1 Spatial Modelling of the Terrestrial Environment - Georeferencial

168 Spatial Modelling of the Terrestrial Environment
very little sediment reaches Lake Victoria, though this lake is known to suffer sedimentation
problems. This seems to illustrate a deficiency in the sediment delivery model when it is
applied to large catchments that contain a considerable amount of flood plain.
Problems with the sediment routing can, to a certain extent, be evaluated by remote
sensing of the lake sediment plumes as it is possible to determine if sediment plumes are
associated with rivers that the sediment delivery ratio method predicts produce no sediment.
AVHRR and ATSR-2 imagery has clearly identified large near-surface sediment plumes
emanating from a number of rivers in the catchement (Figures 8.4–8.6), and in the case of
AVHRR even other rift valley lakes (Figure 8.4). A number of plumes are visible on the
eastern side of the lake (Figures 8.4–8.6). The most extensive being the Malagarasi River
plume, which is observed in all images, and consistently covers hundreds of kilometres.
Patterns displayed by the Malagarasi River plumes closely match the wind-driven lake
currents modelled by Huttula et al. (1997). The sediment yield modelling suggests that the
Malagarasi River provides no sediment inputs to the lake, so there is a big discrepancy
here. The study by Tiercelin and Mondegeur (1991) initially appears to supports that of
the sediment routing. They suggest that Malagarasi River deposits much of its sediment
in upstream swamplands and also traps sediment in its delta system. Thus, the fact that
AVHRR and ATSR-2 imagery clearly identifies many large plumes is somewhat puzzling.
However, Tiercelin and Mondegeur’s (1991) study is based on sediment core analysis of lake
sediments that indicates that the central Lake Tanganyika basin has a lower sedimentation
rate than that of the north where the Ruzizi River flows into the lake. It is possible that there
is not much depth of sediment in the middle basin of the lake because the Malagarasi River
sediment plume is transported towards the north under prevailing near-surface currents,
and dispersed over a wide area, therefore creating less observable deposition in the middle
basin. Thus, it is possible that the large buoyant plumes of the Malagarasi River could cause
an underestimation in the significance of this river as a sediment contributor to the lake
because plume dispersal causes its sediments to be deposited over a wide area.
Only a few plumes can be seen in the northern basin where the highest sediment yields
were predicted. Plumes are detected at the Ruzizi River mouth on only one occasion over
the monitoring period (4 March 1998), whereas plumes can be detected from the Mugere or
Ntahangwa on three dates (22 March 1998, 4 March 1998, 24 April 1998). It is interesting
to note that the Ruzizi River does not have many large visible plumes, while modelling
results estimated that it is responsible for a large amount of sediment input into the lake.
Field measurements suggest that it transports a great deal of sediment (Sebahene et al.,
1999) and that the plume sinks rapidly presumably because its water is more dense than the
lake. Because of this, remote sensing is unlikely to be effective at monitoring the dispersal
of what appears to be the most important source of sediments into the lake.
8.4 Conclusion
The results show that regional scale erosion modelling can provide a general picture of the
source areas of erosion in the Lake Tanganyika catchment, and estimates of the quantity
of sediment transported into the lake. Previous spatial studies of lake sediment inputs
have relied on the interpretation of satellite imagery (and topographic maps) to assess
the amount of forest in a catchment, in order to determine those catchments most likely