Eyes on Earth Episode 34 – Open Training Data

STEVE YOUNG: Hello everyone. Welcome to another episode of Eyes on Earth. Our podcast focuses on our everchanging planet and on the people here at EROS and across the globe who use Remote Sensing to monitor and study the health of Earth. I'm your host Steve Young. Our guest today is Anne Hale Miglarese. The founder of Radiant Earth Foundation. Radiant Earth is a non-profit organization that connects the global development community with the remote sensing tools it needs to tackle social, economic and environmental issues. Researchers and data scientists can use Radiant Earth to access a vast catalog of satellite-based training data such as crop classifications, land cover and the like. It also offers machine learning models that help scientists use that training data and the power of cloud computing to search for trends and change across space and time. Users can also contribute their own training data and coordinate with each other to target parts of the world where more data is needed. Prior to launching Radiant Earth, Anne served as Chief of the National Oceanic and Atmospheric Administration's Coastal Services Center where she directed the Remote Sensing and GIS programs for 10 years. Anne has also worked with the South Carolina Dept of Health and Environmental Control, the Water Resources Commission and the South Carolina Department of Natural Resources. Welcome Anne. ANNE HALE MIGLARESE: Thanks for having me Steve. YOUNG: You bet. Let's start broadly and learn more about you and your lifelong involvement as a leader in advancing earth observation. How did you choose this career focus? MIGLARESE: Well, originally when I went to college, I wanted to be a Marine Biologist and that was until I had my first internship as a Marine Biologist working on the commercial fishing docks outside Charleston, SC. That was hard work. I returned to college at the University of SC the next semester and took my first Geography course and saw my first image of Landsat and from there I knew perhaps I could still do marine and coastal work but do it from satellite imaging and geospatial technologies. So that really was the spark. That first internship that said, "yes, I love the work but this may not be the right fit for me." YOUNG: Describe for our listeners the mission of the Radiant Earth Foundation. MIGLARESE: Our mission is to empower organizations and individuals globally, really. Giving them access to earth observation data, training data, standards and tools to address the world's most challenging problems. We do that really thru three programs. One is and where the majority of our activity has been today is in Radiant ML Hub where one can go and post or discover training data sets for machine learning on earth observation data. The second is to cultivate a community of practice to develop standards for machine learning on earth observation and to expand interoperability's of these tools and data sets. And finally third, raising awareness in the global development sector specifically and with data scientists on the innovation that machine learning and earth observation can bring. YOUNG: So how did you arrive at the Radiant Earth focus on earth observations and machine learning as the pillars of your foundation's emphasis? MIGLARESE: When I originally founded Radiant over 4 years ago now, my hypothesis was that particularly the global development community was having a very difficult time getting their hands on imagery, open imagery and then getting access to the tool set needed to analyze that imagery. Our original mission was to build a cloud based repository of open satellite imagery for the globe and associate that with open source software tools to make it freely accessible to anyone on the earth. About the time we actually built that platform the Radiant Earth platform, eleven commercial companies had actually done the same thing and many of them with a premium access layer and many of them much more robust. What it told me was that our hypotheses was right but it was a commercial marketplace. That is not a good place to spend philanthropic dollars and I think what we have found is as a neutral entity, we are in a place we can help academics, we can help governments where we can help the commercial sector by being this library, if you will this first kind of Guggenheim Library. Specifically for training data of earth observation and I think it's a very nice niche for a neutral entity. We are not a government; we are not a private sector company and it allows us to work with and broker many different types of organizations hopefully to drive innovation. YOUNG: So you are saying that the access to training data became an important part of your emphasis just because of how well it was received? MIGLARESE: Well, no. I mean we are at the cusp of great innovation. The intersection of cloud computing and machine learning and the plethora of data that we have now, we have to use machine learning and AI to analyze all of this data. They very first step in that though, is good high quality training data. It's the bottle neck. It's what's holding this greater innovation. We need a repository so that training data is high quality training data. It's not just used once and discarded but can be shared with others around the globe so that over time we dramatically reduce that bottle neck. YOUNG: I know in a recent article on the topic of training data, you wrote that there is and I am quoting here "a lack of awareness within the funding community about the ways in which training data can buy down future costs and speed the pace of innovation. Instead the geospatial community must leverage lessons learned from the drive for open geospatial data and bring best practices forward by applying the same techniques and policies to open training data." What are some of those lessons and can you incentivize this process? Miglares: This is much of what we have seen for twenty, thirty years and that is when research is funded often times, the data are not openly shared even if that was in the grant language. NASA, USGS, ESA, many of these organizations are very familiar with this issue and are really highlighting how important it is to share. What we are seeing are significant investments by major philanthropies in machine learning on earth observation. They deal with and require data management plans and the sharing of the data is, I think, fundamental to grow the community. Particularly in the global development community. I think the carrots are recognition of best practices. If you are a principal investigator and you share a great data set in your next study, hopefully you will be able to find a data set that allows you to use it in your next research project. Public recognition is also important and hopefully we will be able to find some really wonderful use cases and stress how important that is. YOUNG: So could Radiant Earth have undertaken this whole initiative, earth observations, machine learning, training data sharing. Could they have done this ten years ago? If not, why is it possible now? MIGLARESE: Dramatic acceptance of cloud computing. I would say that's the major initiative that has made this possible. Certainly, distributed computing, cloud storage and machine learning techniques. Now we have a wealth of data. Whether it is Landsat or Sentinel or Planet, or Maxar. The world is wealthy with wonderful earth observation data. Now we need techniques to speed the analysis. YOUNG: So if organizations or individuals want to contribute their training data to the Radiant Earth effort what do they need to do? MIGLARESE: It's really very straight forward. We can host it on an AWS bucket if you would like to keep it in your environment. That is fine you will find instructions at Radiant Earth Foundation Radiant ML Hub on our website. You can sign on, get an account it's all free. If you have a particularly large data set we are happy to consult with organizations to help them register their data. YOUNG: Can you paint a picture if you will of that point in time in which machine learning and earth observations are harnessed and available to the global community for addressing sustainable development goals? MIGLARESE: Oh yes. Absolutely. I think there are a lot of wonderful initiatives underway and the intersection of a tremendous amount of earth observation data with cloud computing with machine learning with a global data science community are going to really drive us to solutions. One that I am particularly interested in and that Radiant is very focused on is in agriculture. Agriculture in the global south primarily. In helping nations and helping regions and helping farmers improve their farming practices by applying machine learning to the earth observation data and giving insights to those regions and those farmers on how to improve their yield. I think we are going to see dramatic improvement in solution development in support of the sustainable development goals in a whole host of arenas. YOUNG: Talk briefly about Landsat. What does Landsat bring to the Radiant Earth table that will make organizations goals a reality? MIGLARESE: I am very excited to see the launch of Landsat 9 and the plans for Landsat Next and the level of collaboration between NASA and USGS and the Copernicus program of the European Union and as I understand it some level of continuity and collaboration there to really maximize the quality of the data and the orbits and the timing. Radiant will invest in building training data sets based on Landsat data and work with the community to bring additional Landsat training data sets to Radiant ML Hub to support researchers around the globe. YOUNG: We've been talking to Anne Hale Miglarese, the founding CEO of Radiant Earth Foundation. Radiant Earth is an organization that connects the global development community with the machine learning resources needed to make decision making that addresses social, economic and environmental issues. Thanks for joining us, Anne. MIGLARESE: Thank you, Steve. It was a pleasure. YOUNG: We hope you come back for the next episode of Eyes on Earth. This podcast is a product of the U.S. Geological Survey, Department of Interior.