Nyarko, Benjamin Kofi
Upper Sakumo. In other words elevation is not the only determinant of high runoff
concentration, but a critical attention should also be paid to the catchment area.
4.5 Flood Risk Areas
The runoff concentration map (Figure 1.5) does not show the areas that are liable to flooding,
though high elevated areas showed a comparatively higher discharge rates than the low lands.
Figure 1.6 Flood risk areas
1 : 1508512.
Using equation (3) for the map combination, it came out that areas within the study area
presented different pixel value ranging from low to high. Upon close examination of the map,
high-elevated areas of 350 m to 1550 m had low pixel value between 0-20 and the low elevated
areas of below 350 m presented high pixel value between 20-87. Based on ground truth, areas
with high elevation have low pixel values indicating that they fall within the low flood risk
zones. However areas with low elevation indicated high pixel values, this falls within locations
that experience periodic flooding at a given rainfall event. This confirmed that areas such as
Aladjo, Ashiaman and Sakumonu fall within the high pixel values experience destructive floods
with high rainfall intensities.
It is important to note that the arithmetic overlay method helped to identify areas even as small
as the area within the Densu flood plain (labeled Island) was isolated and assigned the flood risk
category that it falls in.
4.5.1 Flood risk zoning
Figure 1.6 show pixel values that can be used to represent the several flood risk areas. These
pixel values can be grouped to show a general pattern in the degree of flood intensity of regions
that exist within Accra and its environs.
Thus, with a clearly defined domain of: very high, high, moderate, very low and low risk zones,
the technique of density slicing made it possible to reclassify the flood risk map. This helped to
differentiate between areas that experience different intensity of flood (Figure 1.7).
(
1
\
A
ON de PN o pd
1046 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000.