Pilot on Wildfires in the Aragon Region – Spain



Overall Policy Objective:

Improve the design of emergency policies related to wildfires and management of controlled fires.

The Context.

Forest fires in Aragon have increased significantly over the last two decades, both in number and in the total area covered by them. In addition, farmers need to conduct controlled fires to remove undesirable vegetation debris. Currently, requests are made on-line, but permits are granted for several days (even months) and there is no exact control of when each controlled fire occurs. It is obvious, that this is not a desirable situation, and that having control of the exact moment of a controlled fire, as well as its risk (based on satellite or other data) could improve efficiency in fire prevention. Copernicus ecosystem through DIAS provides an incredible amount of data API and visualizations freely available online, such as the EFFIS Current Situation Viewer. These services provide us a very useful tool to have a deeper understanding about fire risk zones, fire evolution, etc. Based on that, our pilot proposes a more collaborative approach. Also, citizen science methodology is a powerful tool to gather and analyze deep amount of information, through citizen resources and knowledge. Citizens in general could issue alerts about possible fire detection, as well as give useful information about certain elements that can alter the evolution of a fire (such as the presence of fuel in the area).

Pilot Description:

The pilot (named CrowdWall) is a collaborative emergency management pilot that capitalizes on the volunteer work of people affected by a forest fire. It allows citizens to collect data from their environment (such as wind intensity and direction in micro-zones, quality and quantity of fuel in the area, sunlight, temperature, etc.) in order to create a crowdsourced map that facilitates the prediction of fire behaviour by public emergency services. Crowdwall will also provide the opportunity to apply for controlled fire permits (e.g. farmers). Whether or not these permits are granted will depend on the calculated fire risk at that time (through data from the Copernicus ecosystem, citizen data, and data from the Government of Aragon). The person responsible for carrying out the controlled fire will indicate in real time the start and end of the fire, so that this data can be contrasted with the data offered for example by MODIS6 and EFFIS and the Aragon Open Data Repositories7. Furthermore, in the event of a fire, the person in the risk area can be traced, ensuring their safety at all times or they can share their own observations to contribute to the construction of the active fire map. This information is stored anonymously on the cloud and is used to update the map of active fires. This new scenario will contribute in a best design of emergency policies regarding fire alert management. On the other hand with DECIDO, we will research on the correlation between risk zone and actual fire (and its magnitude) using AI techniques through Tensorflow8 and Keras frameworks. In order to work on this, we will look for further datasets – such as OpenWeather. In summary, Crowdwall is a collaborative and data-centric approach to protect the life of our firefighters and avoid as many personal and material losses as possible.

Co-creation and engagement:

In order the pilot to be useful to target groups, their contribution in the design phase is essential. We will develop co-creating workshops in which we will gather information and requirements from all the stakeholders, in particular, we have identified the following target groups:

Affected citizens:

  • Citizen living in risks zones in Aragon
  • People who need to make controlled fires (e.g. farmers)
  • Fire-fighters, to use the data (both from VIIRS and EFFI, and those provided by citizens) for a better management of fire extinction.
  • Policy Makers and local authorities, to use the information provided by the pilot
  • General public, to be aware if they are near a risk zone and/or a fire

Check out the latest newsletter

Skip to content