al. Crop yield prediction using machine learning models: case of irish potato and maize. Agriculture. 2023;13(1):225. DOI: https://doi.org/10.3390/agriculture13010225

5. Piekutowska M., Niedbała G., Piskier T., Lenartowicz T., Pilarski K., Wojciechowski T. et al. The application of multiple linear regression and artificial neural network models for yield prediction of very early potato cultivars before harvest. Agronomy. 2021;11(5):885. DOI: https://doi.org/10.3390/agronomy11050885

6. Vorobyov N. I., Lysov A. K., Kornilov T. V., Hyutti A. V. Computational neural network for processing light-reflective spectra of plants and remote phytosanitary monitoring of potatoes. Agrarnaya nauka Evro-SeveroVostoka = Agricultural Science Euro-North-East. 2024;25(2):283–292. (In Russ.). DOI: https://doi.org/10.30766/2072-9081.2024.25.2.283-292

7. Shanina E. P., Klyukina E. M., Belyaeva N. V., Stafeeva M. A., Kelik L. A., Akhmetkhanov V. F. Testing of potato varieties of different intended uses. Kartofel' i ovoshchi = Potato and Vegetables. 2024;(1):39–43. (In Russ.). DOI: https://doi.org/10.25630/PAV.2024.92.35.003

8. Shanina E. P., Klyukina E. M., Stafeeva M. A., Belyaeva N. V., Gonchar O. N. The comparative geographical analysis of potato varieties from a collection nursery. Dostizheniya nauki i tekhniki APK = Achievements of Science and Technology of AICis. 2020;34(6):75–78. (In Russ.). DOI: https://doi.org/10.24411/0235-2451-2020-10614

9. Cravero A., Pardo S., Sepúlveda S., Muñoz L. Challenges to use machine learning in agricultural big data: a systematic literature review. Agronomy. 2022;12(3):748. DOI: https://doi.org/10.3390/agronomy12030748

10. Lyskova I. V., Permyakov P. V., Kratyuk E. I. The results of study of collection cultivars and potato new breeding numbers according to agronomic traits. Agrarnaya nauka Evro-Severo-Vostoka = Agricultural Science EuroNorth-East. 2024;25(6):1019–1027. (In Russ.). DOI: https://doi.org/10.30766/2072-9081.2024.25.6.1019-1027

11. Wszelaczyńska E., Pobereżny J., Keutgen A. J., Keutgen N., Gościnna K., Milczarek D. et al. Antinutritional nitrogen compounds content in potato (Solanum tuberosum L.) tubers depending on the genotype and production system. Agronomy. 2022;12(10):2415. DOI: https://doi.org/10.3390/agronomy12102415

12. Ding K., Shan Y., Wang L., Zhang Y., Tian G. Transcriptomics combined with physiological analysis and metabolomics revealed the response of potato tuber formation to nitrogen. BMC Plant Biology. 2024;24(1):1109. DOI: https://doi.org/10.1186/s12870-024-05758-2

13. Muleta N. D., Aga M. C. Role of nitrogen on potato production: a review. Journal of Plant Sciences. 2019;7(2):36–42. DOI: https://doi.org/10.11648/j.jps.20190702.11

14. Zebarth B. J., Tai G., Tarn R., de Jong H., Milburn P. H. Nitrogen use efficiency characteristics of commercial potato cultivars. Canadian Journal of Plant Science. 2004;84(2):589–598. DOI: https://doi.org/10.4141/P03-050

15. Frąckowiak K., Potarzycki J., Grzebisz W., Szczepaniak W. Potato nutri-tional status at the onset of tuberisation – a yield prediction tool. Plant, Soil and Environment. 2020;66(2):86–92. DOI: https://doi.org/10.17221/533/2019-PSE

16. do Carmo Milagres C., Fontes P. C. R., da Silveira M. V., Moreira M. A., de Car-valho Lopes I. P. Índices de nitrogênio e modelo para prognosticar a produção de tubérculos de batata. Revista Ceres. 2018;(65):261–270. DOI: https://doi.org/10.1590/0034-737X201865030006