Burning Forests: Instruments for Monitoring and Reporting Wildfire Harm

Should you’ve seen experiences of a wildfire in your area and also you’re on the lookout for open supply information, NASA’s fire-tracking device is commonly the primary place to start out. It gives a warmth signature and an approximate location. However detection is barely step one in understanding what’s occurred. On this information, we discover methods to analyse and report on the dimensions and severity of wildfires, together with these in protected areas the place ecosystems are sometimes most fragile. We additionally look at how usually fires recur in the identical area over a number of seasons, serving to to determine patterns in fireplace exercise as local weather change reshapes fireplace threat world wide. 

Satellite tv for pc imagery from Copernicus Browser might be used to visualise the unfold of the fireplace, and vegetation well being indices to evaluate burn severity. The datasets will then be mixed in QGIS for extra in-depth evaluation. At every stage, solutions might be provided for turning the information into clear, reportable findings.

All through this information, a single case examine might be used: Sicily’s Zingaro Nature Reserve. In 2025, wildfires swept throughout the area, destroying forests, grasslands and croplands. Positioned on the Capo San Vito peninsula, the reserve was so severely affected that sections stay closed immediately. 

Visualising Scorched Earth

When investigating a wildfire, it’s vital to slender down when it occurred and the place it unfold. The Landsat and Sentinel-2 missions are well-suited to this activity, offering common free imagery of a lot of the Earth’s landmass.

Beneath are two units of Sentinel-2 imagery displaying situations shortly earlier than and after a fireplace on July 25, 2025, close to Capo San Vito, Sicily. The highest two photographs are true-colour, just like what could be seen from an aeroplane window. The picture on the highest proper reveals an space of scorched earth on the japanese aspect of the peninsula, however the precise extent of the fireplace is troublesome to find out as a result of the color of the bottom has modified solely barely.

The underside two photographs are false-colour and spotlight the distinction between wholesome vegetation and burned areas. Such imagery is feasible as a result of Sentinel-2 captures bands of sunshine outdoors the seen vary, a method referred to as multispectral imaging. In these photographs, the near-infrared (NIR) band is colored inexperienced, and the shortwave infrared (SWIR) band is colored crimson. Wholesome vegetation primarily displays NIR mild, so it seems inexperienced, whereas burned areas primarily replicate SWIR mild, so they seem crimson.

These photographs had been created with Copernicus Browser, a free browser-based device from the European House Company for accessing and dealing with Sentinel imagery. It permits customers to browse the Sentinel-2 catalogue by date and visualise completely different band mixtures. You don’t want an account to make use of the browser, however signing in permits further options.

Should you’d prefer to attempt Copernicus Browser with out additional clarification, you’ll be able to go straight to the false-colour post-fire picture right here. 

To comply with alongside step-by-step, first open Copernicus Browser. Then to visualise Sentinel-2 imagery:

1. Zoom to the specified space on the map or use the search bar (San Vito Lo Capo, north-west Sicily)
2. Choose the date of curiosity (‘2025-07-27’ chosen beneath).
3. Choose the layer of curiosity (‘True colour’ by default; SWIR chosen beneath).

By figuring out the final out there picture earlier than the fireplace and the earliest picture after it during which the total burn space is seen, it’s doable to determine the situation and timeline of the fireplace.

This permits us to report the next discovering: “Satellite tv for pc imagery reveals the extent of the injury attributable to wildfires throughout the Capo San Vito peninsula on Sicily’s northern coast between July 20 and July 27, 2025.”

Further train: Lookup a latest fireplace (e.g., wildfires close to Penco, Chile in January 2026), navigate to the affected location and attempt to visualise the burned space utilizing Copernicus Browser.

Quantifying the Burned Space

The visibly scorched space might be measured utilizing the Space of Curiosity device (highlighted beneath), which permits customers to attract a polygon on the map and calculate the entire space in sq. kilometres. (As soon as drawn, hold the polygon within the editor, as it will likely be used once more later.)

Reportable discovering: “The wildfire that swept throughout Sicily’s Capo San Vito peninsula in 2025 burned greater than 50 km2 of the peninsula, in keeping with Sentinel-2 information.”

Repeatedly measuring the burned space over time permits you to comply with the development of a hearth. This methodology was utilized by Bellingcat when overlaying the Etosha Nationwide Park wildfire in late September 2025.

