Reforestation: Plant Here, Not There

In this episode we continue our discussion of the reforestation pipeline. Last month, we discussed how reforestation efforts are critical after high severity wildfires, like the ones we have been witnessing across the Southwest. Now we sit down with two researchers who specialize in modeling to learn more about the science of choosing the right spot on the landscape for the best chances of planted seedling survival.

If you’re enjoying this podcast, please consider rating us and/or leaving us a review on Apple Podcasts, Podcast Addict, or Podchaser Thanks!

Follow us on X @RainShinePod

Never miss an episode! Sign up to get an email alert whenever a new episode publishes!

Have a suggestion for a future episode? Please tell us!

Come Rain or Shine affiliate links:

DOI Southwest CASC: https://www.swcasc.arizona.edu/  

USDA Southwest Climate Hub: https://www.climatehubs.usda.gov/hubs/southwest  

Sustainable Southwest Beef Project (NIFA Grant #2019-69012-29853): https://southwestbeef.org/ 

Transcript

Emile Elias: Welcome to Come Rain or Shine, podcast of the USDA Southwest Climate Hub 

Sarah LeRoy: and the DOI Southwest Climate Adaptation Science Center, or Southwest CASC, operated by the USGS. I'm Sarah LeRoy, Research Coordinator for the Southwest CASC. 

Emile Elias: And I'm Emile Elias, Director of the Southwest Climate Hub. Here we highlight stories to share the most recent advances in climate science, weather, and climate adaptation and innovative practices to support resilient landscapes and communities.

Sarah LeRoy: We believe that sharing some of the most innovative, forward thinking, and creative climate science and adaptation will strengthen our collective ability to respond to even the most challenging impacts of climate change in one of the hottest and driest regions of the world.

The contents of this podcast are for informational purposes only and should not be interpreted as endorsement for any of the products, technologies, or strategies discussed.

Today, we are continuing our discussion of the reforestation pipeline. Last month, we discussed how reforestation efforts are critical after high severity wildfires, like the ones we have been witnessing across the Southwest. As a reminder, the reforestation pipeline includes several steps, including conifer seed collection, processing, and storage, nursery production, site selection, preparation, and planting, and post planting maintenance and monitoring.

Each portion of this pipeline is important to the overall reforestation success. Last month, we talked with Josh Sloan and Rachel Foe about the first step in the process, including conifer seed collection, processing, and storage. Today, we're focusing our discussion on site selection and choosing the right trees for the right location.

Joining us are Matt Hurteau, a professor in the Department of Biology at the University of New Mexico, and Chris Marsh, a research assistant professor in the Earth Systems Ecology Lab at the University of New Mexico. Chris and Matt, welcome and thank you for joining us today. Maybe you could just start us off by telling us a little about your research and how it connects with the efforts of the New Mexico Reforestation Center.

Matt, we'll start with you. 

Matt Hurteau: Thanks, Sarah. So our research program is really focused on trying to increase the chance that a planted seedling survives when it gets put in the ground right so we're really interested in understanding those severely burned landscapes and then using that information or making that information available so that it can be used to improve outcomes for post fire reforestation.

Sarah LeRoy: Chris, same question to you. Could you tell us a little bit about your research?  

Chris Marsh: Yeah, sure. Hi. So I'm mainly concentrated on the microclimatic environment of those post fire areas, and we started off trying to classify that in terms of facial heterogeneity within the site and where different areas across a burn scar may offer microclimate refugia for different seedlings that we're putting into the ground and where we might be able to predict where those areas occur in a burn scar based on the biophysics of the system and potentially some vegetation which may exist after a fire as well.

 So there's sort of two levels to it which is the topographical part which we're working on most recently and also the vegetation which is existing which may offer some solar radiation buffering, creating some shade that we can plant those seedlings in.

Emile Elias: So, Matt, can you explain how your work helps guide decisions about what to plant where and why it's so important to, to apply that knowledge? 

Matt Hurteau: Yeah, we, we are really trying to tackle this problem that we see across the Southwest, which is on average about 25 percent survival of planted seedlings. And, you know, that masks a lot of variability.

So there are some places where no seedlings survive when they plant and some places where lots of seedlings survive. And so, you know, our goal at the end of the day is really to get that survival number up. And so, one of the ways we can go about doing that is finding places where, regardless of what happens with temperature and drought and stuff like that, the seedlings just have a better chance of surviving because maybe they're a little more shaded because, you know, there's a slope above the seedling and it's on the north face of that slope.

Uh, or maybe they're in a place where water tends to aggregate. So a little depression or something like that. And so we're really focused on identifying those places on the landscape and then suggesting that, you know, we focus planting in those places so that we end up with overall higher survival rates and, and that has the added benefit of, you know, we're planting out fewer seedlings that die and those seedlings all require a seed, right, which had to be collected in a tree.

