Skip to content
worldfires
MZ · Mozambique · –

Maganja Da Costa fire, 4 September 2021 Maganja Da Costa, Mocubela

Out · last detection 9 Sept 2021, 10:41 UTC

The fire started on 4 September 2021 in Maganja Da Costa, Mozambique, and satellites saw it burning for 6 days. Copernicus’s global GWIS system rebuilt a burnt perimeter of 8,792 hectares from heat detections, spread across 2 municipalities.

It is the 118th largest in Zambézia since 2016 (of 562 with a page). Number 18 in Mozambique in 2021, of 233.

Burnt area
8,792ha
GWIS perimeter (Copernicus)
Duration
6days
4 Sept 2021 – 9 Sept 2021
Municipalities hit
2
districts
Started on
4 Sept
2021
Heat detections
337
in a single hotspot: 110 MW
Danger at start
28
High · FWI index
Average daily power
625MW
detected power per day, on average

Where it burned

HOTSPOTS BY DAY 4 Sept9 Sept
9 Sept 2021

How it evolved, day by day

4 Sept 2021 · 138 MW · 8 5 Sept 2021 · 646 MW · 54 6 Sept 2021 · 1,282 MW · 104 7 Sept 2021 · 1,543 MW · 151 8 Sept 2021 · 90 MW · 13 9 Sept 2021 · 48 MW · 7
4 Sept 7 Sept

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

The weather it burned in

34°high
46km/h · peak gust
20%lowest humidity

It had gone one day without rain (1 mm or more) when it started. Rain came after the fire: 1.4 mm on 18 Sept.

This area had burned before

88% of the perimeter overlaps earlier fires recorded since 2016: 1,877 ha in 2016, 1,019 ha in 2017, 4,576 ha in 2018, 4,380 ha in 2019, 4,500 ha in 2020. Part of it burned more than once, so the yearly figures add up to more than the 7,753 ha that had burned before.

How big is it?

10 km
It burned the equivalent of 5.6% of the area of London (88 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. Maganja Da Costa where it started
    6,516 ha
    2.3 %
  2. 2,318 ha
    0.5 %

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.

People and buildings inside the perimeter

GHSL 2020
Built-up area5.2 ha
Residents≈ 1,200

Inside the perimeter there were about 5.2 hectares of built-up land and about 1,200 residents (GHSL 2020, satellite-based estimate).

This is an estimate on 100 m cells, made from satellite imagery and census data: it does not mean that homes burned or that anyone was harmed. The perimeter also contains land that never burned.

Source: Global Human Settlement Layer (GHS-BUILT-S and GHS-POP R2023A), European Commission Joint Research Centre (JRC), © European Union, CC BY 4.0. Sum inside the perimeter computed by WorldFires. human-settlement.emergency.copernicus.eu

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) · Copernicus Climate Change Service (ERA5)

Day High Min. humidity Peak gust Rain Danger (FWI)
4 Sept 32.7 °C 31 % 32 km/h 0.0 mm High 28
5 Sept 34.1 °C 21 % 40 km/h 0.0 mm Very high 44
6 Sept 33.5 °C 21 % 37 km/h 0.0 mm Very high 44
7 Sept 34.1 °C 20 % 40 km/h 0.0 mm Very high 46
8 Sept 34.1 °C 25 % 44 km/h 0.0 mm Very high 41
9 Sept 31.8 °C 32 % 46 km/h 0.0 mm Very high 43
10 Sept after 28.7 °C 44 % 50 km/h 0.9 mm High 35
11 Sept after 28.7 °C 47 % 36 km/h 0.7 mm High 23
12 Sept after 27.9 °C 59 % 35 km/h 0.2 mm Moderate 20
13 Sept after 32.5 °C 32 % 45 km/h 0.0 mm High 36
14 Sept after 35.6 °C 19 % 36 km/h 0.0 mm Extreme 52
15 Sept after 36.6 °C 18 % 34 km/h 0.0 mm Very high 46
16 Sept after 37.1 °C 23 % 36 km/h 0.0 mm Very high 42
17 Sept after 38.3 °C 24 % 46 km/h 0.0 mm Very high 41
18 Sept after 28.6 °C 57 % 48 km/h 1.4 mm High 25

Its twins: fires that started in the same weather

It started with an FWI of 28.1, a minimum humidity of 31% and 1 day without rain. Fires in the area that started in weather like this since 2016: 15.

