Eyes on Earth Episode 124 – Mapping Fires in Utah

JIM LUTZ:
What’s been striking to me is how naturally resilient forests are to fire. When I say forest, I don’t necessarily mean forests that have been heavily altered by people, but natural forests that have been relatively unaltered. Sure, fire kills a lot of trees, but it leaves a lot of trees alive, and the dynamics of what happens after a fire is truly amazing. If you’re going to study forests in the western United States, you have to study fire. Fire’s been with us forever, and it’s going to be with us, and we’re going to have to deal with it forever.

TOM ADAMSON:
Hello everyone and welcome to another episode of Eyes on Earth, a podcast produced at the USGS EROS Center. 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. My name is Tom Adamson.

Forests are not all the same. Neither are forest fires. And the State of Utah has a unique landscape. So how fire behaves and how forests recover there need to be studied with that in mind. In this episode we’re talking about mapping small, medium, and large fires in Utah and how it’s important for prescribed fire planning and forest management. Our guest is Jim Lutz, professor of forest ecology in the Department of Wildland Resources at Utah State University. Jim, can you start by telling us about your background?

LUTZ:
Well, let’s see. For me, academia is a second career. My first career I worked in engineering and investment banking, and I left those endeavors and went back to school to study forest ecology.

ADAMSON:
OK, now give us a brief summary on what this Utah Fire Atlas is and what you’re trying to find out about fires in Utah.

LUTZ:
The Utah Fire Atlas is a compendium of all the fires, more than 100 acres, that have occurred in Utah since the mid-1980s. What we’re really trying to do is understand the differences that the specific Utah vegetation types have in their response to fire. A lot of the fire research in the West has been done in places that experience very large fires. California, Montana. And the forests in Utah are different. The fires have generally not been as big. But we want to understand the specifics of Utah fires.

ADAMSON:
You’re working with something that we call the Monitoring Trends in Burn Severity, which we call MTBS. Can you describe what that is briefly, what are its benefits, what is that used for?

LUTZ:
Sure, MTBS is an incredibly valuable program that’s run on a nationwide basis. And here in the West, it provides an analysis of fires over 1,000 acres, and using a defined set of algorithms and heuristics, MTBS allows us to look at the effects of fire at a landscape scale. Most people understand the value of MTBS and looking at those larger fires, and we wanted to extend that same value down to a smaller set of medium sized fires.

ADAMSON:
Why do you need to map smaller fires in Utah? Is there something different that they tell us than what big fires can tell us?

LUTZ:
Sure. From a straight fire and forest ecology perspective, smaller fires are different from large fires. Burn severity, the effect of a fire, differs based on its size as well as the vegetation types it burns in. So if we want for some reason to understand how small or medium sized fires behave, we really have to do that analysis. Now, where that’s potentially important is in the realm of prescribed fire planning or in forest management. So in Utah, a lot of prescribed fires or let-burn fires are less than 1,000 acres. So we have a lot of data about how big fires burn, but the ones we might be interested in encouraging on the landscape are the ones we might want to manage below 1,000 acres.

ADAMSON:
Is there something different about Utah, why Utah doesn’t typically get as many larger fires that other states get?

LUTZ:
Well, you know, Utah is very different from a vegetation perspective. We have a lot of the basin and range type of topography that people are familiar with in say, Nevada, where our forested landscapes are generally higher elevation and they’re separated by dissected valleys, so we do have in a lot of the state natural fire breaks that keep fires from getting very, very large, but we also do have large contiguous areas of fires which could lead to large fire—very large fires. We haven’t had many yet though.

What we would really like to do here in Utah, as in most of the West, we have a fire deficit. We have a lot of forests that haven’t burned for a while. We have a lot of fuel buildup for various reasons and the more fire we can put on the ground in a relatively managed way, the healthier the state’s forests are going to be.

ADAMSON:
You mentioned burn severity. Can you briefly define what that is? Why is it important to know about burn severity?

LUTZ:
What we’re really looking at is what’s the effect on the ground of a fire. How many trees died, how many shrubs died? Was the soil affected by the heat of the fire? So we’re looking at the overall effect in various categories of what the fire did. That’s severity. High severity would indicate that there’s been a lot of change from the prefire state. Low severity means not so much change from the prefire state. Satellites measure reflectance. What is reflected from the ground to the satellite, which is quite a long way up. Now the reflectance of, you know, a fire, fire effects—it differs inherently by different vegetation types. Some vegetation types naturally burn in such a way that the satellite records a high severity. Some vegetation types burn in a way that the satellite measures as lower severity, but because this differs, we can’t take the simple number generated by the satellite and use that to assume the effect on the ground was the same everywhere in the West.

