Eyes on Earth Episode 52 – Tracking ‘Gray Ghosts’ with Landsat
The invasive species hemlock woolly adelgid is a threat to eastern hemlocks, filling some southeastern U.S. forests with what are called “gray ghosts” of the trees. Until recently, the insect had stayed in the southern part New York, but late last summer, an infestation was discovered in the Adirondacks. On this episode of Eyes on Earth, a New York university researcher talks about his effort to help detect the insect, which is the size of a poppy seed, by using remote sensing such as Landsat.
JANE LAWSON: Hello, everyone, and welcome to another episode of Eyes on Earth. Our podcast focuses on our ever-changing planet and on the people at EROS and across the globe who use remote sensing to monitor the health of Earth. My name is Jane Lawson, and I'll be your host for today's episode. We'll be talking about an example of university research that applies remote sensing to help a specific problem in the field. The invasive species hemlock woolly adelgid has been killing Eastern hemlocks in the southeastern United States for decades, filling forests with what some call the "grey ghosts" of the trees. New York foresters braced for it to come, and the insect eventually did creep into that state's borders, as well. Until very recently, it had stayed in the southern part of the state, but late last summer, an infestation of 200 acres was discovered in the Adirondacks, along the shores of the treasured Lake George. Detection and treatment plans have been devised to try to contain the insect and the damage it can cause. Our guest today is here to talk about his current work that uses remote sensing, including Landsat data, to try to help detect an insect the size of a poppy seed in rugged terrain. Dr. Andrew Reinmann is an assistant professor of ecology and environmental science at the City University of New York Advanced Science Research Center and Hunter College. He was enlisted by a nonprofit organization in Lake George to help with the woolly adelgid project. Dr. Reinmann, welcome to Eyes on Earth.
ANDREW REINMANN: Thank you for having me. It's great to be here.
LAWSON: First of all, let's talk a little about your path to remote sensing. Your academic background is in field ecology, especially forests, right?
REINMANN: That's correct, right.
LAWSON: And much of your past research has been related to climate and carbon and soil. So how and why did you start looking to the sky to use remote sensing in your research?
REINMANN: That is a great question. And so, yes, my background is as a field ecologist, so I would typically go out to a forest and make physical measurements of different aspects of ecosystem structure and processes, and I guess for a long time earlier in my career, I was a little bit skeptical about what we could really learn about ecosystems from thousands of kilometers up in space. But when I started my post-doctoral research, I was working in a really exciting lab that brought together a variety of different research approaches that bridged the gap between the tree-level measurements I was really familiar with making and the landscape and regional datasets that were really useful in helping us to understand the broader landscape implications of the work that we were doing. I was working with a lot of remote sensing scientists and other folks that had a lot of expertise in geographic information systems. It pretty quickly became apparent to me how powerful of a tool this was in helping us to characterize really what our landscapes look like and to better understand what really the implications of our work were, and our findings from our plot-level measurements and our plot-level experiments, and really, like, what that meant at a much larger spatial scale. And so a lot of the work that we had been doing was on how forest systems respond to climate change and respond to things like urbanization and forest fragmentation. What, in my mind, helped me to realize how important that work was, was when we were able to bring in that remote sensing data and really, just from a much broader spatial scale, understand how prevalent things like fragmentation and urbanization were, and it allowed us to better understand where our forests are, which can really help to inform the broader implications of our field experiments.
LAWSON: How is the Hemlock woolly adelgid affecting Eastern hemlocks, and why is it such a threat?
