<- Back to overview


Wildfires: a Flickering Disaster

An explainer article covering the basics of how wildfires start and spread and what role they play in nature. Published as a longtail piece with a focus on the prevalence of fires in Switzerland and Europe.

Written, researched, and visualized in collaboration with Sven Titz. Read the article in German here and English here
Read a Twitter thread about the process here

We covered:

  1. How do wildfires start? 
  2. What role do they play in nature?
  3. How do we fight them? 👩🏼‍🚒
  4. What’s the deal in Switzerland 🇨🇭
  5. What’s the deal in Europe? 
  6. How do we prevent – or manage them? 

Wildfire types 
While ground fires are most frequent and crown fires are most dangerous, one could consider peat fires the more insidious. These smoulder subterraneously and can even travel below roads.

Before humans made friends with fire, wildfires were ignited exclusively by natural occurrences, like lighting. In the more populated regions of the world however, wildfires are caused by humans.

Most fires are caused by us
Carelessness is commonly the cause: campfires not properly extinguished can spread easily into a forest. Discarded cigarettes are also a frequent ignition source.

32% of Switzerland is covered in forest. For every one Swiss person, 66 surround her. Two-thirds are coniferous trees and the other third is deciduous.

Forest coverage 
Deciduous (light green) or coniferous (dark green)
The three most common species are spruce, fir and beech. The regions of Jura and the southern Alps have especially thick forests.

Wildfires burn most frequently in the southern Alps. Depending on the region, wildfires occur in different seasons. 

More area in the southern Alps burn each year than anywhere else in Switzerland 
Total burned areas (in hectares) in the southern Alp region 

Winter on the southern side of the Alps is extremely dry. The Föhn, a hot southerly wind on the northern slopes of the Alps, plays a deciding factor by drying out the layer of branches and leaves at the bottom of chestnut forests. As a result, it is a haven for forest fires.

At the end of December in 2006 in the Misox Valley
In the southwest of canton Graubünden, there’s still no snow. Then, the forest starts to burn with strong winds that drive the fire up the valley slopes. In just a few days the fire expanded from 20 football fields in size to 100, or 119 hectares.

Flames become particularly problem where the forest comes in contact with human settlements: in the so-called Wildland Urban Interfaces (WUI). Where, for example, trees give shade to houses or hiking paths leading through fire-proned areas. A WUI map, such as the following of the city of Bellinzona, shows where there is a higher risk of man-made forest fires:

Bellinzona Wildland Urban Interface
Red = areas at risk — A map with WUI components (with relevant infrastructure and forest area) is dynamic and can allow foresters and fire managers to easily identify areas of high fire risk and where it is necessary to concentrate financial and technical funds for fire prevention measures.

Read the article in German here and English here. Read a Twitter thread about the process here.

Methodology in Detail

In order to show the forested areas in Switzerland, we visualized data from the National Forest Inventory (NFI), a program from WSL and BAFU. This dataset uses a remote-sensing approach to map deciduous (leafy) and coniferous (needle) trees nationwide with a spatial resolution of ten by ten meters. A single data point thus provides an indication of the likelihood that an area will be leafy or needle-dominated.

Data showing the number of fires, amount of hectares burned and the distribution of wildfire causes is made available by the Eidgenössische Forschungsanstalt für Wald, Schnee und Landschaft (WSL) through their wildfire database (SwissFire).

The Wildland Urban Interface (WUI) map typically shows areas where building and other infrastructure come in contact with natural vegetation or forested areas. In alpine regions, the WUI is computed differently than in North America. Our map highlights areas where there is a large concentration of people and activity that causes fires. To define places with increased risk of ignition (i.e. the radius), we calculated a 100 meters buffer around roads and easily accessible buildings (buildings 100 meters from a drive-able road). How large the buffer is, is dependant on the characteristics and geographic location of the area in question.

In order to show active fires in Europe, we worked with NASA data. This dataset lists fires detected with an algorithm from the MODIS satellite that uses infrared radiation. The algorithm examines each pixel of the satellite image and sorts it into the following classes: missing data, clouds, water, no fire, fire, unknown. The algorithm assigns the following situations to class «fire»:
– intense fire activity on a part of the pixel surface
– fire activity that extends over a larger area.
We display only the «Hotspots» from the data which are categorized with a confidence interval of 80% or more.

Special thanks to Marco Conedera and Boris Pezzatti (WSL)

<- Back to overview