Landlab: Plug-and-play numerical modeling of Earth-surface dynamics

Abstract:
Numerical models are widely used in the environmental sciences. Among these are the sciences that deal with the Earth's surface, including geomorphology, hydrology, sedimentology, glaciology, volcanology, and landscape ecology, among others. Although the scientific questions addressed by these diverse disciplines vary widely, in many cases the development of computational models involves similar programming problems: construction of a grid, calculation of geophysical flows across a 2D topographic surface, conservation of mass, input of parameters and initial conditions, output of calculations, and other tasks. Landlab is a Python-language programming library that takes advantage of these commonalities to help modelers develop, refine, and explore models more efficiently. Landlab provides four general capabilities: easy creation and configuration of a model grid (regular or irregular) and associated data arrays, encapsulation of simulation code into reusable components that can be coupled, a framework for constructing continuous-time stochastic cellular automata, and utilities for handling input, output, and topographic data preprocessing. Landlab is an element of the Community Surface Dynamics Modeling System (CSDMS), and is available at https://landlab.github.io. 

(Poster presented at NSF SI2 PI meeting, February 2017, Arlington, VA)