New moisture sensors could help make wildfire danger ratings better in Arizona

A pilot project in the Tonto National Forest is studying a potential new way to gauge wildfire risk.

It involves monitoring moisture levels in both soil and plants — both living and dead. If it goes well, the sensors could be used across different ecosystems around Arizona and beyond.

Salli Dymond is an associate professor in the School of Forestry at Northern Arizona University and is working on the project.

Full conversation

MARK BRODIE: Salli, can you explain why this is an area that’s ripe for more study, compared to what we know about it so far?

SALLI DYMOND: Yeah. So we know that fire needs some basic elements, right? Needs fuel, heat, oxygen. And in wildfire settings, this fuel is typically in the form of live plants that are close to the ground, as well as dead material like organic material that’s on the forest floor. And so the ability of that material to ignite and carry fire is a function of its moisture content.

Historically, it’s really hard to collect fuel moisture data. People have to go out into the forest and collect it. Usually it’s done every two weeks if we’re lucky. So in our new study, we’re using live and dead fuel moisture sensors coupled with soil moisture sensors, which gives us real-time data on the moisture content of the whole forest.

So really what we’re hoping to do is better understand the drying of the plants, the whole moisture content of the system, and predict their potential to ignite and carry fire.

MARK BRODIE: So when you’re monitoring these, are you able to put a sensor, for example, in the soil or on a plant once and have it continually record data as opposed to having to continue to go to those locations?

SALLI DYMOND: Exactly. So not only are you getting — when you go out and collect the samples, you get essentially what we call a snapshot in time. So these sensors, which the live and dead fuel sensors go right on the plant or the woody material itself, the soil sensors go down in the ground, but we can collect data every 15 minutes, and the really cool thing is that we can access it remotely.

MARK BRODIE: I would imagine that’s really helpful when you’re trying to make decisions based on, for example, lightning strikes or, you know, any other number of factors that might lead to a wildfire.

SALLI DYMOND: Definitely. The nice thing, too, is you can put these in pretty remote locations. And so, you know, if you have a lightning strike nearby, you can look at that soil moisture data, the live or dead fuel moisture data, as well as the weather station data that we collect, and make really, really fine-point scale management decisions about what you’re going to do to fight that fire.

MARK BRODIE: So does that mean then that you can, for example, have particular neighborhoods, or even blocks, you know, in the Ready-Set-Go in terms of evacuation or know where to deploy particular firefighting resources?

SALLI DYMOND: That is the hope, is that we could really use this rather than issuing a really broad-scale, say a Red Flag warning or high or extreme fire danger. We could make it much more localized.

MARK BRODIE: Do you know like what level of moisture or lack of moisture is sort of the sweet spot for a fire igniting? Like do you — I guess I know you know how to use the data, but how specifically are you able to interpret that data?

SALLI DYMOND: Yeah. So the Forest Service now collects live fuel moisture, and so they have very particular thresholds based on that. The fuel moisture sensors and the soil moisture sensors go on a different scale, so we’re learning what that moisture content should be in the fuel moisture sensors.

For soil moisture, we can certainly see when the soil is dry enough that we expect those plants to be stressed. So those are really well-known values, and we’ve certainly seen some of those values in the data this summer.

MARK BRODIE: How have you decided where to put these sensors to begin with?

SALLI DYMOND: Yeah. So we’ve been working very closely with the Tonto National Forest and then Salt River Project, who’s funding this research, to identify sites where they’re actually collecting those fuel moisture samples. So that allows us, at least every two weeks, to compare their data with our data, which is incredibly important anytime you’re putting out a new type of sensor.

And then the Tonto National Forest has been incredibly helpful for helping us understand what sensors, what trees to instrument and where to put our sensors. So we’re really interested in what our information can give the forest managers.

MARK BRODIE: So what do you think is the ultimate potential of being able to do this? Like how far do you think this could go? How much benefit do you think it could offer?

SALLI DYMOND: Yeah, so the technology I’m very excited about, the potential and where we can see it go. So one, as I mentioned, we get this information real-time, so we can see what the system is doing. We can actually — we’ve had fire go over these sensors before, so in terms of understanding how to make decisions on the ground about wildfire or prescribed fire is really, really exciting.

Another thing, too, is just incorporating this information into our new fire danger rating system, so updating those with real-time data so we’re not relying on some of that coarse information about how live, dead fuels, what their moisture content is.

And then another really exciting thing is that when we think about the soil moisture going into the winter or the dormant season, we can better understand the status of those moisture contents and what we think the trees will be doing as essentially they green up. So are they going into that green-up as what we would call healthy or are they going to be vulnerable to any large droughts that we get during the growing season?

MARK BRODIE: I mean, it sounds like there might be some hydrological benefits here, too.

SALLI DYMOND: Ah, there are. And actually my background is forest hydrology. And so understanding that soil moisture component, that live fuel moisture component gives us so much information about the health of the forest, its vulnerability to fire, but also to pathogens and pests.

But we can also get, with that belowground information, a better understanding of hydrology in terms of post-wildfire risk for flooding, any type of monsoon risk for flooding. So really all this information together is going to give us a lot that we can do in terms of protecting the surrounding forests and communities.

MARK BRODIE: Could it also potentially be helpful in terms of trying to prevent fires? Like if you know an area that, you know, the soil moisture is particularly low and there’s maybe you’re noticing more fuels there, more dead fuels there, that you can say, “Hey, you know, somebody should go and clear that stuff out so there’s less fuel if a wildfire were to come."

SALLI DYMOND: Oh, for sure. So, you know, one of the things we’ve talked about is spread and what happens when it’s igniting, but that ability to localize those decision-making tools and say, "OK, yeah, this forest is looking really stressed. Let me prioritize this for one of our thinning operations or one of our fuel treatments."

And same with those dead fuels, like, "Oh yeah, these dead fuels are really, really dry. Let’s go out and clear them," is really — in an ideal world, we’re not getting large wildfires. And so the ability to manage before that happens is much better and more ideal than trying to fight these large fires.

MARK BRODIE: All right, that is Salli Dymond, an associate professor in Northern Arizona University’s School of Forestry. Salli, thanks so much. I appreciate it.

SALLI DYMOND: Thank you for having me.