USGS Multimedia Gallery
This text will be replaced
Information presented is factual at the time of creation.
If no transcript and/or closed-caption is available, please notify us.
Science Thru Imagery
Bill Werkheiser: Welcome to USGS and welcome to our first in a series of public lectures on "Science in Action". And thank you for coming out tonight in this, kind of rainy weather or staying on if you are employees here. I'm just kind of curious. How many people here are USGS employees? Quite a few. So make sure you tell your friends and neighbors for our future series. For those of you I don't know, my name is Bill Werkheiser. I'm Eastern Region Director. And we're very proud to have John Jones here as our first speaker in the series. The idea of these series is, you know, you can imagine we're pretty proud of the work we do, and we'd like to share that work with the community at large and for folks who may not know be aware of what John does other than USGS employees.
So, John is here, he's a research geographer here in Reston, although this week he was out in West Virginia, so thanks John for coming back in on this rainy night, and you're going back out tonight I guess to West Virginia. John is an expert in satellite and aircraft imagery and those applications, and I think he's got a real exciting talk tonight, and I see some show and tell things. So, I think we're going to have a good time and enjoy the talk. So without further ado, I'll pass it on to John.
John Jones: Thank you, Bill. OK, let's swap the slide out. Well, actually, let me say first. I earned my doctorate 7 years, 3-1/2 months ago, and while I wouldn't expect anybody to call me Dr. Jones, I have to point out that still people really get into it when they do. So, I just wanted to leave this slide up for a little bit longer.
But I'm taking that one down and bringing this up. Now, the title of my talk this evening is "Knee-high to Bird's Eye: Multi-scale Remote Sensing of Vegetation Dynamics", which is rather a mouthful. If I do a really good job the next hour, this will all make sense to you. If I don't, then at a minimum, what I'm hoping I can do is give you a deeper appreciation for the information that's in remote sensing imagery and give you just a little slice of the work we do here at the Survey in using that technology, that sort of my bare minimum. So you can let me know how I do in this regard.
My teammates and I have rhyming projects in what the USGS considers the Eastern United States, which is everything east of Mississippi. All of these projects have a few things in common. They all use field-based and aircraft and satellite-based remote sensing technology for the purpose of seeing how the vegetation is changing across space and through time, why that vegetation is changing, and then what impact those changes have on either water supply or habitat. So, what we're trying to get ultimately is information that resource managers can use to improve the health of the environment or to conserve our resources.
Now, I'm not going to go through all of these in detail tonight because it's hard to do that in 45 minutes. Instead, what I want to do is focus on a couple of these projects and give you a better idea of what I'm talking about. So, I'll start with a really basic introduction to remote sensing. I'll focus on two projects in this list and then leave some time for discussion.
So, what's remote sensing? In the broadest sense, remote sensing is a collection of information about something without actually touching it. That's the most basic definition I can give you on remote sensing. So, everybody in this room right now is using remote sensing system. Light is coming out of that projector. It's hitting the screen. Various different colors of light in certain patterns. They're reflecting off of the screen, going through the lens of your eye, hitting your retina, stimulating your rods and cones, traveling through your optic nerve to your brain. Your brain’s putting those pieces of information together to formulate a picture and understand what's on the screen. Remote sensing system.
I intend to use satellite-based systems for example, where the source of light is the sun. That light is being transmitted through the atmosphere, most of it, or some of it, I should say. It's hitting targets on the ground. It's either being absorbed by those targets and then sent out as temperature or different wavelength or it's being reflected back to a remote-sensing satellite. The sensor is on it to detect the different wavelengths of light in different patterns, that's sent down to a computer here on the ground. And then, we analyze that computer information. We got to look visually at that data or we crunch through the numbers the satellite is collecting to get information.
I get grief from my graphs but you're going to get some graphs here today, OK? This, is a depiction of a variety of light energy that's possible. What we're seeing down here is tiny portions. They are everything from gamma rays to x-rays to ultraviolet light. This is why you're wearing sunscreen, right, so you don't get sunburn, into the visible wavelengths, infrared light all the way up to microwave, radar, television, radio waves, and so forth.
So, there are remote-sensing instruments that are designed to measure various types of light energy or various ranges of light energy. Notice, what we can see is just a tiny slip. So, one of the points I'm going to make to you tonight is that remote sensing extends our ability to see these different wavelengths of light and gather more information. It's curious that this area of light is a type of light that the sun puts out the most of. So, we sort of evolved to be sensitive to the maximum wavelength region from the sun.
