Surprising Role of Trees in the Boreal Water Cycle

Video Transcript
Download Video
Right-click and save to download

Detailed Description

Approximately 25 to 50 percent of a living tree is made up of water, depending on the species and time of year. The water stored in trees has previously been considered just a minor part of the water cycle, but a study by University of Alaska Fairbanks scientists with support from the DOI Alaska Climate Science Center shows otherwise. Their research is the first to show that the uptake of snowmelt water by deciduous trees represents a large and previously overlooked aspect of the water balance in boreal watersheds. Calculating the amount of water stored by deciduous trees is important. The area occupied by deciduous trees in the boreal forest (or snow forest) is expected to increase 1 to 15 percent by the end of this century, and the absorption of snowmelt could also then increase. Quantifying tree water storage is important for understanding hydrology, tree response to drought and the related factors of tree water use, soil moisture and climate. Watch the webinar recording to learn more about the methodology and findings from this project! This webinar was conducted as part of the Climate Change Science and Management Webinar Series held in partnership by the USGS National Climate Change and Wildlife Science Center and the FWS National Conservation Training Center. 


Date Taken:

Length: 00:36:56

Location Taken: AK, US

Video Credits

Jessie Young-Robertson and Uma Bhatt, University of Alaska Fairbanks
Elda Varela Minder, USGS National Climate Change and Wildlife Science Center


John Ossanna:  Welcome from the U.S. Fish and Wildlife Service's National Conservation Training Center in Shepherdstown, West Virginia. My name is John Ossanna and I'd like to welcome you to our webinar series held in partnership with the U.S. Geological Survey's National Climate Change and Wildlife Science Center.

Today's webinar is titled "Trees' Surprising Role in the Boreal Water Cycle," and we're excited to have Jessie Young‑Robertson and Uma Bhatt from the University of Alaska Fairbanks with us today.

Let's get started, and to start things off, please join me in welcoming Elda Varela Minder who is a Research Associate with NCCWSC who will be introducing our speakers today. Elda?

Elda Varela Minder:  Thank you, and welcome to everyone. Jessica Young‑Robertson is a Research Assistant Professor at the International Arctic Research Center at the University of Alaska Fairbanks. While she worked on her undergraduate degree in Biology from Fort Lewis College, she was a Field Technician on avian nesting studies in the Southwest U.S., Montana and Louisiana.

She then got her MS and PhD in Ecology and Evolutionary Biology with a minor in Global Change at the University of Arizona. Her research focused on the effects of changes in precipitation and vegetation on cyclical ecological processes in the Sonoran desert.

She did a postdoc at the University of Wyoming where she developed strong modeling skills while analyzing extensive soil carbon data sets from the seven North American deserts.

A second postdoc funded through the National Science Foundation led to her work in Alaska where she's started exploring how boreal forest plants and ecosystems use, move and store water. She is fascinated by the role that tree water use can play in large‑scale processes in the boreal forest.

Uma Bhatt is the chair of the Department of Atmospheric Sciences and Geophysical Institute and the Director of the Cooperative Institute for Alaska Research at the University of Alaska Fairbanks.

After completing her undergraduate degrees in Mechanical Engineering and Russian Language, she serves in the U.S. Peace Corps as a teacher in rural Kenya, where she experienced amazing weather phenomena, as well as the East Africa drought of 1983 through '85. This is what led her to become a climate scientist.

She attended the department of Atmospheric and Oceanic Sciences and the University of Wisconsin‑Madison where her MS focused on Tropical Meteorology and her PhD incorporated local climate models to study mid‑latitude air‑sea interaction.

Since joining the University of Alaska Fairbanks, she teaches and conducts research that focuses on arctic climate issues. Welcome to both of you.

Jessie Young‑Robertson:  Thank you.

Uma Bhatt:  Thank you.

Jessie:  Thanks for joining us. Today we're going to talk about a paper that was recently published in "Nature Scientific Reports" on the surprising role of trees in the boreal water cycle. I'd like to acknowledge my co‑authors, Bob Bolton, Uma Bhatt, Jordi Cristóbal, and Rick Thoman. Rick is with the National Weather Service, and the rest of us are with the University of Alaska Fairbanks.

