Automatic Extraction of Delta Channel Network Topology
Abstract
Populated by nearly half a billion people, deltas are receiving heightened attention due to increasing climatic and anthropogenic stresses. In order to better understand delta change and vulnerability to stressors, we developed a framework that investigates deltas' form and function through the lens of their channel network structure and connectivity. This new framework maps delta channel networks to corresponding graphs then relies on well-developed graph theory to design and compute metrics that differentiate deltas formed under different controlling processes and environments and assess overall and local delta vulnerability. Manually generating graphical representations of deltas is both time-consuming and subjective, so we developed RivGraph, a Python-based tool to automatically extract a delta's topology from its binary image. RivGraph resolves a delta channel network's links and nodes, sets flow directionality for each link, and corrects the resulting topology to ensure no sources, sinks, or cycles are present. An adjacency matrix representation of the delta channel network is created which can be used to assess topologic and dynamic metrics of complexity and resilience. RivGraph thus enables rapid assessment of delta function and structure, paving the way for improved understanding of delta channel networks at a global scale.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMEP31D2383S
- Keywords:
-
- 1615 Biogeochemical cycles;
- processes;
- and modeling;
- GLOBAL CHANGEDE: 1856 River channels;
- HYDROLOGYDE: 1861 Sedimentation;
- HYDROLOGYDE: 4558 Sediment transport;
- OCEANOGRAPHY: PHYSICAL