Eyes on Earth Episode 95 – EROS 50th: Landsat Science Team

JANE LAWSON:

Hello, everyone, and welcome to another episode of Eyes on Earth, a podcast produced at the USGS EROS Center, which celebrates its 50th anniversary this year. Our podcast focuses on our ever-changing planet and on the people at EROS and around the globe who use remote sensing to monitor the health of Earth. My name is Jane Lawson, and I'll be hosting today's episode, where we're talking about the Landsat Science Team's role in influencing the Landsat Program.

The first incarnation of the Landsat Science Team, led by NASA, started with 14 members agreeing in 1996 to a five-year term to encourage research addressing Landsat continuity and the upcoming Landsat 7's Enhanced Thematic Mapper-plus instrument. The USGS asked members of this team for help again in 2003, when Landsat 7's Scan Line Corrector failed. The next three teams, starting in 2006 and including the term that ended this year, were led jointly by the USGS and NASA and had a few more members than the first with a wide breadth of expertise, backgrounds and geographic locations. Members explored strategies for the effective use of archived Landsat data and integration of future data, and they helped identify Landsat user needs for upcoming satellite sensors, including those on Landsat 8 and 9 and the future Landsat Next. Our guests here today are here to talk about their experiences as members of Landsat Science Teams. Curtis Woodcock, a Boston University professor whose research involves forests, urbanization and validation of remote sensing products, has served terms on all four Landsat Science Teams, including as a co-leader. Mike Wulder, a forest research scientist with the Canadian Forest Service who uses remote sensing, has served on the three most recent Landsat Science Teams. And David Roy, a Michigan State University professor whose research involves the integration of satellite data for studying land change, has served on the two most recent Landsat Science Teams, including as a co-leader. Welcome, gentlemen, to Eyes on Earth. Do you want to each describe how you were roped into serving on the Landsat Science Team and what your particular contributions have focused on?

Curtis, would you like to start?

CURTIS WOODCOCK:

Sure. Thanks. I think you got the question backwards. I was really excited about being on the Landsat Science Team. Nothing about being roped into it. It was really, I thought, an incredible opportunity to try to get to use more Landsat data, to start to pursue new applications. You would've had trouble keeping me off it if it was up to me, in terms of making a choice like that.

DAVID ROY:

Yes. This is David. I wasn't roped into it either. I wanted to join the Science Team because I saw where it was going, and I was like Curtis, I wanted the opportunity to participate in that. It was quite exciting. Formally, we all had to actually write proposals, even Mike Wulder, who's international.

MIKE WULDER:

I'll echo the sentiments of Curtis and David. It's really a privilege and an honor to have been on the Science Team and just doing a bit of a retrospective of the science teams themselves - just looking back and seeing that it's, you know, 18 years ago, would have been 19 years ago the proposals were written, and, you know, almost 20 years, that's a pretty good dent into the life of the EROS Center and the Landsat Program overall. So it's been a real privilege to be involved. And at that time, there really wasn't that much information floating around about Landsat, and a lot of it was related to word of mouth if you wanted to find out what the latest and greatest things to do were. So getting involved with a group like this really advanced everyone involved and gave us an opportunity to communicate those types of things to the greater Landsat community.

WOODCOCK:

It was an opportunity to learn more about the program in a way that just wasn't generally available. And as a long time user of Landsat data, that was a big part of my interest in being involved was just understanding better exactly what went on behind the curtain in the Landsat Program.

ROY:

Yeah, absolutely. That's been one of the pleasures of being on the team is actually understanding that. And, you know, several of us on the team kind of geek out, for want of a better term, with some of the engineering, some of the launch stuff, but also just the mechanics of how Earth observation is done by space agencies. I was based at NASA's Goddard and working on MODIS, but in the way Landsat was operated and the data was archived, back when I joined the team, was really evolving quite rapidly, and it was very fascinating to see all of that come back. Back in the day, we had a metric which we actually called the Woodcock Metric, if you remember, Curtis, which was to do with getting images from outside of the United States that are being acquired by other agencies under license, getting them back into a central archive in the U.S., which, and the central Landsat archive is at EROS, that LGAC Archive, repatriation of the data from around the world was a really important thing.

WULDER:

Yeah, just to jump in there, David, on the engineering and the listening and learning part of being on the science team, just discovering how so much of it all got back to how much data storage you could have on the satellite itself, and how would you get the data down from the satellite and how that formed, really, a lot of the limitations to how much data was around and how it is distributed over the globe and really tackling how to get more storage on satellite and how to get it down faster really opened everything up. And, I think, for me personally, a lot of lights went on once you started to understand those things a little bit better. And then a lot of the opportunities we've had in the last few years have really been around overcoming, you know, those two things.

