(47-1) 16 * << * >> * Russian * English * Content * All Issues

Central Russia heavy metal contamination model based on satellite imagery and machine learning
A. Uzhinskiy 1, K. Vergel 1

Joint Institute for Nuclear Research, 141980, Dubna, Moscow region, Russia, 6 Joliot-Curie

 PDF, 10 MB

DOI: 10.18287/2412-6179-CO-1149

Pages: 137-151.

Full text of article: English language.

Atmospheric heavy metal contamination is a real threat to human health. In this work, we examined several models trained on in situ data and indices got from satellite images. During 2018-2019, 281 samples of naturally growing mosses were collected in the Vladimir, Yaroslavl, and Moscow regions in Russia. The samples were analyzed using Neutron Activation Analysis to get the contamination levels of 18 heavy metals. The Google Earth Engine platform was used to calculate indices from satellite images that represent summarized information about sampling sites. Statistical and neural models were trained on in situ data and the indices. We focused on the classification task with 8 levels of contamination and used balancing techniques to extend the training data. Three approaches were tested: variations of gradient boosting, multilayer perceptron, and Siamese networks. All these approaches produced results with minute differences, making it difficult to judge which one is better in terms of accuracy and graphical outputs. Promising results were shown for 9 heavy metals with an overall accuracy exceeding 89%. Al, Fe, and Sb contamination was predicted for 3,000 and 12,100 grid nodes on a 500 km2 area in the Central Russia region for 2019 and 2020. The results, methods, and perspectives of the adopted approach of using satellite data together with machine learning for HM contamination prediction are presented.

heavy metal contamination, modeling, air pollution, biomonitors, prediction, satellite imagery, machine learning, neural architectures, Siamese neural networks.

Uzhinskiy A, Vergel K. Central Russia heavy metal contamination model based on satellite imagery and machine learning. Computer Optics 2023; 47(1): 137-151. DOI: 10.18287/2412-6179-CO-1149.


