CoRD Process

Here we describe what CoRD is doing when it runs. This should be useful for developers looking to modify the code or users wanting to ensure the system is designed properly for scientific use. This will also help when the time comes to use CoRD with a new watershed.

In the most broad sense, CoRD is a two-step model. The first step is to run DFLOW and the second step is to run RipCAS. Breaking this down further, the user first provides a vegetation map (details in the next paragraph) in ESRI .asc format, which is converted by CoRD to a roughness map, using a vegetation-to-roughness lookup table provided by the user. There are many files that DFLOW requires that we will cover later. CoRD submits the DFLOW job to the job queue, observes the job status, and once the job finishes, extracts the shear stress from the DFLOW output netCDF and converts the mesh to an ESRI .asc map of shear stress. Then using the vegetation map, the vegetation zone map, and the shear resistance lookup table, RipCAS calculates the change in vegetation at each point on the ESRI .asc grid given the shear stress output from DFLOW. The output vegetation map is then used as input to the subsequent DFLOW run and the process repeats. This is done for every year of peak flow given in the peak flows file, just a single-column file with flows in units of ??????

To begin, the user provides some parameters: file and directory locations and three numerical parameters. The directory that must be provided is the location where the simulation happens. All input files are copied here so that it is trivial to tell which inputs were used for a particular model run. The files are the initial vegetation map of the watershed, the vegetation zone map for the watershed, an Excel spreadsheet for looking up the shear resistance.