 All right, I'll get started. Greetings everyone and thank you for attending this month's science seminar presented by the NSF's National Ecological Observatory Network, which is operated by Battelle. Our goal with this monthly series of talks is to build community among researchers at the intersection of ecology, environmental science and neon. We are so excited to have Aaron Rooney of the University of Tennessee Knoxville here to present with us today. But before we do, I'll go over a few logistics. So we have enabled optional automated closed captioning for today's seminar. So if you'd like to use that feature, you can find the CC button on your zoom menu. The webinar will consist of a presentation followed by a Q&A session. As you think of questions during the talk, please add them to the Q&A box. We do also have a chat box. You're welcome to just make general comments, share resources, make sure you select chat to everyone if you want to do that. But try to think, you know, put your actual questions to the speaker in the Q&A if you can and no big deal if those get confused. And there's also an opportunity to raise your hand at the end during the Q&A and we can unmute you and you can ask your question over voice if you prefer. Neon welcomes contributions from everyone who shares our values of unity, creativity, collaboration, excellence and appreciation, which is outlined in our Code of Conduct. And these guidelines apply to neon staff as well as everyone participating in neon programs like the seminar today. This talk will be recorded and made available for later viewing on the Science Seminars webpage which I'm sharing here. In our Erin's talk, there will be a link to the recording and you can watch this, share it, enjoy the talk for the future. I'll take us about a week to get that posted. To complement our monthly Science Seminars, we host related data skills webinars on how to access and use neon data. And the registration info for those is down here. In our March version, we are going to do an intro to neon soil sensor data and there will very soon be a link available for you to sign up and register for that talk. And lastly, I just wanted to point out we are soliciting nominations for next year's version of the monthly neon seminar series. So if you would like to nominate yourself or a colleague that's doing exciting research that fits the seminar series themes, please use this button right here and go ahead and get us their information. So with that, I think I can stop sharing my screen and give an introduction for Erin, brief introduction. So Erin has a master's of science in soil science from the University of Wyoming and got her PhD from soil, also in soil science at Oregon State. She studied permafrost in both high alpine as well as Arctic environments, and she's worked a lot with neon scientists during her PhD using neon data sets to study how freeze thought cycles affects, affect permafrost soils in Alaska. With doctoral research focusing on impacts of soil structure and by G chemistry. And she's also done a lot of collaborations with the environmental molecular science lab or MSOL at the PNNL lab. She is currently an NSF office of polar programs postdoc at University of Tennessee Knoxville, and she's collaborating with the multi institution phosphorus iron team as well as some scientists from MSOL to study the influence of freeze thought and redox on iron phosphorus by G chemistry at poor core and hillslope scales in the Arctic tundra. So again we're so excited to have Erin here as our speaker please take it away. All right. Thank you so much, Samantha, and I will share screen. Okay, I'm just going to zoom in unless I hear otherwise that this looks good. All right. Hi everyone. My name is Erin Rooney. I'm a postdoc at the University of Tennessee Knoxville. I am so excited to talk to you all about freeze thought today, and specifically using neon soil and site properties to evaluate cross scale freeze thought disturbance. I'll be presenting some published research from my dissertation, as well as some newer work that I've been doing with Angela passenger, who's an assistant professor at Virginia tech. What I really want to highlight throughout my presentation is the ability to connect cross scale concepts using neon assignable assets. So up at the top, I want to acknowledge do some acknowledgments I want to think neon as well as the soil organic matter mechanisms of stabilization group, which connected me with neon during my PhD as well as neon soil course. And I also want to acknowledge my incredible collaborators without him, this research would not have been possible. And I want to especially highlight Angela passenger, Rebecca Librand, Vanessa Bailey, and Kaiser hotel. I would like to acknowledge our funding from Pacific Northwest National Laboratory Department of Energy and NSF. And then also I'm currently at the University of Tennessee Knoxville, but a lot of this work was conducted at Oregon State University and also PNNL. Okay, so let's dive in. What is freeze thought freeze thought is a type of temperature driven soil disturbance that occurs across ecosystems. And this disturbance can influence a range of soil properties and I will get into those more as we move through the presentation. Now there's no official definition in terms of how long the soil temperature needs to be above and below zero degrees Celsius. Just that the soil needs to freeze and then thought or thought and freeze to constitute a freeze thought cycle freeze thought cycles can be seasonal or even diurnal temperature fluctuations. So, why freeze thought, why study these common temperature fluctuations in the soil. Well I just mentioned that freeze thought can influence soil properties. And that ability is what really interested me in freeze thought in the first place. Specifically, I became excited about freeze thought through the lens of cryogenic processes. Cryogenic processes with freeze thought as an underlying mechanism can create these really visually striking features in soils and landscapes. And just to sort of explain what you're looking at I'll start at the bottom left and go clockwise. So we can see cryoturbation, which is the mixing of soil horizons. We can see frost heave hummocks and then sorted circles. And as we can see cryogenic processes are really interesting and they can occur across the globe, but they are especially prevalent in permafrost landscapes, and we're going to stay in permafrost landscapes for the first about two