 So let's begin. Greetings, everyone, and thank you so much 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 seminar series is to build community among researchers at the intersection of Ecology, Environmental Science, and NEON. You can find a lot more information about the seminar series on the webpage that I'm showing right here. We are very excited to have Dr. Elizabeth LaRue as our speaker today, and before we turn it over to her, I will just go through a few logistics. So we have enabled the optional closed captioning for today's talk. If you would like to use that feature, please find the CC button in the bottom of your Zoom menu bar and go ahead and turn on the subtitles. The webinar will consist of a presentation followed by a Q&A session. If questions come to you while you're watching the talk, please drop them in the Q&A box, and we will moderate discussion at the end of the talk. During the Q&A period, there should also be the opportunity to ask questions over audio if you prefer to raise your hands. NEON welcomes contributions from everyone who shares our values of unity, creativity, collaboration, excellence, and appreciation. This is outlined in our Code of Conduct. The guidelines apply to NEON staff as well as any participants in any programs we are offering as part of NEON. Please do review our Code of Conduct, which can be found on our webpage. And lastly, to compliment these monthly seminar series, we are hosting related data skills webinars where you can learn more about how to access and use some of the NEON data that you might be hearing about in the presentations. Registration for those that are available on this same Science Seminars webpage. These are the talks, and then if you just keep going down here, here is the series of data skills webinar with links to register for anyone who's interested in. So now that we've gotten that out of the way, it is my pleasure to introduce today's speaker, Dr. Elizabeth LaRue, who is an assistant professor in biological sciences at the University of Texas, El Paso. Her research uses big data and remote sensing to uncover mechanisms underlying spatial patterns and plant species distributions and ecosystem structure and function. She's a thought leader in the evolution of the new macro system science discipline, and we are so pleased to have her presenting today. With that, I would like to turn it over to Dr. LaRue. Thank you, Samantha. It is a pleasure to be here. Let's get my screen up. Does that look good? Anybody see? Yes, excellent. Perfect. All right. Well, yes, I'm pleased to be here. Are you seeing the toolbar? Just as a double check. We are seeing the toolbar, yes. Is that looking better? Now we are not seeing your screen. Okay, now we're seeing PowerPoint. So if you go into slide show mode, that should work. Excellent. Take it away. Okay, perfect. We're good now. Okay, yes. So thank you for having me here. So I'm at the University of Texas at El Paso, and I'm happy to share my work on looking at structural diversity as a predictor of ecosystem function across scales and eco climatic domains. So this is probably a very familiar challenge to the neon community, whether you're a current or maybe a potential future neon data user. Collecting data across macro scales as illustrated by the large landscape photo here is very difficult, but thankfully we have resources such as data sharing networks such as neon and also other remote sensing tools that can allow us to get this macro system scale data from local to landscape regional up to larger macro scales, so that we can understand ecological patterns across space. So indeed, as we've increased our capacity for this big data, macro systems biology has become an emerging frontier in ecology. So this was a big literature review in the ecological literature space, looking at the over three decades of abstracts and top ecological journals to show that the in this showing that the literature space in two dimensions using a principal components analysis, we mapped the common ecological themes showing as shown by the stars in the 2d literature space. The look at the bottom part of the graph, the turquoise stars indicate topics that would indicate spatial scale, large scale or things such as climate change or anthropogenic type topics. And then the big yellow star is macro systems biology literature. Many of it coming from the NSF program in macro systems biology from the last decade. So showing that at the edge of this ecological literature space over three decades, we have this new emerging frontier in ecology of large scale ecology, but also macro systems biology, indicating its importance and emergence in our research. So narrowing in so I as an ecologist focus on measuring ecosystem structure with remote sensing and large scale data sets like neon for understanding the functioning of ecosystems from local to landscape to regional to macro scales. So more specifically, I use a type of diversity that we call are calling structural diversity and that's described as the three dimensional volumetric capacity, the physical arrangement and the identity and traits of biotic components within ecosystems. So I have a little cartoon to illustrate what I mean by these different aspects of structural diversity. So the example is two different forests. The one on the left would be that has a larger volumetric capacity filled, but also has a more diverse arrangement of trees that have different heights and