There are two branches under this repository. Master branch (default) holds a set of shell / R scripts for processing GIS data and assemble RHESSys model inputs. Tools are designed to run locally on Mac / Linux. The usage of the scripts are organized by modules documented in workflow guide (links are below). The second branch is "JupyterNotebook" developed and maintained by Lukey (Choi). Lukey's goal is to create a user-friendly workflow interface on JupyterNotebook that utilizes the same scripts held on master branch. JupyterNotebooks have been developed for UVA Rivanna and HydroShare Jupyter App.
- GRASS GIS 7.x.x installed and its associated dependencies and libraries (please see offical GRASS GIS website)
- R 3.6.x and above installed
- Bash scripting environment installed or supported by the OS
- Run this script to test if the required software has been properly installed on Mac OS. (https://raw.githubusercontent.com/laurencelin/GIS2RHESSys/master/workflows_test_script.sh)
- Any experiences in GIS, including finding data sources, pre- and post- processing data
- Experiences in command lines, being comfortable to use commands to operate computer
- Backgrounds in Hydrology and Ecology; After all, RHESSys is an eco-hydrological model used for hydrology and ecosystem studies.
GIS2RHESSys has two parts: GIS setup part (I) and building RHESSys part (II).
The GIS setup part contains GIS (command line) modules to initiate GIS dataset with a desired projection and resolution, import and process raw GIS data, calculated variables and information used to for RHESSys modeling. User has options to perform these tasks on their own or GUI interface as long as user is responsible to understand what variables and information the model requried. A figure below (in data requirement section) summarizes the required GIS information for building RHESSys.
The building RHESSys part is to fill out a template used by the R script to read in maps from an established GRASS GIS database and write out into RHESSys input files, including def files, flow tables, worldfile, and worldfiles header. User is responsible for making the climate data available to the model. Note that vegetation carbon pools in the output worldfile have been initiated, i.e., no need of using LAIread. However, growth simulation is still required spin-up.
- https://nbviewer.jupyter.org/github/laurencelin/GIS2RHESSys/blob/master/GIS2RHESSys_workflow_script_master.ipynb
- https://colab.research.google.com/github/laurencelin/GIS2RHESSys/blob/master/GIS2RHESSys_workflow_script_master.ipynb
- https://nbviewer.jupyter.org/github/laurencelin/GIS2RHESSys/blob/master/GIS2RHESSys_template_script_master.ipynb
- https://colab.research.google.com/github/laurencelin/GIS2RHESSys/blob/master/GIS2RHESSys_template_script_master.ipynb
When setup a RHESSys model for a catchment, we are expecting certain directory structure (Figure below).
- Project directory holds all raw data, GIS dataset, RHESSys model inputs, and observations.
- Raw data directory (named "raw_data" ) holds any data that are used for model setup, including GIS data, vegetation physiological dataset, and soil data.
- GRASS database (named "grass_dataset") holds all the GRASS GIS databases that are developed by the commandline workflow; One GRASS database for one RHESSys model;
- RHESSys model directory (could be named as anything) holds all input files for RHESSys; Sources / binary codes of RHESSys are downloaded and compiled seperatedly.
- Observation directory (named "obs" ) usually holds streamflow time series or other data for later model calibrations.
- Elevation
- Landuse & landcover
- Soil texture
- Climate time series (daily)
Below diagram shows the relationship between GRASS dataset and RHESSys model.
- RHESSysEastCoast https://github.com/laurencelin/RHESSysEastCoast
- RHESSys model calibration https://github.com/laurencelin/R-coded-scripts-for-RHESSys-calibration
- General RHESSys github: https://github.com/RHESSys/RHESSys
- General RHESSys wiki: https://github.com/RHESSys/RHESSys/wiki General RHESSys tools by Will from U.S. west coast https://github.com/RHESSys/RHESSys/tree/master/RHESSysPreprocessing Will's RHESSys tools are written R-studio and designed for general RHESSys model.