Using models to guide field experiments: a priori predictions for the CO2 response of a nutrient- and water-limited native Eucalypt woodland.
Belinda E. Medlyn, Martin G. De Kauwe, Soenke Zaehle, Anthony Walker, Remko A. Duursma Kristina Luus, Mikhail Mishurov, Bernard Pak, Ben Smith, Ying-Ping Wang, Kristine Crous, John G. Drake, Teresa E. Gimeno, Catriona A. Macdonald, Richard J. Norby, Sally A. Power, Mark Tjoelker, David S. Ellsworth.
Global Change Biology, 2016, in press.
Repository containing all the model code (where possible) and associated scripts required to reproduce simulations our Global Change Biology paper.
Information provided to modellers to aid site parameterisation, as well as information on which variables to output, can be found in doc
The met_data folder contains all the raw data given to modellers to run the 4 experiments used carried out in the paper. Note this directory has some large files in it (>50 meg).
For some models it is not possible to share code due to ownership issues, e.g. O-CN. In these cases, models have set up their own repositories and the relevant model sub-directories contains README files which point to these stored repositories.
The top-level scripts directory contains the model checking scripts (check_model_output.py; check_model_output_AVG.py) to make sure post-processed outputs are sensible. There is also a script (generate_pickled_model_output.py) to generate a big (pandas) dataframe and then turn this into a binary object to allow easy access of all of the model output inside a script. Note the checking script require this binary object to be built first.
Model outpus are each of the model/outputs sub-directories, e.g. GDAY.
- Belinda Medlyn: B.Medlyn at westernsydney.edu.au
- Martin De Kauwe: mdekauwe at gmail.com