Eyes on Earth Episode 68 - Tracking Mangroves by Satellite

LOLA FATOYINBO:

Mangroves are kind of seen as these superheroes of climate change, because on the one hand, they protect us from some of the impacts of climate change, like sea level rise and increasing storm surges. And on the other hand, they're also able to absorb a lot of carbon dioxide out of the atmosphere and store it in a place for a really long time.

JOHN HULT:

Hello everyone, and welcome to another episode of Eyes on Earth. We're a podcast that focuses on our ever-changing planet and on the people here at EROS across the globe who monitor and study the health of Earth. I'm your host for today, John Hult. The sturdy root systems of mangrove forests act as buffer zones along the coastlines of some of the planet's most vulnerable communities protecting lives, ecosystems, and property from the rigors of hurricanes and tsunamis. The dual stressors of climate change and manmade changes to the environment such as offshore aquaculture have damaged these critical buffer zones in recent years. Remote sensing scientists are using satellite data to understand the impact those changes will have on the communities they protect as temperatures continue to warm worldwide and extreme weather events become more frequent. Dr. Lola Fatoyinbo is one of those scientists. Lola works at NASA Goddard Space Flight Center, where she studies things like change to mangroves, or the impact of aquaculture to help us understand more about coastal ecosystems and land change. Lola's data sources include optical imagery from sensors on board Landsat satellites, as well as spaceborne light detection and ranging, or LIDAR, and synthetic aperture radar, or SAR. In 2021 Lola was part of a team that investigated the impact of Hurricane Irma on the mangroves of southwest Florida. They found that some 62% of them suffered canopy damage from storm surge. Work like this helps to form the basis for coastal management plans that can boost resilience. We're so pleased to have Lola here to talk about these topics. Lola, welcome to Eyes on Earth.

FATOYINBO:

Thank you for having me.

HULT:

Great to have you. Let's start kind of in the past here, let's start with your path to remote sensing. What drew you to satellite technology, and how did you come to focus on forest ecosystems and coastal regions?

FATOYINBO: 

In my last year of college, or maybe in my first year of graduate school. I had been doing field work as a research assistant for some of my professors and I remember going to one of my professor's offices and standing in front of it and seeing a picture of Africa. But it wasn't a picture, a map like I had seen before. It was actually a huge satellite image, and I was just in awe. This was actually a mosaic that was taken with the MODIS instrument, and I had never seen something like this before. It had so much detail. I remember looking at it and thinking, "when you have data like this, you could go and explore." It's like you're traveling somewhere, but without having to actually travel to that place.

HULT:

You were sort of bit by the bug when you saw this MODIS mosaic.

FATOYINBO:

Yes. So when I saw this mosaic, it just gave me so many ideas. I was thinking, "you know, I could be here in Virginia, studying forests or other ecosystems in west Africa where I grew up or all over the world." It just blew my mind. And I immediately thought of so many studies and possibilities. I also carried out a lot of field work. When I then finally went to graduate school, I spent a lot of time working in mangroves, actually. And working in mangroves is really, really difficult. It's really wet. It's hot. There's lots of mosquitoes. You have to climb over all of these roots and go between trees. It's a swamp. It's not necessarily a really pleasurable environment to be in, let's say. And so, after having done two months of field work in mangroves, and the year before I had done field work in Madagascar and a rainforest, I was really exhausted. And I just thought, you know, I really love this work. I really love to study forest and trees and these coastal areas, but I don't think I want to be spending all of my time doing this really heavy lifting, intensive field work. And so that's also how I got really interested in working with satellite data and remote sensing, because I just thought, you know, this just really allows me to work in remote regions without having to go all the time and spend a lot of time there.

HULT:

Just for clarity's sake ... there aren't any mosquitoes in your office there at Goddard?

FATOYINBO:

No mosquitoes right now.

HULT:

Okay, good, good. Well, that's interesting. So you kind of ... you were sort of taken with this satellite image and the idea of being able to study somewhere without being there, but you also went out into the field. So, you saw it from both sides before you dove deep into the kind of work that you're doing right now.

FATOYINBO:

That's exactly it. I thought that by having the type of work where you were actually working with satellite data, this would allow me to kind of do both. I could still go into the field, because we still need to do field work to validate the results that we have or to get training points for our data, and at the same time, it would allow me to do things at larger scales and maybe in regions that are just not accessible for various reasons.