ROY:

Yeah. And often when you're in the midst of stuff, you don't realize it there in the midst of it, you just kind of go, Oh, this is cool. But a lot of the development we have, in general, in Earth observation is coming out of the heritage of AVHRR and now Landsat. Landsat is the longest terrestrial record on the planet, since 1972. So people - when I talk to students, they often think that Landsat's a given, we take it for (granted). You know, they just assume that someone's going to pay for it, and it's going to continue. And so that continuity of the record is something that the team has really been always reminding folk of and trying to advocate for that. It's so critical given what's happening in the world currently with population changes, land use change, climate change and so forth.

WULDER:

Yeah, I think Curtis will remember that from the first team around, how much writing and support we needed to do around just straight up continuity and ensuring that Landsat 8 wasn't delayed for too long, you know, already having the Scanline Corrector issue and Landsat 5, you know, continuing to go long beyond anyone should realistically have expected it to continue going. So the notion of continuity was a very big part of what the advocacy was in that first science team.

LAWSON:

What are some other key contributions you could talk about that the Landsat teams have made?

WOODCOCK:

Well, if you go back to sort of the early days of the USGS-funded science teams, the last three science teams, one of the initial focuses was on the whole accessibility of the data. You still had to pay for data at that point in time. And the science team, I think, was one of the key groups that pushed hard to try to get better access to the data. And success in that domain has completely changed the whole program. It's gone from a relatively infrequently used program because you had to buy the data, and it was expensive, and so you didn't get to use very much data, to this incredible program where people are doing global-scale studies with Landsat data. So that has to be, I think, near the top of our list of sort of accomplishments in terms of trying to improve the program. The other thing that that first science team pushed hard on, and eventually this had a significant effect, was to add a thermal band back onto Landsat 8. There's always been a thermal band on Landsats in the past, but the design of Landsat 8 had left it off because it was expensive. And that was going to undermine a lot of science, and that was really the first key topic that that science team took up was trying to push to get it to get a thermal capability added back to Landsat 8, which eventually worked and has dramatically improved the value of that mission.

ROY:

I think, from my perspective in the most recent team, we pushed very hard to make the case for collections and reprocessing of the data. Previously, Landsat data was just processed with the best available knowledge from the engineers as people ordered it. As Curtis said, you had to pay for the data, so the data wasn't all just processed and put somewhere on a spinning disk. You'd have to order it. And you were basically almost paying for the cost of the processing and the delivery. And from a science point of view, when we think about climate change, on how people have changed the landscape over the last 50 years, you want to be able to differentiate long term changes from just artifacts in the satellite processing and the data. And so by having reprocessing, we can have a consistent, what we call a collection of data. So a scientist can get a single location for 50 years, and it's fairly consistent, the data they are looking at, notwithstanding differences between sensors. So that was actually really important, and that was something that other science teams, like the MODIS science team, had already done that, but it was very expensive to do that, given the nature of the data processing system at EROS. And the science team really led the, we really led the charge for advocating for that. I remember one meeting where basically we were told it just wasn't possible, it was too expensive. But the management, at USGS EROS in particular, had the foresight to look at the commercial cloud, Amazon Web Services currently, and enable them to reprocess the data a second time very rapidly at a low cost to taxpayers.

WULDER:

So once it became available and people could start to map over larger areas and then, you know, eventually do that through time, it really put a lot of enthusiasm behind Landsat and surely encouraged, you know, the efforts towards Landsat 8.

ROY:

I would also say that the current team, one of the things that we were asked to do, tasked to do, was to provide feedback on the, what we call the science and applications observation requirements for the next Landsat, Landsat Next. And so we - the science team isn't on the engineers. We're not meant to engineer or design a system. We're not told what that system is going to be. We're not really told what the budget is, but we are asked to advocate on the behalf of the Earth observation community, which is a huge community, to advocate for what the science applications kind of observations requirements are. In other words, what temporal repeat, what spatial resolution, what wavelengths and so forth. And they can get quite complicated. And we've formally responded to requests for interest from NASA. We did that as a team collectively, and it was very gratifying because what we advocated for is pretty much the Landsat Next architecture that they're now, that's now out there and the industry is trying to build.

