Цифровое картографирование показателей, определяющих сорбционные свойства почв по отношению к поллютантам, по данным дистанционного зондирования Земли с применением машинного обучения

По данным дистанционного зондирования Земли (ДЗЗ) проведено сравнение точности пространственного прогноза почвенных показателей, определяющих сорбционные свойства по отношению к поллютантам. Для построения пространственных карт изменения свойств почвы использовались методы машинного обучения на основе моделей регрессии опорных векторов (SVMr – support vector machine regression) и случайного леса (RF – random forest). Показано, что методы машинного моделирования с использованием ДЗЗ могут быть успешно использованы для пространственного прогноза содержания гранулометрических фракций, органического вещества, рН и емкости катионного обмена почв на участках небольшой площади. Выявлено, что пространственный прогноз содержания фракции пыли наилучшим образом моделируется с помощью алгоритма RF, тогда как остальные свойства почв, способные определить их собрционный потенциал по отношению к поллютантам, лучше моделируются с помощью метода SVMr. В целом, оба метода машинного обучения дают близкие результаты пространственного прогноза.

1. Цибарт А.С., Геннадиев А.Н. (2013). Полициклические ароматические углеводороды в почвах: источники, поведение, индикационное значение (обзор). Почвоведение, 7, с. 788–802. https://doi.org/10.7868/S0032180X13070125

2. Цифровая почвенная картография (2017). М.: Изд-во РУДН, 152 с.

3. Al-Ruzouq R., Gibril M. A., Abdallah S., Kais A., Hamed O., Saeed Al-M., Mohamad K. (2020). Sensors, Features, and Machine Learning for Oil Spill Detection and Monitoring: A Review. Remote Sensing, 12, 3338. https://doi.org/10.3390/rs12203338

4. Andronikov S.V., Davidson D.A., Spiers R.B. (2000). Variability in Contamination by Heavy Metals: Sampling Implications. Water, Air, & Soil Pollution, 120, pp. 29–45. https://doi.org/10.1023/A:1005261522465

5. Beucher A., Adhikari K., Breuning-Madsen H., Greve M.B., Österholm P., Fröjdö S., Jensen N.H., Greve M.H. (2017). Mapping potential acid sulfate soils in Denmark using legacy data and LiDAR-based derivatives. Geoderma, 308, pp. 363–372. https://doi.org/10.1016/j.geoderma.2016.06.001

6. Biau G., Scornet E. (2016). A random forest guided tour. Test, 25, pp. 197–227. https://doi.org/10.1007/s11749-016-0481-7

7. Caubet M., Dobarco R. M., Arrouays D.,Minasny B., Saby N. (2019). Merging country, continental and global predictions of soil texture: Lessons from ensemble modelling in France. Geoderma, 337. pp. 99–110. https://doi.org/10.1016/j.geoderma.2018.09.007

8. Cho K.H., Sthiannopkao S., Pachepsky Y.A., Kim K.W., Kim J.H. (2011). Prediction of contamination potential of groundwater arsenic in Cambodia, Laos, and Thailand using artificial neural network. Water Res, 45(17), pp. 5535–5544. https://doi.org/10.1016/j.watres.2011.08.010

9. Cortes C., Vapnik V. (1995). Support-vector networks. Mach. Learn., 20, pp. 273–297. https://doi.org/10.1007/BF00994018

10. Cui Y.-Q., Yoneda M., Shimada Y., Matsui Y. (2016). Cost-Effective Strategy for the Investigation and Remediation of Polluted Soil Using Geostatistics and a Genetic Algorithm Approach. Journal of Environmental Protection, 07(01), pp. 99–115. https://doi.org/10.4236/jep.2016.71010

11. Deiss L., Margenot A.J., Culman S.W., Demyan M.S. (2020). Tuning support vector machines regression models improves prediction accuracy of soil properties in MIR spectroscopy, Geoderma, 365, 114227. https://doi.org/10.1016/j.geoderma.2020.114227

12. Einax J., Soldt U., Geostatistical investigations of polluted soils. (1995). Fresenius’ Journal of Analytical Chemistry, 351, pp. 48–53. https://doi.org/10.1016/j.envpol.2012.06.006

13. Grunwald S. (2009). Multi-criteria characterization of recent digital soil mapping and modeling approaches. Geoderma, 152, рр. 195–207. https://doi.org/10.1016/j.geoderma.2009.06.003

14. Güler C., Alpaslan M., Kurt M.A. (2010). Deciphering factors controlling trace element distribution in the soils of Karaduvar industrial-agricultural area (Mersin, SE Turkey). Environ Earth Sci, 60, pp. 203–218. https://doi.org/10.1007/s12665-009-0180-8

15. Ha H., Olson J.R., Bian L., Rogerson P.A. (2014). Analysis of Heavy Metal Sources in Soil Using Kriging Interpolation on Principal Components. Environmental Science & Technology, 48, pp. 4999–5007. https://doi.org/10.1021/es405083f

