Implementation, Validation and Study Case of a Landscape Evolution Model Algorithm

Abstract

This paper presents a landscape evolution model based on physical processes – hillslope processes and fluvial erosion, transport, and deposition – solved by numerical methodology. That is, through the solution of differential equations approximated by numerical methods. In this case, hillslope processes are modeled through the classical diffusion equation, discretized by the finite volume method. Fluvial erosion, transport, and deposition are modeled by the fluvial potential equations (stream power law). For this, the approximation is performed by the finite difference method. The topography – initial condition – is set by digital elevation models, obtained from satellite images. These are Raster datasets, that each cell contains a representative elevation value. The drainage is determined through the classical algorithm D8, which performs a scan on the digital elevation model, tracing routes of greater slopes between the cells. The algorithm execution flowchart is presented, and the model is validated. Finally, a geomorphological study is presented in the Piratini river basin, showing thar developed model mimics largescale natural phenomena of watershed processes.