CSDMS 2016 annual meeting poster AjayLimaye
Extraction of multi-thread channel networks using a reduced-complexity flow model
Abstract:
Channels with multiple, interwoven threads are common features of river valleys and alluvial and submarine fans. The geometry of multi-thread channel networks is a basic constraint for modeling stream flow and sediment transport, and is used in applications including fisheries management and flood and debris flow hazards. Understanding the adaptability of multi-thread channel geometry is also important for interpreting landscape response to ancient and modern climate change. Multi-thread channels have been hypothesized to adjust their planform and cross section geometry to accommodate increases in discharge. However, manually measuring channel geometry (e.g., from aerial photos) to test this hypothesis is often time-consuming and subjective. Existing automated approaches to multi-thread channel mapping identify the channel extent using inundation. I will present an alternative framework to automatically and objectively extract multi-thread channel geometry from topography, provided that the data partially or fully resolves the channel cross section. The approach uses a reduced-complexity flow algorithm, similar to those developed for braided river modeling, to reveal the spatial structure of multi-thread channel networks with locally divergent flow paths. Importantly, the flow model highlights abandoned channels that are common in arid climates and landscapes with shifting channel belts. The channel extraction approach is tested for case studies including an experimental submarine fan; a natural braided river near Flathead Lake, MT; and the large-scale anabranching canyon system of Kasei Valles, Mars. These examples range in spatial scale from 1 m to 100 km, and in digital elevation model resolution from 1 mm to 100 m. By repurposing a reduced-complexity flow model, the new channel extraction approach offers a unified framework for testing how multi-thread channels respond to changing discharge in numerical models, laboratory experiments, and natural landscapes.
* Please acknowledge the original contributors when you are using this material. If there are any copyright issues, please let us know and we will respond as soon as possible.
