top of page

Executive Q&A: Rohan Shivakumar on How Trees Might Help Predict Wildfires Before They Start

When it comes to battling wildfires, every second counts. But what if we could know days in advance where a blaze might spread—because the trees themselves told us?


That’s the groundbreaking idea behind the research of Rohan Shivakumar, a student and environmental science researcher whose latest study is in the process of being published. His work explores how plants release water under stress—essentially “sweating” as fire approaches—and how satellite technology can detect these signals to forecast wildfire behavior.

Rohan Shivakumar.
Rohan Shivakumar.

In this conversation, Rohan shares the spark that inspired his research, what trees have taught him about resilience, and why the discovery could reshape how we fight fires—and how we listen to nature.


Q: Let’s start with the moment of discovery—can you describe what it felt like when you realized trees might be warning us days before a wildfire starts?


When the potential implications of my work started clicking in my head, I remember lying back in my chair amazed by the fact that I just did something that could push the frontiers of environmental science, but more importantly, save lives and homes. My process enabled me to see things holistically, not from just a data perspective, but from the perspective of the trees themselves.


Q: How did your partnership with your professor begin, and what has it been like working so closely on something with such big implications?


I met with Dr. Joshua B. Fisher at a summer program called UCLA COSMOS. We discussed various topics; and, since we are both passionate about environmental science and remote sensing applications, we bonded over a variety of remote sensing instruments and areas of environmental science research.


I feel incredibly lucky to have had Dr. Fisher to mentor me during my latest project, with his years of experience and hundreds of publications under his belt. He offered invaluable advice, from helping design the procedures to providing feedback on my figures and enhancing the rigour of my data analysis. Working with Dr. Fisher has been both educational and reassuring.


Q: Many people think of trees as passive—quiet parts of the landscape. What did you learn about how alive and communicative they actually are?


At first, I didn’t realize how important plants would be in understanding fire spread dynamics either. It was only when I learned that plants release water when faced with a fire threat, did the idea even occur to me that plants could predict fire spread.


In other words, plants “sweat” as fire is heading towards them. Moreover, our research found that by monitoring plants around a fire’s perimeter, and observing how they react, we can absolutely predict the path of a fire, which has huge implications in firefighting.


Q: Was there a specific incident or wildfire that made you want to explore this topic in the first place?


In 2018, when I was ten years old, the Woolsey Fire ravaged my neighborhood. While I was lucky to escape the scene with my house intact, many of my friends and even my family members were not as lucky.


I asked my parents questions about fire spread. Their answers sparked my interest to learn how to help communities and, in high school, I decided that I wanted to pursue my passion for helping people further. Motivated by looking at Cal Fire maps of active fires, and how they progressed day after day, I decided to explore how and why fire spreads. That has led me to my work today.


Q: Can you walk us through how the trees send these early warnings—what are they doing, and how are we able to ‘hear’ or measure it?


Plants release water in response to an approaching fire threat. This release is detectable by instruments from space (e.g. Landsat). From this, we discovered that understanding trends in plant stress, or water release, multiple days before the fire hit that specific patch of vegetation, could help us predict fire spread in advance of it happening. This has the potential to save lives, homes, and communities.


Q: What kinds of tools or technology did you use to track the signals the trees were giving off before a fire?


To track the signals trees were giving off, we used satellites like Landsat, Sentinel-2, GOES, among many others. We accessed the data through OpenET’s models. This enabled us to get evaporative stress (how “stressed” the plants are) data. We analyzed how the evaporative stress on plants changed as a fire approached.


Q: Was there ever a moment where you doubted what you were seeing, or thought, “This might be too good to be true”?


Yes. The project actually started with random data exploration, as I looked through land surface temperature data and explored how we could track fire perimeters. That eventually turned into trying to figure out how plants react to the expanding fire.


Then, I realized that what I found could have huge implications for firefighting and save lives. At first, that was a shock to me, as I never expected to come across something like that; it just happened through my curiosity and exploration.


Q: How has this research changed the way you see forests or nature when you’re out in it now?


My work reminds me of The Giving Tree by Shel Silverstein. In the story, the tree gives everything it has to the boy; from its apples, its branches, and even its trunk. Similarly, I’ve realized that trees have something to tell us too: how a fire might spread, and that warning offers a chance to act before it’s too late.


Q: What have you learned from the trees—not just scientifically, but emotionally or spiritually—from doing this work?


Through my research, I’ve learned to think of trees not just as simply objects but as teachers. Specifically, they’ve shown me how much they can tell us. I’m sure that plants have so much more to tell us, even beyond this specific wildfire application. Who knows, they might be able to help us solve some of the most important problems globally, like drought or climate change, someday.


Q: What’s been the reaction from scientists, environmentalists, or even firefighters when they hear about your findings?


I’ve learned from discussions with fire departments that more than anything, firefighters are looking for integration tools more than new technology or data; they need something to bridge the research to the real world. That’s why we’ve been working with Urban Sky to integrate our findings into their systems.


I’ve contacted researchers around the country for their feedback, and they’ve shown interest in the work, adding credibility to our work. Additionally, we’re working to get published in a peer-reviewed journal.


Q: How do you imagine this discovery could be used in real time to prevent damage from wildfires or save lives?


My mentor has worked with Urban Sky–a startup that launches high-altitude balloons to monitor fires and collect data–and we’re looking to find a way to integrate our findings into their systems. That could be one step in the right direction.


Q: What’s something about this research that most people might not expect or might find surprising?


This entire project was done from my bedroom. Unlike the common portrayal of research as dozens of people in lab coats with test tubes, I was able to complete this entire project on my laptop by using satellite data available online. That speaks to the importance of satellite data in understanding our planet.


Q: Do you think humans have lost some of their ability to listen to nature—and could your research be a way to help us reconnect?


In some ways, we’ve lost our ability to connect with nature because we’re so involved in technology. But technology can actually help us reconnect by helping us understand how fire spreads and how plants are giving us these early-warning signals due to an incoming fire.


Q: What’s next for you—do you plan to keep studying how nature sends warnings, or is there another mystery you're hoping to explore?


Currently, I’m working on building an app that uses machine learning, specifically artificial neural networks, to predict wildfire spread during active fires. This could be useful to firefighters who need to better understand where to allocate resources or order evacuations more effectively. In college, I plan to major in environmental science or environmental engineering to keep exploring wildfires and other interesting topics.


Q: If you could tell the public one simple thing about trees and wildfires that you wish everyone understood, what would it be?


Originally, like many people, I thought that fires were completely random and determined by environmental factors like lightning. While that is true, it’s truly astounding how much we can learn about fires through technology. As in my study, we can use remote sensing technology to understand how fires spread, but there’s also been work done on how we can integrate machine learning and artificial neural networks to predict fire spread or how large a fire will get.


 
 
bottom of page