PICUS

Integrative concept for data fusion and optimisation of forest fire monitoring processes

Overview

Forest fire monitoring has undergone a major technological shift in recent years through the use of analytical methods (artificial intelligence, statistical data analysis) and the application of new technologies (high-resolution satellite imagery, large-scale sensor networks, etc.). However, many challenges remain.

The aim of this project is to create a database and web-based platform to identify effective and spatially extended forest fire monitoring strategies while reducing costs.

Abstract

In Austria, approximately 85% of forest fires can be attributed to anthropogenic factors, whereby models that take anthropogenic factors into account currently only have a maximum accuracy of 60% in the forest fire detection rate. Building on the existing initiatives Waldbrand.at and IFDS (Integrated Forest Fire Danger Assessment System), new methods are used in the PICUS project to determine the information content of the individual parameters and to use artificial intelligence to identify strategically significant areas that have both a high fire potential and pose an increased risk to humans and society in the event of a forest fire (e.g. due to the proximity of critical infrastructure, settlements, hiking trails, etc.).

Based on selected scenarios, the use of different monitoring technologies is evaluated by means of a cost and impact analysis in order to determine an optimum between spatial extent and cost efficiency.

The results of the analyses as well as the interactive presentation of the individual parameters are output via a web-based platform. Through workshops with the respective stakeholders as part of the dissemination of the project results, the project aims to create an inclusive concept for collaborative data collection/use and the bundling of domain-specific know-how in the long term.

Official Project Lead

Universität Wien