Bob Bolton and I have worked together for a very long time. Our primary research goal has been to understand the ecological and hydrological interactions in the boreal forest. The field of ecohydrology has been fairly well developed in other ecosystems, but really not in the far north, so we've been trying to make these connections in the boreal forest.

Working with Jordi, Uma, and Rick, we are all ultimately interested in the interaction with climate, and the feedbacks. Throughout the talk, you're going to see these photographs of the boreal forest that are taken in the autumn. That's because it's the best time to see the striking differences between the different ecotones.

The Alaskan landscape is ecologically diverse and hydrologically complex. On the left‑hand side, you can see Alaska and all the various ecosystem types throughout Alaska. For this study, we were primarily interested in most of interior Alaska, including the western parts of Canada.

The dominant ecosystem types are coniferous and deciduous forests, which you can see in the yellow and green colors on the map on the right. When we talk about scaling later on in the talk, we're referring to this region.

These two main vegetation types cover a significant part of the landscape, and this area is underlain with discontinuous permafrost. What that means is essentially the permafrost that's beneath the soil surface isn't everywhere, not ubiquitous.

Thinking about the map we saw on the previous page, the deciduous‑dominated ecosystems ‑‑ this would be aspen, birch, and also some poplar ‑‑ cover about 61 percent of the landscape. The coniferous‑dominated ‑‑ black spruce, but also white spruce ‑‑ cover about 38 percent of the landscape. These are really interesting ecosystems because they're so different, and they lie right next to each other, just meters apart.

The coniferous‑dominated ecosystems, they're typically found on north‑facing slopes and valley bottoms. The soils are very wet and cold, and they're typically underlain by permafrost. The trees typically have low physiological activity.

This is in direct contrast to the deciduous vegetation. The soils are dry and pretty warm. They're typically not underlain by permafrost, and the trees have very high physiological activity relative to the coniferous trees.

When you take all of these things and put them together, you can see that these ecosystems might have a very different relationship between ecological and hydrological processes.

Here's a drawing of a typical boreal watershed with mixed vegetation. Obviously, this is a cross section.

On the left‑hand side...Let me see if I can use this pointer. That gray blob there is permafrost. You can see up top there, I'd say it's a north‑facing slope or valley bottom. It's where permafrost is found. In the top of that we have seasonally thawed soil layers. On top of that, we primarily have black spruce, coniferous vegetation, and also typically a mossy understory.

The seasonally thawed soil layer is quite thin at the beginning of the summer. Then we reach a fairly deep thaw by autumn. In a lot of the areas we study the thaw is about a meter to a meter and a half deep. These soils have a thick organic layer and then, below that, a mineral layer.

If we go across the valley ‑‑ see if I can get us there ‑‑ we see that we have quite a different ecosystem. There we have a mineral soil, very thin organic layer, there's no permafrost, and there's typically fractured bedrock.

The typical vegetation there is deciduous trees. You will see some conifers, but primarily deciduous vegetation. We're talking birch, aspen, and poplar. These are predominantly south‑facing slopes. You can see these are very different ecosystems.

What we're thinking about, and what has led Bob and I through much of our work and into this paper, is that we want to know where does the water go when it enters an ecosystem. Where is it stored? In the soil, the plants, or even deep in the soil in the aquifer?

Which compartments does it move through, and when? How are the vegetation interacting with other parts of the system to affect where water goes?

Thinking about this ecology‑hydrology interaction, a significant uncertainty in boreal forest ecohydrology, and arguably just boreal hydrology, is the role of vegetation water use in removing and storing water, thereby affecting water availability for streamflow and then ultimately affecting climate dynamics.

If we return to that drawing we had a couple of slides ago, here I have three versions of it. Essentially, it's as the system moves through the seasons. The spring snow‑melt period is the top, the growing season is the center, and autumn is the lower part.

The blue arrows depict where we think water is going. These were our initial hypotheses when thinking about the ecohydrology of these systems.

On the left‑hand side where we have permafrost and coniferous vegetation, we hypothesize that the hydrological processes are going to be dominated by what's going on in that seasonal soils thaw. It's the depth of that thaw that's going to affect where the water is going.

