Validation of height above nearest drainage inundation in South Africa: a case study of Durban and Mthatha flooding events

Flash floods occur frequently in South Africa, causing infrastructure damage, economic losses, and loss of life. Accurate delineation of flood inundation extents is essential for flood risk assessment, disaster management, and mitigation planning. This study aimed to evaluate the performance of the height above nearest drainage (HAND) method in delineating flood inundation extents for two flood events in Durban and Mthatha, South Africa. A 30-m resolution Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) was used to derive simulated flood inundation extents using HAND. The simulated extents were evaluated against observed flood data sourced from Sentinel-2 B and WorldView-3 using overall accuracy (OA), critical success index (CSI), false alarm ratio (FAR), and commission and omission errors. The results indicated flood detection probabilities (“hit”) of 87% for Durban and 90% for Mthatha. Simulated flood extent versus actual flooding extent obtained CSI values of 0.49 and 0.64, respectively. FAR values were 0.46 for Durban and 0.31 for Mthatha, while overall accuracy for extent was 49.9% and 64.3%, respectively. Omission errors of 12.5% and 9.5% and commission errors of 46.2% and 31.0% for Durban and Mthatha were also obtained. Overall, HAND demonstrated potential for preliminary flood hazard screening, with stronger performance across the reported detection metrics in Mthatha, although limitations in accurately delineating observed flood extents remained. These findings highlight the potential of HAND for large-scale preliminary flood hazard screening, particularly in data-scarce regions. However, its application is subject to uncertainties associated with terrain representation and temporal changes in landscape conditions. A static SRTM DEM may not capture changes in land use, drainage networks, and channel morphology between DEM acquisition and the 2022 Durban and 2025 Mthatha flood events, potentially affecting the accuracy of the derived inundation extents. Therefore, HAND outputs should complement locally surveyed flood lines for informing detailed flood risk zoning, infrastructure planning, and disaster management interventions. Using higher-resolution, more recent DEMs may further improve the representation of local terrain features and drainage pathways, potentially enhancing flood inundation delineation.

Authors

Institutions

Publication Details

Journal
Frontiers in Remote Sensing
Published
2026-11-16
DOI
https://doi.org/10.3389/frsen.2026.1993359
Primary Topic
Flood Risk Assessment and Management
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Validation of height above nearest drainage inundation in South Africa: a case study of Durban and Mthatha flooding events

Paidamwoyo Mhangara, Lerato Shikwambana, Siyamthanda Gxokwe, Morwapula Mashalane
2 links
Frontiers in Remote Sensing
Flood Risk Assessment and Management
article

Validation of height above nearest drainage inundation in South Africa: a case study of Durban and Mthatha flooding events

Paidamwoyo Mhangara, Lerato Shikwambana, Siyamthanda Gxokwe, Morwapula Mashalane
article en

Abstract

Flash floods occur frequently in South Africa, causing infrastructure damage, economic losses, and loss of life. Accurate delineation of flood inundation extents is essential for flood risk assessment, disaster management, and mitigation planning. This study aimed to evaluate the performance of the height above nearest drainage (HAND) method in delineating flood inundation extents for two flood events in Durban and Mthatha, South Africa. A 30-m resolution Shuttle Radar Topography Mission (SRTM) digital elevation model (DEM) was used to derive simulated flood inundation extents using HAND. The simulated extents were evaluated against observed flood data sourced from Sentinel-2 B and WorldView-3 using overall accuracy (OA), critical success index (CSI), false alarm ratio (FAR), and commission and omission errors. The results indicated flood detection probabilities (“hit”) of 87% for Durban and 90% for Mthatha. Simulated flood extent versus actual flooding extent obtained CSI values of 0.49 and 0.64, respectively. FAR values were 0.46 for Durban and 0.31 for Mthatha, while overall accuracy for extent was 49.9% and 64.3%, respectively. Omission errors of 12.5% and 9.5% and commission errors of 46.2% and 31.0% for Durban and Mthatha were also obtained. Overall, HAND demonstrated potential for preliminary flood hazard screening, with stronger performance across the reported detection metrics in Mthatha, although limitations in accurately delineating observed flood extents remained. These findings highlight the potential of HAND for large-scale preliminary flood hazard screening, particularly in data-scarce regions. However, its application is subject to uncertainties associated with terrain representation and temporal changes in landscape conditions. A static SRTM DEM may not capture changes in land use, drainage networks, and channel morphology between DEM acquisition and the 2022 Durban and 2025 Mthatha flood events, potentially affecting the accuracy of the derived inundation extents. Therefore, HAND outputs should complement locally surveyed flood lines for informing detailed flood risk zoning, infrastructure planning, and disaster management interventions. Using higher-resolution, more recent DEMs may further improve the representation of local terrain features and drainage pathways, potentially enhancing flood inundation delineation.

Frontiers in Remote SensingVol. 7
South African National Space Agency (ZA), University of the Witwatersrand (ZA), University of Johannesburg (ZA)
South African National Space Agency
Sustainable cities and communities, Climate action
Flood Risk Assessment and Management
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