  1. Uzhinskiy A. Intelligent environmental monitoring platform. CEUR Workshop Proc 2019; 2267: 351-358.
  2. Harmens H, Norris DA, Steinnes E, Kubin E, Piispanen J, Alber R, Aleksiayenak Y, Blum O, Coskun M, Dam M, De Temmerman L, Fernández JA, Frolova M, Frontasyeva M, González-Miqueo L, Grodzinska K, Jeran Z, Korzekwa S, Krmar M, Kvietkus K, Leblond S, Liiv S, Magnússon SH, Mankovská B, Pesch R, Rühling Å, Santamaria JM, Schröder W, Spiric Z, Suchara I, Thöni L, Urumov V, Yurukova L, Zechmeister HG. Mosses as biomonitors of atmospheric heavy metal deposition: spatial patterns and temporal trends in Europe. Environ Pollut 2010; 158: 3144-3156.
  3. Yuan Y, Wu Y, Ge X, Nie D, Wang M, Zhou H, Chen M. In vitro toxicity evaluation of heavy metals in urban air particulate matter on human lung epithelial cells. Sci Total Environ 2019; 678: 301-308.
  4. Jakubowski M. Biological monitoring versus air monitoring strategies in assessing environmental-occupational exposure. J Environ Monit 2012; 14(2): 348-352. DOI: 10.1039/c1em10706b.
  5. Holt EA, Miller SW. Bioindicators: Using organisms to measure environmental impacts. Nature Education Knowledge 2010; 3(10): 8.
  6. Uzhinskiy A, Urošević MA, Frontasyeva M. Prediction of air pollution by potentially toxic elements over urban area by combining satellite imagery, moss biomonitoring data and machine learning. Cienc e Tec Vitivinic J 2020; 35(12): 34-46.
  7. Uzhinskiy A, Ososkov G, Goncharov P, Frontsyeva M. Combining satellite imagery and machine learning to predict atmospheric heavy metal contamination. CEUR Workshop Proc 2018; 2267: 351-358.
  8. Salo H, Mäkinen J. Magnetic biomonitoring by moss bags for industry-derived air pollution in SW Finland. Atmos Environ 2014: 97: 19-27. DOI: 10.1016/j.atmosenv.2014.08.003.
  9. Ben-Dor E, Irons JR, Epema GF. Soil reflectance. In Book: Rencz AN, Ryerson RA, eds. 3rd ed, Vol 3. Manual of remote sensing: Remote sensing for the earth sciences. New York: John Wiley & Sons Inc; 1999: 111-189.
  10. Kemper T, Sommer S. Estimate of heavy metal contamination in soils after a mining accident using reflectance spectroscopy. Environ Sci Technol 2002; 36(12): 2742-2747.
  11. Choe E, van der Meer F, van Ruitenbeek F, van der Werff H, de Smeth B, Kim KW. Mapping of heavy metal pollution in stream sediments using combined geochemistry, field spectroscopy, and hyperspectral remote sensing: a case study of the Rodalquilar mining area, SE Spain. Remote Sens Environ 2008; 112(7): 3222-3233.
  12. Ren H-Y, Zhuang D-F, Singh AN, Pan J-J, Qui D-S, Shi R-H. Estimation of As and Cu contamination in agricultural soils around a mining area by reflectance spectroscopy: a case study. Pedosphere 2009; 19(6): 719-726.
  13. Beloconi A, Chrysoulakis N, Lyapustin A, Utzinger J, Vounatsou P., Bayesian geostatistical modelling of PM10 and PM2.5 surface level concentrations in Europe using high-resolution satellite-derived products. Environ Int 2018; 121(1): 57-70. DOI: 10.1016/j.envint.2018.08.041.
  14. Alvarez-Mendoza CI, Teodoro AC, Torres N, Vivanco V. Assessment of remote sensing data to model PM10 estimation in cities with a low number of air quality stations: A case of study in Quito, Ecuador. Environments 2019; 6: 85. DOI: 10.3390/environments6070085.
  15. Zheng T, Bergin MH, Hu S, Miller J, Carlson DE. Estimating ground-level PM2.5 using micro-satellite images by a convolutional neural network and random forest approach. Atmospheric Environ 2020; 230: 117451. DOI: 10.1016/j.atmosenv.2020.117451.
  16. Muradyan V, Tepanosyan G, Asmaryan S, Maghakyan N, Sahakyan L, Saghatelyan A. Estimating Mo, Cu, Ni, Cd contents in the crop leaves growing on small land plots using satellite data. Commun Soil Sci Plant Anal 2020; 51(11): 1457-1468. DOI: 10.1080/00103624.2020.1784922.
  17. Liu M, Wang T, Skidmore AK, Liu X. Heavy metal-induced stress in rice crops detected using multi-temporal Sentinel-2 satellite images. Sci Total Environ 2018; 637-638: 18-29. DOI: 10.1016/j.scitotenv.2018.04.415.
  18. Amer M, Tyler A, Foudat T, Hunter P, Elmetwalli A, Wilson C, Vallejo-Marin M. Spectral characteristics for estimation heavy metals accumulation in wheat plants and grain. J Product Dev 2017; 22(3): 409-428.
  19. Zhou C, Chen S, Zhang Y, Zhao J, Song D, Liu D. Evaluating metal effects on the reflectance spectra of plant leaves during different seasons in post-mining areas, China. Remote Sens 2018; 10(8): 1211. DOI: 10.3390/rs10081211.
