Eyes on Earth Episode 83 - ECOSTRESS and Burn Severity

NATASHA STAVROS:

There's an inherent fear that we all experience with fire. And when we make decisions from a place of fear, it's very reactionary. That fear doesn't just sit with us as individuals. It sits with us as a society, and it transcends generation to generation. And so it's really on us to change the narrative and the way that we see fire and understand that fire is inevitable. The impact of it can be mitigated. And that's where the value of prediction is so important.

JANE LAWSON:

Hello, everyone, and welcome to another episode of Eyes on Earth. Our podcast focuses on our ever-changing planet and on the people at EROS and across the globe who use remote sensing to monitor the health of Earth. My name is Jane Lawson, and I'll be hosting today's episode where we're talking about wildfires and ECOSTRESS, which stands for Ecosystem Spaceborne Thermal Radiometer Experiment on Space Station. ECOSTRESS provides variables related to plant water stress, including evapotranspiration, evaporative stress index and water use efficiency. We'll take a look at how these variables were useful for research that looked at predictors for spatial patterns of burn severity in recent California wildfires in the Sierra Nevada and the Southern California mountains. The NASA Land Processes Distributed Active Archive Center, or LP DAAC for short, is located at EROS and has provided storage and distribution of ECOSTRESS data for more than four years. Our guests today are here to talk about their work involving ECOSTRESS and burn severity. Madeleine Pascolini-Campbell is a scientist at NASA Jet Propulsion Laboratory who studies hydroclimate variability along with agriculture and wildfire science applications. Fire ecologist Natasha Stavros is director of Earth Lab Analytics Hub at the University of Colorado Boulder and previously worked at the Jet Propulsion Laboratory. First of all, wildfires are inherently complex, and burn severity is just one aspect. Natasha, would you like to start by telling us why it would be beneficial to predict burn severity and how difficult that is?

STAVROS:

Burn severity just tells you a little bit about how much actually burned, from the soils to the understory up into the canopy. So when you say something has high severity, ecologically what that means is that you've burned the canopy. And being able to predict burn severity has a lot of implications from a management perspective because burn severity affects things like our ability, or for the ecosystem to recover. And likely if there's going to be a type conversion. So if you have a really hot fire in an ecosystem that is not used to experiencing hot fire that would have very deep seed reserves, you could burn through all of the seed reserves and make essentially a blank canvas, which allows for invasive species and potentially a type conversion. So what that means is that a forest could become a grassland. It also has huge implications for the amount of carbon loss that we could expect from a more high severity fire. And so this is really important because we can take different management techniques or strategies to reduce the likelihood of an extreme fire event. This can even include introducing fire itself, like a prescribed fire can actually reduce the amount of carbon loss from a wildfire that is not prescribed. It also has impacts for resilience and building resilience mitigation pathways such that communities are less susceptible to what we would call cascading disasters, which are essentially disasters that happen after a fire. So these are things like landslides, mudslides, debris flows, flooding that can happen in the rains the following winter when all the vegetation is removed. So those are just a few of the reasons why it's really beneficial to be able to predict burn severity. And the difficulty - we've been mapping burn severity for a long time, but we have not necessarily been predicting. And so looking at what variables would lead to a higher severity fire than a lower severity fire in the future as opposed to mapping what has happened.

LAWSON:

Madeleine, would you like to add anything?

MADELEINE PASCOLINI-CAMBELL:

Burn severity can change spatially from one spot to the next. And so one of the main motivating parts of this research was also to really get at these very heterogeneous patterns of burn severity and be able to explain them with a number of different predictors that Natasha mentioned.

LAWSON:

Madeleine, do you want to tell us about the variables you took into consideration in this research, in relation to the geographic locations of the six fires you examined. The variables included the ECOSTRESS plant water stress indicators, but also things like topography and vegetation type?

PASCOLINI-CAMBELL: 

Sure. As you mentioned earlier, we used data from ECOSTRESS. And this is an instrument that's on board the International Space Station, which orbits the Earth about 16 times a day and passes over the same spot every 1 to 5 days at different times. So the data that we get from ECOSTRESS can give us really good resolution, about the size of a football field. And ECOSTRESS is essentially measuring the surface temperature of the land surface. And this can be used in order to understand how stressed plants are, because plants will take in water through their roots and release it to the atmosphere through tiny pores on their leaves. And this process is known as transpiration, which helps to cool the plants down. But in hotter, dry weather, plants will actually close their pores to conserve water. And this leads to them heating up. So ECOSTRESS is able to measure these changes in temperature and convert that into these plant water stress measurements. So in this study, we used evapotranspiration, which includes the water being released by plants and can indicate plant vitality. We also used water use efficiency, which is a measure of how much water plants are using per unit of carbon that they store. So this can be used to understand whether plants are using water in an efficient way and can be a good indicator of how susceptible they are to things like drought. And we used evaporative stress index, which is also a measurement of plant stress. We used these variables for the year leading up to the fire season in order to capture the long-term drought conditions that were taking place in the year before the fire season, as well as to understand changes in plant biomass accumulation as these different water variables can indicate both plant stress, but as well as an increase in the amount of vegetation cover or fuel available to burn. In addition to that, lots of studies have shown that topography, as well as the land cover type, are also really important drivers of the patterns of burn severity. So in addition, we also

 used elevation and slope angle as well as aspect and also looked at different types of land cover, including forests and grasslands and savanna-type vegetation, as these have been shown to have different relationships between the different predictors and burn severity.