So it's basically like getting more efficient with the process. 

Emile Elias: Okay, excellent. And so that, that leads to a question I've had, which is how do you decide which areas within a disturbed site? should receive seedlings for regeneration. So if there's a, you know, huge area and you can't plant all of it, so how do you decide where you're going to replant?

And Chris, we'll start with you on that. 

Chris Marsh: So yeah, we've been working on a modeling process to kind of make those decisions for us. Cause when you're sort of ground level of a site in a post fire area, which has had a high severity burn, it can be kind of daunting to look at that landscape and decide where to plant.

They're normally relatively large in area, recently even more so, and in a high severity burn just after maybe a couple of years after that burn's gone through, there's normally 100 percent tree mortality in some of those more severe burns. And when you're confronted in that sort of moonscape, trying to decide where to best put your efforts can be quite challenging.

So some of the modeling that we've done. Take some data which is readily available to the public, basically digital elevation models which are small scale elevation data captured normally by the USGS with LIDAR surveys, and it's those ground hits, and that allows us to get an awful lot of information just from that elevation of a particular square, which is how that data is sort of collected and then rasterized into a form that we can use. So we can process that data and find stuff out like slope and aspect like Matt mentioned, but also where water will likely flow. On that landscape when it, there is precipitation in the system, and where it's shielded from the highest temperatures during the summer, and that's normally a bit of aspect, but also where we'll shield seedlings from high winds over a small radius, normally like 10 square meters shaded in that sort of Immediate divot or it, its elevation relative to a larger landscape, about a hundred meters around from it.

And those five variables, slope topographical, wetness index, heat load index, what we call topographical position and roughness index, which gives us a decent idea of what's protected from the wind, allows us to model planted seedling survival after monitoring a lot of seedlings, planted in the same place, and seeing which survived.

And which didn't, and put those into a machine learning model to make those decisions for us. It gives us a probability map of where those seedlings were most likely to survive in a singular landscape. And that's kind of cool because, as I said, the biophysics of that system um, don't necessarily change.

So we can apply that same modeling framework to different burn scars and have the same sort of model output, which can be used as a guide for land managers when they come to make those reforestation decisions about where to best put their effort, which is most likely to see chances of success.

Emile Elias:  All right, excellent. And Matt, do you have anything to add? 

Matt Hurteau: Yeah, I think, you know, Chris gave a great explanation of what we're looking for in those high severity burn footprints. I think the other thing too is that we're making the assumption that you're not going to target places where you're going to get natural seed rain, right?

So The model is modeling the biophysical environment and not necessarily saying, oh, well, we've got live trees within 50 meters of this location. We don't need to plant here. And so it still does require, you know, that site specific understanding, uh, because it's not, doesn't make a lot of sense to plant trees when you've got a seed source within, you know, dispersal distance.

Emile Elias: Right. Excellent. Thanks. So there may be some fires that go through that aren't quite as intense where you might have some natural regeneration. And so you avoid that spot and prioritize some others. Excellent, thanks for that. And Chris, you may have called on this in your last answer, but I know a couple of years ago, you were the lead author on a paper about modeling the probability of seedling survival.

And so I'm wondering, based upon that work, what would you recommend in terms of site selection and seedling survival rates? And I know you just went through some of those site characteristics, so is there anything else that you would add to that? And specifically our listeners might not all be familiar with aspect, so if you can tell us a little bit more about aspect that could be helpful.

Chris Marsh: Sure, yeah, happy to. So aspect normally describes which way a slope is facing, like a northerly facing slope or a southerly facing slope, maybe east or west. And those are typically like the those points on the compass are normally typically how that's um, that that's classified uh, kind of in a slightly old fashioned way.

What we tend to use now is a heat load index, which just tells you basically the same information but without a categorization, just how much sunlight hits a slope over a year's worth of time on a scale from zero to one. So it becomes quite an easy measure of how hot a particular place on the landscape is going to get.

So that's one of the main variables that we've been used in that modeling process. So as Matt was kind of saying, we can start off a step when we're looking at reforestation or how I would like it to see to be done is using the model to identify places on the landscape where you're going to see the best chance of reforestation success, the most chance of seedlings survival.

But there's also areas around that. One of the main things that we came away from our modeling work, which was like a real eye opener was that only a small proportion of high severity burn areas are actually suitable for replanting. Those overstory canopies, which used to be there, that's the shaded, moist environment under the mother tree.

That's where those seedlings have evolved to establish and grow within those environments. And when you have a high severity burn over large areas, a lot of that shade has sort of gone up in smoke. So it's kind of a, it's a harsh thermal environment that we're asking these seedlings to establish and grow in.