WOODCOCK:

But what the USGS has done is made available to the science team, in essence, the whole Landsat Program and allowed us to make comment and recommendations and suggestions as it relates to pretty much all the phases and operations of the Landsat Program. And that's what's allowed us to do things like focus on the archive and LGAC and to focus on collections and all those kinds of things, is that the whole thing was opened up. We got to see what was going on behind the curtain and make recommendations accordingly.

WULDER:

Yeah, that's right. Curtis. That programwide emphasis really allowed for that freeing people up from responding to day-to-day problems and thinking of the larger program and seeing how these things all fit together. And really what that allowed for, especially back to continuity, once again, you're not thinking of just getting one sensor up and running. It's really about the backwards compatibility. What does it mean once we get this sensor working with the others that are already in orbit? Those types of higher level considerations. So it's certainly a value. So really looking back over the past three teams, I think what we've been drawing out is how, early on, there were some pretty big and fundamental issues we were trying to address around continuity and actual launches of satellites. But over time, you know, we've gotten into a lot more fine-grained questions. You know, problems or issues arise, and then we find ways of dealing with it, whether it's cloud cover or something to do with geometry, just the nature of making the data as usable to the science community and those that want to develop products.

ROY:

I would definitely make the case that we need a global analysis ready data set from Landsat because it enables people to derive high level products. I would also say that, going back to one of the contributions of the teams, was this idea of using data from different agencies together. I think it was Tom Loveland that said that continuity has been incredibly important, but now we're looking at fusion. In other words, we've got - we want to get to a place where users can use satellite data in a sensor-agnostic way. In other words, the most obvious thing to do is take Sentinel-2 data, which is quite similar to Landsat 8 and 9, and combine them. And there are projects to do that. And I think the team has been very much involved in that.

LAWSON:

Curtis, since we're looking ahead, do you have anything you'd like to add about the next Landsat Science Team? I will point out that we don't know yet who will serve on it, but do you have any thoughts on what you think that it should cover as well?

WOODCOCK:

I think the answer that that they need to focus on data accessibility and usability in an era when the data volume is going to go through the roof will be the significant challenge associated with that science team.

WULDER:

For those that are considering applying to the next science team, if there's a takeaway from what we've been talking about, is how your own research and the types of things you work on are very important, but also try to think about how what you're doing is of benefit to others and how it fits in with the more overall program. I think that would really help strengthen any proposal that you're putting in, to be able to contextualize your work into that greater program interest.

WOODCOCK: 

One thing, I think, that remains true, if you were to ask, you know, what has the main role of the science team been over time? It's been to represent the users of the data. It's been to try to make sure that their interests are the ones that are paid attention to and the data is processed as best as possible and that sort of thing. And that will continue. That process doesn't end, and it's been kind of an honor to try to play that role for the larger Landsat user community, because it's so broad and so large and so diverse. But that, in the long run, has been the key value of the Landsat Science Team.

LAWSON:

What have you enjoyed most or find most rewarding about your terms? Do you want to start, Curtis?

WOODCOCK:

What I have found most rewarding is the success of the program as a whole. When I started back in 2006, the Landsat Program was going nowhere fast. Nobody was using it much. The data was expensive. Commercialization had almost killed the program. And it's turned around, and we now have this incredible, thriving, successful program that's serving the needs of many people in many contexts. And so being part of that has really been special, and I've thoroughly enjoyed that.

WULDER:

To build on that, Curtis, it's really been quite the ride from those earlier days in that first science team to now. The ability to just, to get all the information from the team and develop storylines to share with the entire community, I think, has been one of the most rewarding things to me is just being able to be amongst people that know these topics, ask questions, and then at some point in time to communicate this information to the team overall.

ROY:

I've also really enjoyed, as we said earlier, the opportunity to work with the USGS/NASA managers, engineers and support team. They're really stimulating, and it's interesting working with them. So going to meetings and interfacing with those managers and implementers has, from my point of view, has been very rewarding. I don't normally do that kind of thing. I think on a smaller scale, one of the things, one of the most enjoyable things I did was attend the Landsat 9 launch. It got canceled multiple times, but what was super cool is the Ladies of Landsat invited Virginia Norwood, who was really inspirational. She was the first engineer, she WAS the engineer who developed the first Landsat sensor, so the MSS sensor. Virginia was there, and she gave a talk, and it was really cool to see that and to think about the history of the Landsat Program. So on a personal note, I would say that was, for me, very, very enjoyable.