LAWSON:

That's a lot to keep track of.

PASCOLINI-CAMBELL:

Yeah.

LAWSON:

What were the most useful results, or what you consider the key takeaways in your study then, Madeleine?

PASCOLINI-CAMBELL:

So this study was one of the first to use ECOSTRESS in order to predict burn severity. And so I think a key takeaway was that we were able to show that by including information on plant water stress in the year before the fire, it was actually found to be a leading predictor of burn severity. And so this is an important use of the ECOSTRESS data, which hopefully can pave the way for future research in order to understand which regions might be most at risk to severe wildfire burning. We also found that it was very important to differentiate by land cover type or by the type of vegetation present. And we found different relationships in forests compared to grasslands where in forests we found that ECOSTRESS, which showed that if regions were more stressed, this actually was related to higher burn severity. Whereas in other types of ecosystems, such as in grasslands, we found that higher rates of evapotranspiration, which would indicate healthier vegetation, were related to greater burn severity, which was likely due to the accumulation of fuels and a buildup of grasses, therefore providing more fuel to burn. So, yeah, having these different types of information, we found, were very important in order to solve this hard problem of predicting burn severity.

LAWSON:

And dramatically different results than depending on where they were. What kind of future applications or research would you like to see come from this? Do you want to start, Natasha?

STAVROS:

Yeah, I think that this work is really important and sets a foundation for making the connection between research and management practices, explicitly thinking about the impact of how we manage either with or without fire, which can ultimately result in mega-fires or fires with negative impact to the ecosystem; what that means for ecosystem services like carbon storage, habitat and thermal regulation. There was just a paper that just came out showing that many of the carbon emission reductions in California over the last couple decades were, quote, erased by the 2020 fire season. But this is a little bit of a bait and switch. Because if we were to have fires on the landscape through prescribed burning and we were to manage our lands differently, we can actually change the amount and type of emissions that we could have from fires. And so being able to use information like the condition of the stand, how stressed that it is to focus our management practices in a very strategic way, can actually help us to improve things like carbon storage or habitat availability or thermal regulation of the forest on the landscape. So this would really allow us to use observations of plant stress in those decision frameworks to inform priority areas before we see a fire that's outside of our ability to manage.

LAWSON:

Madeleine, do you have anything you'd like to add?

PASCOLINI-CAMBELL:

Just to build on that, other studies in the past have used different measurements of plant stress to build into their models, but I think that a future application of this work is that by using ECOSTRESS, we have this much finer spatial resolution of 70 meters, which really is a big advancement in terms of how well we can detect the heterogeneous changes in plant stress. So I think that having the ability to incorporate this into future models could be very valuable in terms of producing higher resolution, fire risk mapping and predictions of where fires are going to have the greatest impact on the environment.

LAWSON

Madeleine, you've talked about the value of ECOSTRESS data in this study and you've worked with it in the past. Do you have anything else you'd like to add about how ECOSTRESS differs from other information that people can glean from other remote sensing sources?

PASCOLINI-CAMBELL:

Sure. So as I've mentioned, ECOSTRESS can be used to estimate the amount of water being used by plants. So this was useful in this study in terms of characterizing plant stress as it relates to fires. But it can also be used for a wide range of applications. And in past work, I've used measurements from ECOSTRESS to estimate water use due to irrigation. And this is important as it's difficult to come by data in terms of the amount of water that's being used by different agricultural regions. So if we have measurements taken from space, then this can be an advancement in terms of how we can monitor water use, which can be very important, especially in regions where water is a very important resource. Another aspect of that is that given that ECOSTRESS is this really high, 70 meter resolution, it's able to detect field scale measurements of evapotranspiration. And so in this research, we were able to estimate the amount of water being used by individual fields, which can be seen in the images to vary field to field. This is also a potential application of ECOSTRESS. And then in other work that, ongoing in my group, there is a number of different applications of ECOSTRESS data ranging from things such as monitoring the coastal water temperatures to understand algal blooms and habitability of fish species, as well as using ECOSTRESS to look at urban heat islands. So there is just so many different applications of the data.

LAWSON:

With wildfires increasing in frequency and intensity, how important is research like this for future fire management?