Further train: Replicate the evaluation of the Etosha Nationwide Park fireplace from this Bellingcat article.

Assessing Burn Severity 

Some fires solely have an effect on floor vegetation, whereas others scorch the bottom and trigger long-lasting injury. Burn severity might be measured utilizing an index known as the Normalised Burn Ratio (NBR).

Copernicus Browser doesn’t embody a default NBR layer, however it may be added through a customized script, as proven within the screenshot beneath:

1. Choose ‘Customized’ within the layer selector.
2. Swap from the ‘Composite’ to the ‘Customized’ tab.
3. Test ‘Load script from URL’.
4. Paste this URL: https://bellingcat-scripts.ams3.cdn.digitaloceanspaces.com/NormalizedBurnRatio.js 
5. Load the script by clicking the inexperienced round arrows to the correct of the URL.
6. Click on ‘Apply’ (chances are you’ll have to scroll down).

Alternatively, you’ll be able to skip these steps and go straight to the customized NBR post-fire picture right here.

The NBR layer shows constructive values in inexperienced (wholesome vegetation) and detrimental values in purple (burned areas), making the boundary of the scorched space a lot clearer than earlier than.

To calculate the change in NBR in Copernicus Browser, use the Statistical Data device (a free account is required to entry this characteristic).

1. Inside the date selector, select a date a couple of weeks or months after the fireplace.
2. Utilizing the Space of Curiosity polygon, choose the ‘Statistical Information chart’ icon.

3. Set the utmost cloud cowl to round 30% utilizing the slider within the high proper.
4. Choose a date vary that captures the out there information surrounding the fireplace (July 20-27 proven beneath).
5. Determine when the fireplace occurred on the graph (this might be marked by a pointy drop within the NBR, as proven beneath).
6. Hover over the factors on the graph instantly earlier than and after the fireplace to show the imply worth.

On this instance, the pre-fire picture had a median NBR of 0.11 and the post-fire picture had a median NBR of -0.18. The NBR decreased by 0.29, which represents a reasonable burn.

Reportable discovering: In late July, the fireplace, which scorched greater than 50km2 of Sicily’s Capo San Vito peninsula, was deemed reasonably extreme in keeping with the US Forest Service tips. 

Further train: Discover a customized visualisation script of curiosity from this repository and discover what it does.

Wildfires in Conservation Areas

By specializing in protected websites akin to nature reserves and nationwide parks, we will start to evaluate how wildfires have an effect on areas of excessive conservation worth. Managed burns are broadly utilized in agriculture and land administration, however unchecked fires in protected areas threat eroding fragile ecosystems.

The proportion of the Zingaro Nature Reserve that was broken by the fireplace might be estimated by combining the NBR picture created in Copernicus Browser with a dataset from Protected Planet, a world map of protected areas that features nature reserves.

QGIS, a program for working with geographic information, is well-suited for this kind of evaluation. Obtain and set up QGIS in your pc. For assist with this step, seek advice from the QGIS set up information.

To obtain the NBR picture from Copernicus Browser:

1. With the NBR visualisation chosen, click on the ‘Obtain’ icon. 
2. Swap tabs on the high from ‘Primary’ to ‘Analytical’.
3. Change the picture format to ‘TIFF (32-bit float)’.
4. Change the picture decision to ‘HIGH’.
5. Change the coordinate system to ‘Common Internet Mercator (EPSG:3857)’.
6. Toggle the ‘Clip further bands’ change to the off place (see picture beneath).
7. Choose the ‘Customized’ layer verify field (and deselect any others).
8. Click on ‘Obtain’.
9. Wait. It may take a number of minutes for the picture to be generated and downloaded.

As soon as the picture has downloaded, rename it to NBR.tiff to make it simpler to work with. 

Subsequent, open QGIS and click on ‘New Undertaking’ within the higher left. 

Load the picture from Copernicus Browser by dragging and dropping the downloaded file into QGIS.

Processing the NBR Picture

Subsequent, we categorise every pixel within the NBR picture as burned or unburned. 

Earlier evaluation in Copernicus Browser confirmed that the Zingaro Nature Reserve’s NBR worth dropped beneath zero solely after the fireplace (earlier than picture: imply NBR worth on July 20, 0.11; after picture: imply NBR worth on July 27, -0.18).

