Page last updated:
24 June, 2002
|
Abstract of Talk Presented at
7th International Conference on Fluvial Sedimentology
University of Nebraska, Lincoln, USA - August 2001 |
|
A Conceptual
Model of Channel Pattern Change in Confined Mixed Bedrock-Alluvial,
Semiarid River Systems in the Kruger National Park, South Africa
|
Back to Index
page  |
ROUNTREE, Mark W. and ROGERS, K.H.
Centre for Water in the Environment, Department
of Animal, Plant and Environmental Sciences, University of the Witwatersrand,
Johannesburg, South Africa |
Contact the authors:
markr@gecko.biol.wits.ac.za
|
|
Five main channel patterns (viz. alluvial single thread, mixed
(bedrock and alluvium) pool-rapid, mixed braided, mixed anastomosing
and bedrock anastomosing) were identified in a study of channel
pattern changes in a mixed bedrock/alluvial, semi-arid river system
in the Kruger National Park, South Africa. The study made use of
the 50 year aerial photographic record. A channel type classification
system developed for morphologically similar rivers (van Niekerk
et al., 1995) was used to classify the changing channel patterns
between large flooding events. The results showed that there was
frequent alternation between the mixed braided and pool-rapid channel
type states; that the bedrock anastomosing channel type gradually
became more alluviated and that the mixed anastomosing channel type
had a highly stable planform (Rountree et al., 2001). From these
data, a conceptual model (see figure) was developed which explains
the differing rates and pathways of the channel pattern changes
during the recovery period between large stripping events.

The magnitude of a flood required to strip (or 'reset') the different
channel types is the main driver of this model. Very large floods
would be required to exceed the thresholds of change in the wide
anastomosing channel types and they are therefore the most stable
channel patterns. Relatively small events can scour sediment from
the braided and pool-rapid channel types, hence the comparative
instability of these channel patterns. During the "recovery
phase" between large flooding events, sediment deposition and
vegetation establishment in and adjacent to the active channels
increases the stability of the channel pattern. As the recovery
phase progresses, the magnitude of flood required to strip the channel
segment increases with increasing sediment deposition and vegetation
development. A project is currently underway to validate this model.
|
|
|