thirds of this seminar. So just to make sure we're all on the same page permafrost is any ground that remains below zero degrees Celsius for two or more consecutive years. And permafrost occurs below the active layer which is the portion of the soil that seasonally freezes and then thoughts. And when we talk about permafrost, we generally don't talk about it without also talking about permafrost thought. If you've heard about permafrost at this point you've probably heard about permafrost thought. And the reason why is that permafrost thought has the potential to release a lot of carbon into the atmosphere. We're projected to lose around 40% of our global permafrost by the year 2100 and around 50% of global soil organic carbon is stored in that permafrost. This is obviously something that needs a lot of attention, but something that I always get really interested in when I think about permafrost thought is, what is that thought going to look like. And this is a concept that I find really striking, because permafrost thought is not a bimodal transition from a frozen to thought state, but rather exposure to freeze thought cycles, possibly for the first time in millennia. So why does this matter that permafrost thought is exposing soil to freeze thought cycles. Well, as we talked about a couple slides ago when I showed you those pictures, we could see that surface micro morphology and underlying soil structure were influenced by freeze thought. So freeze thought has the ability to alter soil properties like the physical structure of the soil. If we zoom in even further, we can also see that freeze thought can alter the soil properties at a really fine scale. So crystallization pressure can deform soil cores and soil for throats, it can either expand or collapse them. I'm hoping you can see my mouse. I'm going to indicate this would be a soil core. This would be a soil core throat. And we can expect that freeze thought might cause these to expand or if to expand one in between might collapse. So I want to connect this idea to permafrost aggregates and specifically the incipient stages of freeze thought. So freeze thought can change if it can change the core network as permafrost thoughts. This could potentially influence the ability of the soil to hold water against gravity. If we're seeing expansion of soil cores and for throats at too large of a size to hold that water against gravity and this could influence the water holding capacity. I was interested in the ability of freeze thought cycles to transform the soil core network because there could be potential impacts to water holding capacity and also to spatial connectivity and access to substrates and nutrients for microbes. So the driving question before behind this first research project that I'm going to present was can freeze thought change the micro environment of permafrost aggregates. The hypotheses were that smaller porthroats less than 50 microns would hold capillary water. And as that water crystallized those smaller porthroats would undergo expansion following freeze thought. Because those porthroats were expanding. We also anticipated that connectivity of the core network might increase if those porthroats were enlarged from crystallization pressure. To examine this we used neon cores from Tulik, Alaska specifically examining the upper part of the permafrost, which could in the coming years be exposed to freeze thought so we wanted to know how that core network would respond. So look at this we use x-ray computed tomography, which can tell us the high density and low density soil as well as water and air, and really in a really cool method we can differentiate the core network into these individual core regions and pinch points or poor regions between the four regions. So, for our sampling three cores were extracted from Tulik, Alaska by neon, shout out to Mike's and Clements and the Alaska neon team who I believe are currently out in the field doing some more core sampling of permafrost cores. So we used three cores and then we sampled two aggregates per core from the upper mineral permafrost. Half of the aggregates were brought to 16% moisture and the other half to 28%. We then had an initial XCT scan and then a freeze thought incubation. So the aggregates were subjected to five freeze thought cycles. After this we did another XCT scan. This allowed us to compare the scans before and after freeze thought and look at things like poor coordination number, poor volumes and poor throat size distribution, and I'm going to get into what each of these mean as we move through the results. To present our first finding we found that connected water filled poured volume decreased following freeze thought. And I just want to back up for a second when I'm talking about connected pores that means pores that are connected to the rest of the core network, and unconnected pores are not connected to the rest of the core network. So linking this back to our first hypothesis, we expected to see expansion in smaller porthroats, and we did expect those porthroats to be filled with water. So it seems like the pores and the porthroats that we anticipated to be impacted by freeze thought are being impacted we're seeing a decrease in connected water filled pores. And I want to mention it makes a lot of sense that the connected pores would undergo expansion, because they have access to more water and ice throughout the rest of the core network since they're connected to it. So more water can move along the freezing front and deform these pores. But the question is we expected to see expansion in our hypothesis so we know that these pores are being deformed, but are they being expanded or collapsed. At first glance, it could be expansion. If you expand these pores it could make it harder to hold water against gravity. And so as porthroats expand, we could see a decrease in water filled pores and an increase in connected air filled pores. That's kind of what I would expect is an equivalent increase in air filled pores to indicate that expansion is occurring. That's not what we're seeing though. There's no equivalent increase. We see an increase in pretty much every other volumetric pore type, but that includes unconnected pores, and an increase in unconnected pores could indicate that collapse is also occurring. And a collapse could also result in a reduction in the volume of these water filled connected pores. Another reason that I think we might be seeing both expansion and collapse. Is that we saw this