The ESMValTool has the following software requirements (note that specific diagnostics might require additional software packages):
- Unix(-like) operating system
Python version 2.7.x; some diagnostics written in Python (e.g.,the diagnostic "TropicalVariability") require installation of additional Python packages such as the Geometry Engine(GEOS), scientificpython, netCDF4, and cdo, e.g.:
- conda install basemap
- conda install --channel https://conda.anaconda.org/Clyde_Fare scientificpython
- conda install netcdf4
- conda install --channel https://conda.anaconda.org/auto cdo
It is strongly recommended to use the Python distribution Anaconda, as it allows the user to install additional Python libraries and extensions in a simple way and without modifying the installed Python distribution (i.e., without root permissions).
NCAR Command Language (NCL 2014) version 6.2 or higher
- Diagnostics written in R require a working installation of the statistical computing software R and that the executable Rscript is in the default search path. In addition, the netCDF libraries (ncdf / ncdf4) for R are needed. Currently, only the diagnostic “Standardized Precipitation index (SPI)” requires R. More diagnostics written in R might be added in the future.
- Common GNU utilities such as “wc”, “date”, “basename”, and “more”, which are usually part of the standard Linux distribution
Obtaining the source code
The ESMValTool can be downloaded here: Download
In addition, ESMValTool will be further developed in a version controlled repository that is accessible only to the development team. The wider climate community is encouraged to contribute to this effort and to join the ESMValTool development team for contribution of additional more in-depth diagnostics for ESM evaluation. Please contact Veronika Eyring or Axel Lauer if interested.
The tar-ball can be unpacked with the standard tar command, e.g.,
tar -xvf ESMValTool_v1.0.tar
A comprehensive user guide is available here: User Guide
The software design of the ESMValTool and the technical implementation is presented in detail in the Open Access Article:
Eyring et al., ESMValTool (v1.0) - A community diagnostic and performance metrics tool for routine benchmarking and process evaluation of Earth System Models in CMIP, Geosci. Model Dev., 2016.