vestibИзвестия Национальной академии наук Беларуси. Серия биологических наукProceedings of the National Academy of Sciences of Belarus, Biological Series1029-89402524-230XThe Republican Unitary Enterprise Publishing House "Belaruskaya Navuka"10.29235/1029-8940-2024-69-3-237-248vestib-931Research ArticleСтатьиПостроение прогноза накопления 137Cs древесными растениями и сельскохозяйственными культурами с использованием метода решающих деревьевForecasting the accumulation of 137Cs by trees and crops using the decision tree methodhttps://orcid.org/0000-0002-1369-0093НикитинА. Н.NikitinA. N.

Никитин Александр Николаевич ‒ канд. с/х наук, заместитель директора.

ул. Купревича, 2, 220141, Минск

Aleksander N. Nikitin ‒ Ph. D. (Agricult.), Deputy Director, Institute of Microbiology of the National Academy of Sciences of Belarus.

2, Kuprevich Str., 220141, Minsk

nikitinale@gmail.comhttps://orcid.org/0000-0001-6857-2337КудинМ. В.KudinM. V.

Кудин Максим Владимирович ‒ канд. с/х наук, доцент, заместитель директора.

ул. Терешковой, 7, 247618, Хойники

Maksim V. Kudin ‒ Ph. D. (Agricult.), Associate Professor, Deputy Director, Polessie State Radioecological Reserve.

7, Tereshkova Str., 247618, Choiniki

max.kudin@mail.ruКалиниченкоС. А.KalinichenkoS. A.

Калиниченко Сергей Александрович ‒ канд. биол. наук, доцент, заведующий лабораторией.

ул. Терешковой, 7, 247618, Хойники

Sergey A. Kalinichenko ‒ Ph. D. (Biol.), Associate Professor, Head of the Laboratory, Polessie State Radioecological Reserve.

7, Tereshkova Str., 247618, Choiniki

s-a-k@list.ruЛаськоТ. В.LaskoT. V.

Ласько Тамара Васильевна ‒ ст. науч. сотрудник.

ул. Федюнинского, 4, 246007, Гомель

Tamara V. Lasko ‒ Senior Researcher, Institute of Radiobiology of the National Academy of Sciences of Belarus.

4, Fedyuninski Str., 246007, Gomel

t-lasko@yandex.byШуранковаО. А.ShurankovaO. A.

Шуранкова Ольга Александровна ‒ науч. сотрудник.

ул. Федюнинского, 4, 246007, Гомель

Olga A. Shurankova ‒ Researcher, Institute of Radiobiology of the National Academy of Sciences of Belarus.

4, Fedyuninski Str., 246007, Gomel

shurankova@list.ruМищенкоЕ. В.MishchankaE. V.

Мищенко Егор Викторович ‒ заместитель заведующего лабораторией.

ул. Федюнинского, 4, 246007, Гомель

Egor V. Mishchanka ‒ Deputy head of the Laboratory, Institute of Radiobiology of the National Academy of Sciences of Belarus.

4, Fedyuninski Str., 246007, Gomel

egormischenko@gmail.comИнститут микробиологии НАН БеларусиInstitute of Microbiology of the National Academy of Sciences of BelarusПолесский государственный радиационно-экологический заповедникPolessie State Radioecological ReserveИнститут радиобиологии НАН БеларусиInstitute of Radiobiology of the National Academy of Sciences of Belarus202401082024693237248Copyright © Никитин А.Н., Кудин М.В., Калиниченко С.А., Ласько Т.В., Шуранкова О.А., Мищенко Е.В., 20242024Никитин А.Н., Кудин М.В., Калиниченко С.А., Ласько Т.В., Шуранкова О.А., Мищенко Е.В.Nikitin A.N., Kudin M.V., Kalinichenko S.A., Lasko T.V., Shurankova O.A., Mishchanka E.V.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://vestibio.belnauka.by/jour/article/view/931

Изучены закономерности накопления 137Cs в стволовой древесине сосновых насаждений и в урожае сельскохозяйственных культур с использованием метода градиентного бустинга на решающих деревьях и SHAP-анализа. Для сосновых насаждений в зоне отчуждения Чернобыльской АЭС выявлена нелинейная связь коэффициента перехода с высотой над уровнем моря, а также с вегетационными индексами, указывающими на общее состояние насаждений, их биологическую продуктивность и обеспеченность калием. В агроэкосистемах на территории Гомельской и Могилевской областей подтверждено влияние на коэффициент перехода вида растения и концентрации K+ в почвенном растворе. Использование интерпретируемого метода машинного обучения позволило определить характер влияния дефицита почвенной влаги на накопление 137Cs сельскохозяйственными растениями, а также показать сохранение вклада трансфолиарного поступления радионуклида при низких уровнях загрязнения почвы на этапе отдаленных последствий радиоактивных выпадений. Применение метода градиентного бустинга на решающих деревьях и интерпретация модели с помощью SHAP-анализа обеспечили более глубокое понимание сложных взаимосвязей факторов, влияющих на поступление 137Cs в растения, что открывает перспективы для повышения точности прогноза загрязнения растительных ресурсов радионуклидами.

