Free Satellite Data Sources for Mineral Mapping in 2026

Commercial satellite data subscriptions for mineral exploration cost real money. Some run into six figures per year. Most exploration teams do not need them. The major space agencies give away data that does the same job for free, and in some cases does it better.

This is the practical list. Six sources, what they measure, where to download them, and which deposit types they suit. Real URLs, no marketing.

Sentinel-2

ESA's Sentinel-2 constellation is what most people start with. Two satellites, 13 spectral bands from visible to shortwave infrared, 10 to 60 metre resolution depending on band, and a 5-day revisit globally. The L2A product is atmospherically corrected and ready to use.

What you can detect: iron oxides (goethite, hematite) using B04/B02, clay minerals (kaolinite, montmorillonite, sericite) using B11/B12, ferrous iron using B12/B8A, vegetation stress using B11/B08. What you cannot do: separate individual clay species or detect carbonates. The SWIR resolution is too coarse for that. If you need fine mineral discrimination, you need ASTER or hyperspectral data.

Download from dataspace.copernicus.eu. Free registration. Programmatic API available.

ASTER

ASTER is a Japanese instrument on NASA's Terra satellite, launched in 1999 and still running. It has 14 bands: 3 in VNIR (15-30 m), 6 in SWIR (30 m), and 5 in thermal infrared (90 m). The SWIR bands are where ASTER earns its place in any exploration workflow.

ASTER Band 9 is the one that matters most. It picks up the K-OH absorption in alunite and jarosite, both of which are pathfinders for high-sulphidation epithermal gold and porphyry copper. Sentinel-2 has no band in that wavelength range. If you are looking for those deposit types, you need ASTER.

One caveat: new ASTER acquisitions have been limited since 2024 due to instrument issues. The archive going back to 2000 is still available. For most exploration work that historic data is fine — the geology has not moved.

Download from NASA Earthdata. Registration required.

Landsat-9

Landsat-9 launched in 2021. The Landsat program has been running continuously since 1972, which gives you a 50-plus year archive for change detection if you need it. For mineral mapping, Landsat-9's thermal infrared bands (TIRS-2, 100 m resolution) are the main reason to use it. They detect emissivity variations linked to quartz and silicification.

Silicification is a useful gold pathfinder, particularly for Carlin-style and epithermal deposits. For SWIR mineral mapping you would still go to Sentinel-2 or ASTER first. Landsat-9 fills a specific gap in arid regions where silica caps mark the surface expression of mineralised systems.

16-day revisit. Download from USGS EarthExplorer. Also available via Google Earth Engine if you want to skip the download step entirely.

NASA EMIT

EMIT is mounted on the International Space Station, operating since July 2022. 285 spectral bands across 380-2,500 nm at 60 m resolution. It was built for mapping mineral dust source regions, but it is now the best free hyperspectral data available for exploration.

Most satellite mineral detection stops around 2.2 µm because that is where Sentinel-2's SWIR bands end. EMIT goes to 2.5 µm and samples continuously across the whole range. For two commodities this changes everything.

Lithium pegmatites contain spodumene, which has diagnostic absorption at 2,206 nm and 2,308 nm. Sentinel-2 misses both. ASTER catches the 2,206 nm feature with Band 5 but lacks the spectral resolution to be confident. EMIT resolves both clearly. If you are doing greenfield lithium exploration, EMIT is the only free option that actually works.

Rare earth elements are the second case. Several REE-bearing minerals have absorption features in the 500-1,300 nm range. EMIT captures these. Sentinel-2 cannot.

The catch is coverage. EMIT operates from the ISS orbit at 51.6 degrees inclination, so it covers about 60°N to 60°S. No high-latitude data. Acquisition prioritises arid and semi-arid regions where dust source mapping is the primary mission. If your target area is in tropical Africa or the Andes, coverage is good. If you are working in northern Canada or Scandinavia, EMIT is not for you.

Download from NASA Earthdata.

EnMAP

EnMAP is a German hyperspectral satellite launched in April 2022. 228 bands across 420-2,450 nm at 30 m resolution. It samples spectra at 10 nm intervals, which is fine enough to separate minerals that look identical on Sentinel-2.

Where Sentinel-2 might lump kaolinite and montmorillonite together as "clay alteration," EnMAP can tell them apart. That matters because the two minerals form under different conditions and indicate different parts of the alteration system. Distinguishing them moves you from "there is alteration here" to "this is the phyllic zone of a porphyry system."

EnMAP coverage is not global. The German PI tasks the instrument for specific areas, and you can request acquisitions for your area of interest. Download from enmap.org. Registration required.

Copernicus DEM

Not spectral, but you need it. The Copernicus DEM comes from the German TanDEM-X radar mission and provides global elevation at 30 m resolution.

For exploration the DEM gives you structural context. Faults and fractures control fluid flow, and fluid flow controls where deposits form. Run lineament detection on the slope and aspect derivatives, overlay the result on your spectral anomaly map, and the targets that sit on structural intersections are the ones to drill first.

Download from Microsoft Planetary Computer or OpenTopography.

Known deposit databases

These are not satellite data, but you cannot do satellite exploration without them. Every spectral anomaly needs to be cross-checked against known mineral occurrences. The USGS Mineral Resources Data System has over 300,000 records with commodity, deposit type, location, and production history. Download as CSV or query through the API at mrdata.usgs.gov/mrds.

The S&P Capital IQ mining database is the commercial gold standard but it costs serious money. For free alternatives, the SNL Metals & Mining database has limited public access, and the British Geological Survey publishes some regional inventories. None match the global coverage of MRDS.

Which satellite for which commodity

Where free data falls short

Free data has limits. Commercial satellites like WorldView-3 (1.24 m) and GeoEye (0.46 m) give you 10 to 50 times the spatial resolution. For deposit-scale targeting near an existing mine, that resolution gap matters. For greenfield regional screening at the concession scale, it does not.

Commercial providers also let you task the satellite. You ask, you get an image of your area on a specific date. With free data you wait for the satellite to come over you, and you hope it is not cloudy. In tropical regions that can mean weeks of unusable acquisitions before you get a clean scene.

Hyperspectral coverage is the biggest gap. EMIT and EnMAP are the only free options, and EMIT's coverage is biased toward arid latitudes. If you need on-demand hyperspectral over a specific tropical target, you are paying.

For everything else — regional screening, target generation, alteration mapping over hundreds of square kilometres — the free data stack is enough. The savings against a commercial subscription can fund the rest of the program.

Where to download everything

Skip the processing

Downloading the data is the easy part. Atmospheric correction, cloud masking, band ratio computation, SAM classification, anomaly extraction, MRDS cross-referencing — that is where a few weekends disappear. ProspectIQ runs the whole chain. Pick your area, pick your commodity, get results in about a minute.