Box model code, processing routines, and model ensemble data for the paper "Microbial feedbacks optimize ocean iron availability" by Jonathan Maitland Lauderdale, Rogier Braakman, Gaël Forget, Stephanie Dutkiewicz, and Michael J. Follows in Proceedings of the National Academy of Sciences.

Model parameters are set in comdeck.h, while the main model routines are in boxmodel.f. Compile the model with f2py to generate a python module:

>>f2py -c -m nutboxmod --verbose boxmodel.f fe_equil.f transport.f insol.f

To replot the figures from the paper, then run boxmodel.py or the Jupyter Notebook boxmodel.ipynb, which will read model ensemble data (10,000 members) from boxmodel_input.csv and boxmodel_output.csv. Note:

1. Changing RUNMODEL to True will re-run the ensemble and possibly overwrite the input and output files - this has been known to take on the order of 2 days because it is not parallelized.
2. The model is compared to data from external servers. If you do not have files for World Ocean Atlas 2013 annual Nitrate or Phosphate climatologies or the GEOTRACES IDP 2017 v2, then the default values from the paper will be used.

Any questions or comments, please get in contact!