The article provides a profound analysis of the accumulation of the radionuclide 137Cs in the stems of pine trees and harvest of crops, employing decision tree methods and SHAP analysis. In pine forests situated in the Chernobyl exclusion zone, a nonlinear relationship between the aggregated transfer factor and elevation above sea level is identified, along with the influence of vegetation indices pointing to overall stand condition, biological productivity, and potassium deficiency. In agroecosystems situated in Gomel and Mogilev regions, the impact of plant species, K+ concentration in the soil solution on aggregated transfer factor is confirmed. Interpretable machine learning method shows dependence of aggregated transfer factor from soil moisture and the persistence of transfoliar contamination at low soil pollution levels at late stage after Chernobyl catastrophe. The application of decision trees and SHAP analysis offers a deeper understanding of complex interactions in the “soil-plant” system, opening perspectives for effective monitoring and management of radioactive contamination in diverse natural and agricultural environments.

экологические факторыпочвенно-растительная системацезий-137вегетационные индексымашинное обучениерешающие деревьяSHAP-анализecological factorssoil-plant systemcesium-137vegetation indicesmachine learningdecision treesSHAP-analysisРабота выполнена в рамках мероприятий Государственной программы по преодолению последствий катастрофы на Чернобыльской АЭС на 2011–2015 годы и на период до 2020 года, а также Программы совместной деятельности России и Беларуси в рамках Союзного государства по защите населения и реабилитации территорий, пострадавших в результате катастрофы на Чернобыльской АЭС.The work was carried out within the framework of the tasks of the State Program for Overcoming the Consequences of the Chernobyl Nuclear Power Plant Accident for 2011–2015 and for the period until 2020, as well as the Joint Program of Russia and Belarus within the framework of the Union State on the Protection of the Population and the Rehabilitation of the Territories Affected by the Chernobyl Nuclear Power Plant Accident.ReferencesDebeljak, M. Decision trees in ecological modelling / M. Debeljak, S. Džeroski // Modelling complex ecological dynamics: An introduction into ecological modelling for students, teachers & scientists / ed.: F. Jopp [et al.]. – Heidelberg, 2011. – P. 197–209.Debeljak M., Džeroski S. Decision trees in ecological modelling. Modelling complex ecological dynamics: An introduction into ecological modelling for students, teachers & scientists. Berlin, Heidelberg, 2011, pp. 197–209. https://doi.org/10.1007/978-3-642-05029-9_14Zhang, B. Modelling the productivity of naturalised pasture in the North Island, New Zealand: a decision tree approach / B. Zhang, I. Valentine, P. Kemp // Ecol. Modelling. – 2005. – Vol. 186, N 3. – P. 299–311. https://doi.org/10.1016/j.ecolmodel.2004.12.016Zhang B., Valentine I., Kemp P. Modelling the productivity of naturalised pasture in the North Island, New Zealand: a decision tree approach. Ecological Modelling, 2005, vol. 186, no. 3, pp. 299–311. https://doi.org/10.1016/j.ecolmodel.2004.12.016Prediction of phytoplankton biomass and identification of key influencing factors using interpretable machine learning models / Y. Xu [et al.] // Ecolog. Indicators. – 2024. – Vol. 158. – Art. 111320. https://doi.org/10.1016/j.ecolind.2023.111320Xu Y., Zhang D., Lin J., Peng Q., Lei X., Jin T., Wang J., Yuan R. Prediction of phytoplankton biomass and identification of key influencing factors using interpretable machine learning models. Ecological Indicators, 2024, vol. 158, art. 111320. https://doi.org/10.1016/j.ecolind.2023.111320Prediction and analysis of net ecosystem carbon exchange based on gradient boosting regression and random forest / J. Cai [et al.] // Appl. Energy. – 2020. – Vol. 262. – Art. 