Not only is it without the shade of that overstory canopy, but there's a lot of other vegetation which comes back immediately after a burn, which may not be the vegetation that we're after. The Jemez Mountains in the footprint of the Los Conchas burn area, we're seeing a lot of New Mexico locust coming through.

In areas where the model is also suggesting would be a good area to plant, so they kind of cancel each other out. So it's those shoulder areas that we're also looking to plant, because you don't want to just plant a very small area of a burn scar, and then hopefully that will come back within 30 years.

But those shoulder areas is probably where we would look to use existing vegetation, which is on the landscape, and plant in the shade of those areas. Gamble Oak has been, uh, shown quite a lot of promise that when we plant under Gamble Oak, we see a dramatic increase in probability of survival by about 43 percent in one of our experiments.

So we can utilize areas like that in not perfect conditions when we're planting in post fire areas, but also using that model to suggest where we might need to use that vegetation or where we might be able to plant normally in those high probability areas. Did that answer the question? 

Emile Elias: Thanks, that's great.

I appreciate the specific examples of types of trees and also thinking about the landscape and that not the entire, you know, you're not having those mother trees in that shade and so you're really looking for some very specific places where there may be higher probability of success. Yeah, so Matt, I'm wondering if you can provide us a specific example of a group or a forest that you're working with and how the people doing the planting use the information you provide.

Matt Hurteau: Yeah, so we're in early days of the NMRC effort, the New Mexico Reforestation Center effort. And, you know, this whole data driven approach to predicting where to plant, we're slowly working toward integrating that into efforts that are underway. And so right now we're using the model for an NMRC planting effort within the footprint of the Hermit's Peak Calf Canyon fire from 2022.

There's an area where We're working collaboratively with the U. S. Forest Service to plant, I believe the number is 30, 000 seedlings, but you know, we're using the model to predict where to plant in, in that case and, and the interesting thing about this is that, um, the more we do this and then monitor survival, the better predictive power the model will have because it will become more accurate as we feed it more information.

And so that's the first example where we're using the model in a something other than a research context. 

Emile Elias: Excellent. Chris, did you have anything to add? 

Chris Marsh: Yeah, so I guess one of the things that we One of our main collaborators, Owen Burney, is a big fan of nucleation plantings, where effectively you would plant in a good area to replant within the area of a burn area at quite high densities of seedlings.

The idea being that though, as those trees grow, they offer shade to each other and you're creating a sort of nursery environment, which would, once those trees mature, provide seed rain, like natural seed rain, for the rest of the landscape to regenerate. As chance would have it, when we are applying the model to these areas, we see these small pockets of high seedling survival probability on the landscape, which would make perfect areas for those type of nucleation planting.

And that project that Matt was just mentioning there is one of the main experimental aspects that we're doing is planting at different densities to see when the competition between those trees kind of overrides the positive impacts. So with that experiment we're planting at multiple different densities to see where that sort of, you know, Where we're getting that in that facilitator aspect to that high density planning becomes potentially an inhibitor.

And if we can find that sweet spot, which would be a product of how the elevation of that landscape and how much water is in the system. If we can tailor that, then we can have a big impact with nucleation planning in high severity burn areas with relatively little effort, but they will also provide a natural seed rain to restore the rest of the landscape through time.

Emile Elias: Excellent, thanks. Yeah, that wasn't something that I had thought about and so that's an interesting extra piece of research that you're looking at to identify really how to optimize the survival rates out there. Yeah, so my next question relates to the modeling and climate change. So how are you incorporating climate change information into your modeling efforts?

And specifically, are you suggesting plants for conditions projected sometime in the future, like mid century or the end of the century, or are you recommending trees that you think would do well in specific locations now, based on current conditions, or maybe a combination of both? And we'll start with you, Matt.

Matt Hurteau: Yeah, so that's a great question. I think the thing to keep in mind is that if you look at the range of climatic conditions that a tree can tolerate or grow and survive in, they're different depending on the age of that tree, right? And so the range of conditions that an adult tree can survive and grow in is much larger than a juvenile, than a seedling.

And so what, the way we're thinking about it at this point is that if we identify locations where seedlings are more likely to survive now, those are also locations when they grow into more mature trees that they'll be able to persist into the future. And, you know, there, there's no, there's no prediction, uh, in the model that accounts for future climate because we're really looking at that very narrow set of conditions that a seedling can survive in now because that's when we're focused on.

And then we're just making the assumption that if it, if it survives and grows there now, it will be suitable in the future. That being said, one of the things that we do discuss and that we have discussed with land managers and have done some work on, uh, but need to do more work on is the idea of what to plant in, in kind of the, what's often referred to as the trailing edge of a species range.

