Skip to content

KUL-RSDA/RegionalAC_Py

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RegionalAC_Py

This repository provides a basis to run AquaCrop (version 7.0 and higher) in a spatially distributed way, using a Python wrapper. Examples of preprocessing and postprocessing scripts are available. Note that these scripts are only there to serve as examples to the users, and need to be adapted for their own regional simulations.

Regional AquaCrop scripts

The regional AquaCrop is based on de Roos et al. (2021) and originally developed for AquaCropV6.1. The code has been updated to run for AquaCropV7.0 and higher. To spatially run AquaCrop using the Python wrapper, two scripts are required:

  1. AC_exec.py This script executes and contains the parallelization (using the multiprocessing Python package). This scripts prepares the AquaCrop environment and calls the function run_ac_pro_yrs (from AC_PRM.py) to prepare the project file in order to run AquaCrop.
  2. AC_PRM.py This script prepares the project file (.PRM) to run AquaCrop. Note that only the soil files and the climate files are spatially distributed for now. The crop, irrigation, and management files can be used but will be applied uniformly for the whole domain. However, dynamic (in time and space) input can readily be implemented.

The AquaCrop executable is run for each pixel. Note that the code was tested for AquaCropV7.1.

The RegionalAC can be tested by adapting the paths in AC_exec.py to the testace (test_linux or test_windows). Reference output, generated on Linux with AquaCropV7.1 is also provided under OUTPUT_REF.

Preprocessing

In this folder, some example scripts are given to help the users preprocess their data to run the regional AquaCrop. Note that these scripts may not be readily usable and should be adapted to your setup (domain, parameters, options, datasets,...).

  1. AC_SOL_253.py - Creation of soil files for a given grid based on the HWSD (as in de Roos et al., 2021).
  2. prep_CLI.py - Creation of climate files (meteorological input).
  3. COORD.py - Example on how to set the Georeference for AquaCrop and other input data of another resolution (e.g. climate).

Postprocessing

vars_ACout_to_netCDF.py - Example of how to store the AquaCrop output (generated in different directories, one for each pixel) into a netcdf files with dimensions (time, lat, lon).

Citation

  • de Roos, S., De Lannoy, G.J.M., Raes, D. (2021). Performance analysis of regional AquaCrop (v6.1) biomass and surface soil moisture simulations using satellite and in situ observations. Geoscientific Model Development, 14(12), 7309-7328, 10.5194/gmd-14-7309-2021.
  • Busschaert, L., de Roos, S., Thiery, W., Raes, D., De Lannoy, G.J.M. (2022). Net irrigation requirement under different climate scenarios using AquaCrop over Europe. Hydrology and Earth System Sciences, 26, 3731–3752, 10.5194/hess-26-3731-2022.

Releases

No releases published

Packages

No packages published