You can imagine moisture on the left‑hand side. When it comes into the system, it hits an icy layer. It moves laterally through the system and generates streamflow. Then as the seasons progress, as we move into autumn, that seasonally thawed layer is getting deeper and deeper. But the moisture still comes in. It reaches an icy layer and moves laterally out of the system.

If we focus on the right‑hand side of these figures, we have these deciduous forests. In the spring snow‑melt period, we always hypothesized that because there's no leaves on the trees, they're not really participating in the water cycle. The snow melts, it moves down into the soil, generates some streamflow, but largely just moves into deeper soil layers.

During the growing season when the leaves are present, we hypothesize the trees do participate. Rain comes into the system, the trees pick it up, and transpire it to the atmosphere. In the autumn, we suspected, because the leaves aren't quite as active, moisture moves into the system and just keeps moving through.

We suspect that much of the hydrological processes in these deciduous ecosystems without permafrost are probably dominated by vegetation and their activity.

If we think about what we already know about boreal tree water use ‑‑ which, by the way, is not very much ‑‑ John Neary at UAF and his team have shown that deciduous trees rely heavily on snow‑melt water. They did a complete snow exclusion on some plots and a complete rain exclusion on other plots.

They excluded snow for five years, and they showed that it basically resulted in tree death. They excluded rains for 20‑some‑odd years. It didn't kill the trees. It shifted some dynamics in the plots, but it didn't kill the trees.

Now, they didn't measure any of the actual tree water‑balances components, but this sends a pretty clear message that snow‑melt water is very important for, particularly, deciduous trees.

Add to this part, snow‑melt is the most important part of the water‑balance...this is not part of what John Neary's group showed, but this is just, generally, known to the hydrological community and this is important for our story.

What Bob and I have shown ‑‑ I'll show you in a graph in the next slide ‑‑ is that deciduous trees transpire in order of magnitude more water compared to coniferous trees, especially during dry times. This is a loud graph but on the left‑hand side, we see red lines and on the right‑hand side, we see green ones.

On the left‑hand side, these are transpiration rates from deciduous trees, and the different colored lines are from sites that are positioned differently in the landscape but they're all deciduous trees. On the right‑hand side is the transpiration rates from coniferous trees.

The real take‑home point here is that if you look at the Y‑axes on all these graphs, you can see that the deciduous trees are transpiring in order of magnitude more water than the coniferous trees. This is the same slide I showed you just two slides ago, but we added on to it. The other thing that we know is that the soils in deciduous forests, they just don't store water well.

We take all this together. We know that deciduous trees, they do rely on snow‑melt water. We know they use a lot of water. We know they live on soil that doesn't store a lot of water. So, how can these trees sustain high transpiration rates during dry periods? Are they using water they stored in their trunks, that they took up during snow melts? These are some basic questions we just don't know.

The way to think about organisms that have to function in places that are dry is that they all have to have a reservoir they drink from, whether they are their own camel or they're drinking from some other pot of water. It's clear they're probably not using the soil as a reservoir, so, maybe, they're using their trunks as their reservoir.

The primary research questions that are within the paper that we studied are: If these boreal trees are using snow‑melt water stored in their trunks, how much are they using and storing? Do they use the water over the summer to support transpiration during dry times? How do deciduous trees compare to coniferous trees?

Again, I want you to have in the back of your mind, that these organisms are trying to find a way to have a water pool that they sip from in order to support themselves during stressful times. Think of it as that pool of water is a buffering agent. Briefly, what we measured was tree‑water content, soil moisture, tree‑stand characteristics over a couple summers in an interior Alaskan boreal forest watershed.

Then, we scaled these data to landscape levels. I want to talk a little bit about plant water use. There's a lot of terms that float around that are used interchangeably, but they shouldn't be. Evapotranspiration is the water vapor loss from the surface to the atmosphere. Evaporation is the physical process of loss of water from moss, soil, and open water.

Transpiration is, actually, a physiological process that's controlled by living organisms. This is how a tree can actually affect climate, one of the ways it can. Transpiration pulls water from a storage source. So, from the trunk which is, ultimately, pulled from the soil. Evapotranspiration is the sum of these two processes. Now, transpiration is, actually, a flux.