really striking spatial isolation of some parts of the pore network following freestyle. So to give you a little bit of an explanation, the more vivid colors are the connected poor network, and the paler colors are disconnected. So in this aggregate, we had a lot of disconnection following freestyle which would indicate that we're going to disconnect portions of the core network collapse is likely to occur in. So I want to link this back to our second hypothesis where we anticipated we would see an increase in poor connectivity as poor threads enlarged via crystallization pressure. We're not seeing an increase in poor connectivity. We're actually seeing isolations of portions of the poor network following freestyle. I just think this is really interesting and I want to mention it really quickly. The aggregate that I'm showing not all aggregates were this dramatic in terms of the, the decrease in connectivity. But this aggregate is one of our lower moisture ones. And so I want to mention that the lower water content did not seem to reduce the ability of freeze thought to isolate large portions of the poor network. I also just want to give you a little bit of a glimpse into the variability in the poor morphology that we saw. There are six aggregates and you can see there's a ton of variability. And it's almost surprising to me that we see any common responses to freeze thought when the poor networks look this different. But we do see common responses and I'm going to get into another one of those now. So we're going to get back into spatial connectivity and to talk about that I want to explain what poor coordination numbers are. A poor coordination number of three means that that poor is connected to three other pours directly. A poor coordination number of one means that that poor is directly connected to one other poor. And then a poor coordination number of seven means that that poor is really well connected and it's connected to seven other pours directly. So we found that there were more singly connected pours following freeze thought. And this could indicate potentially collapse or expansion of smaller poor throats. It could be collapsed in that we have some poor throats closing a loss of connectivity and it's leaving only one poor directly connected to another poor in some portions of the poor network. So if this poor throat closed, we would now have a singly connected poor right here. We could also be having pores and poor throats that are expanding into resolution, we're only able to see down to a certain resolution down to around 20 microns. And so, if we have pores and poor throats that are smaller than that, which of course we do, and during freeze thought they expand, then they could be expanding into resolution and we're seeing them for the first time. Regardless of whether expansion and collapse is occurring and again it's really seeming like it's both a singly connected poor network is a poor network that is really vulnerable to future disconnections. If you close one poor throat, you could isolate a large section of the poor network. So the last piece of data that I want to talk about in this project is the deformation of poor throat diameters and we're going to look at poor throat diameter distribution. The largest shifts in poor throat diameter distribution occurred in poor throats less than 100 microns. And I want to draw your attention to peak splitting. So we saw this peak that would occur in the before poor distribute poor throat diameter distribution distributions. There was this peak right at around 30 to 30 to 35 microns. And following freeze thought this peak seemed to split into two smaller peaks on either side. So this could indicate that freeze thought is not driving similar deformation type in poor throats of similar sizes we're seeing both expansion and collapse occur. But again, have to note that we might be seeing some peaks expanding into resolution. And from poor throats expanding into resolution right here, it's very suspicious that this peak is occurring right where our resolution starts. So crystallization pressure acted on poor throat diameters of 30 to 40 microns resulting in both collapse and expansion. And this is a critical poor throat range for water holding capacity, depending on texture, impacted by freestyle cycles and permafrost aggregates. I also want to note, if you're looking at this so we have the water content 16% here, 28% here you might be thinking to yourself, okay so some of these lower moisture aggregates responded a little less in the poor throat diameter distributions to freeze with the exception of this one, but I want to bring your attention to RECC 16% moisture, which arguably had some of the most subtle responses. This is the aggregate that I showed a couple slides ago that had the most dramatic spatial isolation following freeze thought. So even if the poor throat diameter distributions don't change that much we can still have very dramatic isolation in terms of the poor throat network. So both expansion and collapse can occur across the same poor throat sizes based on our data. And just to wrap up this section, we did see that freeze thought changed the micro environment of permafrost aggregates. We think both expansion and collapse are likely specifically in 50 micron or less poor throats. Poor throats size did not seem to be driving deformation type, and there was a decreased spatial connectivity of the poor network. So what I want to put forth is that we wondered if initial poor morphology could play a role in driving poor response, as opposed to poor throat size. So something I want to propose is that poor deformation from freeze thought may be dictated by a combination of poor geometry, architecture, and the movement and direction of the freezing front. So this complicates our understanding of poor network response to freeze thought during the incipient stages of permafrost thought, but it's an important finding because with these changes is the power to decrease spatial connectivity, isolate chunks of the poor network, and change water holding capacity in the soil. So if you want to read more about our work you can read our article it's in geoderma I want to again think all of my co authors on this work. Alright, so we talked about freeze thought and how it can alter physical structure, but it can also affect other soil properties and we're going to move into these next. So freeze thought can cause salt to be excluded during ice crystal formation as those ice crystals