Sort through all forecasts to find those that should be generated by the Arbiter from NWP models. The forecast must have a model key in extra_parameters (formatted as a JSON string). If piggyback_on is also defined in extra_parameters, it should be the forecast_id of another forecast that has the same parameters, including site, except the variable.
- forecasts (list of datamodel.Forecasts) – The forecasts that should be filtered to find references.
- run_time (pandas.Timestamp or None, default None) – The run_time of that forecast generation is taking place. If not None, the next issue time for each forecast is added to the output.
pandas.DataFrame – NWP reference forecasts with index of forecast_id and columns (forecast, piggyback_on, model, next_issue_time).