compositer

Functions to prepare input for pywapor.et_look, more specifically to group various parameters in time to create composites.

add_times(ds, bins, composite_type)

Add times to the time coordinates, so that every bin has at least one datapoint.

Parameters:

• ds (xr.Dataset) – Datasat for which to check empty bins.

• bins (list) – List of np.datetime64’s which are the boundaries of the groups into which the variables will grouped.

• composite_type ({"min" | "max" | "mean"}) – Type of composites that will be created based on the data inside each bin.

Returns:

Dataset to which time coordinates have been added to assure no empty bins exist.

Return type:

xr.Dataset

time_bins(timelim, bin_length)

Based on the time limits and the bin length, create the bin boundaries.

Parameters:

• timelim (list) – Period for which to prepare data.

• bin_length (int | "DEKAD") – Length of the bins in days or “DEKAD” for dekadal bins.

Returns:

List of np.datetime64’s which are the boundaries of the groups into which the variables will grouped.

Return type:

list

main(dss, sources, folder, general_enhancers, bins)

Create composites for variables contained in the ‘xr.Dataset’s in ‘dss’.

Parameters:

• dss (dict) – Keys are tuples of (‘source’, ‘product_name’), values are xr.Dataset’s which will be aligned along the time dimensions.

• sources (dict) – Configuration for each variable and source.

• folder (str) – Path to folder in which to store (intermediate) data.

• general_enhancers (list) – Functions to apply to the xr.Dataset before creating the final output, by default “default”.

• bins (list) – List of ‘np.datetime64’s which are the boundaries of the groups into which the variables will grouped.

Returns:

Dataset with variables grouped into composites.

Return type:

xr.Dataset