Conflict-driven rapid forest degradation in an urban riparian ecosystem: High-resolution satellite evidence from Sunut Reserved Forest, Sudan

Ahmed, Abdalrahman és Osman, Abubakr és Czimber, Kornél (2026) Conflict-driven rapid forest degradation in an urban riparian ecosystem: High-resolution satellite evidence from Sunut Reserved Forest, Sudan. TREES FORESTS AND PEOPLE, 26. ISSN 2666-7193

Szöveg 1-s2.0-S2666719326002207-main.pdf Download (10MB)

Absztrakt (kivonat)

Armed conflict is increasingly recognized as a driver of rapid land degradation, particularly in regions where the breakdown of the governance accelerates unsustainable resource extraction. This study quantified conflict-associated forest degradation in the Sunut Reserved Forest, a riparian ecosystem in Khartoum, Sudan, following the escalation of armed conflict in 2023. Multi-temporal PlanetScope imagery from January 2023, 2024, 2025, and 2026 was analyzed to assess vegetation dynamics and land cover transitions. Vegetation dynamics were assessed using the Normalized Difference Vegetation Index (NDVI), and land use/land cover (LULC) classification was performed using an object-based Support Vector Machine (SVM) algorithm. A total of 260–272 stratified random samples per year were used, with 70% allocated for training and 30% for independent validation. The classification performance was evaluated using confusion matrix–based metrics, yielding overall accuracies ranging from 82.09% to 94.10% and Kappa coefficients between 0.64 and 0.88. Forest cover declined from 105.34 ha (58.23%) in 2023 to 20.19 ha (11.16%) in 2026, representing 80.83% reduction over three years. Annual forest loss accelerated from −12.66 ha·yr⁻¹ (2023–2024) to −36.25 ha·yr⁻¹ (2024–2026), and sharply increased to −66.66 ha·yr⁻¹ (≈ −76.75%·yr⁻¹) between 2025 and 2026, indicating a non-linear degradation trajectory. NDVI correlations weakened substantially over time (τ = 0.69 for 2023–2024; τ = 0.20 for 2024–2026; p < 0.001), reflecting the structural collapse of canopy integrity. Analysis of long-term climate data (2015–2025) showed no significant rainfall or temperature anomalies, indicating that climatic variability is unlikely to explain the observed vegetation decline in the area. The classification accuracy exceeded 88% (κ > 0.80), and the uncertainty estimates were considerably smaller than the magnitude of the detected changes. Spatial evidence confirms that direct deforestation activities, rather than climatic stress, are the primary drivers. The near-complete removal of riparian forest cover implies severe degradation of hydrological regulation, soil stability, and biodiversity in urban floodplain settings. These findings demonstrate how conflict can trigger abrupt land system transitions and underscore the need for conflict-resilient land governance, rapid restoration interventions, and satellite-based monitoring frameworks to prevent long-term degradation and loss of ecosystem services.

Tudományterület / tudományág

agrártudományok > erdészeti és vadgazdálkodási tudományok

Kar

Nem releváns

Intézmény

Soproni Egyetem

Mű tipusa:	Cikk
SWORD Depositor:	Teszt Sword
Felhasználó:	Csaba Horváth
A mű MTMT azonosítója:	MTMT:37390943
Dátum:	17 Júl 2026 08:35
Utolsó módosítás:	17 Júl 2026 08:35
URI:	http://publicatio.uni-sopron.hu/id/eprint/4110

Actions (login required)

Tétel nézet