formed they will exclude salt and increase the solute concentration of the soil water which can lice microbial cells resulting in a nutrient influx. And as nutrient influx or availability of substrates can impact biological activity as micro microbes that survived the freeze thought cycles are able to use these substrates that came from the microbial turnover. freeze thought can also affect organo mineral interactions physical deformation to minerals and aggregates can release nutrients into solution or substrates. So we talked about to look and we talked about for deformation, but by studying the neon's tool excite we have the ability to compare it with other neon sites across the climate gradient. So next we're going to compare to look and Healy, and specifically the influence of contrasting freeze by history at these two sites. And while we looked at those physical impacts and the previous section now we're going to talk about soil organic carbon response to freeze thought in terms of compound type and nominal oxidation state of carbon. So the focus of this research was the response of sites with contrasting freeze thought histories to future freeze thought. So our first hypothesis was that we have these two sites we have healing to look, and they have very different prior freeze thought, Healy has over 40 freeze thought cycles and these upper horizons to look has less than 15. So what we did with our first hypothesis was that soils with more prior freeze thought, like it healy would have a greater relative abundance of lignin like an aromatic compounds because there would have been more decomposition and use of more simple compounds during previous freeze thought. For our second hypothesis, we wanted to compare these horizons that actually had very similar prior freeze thought in terms of it being little to no prior freeze thought. And here we expected that following experimental freeze thought in these soils that had seen very little freeze thought previously, we would see a loss of aliphatics following experimental freeze thought and an increase in the oxidation of carbon, as those compounds underwent decomposition. So again, we're using neon cores we're using Healy and two of neon cores. We sampled them by horizon, and then the cores were split into two, they went into half went into a freeze thought experiment incubation. The other half went into freeze only for our control the the soils that went into the freeze thought incubation underwent six freeze thought cycles. We then analyzed FTI CRMS looked at the relative abundance of carbon classes as well as the nominal oxidation state of carbon. We also looked at some soil properties to compare and get some context of how the two sites differed overall, we looked at mineralogy texture peach total carbon and nitrogen and soil moisture. And lastly we integrated some neon assignable assets bringing in iron and aluminum crystallinity and soil temperature. So let's start with some of the neon temperature data which allowed us to quantify freeze star. We were able to see that there was more prior freeze thought and Healy soils, the organic soils soil depths had over 40 freeze thought cycles and the upper mineral soils had around 13 freeze thought cycles. In Tulik, the organic soils had less than 15 freeze thought cycles and upper mineral soils had zero to two freeze thought cycles. The lower mineral soils of both sites had little to no freeze thought. And so in our thinking we would be able to see large differences between these two depths and potentially these two depths, and very similar responses to experimental freeze thought in these two depths. Getting into some of those soil properties we found that mineralogy was pretty comparable across both sites. Healy had more mica and felt sparse whereas Tulik had more quartz. Healy had a higher ratio of poorly crystalline iron and aluminum to crystalline to look at more crystalline. Healy had a little bit more clay and a little bit higher moisture, and to look had a little bit higher sand and lower moisture. So let's get into our carbon results and we're going to start by just talking a little bit about what for your transform ion cyclotron resonance mass spectrometry will tell us. So this is a high resolution analysis of a soil organic matter composition. We can see the hydrogen to carbon and oxygen to carbon ratios of molecules and from this we can delineate them into aliphatic molecules protein lipids carbohydrates and simple structures, lignin molecules which are plant derived complex benzene ring structure and they require specialized enzymes for decomposition, and then aromatics and condensed aromatics again complex benzene ring structure, and again requiring specialized enzymes for decomposition. We can also see the nominal oxidation state of carbon, a higher NOSC is a more thermodynamically favorable molecule, which will require less energy for decomposition, and lower NOSC is a less thermodynamically favorable molecule, requiring more energy for decomposition. So let's just look at the soil organic matter composition without any free spa so we're just looking at the control soils. And just looking at this you can see that the sites have a lot of variability. The red is he leave and the teal is to look and they really overlap. We actually saw the most differences by depth. So the circles indicate organic horizons, and we're seeing a little bit more of an influence of aliphatic compounds here. But again, a lot of variability to start off with before we even introduced experimental freeze thought. We found very few differences in the relative abundance of carbon classes following freeze thought. So again, I just want to show you these organic soils of both sides having a really high dark purple is aliphatics, a really high influence of aliphatic peaks, compared with the mineral soils. So I would characterize that as the opposite of what we expected we expected less aliphatic peaks and soils that had undergone more prior freeze thought. And also for looking this is control and the next to it is following freeze thought. I would also say that I was a little bit surprised that we didn't see more of a response to freeze thought in the relative abundance of carbon classes. These are the two depths that were the most different with prior freeze thought, and I would say they're responding very similarly to experimental freeze thought. The change that we saw following the freeze thought incubation was actually in Healy's lower mineral horizon, which