What we looked at in the study was tree‑water content. What this is, it's the volume of water that's stored in the trunk. It's the net of water that's supplied from the roots and then, transpired to the atmosphere. It's like your checking account. Money comes in, money goes out and what's left over is your net balance in your account. And, that's what we're looking at, is the water content that's stored.

First, I'll talk about our field research that we did, and then I'll talk about our findings as it was scaled to the landscape. We did our work at Caribou‑Poker Creek Research Watershed. This is in interior Alaska. It's part of the LTER network, the Long Term Ecological Research network. We set up four sites in the watershed with mixed vegetation. We set up two sites on a north‑facing slope.

This area has permafrost with coniferous trees with a mossy under story. We set up two sites on a south‑facing slope. That's the one in the purple. This area does not have permafrost. It does have near‑surface bedrock. It has deciduous trees, primarily, aspen and birch. The two positions that we picked within each forest type is both high and low on the hill slope.

That's not, necessarily, relevant for the results we'll show, we just needed you to know that. Our approach for measuring tree‑water content was measured weekly or semi‑weekly for several summers but will only show results for 2013 and 2014. Our instruments were installed in Black Spruce on north‑facing sites and birch and aspen on south‑facing sites. We used a time‑delayed reflectometry approach.

This was, typically, used to measure soil moisture, but we went on ahead and put them in the trees. We lab calibrated the instruments using logs that we cut. We analyzed data from 2013 and 2014. We also conducted additional stand characterization measurements for the scaling, as you'll see. We measured soil moisture at two depths and additional meteorological data that we won't end up talking about.

This figure just steps you through some of the data and how it was all merged together. We developed an allometric relationship between tree height and circumference to determine tree volume and this is throughout the stand. Then, we scaled that to the stand and got stand volume, and we incorporated tree‑water content that we measured on many, many trees. We got stand‑level water volume.

Then, we had a database of snow depth and density throughout Alaska to calculate snow water equivalent and again, we focused on non‑coastal areas and not the north slope. We also got estimates of land area covered by coniferous and deciduous trees. We used a Bayesian statistical analysis approach so that we could take uncertainty from all these different data sets and propagate it throughout our calculations.

In the end, we actually have uncertainty estimates for our scaling data. We do have some assumptions for the scaling that we want to put out there. One is that the tree volumes are calculated as cylinders. Obviously, trees are not cylinders. But, we're lucky with these trees, they have a pretty significant branching pattern up top.

So, while the trunks taper, if we include all the rest of that volume, we make the assumption that it's equivalent to a cylinder. All the deciduous trees outside our research area attain the same level of wood saturation prior to leaf‑out.

What you're going to see in the next graph is these trees get really saturated. We had to make the assumption that all the trees that we're scaling to reach that same level of saturation as deciduous trees.

This is very easy to test because when they get saturated, you can just go hammer a hole in the tree and it gushes out water. That's a very easy thing to do to a lot of trees in an area. We’ve walked around and done that for several different trees in different locations and sure enough, it's pretty universal.

Also, we assumed the density of trees in our research site is representative of stand densities in other parts of Alaska. Again, non‑coastal areas. This one's a bit tricky. We know any ecosystem you go into, it's going to be heterogeneous and we feel fairly confident that our watershed is representative of many parts of the boreal forests. So, that's one of the assumptions.

The final assumption is that the trees are taking up snow‑melt water, not another source of water, prior to leaf‑out. And, the trees are taking up the current year's snow‑melt water rather than stored moisture. There's a few reasons why we're comfortable with this assumption. One is John Neary's study showing their reliance in taking up of snow‑melt water.

Another thing that we rely on is that these soils don't actually store a lot of water. They have to take up whatever water's immediately available. It's that snow‑melt water that actually has enough volume to create situations that we're able to see in the trees.

Now, onto the data. These are the data from our field site. They're both graphs of soil moisture at five centimeters. For each centimeter of data, I went on ahead and removed it if it wasn’t terribly relevant.

On the left‑hand side, we have the deciduous ecosystem full of moisture. On the right‑hand side, we have the coniferous system. I’m showing it for two years 2013 in the black, and 2014 in the green. The asterisks in each case show the end of snow‑melt.

What we see is soil moisture starts out quite low. Then it actually saturates the soil in both cases. That's the snow‑melt. Then we see a drop‑off in soil moisture after that period.