114566. https://doi.org/10.1016/j.apenergy.2020.114566Cai J., Xu K., Zhu Y., Hu F., Li L. Prediction and analysis of net ecosystem carbon exchange based on gradient boosting regression and random forest. Applied Energy, 2020, vol. 262, art. 114566. https://doi.org/10.1016/j.apenergy.2020.114566Predictive models for the effect of environmental factors on the abundance of vibrio parahaemolyticus in oyster farms in taiwan using extreme gradient boosting / N. Ndraha [et al.] // Food Control. – 2021. – Vol. 130. – Art. 108353. https://doi.org/10.1016/j.foodcont.2021.108353Ndraha N., Hsiao H., Hsieh Y., Pradhan A. K. Predictive models for the effect of environmental factors on the abundance of vibrio parahaemolyticus in oyster farms in Taiwan using extreme gradient boosting. Food Control, 2021, vol. 130, art. 108353. https://doi.org/10.1016/j.foodcont.2021.108353Golden, C. E. Comparison between random forest and gradient boosting machine methods for predicting Listeria spp. prevalence in the environment of pastured poultry farms / C. E. Golden, M. J. Rothrock, A. Mishra // Food Res. Int. – 2019. – Vol. 122. – P. 47–55. https://doi.org/10.1016/j.foodres.2019.03.062Golden C. E., Rothrock M. J., Mishra A. Comparison between random forest and gradient boosting machine methods for predicting Listeria spp. prevalence in the environment of pastured poultry farms. Food Research International, 2019, vol. 122, pp. 47–55. https://doi.org/10.1016/j.foodres.2019.03.062Techniques to improve ecological interpretability of black-box machine learning models / T. Welchowski [et al.] // J. Ag., Biol. Envir. St. – 2021. – Vol. 27, N 1. – P. 175–197. https://doi.org/10.1007/s13253-021-00479-7Welchowski T., Maloney K. O., Mitchell R., Schmidt M. Techniques to improve ecological interpretability of black-box machine learning models. Journal of Agricultural, Biological and Environmental Statistics, 2021, vol. 27, no. 1, pp. 175–197. https://doi.org/10.1007/s13253-021-00479-7Konoplev, A. Mobility and bioavailability of the Chernobyl-derived radionuclides in soil-water environment: Review / A. Konoplev // Behavior of radionuclides in the environment II: Chernobyl / ed. by A. Konoplev, K. Kato, S. N. Kalmykov. – Singapore, 2020. – P. 157–193. https://doi.org/10.1007/978-981-15-3568-0_3Konoplev A. Mobility and bioavailability of the Chernobyl-derived radionuclides in soil-water environment: Review. Behavior of radionuclides in the environment II: Chernobyl. Singapore, 2020, pp. 157–193. https://doi.org/10.1007/978-981-15-3568-0_3Мирошников, В. Справочник таксатора / В. Мирошников, О. Труль, В. Ермаков. ‒ Минск: Ураджай, 1980. – 359 с.Miroshnikov V., Trul’ O., Ermakov V. Taxator’s handbook. Minsk, Uradzhai Publ., 1980. 359 p. (in Russian).Радиационный контроль. Обследование лесосек. Порядок проведения: ТКП 239.2010 (02080). Введ. 01.06.2010 / М-во лесн. хоз-ва Респ. Беларусь. – Минск: М-во лесн. хоз-ва Респ. Беларусь, 2010. – 22 с.Radiation control. Inspection of cutting areas. Procedures, 239.2010 (02080). 01.06.2010. Minsk, Ministry of Forestry of the Republic of Belarus, 2010. 22 p. (in Russian).Практикум по агрохимии: учеб. пособие / О. А. Амельянчик [и др.]; под ред. В. Минеева. ‒ 2-е изд., перераб. и доп. – М.: Изд-во МГУ, 2001. – 689 с.Amel’yanchik O. A., Bolysheva T. N., Gomonova N. F., Durynina E. P., Egorov V. S., Egorova E. V. [et al.]. Workshop on agrochemistry. 2nd ed. Moscow, Moscow State University Publ., 2001. 689 p. (in Russian).Gilmore, G. R. Practical gamma-ray spectrometry / G. R. Gilmore. ‒ 2nd ed. – John Wiley & Sons, 2008. – 387 p.Gilmore G. R. Practical gamma-ray spectrometry. 2nd ed. John Wiley & Sons, 2008. 