different structural arrangements. And also, we can consider the identity of individuals such as species in the three dimensional space and ecosystem when we're thinking about structural diversity, it doesn't just have to be things like volume filled or the height of the canopy or the heterogeneity of vegetation, we can also consider species identity or other tree identity. So structural diversity, you may or may not have heard a term or you may have heard terms similar things like structural heterogeneity or structural complexity is not a new idea in ecology. However, we argue that it is a relatively overlooked concept of diversity in ecology. So it has early history in the 20th century with notable scientists such as MacArthur and it also has a strong history in forestry. However, in the part of our new renewed interest in structural diversity is that our technology to measure it across spatial scales has rapidly advanced in the last several decades. But we're still at a point where we do know it's an important conceptualization of diversity for ecosystem pattern and process, but we still are lacking in knowing when it's important, where it's important to just general spatial patterns. So to try to address this gap in the literature in 2020, my colleagues and I had a workshop that was supported by NSF and ended up being virtual because of COVID. Fortunately, we partnered with Neon on this and brought over 70 scientists together to talk about what is a consensus definition for structural diversity, how does that, what does that mean to different, like interdisciplinary backgrounds, like if you're a geologist versus an ecologist or maybe an earth system scientist, and then identifying key research questions and also providing some remote sensing training geared at early career scientists. So as a product of that workshop, we've been working on a special issue for Frontiers and Ecology of the Environment that will address this overarching foundation for quantifying structural diversity and its conceptualization across ecosystems, but also providing ecological examples of ecological applications and new hypotheses for the literature and talking about methods for measuring it. So that'll be coming out soon. So our goal for today is to talk about understanding how, what the ecological role of structural diversity is specifically focusing on ecosystem function. So I want to point out that we can quantify structural diversity from across many different spatial scales and levels of biological organization from molecules, for example, proteins have structure all the way up to populations, communities, and then to landscapes and macro systems. So I will primarily be focusing on the community to macro system scale today. But if this is an interesting topic to you, there's plenty of space in there for new researchers to come in and understand patterns of structural diversity and its ecological role across these different biological levels of organization and scales. So a little roadmap for the topics we're going to talk about today. So first I'm going to describe some tools for measuring structural diversity from a local plot scale up to a larger macro system scale. And then talk about two studies where we show example, my colleagues and I have showed examples that structural diversity is related to ecosystem functions. So first force productivity across North America, and then microbial diversity and the central hardwood region. And then I'm going to talk a little bit about some ongoing cross ecosystem work. So thinking about tools to measure structural diversity in a macro system framework, because you need to go you have to have good quality data, not just at a large scale, but you need to be able to have local and landscape to regional to macro scale coverage of your data. So of course, in all, all things measurement of any type of ecological pattern or process, there's a trade off between spatial scale and resolution with the sensor that you need to measure said pattern. So for structural diversity, that is also very true. So in thinking about measuring structural diversity using remote sensing, there's a trade off between high resolution sensors such as terrestrial or handheld sensors and drones where you have high resolution data, but your spatial extent is quite a bit smaller, compared to things such as aerial or satellite platforms where you have a large spatial extent, but you have typically a coarser spatial resolution in your data. So password has shown that terrestrial lidar sensors are quite sufficient for measuring canopy structural diversity aspects, things such as a backpack TLS unit, which be terrestrial laser scanning illustrated in the picture, the guys wearing the lidar backpack unit. So we know that we can get good resolution of forest canopy with these technologies. So in the study, but we wanted to scale up using neon AOP lidar and make sure that our structural diversity metrics were getting good agreement with our TLS metrics. And there's also a little bit of a difference in the way these two sensors collect data. So the TLS sensor is so you're wearing it and it's going looking up from the under store from the ground up into the understory of the canopy. So you're getting a high density of data points lower underneath the canopy versus the airplane sensors flying across the landscape and providing a high density data product at the surface of the canopy. So this difference in data viewing and