Skip to content
worldfires
MG · Madagascar · –

Antsalova fire, 31 October 2026 Antsalova

Dying down · last detection

The fire started on in Antsalova, Madagascar, and satellites saw it burning for 22 days. Copernicus’s global GWIS system rebuilt a burnt perimeter of 4,106 hectares from heat detections, spread across a single municipality.

We are reviewing this page: some of its data look inconsistent (the place, the dates or the area), so treat them with caution.

GWIS near-real-time perimeter, from the layer that changed in August 2025: this fire is not compared with fires from earlier years, because since then the GWIS layer gives more and larger perimeters.

Is it still active?

No detections since 21 Nov 2026, 11:06 UTC. That usually means no flame is visible from space any more, although hotspots may remain. Satellites pass a few times a day and cloud or smoke can hide the fire: a lack of detections is not an official extinction. Each country’s emergency service declares a fire contained or out.

Wind, humidity and rain for the next 72 h

Burnt area
4,106ha
GWIS perimeter (Copernicus)
Duration
22days
31 Oct 2026 – 21 Nov 2026
Municipalities hit
1
districts
Started on
31 Oct
2026
Heat detections
167
in a single hotspot: 39 MW
Danger at start
15
Moderate · FWI index
Average daily power
130MW
detected power per day, on average

The next 72 hours

The wind is blowing from the west and pushing eastwards; no sharp shifts are expected until Wed 25, 14:00.

ECMWF IFS 0.25° at the centre of the fire · computed 22 Nov 2026, 10:20 UTC

No sharp changes in the next 72 h

No strong wind shifts and no dry nights. No warnings does not mean the fire will go out: in this weather it can carry on as it is.

Checked, no warning: no wind shifts of more than 60° nights recover humidity (peaks of 89 % or more) no useful rain expected (3.6 mm in 72 h)

Hour by hour, local time

Tap the strip to see each hour

Mean wind Gusts where it blows to Relative humidity Night (8 pm to 8 am) Rain (mm) · chance in %

Where is it blowing? The next 48 h

geoBoundaries/Madagascar National Disaster Management Office (BNGRC), OCHA ROSA (Creative Commons Attribution 3.0 Intergovernmental Organisations (CC BY 3.0 IGO))

At this hour

Places the fan reaches

    Is it heading my way?

    We compare your point with the fan for each hour. Your location never leaves your device.

    or tap the map

    This is not an official warning. It is a global model forecast with cells of about 25 km, at the centre of the fire: it cannot see valley winds or the winds the fire itself creates. The fan shows where the wind blows, not where the fire will spread. If in doubt, follow what the authorities say.

    Forecast: ECMWF IFS 0.25° (European Centre for Medium-Range Weather Forecasts), served by our own Open-Meteo instance; the chance of rain comes from the ECMWF ensemble. Variables: wind and gusts at 10 m, wind at 850 hPa, relative humidity and temperature at 2 m, rain. Open-Meteo.com (CC BY 4.0).

    Fire: GWIS / Copernicus (reconstructed from VIIRS) (CC BY 4.0); hotspots from the last 24 h: VIIRS from NASA FIRMS. Boundaries: geoBoundaries/Madagascar National Disaster Management Office (BNGRC), OCHA ROSA (Creative Commons Attribution 3.0 Intergovernmental Organisations (CC BY 3.0 IGO)).

    How it is worked out: wind shift = a change of more than 60° in 6 h or less, with at least 15 km/h before and after; no overnight recovery = peak humidity between 8 pm and 8 am stays below 55 %; useful rain = 2 mm or more in 6 h. The fan opens 18° plus however much the wind varies within ±3 h and reaches 25 km beyond the edge of the fire: one model cell, not a rate of spread. Methodology

    Where it burned

    HOTSPOTS BY DAY 31 Oct21 Nov
    21 Nov 2026

    How it evolved, day by day

    31 Oct 2026 · 0 MW · 0 2 Nov 2026 · 0 MW · 0 4 Nov 2026 · 0 MW · 0 5 Nov 2026 · 46 MW · 4 7 Nov 2026 · 68 MW · 4 8 Nov 2026 · 496 MW · 31 9 Nov 2026 · 225 MW · 20 10 Nov 2026 · 49 MW · 8 11 Nov 2026 · 167 MW · 22 12 Nov 2026 · 104 MW · 13 13 Nov 2026 · 60 MW · 6 14 Nov 2026 · 36 MW · 5 17 Nov 2026 · 11 MW · 2 19 Nov 2026 · 118 MW · 19 20 Nov 2026 · 107 MW · 21 21 Nov 2026 · 209 MW · 12
    31 Oct 8 Nov 11 Nov 17 Nov

    Fire radiative power detected each day, in megawatts, and number of hotspots. The magenta bar marks the peak day.

