Free satellite imagery and remote sensing data sources
For most projects the fastest path is the Copernicus Data Space Ecosystem for Sentinel radar and optical imagery, USGS EarthExplorer for Landsat, and NASA Earthdata Search for MODIS — all free, all requiring only a basic account. For repeated analysis over large areas, Google Earth Engine and Microsoft Planetary Computer put the same archives next to cloud compute so you are not downloading raw scenes yourself. Below is a source-by-source breakdown of what each platform holds, what it costs, and what to check before you build a pipeline on top of it.
The short answer
Start with the agency archive that owns the sensor you need: Copernicus Data Space Ecosystem for Sentinel-1 (radar) and Sentinel-2 (optical), USGS EarthExplorer for the full Landsat archive back to 1972, and NASA Earthdata Search for MODIS. If you need to process many scenes without downloading them individually, use Google Earth Engine or Microsoft Planetary Computer, which host the same data alongside cloud compute. All of the sources below are free; the only recurring cost is your own storage and processing time.
Sources, one by one
1. Copernicus Data Space Ecosystem — Sentinel-1 and Sentinel-2
The Copernicus Data Space Ecosystem is the European Space Agency's current home for all Sentinel mission data, having replaced the old Copernicus Open Access Hub in January 2023. It covers Sentinel-1 C-band synthetic aperture radar (all-weather, day/night) and Sentinel-2 multispectral optical imagery at 10-60m resolution with a 5-day revisit. Data is published under the Copernicus open data policy, which places no restriction on commercial or non-commercial use, only a request to acknowledge the source. Browsing and direct downloads run through the built-in Copernicus Browser; a free registered account is required to download. Formats are SAFE (native) and cloud-optimized GeoTIFF via the STAC API.
2. USGS EarthExplorer — Landsat, and more
EarthExplorer, run by the USGS EROS Center, is the standard portal for the full Landsat archive — every mission from Landsat 1 (1972) to the current Landsat 8/9 — plus aerial photography and SRTM elevation data. Landsat imagery is public domain and free with no usage restriction. You draw an area of interest on the map, filter by date range and cloud cover, and download scenes directly; a free USGS EROS account is required to download (not just to browse). Note that as of August 2024, NASA's own MODIS and other LP DAAC land products were removed from EarthExplorer's catalog and now live only in NASA Earthdata Search (below).
3. NASA Earthdata Search — MODIS and other NASA missions
Earthdata Search is NASA's unified discovery and download tool for its Earth-observing missions, including the two MODIS instruments (Terra and Aqua, daily global coverage at 250m-1km resolution since 1999/2002). A free NASA Earthdata Login is required. MODIS land products are distributed through the Land Processes DAAC, atmosphere products through LAADS, and snow/ice products through NSIDC — Earthdata Search indexes all of them in one place, and near-real-time products are also available within roughly 1-2 hours of acquisition through NASA's LANCE service.
4. Alaska Satellite Facility — Sentinel-1 SAR distribution and tools
The Alaska Satellite Facility (ASF) Vertex portal is NASA's dedicated distributed archive for synthetic aperture radar data, including a complete mirror of Sentinel-1 alongside SAR from other missions. It is a useful alternative to the Copernicus Data Space Ecosystem when you specifically need SAR products, on-demand InSAR processing (via the HyP3 service), or the asf_search Python module for programmatic bulk retrieval. Access requires a free NASA Earthdata Login, the same credential used for Earthdata Search.
5. Google Earth Engine — analysis-ready, planetary-scale catalog
Google Earth Engine hosts a multi-petabyte public data catalog — Landsat, Sentinel, MODIS and dozens of other collections — colocated with cloud compute so you write analysis code (JavaScript or Python) against the archive instead of downloading scenes first. Non-commercial and research use is free; commercial use requires a paid Earth Engine plan through Google Cloud. It is the right choice once your workflow needs to touch many scenes across a large area or time range rather than a handful of individual images.
6. Microsoft Planetary Computer — open-standard alternative
Microsoft Planetary Computer occupies similar ground to Earth Engine — petabytes of environmental data plus cloud compute — but is built on the open STAC catalog standard and standard open-source Python tools rather than a proprietary API, which makes code more portable across platforms. Its catalog includes Sentinel-1/2, Landsat, MODIS, NAIP aerial imagery, ESA WorldCover land cover, SRTM elevation and more. The data catalog itself is free to query; compute through the hosted Hub has usage-based options.