is finally something that we did expect with our hypothesis. We anticipated that we would see a decrease in aliphatics following experimental freeze thought, which we do and an increase in lignin like molecules, which we do see. And interestingly, though, to look with a very similar amount of prior freeze thought is not showing that same change to freeze thought. So we found a loss of carbon molecules during freeze thought in Healy's lower mineral soils, you can see it really well in this Van Kravlin plot, where the orange indicates the lost molecules and the teal indicates the gained molecules during freeze thought. So more peaks lost during freeze thought and gained in this lower depth and very few aliphatic peaks gained. Again, totally what we expected from this lower mineral depth, but it would be have been really great if the tulip one also did it because that was the reason that we expected this one to do it was little prior freeze thought. So in Healy soils, the soil depth that the least prior freeze thought had the most carbon loss. And that is the fundamental oxidation state of carbon now, and how that changed following freeze thought. So the NOSC of gained compounds was higher than the NOSC of lost compounds in all Healy depths, and that is what we expected to occur following freeze thought. The NOSC only increased in the organic depth from tulip. It actually decreased in the upper mineral and there was again no change in the lower mineral depth of tulip. The tulip, lower mineral soils with the most similar prior freeze thought in terms of it being very, very little prior freeze thought had totally divergent responses following experimental freeze thought. So something we were wondering was if a combination of freeze thought history and higher moisture content in Healy soils influenced carbon loss. So the saturated conditions in Healy soils, which could have resulted in a mobilization of iron and carbon as a result of iron reduction during anaerobic metabolism. So the reactive iron mineral dissolution could have resulted in carbon release. So evidence for this is the high water content that we had in those, those lower Healy soils as well as O2 limitations for depolymerization which would favor lignin preservation, which is what we saw in those soils. It would have increased in the relative abundance of lignin and a decrease in the relative abundance of aliphatics. Another question that we had was if contrasting permafrost formation could have altered soil organic matter response to experimental freeze thought. So I'll give two examples epigenetic and syngenetic permafrost and epigenetic permafrost the material is deposited and then the permafrost grows up into it, allowing for decomposition or biogeochemical processes to take place. Whereas in syngenetic, the material is deposited and the permafrost grows at the same time. And so that could result in different soil organic matter composition between the two sites and influence how they responded to future freeze thought. So future freeze thought may result in contrasting carbon losses across sites and depths. A combination of freeze thought history and soil properties dictated soil response to experimental freeze thought. And the site and depth combinations with the least prior freeze thought showed totally diverging responses to experimental freeze thought, potentially due to differences in soil moisture mineralogy and permafrost formation. So if you want to read more about this work, you can check out our article in JGR biogeosciences that we'll get into the details more. And I'm going to move forward now into the last section of my presentation. So sharing across tulik and hilly allowed us to get a sense of not only variable responses to freeze thought, but also what so properties may look like as the Arctic continues to warm will tulik start to look more like hilly. So in terms of freeze thought, we do see increases in freeze thought cycle frequency with depth over pat over the past decades. The ability for freeze thought to influence soil properties is especially important as freestyle cycle frequency changes with global warming, and this is a really great example of that, where you can see through this data from NRCS tulik monitoring freeze thought cycles becoming more frequent at deeper depths across two decades. So the increasing frequency of freeze thought and the Arctic made Angela passenger and I really curious about how freeze thought might differ across a larger subset of neon sites from multiple biomes. We were also curious about what freeze thought and other biomes could tell us about what freeze thought might look like in the Arctic in coming decades. And if I'm being totally honest, we were also just really excited to be able to take a large neon data set and quantify freeze thought across a bunch of different neon sites. So as we started to quantify freeze thought, we wanted to make sure that our freeze thought parameters or the way that we defined a freeze cycle weren't inadvertently excluding anything cool. So we compared rapid and longer duration rapid being four hour longer duration being 12 hour. We compared those different durations of freeze thought cycles across 40 sites and found that despite the majority of those sites showing sort of a one to one relationship. There were certain sites that had more rapid freeze thought cycles. So CPER, which is the Central Plains experimental range. It's located in Weld County near the Pawnee National Grasslands in Colorado, which I think is near where I'm from. And then also North Sterling, which is also in Colorado and Logan County, where some of those sites where we saw a lot of rapid freeze thought cycles and less longer duration freeze thought cycles. We found this really interesting and it made us want to use neon data to go beyond just quantifying freeze thought and instead to see if we could use neon site and sell properties to understand what was driving freeze thought at different sites. But interpreting data for 40 individual sites was not a great approach. So we wanted to see if we could find commonalities in freeze cycle frequency and drivers by biome. So just looking at the relationship between the rapid and longer duration freeze thought cycles, we can see that they're strongly correlated across most biomes. And then the temperate grassland and woodland shrubland is where we see some favoring of those more rapid freeze thought cycles. We also found that seasonal freeze thought patterns differed across