REINMANN: The hemlock woolly adelgid, like you mentioned earlier, is this tiny, sort of poppy seed-sized insect that feeds on sap or the sugars that are being exported from the tree's needles to the rest of the tree. And so it's effectively kind of robbing the tree of that nutrition, which over time can eventually starve it and result in the tree's death, anywhere between maybe 10 and 20 years after infestation is discovered. This is really important because eastern hemlocks have been recognized as being a really important foundation species. In this case, they can provide this really unique and important habitat, and really ecosystem type, that just would not exist in the absence of this species. Throughout much of the eastern U.S., it's really the only conifer tree that tends to grow in pure stands. This landscape is mostly deciduous trees that lose their leaves outside of the growing season. But hemlocks keep their leaves year round, and there aren't very many other species that do that in this region, and certainly not very many that are capable of producing what we would call pure stands or nearly pure stands, meaning that you can find many, many hectares of forest that is dominated by these hemlock trees. And the coniferous nature, the evergreen nature of these really alters some of the microenvironmental conditions of the landscapes that they occupy. For example, during the wintertime, the canopies will intercept a lot of that snowfall, and so then typically the snowfall beneath those canopies doesn't get quite as deep as (it) does in nearby deciduous forest that don't have leaves to intercept that snowpack. This can provide refuge from deep snow for deer and many other animals, (and) that can be important in helping those organisms survive harsh winter conditions. In the summertime, the deep, dense shade that these trees produce can offer a cool microclimate for a refuge, but they can also play a really important role in helping to shade trees, regulate stream flow and regulate stream temperature. In many locations where these hemlock trees grow, they can be found along these ravines right along these streams, and there's certain organisms that occupy these streams that would likely not fare as well in the trees' absence. So one species we hear a lot about, or one group of species that we hear a lot about, is trout, which require high oxygen concentrations in the water, and oxygen concentrations in the water tend to decline as water temperature increases, so having the hemlocks there helping to keep those temperatures cooler can create important habitat for those species. Hemlock are also probably the most shade tolerant trees that we have in the region, and so they're one of the few trees that can reproduce and regenerate beneath its own shade, and so for a variety of reasons and for a lot of different organisms, hemlock forests are really important.
LAWSON: So how did you get involved in studying the hemlock woolly adelgid, and what kind of research is your group doing?
REINMANN: This is maybe sort of a funny, an example of "funny the way the world works sometimes." So in the late 2000s, I was working for a nonprofit organization as a conservation biologist, and we were doing some work in trying to understand how declining hemlock stands might impact biodiversity and different habitat types in southern New York. Fast forward 12 years or so, here I am working with hemlock woolly adelgid and hemlock decline again after, you know, a little more than a 12-year hiatus. So I guess there's been something that keeps drawing me back to this tree and this species. And certainly one thing is, when I was in college, I would go hiking a lot in a large state park a little bit north of New York City, and there were these beautiful old hemlock forests there, and like these were the places that helped to get me to fall in love with nature, and getting me to want to get outside more and more. And then over the years, I could see those and similar forests start to decline because of the hemlock woolly adelgid. Its impact on these forests has really resonated with me in a really important and personal way, so I guess as a result I keep getting drawn back to these sorts of projects. And when I started my position at CUNY a few years ago, I started studying more and more of these urban forests, and what you find out pretty quickly is that it's really difficult to understand how they're functioning and how they're being impacted without also thinking about the wide range of invasive tree species and invasive forest pests that might be adversely impacting them. We got some funding to, among other things, use satellite remote sensing as a tool to try and map out and quantify the distribution of different invasive tree species and different invasive pests and then, as a spinoff of that project, a PhD. student in my lab, Kelsey Parker, got some funding to do some work in the Catskill Mountains about two hours north of New York City looking at the distribution of hemlock decline associated with the hemlock woolly adelgid, specifically using freely available remote sensing products, mostly Landsat. Through that project and some presentations related to it that, I got connected to these groups in the Adirondacks. While they didn't have any stands that were in any sort serious state of decline yet, they very much wanted to stay on top of it, so they were pretty interested in the work that we were doing in the Catskills and were wondering if we could work with them. And I should mention that a hallmark of a lot of this invasive species and remote sensing work that we've been doing is to really kind of give it a foundation in freely available products. We're certainly not the first ones to use satellite remote sensing to try and detect the distribution of invasive species. Here, what we want to do is figure out an approach that could transcend that invisible divide between the academic community and the land manager, conservation biology and practitioner communities so that we create a product that was then easy for them to use, and hopefully do a better job finding locations of the hemlock woolly adelgid.
LAWSON: So you are working with a lot of different groups, or the hemlock woolly adelgid effort involves a lot of different groups: New York State Department of Environmental Conservation, the New York State Hemlock Initiative, the Nature Conservancy, and then the Fund for Lake George, which is the group that brought you in. What are the advantages of being able to tackle a problem like this with such a wide range of groups and of people with different expertise?
