Say Hello to Radiant Earth

Almost exactly six years ago, Geospatial World published a profile on the then-brand-new Radiant.Earth. It’s a great document – full of insights into the minds who recognized the need for our organization:

In defining Radiant.Earth, the team realized three key things: first, open data is not good enough – it must be discoverable, accessible and useful to a very diverse group of users, some sophisticated, some new users. Second, there are similar needs across the global development community and an entity such as Radiant.Earth can help amplify those needs. Third, there is so much change going on in the remote sensing sector right now that such an entity is needed to help guide and connect users worldwide to earth imagery, geospatial data, and analytical resources.

These three things remain just as relevant today, and we’re still hard at work to address them. As we continue this work, we’re making some changes to our brand that will help us expand access to Earth science data.

Source Cooperative is the next generation of Radiant MLHub

Today, just as when we started, open data is not good enough. In December 2019, we launched Radiant MLHub as the world’s first cloud-based open library dedicated to Earth observation training data for use with machine learning algorithms. Today, we’re announcing Source Cooperative, the next generation of Radiant MLHub. Source Cooperative is in beta development, and you can sign up to help us test it by filling out the form at

Our collaborators have told us many times that they have benefitted from having a neutral and trustworthy place like Radiant MLHub to publish data. Because Radiant MLHub isn’t linked to any particular government agency, technology company, or university, it has enabled research across a diverse group of collaborators that include governments, NGOs, academic institutions, and commercial organizations all over the world. Source Cooperative builds upon Radiant MLHub’s legacy as a neutral and trustworthy data publishing platform and will enable the publication of a wider variety of datasets in addition to machine learning training data products and machine learning models. For anyone who has any kind of data or machine learning models that they need to share, Source Cooperative will allow them to upload it, define how open they want it to be, and even charge for it if they want to. By supporting a wide array of data sharing use cases, Source Cooperative will allow us to help solve more issues that get in the way of equitable access to data.

Radiant MLHub is still operational and will remain so until the end of October 2023. All content currently in Radiant MLHub will be migrated to Source Cooperative before we remove access to Radiant MLHub.

Meet the Cloud-Native Geospatial Foundation

The Cloud-Native Geospatial Foundation is a new initiative that we will use moving forward to help guide, connect, and expand our global community of users. Specifically, it will help people adopt patterns and best practices for efficiently sharing Earth science data on the Internet using a cloud-native approach.

The pace of innovation in Earth science data has only accelerated since we got started, and it shows no sign of slowing down. When introducing Radiant.Earth in 2017, Geospatial World reported that there were “around 1,400 satellites orbiting Earth and this number could more than double over the next five years as satellites become smaller, lighter, cheaper and easier to launch.” Last year, they reported there were 4,852 active satellites orbiting the planet as of December 2021 – rather than doubling over five years, the number of satellites more than tripled over four years. When we started, we focused on Earth imagery from drones and satellites. Today, we work with users who need to work with data types beyond imagery, such as vectors, n-dimensional arrays, and point clouds.

A common technology that has enabled much of the progress of the past years is public cloud object storage (e.g., Amazon S3, Azure Blob Storage, and Google Cloud Storage). Competition within the public cloud sector has reliably led to continual reduction in object storage costs, continual improvement in performance, and a commodification of cloud storage services. We believe that public cloud object storage will continue to be the best available data sharing technology for a long time to come, and “cloud-native” refers to methods of data sharing that take full advantage of modern object storage services. Our overarching goals with the Cloud-Native Geospatial Foundation are to make geospatial data easier to work with, more interoperable, and accessible to more people worldwide.

We are already working on new educational materials and other content for the Cloud-Native Geospatial Foundation. If you’d like to be notified about updates on our progress, please sign up at We also invite you to complete our first cloud-native geospatial community survey to help us better understand your needs and preferences. Your feedback will help us develop content that is relevant and useful.

Reintroducing Radiant Earth

Radiant Earth logo

First, we were Radiant.Earth. Then we became the Radiant Earth Foundation. Today, we’re simply Radiant Earth, and we’re still working to ensure that Earth science data is used in service of underrepresented people around the world.

In November, we published a blog post about democratization that included this quote from Geoff Mulgan:

At a global level, there is a striking lack of institutions well-fitted to the big tasks of our times, from carbon reduction to cybersecurity, data to post-conflict reconstruction.

I repeat Mulgan’s observation here because it illuminates why Radiant Earth came to exist in the first place: it was to fill a gap that our founders had identified in the global development community. Filling that institutional void is the ongoing work of Radiant Earth, and I believe it’s our most urgent task. We simply will not address the big tasks of our times without creating new global institutions, and we aspire to be an exemplar of what a new global institution can look like.

We have worked with commercial companies, governments, academic institutions, non-governmental organizations, non-profits, and foundations. All of these groups are brought together by their need for data about our planet. Source Cooperative and the Cloud-Native Geospatial Foundation are the initiatives we’re currently using to help our community collaborate and become something greater than the sum of their parts. As we find new ways to help our global community collaborate, Radiant Earth will create new initiatives in our ongoing effort to increase shared understanding of our beautiful world.