Every active fire seen from space since 2000, in a few lines of Python.

What it does

Small, typed, honest about the data

Built for researchers, data engineers and civil-protection teams who want FIRMS data in a notebook or a pipeline, without the e-mail order form.

Quick start

A country's year, no key needed

pip install firms-client            # the library and the firms command
pip install 'firms-client[mcp]'     # plus the firms-mcp server

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:  # that year is not published (yet)
                continue
            fires = select(rows)  # vegetation fires, nominal confidence or better
            types = dict(sorted(Counter(r.detection_type for r in rows).items()))
            print(product.slug, len(rows), "detections", types, "->", 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 CSVs are now in ~/.cache/firms-client/: run it again and nothing is downloaded.

From a shell

firms archive Italy 2025 --fires --format geojson > fires.geojson
firms anomaly Italy 2017 --month 7
FIRMS_MAP_KEY=… firms area --bbox 6,36,19,47.6 --days 2 --mask Italy --fires

From Claude, or any MCP client

claude mcp add firms -e FIRMS_MAP_KEY=your-key -- firms-mcp

Then ask: "how did the 2021 fire season in Sicily compare with 2025?" The tools answer with digests computed by firms-client, never raw rows.

Or the last days, with a MAP_KEY

The archive runs up to the previous year. For the last days, get a free MAP_KEY and ask the area API:

from firms_client import fetch_area, select

ITALY = (6.0, 36.0, 19.0, 47.6)  # west, south, east, north

rows = await fetch_area(client, MAP_KEY, "VIIRS_SNPP_NRT", ITALY, days=3)
mask = await fetch_static_mask(client, "Italy")  # area rows have no type: learn it from the archives
fires = select(mask.label(rows), min_frp_mw=1.0)

Fire or factory?

Not every hotspot is a fire

A satellite sees heat, not fire. Steelworks, refineries, gas flares, volcanoes and offshore platforms are hot every day of the year, and they come through at high confidence, because they really are hot.

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

How it works

Two doors in: archives and area API

Reference

API at a glance

Every async function takes your httpx.AsyncClient: timeouts, proxies, retries and pooling stay yours.

Full details, recipes (fire history, fire-days per km², GeoJSON, a wildfire trigger) and the traps we already fell into are in the README.

Roadmap

From hotspots to answers

Shipped: fire events, digests for LLMs, the MCP server, the static-source mask and season anomalies. Next: alert bulletins grounded in computed numbers, crop burning told apart from wildfires, and training sets for fire-risk models.