STAVROS:

You know, research like this is really crucial because management happens at a different scale than our observations. Our observations look backwards, and management needs to look forward. And that's kind of a tricky space to do data science in, especially when we have to deal with things like uncertainty and error propagation, and then how that ultimately affects a decision maker's liability for making a decision based on a prediction rather than something that's retrospective and hindsight, right? Hindsight is 20/20. So the more that we do research like this that allows us to understand predictors of things that have impacts on our landscape, the better situated we are to be able to make management decisions that help us to meet what we would call key performance indicators, or to meet our goals and objectives of management on the land in the first place. Otherwise, we're kind of using our instinct. And while that's great, there's also been social science studies that have shown that people's instinct can be colored by their past experience. So if somebody, a decision maker, had a really catastrophic fire in their district, they may be less likely to take a risk. Even if they understand that it's a low risk, they might - their risk posture may be lower in the first place just because of what happened in the past. And so using objectivity of observations to tell us the likelihood of something happening can really help ground us so that when we are faced with making those decisions, we have something that's objective that can ground us in how to how to make that decision in an informed way rather than in a reactionary, emotional way.

LAWSON:

The objectivity of remote sensing is a great point. Natasha, I understand wildfires have been more than just a research topic for you. How have your personal experiences influenced your work?

STAVROS:

When I say that what happened in a district can color the way that we perceive risk and the posture that we're willing to take going forward, I say that from a place of personal experience. Of course there's been research studies that have shown that this is not just exclusive to me, this is part of the human condition, and that we all sort of react similarly. For me, fires have been extremely personal for me. In early 2000, the first mega-fires of the 21st century in San Diego, my family evacuated, and my friends lost their homes. We took a week off of school for smoke week because the air quality was so bad. It happened again four years later. Again, more friends lost their homes. Since then, I have evacuated twice more and I've had close friends lose their homes in other fires here in Boulder, Colorado, with the Marshall Fire this last December. So it's extremely personal. And the more I reflect on it, the more I realize that there's an inherent fear that we all experience with fire. And when we make decisions from a place of fear, it's very reactionary, and it's often not informed. And so that was kind of my motivation for getting into scientific research was to kind of understand the science. And the reality is, is that we need fire on the landscape. And so if we act from a place of fear thinking that fire is bad because we're scared of it, then we actually set ourselves up for more devastation in the future. And so it's better for there to be fire and to understand that not all fires are the same. So prescribed fires that happen on days when conditions are more conducive for being able to manage where the fire goes is not the same as a wildfire event. And that fear does not just sit within us as individuals. We carry it from one generation to the next. When I've thought back and reflected, I've actually realized that I think my fear that I experienced started actually with my grandma, who lost a house to a fire on a farm. And she told my dad and my dad told me, and since I was a little girl, I was told, okay, the house could burn down. Just remember, the house could burn down. And so I think it's important to remember that that fear doesn't just sit with us as individuals. It sits with us as a society, and it transcends generation to generation. And so it's really on us to change the narrative and the way that we see fire and understand that fire is inevitable. The impact of it can be mitigated, and that's where the value of prediction is so important.

LAWSON:

I really appreciate you sharing your perspective. It's so valuable to hear a personal experience and angle versus just, you know, the straight science. So thank you, Natasha.

STAVROS:

Yeah, thanks for asking. I think, you know, one of the things is, is that fire is a very personal thing. It's something that every single human being understands. And it's the only, quote, natural disaster that we try to use as a tool. We don't use hurricanes as a tool, but we do use fire as a tool to cook our food. And in fact, fire is how we evolved as a species. So fire is a thing in the world, and it is personal. And I think that that's what makes it a topic that can really connect people together.

LAWSON:

Madeleine, did you have anything you wanted to add to this particular question?

PASCOLINI-CAMBELL:

So I can't really say I have quite the same personal experiences, but I do live in California, and it is impossible to live here without being very aware of the impact of wildfires. And that includes things both just the month of the year where you can't go outside because there's wildfire smoke and you have to be cognizant of that, to having these evacuation orders just outside of Jet Propulsion Laboratory. So I think that just living in the West gives you this really heightened awareness of wildfires, and for me also makes it something that I'm really interested to learn more about through my research.

LAWSON:

Madeleine, is there anything else you would like to add?

PASCOLINI-CAMBELL:

I would like to add that this research for me demonstrated the importance of these types of measurements that we are able to have with ECOSTRESS, and it really is a unique instrument in terms of providing information at this really high resolution, but also every 1 to 5 days and at different times of day. So by having this additional information, we're able to really map plant stress in a way that we haven't been able to before. And I'm hoping that by having this type of data set available, this can help pave the way for future studies to really understand the impact that vegetation has on how wildfires burn.

LAWSON: Thank you, Madeleine and Natasha, for joining us for this episode of Eyes on Earth, where we talked about how ECOSTRESS can help provide predictors of burn severity. And thank you to the listeners. Check out our EROS Facebook and Twitter pages to watch for our newest episodes, and you can also subscribe to us on Apple and Google podcasts.

VOICE 4:

This podcast.

VOICE 5:

This podcast.

VOICE 6:

This podcast. 

VOICE 7: This podcast is a product of the U.S. Geological Survey, Department of Interior.