CrossBorder Mapping of Prosopis juliflora Invasion in the Dolo Ado and Dolow/Dolo Bay Landscape Using Remote Sensing and Support Vector Machine Classification in ArcGIS

Abstract The Dolo Ado (Somali Region, Ethiopia) and Dolow/Dolo Bay (Gedo Region, Jubaland, Somalia) landscape is shared by both Ethiopia and Somalia along the Dawa and Ganale/Jubba river systems, and is home to large pastoral, agropastoral and refugee-affected populations relying on shared rangeland and riverine resources. Prosopis juliflora was introduced in the wider Horn of Africa during the 1980s for dune stabilization, fuelwood production and land rehabilitation and is today one of the most devastating invasive plant species in this transboundary region but an assessment of its spatial extent across both sides of the border does not exist. In the present study, the coverage of Prosopis juliflora was mapped and mapped distribution for 2015 and 2025 were carried out in the combined Dolo Ado – Dolow landscape with Support Vector Machine (SVM) supervised classification method in ArcGIS. A reference year was chosen that is identical in the two years and the same bands and vegetation indices used in the Sentinel-2 Multispectral Instrument (MSI) were used to ensure the consistency of the imagery. For the 2015 and 2025 reference years, an overall classification accuracy of 84.22% and 86.54% (Kappa = 0.81 and 0.85) was obtained respectively, with an accuracy of 82.83% and 84.43% obtained for the class Prosopis (producer's accuracy) and 83.71% and 83.82% (user's accuracy) in the two years respectively. Across the combined study landscape, Prosopis cover increased from 10.18 km² (3.79% of the landscape) in 2015 to 18.85 km² (7.01% of the landscape) in 2025, representing an increase of 63.55% over the decade (or an average of 8.51% per year), with divergent rates of spread on the Ethiopian (Dolo Ado) and Somali (Dolow) side of the border (around 69.24% and 30.76% per year, respectively). The findings give a first-ever cross-border evidence base to support coordinated planning for Prosopis control, rangeland restoration and natural resource management in refugee-hosting areas in Ethiopia and Somali.

Authors

Institutions

Publication Details

Journal
Research Square
Published
2026-11-11
DOI
https://doi.org/10.21203/rs.3.rs-11175162/v1
Primary Topic
Remote-Sensing Image Classification
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

CrossBorder Mapping of Prosopis juliflora Invasion in the Dolo Ado and Dolow/Dolo Bay Landscape Using Remote Sensing and Support Vector Machine Classification in ArcGIS

Abdi Mohammed, Tesfu Abebe, Mohamed Abdi
Research Square
Remote-Sensing Image Classification
preprint

CrossBorder Mapping of Prosopis juliflora Invasion in the Dolo Ado and Dolow/Dolo Bay Landscape Using Remote Sensing and Support Vector Machine Classification in ArcGIS

Abdi Mohammed, Tesfu Abebe, Mohamed Abdi
preprint en

Abstract

Abstract The Dolo Ado (Somali Region, Ethiopia) and Dolow/Dolo Bay (Gedo Region, Jubaland, Somalia) landscape is shared by both Ethiopia and Somalia along the Dawa and Ganale/Jubba river systems, and is home to large pastoral, agropastoral and refugee-affected populations relying on shared rangeland and riverine resources. Prosopis juliflora was introduced in the wider Horn of Africa during the 1980s for dune stabilization, fuelwood production and land rehabilitation and is today one of the most devastating invasive plant species in this transboundary region but an assessment of its spatial extent across both sides of the border does not exist. In the present study, the coverage of Prosopis juliflora was mapped and mapped distribution for 2015 and 2025 were carried out in the combined Dolo Ado – Dolow landscape with Support Vector Machine (SVM) supervised classification method in ArcGIS. A reference year was chosen that is identical in the two years and the same bands and vegetation indices used in the Sentinel-2 Multispectral Instrument (MSI) were used to ensure the consistency of the imagery. For the 2015 and 2025 reference years, an overall classification accuracy of 84.22% and 86.54% (Kappa = 0.81 and 0.85) was obtained respectively, with an accuracy of 82.83% and 84.43% obtained for the class Prosopis (producer's accuracy) and 83.71% and 83.82% (user's accuracy) in the two years respectively. Across the combined study landscape, Prosopis cover increased from 10.18 km² (3.79% of the landscape) in 2015 to 18.85 km² (7.01% of the landscape) in 2025, representing an increase of 63.55% over the decade (or an average of 8.51% per year), with divergent rates of spread on the Ethiopian (Dolo Ado) and Somali (Dolow) side of the border (around 69.24% and 30.76% per year, respectively). The findings give a first-ever cross-border evidence base to support coordinated planning for Prosopis control, rangeland restoration and natural resource management in refugee-hosting areas in Ethiopia and Somali.

Research Square
University of Nairobi (KE), Jigjiga University (ET)
Remote-Sensing Image Classification
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon