Detecting and Mapping Stoebe vulgaris Post Herbicide Control, Using UAV Imagery and Machine Learning

Abstract

Stoebe vulgaris is a declared indigenous bush encroacher species in South Africa. It has invaded over 11 million ha of grasslands. It is commonly called bankrupt bush due to its ability to outcompete other indigenous forb and grass species, decreasing grazing capacity, biodiversity, and ecosystem functioning, eventually leading to financial ruin for farmers. Landowners are legally required to control the plant. A herbicide trial was set up in a severely encroached camp at Dundee Research Station in KwaZulu-Natal, South Africa, to test the effectiveness of metsulfuron-methyl (50 g active ingredient ha⁻¹) in controlling S. vulgaris. Applying metsulfuron-methyl provided a significant long-term reduction in S. vulgariscover over six years. However, effective monitoring and management strategies depend on knowledge of the spatial distribution and expansion patterns of invasive species. We evaluated the ability of UAV-based imagery and machine learning, using Picterra, to detect and map S. vulgaris, while determining the optimal parameters to maximise detection accuracy. The best season for image acquisition was late summer when vegetation was at peak growth and maturity, providing the best spectral distinction between species, under light overcast and mild wind conditions. We recommend careful consideration of the flight orientation to the solar angle. We achieved 92% detector accuracy, with multispectral imagery enhancing the discrimination of similarly coloured plants. Plants smaller than 10 cm were not detected by the model. Our approach, using high-resolution drone imagery and AI, is capable of individual plant detection suited to a farm scale. This opens the way for using advances in drone technology for targeted, spot-application of herbicide.

Keywords

Seriphium plumosum / Drones / Invasive species / Woody encroacher / Multispectral imagery / Colour imagery / Flight parameters / AI

References

[1] IPBES. Summary for Policymakers of the Thematic Assessment Report on Invasive Alien Species and Their Control of the Intergovernmental Science—Policy Platform on Biodiversity and Ecosystem Services; Roy HE, Pauchard A, Stoett P, Renard Truong T, Bacher S, Galil BS, et al., Eds.; IPBES Secretariat: Bonn, Germany, 2023. DOI: 10.5281/zenodo.7430692

[2] Stafford W, Birch C, Etter H, Blanchard R, Mudavanhu S, Angelstam P, et al. The economics of landscape restoration: Benefits of controlling bush encroachment and invasive plant species in South Africa and Namibia. Ecosyst. Serv. 2017, 27, 193-202. DOI: 10.1016/j.ecoser.2016.11.021

[3] Bakacsy L, Tobak Z, van Leeuwen B, Szilassi P, Biró C, Szatmári J. Drone—Based Identification and Monitoring of Two Invasive Alien Plant Species in Open Sand Grasslands by Six RGB Vegetation Indices. Drones 2023, 7, 207. DOI: 10.3390/drones7030207

[4] Zaka MM, Samat A. Advances in Remote Sensing and Machine Learning Methods for Invasive Plants Study: A Comprehensive Review. Remote Sens. 2024, 16, 3781. DOI: 10.3390/rs16203781

[5] Marquart A, Mangani T, Brown LR. The rise of Stoebe vulgaris (formally treated as Seriphium plumosum) in South Africa: A critical review of nearly a century of research. S. Afr. J. Bot. 2026, 190, 447-456. DOI: 10.1016/j.sajb.2026.01.013

[6] Snyman HA. Control measures for the encroacher shrub Seriphium plumosum . S. Afr. J. Plant Soil. 2012, 29, 157-163. DOI: 10.1080/02571862.2012.745905

[7] Avenant P. Report on the National Bankrupt Bush (Seriphium plumosum) Survey (2010—2012) ; Department of Agriculture, Forestry and Fisheries: Pretoria, South Africa, 2015; pp. 1-48.