density could also influence the agreement between the two types of sensors. So to test this, we looked at the agreement between TLS and ALS across seven neon sites to establish the feasibility of upscaling these structural diversity metrics using the little backpack unit that Franklin Wagner was a master student working on this project. So we had that TLS data from that backpack unit shown across these seven sites. And then of course, NEON's aerial observation platform provides free standardized area lidar across all their sites. So we use that as our ALS data product. So okay, so I keep mentioning structural diversity metrics. So we have five categories of structural diversity metrics that describe different dimensions of the canopy. So things like category of height, so it's the mean or max height of the canopy or the median height, the density of vegetation, so this would be equivalent to something like an LAI metric, covered openness, so how many gaps do you have in your canopy, and then also measures of external and internal heterogeneity. So internal heterogeneity metrics measure how rough the outer surface of the canopy is, whereas internal heterogeneity metrics measure how heterogeneous the internal vegetation heights are. So and if you are interested in using NEON lidar data to measure these structural diversity metrics, there's a tutorial on the NEON website that we created to provide an introduction to these structural diversity metrics using the lidar product. So we could check that out if you're interested in using NEON lidar to measure structural diversity. We'll just give a second in case you want to jot down the URL there. So indeed we did find that there was good agreement between the TLS and ALS technologies, except for one category of metrics, and that was external heterogeneity. So we used Spearman correlation coefficients to test the strength of the linear relationship between the metrics from each sensor, and we counted anything above 0.6 as good agreement, and we counted those as relatively good equivalent that we could scale up from terrestrial to aerial technologies. So we could have a larger macro scale look using those structural diversity. So the category that didn't have good agreement was external heterogeneity. The correlations were either insignificant or below that 0.6 threshold, and we attribute that to the different viewing angle of the sensors like the terrestrial lighters looking upward and the aeroplanes looking downward. Since external heterogeneity would be measuring the surface roughness, we recommend that you would use the aerial lidar for that type of measurement. So another, this paper just came out looking at just a little bit of a caveat that I'll mention briefly, that depending on which metrics you're using, some of them can be more sensitive to low density lidar. So I know what the, Neon has some really great new sensors that they've gotten in the last couple years, which provide typically much higher resolution than eight points per meter squared, but some of the earlier products at different sites can have that maybe like four points per meter squared average. So in the study, if you're interested, you could take a look, but just for time's sake, I'll just mention that some of the metrics were a bit sensitive to those lower densities, but there were plenty of metrics that you could choose from that weren't sensitive to the lower density lidar. So depending on what type of lidar data you're using or where you're getting that or which here's something to keep in mind if you're going to be measuring structural diversity with these data products. Okay, so now we've talked a bit about measuring structural diversity across space and some remote sensing tools to do that. And actually in my next piece of the outline, we're going to talk about using forest inventory data to measure structural diversity and then testing that as a predictor of forest productivity across North America. So in ecology, it's a known pattern that biodiversity typically measured as the number of species in an area is correlated with ecosystem functions. However, as illustrated by the example study shown here, the strength of those relationships can often vary across space or with environmental conditions. So the example here on the slide, the map, the different color shading indicates there's a positive relationship between the number of tree species in an area and the forest productivity. However, the strength of that relationship varies quite substantially across the globe regionally, depending on which region you're in. So my colleagues and I thought, well, maybe structural diversity could be a interesting predictor of things such as forest productivity. And we hypothesized that it might provide a really good proxy and potentially a better proxy of the niche space filled than things such as species richness that's often used in this biodiversity ecosystem function studies. And the rationale for that is if you look at the cartoon on the slide down at the bottom under species diversity, as we increase the number of species on this example, different tree species, those species aren't always functionally distinct on the niche axis that would be promoting our ecosystem function. And because the species aren't always functionally distinct, counting the number of species isn't always a strong predictor of ecosystem function because you're not filling that niche space. In contrast, structural