aerosol-modeling

Aerosol concentration modeling based on satellite data

Run source code:

1. Configure as you wish in configuration.yml
2. Run python main.py

Retrieved data

CAMS and aerosol data zipped inside raw/cams.zip and raw/n100.zip

Development tips

If using conda:

git clone https://github.com/hd4niel/aerosol-modeling.git
cd aerosol-modeling
conda env create -f environment.yml
conda activate inar
jupyter lab
conda deactivate inar

New dependencies

conda env export --no-builds | grep -v "^prefix: " > environment.yml

Remove environment

conda env remove --name inar

In case of problems with the environment:

conda create -n inar python=3.5.6
conda install -c anaconda pyyaml
conda install -c conda-forge cfgrib
conda install -c conda-forge ecmwf-api-client
conda install -c conda-forge jupyterlab
etc ...
conda activate inar

Data Retrieval

1. Log in to https://ecmwf.int

2. Save credentials at https://api.ecmwf.int/v1/key/ to $HOME/.ecmwfapirc

3. Specify load_data in configuration.yml

4. Run python main.py

CAMS Reanalysis

Reanalysis data documentation

https://confluence.ecmwf.int/display/CKB/CAMS%3A+Reanalysis+data+documentation

Data

https://apps.ecmwf.int/data-catalogues/cams-reanalysis/?stream=oper&levtype=ml&expver=eac4&type=an&class=mc

Efficiently retrieve daily data

https://confluence.ecmwf.int/display/WEBAPI/CAMS+Reanalysis+daily+retrieval+efficiency

Atmospheric model (we use 60, i.e. 10m above ground)

https://www.ecmwf.int/en/forecasts/documentation-and-support/60-model-levels

Reference CAMS: 'Generated using Copernicus Atmosphere Monitoring Service Information 2020'.

https://confluence.ecmwf.int/pages/viewpage.action?pageId=58131166

N100

Datasource:

In situ by INAR

Retrieve data on disks

Faster transfer speeds for future reference, released recently

https://ads.atmosphere.copernicus.eu/