biomes and depths. So the majority of winter freeze thought is again occurring in those biomes that have more that favored more rapid freeze thought temperate grassland desert and woodland shrubland. And then the majority of fall and spring freeze thought is occurring in the boreal forest biome and at deeper depths. So in looking at this we started to observe some similarities between certain biomes, for example, temperate grassland desert and woodland shrubland. So we know from the literature what the main drivers of freeze thought are expected to be air temperature precipitation organic math thickness and snowpack. But these typically these drivers are examined within a single biome and not at a cross biome scale. However, we wanted to evaluate these drivers at a larger scale so that we can form a cross biome understanding of how freeze thought cycle frequency and the drivers of freeze thought might change as site and soil properties shift with global warming. So what is the best way to interpret freestyle data at a cross biome scale, and the answer that we came up with for climate groupings. This allowed us to break sites and biomes down to two parameters mean annual temperature and mean annual precipitation, these parameters would influence freeze thought and also shift with climate change. Obviously they're not the only important parameters for freeze thought but by using them to break our sites into categories, we can look for other commonalities and drivers of freeze thought. So we grouped Whitaker biomes. Oh, I'm sorry. We grouped Whitaker biomes into larger climate groups based on adjacent regions in the Whitaker biome diagram. So for cold and dry we now have an end of six, and it corresponds to boreal forest and tundra. For our warm and wet climate grouping we now have an end of 18 and that corresponds to temperate rainforest, temperate seasonal forest and tropical seasonal forest in Savannah. And then warm and dry corresponds to subtropical desert woodland shrubland and temperate grassland and desert, and we have an end of 15 so it, it increases the amount of sites that we're able to look at by grouping them in into groups that have similar characteristics. By interpreting freeze thought data across these climate groupings, we can evaluate these drivers of freeze thought within each individual grouping because they may impact those individual groupings differently. So we can see some additional support for dividing climate groupings in this way when we look at the distributions of variables that we predicted based on the literature would be the most likely to influence freeze thought. So we have mean annual temperature, the difference in temperature maximum to minimum, organic map thickness mean annual precipitation, aridity and precipitation is snow. So this allows us to get a better understanding of the variability within each climate grouping, and also look at how the properties differ across our different climate groupings. So, for example, we can look at organic map thickness which shows a ton of variability in the cold and dry grouping and has a much higher possibility of how much organic map thickness can be present, compared to warm and dry and warm and wet groupings. Precipitation as snow is another interesting one to look at. We can again see some variability in cold and dry and warm and wet and a difference in how much precipitation as snow could be expected in these different climate groupings. So, when we looked at the data as just one large data set without climate groupings precipitation as snow was not significantly related to the presence of freeze thought cycles overall. However, there's a significant interaction between precipitation and snow and climate group, indicating environment specific, an environment specific relationship between precipitation as snow, and the presence of freeze thought cycles. So, currently in this figure we're just looking at the likelihood of freeze thought occurring with a probability between zero and one. And so in cold and dry climates, increasing snow is associated with less likelihood of freeze thought presence. And then the inverse is observed in warm and dry and warm and wet climate groups. So this is super interesting, but we're only predicting the presence and the likelihood of freeze thought. We're not looking at the amount. And this figure also raises questions about how seasonal differences can influence snow as a driver of freeze thought within these different climate groupings. So looking seasonally, we can get more information about the influence of snow as precipitation on the amount of freeze thought within climate groupings. When we're looking at these correlation thoughts, the magnitude and the direction, not necessarily the significance of what we're interested in, we're supporting these contrasts with ongoing statistical analyses, sort of behind the scenes right now. So let's start with precipitation as snow. And let's look at how we can see these differences in the direction of correlation. So the only seasons and climate groupings that were increased snow resulted in more freeze thought was warm and dry climate groupings during winter and spring. Most other groupings and seasons, when there was more snow, there was less freeze thought. We also looked at how mean annual precipitation influences freeze thought. And in the warm and dry climate grouping, a higher mean annual precipitation resulted in less freeze thought in every season. But in cold and dry and warm and wet, more precipitation, mean annual precipitation resulted in more freeze thought. Organic mat thickness was a little tricky because not all of the sites had organic mats to be reported. But we did find that there was. Okay, so this is kind of an interesting finding that the thicker organic mats increased freeze thought and spring for cold and dry. This is honestly unexpected. I would expect a thicker organic mat to buffer frozen soil against warming spring temperatures. I'm a little bit surprised by this, but at the same time, to only six observations, that's kind of one of the issues with the cold and dry climate grouping, and there's still more research that needs to be done here. There's more in line with what I expect, a thicker organic mat in the warm and wet, decreased winter freeze thought. This makes sense because a thicker organic mat would reduce the exposure of frozen soil to air temperature fluctuations during winter. So this direction of correlation, less freeze thought under a thicker