diversity measures a functionally distinct aspect of some structural aspect of the ecosystem, where in the example under the structural diversity cylinders there in blue, as we increase the number of trees of different size, we have functionally distinct individuals that are using some niche axis, maybe something like light competition, and they're filling the niche space differently in such a way that it would promote ecosystem function more than an equivalent number of tree species in the example. So we wanted to test this hypothesis. And to do that, we pulled national forest inventory data from three countries of North America, Canada, the United States and Mexico. And from these inventories, they have, they all measure the diameter of the trees, the height of the trees, and also have species identity provided from the individuals that they've measured. So we took advantage of this. And this is relatively common forest inventory data we wanted to generate metrics that would be relatively easy to use for forest managers with available data that they're used to using. And we generated three measurements of structural diversity, and then we had our tree species richness, which would be our traditional species diversity metric. So the first structural diversity metric was horizontal richness, which would be represented by the number of trees in a sampled area that differed by different diameter classes of the trunk. And then vertical richness would be the number of individuals that had different counting different height classes. And then 3d richness is a combination of the two normalized combination of the two. And then the in the map we have the species richness in the bottom right hand corner. So because these three forest inventories do sample the use similar sampling methods for diameter height, species identity, but they do sample different plot areas. So we had to account for that we used hill numbers to estimate the effective richness value for 10 individuals. So that's why the the scale bars would approximately the values would scale from one to 10 being one being the lowest number you can have one in one species or one horizontal or vertical richness class, filled in those 10 individuals, or you could have up to 10. And then pattern the map is showing a average plots are averaged across the 20 by 20 meter pixel grid. So I want to point out a couple noticeable patterns and differences in structural diversity versus species richness. So both horizontal richness and vertical richness, we see higher values of richness in the eastern United States and the western particularly the Pacific Northwest. Whereas with species richness, we see our highest richness values in the eastern United States, but then southern Mexico and the Yucatan. So we see some pretty large differences in spatial patterns of the larger richness values. And this may translate into relationships with structural diversity versus species diversity and their ability to predict productivity. So to test if structural diversity versus species diversity was a stronger predictor productivity, we use the inventory information when we looked at productivity with a proxy. In this case, basal area increments. So how much diameter did a tree increase each year? And that was our response variable. And then the different four different to first diversity metrics were the predictor variable. So past research, so we're working with a large spatial scale here, and there's a lot of environmental and climate heterogeneity. And past work had shown that climate space is important in determining the strength of the relationship between tree species richness and productivity. So we used a approach used by a past group and split the data into different climate units. So we split the data into precipitation and temperature quantiles 10 each. So we ended up with 10 different climate quantile units. So that represented by 100 different models in each of our diversity productivity regressions. So the way this is summarized here, the graph on the left shows the frequency of those climate quantile units. How many of those diversity predictors of productivity models had a specific r-squared value and showing this as a histogram. And species diversity is the pink histogram. So this is the one here in the far left. And then the structural diversity metrics are the purple, green, and blue. And you'll notice that the purple, green, and blue histograms have a distribution skewed more to the right, where there's higher, more models with higher r-squared values. And if we look at this slightly different way and take the difference for each of those climate quantile units, and we subtract the structural diversity metric r-squared, model r-squared from the species diversity model r-squared across the 100 models, and look at the difference, both the distribution with a violin plot, but also the average difference between those models. So the zero line above the zero line indicates structural diversity, which shows that structural diversity was predicting more variation in productivity than species diversity. If it's below the zero line, species diversity was predicting more variation than structural diversity. So the violin plots are skewed above the zero line, and then the averages are positive, ranging from 0.7 to 4.1 percent average, higher predictive variation in structural, by structural diversity metrics than species diversity. So this analysis was across all of North