[8] Shaik Z, Verboom GA, Oxelman B, Bergh NG. Revision of Seriphium plumosum (Gnaphalieae: Asteraceae) in southern Africa with description of new species in Seriphium . S. Afr. J. Bot. 2024, 165, 367-383. DOI: 10.1016/j.sajb.2023.12.046

[9] Snyman HA. Habitat preferences of the encroacher shrub, Seriphium plumosum . S. Afr. J. Bot. 2012, 81, 34-39. DOI: 10.1016/j.sajb.2012.05.001

[10] Graham SC, Barrett AS, Brown LR. Impact of Seriphium plumosum densification on Mesic Highveld Grassland biodiversity in South Africa. R. Soc. Open Sci. 2020, 7, 192025. DOI: 10.1098/rsos.192025

[11] Marquart A, Slooten E, Jordaan FP, Vermeulen M, Kellner K. The control of the encroaching shrub Seriphium plumosum (L.) Thumb. (Asteraceae) and the response of the grassy layer in a South African semi—arid rangeland. Afr. J. Range Forage Sci. 2023, 40, 316-321. DOI: 10.2989/10220119.2022.2086620

[12] Jordaan D. Bankruptbush (Slangbos)—A silent threat to grasslands? Grassroots Newsl. Grassl. Soc. South. Afr. 2009, 9, 40-42. Available online: https://grassland.org.za/sites/default/files/2023—11/February%202009.pdf (accessed on 27 April 2026).

[13] Snyman HA. Germination potential of Seriphium plumosum (bankrupt bush, slangbos or vaalbos) . Grassroots Newsl. Grassl. Soc. South. Afr. 2009, 9, 43-48. Available online: https://api.semanticscholar.org/CorpusID:131346038 (accessed on 27 April 2026).

[14] Adepoju K, Adelabu S, Mokubung C. Mapping Seriphium plumosum encroachment and interaction with wildfire and environmental factors in a protected mountainous grassland. Environ. Monit. Assess. 2020, 192, 328. DOI: 10.1007/s10661—020—08253—x

[15] Singini EJ, McMillan BM, Newete SW, Witkowski ETF. Ecological drivers of Seriphium plumosum encroachment: Implications for management and conservation. S. Afr. J. Bot. 2025, 176, 21-28. DOI: 10.1016/j.sajb.2024.11.001

[16] Snyman HA. Allelopathic potential, seed ecology, and germination of the encroacher shrub Seriphium plumosum . Afr. J. Range Forage Sci. 2010, 27, 29-37. DOI: 10.2989/10220111003718482

[17] CARA. Conservation of Agricultural Resources Act. Notice No. 883. 1983. Available online: https://www.gov.za/documents/notices/conservation—agricultural—resources—act—declaration—bankrupt—bush—seriphium (accessed on 19 February 2026).

[18] Nkosi SE, Brown LR, Barrett AS. A baseline study for controlling the indigenous encroacher Stoebe vulgaris in the natural grasslands of Southern Africa. Agric. Ecosyst. Environ. 2018, 265, 209-216. DOI: 10.1016/j.agee.2018.06.013

[19] Singh KK, Surasinghe TD, Frazier AE. Systematic review and best practices for drone remote sensing of invasive plants. Methods Ecol. Evol. 2024, 15, 998-1015. DOI: 10.1111/2041—210X.14330

[20] Baena S, Moat J, Whaley O, Boyd DS. Identifying species from the air: UAVs and the very high resolution challenge for plant conservation. PLoS ONE 2017, 12, e0188714. DOI: 10.1371/journal.pone.0188714

[21] Mathews AJ, Singh KK, Cummings AR, Rogers SR. Fundamental practices for drone remote sensing research across disciplines. Drone Syst. Appl. 2023, 11, 1-22. DOI: 10.1139/dsa—2023—0021

[22] Reddy DVSC, Sahoo RN, Kondraju T, Rejith RG, Ranjan R, Bhandari A, et al. Drone—Based Multispectral Imaging for Precision Monitoring of Crop Growth Variables. Biol. Life Sci. Forum 2025, 41, 10. DOI: 10.3390/blsf2025041010

[23] Rominger KR, Meyer SE. Drones, deep learning, and endangered plants: A method for population—level census using image analysis. Drones 2021, 5, 126. DOI: 10.3390/drones5040126

[24] Robinson JM, Harrison PA, Mavoa S, Breed MF. Existing and emerging uses of drones in restoration ecology. Methods Ecol. Evol. 2022, 13, 1899-1911. DOI: 10.1111/2041—210X.13912

[25] Mucina L, Rutherford MC. The Vegetation of South Africa, Lesotho and Swaziland ; Strelitzia 19; South African National Biodiversity Institute: Pretoria, South Africa, 2010; pp. 421-422.