organic mat makes a little bit more sense to me. It's kind of interesting, the highest magnitude correlations between mean annual temperature and freeze thought only occurred in one season per climate grouping. So a higher mean annual temperature increased freeze thought or resulted in increased freeze thought for the warm and wet grouping. And then a higher mean annual temperature was correlated with a decreased freeze thought cycle frequency in spring for the warm and dry climate grouping. And then a higher mean annual temperature was correlated with more freeze thought in fall for the cold and dry grouping. We're going to see some really interesting patterns. For example, the direction of correlation between mean annual temperature and freeze thought and snow and freeze thought are opposites in these individual climate and season groupings. So these correlation plots really highlight the reasoning behind breaking the sites and biomes into climate groupings with freeze thought drivers behaving differently depending on the grouping. Of course, there's variability within each grouping as we can see in this violent plot of freeze thought counts, but in my opinion, we also see a pretty convincing gradient in terms of freeze thought amount across climate groupings. So this analysis is ongoing, but our hope is that understanding drivers of freeze thought in the context of site and soil properties will allow us to understand how seasonal drivers of freeze thought differ at a cross biome scale. And our goal is to contribute to an understanding of what the cross biome scale can tell us about drivers of freeze thought and how those drivers may change during climate change driven shifts in site and soil properties. All right. So thank you for coming along on this journey with me through the cross scale impacts and drivers of freeze thought from physical deformation at the core scale to high resolution characterization of soil organic carbon response at the depth and site to the drivers of freeze thought grouped by climate and the cross biome scale. Neon cores and data drove the connectedness of this research and allowed us to study freeze thought in the same sites at multiple scales and connect the sites to others within Neon's network. So thank you for the opportunity to talk about freeze thought and if there are any questions, I can take them now. Thank you so much Aaron for a wonderful presentation that was so interesting, especially the first one I was just, I felt like I was a microbe I was down in the poor network it was so interesting to think at such a fine scale which is how the microorganisms are experiencing that world and soils that was very, very interesting. So there are a couple of questions in the Q&A. So and if you think of more please just keep adding those to the Q&A and we'll do as many as we have time for. And I'm trying to put them in the Q&A if you can instead of the chat but I'll keep an eye on the chat as well. So someone asked Aaron, I was wondering how did you set the threshold of cold and dry warm and dry and warm and wet in that third study. Yeah. So we really just differentiated those. Let me, let me go back one second. Okay, and then. So we differentiated those in terms of what adjacent winnaker biomes we thought could be grouped together. And so this was the result of like mean annual precipitation and mean annual temperature, and then also something that we were looking at was, and this is a little ongoing, but something else we're looking at is how these different groupings kind of differentiate themselves when we look at these main drivers of freeze thaw. And so that was kind of how we settled on the cold and dry which was tundra and boreal forest. That being said, there are some obviously differences. I would say that that is one of our most variable climate groupings. And so yeah I'm not sure if there's if there's any follow-ups but that definitely interesting and interested in talking about that more. Nice, thank you. I want to add another question by Stephanie Parker. What effects to changes in freeze thaw have on the organisms living in the soil? That is a great question. There's been a lot of research done on this. My understanding as someone who has not done a ton of work in this area is that when you're looking at the effect on microorganisms of freeze thaw, the biggest effects I believe occur during the first freeze thaw cycle. And from that point on, after we have some turnover of the microbial biomass from the microbes that were laced when the solute concentrations became really high in the soil water. After we see that turnover and the response of the microbial community that survives, there seems to be less of a response in continued freeze thaw cycles. So if you're really interested in looking at this, especially in terms of permafrost, you want to really focus on that first freeze thaw cycle, which is so hard, because if you're going to sample soil, you're essentially just doing everything you can to preserve the first freeze thaw cycle so that you can monitor what's happening during it. So that would be my response to that and I can answer any follow-up questions. Great. Yeah, I'm seeing some more questions in the Q&A, so I'll keep going. Also, folks, remember if you would prefer to, you know, if you want to raise your hand through the Zoom software and ask a question over voice, we probably have time to do that, but I'll do a couple more from the chat. Cameron says, I may have missed this, but from the JGI Biology of Sciences, I think the second study. Do you have data to compare the temperature intensity of the freezes, i.e., are the historical freezing temperatures comparable for the permafrost between Healy and Tulik? That's such a good question. No. Yeah, so I can confidently say that we never consider a freeze thaw cycle just above and below zero degrees. We need for that when we did above and below two or minus two. And so we're always trying to get, make sure that we're getting enough so that the soil would be completely freezing and thawing, but we don't break that magnitude into different groupings. Like, so what you're suggesting would be, okay, let's look at freeze thaw cycles that go above and below two degrees plus minus, let's cap those at four and then let's have another grouping that goes above and below four to above and below eight and then like 12. And I think that's a really cool idea and I didn't do it, so you should do it because that sounds awesome. Nice, full. All right, so