HULT:

What sorts of remote sensing data sources do you use and why is it important to look at multiple types of data? Particularly in coastal environments?

FATOYINBO:

I work with Landsat data. I work with Landsat a lot. I also work with lidar data, either from aircraft, so we call that airborne lidar, and also working with space born lidar from the GEDI instrument, and from the ICESat-2 satellite. That type of data gives us three-dimensional information. In my case, what I'm really interested in is better understanding forest structure. That is what height trees grow to, and also what the density is of trees in a forest, because this allows us to get some information on how much carbon is stored in these ecosystems.

HULT:

For clarification, when you're talking about something like lidar, the difference here is with the Landsat imagery habit, you have a flat, two-dimensional image, and with lidar, you're essentially shooting lasers at the ground and measuring how long it takes to get back. You have that sort of height information ... is how you get the 3D information, whether it's from space or airborne. That's kind of the idea, is that right?

FATOYINBO:

Exactly. And there's different ways of getting height information. We could also get it from radar data or synthetic aperture radar, or we can use another technology where we're using stereo imagery. And so really, I work with whichever data is available that will give me the answer to my question. When I first started, there was a lot less data available. And now what's really amazing is that we have this whole suite of different data sets that are available. And really what's so great about it is that we can look at our area of interest, in my case, the coastal zone, in multiple dimensions. Landsat is really important because it gives us that time component. We can go back in time and see what an area let's say a coastal zone or a mangrove forest has been like over time. We can see what it looked like in the 90s and the early 2000s and how that's changed over time to now. With a three-dimensional data from lidar, for example, we get that other dimension, where we can add some type of height or structural characteristic to it and get information on what the structure is of the area. Combining all of these data sets together is really what is key right now. And this is something that by having so many different data sets, in addition to having that long time series of Landsat, that's allowing us to do.

HULT:

I have kind of two questions here. Number one, how important is it to have that collection strategy where all the pixels align back to the Landsat archive? That wasn't always the case, and it is the case now. How important is that to being able to pull these data sources together? And bigger picture, would it even be possible to do the kind of work that you're doing without that historical archive?

FATOYINBO:

For your first question: for us, this is really something that's crucial. We did a study, for example, last year, where we were looking at the main drivers of change in mangrove forests on a global scale. And we would not have been able to do that if it wasn't for Landsat data and the standardized, well-calibrated data sets that come out of Landsat. It's really important to be able to do this time series analysis. Because if you're trying to, let's say, change your behavior, if you're trying to stop the continued loss of mangrove forest or any type of ecosystem, you really have to understand what was causing it. What is causing it now? What was causing it in the past? How has it changed? Where has it changed? Because if you don't know really what the main causes or drivers are, it makes it more difficult to plan to make any changes in the future. 

HULT:

Can we step back and even get bigger picture and talk about the importance of mangroves in general. Why is it so important to understand the dynamics of mangroves? What purpose do they serve? What are the threats they face and what are those drivers of change?

FATOYINBO:

Mangroves are these salt tolerant trees that grow in coastal regions only in tropical and subtropical areas. So, in the U.S., you can find mangroves from Florida and in Louisiana, and then there's some small patches in California. And they are essentially the ecosystem that is equivalent to like a salt marsh in the Northern zones. These are coastal forests, and they're so important for multiple reasons, one of them being that they provide habitat for both aquatic species and terrestrial species. They're really important fishery nurseries. But they also provide this coastal protection. They protect our coast from erosion, from storms, surge, and wind. And they also protect from sea level rise because they essentially serve as buffers for all that water that comes in. Historically mangroves have been really impacted. We've lost about 50% of the original cover in the past, let's say 60 years. And this is because they grow in the coastal zone, which is really prime real estate for humans to live, for hotels. It's also an area where you would put aquaculture. Over time we've lost a lot of our mangroves and they were not always appreciated. As it turns out they're also really good at absorbing carbon from the atmosphere and storing it in their soil. There has been an increasing interest in protecting mangroves and preserving them, and also in trying to regain some of that loss, and this is kind of a global effort that is happening right now. But before we're able to really carry out a lot of these projects of protection and reforestation, we want to make sure that we really have a good understanding of what some of those main drivers of change were that I mentioned before. So we did this study looking at Landsat data. We used the time series from the 2000 to 2016, and then we have baseline mangroves values from the 1990s. And then looked at changes in the Landsat data and the NDVI over time. What we found when looking at areas of change and comparing those with other data sets was that there were two sorts of main drivers in mangroves. So you can have the natural drivers of change, erosion or a storm surge or drought, any climatic change that would result in changes to the environment, changes to the water availability and mangroves. The real main drivers of change of mangroves are actually aquaculture and agriculture, so mangroves being converted to shrimp ponds or fishponds or to rice fields and other types of agricultural areas. Mangroves have also been converted to urban areas and to other uses, for example, just deforestation for fuel wood or deforestation for other types of agriculture or forestry. What we're noticing is that there is an increased awareness of the importance of mangroves, and what we actually found in our study is that there has been a decline on a global scale in deforestation rates and mangroves. It's actually quite encouraging to see that a lot of this work and research and studies have come out, and I'm hoping that some of the work that we've been doing is also responsible for this. It's actually resulting in some changes to policies, changes in management practices and general knowledge of the importance of these ecosystems by the general public.