So, the instruments I want to talk to you tonight are primarily in the visible and up into the infrared portion. Here at the Eastern Geographic Science Center, we have instruments that measure at other wavelengths. And I'll talk a little bit more about some of those but not all. I do want to make it clear; I'm not talking about all the remote sensing of the USGS. I'm focusing on a portion of the remote sensing that I do under my area of work.
OK. Sorry, this picture is about 10 years old. This is me when I had more hair, and more of the hair I had was not white. This instrument, this is a ground-based remote sensing instrument. It's like an old aircraft carrier. You keep refitting it and rearming it and probing it through time and using it over and over. This is actually the same instrument here. The internals are all the same. What's inside this box down here is still inside there, but we made, I think, some improvements on how you carry it, for example.
Now, why do I have a 10-year-old picture? Because it's really hard to get people to go out with me in places I take this instrument to take my picture. So, I'm stuck with this one from 10 years ago with somebody to stand on the side of the road and go "OK". More often, what happens is, I get dropped off by a helicopter. And on the day I took this picture, I had my instrument in a watertight cooler. I landed in a helicopter. I put the cooler on the pontoon. I stepped onto the pontoon. And I looked out and all my coolers were shin-deep in water. I stepped off of the pontoon and caught it before my armpits went under. We had landed over a nice solution hole. The helicopter pilot, he thought it was hysterical, but he realized I had to get back in the helicopter to go back to our launching pad.
So, this instrument, believe it or not, can't get wet. And when I called the folks who designed it and I said, "Hey, I'm taking this down to the airbase," they said, "What? You know this can't get wet, right?" Which I did know. I don't talk to them about this so often anymore. But I have this mantra “protect the instrument.” It's constantly in my head. So, the reference here is to the fact that they would stand there and watch me get nibbled on by things I couldn't see underneath the water. And while I'm balancing, trying to make this instrument stay dry, they go, "Protect the instrument!" These are the kind of friendships you build when you're around.
So, what kind of data does this instrument collect? In this graph, what we're seeing on this line are the wavelengths of light. So, here's the blue, here's the green, here's the red, here's the infrared, and we go further along through the infrared. On this axis, we're seeing reflectance, how much light is being reflected as a function of that type of light. Here, we're seeing data collected by this instrument for three different surfaces. This blue one is the water, and more truthfully, it's a puddle on the airport, at Griffin Airport, Southeast Atlanta. They're very nice down there in Atlanta. They let you walk out there they don't run you over when you’re collect measurements. This is some grass, like you'd see in your yard. Here's some soil. So, we can get very detailed information about how much light’s being reflected at very specific wavelengths using this instrument.
Now, the satellite can't quite get as much information. It's too expensive to put something like this up into space. But I can use this to see what sort of information the satellite will give me. And I'm going to use it to show you what sort of information the satellite can give me.
So, the satellite is going overhead and for areas on the ground, it's collecting, reflecting and assessing function of the type of light and storing it. So, I end up with an image that shows how much light is reflecting at each point on the ground. I can tell the computer to take the blue light that the satellite had sensed and display it as blue on the computer. Take the green light that the satellite had sensed and display that as green light on the computer. Take the red light the satellite had sensed and display that as red on the computer. And this is what I get.
So, this is a satellite image from the Landsat Satellite System taken from about 400 kilometers in space. And this is the Blue Ridge running here. This is Route 50 coming eastward toward the District of Columbia. This is the Shenandoah River. So, you can see what bridges are crossing the river. Let's see. Right in about here is Paris, Virginia. Middleburg would be somewhere around here. Front Royal would be somewhere down in here.
So, this looks a lot like the image. How many of you use Google Earth? This is a lot like the sort of imagery you'd see in Google Earth. You know that green is vegetation. This is probably trees because it's so green and mottled-looking. This is also what you sort of see, looking over the airplane window, [unclear] staring out the window.
But I said to you already, these satellites can measure light we can’t see. So, instead of just showing a true-color deposit, let's still keep the blue light displayed as blue so the water looks kind of blue. But let's take the green light and the green display on the computer and use it to show the red light. And now, let's get a little fancier and take the infrared light we can't see and display it as red. What happens?
Sorry, sorry, sorry, just keeping you awake. The vegetation is red. What is up with that? Well, not only is it red but it's showing all sorts of variation that we had a much harder time seeing in that true-color picture. There’s a couple of reasons for that, and some of them we'll get to in a minute.
