nc2atmodat

This repository contains processing scripts to apply the AtMoDat standard to obstacle resolving model (ORM) datasets. Requirements are netCDF datasets as input files. Code is written in python. This code is mainly adapted to MITRAS model datasest (https://www.mi.uni-hamburg.de/en/arbeitsgruppen/memi/modelle.html). Adaptations towards datasets from other ORMs have to be done by the program user.

How to process datasets:

1. Download this code from GitHub with:

git clone [email protected]:AtMoDat/nc2atmodat.git

2. Go to directory process_model_data

cd process_model_data

You find here the following files:

• process_model_data.py (main programm)
• dict_OUTPUT_variables.py (pre-selected output variables)
• dict_custom_variables_attributes.py (additional variable information)
• Metadata_for_atmodat_standard.xlsx (metadata information)

3. open "Metadata_for_atmodat_standard" with Microsoft Excel and modify the sheet

4. open "process_model_data.py" with an text editor:

5. set 'path', 'file', 'Metadata_file' in the user input section.

6. set switches:

• if you use your own reference point
• or if you want to crop your data to an certain extend.

7. If you want to write specific variables add them to 'custom_variables' or add variables to "dict_OUTPUT_variables.py".

8. Run script with python

module load python3 ipython process_model_data.py

or

ipython process_model_data.ipynb

or

python process_model_data.py

to translate process_model_data.ipynb to an python file:

"jupyter nbconvert --to script process_model_data.ipynb"

Required Python Modules:

• xarray
• netCDF4
• numpy
• pyproj
• pandas
• dask

Description of files:

Directory process_model_data:

If you want to standardise your data go to process_model_data This directory contain the following files:

• process_model_data.py (Main script to run the code)
• dict_OUTPUT_variables.py (Pre-Selected output)
• dict_custom_variables.py (Additional variable information)
• functions.py (Functions to run the code)
• Metadata_for_atmodat_standard.xlsx (Metadata Information)

process_model_data.py

• main program
• add here path, input-file, output-file and metadata-file
• rewrite coordinates and dimensions
• add georeferencing to dataset by adding geographical and utm coordinates
• removes variables that are not meant for publication.

functions.py

• contains all functions used for process_model_data.py

dict_OUTPUT_variables.py

• contains dictionaries with all variables that should be published

dict_custom_variables_attributes.py

• contains dictionaries with variables and attributes.
• user specific descriptions for variables can be added here. e.g. tracer = pm10, pm2.5 or comments on the variables (references, explanations)

Metadata_for_atmodat_standard.xlsx

• contains global metadata that should be added to the file.
• must be filled by the producer for each dataset individually