Skip to content

Explainable ML to understand wildfire burn severity

Notifications You must be signed in to change notification settings

minhokim93/burnseverityML

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

32 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Mapping Wildfire Burn Severity Using Machine Learning 🔥

Collaborators: Minho Kim, Adrienne Dodd, Xihan Yao, John Radke, Matt Kondolf

Background

With the exacerbation of climate change effects, communities in the Wildland Urban Interface (WUI) are at risk of being devastated by higher severity and more frequent wildfires over time. However, WUI landscapes are heterogeneous and mixed with highly complex features. In response, there is an urgent need to understand the wildfire-causing factors towards the development of effective and sustainable wildfire mitigation. However, there is a disconnect between our understanding of landscape patterns and wildfire burns, especially at a high spatial resolution needed to resolve the highly heterogeneous and complex WUI landscapes. We aim to develop a data-driven, machine learning approach to assess wildfire burn severity with high resolution remote sensing data and generate an understanding of burn severity with different landscape-related variables.

Requirements

  • numpy
  • scikit-learn
  • matplotlib
  • pandas
  • rasterio
  • seaborn
  • jupyter

Usage

  1. Install a new conda environment
$ conda env create --name burnseverityML --file environment.yml
  1. Activate the new environment and navigate to the "src" folder
$ conda activate burnseverityML
  1. Start notebook
$ jupyter notebook

Citation

Please cite the journal paper if this code is useful and helpful for your research.

About

Explainable ML to understand wildfire burn severity

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published