Introducing Aether: a new data platform for painlessly accessing satellite imagery.
- you can deploy operational applications derived from satellite and geospatial datasets.
- your analytics teams can prototype models with combinations of machine learning and remote sensing knowhow.
- you do not need a post-graduate GIS degree to understand what to do.
Maybe it’s me. I think the industry believes that because the launch of a satellite is a heroic technical feat, accessing satellite data needs to be one too.
Aether sets out to refocus the business of satellite analytics on the knowhow of making data important and connecting to business where it matters.
The Aether platform is a system of applications and utilities for developers to rapidly and easily build algorithms that use satellite and geospatial data. The Aether platform is accessible by APIs and python, but operates in the cloud using deferred graphs. This will allow developers to build and execute applications with processing abstracted away and minimal data transfer to mobile and user devices. This design choice leads users of the Aether platform into an ecosystem of geospatial data and cloud algorithms deployed as portable and lightweight mobile or web applications, accessed by URLs.
Introduction to Satellite Imagery
Developers today can download from two freely available satellite Earth observation imagers, and one foundational agricultural data layer:
- the NASA LandSat program, providing wide spectrum imagery of the entire Earth biweekly at 30m pixel resolution stretching back to 1982,
- the ESA Sentinel-2 program, providing wide spectrum imagery plus special chemical imagery optimized for agriculture of the entire Earth at 10m pixel resolution from early 2016 onward, and
- the USDA Cropland Data Layer, a 30m per pixel map of the US categorizing the agricultural land use decisions annually. Each color represents one specific crop, such as cotton (red), alfalfa (pink), or urban land use (gray). Color tables are included and available here as well.
The long historical archive of LandSat imagery is particularly valuable because remote sensing specialists have been developing applications and analytics for decades, such as this agricultural use case by the USGS.
Using repeated observations of Iowan crop fields, analyzing the imagery to generate fertilizer recommendations to farmers, and applying the fertilizer according to those schedules would generate $858M corn and soybean per year, lower fertilizer costs from farmers, and give cleaner drinking water.
Aether Lightning Quickstart
Our Aether Lightning Quickstart aims to have you downloading satellite data to your python developer environment for data science and running a hosted web app in under fifteen minutes: http://docs.runsonaether.com
The article examples were generated using the Aether platform. Similar query parameters search for other data resources, non-visible spectrum imagery, and locations.
We hope this article will leave you with a clear and concrete understanding of what the Aether platform provides developers and users, and how we work to bring data to users and businesses alike.
Post a comment to this article to discuss remote sensing applications and email us directly to discuss how to get your business connected with Aether.