Landsat on AWS: Half a Year, Half a Billion Requests

A few weeks ago, we had the chance to attend the world’s largest gathering of earth scientists at the American Geophysical Union’s Fall Earth Sciences Conference.

More and more, research in the areas of climate change, agricultural resilience and space exploration rely on access to computing resources in the cloud. Because the cloud makes it easy to share massive amounts of data and allows them to only pay for the computing resources they need, they can accelerate their pace of research while reducing costs.

Last year at the AGU Conference, we announced Landsat on Amazon Web Services (AWS), a service to make Landsat data available for anyone to access from Amazon Simple Storage Service (Amazon S3). Today, over 250,000 Landsat 8 scenes are freely available from Landsat on AWS. All Landsat 8 scenes from 2015 are available, along with a selection of cloud-free scenes from 2013 and 2014. All new Landsat 8 scenes are made available each day (~680 per day), often within hours of production.

Available near on-demand IT resources

Landsat on AWS is designed to allow fast access to Landsat data via a RESTful interface, reducing the time required for analysis. The data shared via Amazon S3 can be transferred programmatically and quickly to AWS cloud computing resources, and researchers can analyze data without needing to download it or store it themselves.

Landsat on AWS makes each band of each Landsat scene available as a stand-alone GeoTIFF, and scene metadata are available as text and JSON files. These individual files allow efficient and targeted data access. Landsat on AWS GeoTIFFs have “internal tiling,” which allows users to use HTTP range GET requests to access 512-pixel squares within each scene. This allows highly targeted access to data based on geography.

Half a year, half a billion requests

Within the first 150 days of the launch of Landsat on AWS (19 March 2015 to 16 August 2015), Landsat imagery and metadata were requested over 500 million times, globally.

The most requested WRF PATH/ROW combination is 040/036, which includes the southern California high desert and the location of the 2015 Lake Fire. The scar of the fire is rust colored in the visualization below, which is based on data acquired on 15 July 2015. This false color composite visualization was made in minutes with Snapsat, a web application built on AWS.

AGU attendees who learned about Landsat on AWS were eager to start using it themselves, sharing it with their students, or in using Amazon S3 to share similar data sets.

Learn how to access Landsat on AWS at http://aws.amazon.com/public-data-sets/landsat/