Skip to content
sean p. may
← all projects
Wildfire GOFER confidence heatmap viewer

Wildfire Modeling

Updated Feb 2026

Git Repo(opens in new tab)
PythonJavaScriptREarth Enginescikit-learn

this is a research workflow for collecting wildfire geospatial data, converting it into GOFER-style JSON, and running local regression analyses on fire spread and continuation. it basically sits between raw remote-sensing data and whatever modeling experiments i want to run downstream.

earth engine download and normalization pipeline

locational spread + continuation regression workflows

lightweight local viewer for GOFER-style JSON outputs

what i built

  • CLI scripts for downloading GOES confidence stacks and aligned RTMA meteorology
  • conversion tools that normalize geospatial outputs into modeling-friendly JSON
  • regression workflows and notebooks for single-fire and multi-fire analysis

how it works

  1. 1download wildfire confidence stacks and aligned meteorology
  2. 2convert outputs into GOFER-style JSON time series
  3. 3run locational regressions and inspect outputs through notebooks or the local viewer

results

  • ✓repeatable research workflow from earth engine exports to regression outputs
  • ✓supports both single-fire analysis and broader multi-fire experiments

what's next

  • tighten environment bootstrapping and add clearer example outputs
  • expand the multiresolution data path and result visualization