[26] Jewitt D. Vegetation type conservation targets, status and level of protection in KwaZulu—Natal in 2016. Bothalia 2018, 48, a2294. DOI: 10.4102/abc.v48i1.2294

[27] du Toit JCO, Cronje WB, Trollope WSW. Towards a low—input control slangbos (Seriphium plumosum)—Quality and grazing interaction hypothesis. Grootfontein Agric. Dev. Inst. 2013, 13, 21-25. Available online: https://scholar.google.com/citations?view_op=view_citation&hl=en&user=7LEIzVgAAAAJ&cstart=20&pagesize=80&citation_for_view=7LEIzVgAAAAJ:LkGwnXOMwfcC (accessed on 27 April 2026).

[28] Clark MD, Wonkka CL, Kreuter UP, Rogers WE. Interactive effects of prescribed fire and livestock grazing on Seriphium plumosum in South African sour bushveld. Afr. J. Range Forage Sci. 2020, 37, 278-285. DOI: 10.2989/10220119.2020.1751287

[29] Goodall JM, Witkowski ETF. Selective herbicides extirpate forbs more effectively than alien plants, facilitating dominance of aggressive low grazing value grasses. S. Afr. J. Bot. 2025, 181, 523-536. DOI: 10.1016/j.sajb.2025.04.018

[30] du Toit JCO. Problems with using the herbicide Tebuthiuron for controlling Seriphium plumosum . Grootfontein Agric. Dev. Inst. 2012, 11, 31-40. Available online: https://www.researchgate.net/profile/Justin—Du—Toit/publication/261213758_du_Toit_JCO_2012_Problems_with_using_the_herbicide_tebuthiuron_for_controlling_Seriphium_plumosum_Grootfontein_Agric_1131—40_31/links/0046353396a9b48d69000000/du—Toit—JCO—2012—Problems—with—using—the—herbicide—tebuthiuron—for—controlling—Seriphium—plumosum—Grootfontein—Agric—1131—40—31.pdf (accessed on 9 March 2026).

[31] du Toit JCO, Sekwadi K. Tebuthiuron residues remain active in soil for at least eight years in a semi—arid grassland, South Africa. Afr. J. Range Forage Sci. 2012, 29, 85-90. DOI: 10.2989/10220119.2012.705325

[32] Pule HT, Tjelele JT, Tedder MJ. The effects of abiotic factors in South African semi—arid grassland communities on Seriphium plumosum L density and canopy size. PLoS ONE 2018, 13, e0202809. DOI: 10.1371/journal.pone.0202809

[33] Maes WH. Practical Guidelines for Performing UAV Mapping Flights with Snapshot Sensors. Remote Sens. 2025, 17, 606. DOI: 10.3390/rs17040606

[34] Leem J, Mehrishal S, Kang I—S, Yoon D—H, Shao Y, Song J—J, et al. Optimizing Camera Settings and Unmanned Aerial Vehicle Flight Methods for Imagery—Based 3D Reconstruction: Applications in Outcrop and Underground Rock Faces. Remote Sens. 2025, 17, 1877. DOI: 10.3390/rs17111877

[35] Ahmed S, El—Shazly A, Abed F, Ahmed W. The Influence of Flight Direction and Camera Orientation on the Quality Products of UAV—Based SfM—Photogrammetry. Appl. Sci. 2022, 12, 10492. DOI: 10.3390/app122010492

[36] Lu X, Xie J, Yan J, Zhou J, Cen H. Precise Image Color Correction Based on Dual Unmanned Aerial Vehicle Cooperative Flight. Plant Phenomics 2025, 7, 100101. DOI: 10.1016/j.plaphe.2025.100101