7. AWS Open Data Registry — S3-hosted mirrors
Both Landsat and Sentinel-2 are also mirrored on AWS as part of the AWS Open Data Registry, stored as cloud-optimized GeoTIFF in public S3 buckets. This is useful when your pipeline already runs on AWS and you want to avoid downloading through a portal at all — you can read directly from S3. There is no charge for the storage or the data itself; you pay only for compute and any cross-region data transfer on your own AWS account.
Sources side by side
| Source | Best for | Access requirement | Typical format |
|---|---|---|---|
| Copernicus Data Space Ecosystem | Sentinel-1 radar, Sentinel-2 optical | Free account, no cost | SAFE, cloud-optimized GeoTIFF |
| USGS EarthExplorer | Full Landsat archive (1972-present) | Free USGS EROS account | GeoTIFF |
| NASA Earthdata Search | MODIS and other NASA missions | Free Earthdata Login | HDF, NetCDF, GeoTIFF |
| ASF Vertex | Sentinel-1 SAR, InSAR processing | Free Earthdata Login | SAFE, GeoTIFF |
| Google Earth Engine | Large-area, multi-scene analysis | Free (research); paid for commercial | Analysis-ready via API |
| Microsoft Planetary Computer | Open-standard STAC access, portability | Free catalog; usage-based compute | Cloud-optimized GeoTIFF via STAC |
| AWS Open Data Registry | Direct S3 access inside AWS pipelines | Free data; pay for your own compute | Cloud-optimized GeoTIFF |
How to choose
Match the source to your task, not the other way around. If you need a handful of scenes over a known area and date range — say, before-and-after imagery for a specific event — go straight to the Copernicus Browser or EarthExplorer and download directly. If you need a time series or a mosaic across a large region, that same download-and-stitch approach becomes painful fast, and you are better off writing your analysis against Google Earth Engine or Microsoft Planetary Computer so the platform does the tiling and reprojection for you. If you specifically need radar (SAR) — for example to see through cloud cover — go to ASF Vertex rather than a general optical archive. And if your infrastructure already lives on AWS, the Open Data Registry mirrors save a data-transfer hop.
Whichever source you use, check three things before building anything on top of it: the exact license terms for your use case (open data policies vary in whether attribution is required), the sensor's revisit frequency and resolution against what your analysis actually needs, and whether the coverage for your area and date range is complete — cloud cover, orbit gaps and sensor outages are common enough that you should verify before committing to a workflow.
Frequently asked questions
Is Sentinel and Landsat data really free for commercial use?
Yes. Sentinel data is published under the Copernicus open data policy with no restriction on commercial or non-commercial use, and USGS Landsat data is in the public domain. Cloud-hosted copies on Earth Engine, Planetary Computer or AWS carry no additional license beyond the source agency's terms, though each cloud provider's own terms of service still apply to how you use their compute.
What happened to the Copernicus Open Access Hub?
It was retired. Since January 2023 all free Sentinel data and browsing tools moved to the Copernicus Data Space Ecosystem, which also added the browser-based Copernicus Browser viewer. Old bookmarks or scripts pointing at the previous scihub domain no longer work.
Which source should I start with if I just need a few scenes?
For a handful of scenes over a known area and date range, USGS EarthExplorer or the Copernicus Browser are the fastest path — both let you draw an area of interest, filter by cloud cover and date, and download directly after a free account signup. For repeated or large-area work, move to Google Earth Engine or Microsoft Planetary Computer so you are not downloading and reprocessing raw scenes yourself.
Can't find the exact coverage you need? Leave it to us
If you are not sure which archive actually has usable coverage for your area, date range and cloud-cover tolerance, you can give us the task and the conditions it must meet, and we run a data availability assessment (3 free credits on sign-up) first — searching reputable data platforms for real and judging matches and gaps against each required item. Even when no perfectly fitting dataset is found, we present the search directions, approximate sources and item-by-item findings honestly for your reference.
New accounts get 3 free searches; every result comes with verifiable source links.