Mean Climate

Overview

The mean climate summary statistics are the most routine analysis available from the PMP. Because they are quasi-operationally applied to large numbers of simulations and under different conditions, the current mode of operation is fairly general. Before it can be applied some preparation is needed including:

  • Setting-up observational climatologies

  • Construction of model climatologies

  • Construction of an input parameter file to run the desired operations

Each of these steps is included in the following demo notebooks along with a series of examples that demonstrate the options. These steps are also summarized below.

Demo

Observational climatologies

A subset of the observational climatologies used for the PMP’s mean climate metrics is available via a jupyter notebook demo. Once you have run this demo or downloaded this demo data you can interactively run the mean climate and other demos. The complete database of observational climatologies is available to users of the PMP. To obtain this, please contact the PMP user group (pcmdi-metrics@llnl.gov) and you will be promptly provided with the database.

The PMP’s mean climate summary statistics can be applied to many fields and in most cases there is more than one reference data set available. To accommodate this, the observational climatologies used by the PMP are managed via a simple catalogue in the form of a JSON file. For many of the variables there are ‘default’ and ‘alternate1’ datasets and for some there is also an ‘alternate2’. To simplify the use of the different options in the mean climate, the mean_climate_driver.py (see below) expects to be pointed to observational catalogue. Currently, if a user wants to add additional observational data this can be done by including it in the JSON catalogue. However, this must be done carefully to ensure the file retains JSON compliant structure.

A recent observational climatology catalogue is included as part of the PMP as a default, so it does not need to be explicitly identified when using the mean_climate_driver.py (unless the catalogue has been modified to include new observations). However, as described below, the user must provide the base path to the observational database. As indicated in the catalogue, the actual database does incorporate further directory structure and defined filenames which should not be modified. If changes are made to the catalogue, this can be done with input parameter settings (below) using the “custom_observations” option.

Preparation of model climatologies

Sample model climatologies are available as part of the PMP demo database noted above and are used for the mean climate notebook. However, if a user wants to create and use their own model climatologies with the PMP a simple example is provide in a stand alone climatology notebook, via the mean climate metrics notebook, or the PMP github repository.

Construction of an input parameter file

The PMP mean climate metrics can be controlled via an input parameter file, the command line, or both. With the command line only it is executed via:

mean_climate_driver.py  -p basic_param.py

or as a combination of an input parameter file and the command line, e.g.:

mean_climate_driver.py  -p basic_param.py --vars rlut pr

where the list of variables (vars) to run the analysis on includes ‘rlut’ (outgoing TOA longwave radiation) and ‘pr’ (precipitation). The following parameters are required to be set by the user either in a parameter file or on the command line:

  • vars: a python list of variables to apply the summary statistics, e.g., [‘pr’, ‘rlut’, ‘tas’]

  • test_data_set: a python list of runs or models, e.g., [‘ACCESS-1-0’, ‘CESM1’]

  • filename_template: template that is applicable for the runs in test_data_set, e.g., “CMIP5.historical.%(model_version).r1i1p1.mon.%(variable).198101-200512.AC.v20190225.nc” where “model_version” and “variable” will be analyzed for each of the entries in test_data_set and vars.

  • test_data_path: the path/template where the test_data resides, e.g.:

  • reference_data_set: a python list that specifies ‘default’, ‘alternate1’, ‘alternate2’ or ‘all’, e.g., [‘default’]

  • reference_data_path: the root path to the observational climatology database, e.g., ‘~/demo_data/PCMDIobs2_clims/’

  • target_grid: a string giving the desired horizontal resolution in degrees following the form ‘LATxLON’, e.g. ‘2.5x2.5’

  • regrid_tool: options include ‘esmf’ and ‘regrid2’

  • metric_output_path: the full path to the metrics output in JSON files, e.g., ‘~/demo_data/PMP_metrics/’

In addition to the above required input parameters, if the default catalogue of observational climatologies is not being used its replacement needs to be specified, e.g.:

custom_observations = './pcmdiobs2_clims_byVar_catalogue_v20200615.json'

The output of the mean climate summary statistics are saved in a JSON file. An example result demonstrates that multiple statistics are computed for different conditions including regions and seasons. The resulting JSON files include the data, software and hardware information on the summary statistics.

In addition to the minimum set of parameters noted above, the following additional options can be controlled for the mean climate:

  • cmec Flag to save summary statistics into a CMEC compliant JSON file or not.

  • region_specs Define a different set of domains to compute the statistics (python dictionary).

  • regions Specify which domains are applied to which variables (python dictionary).

  • sftlf_filename_template Provided a land-sea mask to be used in defining regions.

  • generate_sftlf Estimate a land-sea mask.

  • save_test_clims Select to save (or not) interpolated climatologies including masking

  • case_id Save JSON and netCDF files into a subdirectory so that results from multiple tests can be readily organized