Assessment of the distribution of Sosnowsky’s hogweed (Heracleum sosnowskyi Manden.) habitats using automated classification based on aerial photography data

Invasive plants pose a significant threat to local ecosystems and agriculture, and their rapid spread necessitates the development of effective methods for detection and control, one of which is mapping based on remote sensing data. Geoinformation mapping of invasive plant species, particularly Sosnowsky’s hogweed (Heracleum sosnowskyi Manden.), is a pressing task for ecological monitoring. This study examines modern methods for analyzing remote sensing data to identify areas occupied by Sosnowsky’s hogweed. The research was conducted in the Leningrad Region, where the spread of this plant is a critical issue. The input data consisted of aerial photographs and open-source data from the Yandex.Maps cartographic service with ground sample distance of 3, 15, and 30 cm/pixel, which depicted territories overgrown with Sosnowsky’s hogweed. The objectives of the study were to develop an image processing technique for detecting Sosnowsky’s hogweed based on the use of automated classification algorithms. The data processing was performed using the QGIS geoinformation system and its additional plugins, «dzetsaka» and «Orfeo Toolbox». The accuracy of the research results was verified using data obtained through expert classification and manual vectorization. As a result, 24 classification maps were generated using different methods of automated classification. Accuracy analysis produced error matrices and tables showing the deviation in the areas occupied by Sosnowsky’s hogweed. Nevertheless, the accuracy assessment confirmed the high reliability of the presented classification methods. The algorithms developed in this study can be applied to detect and monitor Sosnowsky’s hogweed in the Leningrad Region.

1. Aegopodium podagraria // Plants for a future. URL: https://pfaf.org/user/plant. aspx?LatinName=Aegopodium+podagraria (accessed: 05.05.2025).

2. Basheer S., Wang X., Farooque A.A., Nawaz R.A., Liu K., Adekanmbi T., Liu S. Comparison of Land Use Land Cover Classifiers Using Different Satellite Imagery and Machine Learning Techniques. Remote Sensing, 2022., vol. 14, no. 19, art. no. 4978.

3. Bogdanov V.L., Osipov A.G., Garmanov V.V. Methodology for monitoring land contamination with Sosnowsky's hogweed according to remote sensing data. Izvestiya Saint-Petersburg state agrarian university, 2020, no. 59, pp. 69-74. (In Russ.)

4. Bogdanov V., Osipov A., Garmanov V., Efimova G., Grik A., Zavarin B., Terleev V., Nikonorov A. Problems and monitoring the spread of the ecologically dangerous plant Heracleum sosnowskyi in urbanized areas and methods to combat it. E3S Web Conf. EDP Sciences, 2021, vol. 258, art. no. 08028.

5. Congalton R.G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sensing of Environment, 1991, vol. 37. pp. 35-46.

6. Dalke I.V., Chadin I.F., Zakhozhy I.G., Malyshev R.V., Maslova S.P. Modeling of geographical distribution limits of Heraculum sosnowskyi Manden. in the taiga zone of the European part of Russia. Ecology and geography of plants and plant communities: proceedings of the IV int. sci. conf. Yekaterinburg, 2018, pp. 219-224. (In Russ.)

7. Dueñas M.A., Hemming D.J., Roberts A., Diaz-Soltero H. The threat of invasive species to IUCN-listed critically endangered species: A systematic review. Glob. Ecol. Conserv., 2021, vol. 26, art. no. 01476.

8. Jahodova S. Heracleum sosnowskyi (Sosnowskyi’s hogweed). CABI Compendium, 2009, no. 108958. DOI: 10.1079/cabicompendium.108958.

9. Gurchenkov A.A., Murynin A.B., Trekin A.N., Ignatiev V.Yu. Object-oriented classification of substrate surface objects in Arctic impact regions aerospace monitoring. Herald of the Bauman Moscow State Technical University. Series: Natural Sciences, 2017, no. 3, pp. 135-146. (In Russ.)

10. Koldasbayeva D., Tregubova P., Shadrin D., Gasanov M., Pukalchik M. Largescale forecasting of Heracleum sosnowskyi habitat suitability under the climate change on publicly available data. Scientific Reports, 2022, vol. 12, art. no. 6128.

11. Marpu P.R., Niemeyer I., Nussbaum S., Gloaguen R. A procedure for automatic object-based classification. Lecture Notes in Geoinformation and Cartography. Berlin, 2008, chapter 2.4, pp. 169-184.

12. Netsvetova E.V., Salovarov V.O. The fight against invasive alien species in the concept of the Convention on Biological Diversity. Vestnik IrGSHA, 2022, no. 112, pp. 121-136. (In Russ.)

13. Norovsuren B., Tseveen B., Batomunkuev V., Renchin T., Natsagdorj E., Yangiv A., Mart Z. Land cover classification using maximum likelihood method (2000 and 2019) at Khandgait valley in Mongolia. IOP Conference Series: Earth and Environmental Science, 2019, vol. 381, no. 1, art. no. 012054.

14. Osipov A.G., Dmitriev V.V., Kovyazin V.F. Methodology for assessing and mapping the natural and agricultural potential of landscapes. Geodesy and Cartography, 2021, vol. 82, no. 9, pp. 11-20. (In Russ.)

15. Rutman V.V., Kantor G.Ya., Adamovich T.A., Tovstik E.V., Ashikhmina T.Ya. Identification of thickets of Sosnowsky's hogweed according to remote sensing of the Earth. Ecology of the native land: problems and ways to solve them: mat. of the XIII All-Russ. sci.-pract. conf. with int. part. Kirov, 2018, pp. 81-85. (In Russ.)

16. Satellite map of St. Petersburg. Yandex.Maps. URL: https://yandex.ru/maps/2/saint-petersburg/sputnik/ (accessed: 01.06.2024). (In Russ.)

17. Shetty S. Analysis of Machine Learning Classifiers for LULC Classification on Google Earth Engine: master’s thesis. Enschede, The Netherlands: University of Twente, 2019. 73 p.

18. Talukdar S., Singha P., Mahato S., Fahad S., Pal S., Liou Y-A., Rahman A. Landuse land-cover classification by machine learning classifiers for satellite observations – a review. Remote Sensing, 2020, vol. 2, no. 7, art. no. 1135.

19. Xie Z., Chen Y., Lu D., Li G., Chen E. Classification of land cover, forest, and tree species classes with Ziyuan-3 multispectral and stereo data. Remote Sensing, 2019, vol. 11, no. 2, art. no. 164.