387 p.Tarsitano, D. Evaluating and reducing a model of radiocaesium soil-plant uptake / D. Tarsitano, S. Young, N. Crout // J. Environ. Radioactiv. – 2011. – Vol. 102, N 3. – P. 262–269. https://doi.org/10.1016/j.jenvrad.2010.11.017Tarsitano D., Young S., Crout N. Evaluating and reducing a model of radiocaesium soil-plant uptake. Journal of Environmental Radioactivity, 2011, vol. 102, no. 3, pp. 262–269. https://doi.org/10.1016/j.jenvrad.2010.11.017Deriving probabilistic soil distribution coefficients (Kd). Part 2: Reducing caesium Kd uncertainty by accounting for experimental approach and soil properties / Ramírez-Guinart O. [et al.] // J. Environ. Radioactiv. – 2020. – Vol. 223–224. – Art. 106407. https://doi.org/10.1016/j.jenvrad.2020.106407Ramírez-Guinart O., Kaplan D., Rigol A., Vidal M. Deriving probabilistic soil distribution coefficients (Kd). Part 2: Reducing caesium Kd uncertainty by accounting for experimental approach and soil properties. Journal of Environmental Radioactivity, 2020, vol. 223–224, art. 106407. https://doi.org/10.1016/j.jenvrad.2020.106407Sheppard, S. C. Robust prediction of Kd from soil properties for environmental assessment / S. C. Sheppard // Hum. Ecol. Risk Assess. – 2011. – Vol. 17, N 1. – P. 263–279. https://doi.org/10.1016/j.jenvrad.2020.106407Sheppard S. C. Robust prediction of Kd from soil properties for environmental assessment. Human and Ecological Risk Assessment, 2011, vol. 17, no. 1, pp. 263–279. https://doi.org/10.1016/j.jenvrad.2020.106407Sentinel-2: ESA’s optical high-resolution mission for GMES operational services / M. Drusch [et al.] // Remote Sensing Environ. – 2012. – Vol. 120. – P. 25–36. https://doi.org/10.1016/j.rse.2011.11.026Drusch M., Del Bello U., Carlier S., Colin O., Fernandez V., Gascon F. [et al.]. Sentinel-2: ESA’s optical high-resolution mission for GMES operational services. Remote Sensing of Environment, 2012, vol. 120, pp. 25–36. https://doi.org/10.1016/j.rse.2011.11.026ALOS World 3D 30-meter DEM. V3.2 [Electronic resource] / JAEA 2021. – Mode of access: https://doi.org/10.5069/G94M92HB. – Date of access: 17.11.2021.ALOS World 3D 30-meter DEM. V3.2. JAEA 2021. Available at: https://doi.org/10.5069/G94M92HB (accessed 17.11.2021).Version 4 of the SMAP level‐4 soil moisture algorithm and data product / R. H. Reichle [et al.] // J. Adv. Model. Earth Syst. – 2019. – Vol. 11, N 10. – P. 3106–3130. https://doi.org/10.1029/2019MS001729Reichle R. H., Liu Q., Koster R. D., Crow W. T., De Lannoy G. J. M., Kimball J. S. [et al.]. Version 4 of the SMAP level‐4 soil moisture algorithm and data product. Journal of Advances in Modeling Earth Systems, 2019, vol. 11, no. 10, pp. 3106–3130. https://doi.org/10.1029/2019MS001729Lightgbm: A highly efficient gradient boosting decision tree [Electronic resource] / G. Ke [et al.] // Advances in Neural Information Processing Systems 30 (NIPS 2017) / ed. by I. Guyon [et al.]. – Mode of access: https://proceedings.neurips.cc/paper_files/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf. – Date of access: 31.05.2024.Ke G., Meng Q., Finley T., Wang T., Chen W., Ma W., Ye Q., Liu T. LightGBM: A highly efficient gradient boosting decision tree. Advances in Neural Information Processing Systems 30 (NIPS 2017). Available at: https://proceedings.neurips.cc/paper_files/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf (accessed 31.05.2024).Snoek, J. Practical bayesian optimization of machine learning algorithms [Electronic resource] / J. Snoek, H. Larochelle, R. P. Adams // Proceedings of the 25th International conference on neural information processing systems: in 2 vol. – Lake Tahoe, Nevada: Curran Associates Inc., 2012. – Vol. 2. – P. 2951–2959. Mode of access: https://dl.acm.org/doi/10.5555/2999325.2999464. – Date of access: 31.05.2024.Snoek J., Larochelle H., Adams R. P. Practical Bayesian optimization of machine learning algorithms. Proceedings of the 25th International conference on neural information processing systems: in 2 vol. Lake Tahoe, Nevada, Curran Associates Inc., 2012, pp. 2951–2959. Available at: https://dl.acm.org/doi/10.5555/2999325.2999464 (accessed 31.05.2024).Explainable machine-learning predictions for the prevention of hypoxaemia during surgery / S. M. Lundberg [et