Execute a dry-run simulation

Follow the steps below to perform a GEOS-Chem Classic dry-run simulation:


Also be sure to watch our video tutorial Using the updated dry-run capability in GEOS-Chem 13.2.1 and later versions at our GEOS-Chem Youtube Channel, which will guide you through these steps.

Complete preliminary setup

Make sure that you have done the following steps;

  1. Downloaded GEOS-Chem Classic source code

  2. Compiled the source code

  3. Configured your simulation

Then doublecheck these settings in the following configuration files:

  1. start_date: Set the start date and time for your simulation.

  2. end_date: Setthe end date and time for your simulation.

  3. met_field: Check if the meteorology setting (GEOS-FP, MERRA2, GCAP2) is correct for your simulation.

  4. root_data_dir: Make sure that the path to ExtData is correct.

  1. Set the frequency and duration for the HISTORY diagnostic collections to be consistent with the settings in geoschem_config.yml.

  1. Check the Settings section to make sure that diagnostic frequency DiagnFreq: is set to the interval that you wish (e.g. Monthly, Daily, YYYYMMDD hhmmss, etc).

  2. Check the Extension Settings section, to make sure all of the required emissions inventories and data sets for your simulation have been switched on.


You can reduce the amount of data that needs to be downloaded for your simulation by turning off inventories that you don’t need.

Run the executable with the --dryrun flag

Run the GEOS-Chem Classic executable file at the command line with the --dryrun command-line argument as shown below:

$ ./gcclassic --dryrun | tee log.dryrun

The tee command will send the output of the dryrun to the screen as well as to a file named log.dryrun.

The log.dryrun file will look somewhat like a regular GEOS-Chem log file but will also contain a list of data files and whether each file was found on disk or not. This information will be used by the download_data.py script in the next step.

You may use whatever name you like for the dry-run output log file (but we prefer log.dryrun). You will need this file to download data (see the next chapter).