16. Harrell F.E.Jr. (2001). Regression Modeling Strategies. With Applications to Linear Models, Logistic Regression, and Survival Analysis. Springer, 507 p. https://doi.org/10.1007/978-1-4757-3462-1

17. Hengl T., Nussbaum M., Wright M.N., Heuvelink G.B.M., Gräler B. (2018). Random forest as a generic framework for predictive modeling of spatial and spatio-temporal variables. PeerJ, 6, e5518. https://doi.org/10.7717/peerj.5518

18. Hooda, P.S., Glavinandp R.J. (2005). A Practical Examination of the Use of Geostatistics in the Remediation of a Site with a Complex Metal Contamination History. Soil and Sediment Contamination, 14, pp. 155–169. https://doi.org/10.1080/15320380590911814

19. Juang K.-W., Liao W.-J., Liu T.-L., Tsui L., Lee D.-Y. (2008). Additional sampling based on regulation threshold and kriging variance to reduce the probability of false delineation in a contaminated site. Science of the Total Environment, 389, pp. 20–28. https://doi.org/10.1016/j.scitotenv.2007.08.025Kabata-Pendias A. (2000). Trace Elements in Soils and Plants. CRC Press, 403 p. https://doi.org/10.1201/9781420039900

20. Laborczi A., Gábor S., Kaposi A., László P. (2018). Comparison of soil texture maps synthetized from standard depth layers with directly compiled products. Geoderma, 352, pp. 360–372. https://doi.org/10.1016/j.geoderma.2018.01.020

21. Levy D.B., Barbarrick K.A., Siemer E.G., Sommers L.E. (1992). Distribution and partitioning of trace metals in contaminated soils near Leadville, Colorado. J Environ Qual, 21, pp. 185–195. https://doi.org/10.2134/jeq1992.00472425002100020006x

22. Lin Y.-P., Cheng B.-Y., Chu H.-J., Chang T.-K., Yu H.-L. (2011). Assessing how heavy metal pollution and human activity are related by using logistic regression and kriging methods. Geoderma, 163(3–4), pp. 275–282. https://doi.org/10.1016/j.geoderma.2011.05.004

23. Lin Y.-P., Chu H.-J., Huang Y.-L., Cheng B.-Y., Chang T.-K. (2010). Modeling Spatial Uncertainty of Heavy Metal Content in Soil by Conditional Latin Hypercube Sampling and Geostatistical Simulation. Environmental Earth Sciences, 62, pp. 299–311. https://doi.org/10.1007/s12665-010-0523-5

24. Loiseau T., Arrouays D., Richer-de-Forges A., Lagacherie P., Ducommun C., Minasny B. (2021). Density of soil observations in digital soil mapping: A study in the Mayenne region, France. Geoderma Reg., 24, e00358. https://doi.org/10.1016/j.geodrs.2021.e00358

25. Lv J., Yang L., Zhang Z., Dai J. (2013). Factorial kriging and stepwise regression approach to identify environmental factors influencing spatial multiscale variability of heavy metals in soils. Journal of Hazardous Materials, 261(15), pp. 387–397. https://doi.org/10.1016/j.jhazmat.2013.07.065

26. Mahmoudzadeh H., Matinfar H.R., Taghizadeh-Mehrjardi R., Kerry R. (2020). Spatial prediction of soil organic carbon using machine learning techniques in western Iran. Geoderma Reg., 21, e00260. https://doi.org/10.1016/j.geodrs.2020.e00260

27. Matinfara H. R., Maghsodi Z., Mousavi S. R., Rahmani A. (2021). Evaluation and Prediction of Topsoil organic carbon using Machine learning and hybrid models at a Field-scale. Catena, 202, pp. 105258. https://doi.org/10.1016/j.catena.2021.105258

28. McBratney A.B., Mendonçа Santos M.L., Minasny B. (2003). On digital soil mapping. Geoderma, 117, pp. 3–52. https://doi.org/10.1016/S0016-7061(03)00223-4

29. Milillo T.M., Sinha G., Gardella J.A. (2012). Use of Geostatistics for Remediation Planning to Transcend Urban Political Boundaries. Environmental Pollution, 170, pp. 52–62. https://doi.org/10.1016/j.envpol.2012.06.006

30. Mishra R., Naseer M., Nilanjan R. (2016). Soil pollution: Causes, effects and control. Van Sangyan, 3, pp. 1–14.

31. Pahlavan-Rad M.R., Dahmardeh K., Brungard C. (2018). Predicting soil organic carbon concentrations in a low relief landscape, eastern Iran. Geoderma Reg., 15, e00195. https://doi.org/10.1016/j.geodrs.2018.e00195

32. Pasolli L., Notarnicola C., Bruzzone L. (2011). Estimating soil moisture with the support vector regression technique. IEEE Geosci. Remote Sens. Lett., 8, pp. 1080–1084. https://doi.org/10.1109/LGRS.2011.2156759

33. Paterson S., Minasny B., Mcbratney A. (2018). Spatial variability of Australian soil texture: A multiscale analysis. Geoderma, 309, pp. 60–74. https://doi.org/10.1016/j.geoderma.2017.09.005