We can see a lot more difference in the agricultural fields that are out here because of this missing piece of information. In that infrared band, there are changes in vegetation that occur as a function of the health of the vegetation and the amount of vegetation that we can't see with our eyes, but the satellite can see. So, we can see stress, for example, in vegetation with the satellite remote sensing them before we would when we're standing on the ground.
Let's go a little further. Let's go to the blue up to the red. Let's use the infrared as green. So now, what color is our vegetation going to appear? A lot of vegetation. Green, thank goodness, right? Or I'd be screaming at you again.
Now, let's get really fancy. Let's take the red and move it out here to a portion of the light spectrum where the reflectance is affected by the amount of water in the leaves or the amount of water in the surface of the soil. And this is what it looks like. So, now I can still see a lot of that variation in the vegetation. But I'm also picking up even more variation in those agricultural fields, how much moisture is in the vegetation or the soil. I can really see the river up here much more easily, and see Route 50. Here's Route 17 coming down here. So, you can see quite a difference, right? So, we've really leveraged the information that we can gain about the Earth's surface by using these other portions of the spectrum.
Now, the reason why we’re looking at this satellite image is because Sky Meadows State Park in Virginia is right in here. This is actually the farmhouse right about here, and Blue Ridge Middle School out in Loudoun County is participating in a National Park Service program called "A Trail in Every Classroom." And so, this very day they went for a hike. They go down the road here and up to the farmhouse. They walked down this gravel road and went all the way up to the Appalachian Trail. A bunch of brave teachers and 120 kids did this, sixth graders.
And I went out last Friday and talked to them about how the AT appears from space and how, where they were going to be walking, is in this greater area that affects what they were going to see. And the night before that presentation, I went to this website at home and I downloaded all of this imagery. So, since the start of January, all of the imagery that’s been collected by this satellite and another satellite and one of its predecessors from 1973 on to present around the globe is now available for no fee to anybody with an Internet connection.
In 1986, I was a graduate student at the University of Maryland, and I had to win a $5000 grant to get one satellite image for my thesis. And now, this isn't one of those "Boy, I had it so bad when I was a kid." No, my point is, we're really excited because now, any graduate student anywhere in the world with a connection to the Internet can download more data than they’d ever want to process, at no additional fee. Taxes involved, tax money and people’s effort, but you don't have to pay to download it. So, we're very, very excited about that and I wanted to call your attention there.
So far, we have been aligning our mission to interpret this imagery taking light that we can't see with our eyes, putting it into light we can see with our eyes effectively, allowing us to interpret what we're seeing on the screen. But to the computer, these are all just numbers. So, let's talk a little bit about what we can do, analyzing these numbers to gain even more information.
So, we're back to the graph similar to the one I showed you earlier. Here's a wavelength that does not go far on out in the spectrum of this particular image. Here's reflectance still. Here's our vegetation curve that's showing how the vegetation reflects. So, what type of light do thinks around here? Green. All you guys have to do is keep saying "green" and you'll be good to go. Green light right here. So, vegetation occurs as green to us because it's reflecting the most when it's healthy, in the green portion of light.
These boxes that you see here are actually the wavelengths or the types of light that the Department of Commerce satellite senses. That satellite was put up actually to map ocean temperature for climate modeling and monitoring and things of that sort. But they also, in their wisdom, put two sensors on it that collect information in the red and the infrared. And folks who work with the satellite data that I just showed you thought, "Aha! We can capitalize on this system and get information about the vegetation." The reason why it's attractive is because this particular instrument makes a measurement every day.
Now, it makes measurements over larger areas, which is good. But with each individual measurement also covers a larger area. So, there's trade-offs that we're always going through in remote sensing. How many pieces of light do we get? How small an area do we resolve so we can pick out the individual houses or maybe just a football field? How often do we collect those data? We have to balance all these things because of cost, primarily. And so, the reason why it I use all these different instruments in my projects is because I'm trying to gain the best information from each one as I pull those pieces of information together.
So, this portion of the spectrum here, of this type of light, red light, how much reflectance occurs here is dependent upon how much the vegetation is growing essentially. If you think back to your biology in school, when you talk about photosynthesis, when you talk about chlorophyll, this is a chlorophyll absorption. All I'm saying here is that if there is a lot of healthy vegetation, less and less light will be reflected back to the satellite here because the plants are using it.
This portion out here is affected by the number of layers of leaves. The more leaves there are there, the more light gets sent back to the satellite, OK? So, here with this soil, it's not absorbing it, has no chlorophyll or very little. It's not absorbing much light. It doesn't have many layers of leaves on it. It's not reflecting a lot of light. But this place has vegetation with layers of leaves.
