AIRMET/SIGMET

AIRMETs and SIGMETs are official aviation weather advisories for hazardous en-route conditions. AIRMET (Airmen's Meteorological Information) covers moderate hazards; SIGMET (Significant Meteorological advisory) covers higher-severity events. This endpoint returns every active advisory whose affected area intersects your bounding box — use it for en-route hazard overlays on maps and pre-flight briefings. Filter with type=airmet or type=sigmet when you only need one severity class. Empty results (no active advisories) are normal and very common in good weather.

Request

Send a GET with a geographic bounding box and optional type filter.

Bounding box tips. Keep boxes route-sized — very large bboxes can stress upstream processing. An empty reports array means no active advisories intersect the area (common in fair weather).

Response

{
  "bbox": "38.0,-90.0,45.0,-80.0",
  "filter_type": "airmet",
  "reports": [
    {
      "raw": "WAUS46 KKCI 151445 WA6S AIRMET SIERRA ...",
      "bulletin_type": "AIRMET",
      "report_type": "AIRMET",
      "country": "US",
      "issuer": "KKCI",
      "area": "S",
      "region": "KZAU",
      "start_time": "2025-01-15T14:45:00Z",
      "end_time": "2025-01-15T20:45:00Z",
      "body": "AIRMET IFR...",
      "observation": {
        "type": "IFR",
        "intensity": null,
        "floor": "SFC",
        "ceiling": "FL180",
        "coords": [
          { "lat": 41.5, "lon": -87.5 },
          { "lat": 42.0, "lon": -86.0 }
        ],
        "movement": null
      },
      "forecast": null
    }
  ],
  "total": 1
}

When no advisories intersect the bbox, the API returns "reports": [] and "total": 0.

Report

JSON path: reports[]. Only raw is guaranteed non-null. All decoded fields are null when not parsed from the advisory.

Object schemas and nullability

Every key in a Report object is always present — nothing is omitted when a section is not decoded. The observation and forecast objects themselves may be null, but when present every key inside them is always included.

AIRMET families

SIGMETs cover higher-intensity en-route hazards. Convective SIGMETs are issued for severe thunderstorms, hail, and tornadoes over the contiguous US.

Client types

"""AIRMET/SIGMET response types - matches airsigmet field tables."""from dataclasses import dataclass, field@dataclassclass AirSigmetCoord:    lat: float    lon: float@dataclassclass AirSigmetObservation:    type: str | None = None    intensity: str | None = None    floor: str | None = None    ceiling: str | None = None    coords: list[AirSigmetCoord] = field(default_factory=list)    movement: dict | None = None@dataclassclass AirSigmetReport:    raw: str    bulletin_type: str | None = None    report_type: str | None = None    start_time: str | None = None    end_time: str | None = None    body: str | None = None    observation: AirSigmetObservation | None = None    forecast: AirSigmetObservation | None = None@dataclassclass AirSigmetResponse:    bbox: str    reports: list[AirSigmetReport]    total: int    filter_type: str | None = None

Integration

Fetch active advisories for a route bbox and build polygon overlays.

The example queries a corridor bbox, caches for 15 minutes (advisories update on a fixed schedule), and extracts advisory polygons with valid time windows for map rendering. Empty results are expected in VFR conditions — render "no active advisories," not an error.

import osimport timeimport requestsHEADERS = {    "x-api-key": os.getenv("SKYLINK_API_KEY", "YOUR_API_KEY")}BASE = "https://data.skylinkapi.com/v3"CACHE: dict[str, tuple[dict, float]] = {}AIRSIG_TTL = 900  # 15 mindef fetch_airsigmets(bbox: str, advisory_type: str | None = None) -> dict:    cache_key = f"airsig:{bbox}:{advisory_type or 'all'}"    cached = CACHE.get(cache_key)    if cached and time.time() < cached[1]:        return cached[0]    params: dict[str, str] = {"bbox": bbox}    if advisory_type:        params["type"] = advisory_type    r = requests.get(        f"{BASE}/weather/airsigmet",        headers=HEADERS,        params=params,        timeout=(10, 25),    )    r.raise_for_status()    data = r.json()    CACHE[cache_key] = (data, time.time() + AIRSIG_TTL)    return datadef active_polygons(bbox: str) -> list[list[tuple[float, float]]]:    """Return coord rings for map overlays (may be empty)."""    payload = fetch_airsigmets(bbox, advisory_type="airmet")    rings: list[list[tuple[float, float]]] = []    for report in payload.get("reports", []):        obs = report.get("observation") or {}        coords = obs.get("coords") or []        if coords:            rings.append([(c["lat"], c["lon"]) for c in coords])    return ringsif __name__ == "__main__":    import json    bbox = "-74.5,40.3,-73.5,41.0"    payload = fetch_airsigmets(bbox, "airmet")    print(json.dumps(payload, indent=2, default=str)[:3000])

Implementation notes

Empty is normal. Fair-weather routes often return total: 0. Your map layer can hide overlays when empty.

Validity windows. Use start_time and end_time to expire polygons client-side. Do not show advisories past end_time even if cached. If end_time is null, treat the advisory as active until the next poll.

Polygon rendering. Draw observation.coords (or forecast.coords) as closed polygons. When coords is [], show the body text in a list view instead of a map overlay. Always close the polygon ring (last vertex → first vertex) before passing to a mapping library.

Filter strategically. Pre-flight briefing UIs often fetch type=airmet for dispatch and type=sigmet for higher-severity overlays. Omit type when you need the full picture.

Large bboxes. Size the query to your route plus a modest buffer (approximately 1–2° latitude/longitude) and split long routes into segments if needed.

Error responses

Endpoint-specific failures only. For auth, validation, rate limits, and server errors, see Error Handling.

400 — invalid bbox format

{
  "detail": "Invalid bounding box format. Expected 'lat1,lon1,lat2,lon2'"
}

503 — upstream unavailable

Returned when the advisory feed is temporarily unreachable or the bbox is too large for upstream processing. Retry with exponential backoff and a smaller bbox.

{
  "detail": "Upstream AIRMET/SIGMET service unavailable"
}

curl -X GET "https://data.skylinkapi.com/v3/weather/airsigmet?bbox=string"

{
  "bbox": "38,-90,45,-80",
  "reports": [
    {
      "raw": "WAUS46 KKCI 151445 WA6S AIRMET SIERRA ...",
      "bulletin_type": "AIRMET",
      "report_type": "AIRMET",
      "area": "S",
      "body": "AIRMET IFR...",
      "observation": {
        "type": "IFR",
        "floor": "SFC",
        "ceiling": "FL180"
      }
    }
  ],
  "total": 1
}

{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string",
      "input": null,
      "ctx": {}
    }
  ]
}

Related: Weather hub · PIREPs · METAR · AIRMET route use case