America. So we did a second complementary analysis, this, what took stand age into account, because we would expect that structural diversity might increase with stand age. So we included it as a covariate in our analysis. However, we had to, we could only, we only had data to do this with the United States. So this analysis is covering the extent of the USA. And similar, or same setup with the regression models with diversity predictor of productivity, in this case periodic annual increments of biomass. So between two time periods, how much biomass did the trees in, in the plot increase on an annual basis. And we see a similar pattern where the, you look at the histogram on the left, the R square, the adjusted R squared histograms are skewed more to the right for the structural diversity versus the species diversity. And in the violin plots with the difference, we see that structural diversity had higher predicted more variation in productivity than the species diversity metric. And that range from 1.7 up to 8.3 percent higher depending on the metric. So this suggests that across this heterogeneous forest ecosystems across North America, that structural diversity could provide a complementary tool for forest management to increase or enhance and manage ecosystem function like productivity. So just another tool in the forest ecologist or forest managers arsenal for managing ecosystems for functions. So I want to zoom in a little bit to a smaller regional scale thinking about structural diversity and how that might look or how that might relate to much smaller organisms soil microbe communities in the central hardwood region. So this was a, so I'm more of a forest ecologist or ecosystem ecologist and remote sensing scientist. So me and some of my other colleagues teamed up with some soil microbiologists and biogeochemists to test the hypothesis whether forest structural diversity would be a significant predictor of the microbial diversity of forest soil communities because we expect that with higher structural diversity that would provide an indicator of greater microbial habitat. So forest to microbial trophic interactions. So to test this hypothesis or to test the prediction for our hypothesis we had data from the central hardwood region in the Midwest and we had 38 plots where we obtained forest inventory data from the Indiana DNR and they sample the trees over a certain size in a seven meter radius circle and from that they are Indiana DNR collaborator collected soil cores for us on the east and west side of the pot and which was homogenized and sent back to Stephanie Kivalin's lab for analysis for the microbial community and to Rich Phillips lab at IU to get analysis for soil nutrients and pH and then we extracted the USGS three depth LiDAR from around the spatial extent of these plots and calculated several structural diversity metrics like I was showing earlier with the tools for measuring structural diversity with the neon data so similar metrics there and Ashley Lang is a postdoc at IU and she was instrumental in leading this effort was a big group of us and she wrangled us all together and help us test this hypothesis. So we looked at two different aspects of microbial diversity. First we looked at alpha diversity of the bacteria total fungal community and then the ecto mycorrhizal fungi community and the our muscular mycorrhizal fungi community. So we did not find that structural diversity was a particularly important predictor or a correlate of the soil microbial richness. However we did find a significant positive relationship with vegetation area index which would be how dense the vegetation is as that increased we saw higher ecto mycorrhizal fungi diversity. So indicating that there is a relationship it just it wasn't a predominant structural diversity was a predominant correlate of the soil microbial richness of these groups and then we also looked at soil microbial community composition as beta diversity so how different is the composition of the microbial communities among plots and we looked again at bacteria total fungi our muscular and ecto mycorrhizal fungi. So we looked at the relative predictive ability of structural diversity but also tree diversity soil properties such as soil nutrients and pH and also stand productivity and age because these are other factors that have been shown to be important predictors of microbial community composition and diversity too. So we did find that structural diversity was a significant predictor of microbial community composition for every group except the AMF fungi. However structural diversity in comparison to particularly soil properties so things like soil pH soil carbon and nitrogen the soil properties were much stronger predictor of the soil or the community composition whereas structural diversity was a weaker predictor. So this is a as far as we're aware there haven't been many studies looking at vegetation structural diversity and how that relates to microbial communities. So we thought this was a good first step and we're excited to keep working as a group and understanding how structural diversity may relate to the microbial communities. So I've showed you that structural diversity in two instances can be a correlate of forest productivity but also of microbial diversity and then so the examples I've showed you this far have focused on forests and that's been predominant in the literature. Most researchers have been focusing on forests and