So, if I look at the difference between this and this, that gives me some idea of how much vegetation is in that spot on the ground. And we can use the computer to take every measurement that satellite needs and calculate numbers that we can relate to the amount of vegetation that's on the ground. So, in this particular equation, if I end up with a 1, it means I've got lots of healthy vegetation. Now, if I end up with something like a 0 or below 0, it means I don't have vegetation or I have something like bare soil or water.
Here are some maps from that instrument. You can see the entire United States in each image. And the colors that are being shown here are ranging from 0 basically or water or all the way up to about 0.65. And the darker green something gets, the more vegetation is there, the healthier the vegetation is. So, if you look up in this corner, you can see that in April of 1995, the Southeast and Central Mississippi Valley are the only places where the weather has been warm enough for the vegetation to be growing any great amount.
If we go into May, however, we can see that the leaves and everything are out and just cranking their way through to Virginia, all right? Some parts of New England, not so much, not yet. We go into June, and really the whole East and up into the center of the country and Pacific Northwest, all these areas have some vegetation growing or agriculture taking place. Now, watch what happens when you start getting deeper into the summer. You see down here in the Coastal Plain in Florida? We're seeing variations in that green that are a function of, rainfall.
So, now suddenly we have some idea on how well the vegetation is doing as we get variations in temperature and in rainfall through the year. Scientists in the Survey use this sort of data to try and monitor for drought and maybe agencies that calculate how much food will be available for international aid or where there are problems as they are occurring or before they occur.
But I use these data for something really different, different type of graph. In this case, down here we're seeing time of year. So, here's January and we're moving through the year all the way to December. Here is that number between 1 and 0 for vegetation health, vigor, and growth. So, in the winter, we bounce along here, this is a dot every two weeks. When we bounce along, hovering around 0, things are too cold for the leaves to start to grow.
But once these get warm enough, boom! This number starts to rise. And in the absence of rainwater, it will go all the way up to some peak. Just keep watering it and watering it and giving it lots of sunshine. It will get to this level of production and growth that it just stays at and cranks along until the sun starts to set a little earlier in the day, right? It gets a little less light energy or it has a little less water. And then, boom, the leaves will start to change and fall off the trees until it's bare again. So, we've gone now from "Look, you can see how healthy the vegetation is from one place to another" to "Look, you can see how healthy the vegetation is through time from one place to another".
So, what would I use this for? Well, we have a project in the Shenandoah National Park. And the primary scientific question we're asking at this point in that project is can we see evidence of climate change in the Shenandoah National Park? Our objectives are to track those vegetation changes over large areas for many years. We want to explore why that change is occurring and correlate that with weather, correlate it with other factors that would affect vegetation. And then, this is where I was really interested and got involved in this project, actually, we want to relate those changes in vegetation to water flow and habitat conditions for all the streams that get their water from the Shenandoah National Park in the Blue Ridge.
So, here's the satellite image, a Landsat Satellite Image again here. Actually, this is similar to the one that I showed you earlier. We were looking right about up here, in that tiny little spot. This is Massanutten Mountain. I impress all my friends because I look at the satellite image and go, "Why, yes, the Blue Ridge is right there." And I'm just looking for Massanutten Mountain to find the Blue Ridge. "No, Blue Ridge is right there."
So, there's the [unclear] Skyline Drive. Here's a familiar graph. I said to you we can make estimates of when the leaves come out, how quickly things green up, what the maximum amount of green is this, how quickly things change color and the leaves drop off. This is called, by the way, "phenology". Phenology is the study of the timing of biological events as they relate to climate. So, when birds migrate, you study that in relation to climate. That's phenology, OK? In my case, it's when the leaves come out and so forth that I am interested in.
Now, let's just take one variable. If we total all the area under this curve, we get some idea of the overall growth and productivity of a spot on the ground in the course of the year, right? What if we total that up for every spot on the ground for every year for which we have a measurement, and then we look to see whether that is increasing or decreasing through time? And this is what we get.
So, this map, right here is Massanutten Mountain. Here's Loudoun County, Virginia and, of course, the state boundaries, the Chesapeake Bay. Everywhere you see green here, we're seeing an increase in the total amount of vegetation across the course of the year in general, an increase in trend through time. Everywhere you see these reds and yellows like down here in the Coastal Plain, we're seeing a decrease in time.
