Predicting Flood Inundation in Owerri Imo State Using Hydraulic Model

Flooding in Owerri, Imo State, is increasingly frequent and destructive, causing economic losses, displacement, and infrastructure damage. The city's low-lying topography, poor drainage, and rapid urbanization, which encroaches on natural floodplains, exacerbate the problem, revealing the inadequacy of current flood management efforts. Therefore, this study aimed at predicting flood inundation in Owerri Imo State using hydraulic model. The objectives of the study are to: examine the pattern and extent of landcover/landuse within the study area; determine the roughness coefficient of the various landcover/landuse; assess the surface runoff potential across the study area; and model the extent of flood vulnerability in the study area using HEC-RAS. This study employed a geospatial-hydraulic modeling approach to predict flood inundation across Owerri Municipal. Land cover/land use classification was derived from Sentinel imagery using supervised classification techniques in QGIS, achieving high accuracy with an overall classification accuracy of 96.6% and a Kappa coefficient of 0.949. Each land cover class was assigned a corresponding Manning’s roughness coefficient and rasterized for hydraulic simulation. Surface runoff potential was estimated using the SCS-Curve Number method, integrating land cover data and soil hydrologic groups (HYSOGs), while a high-resolution Digital Elevation Model (DEM) supported terrain analysis. Flood modeling was implemented using HEC-RAS 2D rain-on-grid simulations based on an extreme rainfall scenario of 3391.2 mm. Flood depth outputs were further classified into vulnerability zones. Exposure analyses were conducted by overlaying the vulnerability map with road networks and land use categories to assess infrastructure susceptibility. The study found that built-up areas dominated land cover, occupying 61.07% of the area, while vegetation and water bodies accounted for only 5.91% and 0.81%, respectively. Manning’s roughness mapping showed that low-resistance surfaces prevailed, contributing to rapid runoff. Using the SCS-Curve Number method, the average runoff depth from a 3391.2 mm rainfall event was estimated at 536.63 mm, with localized peaks reaching 3354.21 mm. HEC-RAS 2D simulation identified widespread inundation, with high vulnerability zones concentrated in low-lying urban centers. Exposure analysis showed that over 20 km of roads and more than 2700 hectares of land, particularly residential and bare land, fall within moderate to high flood risk areas. Given that only 5.91% of the study area is covered by vegetation, while over 61.07% is built-up, it is recommended that urban greening programs be implemented. Establishing vegetative buffers along waterways, open parks, and green belts in flood-prone zones will increase surface resistance (Manning’s n) and reduce the velocity and volume of surface runoff. These measures would also enhance infiltration and provide ecological co-benefits.