firms-client

Async Python client for NASA FIRMS active-fire detections (MODIS since 2000, VIIRS since 2012): the open per-country archives, the near-real-time area API, and the parsing that tells a vegetation fire from a steelworks.

One dependency: httpx.

Why a library

On Italy, about a quarter of the VIIRS detections are not vegetation fires. Without type == 0, the most "burnt" square kilometre in the country was the ILVA steelworks in Taranto, with 3,308 fire-days.

All checked against the live service on 3 November 2026.

Install

pip install firms-client

Use

The yearly archive of a country, no key needed:

import asyncio
from collections import Counter

import httpx

from firms_client import ARCHIVE_PRODUCTS, fetch_archive, select


async def main() -> None:
    async with httpx.AsyncClient(timeout=120) as client:
        for product in ARCHIVE_PRODUCTS:
            rows = await fetch_archive(client, product, 2025, "Italy")
            if rows is None:
                continue
            fires = select(rows)  # vegetation fires, nominal confidence or better
            print(
                product.slug,
                len(rows),
                "detections",
                dict(sorted(Counter(r.detection_type for r in rows).items())),
                "->",
                len(fires),
                "fires",
            )


asyncio.run(main())

modis 5076 detections {0: 3414, 1: 21, 2: 1635, 3: 6} -> 3315 fires
viirs-snpp 23349 detections {0: 17533, 1: 633, 2: 4277, 3: 906} -> 14880 fires

The archive runs up to the previous year or so; for recent days use the area API with a free MAP_KEY:

from firms_client import fetch_area, select

rows = await fetch_area(client, MAP_KEY, "VIIRS_SNPP_NRT", (6.0, 36.0, 19.0, 47.6), days=3)
fires = select(rows, min_frp_mw=1.0)

Selecting fires

select(rows, vegetation_only=True, min_confidence="nominal", min_confidence_pct=None, min_frp_mw=0.0):

• vegetation_only drops volcanoes, industry and offshore (type 1-3); a row without type is kept, since it cannot be judged;
• VIIRS rows are judged on their level; MODIS percentages on min_confidence_pct when given, otherwise mapped to levels with the FIRMS guidance (below 30 low, 30-79 nominal, 80+ high);
• a missing confidence is kept: losing a real fire is worse than keeping a doubtful pixel.

Two things no filter fixes: crop-residue burning is a real vegetation fire (type 0), and FIRMS is presence-only, so no detection does not mean nothing burnt.

Bring your own HTTP

archive_url, area_url, is_hotspot_csv and parse_csv are pure.

Data

FIRMS data are provided by NASA LANCE / FIRMS and are free to use; please acknowledge "NASA FIRMS" in anything you publish. The test fixtures are a few rows of the 2025 Italian archives.

Extracted from Limen, which uses the archives as the fire history of Italy and the area API as a wildfire trigger.

Development

uv sync
make check

License

Code: Apache-2.0.