UN-SPIDER maps burn severity in Greece and Portugal

Landsat 8 true colour image of the fire in Mati, Greece. Image: USGS.

Countries all over Europe are fighting wildfires that are engulfing forests and endangering population centres. The death toll of a wildfire that raged on 23 July close to Mati, a seaside resort near Athens, Greece, was raised to 91. In early August, a large wildfire in the region of Monchique, south of Lisbon, was brought under control after over 40 people had been injured, hundreds had been forced to evacuate from their homes and about 235 000 square kilometres were burned. As wildfires continue to rage in multiple regions of Portugal, monitoring their spread and effects is crucial for disaster management and emergency response efforts.

Freely accessible satellite imagery can be used to map the extent of the damage caused by forest fires. To support disaster managers in making use of space-based data in such a situation, UN-SPIDER developed a step-by-step procedure on how to obtain, process and map data about burn severity to show how fires affect the functioning of a local ecosystem. Maps illustrating the severity of burns can support the preparation of plans for emergency rehabilitation and restoration post-fire. To do so, satellite images at certain wavelengths before the fire are compared to those during and/or after the fire. In this way, the development and total impact of the fire can be monitored.

Using its Recommended Practice for burn severity mapping, UN-SPIDER mapped the aggregate effect of the forest fire in Mati, Greece. This was done by comparing Landsat 8 images from 2 August 2018 (post-fire) to the state of the vegetation on 3 June 2018 (pre-fire). Figure 1 illustrates the resulting severity of burns in the region using QGIS, a free and open source desktop geographic information system (GIS) application.

Burn Severity Map Attica, Greece created with the UN-Spider QGIS recommended practice to assess burn severity.

Figure 1. Burn Severity Map Attica, Greece created with the UN-SPIDER Recommended Practice to assess burn severity.

In the practice, burn severity is calculated through an index called the Normalized Burn Ratio (NBR). This index exploits the difference in temperature radiance captured by satellites of a vegetated and burnt area, and is created by taking a ratio of the difference between the near-infrared (NIR) and shortwave infrared (SWIR) wavelengths. The results show that high severity burns covered an area of 48,602 square meters and indicate the part of Mati that was most affected.

Due to the revisit time of the Landsat 8 satellite used for this Recommended Practice, the effects of the fire in Monchique, Portugal, which was still ongoing during the time of mapping, are illustrated for the effects that were measurable on 5 August 2018 compared to 20 July 2018. An update will be presented when the first Landsat 8 satellite imagery after the extinction of the fire becomes available.

Burn Severity Map of Monchique, Portugal created with the UN-Spider R recommended practice to assess burn severity.

Figure 2. Burn severity map Monchique, Portugal created with the UN-SPIDER Recommended Practice to assess burn severity.

Figure 2 represents the burn severity as mapped using the R programming language in RStudio. Both approaches are comparable as the burn severity classes are strictly defined according to the USGS standard table. The technique used depends on the preference of the user.

Publishing Date: 

10/08/2018
Zircon - This is a contributing Drupal Theme
Design by WeebPal.