So, some scientists use this to monitor the health of rangelands, for example. We have folks in the USGS who do that. But as I said, I'm really interested in what impacts these changes have on the streams that flow out of here. So, we know vegetation needs water to grow. If it has warmer temperatures and more carbon dioxide and it's growing more and the water doesn't change, then that is really important if you are a fish in this region of the Shenandoah because it means less water is coming to you.
So, the idea then is to look for these changes and try and relate them to their causes and their impacts. Now, what could cause changes here besides changes in climate? Any guesses?
Whistles “Jeopardy” theme
You can even phrase your answer in the form of a question.
Audience 1: Is it people?
John Jones: People, yeah. How would people do that?
Audience 1: Well, cultivating the land, building houses in different places, development.
John Jones: Absolutely. And that's one of the reasons why the answer was people, who are cultivating land, building houses. Absolutely. That's why we're working in the National Park. That's the real benefit of the National Park because nobody is cutting the trees down or building houses in the National Park. But other things are defoliating those trees. Gypsy moths, you've seen evidence of that type along the Appalachian Trail or other pests like wooly adelgid or fire, for example.
So, it's entirely possible, there could have been a fire here. And we're just seeing this increase, the first few years it was recovering from the fire, and now it has really come back. So, we have to go through a lot of effort pulling in other information to try and tease out the places that are changing maybe because of the climate as opposed to some other cause. So, to help with that, we also put out a series of weather stations throughout the park. And we used a very scientific approach to decide where we put them. We put them 200 meters or more away from any sensitive features or the Appalachian Trail. We camouflaged them so they're hard to see. And they look something like this.
Now, this is not an official USGS photo. We do not recommend that you do that. I wasn't there for this particular work. And I have a feeling Sharon is standing here saying "Boy, he's talking. He's really mad at me, standing out, recording climate on his weather station." But the fact of the matter is, I love the 10-foot ladder on some of these sites and I climb up the weather stations. And the reason we have to climb up a weather station, I'll get to that, we have all sorts of things that we need to deal with that are important.
So, we have instruments on here that measure, for example, the amount of light that has made it through the canopy, the amount of light that reaches the ground and is coming back up. So, here's another remote sensing instrument. We measure things like the temperature and the humidity, wind direction. We have things planted in the ground so we know what the temperature of the soil is. It's really the soil temperature that plants really care about. When the soil gets warm enough, they start moving fluid through their roots and up into the tree and they start to grow.
So, I said there are a lot of challenges in this. One of them is animals. Actually, one of them is a particular animal. This is a bear that sat in a tree, eating some apples. That's Cindy Cunningham, sort of the backbone of keeping these systems up, was about to repair our weather station that some bear had decided to climb on, play on and eat. It was repaired for two or three days before some bear decided to climb on it, chew on it and play. So, we have a real problem trying to keep these instruments up and running in the park.
What are some other challenges? People.
Audience: LaughterSo, if you see a system like this out in the woods, please, please don't mess with it, don't use it for geocaching. Thank you.
This is actually Jasper playing with another instrument we use in the field. This is a camera with a fisheye lens. And what we actually do is point it up at the sky and take a picture when there’s no leaves on the trees. Then we go back out later to point it to the sky, and take a picture where there are leaves on the trees the same way. And we use that to determine how many leaves were there at that particular time. So, we have a measurement against the one we're making from 400 kilometers or 900 kilometers in space.
We have another camera out there called the "phenocam". So, here I am with a National Park Service collaborator, putting this instrument out in the National Park on the edge of a cliff, which is a cheap thrill. And it collects images like these every 15 minutes. And we have, in fact, put this up on the website. So, you can go up on the website and check the conditions in the park and we’ve animated it, for example, the following sequence.
So, we have a specific scientific purpose for this camera. We know where that is on the ground and where it is in our satellite images. So, as we go along calculating that value and the condition of the leaves and we're saying, "Based on our analysis, the leaves started turning brown on this date," we have something on the ground we can use to check on against. So far, so good?
So, this project is certainly a work in progress. One of the things I had pointed out is right now we have 18 years of measurement. And when you're a scientist, you never have enough data, but 18 years isn't a whole lot of data to start drawing lines through and drawing conclusions from it. So, we're constantly working on expanding that database.
One of the things we're doing is adding that 30, 40-year record from that other instrument. We're trying to use those two instruments together. But I think this is a really good example of local, with the utility of remote sensing to measure what can be seen over large areas through time.
