An interactive web map of active fire detections across Indonesia for the whole of 2023, built on NASA FIRMS satellite thermal-anomaly data. Pick a date and the map redraws the hotspots detected that day, with the underlying detection records listed beneath it.
Indonesia's peatland and forest fires are a recurring transboundary haze and carbon-emission problem. This is a lightweight, dependency-free way to look at where and when the satellites saw fire — the kind of exploratory view that comes before any modelling.
assets/data/indonesia2023.csv (4 MB) and assets/data/indonesia_fires_2023.geojson (8.5 MB) hold a
full year of detections in the standard NASA FIRMS active-fire schema:
| Field | Meaning |
|---|---|
latitude, longitude |
Centre of the ~1 km detection pixel |
brightness, bright_t31 |
Brightness temperature (K), channel 21/22 and channel 31 |
frp |
Fire Radiative Power (MW) — a proxy for fire intensity |
acq_date, acq_time |
Acquisition date and UTC time of the overpass |
satellite, instrument |
e.g. Aqua / MODIS |
confidence |
Detection confidence, 0–100 |
scan, track |
Pixel footprint size — grows toward the edge of the swath |
daynight |
D daytime or N night-time overpass |
A caveat worth stating. These are thermal anomalies, not confirmed forest fires. Volcanoes, gas flares and industrial heat sources also trigger detections, cloud cover hides fires entirely, and a low
confidencevalue means what it says. Filter onconfidenceand cross-check before drawing any conclusion from a hotspot count.
- Leaflet map over OpenStreetMap tiles, centred on Indonesia
- Date filter across the full 2023 range — the map and table update together
- Detection table showing latitude, longitude, brightness, date, FRP, satellite and instrument
- No build step and no API key — plain HTML, CSS and JavaScript
The GeoJSON is fetched with fetch(), so it needs to be served over HTTP rather than opened as a
file:// path:
git clone https://github.com/apkirana/project_forestfire.git
cd project_forestfire
python3 -m http.server 8000Then open http://localhost:8000. The first load pulls the 8.5 MB GeoJSON, so give it a moment.
index.html - the map application
js/main.js - Leaflet setup, date filtering, table rendering
css/styles.css - layout and map styling
assets/data/indonesia_fires_2023.geojson - full-year detections as GeoJSON features
assets/data/indonesia2023.csv - the same detections in FIRMS CSV form
all_data.html, simple_table.html - standalone table views of the dataset
apps.ipynb - notebook used to prepare and inspect the data
Active fire data from NASA FIRMS (Fire Information for Resource Management System), MODIS collection aboard Aqua/Terra. NASA FIRMS data is in the public domain; please credit NASA FIRMS if you reuse it. Basemap tiles © OpenStreetMap contributors.
- project_polusikalimantan — PM2.5 across Kalimantan, the region these fires most affect
- project_eucairpollution — agentic multi-agent air-quality forecasting
My published work on peatland hotspots takes the same data further: Hotspot pattern distribution in peatland area in Sumatera based on spatio-temporal clustering (Procedia Environmental Sciences, 2016) and Poisson clustering process on hotspot in peatland area using Kulldorff's Scan Statistics (TELKOMNIKA, 2015).
MIT — see LICENSE. The FIRMS data retains its original NASA terms.
Annisa Puspa Kirana — PhD researcher, Faculty of Geo-Information Science and Earth Observation (ITC), University of Twente. Research on agentic AI and LLM-driven workflows for Earth observation.
Google Scholar · ORCID · LinkedIn · GitHub