Now, what I'm showing you is the tree water content. It's a volumetric water content. The circles are the deciduous trees. The triangles are the coniferous trees. I've indicated on the X‑axis when leaf‑out occurred in each year. Each of those points actually do have error bars that data is nicely constrained.

What we see is that if we focus on the deciduous trees, they start out at about 35 to 40 percent water content. Then they start to get saturated. They increase in water content to a point until they leaf‑out, and then water content declines.

So what you can see in 2014 is the water content actually stayed pretty high. That's because 2014 was actually one of the wettest years on record. They are taking up rainfall later on there.

What we see with the coniferous trees and the triangles is they actually do take up water. They're early in the season, and then they stay pretty constant about 50 percent water content. We can already see some pretty significant differences between these two tree types.

The deciduous ones get saturated and they stay very, very wet. Coniferous ones, they stay at about 50 percent water content. They don't seem to be responding that much to variability in climate, which is what we think is driving the variability in the water content data for the deciduous trees.

What we see for the deciduous trees is they reach saturation before they even leaf‑out. They're moving that water before the leaves are on the trees. We'll follow up with this, but you'll also notice that 10 to 20 percent of the stored water is transpired to the atmosphere in the first one to two weeks after leaf‑out.

If you see that ‑‑ I can't get the arrow to work. You see this kind of peak time in both years, they transpire quite a bit of water in the first one to two weeks.

This is two graphs from the prior slide. The top one is soil moisture from the deciduous system. The bottom one is water content from both systems. You'll notice that I put those two red lines in there, and what they indicate is where the peak of the tree water content occurs for each year. I lined it up with the slow moisture graph above it. So, what you can see is slow moisture increases quite a bit.

At this point, we know that the trees are taking up the snow‑melt water. We went on ahead, and did is calculated the percent snow‑melt water actually taken up within each of these ecosystem types.

On the left‑hand side, we have the deciduous ecosystem. On the right‑hand side, we have the coniferous ones for both years, and what we see ‑‑ the error bars are actually the 95 percent credible intervals.

What we see is that the deciduous trees are taking up 21 to 25 percent of the snow‑melt water within their given ecosystem. The coniferous trees are taking up less than one percent. We found this to be pretty mind‑blowing. It's a really significant difference for one thing, but the other part of this is that this has so far gone unquantified.

Next slide. We should have the graph up of the landscape scaling results. On the Y‑axis, we have the tree water in cubic meters, and this is the volume. The seasonal pattern will look the same as the field data that we collected, but what's interesting here is the sheer volume of water that we're talking about.

So again, the triangles are the coniferous trees, and I expanded that graph below so that you could see what the pattern is. The volume is tremendous. If we go to the next slide, again, reiterating, the plants are transpiring about 10 percent of the snow‑melt water in the one to two weeks after leaf‑out. That's about 0.4 to 2.2 billion cubic meters of water.

From that period until the mid‑summer, that's an additional 2 to 5.2 billion cubic meters of water. This is a tremendous amount of water. You'll see that especially when we go to the next slide.

This slide should be snow‑melt water uptake at the maximum water content. What you saw in this prior graph, you saw a peak in the water content. This is the volume of water at that peak.

Again, we can see huge differences between deciduous and coniferous trees. This is one of our scaling numbers. What we see is the deciduous trees are taking up at that peak, they have about 17 to about 21 billion cubic meters of snow‑melt water that they've taken up. The coniferous is about 32 to 33 million cubic meters.

This is a pretty vast difference between these two tree types. In part because the deciduous trees cover more land area, but really this is because they take up and move a whole lot more water.

Next slide. The next question that we ask with this analysis is how can disturbance change the story? We know that across the globe, basically, we're seeing vegetation change, all sorts of things, including shrub encroachment. What we're seeing up here is actually a deciduous tree expansion.

If we move into the next slide, we see that deciduous trees and shrubs, they cover about 61 percent of the vegetated Alaskan landscape. They're expanding in area. If we go to the next slide, the landscape level phenomenon. It's actually related to disturbance.

If we have disturbance to increase deciduous plant cover. Here, I have just some things shown. We have thermokarst, hillside detachment, active layer destabilization, fire. On the north slope, we even have shrub encroachment related to thawing ice ledges, and warming soil.