[37] Tocci F, Figorilli S, Vasta S, Violino S, Pallottino F, Ortenzi L, et al. Advantages in Using Colour Calibration for Orthophoto Reconstruction. Sensors 2022, 22, 6490. DOI: 10.3390/s22176490

[38] Tu YH, Phinn S, Johansen K, Robson A, Wu D. Optimising drone flight planning for measuring horticultural tree crop structure. ISPRS J. Photogramm. Remote Sens. 2020, 160, 83-96. DOI: 10.1016/j.isprsjprs.2019.12.006

[39] Asner GP, Martin RE. Spectral and chemical analysis of tropical forests: Scaling from leaf to canopy levels. Remote Sens. Environ. 2008, 112, 3958-3970. DOI: 10.1016/j.rse.2008.07.003

[40] Nkhwanana H, Adam E, Ramoelo A. Assessing the utility of Sentinel—2 MSI in mapping an encroaching Seriphium plumosum in South African rangeland. Appl. Geomat. 2022, 14, 435-449. DOI: 10.1007/s12518—022—00423—5

[41] Wang N, Guo Y, Wei X, Zhou M, Wang H, Bai Y. UAV—based remote sensing using visible and multispectral indices for the estimation of vegetation cover in an oasis of a desert. Ecol. Indic. 2022, 141, 109155. DOI: 10.1016/j.ecolind.2022.109155

[42] Papakonstantinou A, Stamati C, Topouzelis K. Comparison of True—Color and Multispectral Unmanned Aerial Systems Imagery for Marine Habitat Mapping Using Object—Based Image Analysis. Remote Sens. 2020, 12, 554. DOI: 10.3390/rs12030554

[43] Mazurek K, Zając Ł, Suchocka M, Jelonek T, Juźwiak A, Kubus M. UAV—Based Multispectral Imagery for Area—Wide Sustainable Tree Risk Management. Sustainability. 2025, 17, 8908. DOI: 10.3390/su17198908

[44] Rominger KR, Meyer SE. Application of UAV—based methodology for census of an endangered plant species in a fragile habitat. Remote Sens. 2019, 11, 719. DOI: 10.3390/rs11060719

[45] Urban M, Schellenberg K, Morgenthal T, Dubois C, Hirner A, Gessner U, et al. Using Sentinel—1 and Sentinel—2 Time Series for Slangbos Mapping in the Free State Province, South Africa. Remote Sens. 2021, 13, 3342. DOI: 10.3390/rs13173342

[46] Tesfamichael SG, Newete SW, Adam E. Dubula B. Field spectroradiometer and simulated multispectral bands for discriminating invasive species from morphologically similar cohabitant plants. GIScience Remote Sens. 2018, 55, 417-436. DOI: 10.1080/15481603.2017.1396658

[47] Dubula B, Tefsamichael SG, Rampedi IT. Assessing the potential of remote sensing to discriminate invasive Seriphium plumosum from grass. South Afr. J. Geomat. 2016, 5, 201-213. DOI: 10.4314/sajg.v5i2.8

[48] Nobre FLDL, Santos RF, Herrera JL, Araújo AL, Johann JA, Gurgacz F, et al. Use of drones in herbicide spot spraying: A systematic review. Adv. Weed Sci. 2023, 41, e020230014. DOI: 10.51694/AdvWeedSci/2023;41:00028

[49] Guebsi R, Mami S, Chokmani K. Drones in precision agriculture: A comprehensive review of applications, technologies, and challenges. Drones 2024, 8, 868. DOI: 10.3390/drones8110686

[50] Sühs RB, Ziller SR, Dechoum M. Is the use of drones cost—effective and efficient in detecting invasive alien trees? A case study from a subtropical coastal ecosystem. Biol. Invasions 2024, 26, 357-363. DOI: 10.1007/s10530—023—03190—5

[51] James K, Bradshaw K. Detecting plant species in the field with deep learning and drone technology. Methods Ecol. Evol. 2020, 11, 1509-1519. DOI: 10.1111/2041—210X.13473