that in part has been a technology challenge but as we have much better access to things like UAVs and also like handheld sensors like my iPhone has a LiDAR sensor on it now that's now readily available to many people. So with this advanced technological advances my lab group at UTEP we're going to be expanding to consider how structural diversity across grassland shrubblings and forests is looking at 15 neon sites and the southwest and western united states we're going to be looking at how structural diversity ecosystem function relationships vary across these different ecosystem types that have very different vegetation volume and just a little bit of a preliminary look comparing structural diversity in a grassland versus a shrublin so this is at ornata experimental range this is a neon site in southern new mexico chihuahuan desert habitat but we have so we have a lot of shrub encroachment there and also native grasslands so this is data looking at rumple which is a measurement of the roughness of the canopy surface and it's twice as high in the shrubland as it is in the grassland so my my group some new grad students working in the lab are going to be working with me and we're going to be expanding this work to look more extensively at how structural diversity patterns vary from grasslands shrubblings to forest and also how that links to different ecosystem functions so stay tuned exciting research coming soon and then as i wrap up today so structural diversity is not a new idea in ecology but it's an up and coming new area of research that has been fueled by new technological advances particularly in remote sensing and data sharing networks that now allow us to measure structural diversity in three-dimensional aspects of ecosystems and communities across different spatial scales so now that we have better tools that are improving every day we can start to investigate how structural diversity is related to different ecosystem processes so today i showed you some examples where structural diversity could promote ecosystem function as well as or perhaps better than biodiversity suggesting that structural diversity has potential to be a new monitoring or management tool and may help us better understand ecological patterns and services under global climate change and other global change issues so not saying that we shouldn't use things like traditional metrics like species diversity or species richness as a specific example but structural diversity can provide something in addition to that and may help us better manage our ecosystems so feel free to reach out so my email and twitter's there on the left and i would like to thank the landscape ecology member lab members at utub lots of great early career scientists working with me there at utub and then as many of you know with this large-scale macro systems research we have lots of collaborators that go into making this type of research possible so um i'd like to thank some of my um collaborators but also many more that i didn't have time or space to list today um but and also of course um my funding sources that supported the research that i showed you today see believe we probably would move to q&a no absolutely thank you for a wonderful talk um so much interesting science there and food for thought so we've got a few questions in the q&a box that i can read to you elizabeth if that sounds good and then maybe we can do an order we could take um one or two of the written questions and then if anyone does want to ask a question by unmuting and just asking it over voice you could do raise your hands and we could potentially alternate between written and or all questions if anyone wants to um to ask one so it's the first one um that came in kind of in the middle of your talk how do you measure the sensitivity of the lidar pulses if it's eight points per meter squared versus four points per meter squared yes so that particular study we looked at we took neon data that neon lidar data that was um high density over at least 25 points per meter squared and we simulated random thinning of the point cloud and then we measured the same metrics across that two points per meter squared up to 25 points per meter squared and then looked at the stability of those metrics so if they fluctuated a lot at the low density that was when we indicated that oh this metric is probably going to be pretty sensitive to that low point density because you have a lot of randomness about where the laser hits an object and then return some information so neat and then there's another written question here aside the biodiversity productivity relationship which as far as I know is debated in certain circumstances do you have plans to test the responses of other ecosystem functional properties to structural diversity yes so we have productivity and now that I'm in the southwest um I'm talking with lots of rangeland managers and also thinking about things like shrub encroachment and restoration and other ecosystem services that would be less about how much biomass is in the like forest example so living in the midwest you have I don't know I often would think more about forest but now living here in the desert southwest trying to strike up some local collaborations but then also you can think about things like wild wildlife habitat provisioning excellent um let's see I am seeing somebody raising their hands um are we able to unmute Ebenezer to ask the question please go ahead yeah that's a very