  20. Yu K, Van Geel M, Ceulemans T, Geerts W, Ramos MM, Serafim C, Sousa N, Castro PML, Kastendeuch P, Najjar G, Ameglio T, Ngao J, Saudreau M, Honnay O, Somers B. Vegetation reflectance spectroscopy for biomonitoring of heavy metal pollution in urban soils. Environ Pollut 2018; 243(B): 1912-1922. DOI: 10.1016/j.envpol.2018.09.053.
  21. Bjerke JW, Tømmervik H, Finne TE, Jensen H, Lukina N, Bakkestuen V. Epiphytic lichen distribution and plant leaf heavy metal concentrations in the Russian–Norwegian boreal forests influenced by air pollution from nickel–copper smelters. Boreal Env Res 2006; 11: 441-450.
  22. Khosropour E, Attarod P, Shirvany A, et al. Response of Platanusorientalis leaves to urban pollution by heavy metals. J For Res 2019; 30: 1437-1445. DOI: 10.1007/s11676-018-0692-8.
  23. Alahabadi A, Ehrampoush MH, Miri M, Ebrahimi Aval H, Yousefzadeh S, Ghaffari HR, Ahmadi E, Talebi P, AbaszadehFathabadi Z, Babai F, Nikoonahad A, SharafiK, Hosseini-Bandegharaei A. A comparative study on capability of different tree species in accumulating heavy metals from soil and ambient air. Chemosphere 2017; 172: 459-467. DOI: 10.1016/j.chemosphere.2017.01.045.
  24. Terekhina NV, Ufimtseva MD. Leaves of trees and shrubs as bioindicators of air pollution by particulate matter in Saint Petersburg. Geogr Environ Sustain 2020; 13(1): 224-232. DOI: 10.24057/2071-9388-2019-65.
  25. Lyanguzova I, Yarmishko V, Gorshkov V, Stavrova NN, Bakkal I. Impact of heavy metals on forest ecosystems of the European North of Russia. In: Saleh HEM, Aglan RF, eds. Heavy metals. London: IntechOpen; 2018. DOI: 10.5772/intechopen.73323.
  26. Lassalle G, Fabre S, Credoz A, et al. Mapping leaf metal content over industrial brownfields using airborne hyperspectral imaging and optimized vegetation indices. Sci Rep 2021; 11: 2. DOI: 10.1038/s41598-020-79439-z.
  27. Liu Z, Lu Y, Peng Y, Zhao L, Wang G, Hu Y. Estimation of Soil heavy metal content using hyperspectral data. Remote Sens 2019; 11(12): 1464.
  28. Gholizadeh A, Saberioon M, Ben-Dor E, Borůvka L. Monitoring of selected soil contaminants using proximal and remote sensing techniques: Background, state-of-the-art and future perspectives. Crit Rev Environ Sci Technol 2018; 48(3): 243-278.
  29. Gholizadeh A, Coblinski JA, Saberioon M, Ben-Dor E, Drábek O, Demattê JAM, Borůvka L, Němeček K, Chabrillat S, Dajčl J. vis-NIR and XRF data fusion and feature selection to estimate potentially toxic elements in soil. Sensors 2021; 21(7): 2386. DOI: 10.3390/s21072386.
  30. Ahado SK, Nwaogu C, Sarkodie VYO, Borůvka L. Modeling and assessing the spatial and vertical distributions of potentially toxic elements in soil and how the concentrations differ. Toxics 2021; 9(8): 181. DOI: 10.3390/toxics9080181.
  31. Michaelides S, Paronis D, Retalis A, Tymvios F. Monitoring and forecasting air pollution levels by exploiting satellite, groundbased, and synoptic data, elaborated with regression models. Adv Meteorol 2017; 2017: 2954010.
  32. Foldi C, Sauermann S, Dohrmann R, Mansfeldt T. Traffic-related distribution of antimony in roadside soils. Environ Pollut 2018; 237: 704-712.
  33. Goddard SL, Williams KR, Robins C, et al. Determination of antimony and barium in UK air quality samples as indicators of nonexhaust traffic emissions. Environ Monit Assess 2019; 191(11): 641.
  34. Fang Y, Xu L, Peng J, Wang H, Wong A, Clausi DA. Retrieval and mapping of heavy metal concentration in soil using time series landsat 8 imagery. ISPRS International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 2018; XLII-3: 335-340.
  35. Xu X, Chen S, Ren L, Han C, Lv D, Zhang Y, Ai F. Estimation of heavy metals in agricultural soils using vis-NIR spectroscopy with fractional-order derivative and generalized regression neural network. Remote Sens 2021; 13: 2718. DOI: 10.3390/rs13142718.
  36. Pyo JC, Hong SM, Kwon YS, Kim MS, Cho KH. Estimation of heavy metals using deep neural network with visible and infrared spectroscopy of soil. Sci Total Environ 2020; 741: 140162. DOI: 10.1016/j.scitotenv.2020.140162.
  37. Schroff F, Kalenichenko D, Philbin J. FaceNet: A unified embedding for face recognition and clustering. arXiv Preprint. 2015. Source: <https://arxiv.org/abs/1503.03832>.
  38. Cheng D, Gong Y, Zhou S, Wang J, Zheng N. Person re-identification by multi-channelparts-based cnn with improved triplet loss function. 2015 IEEE Conf on Computer Vision and Pattern Recognition (CVPR) 2015: 1335-1344.
  39. Hermans A, Beyer L, Leibe B. In defense of the triplet loss for person reidentification. arXiv Preprint. 2017. Source: <https://arxiv.org/abs/1703.07737>.
