Setting Up Your Dataset

climpred relies on a consistent naming system for xarray dimensions. This allows things to run more easily under-the-hood.

Prediction ensembles are expected at the minimum to contain dimensions init and lead. init is the initialization dimension, that relays the time steps at which the ensemble was initialized. init must be of type int, pd.DatetimeIndex, or xr.cftimeIndex. If init is of type int, it is assumed to be annual data. A user warning is issues when this assumption is made.

lead is the lead time of the forecasts from initialization. The units for the lead dimension must be specified in as an attribute. Valid options are years, seasons, months, weeks, pentads, days, hours, minutes, seconds. If lead is provided as pd.Timedelta up to weeks, lead is converted to int and a corresponding lead.attrs['units']. For larger lead as pd.Timedelta (months or years), there is no conversion possible.

Another crucial dimension is member, which holds the various ensemble members. Any additional dimensions will be passed through climpred without issue: these could be things like lat, lon, depth, etc.

If the expected dimensions are not found, but the matching standard_name in a coordinate attribute, the dimension is renamed to the corresponding climpred ensemble dimension.

Check out the demo to setup a climpred-ready prediction ensemble from your own data or via intake-esm from CMIP DCPP.

Verification products are expected to contain the time dimension at the minimum. For best use of climpred, their time dimension should cover the full length of init and be the same calendar type as the accompanying prediction ensemble, if possible. The time dimension must be of type int, pd.DatetimeIndex or xr.cftimeIndex. time dimension of type int is assumed to be annual data. A user warning is issued when this assumption is made. These products can also include additional dimensions, such as lat, lon, depth, etc.

See the below table for a summary of dimensions used in climpred, and data types that climpred supports for them.

List of climpred dimension and coordinates

Short Name	Types	Long name	CF convention	Attribute(s)
lead	int (or pd.Timedelta for units up to weeks)	lead timestep after initialization init	forecast_period	units (str) [years, seasons, months, weeks, pentads, days, hours, minutes, seconds]
init	int, pd.DatetimeIndex, xr.CFTimeIndex	initialization as start date of experiment	forecast_reference_time	None
member	int, str	ensemble member	realization	None