REINMANN: There's a lot of virtues to this. First, there's an increased network of resources to draw from, certainly financially, but also resources in terms of expertise. And when you're trying to find invasive species, you also need a network of people who know where these things already are: state agencies, The Nature Conservancy, the Fund for Lake George, and the other scientists that they partner with really allows us to take a fairly comprehensive and holistic approach to the broader work that we're trying to do. The other piece that I think is so important about this work is that it's all too easy for academics to kind of stay in their academic circles and do this research that we think is important, but I think ultimately a lot of us want our work to be useful to society and useful for solving problems. When we have opportunities to work closely with the people that are actually using our science to tackle a problem, it provides really invaluable insight into what the needs of those communities are, how they're using the data and tools and information that we're producing, and it also, like, can be humbling for us when we realize that, you know, we don't know as much as we think that we know. And these people that are kind of like on the ground, on the front lines of dealing with some of these issues, can impart a whole lot of information that can not only help us to advance our science, but help us to help them do the great work that they do. This goes in two directions. These sorts of partnerships also allow us to inform them of really what the capacity is, of the state of the art of the science is, to address some of the problems they confront, right? Just making them aware of the fact that, "Hey, we have these satellites that can do things like detect declining forest health, and that can be a really important tool for finding invasive species." Typically these communities, they rely on field surveys, which are probably the most sensitive way to detecting these sorts of invasive species, but when we're thinking about, in this case, an area the size of the Adirondacks, that's really a pretty vast wilderness, certainly by eastern U.S. standards. There's just certain areas that aren't feasible to send field crews to. What's more is that it's not cheap to send field crews everywhere. These sorts of partnerships allow us to leverage the tools that the scientific community has with the amazing intellectual resource and on-the-ground resource that these land managers, practitioners, other sorts of stakeholders have to increase the effectiveness of invasive species detection and also increase the efficiency in which these resources get used. We started working with their field crews to help them figure out the types of measurements they can make when they're visiting their field plots that would help us to make better use of the satellite imagery. The satellite imagery is really only as good as the field validation that we have to pair it with, and it can be expensive to send field crews out to validate satellite imagery, or just to collect baseline data about the structure of an ecosystem or the health of an ecosystem to pair with the satellite imagery. And so if you have field crews that are going out to these locations anyway, if we can train them to make measurements that can help us make better use of the satellite imagery, that just winds up being beneficial to everybody involved. That's what we're starting to do. It's already starting to show clear benefits of doing that, so we've been able to give them a map of places they might want to visit because we're seeing signs of declining forest. When they go to those locations, they can see physically what the signs of decline are. That data can then be given back to us for us to see if we can further refine some of the remote sensing algorithms we're using to not only detect change but then to parse out what some of the drivers of change might be. We can maybe better detect or categorize a declining hemlock stand that is declining because of a tree that blew over versus insect damage and things along those lines.
LAWSON: Lots of mutual benefits.
REINMANN: Indeed, yeah. LAWSON: So you weren't trained as an expert in remote sensing, you said, but you've come to appreciate the benefits of working with satellite data. How does it help you most? What are some of the biggest benefits of something like Landsat?
REINMANN: I'm constantly discovering new benefits of Landsat, almost every day. But there's now over 30 years of satellite imagery from Landsat at a 30-meter spatial resolution, so our capacity to go back in time and understand changes in land cover, changes in distribution and configuration of forests is really an amazing asset to have. A lot of the work that my group does is trying to understand how land cover change and forest fragmentation is altering carbon sequestration at large spatial scales, and having this long record of, essentially, change on our landscape, we can go back in time and really quantify what those changes are, where they're happening, what the reasons for those changes might be and better understand what our field data are telling us and what they mean at landscape scales. We can also use these data to understand where there are forests that are stressed in a way that greatly magnifies what we could ever do using field-based approaches alone. And then there's always new satellites and new products coming online. We've started using space-based instruments like
ECOSTRESS, which is a thermoradiometer on the space station, to better understand spatial and temporal variations in temperature at the Earth's surface, which is really helpful in helping us to understand relationships between vegetation cover and the urban heat island effect and things along those lines. And having access now to Sentinel-2 data from the European Space Agency, collects a lot of similar data as Landsat, but with a little bit more frequent passes and a finer spatial resolution. Those two satellites together are really starting to revolutionize some of the work that we're doing. And other, newer instruments that I'm really excited about are space-based approaches to quantifying forest structure and biomass using lidar. There's all of these tools out there that wonderfully complement the field measurements that we make and allow us to say things at spatial scales that we would never be able to really say anything about in their absence. I went from skeptical to totally sold on the utility of these different approaches to doing science.
LAWSON: That's wonderful. The more data, the better.
REINMANN: That's right.
LAWSON: In this episode of Eyes on Earth, we've been talking about the use of satellite imagery to help fight an invasive species in forests. Thank you, Dr. Reinmann, for joining us.
REINMANN: Thank you for having me.
LAWSON: Check out our EROS Facebook and Twitter pages to watch for our newest episodes. You can also subscribe to us on Apple Podcasts. This podcast is a product of the U.S. Geological Survey, Department of the Interior.