And so you can think of a widespread species like Ponderosa pine. And here in the southwest, it might grow as low as, you know, 4, 500 to 5, 000 feet elevation, depending on what the topography is like, and, you know, those are not, that elevation is going to become hotter and drier pretty quickly in the, you know, it already is hotter and drier than it has been, and it's going to be even hotter and drier in the future.

And so, We're unlikely to get Ponderosa Pine to re establish in a place like that at this point, and even if we did get a few seedlings to survive, it's going to quickly get to the point where it's not great growing environment. And so, one of the things that we need to consider is, you know, what other species do we have in terms of the suite of species that are native to the Southwest that are capable of surviving in those conditions?

And I think that one of the things we need to start doing now is having conversations about what are our level of, what's our level of tolerance for moving species around on the landscape? And it's not just locally moving things up in elevation, you know, the southwest is really cool because you get down into the Sky Islands and in southwestern New Mexico, southeastern Arizona, and you're coming up to the northern end of the range of a whole bunch of species that the main part of their range is in Mexico, right? 

And there's a whole lot of, you know, genetic material there that may be more suitable for our air, you know, our landscapes in the future. And so we, we need to start doing work to understand the potential for moving those species around and what impacts that might have on our landscapes and our ecosystems.

Emile Elias: To add on to that, I'm curious, Do you know of anybody trying to do that, trying to think about, you know, bringing species from Mexico or from other climates? It seems like there may be a lot of hoops to jump through, uh, to, to try to do that along with a lot of, um, ecological thoughts about doing that. So I'm curious, and Matt, this is a question for you, is that happening that you're aware of?

Matt Hurteau: Yeah, so there's already been some work that's been implemented looking at, you know, seed source and from a widely dispersed species getting seeds from different populations. In fact, uh, Owen Burney at NMSU has an experiment like a provenance trial where they've got seeds from, you know, the southern end of the range of ponderosa pine in the southwest all the way up into Canada and they've planted them together.

And are looking at how those trees perform in the climate of at Mora at the John T. Harrington Forestry Research Center. We've done some work where we've moved species up and down in elevation seedlings to look at survival and growth. And, you know, those are all, uh, native seeds, native seedlings from that exact geographic area.

But one of the things that I think is interesting, and this is an area where we do need to do some more work and we're starting to think about that experimentally, is there's a lot of those species that the main part of their range is in Mexico, but they do occur in the U. S., right? Like a great example is Chihuahua Pine, and it re sprouts after, you know, it gets top killed by fire and it re sprouts, and that might just be a species that is dynamite in a more flammable future, right?

And so you know, there's, I don't think of this so much as like, you know, totally modifying an ecosystem, this is really more about, okay, we have an ecosystem now where we have woody cover, you know, we've got trees, and there are a certain amount of canopy cover and stuff like that, and there's a whole bunch of animal species that are dependent on that, and a whole bunch of ecosystem process that's dependent on that.

And maybe the main player, you know, a particular tree species is no longer capable of occupying that role, but we've got another tree species that grows in the area, not that far away, that can occupy that role. And, you know, if you start to think about it that way from an ecosystem function perspective.

It’s a lot less sticky I think in terms of, you know, how we go about doing these things. 

Emile Elias: Excellent, thanks. Yeah, that's a great example. So Chris, I'm curious about if you rely on information from people who grow the trees. We spoke with Owen Burney and you've mentioned him here in this conversation. And so, does available seed stock or even seeds that are out there factor into your modeling in any way or the work that you're doing?

Chris Marsh:  Yeah, so this first round of modeling that we've done, we've used locally available seed sources under the genetic guidelines set down by the USDA. Ideally, we would produce multiple models with different monitoring data, survival data based on planted seedlings across a wide range of the Southwest to have different layers of a model, which will also help us make decisions on what tree is best placed in a certain area.

And largely we've been looking at Ponderosa Pine because that's the most commonly planted and probably most common forest type that has been at the heart of a lot of large scale wildfire in the Southwest. But we could definitely expand out our modeling approach to include different species at different elevation ranges, including some stuff that we might migrate a bit further north as time goes on.

But Owen's also been doing quite like a lot of interesting work with drought conditioning of seedlings, where he sort of reduces the amount of water that they receive in the greenhouse, which can make them a bit hardier when they're outplanted. And that could make a big improvement. When we're looking at a mean rate of survival of 25 percent of planted seedlings, even if we would bump that up by another 10 percent by having experimentally drought stressed seedlings that we're putting onto the ground, that could make a big impact in a relatively short period of time.

But yeah, that's definitely something we would like to look at into the future by expanding the modeling process out to different seed sources. Cause we could go really go down to a granular level and different species as well. 

Emile Elias: Matt, anything to add? 

