Forecasts — client.forecasts

SJVAPCD daily air quality forecasts, one record per San Joaquin Valley county zone. Each record embeds its Region (with boundary geometry) so map layers don’t need a second request per zone.

from sjvair import SJVAirClient

with SJVAirClient() as client:
    upcoming = list(client.forecasts.list())
    print(upcoming[0])

{
    "id": "abc123",
    "region": {"id": "r1", "name": "Fresno", "type": "county", "boundary": {...}},
    "zone_name": "Fresno",
    "forecast_date": "2026-07-13",
    "issued_date": "2026-07-12",
    "published_at": "2026-07-12T14:31:09-07:00",
    "aqi_value": 101,
    "aqi_category": "Unhealthy for Sensitive Groups",
    "pollutant": "O3",
    "burn_status": "Discouraged",
    "burn_status_text": "Discouraged: Burning Discouraged",
    "air_alert": False,
    "air_alert_start": None,
    "air_alert_end": None,
}

Methods

Method	Signature
list(**params)	Iterate forecasts across zones. Defaults to current + future (forecast_date >= today, server-side) if no forecast_date filter is given.
get(forecast_id)	Get a single forecast record by ID.

list() accepts region_id, forecast_date/forecast_date__lt/__lte/__gt/__gte, and issued_date/issued_date__lt/__lte/__gt/__gte.

# Tomorrow's forecast for one zone
region = client.regions.search('Fresno', type='county')[0]
tomorrow = list(client.forecasts.list(region_id=region['id'], forecast_date='2026-07-13'))

# Every forecast issued on a specific day (today + tomorrow rows for every zone)
issued = list(client.forecasts.list(issued_date='2026-07-12'))

Two rows are written per zone on each daily ingestion run — one for forecast_date == issued_date (“today”) and one for forecast_date == issued_date + 1 (“tomorrow”) — so full forecast history accumulates over time rather than being overwritten.