DO3SE web service
This repository contains a web interface for the DO3SE model to estimate ozone deposition on crops and other plant species. The web interface is a part of the Tropospheric Ozone Assessment Report (TOAR) data infrastructure (see https://toar-data.fz-juelich.de for more information).
DO3SE has been developed by SEI in York.
The full workflow is shown in the flowchart below.
Information
Please refer to...
- The mapping manual (Mills et al., 2017) for information on flux based ozone vegetation metrics, flux-response-relationships and tree growing seasons. Also details on the trapezoids for LAI, fphen and leaf_fphen are detailed there.
- A GUI for the DO3SE model which contains the same Fortran code and should produce the same results as this model
- Crop growing seasons from GGCMI
Getting started
To start this web service, the Fortran code has to be compiled and binded. Then the web service is started
- Requirements: Fortran compiler, venv for python 3. If this software is not available on your system, run
sudo apt-get install python3-venv
sudo apt-get install gfortran
- Run setup to create virtual environment, and to compile fortran code, for _model.pyf
source prepare.sh
- Start the web application
./run.sh
First productive version
Try this link:
https://toar-data-testing.fz-juelich.de/do3se/api/v1/
Adding new vegetation / crop types
- Add a new line to resources/presets.csv
- Specify the growing season source in webdo3se/parameters.py, 'Parameters.get_growing_season()'
Known issues
- There exist more than one overlapping ozone time series for some stations, e.g. DENW074. We should clarify which one to use.
- The vapor pressure deficit calculation in utils.py still does not entirely match up with reference calculations.
- Some code is from Lisa Emberson and her team, namely F/*, setup.py, webdo3se/model.py, webdo3se/modeldata.py. This has to be licensed properly and referenced better.
- If wanted by the ozone researchers, ozone time series with gaps could be interpolated. The option is now disabled in toar_access.py.
- Outputs are explained in webdo3se model.py, but the format is a bit ugly.
- Parameterizations are not validated, except the ones used in the upcoming publication by Emmerichs et al.