Matt Hurteau:  Yeah, I, I think, I think the thing too, that we need to, you know, mention here is that as we were working on building the model, one of the things is there's, there's, there is all this information that's could be available, right.

As it, as it starts to come into play. And one of the things that we're interested in doing in the, in the way we structured the workflow is really about being able to take in all kinds of different information about trees, right, and understand, you know, where, or use that information to make new predictions.

And so that might be seed source. It might be, you know, the, the drought stress conditioning that Chris was talking about. So the, the modeling framework is flexible enough that we can give it that information and update predictions. Another thing that is really important that we have What we've been discussing and also need to get out experimentally is, is basically finding places where on the landscape where there are mature trees that are currently capable of reproducing in those places, but the conditions are bad, right?

So it's a droughty hot place, but yet there are trees that are regenerating and that might give us an idea of, you know, that there's a potential that there's a, uh, Some genetics that that population has that allows it to grow in that environment. And so, you know, all of these different opportunities to improve replanting outcomes.

Uh, exist. And then as we tap into those, we can assimilate that into the model. 

Emile Elias: Excellent. Thanks. So thinking about the trees, the trees that do well in really harsh places, especially, and, um, seed, getting seeds from there. And so that's, you know, we talked about factors related to climate, factors related to trees.

Do you also think about external factors, things like land use change or power lines or lightning strikes. And maybe that's not what you think about in your modeling. Maybe that's what people doing the planting think about. But, but Chris, we'll start with you. Is that something that, where in the process do these thoughts come in?

Chris Marsh: So in terms of land use change, one of the biggest challenges we have in older burn areas is It's basically a natural land use change where it used to be a forested system and now it supports a shrub system or grasslands. And that can be a real challenge, specifically with grasses when we've tried to plant in areas like the Jemez Mountains.

We have a lot of competition from grasses, which are in direct competition with the planted seedlings, that root layer, which only goes down, say, six inches for those planted seedlings, it's in direct competition for the moisture availability with different grass sources. And we see the same thing kind of with New Mexico locust.

One of the sort of we call nurse shrubs effects that we do see with the Gambo locust because they're using a water source which is much deeper, so they might not be in direct competition for water use but in terms of land use, that's mainly how I view it from a reforestation perspective. Power lines, maybe Matt's better suited to answer that question in terms of potential hazards, I guess.

Matt Hurteau: Yeah, so I think that it's not so much power lines or lightning strikes. I mean, those are all, you know, potential ignition sources, right? And abandoned campfires and, and the like, but we have had a lot of discussions about the risk that subsequent fires pose to these planted seedlings, right? So we basically, when you take, for example, a ponderosa pine tree, you know, we need 20, 25, 30 years for that tree to get large enough that it can survive a surface fire.

When they're growing in these kind of harsh environments, they're not growing real fast. And, and for a, for a ponderosa pine, which, you know, is a fire resistant tree, it's got to get large enough that its branches are up in the air, high enough above flame lengths, and its bark has had a chance to grow thick enough that the heat doesn't kill it.

You know, and, and we already see examples of fire basically undoing some of the reforestation good that we're trying to do. So we had, part one of our experiments was planted out in the footprint of the 2011 Las Conchas fire. And then, you know, we had this, so this is what we initially built the model from are these data that we planted, these trees that we planted experimentally.

And we had this really, you know, this really sweet spot on the landscape where there was something about the slope, you know, that the aspect, so it was more northerly facing. The slope position so that where we planted just happened to be maybe three fourths of the way down this slope and so everything about this site we had really high survival and then the crazy thing was how quickly these seedlings grew right? So the the grand champion after four years was like three and a half feet tall and you know. So it was like this I was like this is this is a tree I'm going to keep measuring every every year you know I'm going to make a special trip out here for this little patch of trees.

And then the 2022 Cerro Palado fire in the Jemez started, and it ended up getting pushed into the footprint of Los Conchas fire and, you know, it burned up that patch, right? And so it's one of those things where even if you've got great survival and great growth, if you have subsequent fire too soon, that's problematic.

The other piece that I think is really important to managing for, you know, this subsequent fire hazard to these seedlings is Chris mentioned earlier, the idea of nucleation planting. So planting in places where you're likely to have high survival and then those trees will grow up and become the seed source.

And they'll kind of propagate out from there. And so one of the things that we need to think about as we work to reforest these landscapes is exactly that, identifying the places where they're survived, thinking about how continuous the vegetation is, because you need continuous vegetation to support fire spread.

So ideally we have some, some separation amongst these clumps of trees, these nucleation sites that we're planting on the landscape. And then the other thing, you know, we have to just accept and expect that we're going to lose some of these and be prepared to make the investment to go back in and replant them afterward if we're having good survival. 

