Skip to main content
U.S. flag

An official website of the United States government

Eyes on Earth Episode 23 – Ecosystem Monitoring

Right-click and save to download

Detailed Description

Landsat satellites offer a wealth of information to scientists studying ecosystem health and recovery. Data products derived from Landsat open even more areas of inquiry. Landsat Burned Area products, for example, can help researchers identify previously unknown fires that took place from 1984 through the present. In this episode of Eyes on Earth, we talk to one of the scientists behind that product about Landsat’s role in monitoring ecosystem health and recovery.




Public Domain.


Hello everyone. Welcome to this episode of Eyes on Earth. Our podcast focuses on our ever-changing planet, and on the people here at EROS and across the globe who use remote sensing to monitor and study the health of Earth. Iím your host, Steve Young.
Todayís guest is Melanie Vanderhoof, a research geographer in the Geosciences and Environmental Change Sciences Center in Denver, Colorado. Melanie uses satellite imagery to help us understand how ecosystems change over time, and how they respond to extreme events.
Welcome Melanie.
Your focus seems to be on how ecosystems change over time following droughts, floods, and fires. How did you get interested in that kind of work?
Iíve always been really interested in how ecosystems work. I spent a lot of time wandering through the forests of New Hampshire as a kid, and most ecosystems are such dynamic and resilient environments. I think one of the best ways to see that in action is by studying how they respond to a large disturbance event, or a substantial change, like an insect outbreak, a fire, or an extreme climate event.
Why is that work important?
Well, in addition to having profound impacts on how these ecosystems function, these disturbances can be massively expensive for local communities. For example, wildfires can threaten and burn houses in the wildland-urban interface. Floods can damage housing, and delay or damage crops. And, understanding the disturbance history or the natural range of variability can help communities evaluate risk, and help protect themselves against future threats.
When you say youíre researching post-fire regeneration across the western U.S., what specifically are you looking at and why?
Over the last five to 10 years, there have been a number of field-based studies in the U.S., most often across the southern Rockies, in New Mexico, Arizona, and Colorado, that have found limited or no regeneration of some conifer species following fire events. And thereís increasing concern that as climate patterns change, more forests may transition to grassland or a shrub system, which could have pretty dramatic impacts for wildlife, carbon storage, and climate vegetation feedbacks. We are interested if this concern is just limited to the southern range of some of these conifer species, or if the pattern persists at the scale of the western United States. So, to look across that big of an area, weíre relying on the Landsat archive. And weíre using imagery from the growing season, together with imagery from when snow cover is present to isolate the regenerating conifers from grass, forests, and shrubs that might be structurally confused with regenerating conifers. 
You co-authored the Landsat Burned Area product. What is this product and what is the value of using Landsat to produce it? 
The Landsat Burned Area Product uses the entire Landsat archive, so 1984 to present, across the conterminous United States, to map and track burned area extent every eight to 16 days, at 30-meter resolution. And because the product generation is largely automated, we now have a better understanding about how fire extent may have changed over the last 35-plus years. Prior to our product, there were and continue to be a number of other U.S. and global fire products, and each one kind of has its own contribution and value. But one of the challenges that U.S. fire products have faced is that they often rely on fires being reported, and reporting efforts can be quite uneven over time. And that means it can be harder to understand at a national scale how fires are changing. Global products on the other hand typically rely on satellites like MODIS that are coarser in resolution, so about 500 meters to one kilometer in resolution, and they tend to have a shorter temporal record relative to Landsat. Because we have 30-meter resolution with Landsat, we can map heterogeneity within a given fire, so we can distinguish relatively small unburned patches within a fire that might serve as a feed source and wildlife refuge as the ecosystem recovers.
If you understand that, if you understand the where and the why, how does that become valuable to forest managers?
Well, it helps you understand what the current condition is of the forest, and that enables you to have more informed, and more data to be able to make decisions on how to manage and move forward. It also helps you to better predict the risk of that area burning again, particularly when youíre looking at areas that are close to communities. 
That means then potentially going into those high-risk areas and doing prescribed burns to remove fuel loads, things like that?
Exactly, yeah. A lot of these ... at a state level, a local level, ultimately managing forests, actively doing prescribed burns, understory clearing ... can be really expensive. And so with limited resources, we need to be able to figure out how to prioritize and most efficiently use that funding to make our forests healthier.
And that probably answers the next question I have, but I have read where you have indicated that the analysis of satellite imagery enables cutting-edge science in a cost-efficient manner. How is it cost efficient?
Well, field-based data collection and field-installed sensors will always be a critical component of researching ecosystems. But field work is labor and time intensive, and it can be really hard to generalize our understanding from a few select sites to a regional or a national scale. Using satellite imagery has allowed us to take a birdís eye view for decades. But more recently, as cloud-based data platforms are starting to become more widely available, I can now analyze thousands of Landsat images in less time than it took me to analyze 10 Landsat images 10 years ago. And this new capability really opens the door for big data approaches to help provide decision makers with the data they might need. Having said that, I think having local partners and stakeholders involved in each project is so critical to make sure that any remote sensing effort makes sense at a localized scale. And that we as remote sensors are correctly considering sources of uncertainty and correctly interpreting our findings.