  40. Dong X, Shen J. Triplet loss in Siamese network for object tracking. In Book: Ferrari V, Hebert M, Sminchisescu C, Weiss Y, eds. Computer Vision – ECCV 2018. Cham: Springer Nature Switzerland AG; 2018: 472-488.
  41. Puch S, Sánchez I, Rowe M. Few-shot learning with deep triplet networks for brain imaging modality recognition. In Book: Wang Q, Milletari F, Nguyen HV, Albarqouni S, Cardoso MJ, Rieke N, Xu Z, Kamnitsas K, Patel V, Roysam B, Jiang S, Zhou K, Luu K, Le N, eds. Domain adaptation and representation transfer and medical image learning with less labels and imperfect data. Cham: Springer Nature Switzerland AG; 2019: 181-189.
  42. Anshul T, Daksh T, Padmanabhan R, Aditya N. Deep metric learning for bioacoustic classification: Overcoming training data scarcity using dynamic triplet loss. J Acoust Soc Am 2019; 146: 534-547.
  43. Zhang J, Lu C, Wang J, Yue X, Lim S, Al-Makhadmeh Z, Tolba A. Training convolutional neural networks with multi-size images and triplet loss for remote sensing scene classification. Sensors 2020; 20(4): 1188.
  44. Rühling A, Tyler G. An ecological approach to the lead problem. Botaniska Notiser 1968; 121: 321-342.
  45. Markert BA, Breure AM, Zechmeister HG. Definitions, strategies and principles for bioindications/biomonitoring of the environment, In Book: Markert BA, Breure AM, Zechmeister HG, eds. Bioindicators & biomonitors: Principles, concepts and applications. Vol 6. Pergamon; 2003: 3-39.
  46. Frontasyeva MV. Neutron activation analysis in the life sciences. Phys Part Nucl 2011; 42: 332-378. DOI: 10.1134/S1063779611020043.
  47. CLRTAP. Manual on methodologies and criteria for modeling and mapping critical loads and levels and air pollution effects, risks and trends. UNECE Convention on Long-range Transboundary Air Pollution; 2015. Source: <https://icpvegetation.ceh.ac.uk>.
  48. Yin F, Lewis PE, Gomez-Dans J, Wu Q. A sensor-invariant atmospheric correction method: application to Sentinel-2/MSI and Landsat 8/OLI. EarthArXiv Preprint. 2019. Source:              <https://eartharxiv.org/repository/view/1034/>. DOI: 10.31223/osf.io/ps957.
  49. Friedman JH. Greedy function approximation: A gradient boosting machine. Ann Stat 2001; 29(5): 1189-1232.
  50. He H, Bai Y, Garcia EA, Li S. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. IEEE Int Joint Conf on Neural Networks 2008: 1322-1328. DOI: 10.1109/IJCNN.2008.4633969.
  51. Chen C, Liaw A, Breiman L, et al. Using random forest to learn imbalanced data. Berkeley: University of California; 2004.
  52. Chen T, Guestrin C. XGBoost: A scalable tree boosting system. KDD '16: Proc 22nd ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining 2016: 785-794. DOI: 10.1145/2939672.2939785.
  53. Hertz J, Krogh A, Palmer RG. Introduction to the theory of neural computation. 1st ed. CRC Press; 1991. DOI: 10.1201/9780429499661.
  54. Nair V, Hinton G. Rectified linear units improve restricted boltzmann machines. Proc 27th Int Conf on Machine Learning 2010: 807-814.
  55. Kingma DP, Ba J. Adam: A method for stochastic optimization. 3rd Int Conf on Learning Representations (ICLR). 2015. Source: <https://arxiv.org/abs/1412.6980>.
  56. Abdollahi J, Emrani N, Chahkandi B, at al. Environmental impact assessment of aluminium production using the life cycleassessment tool and multi-criteria analysis. Ann Environ Sci Toxicol 2021; 5(1): 059-066. DOI: 10.17352/aest.000038.
  57. Paraskevas D, Kellens K, Van de Voorde A, Dewulf W, Duflou JR. Environmental impact analysis of primary aluminium production at country level. Procedia CIRP 2016; 40: 209-213.
  58. Nkansah MA, Agorsor PI, Opoku F. Heavy metal contamination and health risk assessment of mechanically milled delicacy called fufu. Int J Food Contam 2021; 8: 6. DOI: 10.1186/s40550-021-00085-y.
  59. Zhang X, Gao S, Fu Q, Han D, Chen X, Fu S, Huang X, Cheng J. Impact of VOCs emission from iron and steel industry on regional O3 and PM2.5 pollutions. Environ Sci Pollut Res 2020; 27: 28853-28866. DOI: 10.1007/s11356-020-09218-w.
  60. Wang R, Balkanski Y, Boucher O, Bopp L, Chappell A, Ciais P, Hauglustaine D, Peñuelas J, Tao S. Sources, transport and deposition of iron in the global atmosphere. Atmospheric Chem Phys 2015; 15: 6247-6270. DOI: 10.5194/acp-15-6247-2015.

© 2009, IPSI RAS
151, Molodogvardeiskaya str., Samara, 443001, Russia; E-mail: journal@computeroptics.ru ; Tel: +7 (846) 242-41-24 (Executive secretary), +7 (846) 332-56-22 (Issuing editor), Fax: +7 (846) 332-56-20