You know, that's kind of a reactive thing. The other piece, and I've talked to folks in that work in fire management about this is we probably need to have a conversation at some point about what kind of work we're going to do on the front end to keep the seedlings from burning.

And so that might be something along the lines of site prep and dealing with fuels before we plant. So if you've got big down logs, maybe we've got to burn beforehand. Maybe they need to be hauled out if it's a you know, one acre nucleation site. And then the other piece is what to do when you have an active fire.

Do we treat a nucleation site like a, you know, a high value spot on the landscape that we then allocate firefighting resources to to dig line around to keep it from burning? I mean, these are conversations that you know, that, that we need to have in terms of how do we want to allocate resources to ensuring the establishment of trees in these areas.

Emile Elias: Excellent. Thank you. Yeah, some deeper management questions that are coming up related to the science and, and some need for some conversations about how those two might interact with each other. So I have a final question about your modeling, and it relates to the limitations. So what are the limitations of what you can model or predict?

And I think we've talked about some of those, or in other words, what can you not predict, that perhaps you wish you could predict? And this is a huge question, I know, as a former modeler. But I'm going to throw it out there anyway and see what comes back. And Chris, we'll start with you. 

Chris Marsh: Okay, thanks.

Matt, you're going to have to correct me if I'm wrong on getting this quote right, but typically we approach modeling as, uh, all, all models are wrong and some of them are useful. This one is useful in a very limited scope. And currently it is useful for predicting the probability survival of planted ponderosa pine seedlings within the Las Conchas burn area.

We can apply the same process to guide where we would plant in different burn areas, but our main limitation at the moment is monitoring data. We don't have as much monitoring data as we do planning efforts in the Southwest. The monitoring isn't quite up to scratch in the way that we would like to use it in our modeling processes.

We would like a really precise GPS location of each individual planted tree, and we would like to know if it's alive or dead every year, and how much it's grown every year. And that's really hard to do at scale. We can do that in little pockets, in an experimental fashion, and we're slowly increasing like the order of magnitude that we do that monitoring of and growth and survival monitoring over periods of time. We started off with a hundred trees in the Las Conchas and then a thousand and now we're up to ten thousand across three different burn areas at different elevations.

But ideally we'd like to add another zero to that in the next effort of model version 2.0 and roll it across different species and A lot of different areas as well, because we've primarily been looking at northern New Mexico, but ideally the same modeling process could be applied to the four corner states, if we had that monitoring data. So it's severely limited at the moment, but version two, which is kind of on the horizon, we could model growth pretty well, I think.

Well, our current effort is those 10, 000 seedlings that we're monitoring every year, survival and growth across three different burn areas. So growth, we could have a decent idea of in a year or so. Once we've done the analysis and get all the data in, but we've also, in each of those burn areas where we've got those 10,000 seedlings put in, we've got microclimate data loggers.

So they all give us an hourly measure of soil moisture, air temperature, relative humidity and light levels for those burn areas, which are relatively small sites. But we're hoping we can use that data to make a dynamic model of microclimate across those burn sites so we could downscale regional weather station data to the seedling level, and that will give us an idea of what influences that seedling growth and survival at the, at a small scale through time. And then we could use that data to downscale regional weather station data across much wider areas and apply it to areas which are further away from our planning sites.

So, whilst this is a first step, it's relatively basic, the current model that we've, we've published on, but we're hoping that we can diversify that across geographical ranges, different species, different seed sources, even, and take into consideration weather effects, which is something that we can't do right now.

So once we have those microclimate data, we'll be able to see how much influence a very dry summer has on seedling survival and growth, or how a particularly wet year might give us a bumper, you know, growth spur for all of our seedlings and the primary reason behind that, whilst it would be interesting to look at the long term survival and growth of those seedlings, is we could predict when to plant.

Across the southwest, that's a big problem. We have lots of different views if it's best to plant in the spring, the summer, or even the fall, depending if it's going to be a dry or a wet year, no one really knows when to plant. And so the next sort of issue that we're trying to tackle with our microclimate modeling is if we can have a good, well statistically accurate way of telling land managers when is best to plant their seedlings, depending on the region. 

With some of the conversations that we've had with land managers that seems to be one of the main causes of anxiety is, are you going to plant and have, you know, three weeks without rain, but if you can look on past climate data and give statistically this is the best point in the year to plant after wildfire, then, then it can give some sort of reassurance and potentially spur some reforestation action on those post wildfire sites.

Emile Elias: Excellent, thanks. Matt. Anything you'd like to add? 

Matt Hurteau: Yeah, I think Chris laid out a nice research program that we hope to pursue to improve the model. But I think the, the point that kind of got lost in there is that. Well, this is really a 1. 0 version, you know, and that we're working to improve things.