I understand that youíve been doing some recent research in southwestern Montana. What exactly are you looking at, what kind of changes are you seeing there, and why is that an important place to study?
One of our recent research projects was focused on the upper headwaters of the Missouri River. So, itís just in kind of southwestern Montana. At the outlet of the basin, three major rivers combine to form the Missouri River. Itís a fascinating area. Fly fishing is an important economic activity. And for the fish to survive, there has to be cool water in the rivers throughout the entire year, including in the late summer, when we have the lowest flows. Thereís also a lot of irrigated agriculture thatís pulling surface waters from  the rivers. And the region has experienced several major droughts over the past several decades. And there is a lot of concern that as more precipitation comes as rain instead of snow, that the timing of that water will skew earlier in the growth season, and make late season flows more limited. There is a lot of active stakeholders in the area, so thereís nonprofits that are working with farmers, and researchers from the University of Montana. Theyíve got state and Federal partners involved. And one of the actions that nonprofits are taking is to use beaver mimickry to restore headwater streams. So, that idea is to help reconnect these streams with their flood plains in order to help the headwaters act as a sponge to slow the flow of water, particularly in the spring months. And we partnered with the Nature Conservancy and used high-resolution imagery to help quantify the impact of their restoration activities by examining how the sites have changed between pre- and post-restoration. 
And these restoration activities are meant to impact the larger downstream rivers?
That is a major goal of some of the nonprofits working in the area. And thatís why we also looked at long-term trends in the riparian condition along the major rivers downstream from these restoration areas. And we are interested in how land use activities, like changes in the irrigated agriculture, might be influencing the condition of the riparian areas downstream of the restored streams. And so, we used Landsat imagery from 1984 to present to look for changes in the wetness of riparian quarters over time. And how this might be related to both variability in climate as well as changes in agricultural activities. 
Youíre saying that riparian corridors important? Why is that?
Riparian corridors are incredibly important ecologically. They store water, nutrients and sediments. They reduce downstream flood impacts and hold non-point source pollution. They provide corridors for animals to move and migrate, particularly through environments that can be hostile to wildlife, such as arid, urban, or agricultural areas. And they sustain fish habitat by lowering stream temperatures and contributing woody debris. But globally, the degradation of riparian areas is a major issue. As rivers become more regulated through diversions and pumping, these activities tend to alter the timing and the magnitude of flow, and often end up degrading the riparian corridors. 
Youíve also done a certain amount of work on the extent of surface water, the movement of water, how itís stored and distributed. Why is that information important?
Information on our surface water is really fundamental to the economy, to protecting life and property, and to help us effectively manage the nationís water resources. But, the amount of surface water and how itís distributed is in constant flux, and reacts to episodic events like rainfall, or drought, seasonal patterns in rainfall and snowmelt, as well as annual variability in precipitation timing and amount. Remote sensing plays a really important role in that it can help us understand the variability in surface water extent over time, as well as help us understand how sensitive watersheds are to that variability and climate. 
Youíre also interested in temporary wetland loss versus permanent loss? Why? What are the implications of permanent wetland loss?
So, wetlands are similar to ... have a lot of similar functions to riparian areas in that they help abate floods, control erosion, sequester carbon, support biodiversity, improve water quality. So, lots of important ecosystem services. But over the last 300 years or so, weíve lost almost 90 percent of wetlands globally, and about 30 percent of wetland areas since 1970. And wetlands can be degraded or lost through a variety of changes. But one of the major causes is changes in land use, for example, expansion of agriculture, or urban development. And these changes make it essential for us to not only monitor wetland extent and condition over time, but to also enhance our understanding of the causes and the spatial distribution wetland loss and gain. Weíre now getting to the point where we can remotely track surface water extent over time. And I see the next step is in being able to separate temporary loss in surface water extent which might be attributable to, letís say a drought year, but which can be expected to return when conditions become wetter from a permanent loss, for instance, when a development occurs. And in one of our ongoing research projects, weíre using the Landsat archive to concurrently track surface water extent and land disturbance with the idea that most permanent wetland loss caused by land use change is likely to be associated with a disturbance event that we can also detect remotely.
Weíve been talking to Melanie Vanderhoof, a research geographer in the Geosciences and Environmental Change Sciences Center in Denver, Colorado.. Thanks for joining us Melanie.
We hope you come back for the next episode of Eyes on Earth. This podcast is a product of the U.S. Geological Survey, Department of the Interior. Thanks for joining us.

Show Transcript