good presentation from Laurie uh I have some few questions here that I would like you to share more lights too now if I have to extract data from neon sites and in relation to structural diversity now do I have to apply remote sensing to such data I mean when I'm not going physically to the site to collect the data how do you maneuver secondary data I mean pre-existing data in consonants with LiDAR and then the second one if I want to analyze structural diversity what do I need to collect what type of data do I need to extract from forest inventory and aside productivity which you actually measured what other ecosystem services do you think one can test for yeah those are great questions so the way at least with neon they have depending on which data product you're interested in working in it's typically geo reference so the way I typically work with the those data products things like the they have like base plots um I usually just go in and extract I'll just cut out the LiDAR data that I need and then process it so that and I use R typically I know some of my collaborators use Python it just depends upon which geo spatial software you you're most comfortable with but you can set up those workflows to select your locations that you want to pull data from like if you're trying to match it with some other neon data product and then just select the LiDAR data that you need but you might also be interested in landscape level structural diversity patterns and then you probably would take a different approach where you're just working with the entire extent of the neon LiDAR footprint and the AOP and then with forest inventory so you asked about forest inventory data so I pulled species identity and then also typically there's diameter so they usually measure that at about 1.3 meters height they have some sort of measuring tape and then they go and measure the diameter of the trunk of the tree stem that's really common and then a lot of inventories at least the ones I worked with will also have height data for some of their trees or all of their trees and you could extract even if you just had diameter you could measure some sort of metric and you didn't have to necessarily do it the way I did but the way I did it was I split the diameter data into different size classes like 5 to 10 centimeters 10-15 centimeters and so on and then counted those similar to species richness where I counted if we had a tree of a certain diameter that was one value of one for richness and then so on and so forth and you can kind of almost just treat it like species richness in that case it's a little bit different depending on the type of data that you have and really it just it's rather up and coming and we don't have a necessarily established specific structural diversity metrics there's a number of groups that have been working on these studies but maybe you have some creative ideas about your study system and what structural aspects that you think would relate to ecosystem functions or maybe you're looking at wildlife habitat or maybe more ecosystem services or human populations it just sort of depends upon what you're interested in awesome thank you so much and thank you for asking the question Ebenezer we got a few more that came through through the Q&A so I'm happy to read those so Zoe Leib says such interesting research thank you are you aware of any other specific labs or universities engaging on this topic I'm interested in exploring structural diversity as a PhD research topic is UTEP leading on this topic I don't know if you want to say something briefly but you know maybe Zoe should email you or tweet at you I think that would be a good we could set up a chat Zoe there are other groups but it yeah so like the lab that I did my postdoc in is at Purdue and they're doing a lot of the they're calling it digital forestry and they're doing a lot of work focused on forests and how the structure and different remote sensing platforms could be used to improve forest management and forest ecology but there are other groups there's some groups that are doing coral reef let's see Greg Asner say Kim Calder's and Josh Madden are all groups that are working on coral reef and thinking about structural diversity they measure it with drones using structure for motion those those are just a quick couple quick examples but if you want a more in-depth answer feel free to reach out we can have a chat awesome thank you so another one thanks for this great talk could you expand on how to upscale field measurements to larger spatial scales when you also use satellite data that has different temporal scales yes and that there are groups I haven't worked as extensively with satellite data I've stuck more with the the aerial idar versus the terrestrial idar and then getting into the drones but there are a lot of cool group or a lot of good groups doing cool research with Jedi data if you're familiar with that's a NASA sensor up on the International Space Station and there I know they're doing a lot of upscaling so the Jedi footprint is 25 meters and they're trying to upscale to use things like Landsat is a proxy to get like global biomass maps or other like forest height or structural aspects so I think that would be a good place to look into if you're interested in that further beyond what I can provide as an answer awesome um not see any hands so I'll keep going there's a few more in the q&a um somebody asked is the structural are your