Configuration

The only conguration you have to do is to build your request. find_agg works with a JSON header file enclosing your requirements. Please use the command-line help to find the embedded template (see Usage).

find_agg uses a combinatory of HTTP requests or XML paths to check if an aggregation exists. The aggregation URLs/paths are rebuilt following ALL requirements from your JSON file in a sense of AND conditions/logical operators with one element of each facets. Consequently, all fields in the JSON header file are required.

Build your request

Edit your requirements.json file defining your requested experiments, ensembles and variables according to the CMIP5 vocabulary and keeping the correct JSON syntax (see template below):

{

   "_help" : "Set below your required variables, experiments and ensembles following the CMIP5 vocabulary. Be sure to keep the correct JSON syntax (brackets, braces, comma) following this template. This comment line is ignored by Python script and can be deleted if desired",

   "variables":
      {
      "pr":  ["mon", "atmos", "Amon"],
      "tas": ["mon", "atmos", "Amon"]
      },
   "experiments": ["piControl", "1pctCO2", "rcp26", "rcp45", "rcp85"],
   "ensembles":   ["r1i1p1", "r2i1p1"]

}

_help is an omitted section by find_agg. You can use it as commentary describing your request.

variables delares the requiered variables with their CMIP5 name and corresponding tuple ["frequency", "realm", "CMOR table"] in that order. Requesting several frequencies or CMOR tables for the same variable requires as much lines. See following example:

"variables":
   {
   "tas": ["day", "atmos", "day"],
   "tas": ["mon", "atmos", "Amon"]
   },

experiments lists the required experiments coma-separated. ensembles lists the required ensembles coma-separated.

Warning

find_agg supports Unix wildcards only for ensembles/members. As for example:

"ensembles": ["r[12]i1p1"]

or

"ensembles": ["*"]

Table Of Contents

Previous topic

Installation

Next topic

Usage

This Page