Get Continuous Forecast
Retrieve a continuous forecast series for H3 cells.
Stitches multiple consolidated forecast runs together. For each day in
the date range, selects the forecast published at latest_hour with
lead time days_ahead, then extracts that day’s data.
Access: Requires advanced subscription level.
Authentication
API key for authentication. Can also be provided in the api_key query parameter.
Path parameters
Query parameters
The local hour (0–23) to use for selecting the model run each day.
Response
Identifier of the weather model used for this forecast (e.g., ‘optimized’, ‘iso’). Null for historical actuals or composite responses.
Timestamp (UTC) when this forecast was created/published. Represents the model run time. Null for continuous/stitched forecasts or historical actuals.
Measurement units for all data values in the response. Common values: ‘MW’ (megawatts), ‘°C’ (degrees Celsius), ‘USD/MWh’, ’$/MMBtu’.
IANA timezone identifier used for the time_local field. Typically matches the region’s local timezone.
Array of timestamps in Coordinated Universal Time (UTC). Each entry corresponds to one row in the values matrix. Always timezone-aware.
Array of timestamps in local time (timezone specified in timezone field). Each entry corresponds to one row in the values matrix. Parallel to time_utc.
Metadata describing each column in the values matrix. Each entry is a list of one or more level labels — single-level columns are returned as length-1 sublists (e.g. [‘pjm_total’]); naturally multi-level data uses one entry per level (e.g. [‘pjm_total’, ‘demand’]). Clients can rebuild a pandas MultiIndex via pd.MultiIndex.from_tuples(columns).
2D matrix of numeric forecast or actual values in column-major order. Each inner list represents a series (corresponding to columns), and each element within that list corresponds to a timestamp (corresponding to time_utc/time_local). Missing or unavailable data is represented as null.
