How CLIX Works

How CLIX Works#

The conceptual foundation of CLIX relies on the xclim library and its extensive collection of climate indicator functions. These functions are the core computational engine for CLIX. You can find detailed documentation for xclim indices here.

Indicator Definitions#

CLIX uses xclim functions to compute climate indicators. Each indicator is defined with its respective default parameter values, which may sometimes differ from the default values used directly in xclim to align with DWD’s specific requirements.

The mapping between indicators and their corresponding xclim functions, along with their default parameters, is stored in a YAML configuration file (e.g., climate_indicator.yaml).

If a new indicator needs to be permanently integrated into CLIX, a new entry for it must be added to this climate_indicator.yaml file.

Example: Definition for Summer Days (txge25)

Below is an example of an indicator definition block for “summer days” (txge25) within the YAML configuration:

txge25:
    standard_name: txge25
    long_name: Number of summer days above or equal 25 degC
    units: days
    xclim_function: xclim.indices.tx_days_above
    default_parameters:
        threshold: 25 degC
        op: '>='

Each indicator definition block includes:

  • standard_name: A short, unique identifier for the indicator.

  • long_name: A descriptive, human-readable name for the indicator, often used for output file metadata.

  • units: The units of the calculated indicator, also for output file metadata.

  • xclim_function: The full path to the xclim.indices function used for the calculation.

  • default_parameters: A dictionary of parameters passed to the xclim function if not explicitly overridden by the user.

Custom Indicators#

While xclim provides a wide range of specific indicator definitions, there are cases where custom or more complex indicator definitions are required. For these situations, custom indicator functions can be included within the clix_custom_indicators.py file. These functions are then linked using the climate_indicators.yaml configuration file to make them available to the public. The exact function structure is explained on the Adding New Indicator page. It should be noted that xclim offers powerful generic functions that can be configured to tackle a broader set of problems.