We will use this evaluation to set a threshold; something beneath zero might be categorised as burned.

The QGIS Raster Calculator lets us apply our threshold to the NBR picture and create a brand new layer. 

Open the Raster Calculator by deciding on ‘Raster > Raster Calculator…’ from the menu bar on the high.

The Raster Calculator lists the raster bands out there within the challenge. On this instance, there are 5. These bands are set by the customized script we utilized in Copernicus Browser and are numbered as follows:

1. Pink
2. Inexperienced
3. Blue
4. Pixel validity (not used on this instance)
5. NBR index

To create a brand new raster layer that applies our threshold on the NBR index band:

1. Double-click the fifth band (ending ‘@5’) so as to add it to the expression field on the backside. 
2. Add < 0 utilizing your keyboard (proven above).
3. Choose the ‘Create on-the-fly raster as an alternative of writing layer to disk’ checkbox.
4. Click on ‘OK’.

The expression NBR@5 < 0 tells QGIS to classify NBR index values as burned if they’re lower than zero. 

The brand new layer reveals burned areas as white (a price of 1), and unburned areas as black (a price of 0).

Further train: Obtain an NBR picture captured earlier than the fireplace. Use the Raster Calculator to create a brand new layer that reveals burn severity.

Including Conservation Space Knowledge

Obtain the Zingaro Nature Reserve dataset from Protected Planet by deciding on ‘Obtain > File Geodatabase’.

As earlier than, drag and drop the downloaded file into QGIS. This time, the obtain is a zipper file and accommodates many PDF information in addition to the geodatabase file of curiosity. Scroll right down to the underside of the listing and choose the ‘gdbtable’ file with a polygon icon on the left aspect (see the blue highlighted row beneath), then press ‘Add Layers’.

This provides the character reserve polygon as a layer in QGIS (and provides it an arbitrary color). The character reserve is sort of utterly contained throughout the white burned space, indicating it was closely affected by the wildfire.

Quantifying the Burned Space within the Nature Reserve

To measure the proportion of the character reserve that was burned by the wildfire, we are going to use the Zonal Histogram device from the QGIS Processing Toolbox to depend the variety of unburned and burned pixels throughout the reserve polygon.

Open the toolbox with ‘Processing > Toolbox’, and a pane ought to open to the correct. Within the Processing Toolbox search subject, search for ‘Zonal Histogram’ and double-click the outcome to open the device.

To create a brand new layer:

1. Set the ‘Raster layer’ to the brink burn space layer (NBR@5 < 0)
2. Set the ‘Vector layer containing zones’ to the character reserve polygon layer (ought to begin with ‘WDPA_’).
3. Click on ‘Run’
4. Click on ‘Shut’

It will create a brand new layer known as ‘Output zones’, which is a duplicate of the character reserve polygon with pixel counts added.

Choose the output layer within the decrease left and click on ‘Attribute Desk’ within the higher proper. (The attribute desk is a spreadsheet-like view of the information contained in a layer.) 

For the output layer, there is only one row as a result of there is just one polygon. If the layer contained many polygons, there could be many rows.

The newly calculated counts are added to the tip of the desk, so scroll all the best way to the correct. Search for fields beginning with ‘HISTO_’. Right here, HISTO_0 is the depend of unburned pixels (worth of 0), and HISTO_1 is the depend of burned pixels (worth of 1).

To calculate the proportion of burned space, the variety of burned pixels is split by the entire variety of pixels.

Proportion = 57413 / (57413 + 2195) = 0.96318…

A price of 0.96318 implies that simply over 96.3% of the character reserve burned.

Reportable discovering: In late July, greater than 95% of the Zingaro Nature Reserve burned in a wildfire, in keeping with Sentinel-2 satellite tv for pc imagery and Protected Planet information.

Monitoring Previous Wildfires

To evaluate the importance of an ongoing wildfire, it is very important place it in historic context. How does it examine with earlier fires in the identical space? Is it a part of a seasonal sample, or does it signify an unusually extreme occasion?

With protection relationship again to 2008, the European Forest Fireplace Data System (EFFIS) routinely maps wildfires throughout Europe, North Africa, and elements of the Center East.

