Diagnostic assessment on urban floods using satellite data and hydrologic models in Kigali, Rwanda

M. Manyifika, T.H.M. Rientjes

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

The Nyabugogo commercial center of Kigali city is geographically located between the Mont Kigali and Jali forming a bottleneck. Frequent flooding, of which no records are available, have characterized the area for the past few years. Additionally, the area hydrological behavior knowledge was lacking. This study provides an assessment of the frequent flooding in the area using hydrologic models and satellite data. CMORPH 8Km 30min, time variably bias corrected, is used to assess the rainfall events likely to cause flooding with their patterns. The HEC-HMS Curve Number (CN), Muskingum routing and base flow recession sub models are used to estimate the upstream area runoff later serving as upstream boundary conditions to the 1D2D flood model of the area set up with the national 10X10 m DTM and measured river cross sections developed in SOBEK 1D2D. Local regionalization of optimum parameters from gauged sub catchments rainfall-runoff models, using area and main channel length conversion with their proximity factors, is applied to model the runoff from ungauged systems. The Peak weighted root mean square error (PWRMSE), relative volumetric error (RVE) and Nash-Sutcliff (NS) are used as objective functions for the rainfall-runoff model calibration. A PWRMSE of 3.4, RVE of -4.9 and NS of 0.6 were obtained after calibration. 4 extreme rainfall events were detected with a spatial-temporal pattern exhibiting a horizontal movement from east to west with higher amounts observed during March-April-May and October-November-December. Results indicated flash floods in the area with 7 sub catchments found to contribute much runoff in the Nyabugogo River during flooding. Also, the flood extent, depth and velocity were affected by the DEM resolution, building representation and significantly by the surface roughness. Strong backwater effects at the confluence points of the Nyabugogo River and its tributaries Yanze and Mpazi act as triggers to flooding during heavy rainfall.
Original languageEnglish
Title of host publicationProceedings of GeoTech Rwanda 2015, 18-20 November 2015, Kigali, Rwanda
Place of PublicationKigali
Pages1-8
Publication statusPublished - 18 Nov 2015
EventGeoTech Rwanda 2015 - Kigali, Rwanda
Duration: 18 Nov 201520 Nov 2015

Conference

ConferenceGeoTech Rwanda 2015
CountryRwanda
CityKigali
Period18/11/1520/11/15

Keywords

  • METIS-314201

Fingerprint

Dive into the research topics of 'Diagnostic assessment on urban floods using satellite data and hydrologic models in Kigali, Rwanda'. Together they form a unique fingerprint.

Cite this