FWI at the start 28.1
Minimum humidity 31 %
Days without rain 1
this fire chosen twins rest of the area
Almost identicalSimilar

Outlines of the closest twins

Mocuba fire, September 2020
3,278 ha 20 days FWI 26.9 Min. humidity 32 % 86 km away, in 2020

First rain (1 mm or more) after it ended: none in the following ten days.

What happened to them

16 days
median duration
6,455 ha
median area
33 %
went past 10,000 ha
3.7 mm
rain at the end (median; 9 of 15 had it)

It burned an area similar to its twins (8,792 ha against a median of 6,455 ha) and lasted 6 days, against 16 days.

In 6 of 15, not even 1 mm fell in the ten days after the end: they went out without rain. In the rest, the first rainy day brought a median of 3.7 mm.

This is a bank of analogues: it shows what usually happens, not what will happen. With fewer than 15 twins we give no figures. How we calculate it

How it ended

1.4 mmfirst rain
18 Sept9 days after it ended
46km/h · max. gust

It was considered over on 9 Sept, and on 18 Sept 1.4 mm fell. We do not know whether the rain or the firefighters put it out. Twins with rain when they ended: 9 of 15, with a median of 3.7 mm.

Weather on this page: Open-Meteo.com (CC BY 4.0) · Copernicus Climate Change Service (ERA5)

How a twin is chosen

  • Same area: fires that have already ended, less than 250 km from this one, since 2016.
  • Same start: we compare the FWI on the first day, the minimum humidity that day and the run of days without rain before it. The slider sets how much they may differ.
  • Only pages with final weather data (ERA5 reanalysis) calculated in the same way as this fire’s; agricultural and grass burns and pages with data under review are left out.
  • Fires burning at the same time as this one (a week before or after) are left out: they share the same spell of weather and do not count as a precedent.
  • They are sorted by similarity, and the outlines of the first five are drawn.

How we calculate it

Before and after, from space

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

Before · 28 Aug 2021
After · 16 Sept 2021

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

Detections table · 337Every 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 Mozambique

Frequently asked questions

How long did the Maganja Da Costa fire last?
According to the heat detections seen by satellite inside the perimeter, it burned from 4 September 2021 to 9 September 2021 (6 days). The perimeter dates it from 4 September 2021 to 8 September 2021.
How much area did it burn?
The perimeter GWIS (Copernicus) rebuilt from heat detections is 8,792 hectares (87.9 km²). It is an estimate: official national figures may differ.
Which municipalities burned?
The fire entered 2 municipalities: Maganja Da Costa (6,516 ha, 2.3 %); Mocubela (2,318 ha, 0.5 %).
What was the weather like?
During the fire the temperature reached 34.1 °C, humidity dropped to 20 % and gusts hit 46 km/h. The fire danger index (FWI) reached 46, class Very high.
Where does this data come from?
From the GWIS (Copernicus) burnt perimeter, NASA FIRMS VIIRS hotspots and the administrative boundaries of Mozambique (geoBoundaries). It is not an official report.
How did the fires that started in similar weather in this area end?
Fires less than 250 km away that started since 2016 with a similar FWI, humidity and dry spell: 15. Their median duration was 16 days and their median area 6,455 ha; 33 % went past 10,000 ha. In 6 of 15, not even 1 mm fell in the ten days after the end: they went out without rain. In the rest, the first rainy day brought a median of 3.7 mm.

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 · 1 Changes in area, dates, name, official record or merged pages, compared every night since 24 October 2026.
  1. Last day with fire: 8 September 2021 → 9 September 2021.

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.
  • Dates adjusted with the VIIRS detections inside the perimeter, which said 4 September 2021 to 8 September 2021. 337 Suomi-NPP VIIRS detections on those dates.
  • We have not matched this page with any official record.
  • Weather: Copernicus ERA5 reanalysis at the fire location, in cells of about 25 km.
  • Built-up area and population come from the JRC Global Human Settlement Layer (epoch 2020, 100 m cells), summed inside the perimeter: an estimate of what was there, not a damage count.

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.