Summary
The SeasFire project aims to explore and capture potential spatio-temporal asynchronous links happening between pre-occurring and non-overlapping atmospheric conditions and European fire regimes to predict the seasonal burned areas sizes in Europe by leveraging two major advancements of our time: a) the availability of a huge amount of satellite data with a good spatio-temporal resolution, which will be used as fire drivers called the Earth system variables, and b) the progress in Deep Learning (DL) and especially Graph Neural Networks (GNNs) have proved capable of capturing the spatio-temporal interactions of the Earth System variables. The project will develop a first-of-its-kind prototype system that predicts the seasonal burned area sizes for Europe, using global environmental variables and terrestrial ecosystem modelling, and simulate their impact on local fire regimes.
The project will address the following scientific questions:
- What is the spatiotemporal contribution of the different fire drivers in the seasonal fire patterns in Europe and how do those fire drivers interact?
- How much do teleconnections enable us to anticipate seasonal fire patterns with high confidence in comparison to merely climate forecasting strategies?
- Can we use modern DL architectures to learn memory/lag effects in fire regimes and the possible teleconnections?
In this context, the project aims to:
- Create and publish Global Data Cube with Analysis Ready Data (ARD) that includeGlobal predictor variables related to fire drivers; and the predictand, i.e. theaggregated burned area sizes.
- Develop a model showing the predicted burned areas in Europe for the coming fire season and explain the model’s predictions
- Analyse the SeasFire datacube and use the developed model to Identify the spatiotemporal contribution of the different fire drivers in the seasonal fire patterns in Europe and how those fire drivers interact; detect the contribution of teleconnections and if they enable us to anticipate the seasonal fire patterns with high confidence; Learn memory/lag effects in fire regimes and the possible teleconnections