We can do a lot better with this model than you can just by going out and planting by chance. Right. And so we've got about 63 percent accuracy in our predictions. And, you know, as we assimilate more data, as we get more monitoring data, like Chris was saying, we'll improve, you know, and, and we would expect when you move to a new area, that would, accuracy would decrease some until you get data from that place.

But I think that the big thing is, is the model is not, it's relying on characteristics about the environment to predict a particular species that we've got a fair amount of planting data on. And so it's not taking, you know, weather predictions or anything like that into account. It's just topography.

It's just an understanding of that environment. And is actually, those are pretty good predictors because they modify that climatic condition around there so you can find those wetter, cooler spots to plant trees. 

Emile Elias: Excellent. So Matt, what has surprised you most about your work over the years? 

Matt Hurteau: Wow. What surprised me most about this? I would say that one of the things that is, I guess, I'm most surprised by the fact that, you know, if we, if we look back over the history of our work in reforestation, you see that there were, there were big investments in understanding how to grow seedlings after, in most cases cutting, right after harvest.

So if you go back to the literature from the 1950s, you see a lot of this post logging management kind of stuff and planting and post disturbance management and planting. And, and, you know, that, that work continued for a while. And with it, there was also a pretty significant investment by U. S. taxpayers in establishing nurseries to grow seedlings to do that work. And then, you know, we basically stopped investing in that, right? And so we have this period in the eighties and nineties and early two thousands where, you know, it's really a lull. 

And I think the thing that is most surprising to me is that in this period, particularly starting about the late nineties in the Southwest and, and into the, the two thousands, that we didn't see a big recognition by everybody that getting trees to establish in these post fire environments was going to be a real problem, you know, and that we're kind of we're coming late to the party and, and, you know, that's us too. I mean, we're none of us have great crystal balls. Right. But it was really unfortunate to me that we weren't thinking about that and making those investments so that we're better positioned today.

Or it could be better positioned today than we are. 

Emile Elias: Thanks, Matt. Chris, same question to you. What has surprised you most in your work? 

Chris Marsh: This is probably a slightly strange answer to this question. But just how extreme the thermal environments in post wildfire areas are. We've been measuring near ground temperatures in like some of the hottest months of 53 degrees Celsius, which is over 100 degrees Fahrenheit.

And that's the environment that we're asking these seedlings to grow in. And the difference between you know, 50 degrees, where you would definitely, you know, the seedling's definitely not going to survive there, and a sort of more comfortable 40 degrees, sort of, what's that, 90 degrees ish Fahrenheit, can be one meter apart, and it can be the shade from a shrub, which causes this much less climactically extreme environment.

Thank you, Matt. So 53 degrees Celsius is 127 degrees Fahrenheit, so really, really hot temperatures, and that's in these post wildfire burn areas once they've lofted all those canopies, and it's not, it's not crazy to think about that you would struggle to establish trees in that environment, but finding those little pockets is quite challenging, and it takes quite a lot of instrumentation to monitor those temperatures over long periods of time, and to understand the spatial organization of those climactic refugia which may exist on those burn areas is, it's quite involved and not necessarily totally intuitive.

Like, you can go to those landscapes and be like, oh nothing will grow here. But there is a few, a few pockets within even those moonscapes like we see in the Jemez Mountains and parts of the areas in the Hermit’s Peak Calf Canyon. So it can be really demoralizing when you look at those vast swathes of burn area with just dead trees in the landscape.

But once you've sort of seen a bit of the modeling output and understand how existing vegetation can act as facilitators to those seedlings, you can look on those landscapes in quite an optimistic way. Like, we can do this. We have the, you know, what's the million dollar man thing? We have, we have the technology.

We can, um, that, that's something we can really tackle. So that sort of switch around from being really pessimistic to something that's achievable with fairly limited resources, I guess that's been most surprising to me. 

Emile Elias:  So we always like to ask the experts that we get to talk with on, in these conversations, what gives them hope?

And Chris, I'm going to start with you because I feel like you are already heading in that direction. 

Chris Marsh: Okay, New Mexico is a strange place. I'm not from here originally, as you might be able to tell by my accent. I'm from England. But, um, so it's got a really low population density, but there's a few very passionate people with really good knowledge sets.

And they can tackle this reforestation program from multiple different angles using all of their past experience, and their ambition and the tools that are given to them. And it's genuinely inspiring how a lot of people have come together from across different institutions, different universities, federal and state agencies, and have tried to tackle the reforestation issue as a collective and for the greater good of a state. 

It will do no one person any good to plant a bunch of trees which maybe in 30 or 40 years will resemble the forests that they were designed to regenerate or restore. Most of the people who were involved in the reforestation of New Mexico in those burn areas won't be alive to see that and it is just through an understanding of the ecosystem services and societal good, which are a big motivator for those people.