I'm just going to keep going down here because I don't see anyone raising their hands with people are using the Q&A, which is awesome. I'm going to jump to the bottom here because this is kind of one of my questions then we'll go back to some of the ones above. Courtney Meyer is asking thinking about freeze thaw and all the factors that influence it what is the intersection with effects on soil carbon release. And I was kind of going to be my question from your first study was like if you have this more disconnected poor network does that mean actually soil carbon gets protected from thaw or you know what are kind of like some of the integrated implications for carbon? Totally, yeah, such a good question. So, we looked at just the incipient stages of freeze thaw. So, after five freeze thaw cycles but obviously there's going to be more than that. And other studies if anybody's interested in this I really recommend pretty much anything by looking at what's going to happen in soil poor networks beyond like up to 20 freeze thaw cycles. They're not looking at permafrost but they're looking at non permafrost soils. What we find is that what we saw was sort of this like unpredictable response following five freeze thaw cycles and somewhere around seven that response starts to become a little bit more predictable, just in terms of the poor network kind of loosening and pores that are expanding starting to really expand. And so they don't answer the question I believe in terms of looking at like the ability of the poor network to spatially isolate. So my response would be at some point as freeze thaw cycles continue there's a point at which the response of the poor network becomes a little bit more predictable, but it is so easy to disconnect a portion of the poor network by closing just one core throat. And so I think that when we're thinking about carbon response, and we're looking at the poor network, at least in our study, it seemed like it was really easy to spatially isolate some of the poor network, and potentially protect some of the carbon and then I would assume vice versa because it would be really easy to do something else that I have to mention though is that we were only looking to like 25 microns and a microbe could be down to like five microns we can have smaller pores and poor throats, and we did not target that portion of the poor network in our analysis. And so there's still a lot more that needs to be learned in terms of what that connectivity looks like at a smaller scale. I'll just ask another one from the Q&A because I think it was a follow up to that. Do you think more isolated micro pores after freeze thaw would lead to higher microbial activity? Thank you, MTAZ. MTAZ is in my group, my research group. Hello. So more isolated micro pore after freeze thaw cycles would increase, oh no the question went away. I'm so sorry. More disconnected poor network increased microbial activity, which was kind of the opposite of what I was saying would you have decreased microbial activity and like protection of organic matter so they were kind of asking. Ooh, what an interesting question. I'm not going to really definitively say it because I'd want to see the experiment that would give us the answer to that, but I see the reasoning behind it. If you have a simplified poor network, I guess it depends on where the substrates wound up and where the microbes wound up, you could potentially have a more simplified poor network with a lot of substrates and a lot of microbes all in the same area, not having to move very far to get to the carbon that they want. So yeah, it seems, I mean, I feel like the big message of that first section was like variability and kind of the second one too. But I think that I think that's a really interesting question and something that would be really cool to research further. All right, I took the question away from you a little early there. Let's see if we can get through maybe one or two more and then we'll, we'll end. Since we've been talking about the poor network study is someone asked if you could briefly explain the methods in kind of quantifying the poor connectivity. Of course, you know, they can contact you after you could share papers or what would be like the one minute or two minute methods summary. That's not a really good question because poor connectivity week we targeted looking at the poor coordination numbers. So, and then also like we had the poor volumetric fraction so we didn't I didn't present like this was how connect like this is the connectivity percentage of the poor network. What we looked at was the poor volumetric fractions in terms of air and water filled connected pores connected to the rest of the four network and air and water filled pores so we looked at the volumes of those. And then we also looked at just poor coordination numbers which are a measure of how many pores, one pores directly connected to. And then we can look at the distribution of those poor coordination numbers and how they change before and after freestyle. And please feel free to contact me if any of that doesn't make sense and you can also check out the paper in geoderma. Okay, so I think we have time for one more let's put this last one in. Kelly says really fascinating do you know which sites and years soils freeze before snowback when evaluating each site and thinking this is to do with the third study. Oh, oh, oh, oh, okay. Hmm, I don't. That's a great question. I feel like that's a whole other paper though, like looking at freestyle cycles, what like looking at the snowpack and then looking at when freeze freestyle, or freezing occurs relative to when that first snowpack is what I'm getting from this question. So, yeah, I don't know. I don't know anything about that that's not a timing that I've looked into but it's a really good question and I do think there are papers that look at that timing. That are a lot more focused on like soil freeze days relative to when snowpack occurs. If you want to reach out to me I can try to find that paper for you I don't know what it is off the top of my head. Well, thanks again so much and we're giving you a virtual applause. This was a wonderful talk. Thanks to everyone for joining and asking great questions will have another oh and I forgot to say happy Pi day. 314. So we'll have another science seminar in four weeks and then we've got a data skills webinar and using the neon soil sensor data. Very relevant. Thank you everyone and we shall see you next time. Take care.