Wildfires are increasingly exacerbated as a result of climate change, necessitating advanced proactive measures for effective mitigation. It is important to forecast wildfires weeks and months in advance to plan forest fuel management, resource procurement and allocation. To achieve such accurate long-term forecasts at a global scale, it is crucial to employ models that account for the Earth system’s inherent spatio-temporal interactions, such as memory effects and teleconnections. We propose a teleconnection-driven vision transformer (TeleViT), capable of treating the Earth as one interconnected system, integrating fine-grained local-scale inputs with global-scale inputs, such as climate indices and coarse-grained global variables. Through comprehensive experimentation, we demonstrate the superiority of TeleViT in accurately predicting global burned area patterns for various forecasting windows, up to four months in advance. The gain is especially pronounced in larger forecasting windows, demonstrating the improved ability of deep learning models that exploit teleconnections to capture Earth system dynamics.
Dr. Dimitrios Michail (HUA) is a Professor at the Department of Informatics and Telematics at the Harokopio University of Athens. He holds a Diploma in Electronics and Computer Engineering from the Technical University of Crete, and an MSc (Computer Science) and PhD (Algorithms) from the Max-Planck Institute for Informatics, all with distinction. He has also conducted post-doctoral research at the Max-Planck Institute for Informatics in Germany as well as at the INRIA research institute in Sophia-Antipolis, France. His main research is focused on the efficient implementation of computer algorithms in modern models of computation. Topics that he often touches are graph algorithms, graph mining, machine learning and computer vision. He has published
numerous articles in well-known international journals related to algorithms such as Algorithmica and ACM Transactions on Algorithms. He has also published works related to remote sensing in journals like ISPRS Journal of Photogrammetry and Remote Sensing and IEEE Geoscience and Remote Sensing Letters. Dr. Michail has participated as a researcher in a number of R&D projects at European and national level related to management of Big Data in the Cloud (e.g. FP7 TELEIOS, FP7 Fortissimo, H2020 AfarCloud) and Machine Learning (e.g. H2020 Teaching, H2020 DeepCube).