LAWSON:

David, I did want to touch on the fact that, for more than a decade, you were a professor at South Dakota State University, which is less than an hour away from EROS. And then you went to Michigan State. Do you want to share anything about your relationship with EROS through those years?

ROY:

Yeah, sure. I was a researcher at the University of Maryland and working at NASA Goddard, and I was invited. I got a phone call from Matt Hansen, who's on the current team, and he said that himself and Tom Loveland from USGS EROS were going to set up a new Center of Excellence at South Dakota State, and would I consider applying? I remember the first time I went to EROS was - it must have been about 30 years ago. It was pre-MODIS launch, so in the late '90s, and we were going there, I was going out with my boss, Chris Justice, to discuss the data archiving of the MODIS. And I remember I met Tom, and he showed me what he was doing, land cover mapping. And it was really cool. And then as a result, I got to know Tom and Matt a lot, and I already knew them and immensely enjoyed working them and respected them, quite frankly, quite, you know, greatly. But my reason for coming was not just to work with good people, but because, you know, the selling point was that South Dakota State, as you said, is just an hour, one-hour drive from EROS. And the point was EROS was the data archive, right. It's got all the Landsat data. It's also got all the terrestrial remote sensing data, pretty much, is archive at EROS. So the opportunity to work with people at EROS, and the engineers in particular, and get hold of the data, for me, was really attractive. And so I would drive down to EROS really quite frequently, especially in the first few years, to visit Tom, the EROS staff, scientists and engineers. It was and is a very friendly place, especially because, you know, our program generated a bunch of Ph.D. students. Many of them actually now work at USGS EROS.

WOODCOCK:

Well, I'd like to add that one of the most rewarding parts of being on the science team, from my point of view, was the opportunity to work closely with Tom Loveland. He was the agency lead, and I was the science team lead for a lot of years. And so we we spent a lot of time strategizing on how to improve the program and how to move things forward. And he was such a joy to work with that that's been one of the highlights of my time associated with Landsat.

LAWSON:

Does anyone have any closing thoughts about the Landsat Science Team's role with the Landsat Program?

WULDER:

With the science team, a key thing that was offered by the USGS leadership, the partnership with NASA, and then the regular meetings at the EROS Center, was really having a lot of the people there that could answer questions directly. So we could come in, articulate a vision, really, about where we want to be, not necessarily where we are. We could get a better understanding of what the pinch points were that might stop us from getting to that more idealized vision, and then working together in that way to resolve those issues, or to understand better what needed to be better understood or resolved to move that forward.

ROY:

Yeah. And there is a role for the engineering obviously, it's critical, but there's also a role for that visioning and the implications of decisions made in terms of science and applications, which is quite important. And having a science team in the room enables you to have an open discussion about those things. And very often when there's an open discussion, new issues arise. Whereas if you just ask individuals via email, you just don't see that. You don't have that conversation. The Landsat Science Team is composed of U.S. federal, academic, commercial members, but it includes representation from non-U.S. institutions. And that's really critical to ensure, like, an international perspective. You know, satellites are not just U.S.-focused, they're global systems, right? And many of our problems are global in nature, whether they are political or macroeconomic. But just, like, from the climate perspective, you know, teleconnections between deforestations in Africa and South America affect the United States directly. And so having an international membership is really important. Some of those members are, in fact, representing space agencies. Some of them are representing (unclear) large users. And some of them are members of very significant users of satellite data for natural resource management.

WOODCOCK:

The beauty of getting together is that after hearing all the information about the program and learning all the stuff that you could never hear any other place, we get a chance to sit around and discuss it and try to filter it through and turn it into a path forward. It's a fun and interesting process, and to get to do it with a bunch of really smart people, it's been a joy.

LAWSON:

Thank you, Curtis, Mike and David, for joining us for this episode of Eyes on Earth, where we talked about the Landsat Science Team's role in remote sensing. And thank you to the listeners. Check out our EROS Facebook and Twitter pages to watch for our newest episodes. You can also subscribe to us on Apple and Google Podcasts. 

VARIOUS VOICES:

This podcast, this podcast, this podcast, this podcast, this podcast is a product of the U.S. Geological Survey, Department of Interior.

WULDER:

And in general, being on a shared interview with David, I'm just so happy that I got to say anything, so.

ROY:

(Laughing) Yeah, well, you can - but the thing is, I am sure Jane is going to edit me down. That's why I kept saying, "Oh, that's no good. Remove that."

WOODCOCK:

We can only hope. 

ROY: 

Yeah, exactly. You know, actually, I can only hope, too.