Geometry and topology of estuary and braided river channel networks extracted from topographic data
| Autor: | Matthew Hiatt, Willem Sonke, Elisabeth Addink, Wout van Dijk, Kreveld, Marc J., Tim Ophelders, Kevin Verbeek, Joyce Vlaming, Bettina Speckmann, Kleinhans, Maarten G. |
|---|---|
| Přispěvatelé: | Applied Geometric Algorithms |
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: | |
| Zdroj: | AGU Fall Meeting 2018 Pure TUe |
| Popis: | Channels are ubiquitous features of Earth's surface that are important pathways for the transport of water, solids, and solutes across landscapes, provide a range of ecosystem services, and support economic activity. Networks are mathematical representations of the connections among a set of objects and are useful representations of topology, geometry, and connectivity in channelized environments. However, objective and automatic extraction of channel networks from topography in multi-channel systems like braided river and estuaries has remained elusive. We present a mathematically-rigorous framework from extracting network topology and geometry from digital elevation models (DEMs) of braided rivers and estuaries. The concept of the “sand function” is introduced, which quantifies the volume of material separating channels and is a useful metric for identifying the relative scales of channels in the network. Four case studies are included: DEMs from the Western Scheldt estuary (Netherlands) and the Waimakariri River (New Zealand), as well as DEMs generated by numerical models of the morphodynamics in a braided river and an estuary. We show that larger scale channels (with higher sand function values) in the estuaries tend to be significantly deeper than smaller scale channels in the network. The results suggest that the main channel in an estuary is significantly deeper than the rest of the network, while the braided rivers tend to have channel depths that are evenly-distributed across channel scales. In all cases, the length of channels relative to system size scales with sand function scale to the power of 0.24-0.35, while the number of nodes against system scale does not exhibit a power-law relationship. The methods and results presented in this study provide a benchmark for evaluating both geometric and topologic characteristics of multi-threaded channel networks across scales. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|