Free Satellite Imagery: Where to Download Sentinel and Landsat Data
- Primary Signal: Free Satellite Imagery: Where to Download Sentinel and Landsat Data
- Overview: Sentinel-2 and Landsat offer free, high-quality satellite imagery for the whole planet. This guide explains where to download it, what each dataset covers, and how to use it in QGI...
- Verification: Analyzed and compiled by FreeSky GeoData editorial monitoring desk.
Satellite imagery used to be expensive, restricted and hard to obtain. Today, the two most important Earth observation programmes β Europe's Sentinel and the United States' Landsat β give their data away for free, covering the entire planet in high resolution and updated every few days. Anyone with an internet connection can download imagery that would have cost thousands of dollars a decade ago, and the only barrier is knowing where to look and how to use it.
Sentinel-2: Europe's free workhorse
Sentinel-2 is the Copernicus programme's multispectral imaging mission, and it is the most widely used free satellite dataset in the world. It revisits every point on Earth every five days, captures 13 spectral bands from visible to shortwave infrared, and offers 10-metre resolution in its visible and near-infrared bands. That combination β frequent, multispectral, free β makes it ideal for agriculture, forestry, urban mapping and environmental monitoring. The data is available through the Copernicus Data Space Ecosystem, which offers a browser, an API and a cloud-based processing environment, all free for registered users.
Landsat: the longest record on Earth
Landsat is the longest-running Earth observation programme, with continuous imagery since 1972. Landsat 8 and 9, the current satellites, offer 30-metre multispectral imagery with a 16-day revisit time, and the archive is the only dataset that lets you compare a landscape across five decades. That depth of history is Landsat's unique value: no other free dataset can show you how a forest, a city or a coastline has changed since the 1980s. The data is available free through the USGS EarthExplorer portal and through cloud platforms like Google Earth Engine and Microsoft's Planetary Computer.
How to download imagery for your area
The simplest route is a browser-based portal: the Copernicus Browser for Sentinel data, or EarthExplorer for Landsat, where you draw your area of interest, pick a date range and a maximum cloud cover, and download the scenes you need. For larger projects, use an API: the Copernicus Data Space API and the USGS API let you search and download programmatically, which is essential when you need dozens or hundreds of scenes. For analysis without downloading, use Google Earth Engine, which hosts the full archives of both programmes and lets you process them in the cloud, in your browser, for free.
What to do with the imagery once you have it
Raw satellite imagery is a starting point, not a finished product. In QGIS, you can open the downloaded bands directly, create a true-colour composite, and calculate indices like NDVI β the vegetation index β from the red and near-infrared bands. That single calculation turns raw pixels into a map of plant health, which is useful for agriculture, forestry and environmental work. More advanced users can classify land cover, detect change between dates, and build time series that show how an area evolves. The imagery is free, the tools are free, and the only cost is the time to learn them.
The licences and the limits
Both Sentinel and Landsat data are free for any use, including commercial use, with only a requirement to credit the source. That is an extraordinary deal, and it exists because both programmes are publicly funded and their data is treated as a public good. The limits are practical, not legal: the data is large, processing it takes computing power, and cloud cover can ruin a scene. But those are manageable problems, and the cloud platforms solve most of them. Free satellite imagery is one of the best deals in geospatial work β and in 2026, it is better and easier to use than it has ever been.