earthchange

Multipurpose satellite environmental monitoring and change detection, in pure Python. 24 scenarios from free public archives: fire danger (Canadian FWI on BMKG thresholds), smoke exposure in person-days, air-parcel trajectories (HYSPLIT), drought, land-surface heat, floods, deforestation, mining, urbanisation and surface-water change — via Google Earth Engine or Microsoft Planetary Computer (no account needed). Export georeferenced GeoTIFFs, quick-look PNGs, statistics, print-ready A4 maps, citable Markdown records, and a one-file brief that assembles a whole assessment. Every output carries its own limits.

📖 Full hands-on tutorial (English & Bahasa Indonesia): https://firmanhadi21.github.io/rs-change-detection/
💻 Source, examples & case study: https://github.com/firmanhadi21/rs-change-detection

Install

Heavy dependencies are optional extras, so the install stays lean:

pip install 'earthchange[gee]'       # Google Earth Engine backend (free account)
pip install 'earthchange[mpc,maps]'  # Planetary Computer + maps (no account)
pip install 'earthchange[all]'       # everything

Quick start

# Deforestation around a coordinate, with a finished map
earthchange -s deforestation --lat -3.333 --lon 122.25 --radius 6 --map

# Flood extent from Sentinel-1 SAR — no Earth Engine account needed
earthchange -s flood --lat 27.2 --lon 68.3 \
    --pre 2022-07-01:2022-07-25 --post 2022-08-20:2022-09-10 --backend mpc

# Urban growth timing across 2010/2015/2020 (Landsat 5/8/9)
earthchange -s urban-trend --lat -6.30 --lon 107.15 --map

# List everything
earthchange --list

Each run writes a self-contained output/<run-id>/ folder containing the PNG, GeoTIFF, statistics JSON, metadata, and any maps.

Scenarios

Scenario	Method	Sensor
`deforestation`	NDVI loss	Sentinel-2
`urbanization`	Built-up gain — `NDBI` (default), `UI`, `BU`, `IBI`, or thermal `NDISI`/`EBBI` via `--method`	Sentinel-2 / Landsat
`water`	NDWI change	Sentinel-2
`burn`	dNBR severity	Sentinel-2
`mining`	SIRAD radar temporal + NDVI loss	Sentinel-1 + S2
`flood`	SAR water extent (event vs baseline)	Sentinel-1
`urban-trend`	NDBI at 3 epochs → RGB growth-timing map	Landsat 5/8/9
`coastline`	Sea boundary + shoreline change (erosion/accretion) + retreat rate m/yr	S1 / S2 / Landsat
`transit-access`	% population with access to public transport (SDG 11.2.1)	WorldPop + OSM
`island-heat`	SST + LST + wet-bulb (humid heat) trends for small islands	OISST / Landsat / ERA5
`urban-heat`	Urban heat island (SUHII) + hot-spot map + decadal trend	GHSL + Landsat + MODIS
`forest-history`	Multi-epoch deforestation: year-of-loss map + forest-area trajectory	S2 / Landsat NDVI
`population-change`	Two-epoch population change: Miloš-style spike poster + map + optional true GPU 3D via forge3d	GHSL GHS_POP
`haze`	Smoke & air quality during fires: PM2.5 (ISPU), aerosol index, hotspots	CAMS + Sentinel-5P + FIRMS
`fire-history`	Multi-year fire record: burned area per year (peat vs mineral), recurrence map, fire season	MODIS MCD64A1 + FIRMS

Two backends

`--backend`	Data source	Account?
`gee` (default)	Google Earth Engine	free account + `earthengine authenticate`
`mpc`	Microsoft Planetary Computer (STAC)	none — streams COGs, processes locally

Optical scenarios build cloud-masked median composites; radar scenarios auto-select the Sentinel-1 orbit with coverage. The AOI is a square centred on your coordinate. Landsat 7 is skipped (SLC-off gaps).

Make maps from a finished run

earthmap output/<run-id>            # render A4 map sheets offline (no GEE)

Map sheets include an OpenStreetMap basemap, the change layer, legend, a statistics panel, a location inset, coordinate grid, scale bar and north arrow.

License

MIT © Firman Hadi. Data: Copernicus Sentinel (ESA) and Landsat (USGS/NASA), via Google Earth Engine or Microsoft Planetary Computer.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

earthchange-0.1.83.tar.gz (278.9 kB view details)

Uploaded Aug 11, 2026 Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

earthchange-0.1.83-py3-none-any.whl (283.0 kB view details)

Uploaded Aug 11, 2026 Python 3

File details

Details for the file earthchange-0.1.83.tar.gz.

File metadata

Download URL: earthchange-0.1.83.tar.gz
Upload date: Aug 11, 2026
Size: 278.9 kB
Tags: Source
Uploaded using Trusted Publishing? No
Uploaded via: twine/7.0.0 CPython/3.13.13

File hashes

Hashes for earthchange-0.1.83.tar.gz
Algorithm	Hash digest
SHA256	`371d5a3fc5fe7f78dbab0373f44670bb5db13b51c83ee19c1fe164a463fd7c42`
MD5	`c7ddd0e5b212d7734ec5a89618df057a`
BLAKE2b-256	`750048f174ab9192030fbc2ed295992682eda09f9437d58889e929a33338749b`

See more details on using hashes here.

File details

Details for the file earthchange-0.1.83-py3-none-any.whl.

File metadata

Download URL: earthchange-0.1.83-py3-none-any.whl
Upload date: Aug 11, 2026
Size: 283.0 kB
Tags: Python 3
Uploaded using Trusted Publishing? No
Uploaded via: twine/7.0.0 CPython/3.13.13

File hashes

Hashes for earthchange-0.1.83-py3-none-any.whl
Algorithm	Hash digest
SHA256	`110c4884e8c6135ce3085e1aa97114667e7e72baf8a36722fb9b12e6635b31b7`
MD5	`d5158acca0860cb32c79b4bdae858f79`
BLAKE2b-256	`79859bc7d08820651938395e3771767e1033ccaab1b76a6bf31c1792c0c92274`