You know one of the difficulties with MTBS is, you know, the standard MTBS map for a fire is, you know, red, yellow, and green—high, medium, or low severity, and these colors and these severity classifications take on a totemic importance. People are looking at patches of red and say, oh no, everything is dead and that’s just not true. And people are looking at patches of green and saying everything must be fine. That’s probably not true either. So we need a more nuanced understanding of what that satellite data is telling us before we go off and make, you know, big plans just based on the red, yellow, green.

ADAMSON:
Is it fair to say that the Utah Fire Atlas kind of builds on how MTBS does this kind of mapping?

LUTZ:
Absolutely. For almost all the fires that were larger than 1,000 acres, we just used the MTBS data, and then we individually analyzed the fires that were between 100 and 1,000 acres, and for some of the MTBS fires, we thought that we could improve the accuracy by revising them, but generally speaking we just use the MTBS data.

ADAMSON:
There are some other data sources that you’re using, right, besides MTBS? You’re using Landsat, you’re also using, as I understand it, NAIP imagery. That’s another acronym for us to define. N-A-I-P—National Agriculture Imagery Program. And that’s a program of the U.S. Department of Agriculture. That’s aerial photography flown pretty systematically over the lower 48 States. So how are you using those different data sources?

LUTZ:
MTBS, of course, relies on Landsat data. The other big dataset that we use is the LANDFIRE dataset of vegetation types. So since we know that the effects of fire differ based on vegetation type, we want some way that we can divide the State of Utah into these categories of vegetation type so we can analyze fires in those vegetation types separately. And yeah, you know, like any other large dataset, it has its uncertainties. But to a first approximation, we’re able to look at, for example, forested versus non forested, or Douglas fir versus Aspen, or pinion juniper versus sagebrush. And it turns out that each of these vegetation types has characteristic severities, so we wouldn’t want to lump them all together, which would obscure what might be useful to managers. This is something we do that MTBS doesn’t usually do for the fires, since we’re interested in medium-sized fires. Sometimes it’s a little difficult to tell from Landsat where the boundary of the fire was, so we can use the NAIP imagery, which is much higher resolution. It’s a resolution where you can see the trees. We use the NAIP imagery to help us adjust the fire perimeters, you know, when they’re a little approximate.

ADAMSON:
OK, that sounds good. That higher resolution helps you understand, well, what type of vegetation you have there, what type of tree is in that particular forest?

LUTZ:
Sure, this is really important. You know, when we look at satellite source data, Landsat. OK, Landsat has got a 100 foot by 100 foot pixel. That’s really big. You can hide a lot of stuff in one Landsat pixel. Which means it’s a little uncertain and you have to deal with that uncertainty and potentially use other data sources to reduce that uncertainty. And NAIP is just one of the tools that we use to try to reduce the uncertainty that’s present in all uses of satellite data.

ADAMSON:
How does past fire activity inform our understanding of present and future fire dynamics?

LUTZ:
One of the things we want to understand is sort of what is the characteristic fire activity in any particular area. Here, you know we’re interested in the State of Utah. And if we have a good idea, as we do now, what’s been happening over the 30 years, we have a basis for discussion. That’s the important point because all over the West people are concerned about fire, and we have a lot of people who talk about fire, and sometimes this gets sensationalized that such and such a year, usually this year, is a very severe fire year, that the fires are bigger, that we’re losing more and more timber. And a lot of these statements are not really based on data. So what we want to do here is provide that data, divide that data by vegetation type so that we can start to have intelligent discussions about what we might do to improve outcomes.

ADAMSON:
Was there a reason that this goes back to the mid-1980s?

LUTZ:
Yeah, the modern Landsat satellites were launched then. We can get a consistently measured set of data from the mid-80s to present. I mean, it’s possible to use the earlier Landsat data, but it’s not directly comparable, and, you know, if we think about now, we have more than 30 years—that sort of gives us a way to look at what I call the fire normals. You know, we have climate normals. You know, if you look up on Wiki, what is the climate for a particular area, that relies on 30 years of climate data, and people respect 30 years as being enough to include big years and small years, I mean, you know, wet and dry. So I sort of consider fire in the same way that you need a lot of years of data to determine a baseline. And with Landsat, we’ve already got that.

