mirror of
https://github.com/wassname/spaceapps2017_matches.git
synced 2026-06-27 17:17:25 +08:00
53ba6f9a1688fa98c4acfdd5f8a1866518086fd8
Running a bushfire risk simulation using satellite images.
Risk of starting->monte carlo risk of spreading
TODO:
- uncertainty
- elevation and slope
- cumulative risk
data sources:
- auscover http://www.auscover.org.au/browse-all-data/
- satellite:
- sentinel 2
- landsat 8
- GLCF: Landsat Tree Cover Continuous Fields https://explorer.earthengine.google.com/#detail/GLCF%2FGLS_TCC
- Normalised Difference Vegetation Index providing a measure of vegetation density and condition. It is influenced by the fractional cover of the ground by vegetation, the vegetation density and the vegetation greenness. It indicates the photosynthetic capacity of the land surface cover.
- EVI Enhanced vegetation index
- slope
model: start at random pixel spread based on vegetation in nearby cells also die out in each cell remember risk of fire from all simulations
steps
- download tifs of
- l8
- EVI or other veg index
- elevation or slope
- load into js
search
- A stochastic model for assessing bush fire attack on the buildings in bush fire prone areas http://mssanz.org.au/modsim09/A4/tan_z.pdf
- no listed predictive ability
- estimate of loss http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.542.453&rep=rep1&type=pdf
Description
Languages
Jupyter Notebook
91.7%
HTML
5.7%
Python
2.5%
JavaScript
0.1%