Let me give you another one. This is South Florida, Florida Everglades. So, here's Lake Okeechobee. And historically, water has flowed from Lake Okeechobee down to the Everglades and out into the bay. Up here, you're seeing all the sugarcane and other agricultural areas that have been put in place up here. Wherever you see a line here, it's a canal or a levee that’s used to convey water from here on out into the ocean so that either there's enough water to drink here or this area doesn't get flooded when there's something like a hurricane.
So, all of these levees and canals were put in place. And this, by the way, is Tamiami Trail, which runs from Miami to Tampa. And as a result of these features, that flow of water has been interrupted and the chemistry of the water has been changed. So, wherever you see these orange colors, for example, that's where nutrients are leaking out of the canal and changing the composition of the vegetation in those locations.
So, right now, a 20-year, $7-billion effort is underway to restore the health of the Florida Everglades by getting the flow of the water to have the same sort of timing and the same sort of quantity that it has had historically. And as things like sea level rise occur, send sufficient water down here so that we don't have a lot of saltwater moving back up into the interior of the state.
South Florida is really, really flat, I want you to know that. But our measurements show that the average slope here is about an inch and a half per mile. We're talking about an inch and a half per month. The only way I can give you a sense of that is to turn the camera on at an angle before I take a picture. So, the water is all going like this at a very slow rate. It's going so slowly that one of the things that really affects where the water flows is vegetation.
So, my role in this project is to help build models that show us where the water will flow by incorporating into them vegetation using remote sensing. So, I mentioned Tamiami Trail, that's here. Here's the canal that runs down. We have devices, when I say "we", there are a lot of collaborators on this project, right. And we're measuring the amount of water that flows through holes underneath the road. And we're measuring the amount of water that comes out of these little streams at the bottom end. And with those two pieces of information, we're trying to build a computer model that will move water through here very realistically, OK?
Now, my collaborators began to look at how this vegetation affects flow. So, they went out in the field and started making measurements. They also built a flume. They built a big bathtub, basically. And they grew sawgrass in it and they tilted it on a lift and they measured how quickly the water moves through different vegetation.
Well, I took a slightly different approach. I went out in the field with some friends, the ones who were still willing to go out in the field with me. And I used another remote sensing instrument like this. This is called a "ceptometer". So, this actually has a little computer inside of it and all along here are light sensors. And what I do with this instrument is I hold it up above the vegetation and I press a button after I told it what type of vegetation I want. And it actually runs a little calculation and tells me how much light has come in at that place. Then, actually it's more like 'slurp-pop, slurp-pop'. And I go, "Whoa," while Greg listens for the gators and the moccasins. And I press the button again and it sees how much light was intercepted or kept from getting to the sensor by the canopy or the vegetation, and it tells me an estimate of how much vegetation there is.
So, unlike my collaborators who have to cut everything, weigh it, or put it on a table and measure how large the blades of every grass stalk are, I make a measurement, make a measurement, and then move to the next spot and make a measurement, without actually affecting the vegetation.
So, I collect data like that and I look at that vegetation index from imagery. So, I have an air photo that I can calculate that number from and I compare it against my measurements of how much vegetation is there, and I develop some relationships that I then take from aerial photographs on up to satellites.
The other thing I do is I get on airboats and I travel around with my computer in a box, with my satellite image on the computer and with GPS, Global Positioning System, connected to that laptop. So, as I move around in that airboat, I can literally see the image and change where I am on the image that has been collected from 400 kilometers in space.
Now, anybody here old enough to remember the cartoon show "Ed, Edd and Eddy"? This guy's name is Ed. This guy's name is Eddy. And I told them if I ever show up in the morning and there's another Ed in this boat, I am not getting on it.
John Jones: So, one day we're out there and Horatio Frank, another collaborator, says to us, "Boy, it would be really cool if we can put an instrument in a spot like this." And we know we need big areas of the same sort of vegetation if we're going to get a proper measurement. We don't want to be on the edge of one type of vegetation and another, making measurements, because we don't know where our measurements are really coming from.
So, I looked in this box and I said, "You know, if we go about 10 kilometers in this direction, we're going to go from really dense saw grass into a huge patch of this sort of brush of sparse saw grass and leaves." And I stuck my head back in the box. We started flying on the airboat. I watched this go along. I saw us go from one area into another, put my hand up like this. The boat came to a stop. I looked at everybody and they all stood there with their mouths just agape. We were in this huge patch of sparse brush and sawgrass under us. Always the scienties, Horatio, says, “Do that again.”