34. Platenburg R.J.P.M., Tuinhof H., Bot A.P., Iwaco B.V. (1988). Geostatistics in Soil Pollution Research. Contaminated Soil ‘88. Springer, Dordrecht, pp. 209–211. https://doi.org/10.1007/978-94-009-2807-7_32

35. Ryazanov S. S., Ivanov D. V., Kulagina V. I. (2019). Heavy metals in topsoils of the Republic of Tatarstan. Russian Journal of Ecosystem Ecology, 4(3), pp. 1–14. https://doi.org/10.21685/2500-0578-2019-3-4

36. Saby N., Thioulouse J., Jolivet C., Ratie C., Boulonne L., Bispo A., Arrouays D. (2009). Multivariate analysis of the spatial patterns of 8 trace elemets using the French monitoring network data. Science of the Total Environment, 407, pp. 5644–5652. https://doi.org/10.1016/j.scitotenv.2009.07.002

37. Sakizadeh M., Martín J.A.R. (2021). Spatial methods to analyze the relationship between Spanish soil properties and cadmium content. Chemosphere, 268, 129347. https://doi.org/10.1016/j.chemosphere.2020.129347

38. Schneckenburger T., Thiele-Bruhn S. (2020). Sorption of PAHs and PAH derivatives in peat soil is affected by prehydration status: the role of SOM and sorbate properties. J Soils Sediments, 20, pp. 3644–3655. https://doi.org/10.1007/s11368-020-02695-z

39. Sergeev A.P., Buevich A.G., Baglaeva E.M., Shichkin A.V. (2019). Combining spatial autocorrelation with machine learning increases prediction accuracy of soil heavy metals. Catena, 174, pp. 425–435. https://doi.org/10.1016/j.catena.2018.11.037

40. Shi B., Ngueleu S.K., Rezanezhad F., Slowinski S., Pronk G.J., Smeaton C.M., Stevenson K., Al-Raoush R.I., Van Cappellen P. (2020) Sorption and Desorption of the Model Aromatic Hydrocarbons Naphthalene and Benzene: Effects of Temperature and Soil Composition. Front. Environ. Chem, 1, 581103. https://doi.org/10.3389/fenvc.2020.581103

41. Shi T., Yang C., Liu H., Wu C., Wang Z., Li H., Zhang H., Guo L., Wu G., Su F. (2021). Mapping lead concentrations in urban topsoil using proximal and remote sensing data and hybrid statistical approaches. Environmental Pollution, 272, 116041. https://doi.org/10.1016/j.envpol.2020.116041

42. Smola A.J., Scholköpf B. (2004). A tutorial on support vector regression. Stat. Comput., 14, pp. 199–222. https://doi.org/10.1023/B:STCO.0000035301.49549.88

43. Taghizadeh-Mehrjardi R., Schmidt K., Toomanian N., Heung B., Behrens T., Mosavi A., Band S.S., Amirian-Chakan A., Fathabadi A., Scholten T. (2021). Improving the spatial prediction of soil salinity in arid regions using wavelet transformation and support vector regression models. Geoderma, 383, 114793. https://doi.org/10.1016/j.geoderma.2020.114793

44. Tarasov D.A., Buevich A.G., Sergeev A.P., Shichkin A.V. (2018). High variation topsoil pollution forecasting in the Russian Subarctic: Using artificial neural networks combined with residual kriging, Applied Geochemistry, 88, Part B, pp. 188–197. https://doi.org/10.1016/j.apgeochem.2017.07.007

45. Vincent S., Lemercier B., Berthier L., Walte C. (2018). Spatial disaggregation of complex Soil Map Units at the regional scale based on soil-landscape relationships. Geoderma, 311, pp. 130–142. https://doi.org/10.1016/j.geoderma.2016.06.006

46. Were K., Bui D.T., Dick Ø.B., Singh B.R. (2015). A comparative assessment of support vector regression, artificial neural networks, and random forests for predicting and mapping soil organic carbon stocks across an Afromontane landscape. Ecol. Indic., 52, pp. 394–403. https://doi.org/10.1016/j.ecolind.2014.12.028

47. Yuan G., Sun T., Han P., Li J. (2013). Environmental geochemical mapping and multivariate geostatistical analysis of heavy metals in topsoils of a closed steel smelter: Capital Iron and Steel Factory, Bejing, China. Journal of Geochemical Exploration, 130, pp. 15–21. https://doi.org/10.1016/j.gexplo.2013.02.010

48. Zhang L., Liu Y., Li X., Huang L., Yu D., Shi X., Chen H., Xing S. (2018). Effects of soil map scales on simulating soil organic carbon changes of upland soils in Eastern China. Geoderma, 312, pp. 159–169. https://doi.org/10.1016/j.geoderma.2017.10.017

49. Zwolak A., Sarzyńska M., Szpyrka E., Stawarczyk K. (2019). Sources of Soil Pollution by Heavy Metals and Their Accumulation in Vegetables: a Review. Water, Air, & Soil Pollution, 230(164). https://doi.org/10.1007/s11270-019-4221-y