Microplastic Contamination Assessment in Urban Stormwater Runoff UsingHyperspectral Imaging and Machine Learning: A Case Study of the İzmirMetropolitan Drainage Network, Turkey
Keywords:
Microplastics, Hyperspectral Imaging, Random Forest, Polymer Classification, Water QualityAbstract
Microplastic (MP) pollution in urban stormwater systems represents a critical but undercharacterised environmental pathway, with stormwater runoff constituting a primary vector transporting terrestrially originating MPs to receiving water bodies. This paper presents a hyperspectral imaging and machine learning framework for rapid, non-destructive quantification and polymer-type classification of microplastics in stormwater samples collected from the İzmir Metropolitan Municipality drainage network, Turkey. A shortwave infrared (SWIR, 900 - 1700 nm) hyperspectral camera coupled to a conveyor-belt scanning platform acquires spectral data cubes from filtered stormwater membrane samples. A random forest (RF) classifier trained on spectral signatures of seven reference polymer types achieves a mean classification accuracy of 91.3% and a mean F1-score of 0.893 on held-out test samples. Gradient-boosted tree regression
models trained on environmental and catchment-scale predictor variables including impervious surface fraction, traffic density, and antecedent dry period explain 76% of variance in MP particle concentration across 38 drainage sub-catchments (R² = 0.76, RMSE = 412 particles/L). SHAP analysis identifies impervious surface fraction and commercial land-use proportion as the dominant drivers of MP concentration. The framework enables basin-scale MP load mapping at operational throughput (80 samples/day) and provides decision-support outputs for targeted source-control intervention.
References
[1] Y. Xu, Q. He, Z. Li, C. Xu, X. Zhao, Z. Mao, and X. Shi, "Advances in machine learning for the detection and characterization of microplastics in the environment," Front. Environ. Sci., vol. 13, pp. 1573579, Mar. 2025.
[2] S. Ziajahromi, P. Kumar Neale, and F. D. Leusch, "Microplastics and tire wear particles in urban stormwater: Abundance, characteristics, and potential mitigation strategies," Environ. Sci. Technol., 2023.
[3] W. Ai, S. Liu, H. Liao, J. Du, Y. Cai, C. Liao, H. Shi, Y. Lin, M. Junaid, X. Yue, and J. Wang, "Application of hyperspectral imaging technology in the rapid identification of microplastics in farmland soil," Sci. Total Environ., vol. 807, pp. 151030, Feb. 2022.
[4] M. Serry, A. Dujardin, and C. Schwarz, "Assessment of environmental and socioeconomic drivers of urban stormwater microplastics using machine learning," npj Clean Water, 2025.
[5] B. Shi, M. Patel, D. Yu, J. Yan, Z. Li, D. Petriw, T. Pruyn, K. Smyth, E. Passeport, R. J. D. Miller, and J. Y. Howe, "Automatic quantification and classification of microplastics in scanning electron micrographs via deep learning," Sci. Total Environ., vol. 825, pp. 153903, Jun. 2022.
[6] M. Stuart, M. Davies, M. Hobbs, T. Pering, A. McGonigle, and J. Willmott, "High-resolution hyperspectral imaging using low-cost components: Application within environmental monitoring scenarios," Sensors, vol. 22, no. 12, pp. 4652, Jun. 2022.
[7] N. B. Beni, S. Karimifard, J. Gilley, T. Messer, A. Schmidt, and S. Bartelt-Hunt, "Higher concentrations of microplastics in runoff from biosolid-amended croplands than manure-amended croplands," Commun. Earth Environ., vol. 4, no. 1, pp. 42, 2023.
[8] X. Hu, C. Xie, Z. Fan, Q. Duan, D. Zhang, L. Jiang, X. Wei, D. Hong, G. Li, and X. Zeng, "Hyperspectral anomaly detection using deep learning: A review," Remote Sens., vol. 14, no. 8, pp. 1973, Apr. 2022.
[9] International Commission for the Protection of the Danube River, Danube River Basin Management Plan 2021 Update. Vienna: ICPDR, 2021.
[10] A. Turkyilmaz, M. Ozkan, and H. Engin, "Assessment of plastic pollution in Turkish coastal waters using Raman spectroscopy," Environ. Pollut., vol. 287, pp. 117629, Oct. 2021.
[11] B. B. Sezer, H. Turkmen, and U. Nuriyev, "PPFchain: A novel framework privacy-preserving blockchain-based federated learning method for sensor networks," Internet of Things, vol. 22, pp. 100781, Jul. 2023.
[12] K. Luo, M. Tang, C. Luo, and J. Guo, "Identification of microplastics in environmental matrices using hyperspectral imaging: Optimizing pre-processing and spectral wavelength selection," J. Hazard. Mater., vol. 421, pp. 126684, Jan. 2022.
[13] European Parliament, Directive 2000/60/EC Establishing a Framework for Community Action in the Field of Water Policy. Luxembourg: Official Journal of the European Communities, 2000.
[14] H. Arp, C. Bogdal, and K. Oetken, "Microplastic levels in stormwater show high variability between events and catchments," Sci. Total Environ., vol. 834, pp. 155166, Aug. 2022.
[15] M. Schwarzer, J. Brehm, M. Vollmer, J. Jasinski, C. Xu, S. Zainuddin, T. Fröhlich, M. Schott, A. Greiner, T. Scheibel, and C. Laforsch, "Shape, size, and polymer dependent effects of microplastics on Daphnia magna," J. Hazard. Mater., vol. 426, pp. 128136, 2022.
[16] P. Corcoran, C. Biesinger, and M. Grifi, "Plastics and beaches: A degrading relationship," Mar. Pollut. Bull., vol. 58, no. 1, pp. 80–84, 2009.
[17] R. Coyle, M. Service, U. Witte, G. Hardiman, and J. McKinley, "Modeling microplastic transport in the marine environment," ACS ES&T Water, vol. 3, no. 4, pp. 984–995, Apr. 2023.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 BAUDER Press
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Authors submitting a manuscript retain copyright of their work upon acceptance for publication in Aintelia Science Notes Journal (ASNJ). By submitting, authors agree that:
- Copyright Retention: Copyright of the published article remains with the author(s). BAUDER Press is granted a non-exclusive license to publish, reproduce, distribute, and archive the work.
- Licensing: The article is published under CC BY-NC 4.0, allowing third parties to share and adapt the work for non-commercial purposes, provided appropriate credit is given to the author(s) and the journal.
- Author Rights: Authors are free to deposit, distribute, and reuse their article in any medium (repositories, theses, conferences, teaching materials) without restriction, provided the journal is cited as the original publication venue.