Compare topology in natural rivers, optimal channel networks, and random-walk trees : a graph theory approach
Abstract
Through a graph theoretic formulation, we investigate the eigenvalue spectrum of the adjacency matrix of natural rivers, optimal channel networks (OCNs), and random-walk trees. In order to investigate river network simulator, we consider drainage networks generated on two-dimensional lattices through an optimal channel network (OCN) model. In order to create random tree networks, we first generate a uniform grid of nodes and determine one of the nodes as the outlet. The random tree is generated by randomly selecting nodes and adding links such that there is a path from each node of the grid to the outlet. The real river networks are derived from NHDPlus. The T-Test which is an statistical examination of two populations through their t-statistic, t-distribution and degrees of freedom, is performed on the spectral signatures of large populations of river networks in order to find the probability of difference between three populations from three different networks.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMEP51E2126A
- Keywords:
-
- 1824 Geomorphology: general;
- HYDROLOGY;
- 1825 Geomorphology: fluvial;
- HYDROLOGY;
- 1830 Groundwater/surface water interaction;
- HYDROLOGY;
- 1856 River channels;
- HYDROLOGY