ADAMSON:
That’s enough of a time scale for what you’re working on.

LUTZ:
I think it’s a really good start. You know, I think 30 years is enough to, it’s enough to have a discussion. Where it would be more desirable is in forest types or vegetation types where fire is very infrequent. So Utah does have a lot of high elevation forest types, spruce fir forests, and their natural fire return interval is rather long. So you know, we might not be catching, you know, some fire behavior that you know might happen next year or in 10 years. But you know, that is perfectly—quote—natural—unquote. And we just don’t have enough time to really see that. But I think for a lot of the areas that we’re interested in, you know, including WUI, the wilderness urban interface, or lower elevation, you know that 30 years is, you know, is really data that we can build off of.

ADAMSON:
Who is going to benefit from your information, or how do you hope it will be used?

LUTZ:
Again, we hope forest and fire scientists know that we need more fire on the landscape, but we need to put that fire on the landscape in a way that does not risk people or structures or other things that we value. So any information that can help us get more comfortable with fire that we might put on the landscape I think can be helpful. Of course, this is just one, you know, the information that we’re providing here is just another piece of information that our forest and fire managers can use. And, you know, here in Utah, we have very good forest and fire managers. They know a lot about fire, and we want to help them do their job with as much data as we can provide.

ADAMSON:
Give them another tool.

LUTZ:
Right, forest environment managers, you know, they’re always considering, you know, how to handle fires that have been ignited or how to, you know, conduct prescribed burns and these models to do that. And those models are parameterized with data. So the data goes into the models and if we have models that are based on places other than Utah. Oh, we have models of big fires instead of medium sized fires. We can be getting an answer that’s not necessarily the right answer for Utah.

ADAMSON:
What do you think will be the most important thing about the Utah Fire Atlas?

LUTZ:
Well, I mean it provides a lot of data for Utah and people looking at Utah forests. But I think beyond that, it gets to the idea that fire is different in different vegetation types and probably, you know, this kind of analysis would be good for others to repeat in their own area. You know, if we looked at how fires behave in Montana, probably very, very differently than how they behave in Utah. So as fire information, you know, like MTBS or, you know, other fire information becomes more and more available, I think it’s important for researchers to take a little bit more fine-grained approach to understand the specifics of their local area a little better.

ADAMSON:
Since you’re using satellite data to make these maps, does that mean field work isn’t necessary anymore?

LUTZ:
Nothing could be further from the truth. So satellite data, like I said, there’s a lot of things can go on in a 100 foot by 100 foot area. When we talk about the effects of fire on a landscape that has vegetation, we’re really interested in how many of those trees died and the satellite gives us an idea about how many trees died. But it’s very, very imprecise and it’s subject to a lot of misinterpretation. And satellites are really good at finding places where the severity was very high and very low. In that middle ground, they’re very, very imprecise. We quantified the uncertainty in satellite data with respect to actual tree death and it’s very high. You wouldn’t want to make a lot of management decisions solely based on the satellite data.

ADAMSON:
Is the Utah Fire Atlas available to the public yet?

LUTZ:
Yes, it is. We have a paper that we published open access, so anybody can download that. As part of that paper, we did something that’s a little a little unusual for academic papers. We published all of the summary data, so anybody can use that summary data. Fires by size, by vegetation type, by year for whatever they want. The actual satellite data analysis of all of the fires is available to those, you know, who need it, but it’s about two terabytes, so—

ADAMSON:
OK, sure.

LUTZ:
But the summary data, what a policymaker, what somebody looking at fire trends, all of what we think most people would want to get out of the fire atlas we published the numbers.

ADAMSON:
Is this ongoing work? Is there more to come in the future?

LUTZ:
Absolutely. We’re continuing to look at the dynamics of fire in Utah and elsewhere in the West because you know now that we understand what the baseline is for fire, oh, now we want to look at you know the different things that might influence that fire weather, you know, human influence, all these things could potentially be driving what we see in the fire data and we want to understand that and we are working on that actively.

ADAMSON:
I’d like to thank Jim Lutz for joining us on this episode of Eyes on Earth, where we talked about the challenges of mapping burn severity and fire recovery in Utah.

Check out our social media accounts to watch for all future episodes. You can also subscribe to us on Apple and YouTube podcasts.

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