One thing a lot of these disturbances have in common is that they affect the soil thermal and moisture regime. It creates a really nice environment for deciduous vegetation.

Fire's a big one. Fire's a really important process in the boreal forest. It's predicted to increase in intensity and extent. What occurs after fire, almost unanimously, is that we get ‑‑ or ubiquitously, I should say, is that we get these deciduous saplings. In the bottom right‑hand corner, it's dense with aspen and birch saplings. That's a major agent of change.

If we go to the next slide, we asked this question with our data and we said "What if we get an increase in deciduous cover by 1 percent, 5 percent, 10 or 15 percent." These are well within the possibility associated with wildfire, and the changes that they can cause. Our calculations are that it could increase the snow‑melt water uptake by 0.3 to 3 billion additional cubic meters of water.

Next slide. We're onto our conclusions. I think we've shown here that the deciduous trees in the boreal forest, they play a huge role in the water cycle. This was previously unquantified. Tree water storage is typically ignored in hydrology models. It's actually usually part of a big error term.

Snow‑melt, at least in the deciduous systems, is typically assumed to primarily recharge ground water, and what we're showing here is we need to rethink that, at least in the boreal forest where there's no permafrost. We have deciduous vegetation.

We showed that the trees take up 25 percent of the snow‑melt water within their ecosystems. This is about 15 to 20 billion cubic meters of water. This is actually equivalent to 8.7 to 10 percent of the Yukon River's discharge.

It's a large amount of water. They transpire 10 to 20 percent, or 0.4 to 2.2 billion cubic meters of water in the one to two weeks after leaf‑out. They also transpire an additional two to five billion cubic meters of water between leaf‑out in the mid‑summer period.

It appears that they're utilizing this stored snow‑melt water and the chunks over the summer. Rain does not appear to be as important as snow. We also show that the increased distribution of deciduous vegetation will likely increase the amount of snow‑melt water used by boreal forests.

There's still a lot more work to do. We have some ongoing work now to explore this deciduous water use story a bit more. We can see that, at the tip of the iceberg, this is pretty significant. Of course, if they're transpiring that much water to the atmosphere, especially in such a short amount of time right after leaf‑out. We have to ask, "What is the impact on climate?"

If we go to the next slide, this is right before I hand it off to Uma. Transpiration in climate. On the left‑hand side of this slide, we see some figures from Desheko. They did a meta‑analysis of transpiration from as many sites as they could around the globe. Relative to evaporation, transpiration is fairly significant on the global scale.

If you look at the top panel of that figure of the globe with a lot of the blue shading, you see that transpiration can make up 90 percent of the total evapotranspiration's flux. This is extremely significant. The vegetation are playing a really big role, possibly in climate.

Transpiration can impact climate. A good case study is the Amazon. Evapotranspiration can recycle 20 to 35 percent of the rainfall. Essentially, rain falls on the system, the plants take it up. It also gets evaporated from surface waters. It goes back into the atmosphere, and then it rains back down. A significant part of this is transpiration.

Transpiration can impact climate anywhere plants flux a lot, particularly if its a dry climate, or drier time of the year. I'm going to hand it off to Uma, and if you could go to the next slide, please.

Uma:  We're going to look at some of our baseline calculations of how these large amounts of moisture can potentially impact the atmosphere. The two plots that I've shown here, the dark green line showed hourly temperature at Fairbanks airport for four days. I've highlighted where greenup occurred.

The left plot is 2013, and the right plot is 2014. In light green, I'm showing the dew point temperature. Dew point temperature tells us how much moisture there is in the atmosphere, the temperature at which all the moisture would condense. The higher the dew point, the more moisture there is.

One thing the weather service folks have noticed is as soon as things green up, the dew point goes up. It's a little bit hard to tease from this hourly data, but it shows up really nicely in 2013. It doesn't show up quite as well in 2014 because it occurred, actually, much earlier when it was generally cooler, and there wasn't as strong a diurnal cycle in the temperatures.

One of the things folks noticed is there's puffy clouds soon after green up. We're seeing, potentially, that our hypothesis is this is linked to green up and this respiration, now that we know what these numbers are.

