P18 : Wildfire Detection and Prediction system using a Sequential Transformer model for improved accuracy


Students Jasmine Wang
School HDSB - Iroquois Ridge High School - Oakville
Level Intermediate 9/10 - Grade 10
Group Group 8 - Engineering and Computing II
Abstract The purpose of this project is to improve wildfire detection and prediction by developing a transformer-based sequential model that leverages long-term environmental trends. By addressing the limitations of single-day data analysis, the model enhances accuracy in wildfire prediction and detection. Additionally, I created a scalable, open-source data pipeline using Google Earth Engine, enabling future researchers accesss to cost-effective wildfire datasets.
Awards
Group Award Prize
Merit AwardsGold Merit Award$ 100
Canada-Wide Science Fair Trip AwardsCanada-Wide Science Fair Trip Award
Canadian Meteorological and Oceanographic Society AwardsCanadian Meteorological and Oceanographic Society (CMOS) Award - Second$ 50
one year Free membership to CMOS