    Fire front film How it spread from one satellite pass to the next (33 passes) and with what wind. Watch the film

    The weather it burned in

    38°high
    47km/h · peak gust
    29%lowest humidity

    It had gone 0 days without rain (1 mm or more) when it started. Rain came after the fire: 1.0 mm on 22 Nov.

    This area had burned before

    72% of the perimeter overlaps earlier fires recorded since 2016: 626 ha in 2021, 5,639 ha in 2025. Part of it burned more than once, so the yearly figures add up to more than the 2,950 ha that had burned before.

    How big is it?

    10 km
    It burned the equivalent of 2.6% of the area of London (41 km² against 1,583 km²).

    London boundary: geoBoundaries · Office for National Statistics (Open Government Licence v3.0).

    This compares size, not location: both shapes are at the same scale, north up, centred on each other.

    What happened to each town

    1. Antsalova where it started
      6,211 ha reached on 31 Oct
      0.9 %

    Burnt area inside each municipality, from crossing the GWIS perimeter with administrative boundaries. The distance runs from the point marking the centre of each area to the nearest fire: staying a kilometre away does not mean it reached the houses.

    The weather day by day

    ECMWF ERA5 reanalysis at the fire location, the standard reference for past weather, and the Canadian Fire Weather Index (the same system Copernicus uses), computed by WorldFires from that data. Days after the fire run until the first rain.

    Open-Meteo.com (CC BY 4.0) · forecast models, provisional until the ERA5 reanalysis arrives

    Day High Min. humidity Peak gust Rain Danger (FWI)
    31 Oct 30.3 °C 41 % 39 km/h 0.0 mm Moderate 15
    1 Nov 33.0 °C 37 % 44 km/h 0.0 mm Moderate 16
    2 Nov 34.5 °C 36 % 47 km/h 0.0 mm Moderate 20
    3 Nov 35.8 °C 35 % 45 km/h 0.0 mm High 22
    4 Nov 34.1 °C 46 % 40 km/h 1.9 mm Moderate 20
    5 Nov 35.7 °C 43 % 38 km/h 3.2 mm Moderate 13
    6 Nov 36.5 °C 39 % 36 km/h 1.9 mm Low 10
    7 Nov 36.8 °C 33 % 41 km/h 1.2 mm Moderate 16
    8 Nov 38.3 °C 29 % 46 km/h 0.1 mm Moderate 21
    9 Nov 37.6 °C 37 % 39 km/h 2.2 mm Moderate 13
    10 Nov 34.8 °C 48 % 45 km/h 4.3 mm Moderate 12
    11 Nov 32.9 °C 52 % 38 km/h 5.6 mm Low 6
    12 Nov 34.6 °C 37 % 36 km/h 0.5 mm Moderate 13
    13 Nov 37.1 °C 33 % 44 km/h 0.8 mm Moderate 18
    14 Nov 36.0 °C 39 % 34 km/h 4.0 mm Moderate 16
    15 Nov 31.0 °C 44 % 18 km/h 8.8 mm Very low 0
    16 Nov 35.8 °C 39 % 34 km/h 3.0 mm Low 5
    17 Nov 36.5 °C 39 % 35 km/h 3.1 mm Low 9
    18 Nov 34.9 °C 44 % 35 km/h 0.8 mm Low 9
    19 Nov 35.0 °C 39 % 40 km/h 0.1 mm Moderate 17
    20 Nov 37.2 °C 29 % 44 km/h 0.0 mm High 23
    21 Nov 37.7 °C 34 % 40 km/h 2.1 mm High 28
    22 Nov after 37.4 °C 37 % 38 km/h 1.0 mm Moderate 16

    Before and after, from space

    Drag the line to compare the same ground before the fire and after it.

    Before · 24 Oct 2026
    After · 21 Nov 2026

    If the day is cloudy, change the date: each satellite passes once a day. In false colour the burnt ground looks brown or red.

    The smoke

    CAMS · Copernicus

    On 8 November 2026 the air beside the fire reached 21 µg/m³ of PM2.5, but no daily average passed the WHO guideline (15).

    PM2.5 peak21 µg/m³
    Days over the guideline0 days
    PM1026 µg/m³
    Optical depth0.53
    29 Oct2 Nov6 Nov10 Nov14 Nov
    Good Fair Moderate Poor Very poor Extremely poor

    Particles from the Copernicus CAMS model at the centre of the perimeter; these are not measurements in each town. The pale bar is the worst hour, the solid one the daily average.

