Summary of Deep Learning Surrogate Models Of Jules-inferno For Wildfire Prediction on a Global Scale, by Sibo Cheng and Hector Chassagnon and Matthew Kasoar and Yike Guo and Rossella Arcucci
Deep learning surrogate models of JULES-INFERNO for wildfire prediction on a global scale
by Sibo Cheng, Hector Chassagnon, Matthew Kasoar, Yike Guo, Rossella Arcucci
First submitted to arxiv on: 30 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed work develops two data-driven models based on Deep Learning techniques to surrogate the JULES-INFERNO global vegetation and fire model, addressing the computational bottleneck in wildfire risk forecasting. The machine learning models take inputs such as global temperature, vegetation density, soil moisture, and previous forecasts to predict subsequent global area burnt iteratively. Evaluation metrics include Average Error per Pixel (AEP) and Structural Similarity Index Measure (SSIM), demonstrating strong performance with AEP under 0.3% and SSIM over 98% compared to JULES-INFERNO outputs. The proposed models achieve computational efficiency, taking less than 20 seconds for 30 years of prediction on a laptop CPU. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper develops new machine learning models to help predict wildfires faster and more accurately. These models use information like temperature, vegetation density, and past forecasts to make predictions about where and how much area will be affected by fires in the future. The models are tested and found to be very accurate, with a small margin of error compared to existing methods. This work is important because it could help firefighters and emergency responders prepare for wildfires more effectively. |
Keywords
» Artificial intelligence » Deep learning » Machine learning » Temperature