And it's genuinely inspiring. And it's only when you have a bunch of people who are mid career, who are looking at that progress. And then you introduce some people who are early career to that and say, see the enthusiasm that they have to carry on that work and expand the horizons, which, it, yeah, is, is genuinely inspiring and it gives me hope in terms that it's sustainable, right.

And it's not just monetary resources, but it's people, people will make it sustainable and, yeah, that's pretty cool. 

Emile Elias:  Excellent. Thanks, Chris. Matt, same question to you. What gives you hope? 

Matt Hurteau: Yeah, I'm going to build on what Chris said. So, when I first moved to New Mexico, Craig Allen took me out in the footprint of the Las Conchas Fire and remarked about how often the agencies had tried to plant seedlings out there and, you know, you could basically count on a couple of hands how many of them had lived.

And all I could think is, we've got to be able to do better than this, right? And, and here we are nine years later, you know, and, and we've done some work, others have done some work and we're able to do much better, you know? And so the thing is like Chris was saying is that, you know, we're, we're able to bring, a range of different expertises to bear on this problem.

And, in a very collaborative manner. It's not a lot of times science is you're racing to, you know, get the answer or whatever. And there's a lot of competition in science and that's good. But in this particular case, you've got a lot of people who have complementary expertise, working to tackle a complex problem and making headway on that problem.

And I think that's the, the thing that gives me a lot of hope. And it also, I mean, it's, you know, it's, it's what makes me like coming to work in the morning, you know, it's the fact that we get to work on important stuff with other people who really care about working on important stuff. And the net effect as someone who loves trees will be that there are more trees on these landscapes.

Emile Elias: Excellent. So if people remember just one thing from our conversation today, what's the one thing that you'd like them to remember? And Chris, we'll start with you. 

Chris Marsh: That's a real tough one. All I've got rolling around in my head is that Sesame Street thing. Plants need water, man. And I'm like, yeah, that's, that's it.

Matt, have at it. 

Matt Hurteau: All right. You know, I think the one thing that I'd really like people to remember is that we are wholly dependent on these forested ecosystems. For a whole host of things, right? Like, I turn on the tap at my house and that water came from a forested watershed, right? It came down the Rio Grande River from the forested watersheds in southern Colorado and northern New Mexico.

And, you know, that's the lifeblood of this region, right? Is the water flowing through the river. And anything we can do to keep tree cover on the landscape benefits these sorts of things, right? So it, it keeps, for somebody like me who loves forests, it keeps forests intact and stuff like that. But even if you don't ever go outdoors, right?

It benefits you that we have forested watersheds in terms of the water flowing from your tap. And so I think that, the thing to remember is that in an ideal world we make investments to keep these forests from burning down in the first place, so we do some restoration work to decrease the chance that we have these massive tree killing wildfires, but you know it in we're still going to lose forest to wildfire.

And, you know, we need to continue to make the investment and improve on our ability to reforest these severely burned landscapes. And that's critical to the continued provision of things like water in our taps. 

Emile Elias: Chris, what's the one thing you'd like people to remember? 

Chris Marsh: Yeah, so we're talking about microclimate a lot, and how we can find places on the landscape which are slightly cooler and slightly wetter than the prevailing climate of that area, and we can plant trees in those areas.

But if we do nothing, climate change will get to a point where those microclimate refugia won't be cool enough or wet enough to sustain those planted trees. So there's a time limit to our reforestation efforts, and it may not be apparent right now. But given 10 or 20 years, we wouldn't be able to reforest those low elevation sites.

There would be large swaths of the Las Conchas burn area or the Hermit’s Peak Calf Canyon fires that we won't be able to reforest. So there should be a kind of motivating factor to this. And if we do nothing, then that decision will have already been made for us. And it will be that we can't reforest those sites.

So yeah, there's that part of it, which I think is important. Whilst there's been a lot of lack of investment in the reforestation pipeline for many years, if we continue to choose not to fund that reforestation pipeline, then we won't be able to call back on this in 50 years and change our minds.

A lot of that climactic least suitable space for planted seedlings won't exist anymore. 

Emile Elias: Matt Hurteau, Chris Marsh, thank you so much for talking with us. 

Matt Hurteau: Thank you. 

Chris Marsh: Yes, thank you.

Emile Elias: Thanks for listening to Come Rain or Shine, podcast of the USDA Southwest Climate Hub 

Sarah LeRoy: and the USGS Southwest CASC. If you liked this podcast, don't forget to rate or review it and subscribe for more great episodes. A special thanks to our production crew, Skye Aney and Reanna Burnett. If you want more information, have any questions for the speakers, or would like to offer feedback, please reach out to us via our website.