    Open-Meteo.com (CC BY 4.0) · Copernicus Atmosphere Monitoring Service (CAMS)

    Detections table · 167Every satellite heat detection, one by one, with time, satellite, power and coordinates.
    ⬇ Download CSV Times in UTC. Each detection is a 375 m pixel: a large fire yields hundreds per pass.
    Date and timeSatelliteFRP (MW)LatLon

    Other fires nearby

    All fires in Madagascar

    Frequently asked questions

    How long did the Antsalova fire last?
    According to the heat detections seen by satellite inside the perimeter, it burned from 31 October 2026 to 21 November 2026 (22 days). The perimeter dates it from 5 November 2026 to 21 November 2026.
    How much area did it burn?
    The perimeter GWIS (Copernicus) rebuilt from heat detections is 4,106 hectares (41.1 km²). It is an estimate: official national figures may differ.
    Which municipalities burned?
    The fire only entered Antsalova (6,211 ha, 0.9 %).
    Is the Antsalova fire still active?
    No fire detections since 21 Nov 2026, 11:06 UTC. A lack of detections is not an official extinction.
    What was the weather like?
    During the fire the temperature reached 38.3 °C, humidity dropped to 29 % and gusts hit 47 km/h. The fire danger index (FWI) reached 28, class High.
    What weather is expected at the fire over the next 72 hours?
    The wind is blowing from the west and pushing eastwards; no sharp shifts are expected until Wed 25, 14:00. No strong wind shifts, dry nights or useful rain are expected. Times are local time at the fire. ECMWF IFS forecast (Open-Meteo) computed 22 Nov 2026, 10:20 UTC; this is not an official warning.
    Where does this data come from?
    From the GWIS (Copernicus) burnt perimeter, NASA FIRMS VIIRS hotspots and the administrative boundaries of Madagascar (geoBoundaries). It is not an official report.

    Where to verify it

    A satellite detection is not an official report. Final figures and the fire’s status are published by each country’s emergency service.

    Satellite cross-check

    History of this page · 10 Changes in area, dates, name, official record or merged pages, compared every night since 11 November 2026.
    1. Area: 3,724 → 4,106 ha.
    2. Last day with fire: 10 November 2026 → 14 November 2026.
    3. Area: 3,595 → 3,724 ha.
    4. Mapped perimeter arrives: from ≈ 2,716 ha estimated from hotspots to 3,595 ha.
    5. Start date: 3 November 2026 → 31 October 2026.
    6. Area: ≈ 2,548 → ≈ 2,716 ha.
    7. Start date: 2 November 2026 → 3 November 2026.
    8. Area: ≈ 2,044 → ≈ 2,548 ha.
    9. Start date: 1 November 2026 → 2 November 2026.
    10. Page published (≈ 2,044 ha).

    RSS Changes to this page With cross-checks and a citable version

    Does something look wrong on this page?

    This page is built from satellite data and official records, and it sometimes gets things wrong: the wrong dates, the wrong town, two fires merged into one, or something that was not a wildfire at all. If you know this fire first-hand (you live nearby, saw it, fought it or reported on it), tell us what really happened and we will check it.

    Report an error
    What do you think is wrong?

    We will only use your email to reply about this report. Privacy.

    How we know this
    • Perimeter: GWIS (Copernicus), rebuilt from heat detections rather than mapped on imagery; 2 polygons grouped in this page.
    • Dates adjusted with the VIIRS detections inside the perimeter, which said 5 November 2026 to 21 November 2026. 167 Suomi-NPP VIIRS detections on those dates.
    • We have not matched this page with any official record.
    • Weather: ECMWF IFS model, provisional until the ERA5 reanalysis arrives.

    How we build fire pages: methodology

    Where this data comes from

    The perimeter comes from the Copernicus Global Wildfire Information System (GWIS), which rebuilds it from MODIS and VIIRS heat detections and publishes it under CC BY 4.0. It is not measured on imagery as in Europe, so it is an estimate. Heat detections come from the VIIRS sensors aboard Suomi-NPP and NOAA-20, distributed by NASA FIRMS at 375 m resolution. Municipal boundaries come from geoBoundaries ADM2 (Creative Commons Attribution 3.0 Intergovernmental Organisations (CC BY 3.0 IGO)). The area per municipality is computed by WorldFires by crossing the perimeter with those boundaries, so it may differ from each country’s official figure.