Determination of flood risk zones in Accra and Tema Metropolis

Abstract

Accra has been experiencing periodic flooding that affects properties and lives. For instance on 4th and 5th June 1994, 8 persons lost their lives and 80 million cedis worth of properties belonging to Paloma Company were destroyed due the floods that occurred at that time. The government seeing the dangers involved, commissioned institutions such as Ministry of Works and Housing, Town and Country Planning and City Engineers to identify such areas and adopt measures that will help reduce the effect of this unexpected extreme event. The measures used by these institutions in order to identify and prevent the occurrence of such extreme event have been woefully inadequate and inefficient. Hence this thesis examines new methods of identification of potential flood risk zones stressing the possible combination of hydrological models with remote sensing and geographic information system models. The main objective of the study is to determine flood risk zones in Accra and its environs using a hydrological model and a geographical information system model. Specifically, the study aims at assessing how the urban watershed has been modified in-terms of drainage networking, soil texture, runoff and their influences on flooding. A flood risk zones map and the paleogeomorphology of the study area were also constructed. Data for the study were from secondary and primary sources. The secondary sources included rainfall, discharge, and topography data, which were collected through literature search from institutions such as Hydro Division of Ministry of Works and Housing, Survey Department, Meteorological Service and Water Research Institute. The primary data collected included drainage cross sectional measurement and soil characteristics. These data were collected using field measurements, field observations, satellite image interpretation and laboratory analyses. The analyses of the data were based on the integration of the hydrological model into the Geographic Information System model using an overlay operation. The result of the research showed that potential areas likely to experience periodic floods with a given input of rainfall (140.2 mmlhour) are mostly below the 350-meter contour and about 45 percent of the study areas fall within flood risk zone. It was also noted that the flood experienced by an area is mostly dependant on rainfall intensity no matter the catchment area. However other factors such as, landuse, storage and runoff coefficient were identified as contributory factors to flooding in the study area . In search for a method to determine flood risk zones, the use of a hydrological model within a geographic information system model is very effective if only the appropriate decision rule is defined. The studies will enable the supervisory agencies such as Ministry of Works and Housing, Town and Country Planning and City Engineers enforce landuse policies to prevent people from developing areas prone to floods.