When you put moisture into the atmosphere, you're putting energy into the atmosphere, because as those moist air parcels move around, rise up in the atmosphere and they cool, that water condenses. That releases heat into the atmosphere. That heat can then promote convective activities.

Rising motion, we get our beautiful convective clouds. On the next slide, these convective clouds have implications for the climate. This is a seasonal cycle from January to December of a long‑term climatology of precipitable water. It's from one data source, the NCEP/NCAR Reanalysis. It's a combination of model and observations.

What this shows is how much water is there in a column above a box over Alaska. Not exactly the box Jessie used, but a box over the boreal forest. I've highlighted here May, which is the time of year when we typically have greenup.

The values of all the moisture in the column, pretty much if you scale it, adds up to about eight billion meters cubed of water. You can see that the values that Jessie showed of 0.4 to 2.4 billion cubic meters being released at green up, and then throughout the summer when it peaks, 2 to 5 billion.

These are a substantial portion of this eight billion in May, and roughly twice as much during the peak of the season. Again, this moisture that's getting injected in the springtime, depending on when it gets injected, the atmosphere is relatively dry at that time of year, so it's not likely to really lead to big thunderstorms.

When we think about the climate warming, and if you shift this seasonal cycle a bit, if there's a generally warmer climate, and when the plants greenup, there's the potential to start convective activity. The reason the convective activity feeds back, potentially, onto the vegetation is that convective activity is associated with lightning. It poses an ignition risk for wildland fire.

Having these observations really highlights a potential for a rich interaction between the vegetation and the climate now, as well as, as it changes. Thank you for your attention.

Jessie:  Thank you.

John:  Thank you, Jessie and Uma. At this time, if you guys have any questions, feel free to enter them into the chat box.

Toni:  This is Toni Lyn. Sorry, I'm in a noisy place. That was great. I thought maybe I could push you a little bit further about how this might be used.

Not just for how the results might be used, not just for informational purposes, for identifying places, but potentially on this cutting‑edge that Jessica Lundquist’s group is leading us all into thinking about - patching up forests and maybe keeping heterogeneity of moisture on the lands, or forcing heterogeneity of moisture on the landscape. Is that something that you might speak to? [laughs]

Jessie:  I don't know much about it. Can you elaborate a little and we can give it a try?

Toni:  Just to say, do you think that this could maybe lead to some adaptation action of how you might manage the forest differently? In trying to encourage different forest types, given the information of how these trees hold moisture in different seasons.

Jessie:  That's a good question. What's interesting, to back up a little bit, is there's not a lot of forest management throughout interior Alaska, in part because we don't have much of a forestry industry.

In terms of thinking hypothetically about it, one thing that we can kind of speak to is that ‑‑ and this is a question that we've gotten ‑‑ if we have more deciduous vegetation and they add to the convection of the atmosphere, we might have more lightening strikes and wildfire.

What's interesting feedback is that deciduous vegetation are very, very wet. They're actually hard to burn, and so that could change the fire dynamics within the boreal forest. That's not human management, but it's potentially a different trajectory that the forest could be on if we get more fires. That we might actually get less fires if we get a vegetation change.

In terms of management, this isn't directly forced management, it's more thinking about using engineered ecosystems for human purposes. I have heard of people trying to use deciduous vegetation to help dry the soil in certain places.

One use of that was, a group was looking at clay‑capped landfills or something, and it was getting too wet. They were planting deciduous vegetation to help dry the soil. In terms of managing soil moisture, like in a human‑engineered system, we could use deciduous vegetation.

In terms of having these two forest types relative to each other on the landscape, I don't have a direct answer for that one. One thing that is interesting to think about, at least their proximity next to each other in the landscape heterogeneity moisture and vegetation type, is that I get the feeling that the black spruce forests tend to have a lot of sensible‑heat flux.

Deciduous forests tend to have more latent heat flux. When you put those two forest types together, they can at least affect the atmosphere because sensible heat could provide lift for the latent heat, or for the moisture moving to the atmosphere. It's a roundabout answer with a lot of different components to it, but hopefully, that helps a little bit. [laughs]

Toni:  Yeah, thanks.

John:  If there aren't any more questions, I'd just like to thank everyone here for participating. Once again, thank you to Jessie and Uma for a great presentation.

Jessie:  Thank you.

Uma:  Thank you.