Fireplace information might be requested instantly from EFFIS utilizing internet kind, with outcomes delivered by e mail. For ease, you too can obtain Bellingcat’s archived copy of EFFIS wildfire information for Italy overlaying 2015–2025.

For this part, it’s best to open a brand new QGIS challenge.

To view and analyse historic wildfires within the Zingaro Nature Reserve utilizing EFFIS information:

1. (Non-obligatory) Add the OpenStreetMap layer from the XYZ Tiles class by double-clicking it.
2. Load the EFFIS information into QGIS. If prompted to pick a coordinate transformation, click on ‘OK’ to just accept the default possibility. 
3. Load the Protected Planet Zingaro Nature Reserve polygon as described earlier. 
4. Open the Vector Intersection device by deciding on ‘Vector > Geoprocessing Instruments > Intersection…’ from the menu bar on the high.

The Intersection device creates a brand new layer containing solely the fires that affected the Zingaro Nature Reserve. To create the brand new layer:

5. Set the ‘Enter layer’ to the EFFIS fires layer.
6. Set the ‘Overlay layer’ to the Zingaro Nature Reserve polygon layer.
7. Click on ‘Run’.

QGIS performance might be prolonged by way of plugins, together with Knowledge Plotly, which provides information visualisation instruments. To put in Knowledge Plotly, open the Plugin Supervisor by deciding on ‘Plugins > Handle and Set up Plugins…’ from the menu bar, then:

8. Seek for ‘Knowledge Plotly’ within the out there listing.
9. Choose the plugin from the search outcomes.
10. Click on ‘Set up Plugin’ to obtain and set up it. 

As soon as put in, open the Knowledge Plotly panel with ‘View > Panels > DataPlotly’. The panel ought to seem on the right-hand aspect of the QGIS window. 

To plot a graph of historic wildfire exercise throughout the nature reserve, configure Knowledge Plotly as follows:

1. For ‘Plot sort’, select ‘Bar Plot’.
2. Set the ‘Layer’ to the newly created ‘Intersection’ layer.
3. In ‘X subject’, sort “yr(initialdat)”. This expression extracts the yr from the fireplace’s approximate begin date, permitting occasions from the identical yr to be grouped collectively. 
4. In ‘Y subject’ enter “$space/1000000”. This expression calculates the burned space throughout the nature reserve in sq. kilometres.

Notice: EFFIS information present preliminary and closing dates for every fireplace, that are approximate as a result of they rely on the supply of satellite tv for pc imagery. These dates ought to be handled as bounds for when a fireplace occurred, fairly than because the dates when it began and ended.

Subsequent, change to the Format tab in Knowledge Plotly:

5. Untick ‘Present Legend’. Solely do that for easy plots the place a legend just isn’t required. 
6. Add a title and labels for the X and Y axes.
7. Lastly, click on ‘Create Plot’ and wait a couple of seconds for the chart to be generated.

The chart reveals that the Zingaro Nature Reserve has skilled a number of vital wildfires over time. Nevertheless, in 2025, the information present that the fireplace burned a bigger space throughout the reserve than the main fires recorded in 2020 and 2017.

Reportable discovering: The Zingaro Nature Reserve has skilled three main wildfires since 2015. Of those, the 2025 fireplace burned a bigger space throughout the reserve than these recorded in 2020 and 2017.

The instruments and strategies on this information might be utilized to wildfires in lots of different areas. By combining satellite tv for pc imagery with environmental and historic datasets, it’s doable to maneuver past detection and start to quantify a fireplace’s influence. In doing so, you too can place particular person incidents in context, revealing whether or not they’re a part of a recurring sample or an unusually extreme occasion.

To be taught extra about fireplace detection, see Bellingcat’s information to NASA FIRMS.

To discover QGIS additional, go to the Bellingcat toolkit entry on QGIS.

Merel Zoet and Claire Press contributed to this report.

This information accommodates modified Copernicus Sentinel information (2025), processed with Copernicus Browser, in addition to information from the European Forest Fireplace Data System (EFFIS) of the European Fee Joint Analysis Centre.

Bellingcat is a non-profit and the power to hold out our work depends on the sort help of particular person donors. If you need to help our work, you are able to do so right here. You can even subscribe to our Patreon channel right here. Subscribe to our Publication and comply with us on Bluesky right here, Instagram right here, Reddit right here and YouTube right here.