HULT:

We've talked a lot about the coasts, about mangroves, but your work isn't just there, right? Tell us a little bit about your recent study of gold mines in Ghana.

FATOYINBO:

Essentially, we were reproducing some of the methodologies that we had developed for the mangroves and working with different partners who are in Ghana. We started looking at the issue of illegal goldmining. Ghana is one of the world's leading gold producers. And over the past few years, that has been really an increase in artisanal mining in the gold-producing regions. What we were trying to see with our study was whether we would be able to pick up on changes in forest cover and increases in areas that looked like mining using Landsat data. We carried out a NDVI anomaly analysis and trying to see whether we could track how some areas, where there used to be forest would get deforested and converted to bare Earth. And also, maybe some water. This gave us an indication that this would be where you would see an increase in artisanal mining.

HULT:

You sort of sifted through the Landsat archive for signs of change to vegetation health. And doing that you were able to pinpoint where these things were happening?

FATOYINBO:

Exactly. So we did notice then that there was an increase in the rate of deforestation in specific regions. These are areas that are close to rivers, and oftentimes these patches are kind of like squares or have a shape that really tells you, okay, this is most likely an area where you're having mining for gold.

HULT:

What were the top line results that you recorded? 

FATOYINBO:

We did notice that there had been an increase in artisanal mining in the second half of the 2000 to 2020 timeframe, starting in about 2015. There has been since then an increase in the rate of deforestation. When we mapped out the areas, we found that most of that deforestation in this particular region around Kumasi in Ghana was experiencing an increase in gold mining,

HULT:

47,000 hectares of vegetation converted, I believe?

FATOYINBO:

It is. It really is a really big area. And it's something that you can see even when you're looking at the satellite data or in Google Earth. What was really surprising was the rate at which a lot of these changes were happening. But it was encouraging that we would be able to monitor these. What was interesting here is that we were able to detect areas of gold mining and of deforestation with the Landsat data. But by combining it with ancillary data sets, that gave us some idea of areas that might be protected or areas that were open mining concessions. We could then also separate the types of mining into what was most likely illegal mining areas and areas that were legal, mining areas.

HULT:

Lola, do you have any closing thoughts? Anything you'd like to leave folks with if they're more interested in this topic or just in general?

FATOYINBO:

My closing thoughts would just be to really highlight how some of this data that we're using is actually being used in real world applications and is being used to make change to some of our policies and how we are interacting with our environment. As I mentioned earlier, I think for me, it's really encouraging to see that some of the work that we're doing is actually resulting in changes on the ground. So being able to produce maps with Landsat data that show how things have changed over time, and then actually seeing how they are improving and how we are losing less of these really critical and important ecosystems to me is really encouraging. And I'm really excited to see where the Landsat program is going in the future. And also so excited to start using some of that new Landsat 9 data that's coming down the pipeline soon.

HULT:

We've been talking to Dr. Lola Fatoyinbo of NASA Goddard Space Flight Center about coastal ecosystems land change and satellite technology. It's been a fascinating conversation, Lola, thank you so much for joining us.

FATOYINBO:

Thank you for having me. It's been a pleasure.

HULT:

And thank you to the listeners as well. Be sure to join us for the next episode of Eyes on Earth and be sure to check our audio archive.

If you're new to the show, you can find all our shows at USGS.gov/eros. That's USGS-dot-gov, forward slash E-R-O-S. You can also subscribe on Apple podcasts or Google podcasts. This podcast is a product of the U.S. Geological Survey, Department of Interior.