So, we spent an entire day going around and showing that, "Yeah, what I'm seeing on the screen here, what I'm calculating to be the density and the type of vegetation is really what's there." And what we produced is a map like this. I'm not so concerned, if you look at the classes. The point to get from this map is we have groups of vegetation that are grouped together by their density and their shape so that we can alter our model and the flow of water within our model as a function of those classes. I always have to have an alligator picture and I think people get upset. This was a guy who was waiting by one of our sites. I was going to get out and make one of those measurements. The guy driving the air boat said, "You're not going to get out, are you?" And I said, "No need. When they bellow at you it’s a really cheap thrill.
OK, so here's our model. Remember Tamiami Trail and the canal boundary along the edge? Here's the ocean on this side. What we're seeing in colors is the calculated depth of the water. So, we start out by flooding the place and we start running a model. And the model starts using how much light is coming in, what the temperature is, what vegetation is there to start drying the system now. These arrows show the direction in which water is flowing. The length of the arrows, it would be nice to change them, their length according to how fast the water is moving, but the difference is so great across this that a bunch of them become dots. So, we can't really show the speed.
But what's going on along here? Anybody? Tides, tides. We took a model that was developed from the Chesapeake Bay, which is not an inland water, because it handles tides very well. And we adapted it to work here in the Everglades because we wanted to get that mix of saltwater along the coast in our model. Some of the things we found out, water flows uphill in the Everglades because of wind. So, if the wind is strong enough and the vegetation isn't thick enough, it will actually change the direction of the flow of the water.
The other thing we found out was before we put vegetation into this model, when they just had incoming water and outgoing water, they had to really make believe about how much water evaporated off the surface. They had put completely outrageous numbers in there for the amount of water that was taken up into the sky because, otherwise, too much came out of the bottom. So, when you put that vegetation in there and slow the water down, suddenly they can change those evaporation numbers in the model into something that we are actually measuring on the ground.
Why do we need a model it like this? Because, again, the Corps wants to do things like punch bigger holes underneath this room or raise the road. And the question is how much of that road do you need to raise to get the types of flows that we want here? If we raise it this much, how much flow will we get?
All right. [inaudible] So, as a result of this, remote sensing is allowing us to include vegetation into these everglades water flow models. We're getting much better models and much more like the real world, which means we have a lot more confidence or faith that these models are going to be showing us what might happen if we start changing things. It's really the reason why we want to build these models.
What I'm working on right now is taking those changes to vegetation through time and putting them into this model. Fire is a very big thing down in the Florida Everglades. Lightning strikes that sawgrass and they burn large areas of it. Right now, we don't account for that. So, a large area of sawgrass could be burned during a time that we're simulating and we don't account for it. So, we want to use remote sensing through time to take care of that issue.
This is my reminder just to point out to you that there are other types of instruments that we use. Some of them are active systems, meaning we don't rely on the sun for energy. We actually send the energy out from the instrument itself. One of the types of data that I'm really excited about using more and more is called LIDAR. So, we have an airplane that has lasers, shooting lasers down toward the ground as it flies along. And it's recording the amount of time it takes for that light to go out, hit a surface and come back.
So, that's how precise the clock we have and the global positioning system we have. And when we know how long it takes for the light to go from the plane to the top of the vegetation or to the ground and we know very precisely where the plane is, we can figure out how high that tree is and how high that ground is from the center of the Earth. We get data like this.
So, out here is how we used to use remote sensing to look at the elevation surface. We used to use aerial photographs and some other types of processing. This is what an image looks like from one of those LIDAR systems. You can actually see the channels of the streams running through here. You can actually see the footprints of the houses being built, all the roads. You can see the level of the ponds that are collecting storm water that runs off of these buildings. So, the experiment for me is to look at the vegetation, but the original purpose for these systems was to look at the ground. So, my research is teasing out vegetation, which everybody else wants to throw out.
Of course, we have to check these things on the ground. So, here I am with Dr. Hogan. We have what, a couple of years ago, was the state of the art for surveying instruments. It was a Total Station. And we've got some other friends and colleagues running around in the woods. And we're spending the day collecting about 300 points on the ground very accurately so we can see how good a job we're doing, finding the ground in this imagery.
Last Thursday, right in the center of this room, I set up this instrument. It collects millions of points with the same accuracy as this in about 10 minutes. There's a good and bad to that, right? Now, it only takes me 10 minutes to get millions of points. Now, I have millions of points across this. So, for me, a lot of the effort is to figure out how this particular system works and how I can use it to look at the vegetation and other features that I'm getting from that airborne imagery.
