Model:Erode

Erode

Model:	Erode
Contact person:	Scott Peckham
Institute:	CSDMS, INSTAAR, University of Colorado
City:	Boulder, CO
Country:	USA
Email:	Scott.Peckham@colorado.edu
2nd person involved:	--
3rd person involved:	--

Model type:	Modular model for the terrestrial domain.
Description:	Erode is a raster-based, fluvial landscape evolution model written in IDL.

Supported platforms:	UNIX, Linux, Mac OSX, Windows
Programming language:	IDL
Model development started at:	2003 and development still takes place.
To what degree will the model become available:	Source code will be available. Model also available as teaching tool, and as IDL SAV file (can run with free IDL VM)
Current license type:	Apache public license
Memory requirements:	Standard
Typical run time:	Hours to days

Input parameters:	Initial land surface (several built-in options), number of timesteps, DEM grid dimensions, DEM grid cell dimensions, R = "geomorphic" rainrate (m/yr), U=uplift rate (mm/yr), BLR = base-level lowering rate (mm/yr), Kf="erodibility coefficient (m^3/yr)^(1-m), m = area/discharge exponent, n = slope exponent, p = area-discharge exponent, toggles for different types of boundary conditions (e.g. periodic), DEM georeferencing info (bounding box, pixel geometry, etc.)
Input format:	Command line
Output parameters:	A sequence of grids that represent DEMs at different times in the evolution. Saved in RTS (RiverTools Sequence) format with RTI file for georeferencing.
Output format:	Binary
Post-processing software (if needed):	Yes, RiverTools or a similar program can be used to create animations of the grid sequence.
Visualization software (if needed):	Yes, RiverTools

Processes represented by model:	Sediment transport (parameterized with slope and contributing area grids), rainfall, uplift, base-level lowering.
Key physical parameters & equations:	The main equations are: Q = R * A^p Qs = Kf * (Q^m) * (S^n), 2D mass conservation equations for water and sediment
Length scale & resolution constraints:	Typical grid cell dimensions are 10 to 500 meters.
Time scale & resolution constraints:	Typical simulated time is 1000 to 100,000 years.
Numerical limitations and issues :	D8 flow codes are used to compute contributing areas. Would be better to use D-Infinity or the Mass-Flux method.

Available calibration data sets:	None
Available test data sets:	None
Ideal data for testing:	Same as used for other LEMs, like CHILD and MARSSIM.

Currently or plans for collaborating with:	See comments below.

Comments:	I currently have an NSF-CMG grant to work with Greg Tucker, Tom Manteuffel and Steve McCormick to find faster algorithms for this type of model.