al.] // Nat. Biomed. Engin. – 2018. – Vol. 2, N 10. – P. 749–760. https://doi.org/10.1038/s41551-018-0304-0Lundberg S. M., Nair B., Vavilala M. S., Horibe M., Eisses M. J., Adams T., Liston D. E., Low D. K., Newman S., Kim J., Lee S. Explainable machine-learning predictions for the prevention of hypoxaemia during surgery. Nature Biomedical Engineering, 2018, vol. 2, no. 10, pp. 749–760. https://doi.org/10.1038/s41551-018-0304-0From local explanations to global understanding with explainable AI for trees / S. M. Lundberg [et al.] // Nat. Machine Intel. – 2020. – Vol. 2, N 1. – P. 56–67. https://doi.org/10.1038/s42256-019-0138-9Lundberg S. M., Erion G., Chen H. DeGrave A., Prutkin J. M., Nair B., Katz R., Himmelfarb J., Bansal N., Lee S. From local explanations to global understanding with explainable AI for trees. Nature Machine Intelligence, 2020, vol. 2, no. 1, pp. 56–67. https://doi.org/10.1038/s42256-019-0138-9Анализ возможности использования спутниковых данных для прогноза параметров перехода радиоактивных изотопов цезия из почвы в древесные растения / А. Н. Никитин А. [и др.] // Восьмой белорусский космический конгресс: материалы, 25‒27 окт. 2022 г., г. Минск: в 2 т. / науч. ред.: С. В. Кругликов, И. В. Филипченко. – Минск, 2022. – Т. 1. – С. 245–249.Nikitin A. N., Kudin M. V., Cheshik I. A., Shurankova O. A., Mishchenko E. V., Sukhareva D. V., Zhukovskaya E. V., Blinova N. V. Analysis of the possibility of using satellite data to predict the parameters of the transfer of radioactive cesium isotopes from soil to trees. Vos’moi Belorusskii kosmicheskii kongress: materialy kongressa, 25–27 oktyabrya 2022 goda, Minsk. Tom 1 [Eighth Belarusian Space Congress: materials of the congress, October 25–27, 2022, Minsk. Volume 1]. Minsk, 2022, pp. 245–249 (in Russian).The high affinity K+ transporter AtHAK5 plays a physiological role in planta at very low K+ concentrations and provides a caesium uptake pathway in arabidopsis / Z. Qi [et al.] // J. Exp. Bot. – 2008. – Vol. 59, N 3. – P. 595–607. https://doi.org/10.1093/jxb/erm330Qi Z., Hampton C. R., Shin R., Barkla B. J., White P. J. The high affinity K+ transporter AtHAK5 plays a physiological role in planta at very low K+ concentrations and provides a caesium uptake pathway in arabidopsis. Journal of Experimental Botany, 2008, vol. 59, no. 3, pp. 595–607. https://doi.org/10.1093/jxb/erm330Экспериментальная оценка влияния режима увлажнения почвы на накопление 137Cs побегами яровой пшеницы / А. Н. Никитин [и др.] // Вес. Нац. акад. навук Беларусі. Сер. біял. навук. – 2020. – Т. 65, № 2. – P. 229–238.Nikitin A. N., Shurankova O. A., Cheshik I. A., Leferd G. A., Mishchenko E. V., Zhukovskaya E. V., Sukhareva D. V. Experimental assessment of influence of soil moisture on the 137Cs accumulation in shoots of spring wheat. Vestsi Natsyyanal’nai akademii navuk Belarusi. Seryya biyalagichnykh navuk = Proceedings of the National Academy of Sciences of Belarus. 2020, vol. 65, no. 2, pp. 229–238 (in Russian).Nikitin, A. N. Impact of soil moisture on cesium uptake by plants: Model assessment / A. N. Nikitin // J. Environ. Radioactiv. – 2021. – Vol. 240. – Art. 106754. https://doi.org/10.1016/j.jenvrad.2021.106754Nikitin A. N. Impact of soil moisture on cesium uptake by plants: Model assessment. Journal of Environmental Radioactivity, 2021, vol. 240, art. 106754. https://doi.org/10.1016/j.jenvrad.2021.106754Smolders, E. Concentrations of 137Cs and K in soil solution predict the plant availability of 137Cs in soils / E. Smolders, K. Van den Brande, R. Merckx // Environ. Sci. Technol. – 1997. – Vol. 31, N 12. – P. 3432–3438. https://doi.org/10.1021/es970113rSmolders E., den Brande K. V., Merckx R. Concentrations of 137Cs and K in soil solution predict the plant availability of 137Cs in soils. Environmental Science & Technology, 1997, vol. 31, no. 12, pp. 3432–3438. https://doi.org/10.1021/es970113r

The authors declare that there are no conflicts of interest present.