The exciting thing for me is remote sensing is constantly evolving. The challenge for me is remote sensing is constantly evolving. The point is we want to take these various systems and bring that information together so we can see how the surface of the Earth is changing through time and over space. So, there's a great deal more in remote-sensed imagery that just beautiful pictures. I hope I conveyed that to you.
Remote sensing helps us measure what can't be seen, track the health of your resources, and understand how and why that health may be changing. To do this well, I truly believe you have to be on the ground making measurements so you understand the system. You have to be on the ground making measurements, controlled measurements so that you can tease out these various satellites and airborne systems as much information as possible.
Here are some websites that I can put up later on if you are interested in looking for additional information. Dr. Slonecker and Ms. Aiello have agreed to hang out by our field spec instrument over here. They got some plants. And there's a geologist in the building who carted some rocks over here. Yeah, just show them some rocks.
So, there are a couple of rocks over there. Behind the slide projector, we have that LIDAR system set up. I don't think it's activated and running. So, if you're really interested in that, you can look into doing that. I'm more than willing to entertain some questions for a while and hang around for a while. So, I have to bring up the final slide that shows you our schedule for remaining talks and then I will also at the same time entertain questions.
Bill Werkheiser: Thank you, John. Any questions for John? I think you can all see why he was chosen as our first speaker. This is a fascinating, fascinating talk. OK, so here's the quiz. John mentioned phenology earlier. Who remembers what that was?
John Jones: Gadzooks help me out here.
Bill Werkheiser: The good news is, we're all going to hear all about it at our next talk and why it's important, why we should care about it, and what you can do to help phenology. And that talk, what's the date of it?
John Jones: It's always...
Bill Werkheiser: May 6th.
John Jones: Yeah. It’s always the first Wednesday of every month at 7:00 p.m. I know Jake, Jake runs the National Phenology Network, and actually, the Shenandoah National Park project. It’s part of that network. We are the mid-Atlantic sub-region for that network. We're working on a workshop. We're putting a workshop together this summer, where scientists and citizen volunteers make measurements in the field along with Blue Ridge. I'm hoping we'll do that in the park. So, we'll try and get some announcements out. So, I can tell you Jake is an excellent speaker. I'm recommending him from there so somebody can tell you what phenology is and remember it.
You could do me a favor if you're just being shy. He says, "Anybody knows what phenology is?" And I'll raise your hand and tell him. Other questions or just come on up.
Audience 5: I remember on your NDDI, there was a point where the, I think, soil, there was, I think, a peak in where it was refracting some of the light in that peak was there the grass in there?
John Jones: Absolutely.
Audience 5: You don't have any --?
John Jones: Yes, that is an excellent question. So, there are places where reflectance of two different materials will be exactly the same. So, what we're trying to distinguish to tell what is soil and what is vegetation. We can't make a measurement there, OK? So, that's part of the power of these remote sensing systems, when you're making measurements and all these different points, we know "Aha!” If we want to tell the difference between soil and vegetation, we don’t want to use this part, this kind of light, because they can look a lot the same. If we really want to see the water, we want to go here and here because the water looks really different than the vegetation and the soil.
So, that's part of that art and the science as well. And so, we're looking for those combinations of the light we look at, and the target we're after. One of the things I often tell people when they come to talk to me, "Can I use remote sensing to do this and that?" What I relate to them is, "Can you separate the target, that you're interested in, from the bed around you?" If you can't do that, remote sensing won't work for you. So, you have to find the technology that allows you to do that. So, that's an excellent question. It wasn't planted.
Anything else? Well, you're welcome to come on up afterward and visit with me, visit with Dr. Slonecker and visit with Ms. Aiello. And I thank you all for coming.
Title: USGS Public Lecture Series: Science Through Imagery
Knee-high to Bird's Eye: Multi-scale Remote Sensing of Vegetation Dynamics. Dr. John Jones, an expert in remote sensing, discusses several projects in the Shenandoah National Park and the Everglades. Learn how science from satellites can help decision-makers address issues related to climate change, water resources, and habitat conditions.
Transcript in text-only format below
Location: Reston, VA, USA
Date Taken: 4/1/2009
Note: This video has been released into the public domain by the U.S. Geological Survey for use in its entirety. Some videos may contain pieces of copyrighted material. If you wish to use a portion of the video for any purpose, other than for resharing/reposting the video in its entirety, please contact the Video Producer/Videographer listed with this video. Please refer to the USGS Copyright section for how to credit this video.
Suggest an update to the information/tags?
* DOI and USGS link and privacy policies apply.