wildfire-data-analysis

Professor Oversite

Dr. Ilkay Altintas altintas@sdsc.edu

Synopsis

UCSD Data Science Masters Capstone Project. The goal of this project is to produce surface fuel maps of San Diego County and other regions in Southern California using data from satellite imagery. Accurate and up-to-date fuel maps are critical for modeling wildfire rate of spread and potential burn areas. The best available fuel maps are from USGS LANDFIRE, but these are only released every two years (the latest is for 2014); we would like to create fuel maps at a higher temporal frequency.

alt text

Data:

Satellite imagery from Planet Labs and USGS Landsat, existing fuel maps from LANDFIRE..

Code Example

*to fill in later

Motivation

The wildfire analysis application uses GIS mapping to estimate fuel build up in burnable areas surrounding San Diego. Currently the model runs off of older static records, but future improvements will build in a more dynamic working model which includes computer vision and common machine learning tools such as neural networks and spark.

Installation

http://wifire.ucsd.edu/

API and Amazon References

Depending on the size of the project, if it is small and simple enough the reference docs can be added to the README. For medium size to larger projects it is important to at least provide a link to where the API reference docs live.

Core Contributors

FirstName LastName Email
Megan McCarty mmccart@eng.ucsd.edu
Pooja Palkar ppalkar@eng.ucsd.edu
Syed Sadat Nazrul ssnazrul@eng.ucsd.edu
Parinaz Azadvari pazadvar@eng.ucsd.edu
Ryan Riopelle rriopell@eng.ucsd.edu

Technical Advisors:

FirstName LastName Email
Daniel Crawl crawl@sdsc.edu
Mai Nguyen mhnguyen@sdsc.edu
Jessica Block j.block@eng.ucsd.edu

License

May assume licenses for open source application as project progresses: including Apaches Spark, MLLib, Hadoop, and other related modules.

Example: https://github.com/apache/spark/blob/master/LICENSE