| dc.description.abstract | The study is based on the upper Athi River sub-catchment, which comprises an
area of approximately 5 590 knl and is bounded by the parallels of latitude 0 050 I
S and 1 050 I S, and the meridians 36 030 ' E and 37 0 15 ' E. There are two high
points making the sub-catchment. That is, the southern slope of the Aberdare
ranges together with the flanks of the rift valley to the south and Ngong hills,
which is to the west. Sediments, which are by-products of runoff, reduce soil fertility and the economic life of reservoirs. In the Athi river catchment, sediment transport estimates from 1965 and earlier indicate that this catchment lost 55 000 tonnes of sediment annually. Tana and Athi River Development Authority (TARDA) carried some work on the river in 1983 and noted that this figure is increasing annually. With this increase, the following sediment problems are experienced in the catchment; reduction in reservoir life due to sedimentation, turbine operation being adversely affected, water pumps are adversely affected, intake structures and irrigation canals have _silted up and river channel sediment deposition clogs the rivers causing floods during rainy seasons and shortage of water during dry seasons. This results in fish production being drastically reduced while the deposition of fine sediment at the river mouth results in mud floods. These mud floods are colonized by mangrove vegetation. Malindi Beach has been extended in the last 20 years as a result of this deposition. For much of the year, the water of the beach is too loaded with fme silt, making it less attractive for beach tourism.
The above problems call for sediment studies in the catchment. The present study
is based on suspended sediment load using rainfall, runoff and suspended sediment
data -obtained during the period 1980 to 1981. All the data used were first subjected to statistical tests in order to verify their homogeneity and validity. Forecasting the suspended sediment yield during the period of study--was done using the Instantaneous Unit Sediment Graph (IUSG) model, which is a conceptual multi-reservoir cascading system that was developed by Professor Nash in 1957. The IUSG model was changed to T hr USG using the convolution integral method. A sediment rating equation was also developed using the ordinary least squares method. A rating equation was obtained by considering the lognormal distribution of the error component, which represents the state-of-the-art in rating equations. For the period of the study, the total estimated sediment yield from the subcatchment was obtained as 1.05 million tonnes compared to 660 000 tonnes by Wain in 1983. This translates to one eighth of the sediment yield from the entire
catchment monitored at Sabaki River by Mansell-Muollin in 1973. The mean rate
of sediment production was obtained as 187.95 tlkm2/yr, which is high compared
to 118 tlkm2 /yrobtained by Wain in 1983. Wain used the flow duration-sediment .
rating curve, which is known to be inaccurate when compared to the IUSG model.
Due to lack of resources, the catchment was not gauged for present runoff and
sediment data. Therefore, the forecasted sediment yield does not reflect the present
status of sediment 'yield in the catchment. It is, therefore, recommended that the
catchment be gauged for up to date runoff and sediment discharge data, as the catchment has .undergone various land use changes that have increased the sediment yield, thus compounding the sediment problems discussed above. | en |