structural diversity maps based on forest inventory um so are the displayed indices only for forest ecosystems everywhere so I'm thinking referring to the map I showed of North America so yes that is forest inventory data and that was based upon the like Canada and United States and Mexico they have like a sampling grid where they have different plots across their whole country area and so that's extrapolated out it's averaged out to show like a nice visual but that would not include things like rangelands or wetlands or grasslands for example so that thus far is forests awesome and um actually one of our neon staff members says thanks for an excellent and interesting talk what ground collected data could neon ideally offer to support the work you are doing obviously besides what we are doing or just maybe what are your favorite ones of what we're doing already I mean I really love it when there's good geo reference data to be able to link the um other data like the to s products um to the aop products that really helps um I know I haven't worked on this so much but I know some people have looked at like there's the vegetation structure data where they have like different um there's some tree height data in there too but they also geo reference the larger trees above a certain size like that's pretty helpful if you want to try to get like a really good connection between the lidar or other aop products and the forests interesting thank you um by the way Zoe says thanks um I will drop you an email so yeah and um Sarah asks are any of the neon sites wetlands and if so is anyone looking at applying similar methods asking similar research questions I'm a phd student looking at these kinds of questions in Canada but having a hard time finding good lidar or other ground data in wetlands I do believe some of the many of the neon sites do have wetland land cover you probably would need like a land cover map um I haven't specifically looked at wetlands maybe this is a also a good question for some of the neon folks um but there's also right um aquatic sites and there's like riparian habitat so I would assume yes yeah we certainly do have some sites that have at least a wetland component whether they're a femoral um seasonal wetlands we have a bunch of sites with woody wetlands the northern forest as well as some down in the south southeast so there are different kinds of wetlands no coastal wetlands but there are some interior wetlands represented um within the neon data and you would of course have all of the normal data that we measure lidar air um airborne remote sensing as well as to us data so yeah anyone who's interested in wetlands I think we have we're not a wetlands focused network but wetland plots are represented in our sites and if that's of interest I encourage you to check it out thanks for the question um let's see so there's one more question about um measuring structural diversity for a plantation like a mono-dominant plantation if you want to try to have proxies for things like carbon sequestration or pollution removal um how do you think about structural diversity in plantations second question um I would be curious to know the answer the more detailed answer to that question but I guess I would think about things like sometimes in a plantation you might have different like micro habitat differences that might cause even if you have the same species you might have differences in individual tree architecture so maybe in that sense you might be thinking more of an individual level um getting into some I know there's some groups doing every kim called there's somebody who's doing some cool temporal monitoring of forests with terrestrial lidar and looking at how the individuals change like their branching patterns and growth patterns change over time I think that would be a cool connection with some of these other things like carbon sequestration or air pollution I don't know of anybody who's specifically linked air pollution to plantations and structure but it'd be interesting great and yeah someone used the q and a to say I have similar questions but I'm thinking of incorporating flux data which would make a lot of sense if that's an option um does anyone else I think that's I'm seeing we've covered everything in the q and a does anyone else from neon or external members want to unmute and ask a question or make a comment to Elizabeth um actually I saw one more question came through in the chat so maybe we'll do one more over q and a and then um and then we can wrap it up because we're almost out of time so um Javier said thanks Elizabeth could you tell us about what is the level of correlation between the diversity metrics you computed and if you dealt with that in any way yes so for the north america forest productivity study there was a correlation between structural diversity metrics and species richness of trees however the relationship was non-linear in the sense that as you got to higher values of structural diversity the species richness leveled off so they weren't increasing at the same rate so it's there's a relationship there but it's not a perfect linear relationship great I think we're out of time I think we should call it that was a wonderful presentation and a rich discussion thank you so much for being here Elizabeth and thank the rest of you for attending asking great questions um we will be back next month with another very interesting we hope talk and we look forward to seeing you then