 Our next speaker is Dr. Pei. Dr. Pei is currently a postdoc at University of North Dakota in the Earth's System Science and Policy Department. And she recently completed her PhD at NDSU in the Rain Science Program. Her dissertation work was a project that surveyed North Dakota grasslands for bee and plant communities. And she's centered her work around the relationships between exotic species and bee communities. So let's welcome Dr. Pei. All right, so yeah, my name is CK Pei. I am currently a postdoc at UND Aerospace. And I recently did graduate from here and I'll be sharing one of the papers that I did over looking at potential relationships between invasive grasses and bee communities on the grasslands here in North Dakota. So this is a more ecology-based paper study. It's not gonna be talking about how to manage grasses or the extent of grasses or anything like that. But maybe it's nice to chat about bees, especially before a break comes. So before I came up here to North Dakota, I worked in the Flint Hills region of Kansas at some really nice native prairie, which is really great to work there, except for it gave me kind of rose-colored glasses in terms of what contemporary grasslands actually are. So when I came up here to North Dakota and started looking at the grasslands here, I was kind of like forward by the amount of exotic plants that there are in the landscape. And so I centered my research on exotic plant interactions with the bee community. Usually the research involving exotic plants and bees has to do with exotic floral resources and how those explain bee communities. And that's really important to look at because bees require floral resources. They're really tightly linked to floral resources. So that's an obvious thing to look at. However, bees do have different needs. So they have nesting needs, which is not part of the floral community. Also other things, it's not in close system. Other things impact the amount and diversity of those floral resources. So I kind of want to look at non-foral components of the plant community and how they affect and explain the bee community. So in terms of the non-foral components of the plant community, I thought was most relatable to my study. First structure, overall plant structure. I knew that some bees require access to the soil. Some bees like more standing litter around. And so structure is really important and it could filter out species. Another thing is just general plant community and plant community diversity. Now I am talking about like the non-foral plant community. And yes, I know grasses flower. I'm talking about the plants that are most used by bees. Bees do use grasses or can use grass pollen, but it's not a preferred resource. It's kind of like, God, I'll take it if it's there, but I'm not gonna want it. And so yeah, so this is inclusive of that grass community. And as we know, there are two species of exotic cool season grasses that are quite prevalent across North Dakota, Kentucky bluegrass and smooth brome. And those are the species that I'll be focusing on in this study. And those are kind of known to, well, naturally the them being present replaces other plant species. They're also known to be really good at creating their own environment that kind of favors them and maybe not so much the other species. And so I'd expect them to have some influence over things that need that plant community like bees and pollinators. Another big thing that kind of encompasses everything is just general management. Unfortunately, my study and my data set doesn't address management specifically, but it does can provide some sort of ecological basis for informing management actions. So, like I said, invasive grasses have the potential to impact biodiversity. But when you're considering diverse organisms such as insect pollinators, it's kind of unreasonable to expect that invasive grasses or other environmental changes will affect bees all in the same way since bees are very diverse. So it's kind of like a filtering system. So obviously there's a lot of things that happen that make exotic grass invasions present. So there's probably a lot more factors than the ones I have up there. But like I said, those exotic grass invasions could impact the plant community, decide what kind of forb species are there, which then will determine what bee species can be there because which are reliant on those plant species. So that's an indirect relationship from exotic grasses to bee species. However, there also are direct, I think direct associations that could happen as well from exotic grasses to bees. And it's kind of through that structure. So under certain circumstances, especially if exotic grass invasions are left idle and unmanaged and undisturbed, they're allowed to build up these really thick litter layers. There are these two species connected to grass, but we're really good at that. And so that would be kind of a direct relationship to bees because it can determine which bees are there who has nesting resources who can access the ground. So that can also filter out bee species. And the reason what kind of bee species are there is because that affects the ecosystem services that come from those bees. So pollination services are the main ecosystem service that comes from bees. And so determining how what kind of bee species is really important, especially for what it means to bee function. And bee function is actually, so bee functional diversity is what determines their pollination services. So I like to use this little chart here, not chart, figure here, just shows like crop yield will increase with the more diverse set of bees or pollinators that you have. And so that's focusing on crop systems, but it's the same principle out in natural systems. The more types of bees you have, the more types of pollination services you have to fit many different types of plants. And so it's really important to keep that functional diversity present on grasslands. And so that's why I'm looking at how exotic grasses will impact those services. So some of the ways bees lend their services is through their different functional traits. So I'll be focusing on four main groupings today, body size, diet, sociality, and assuming habit. For body size, these can range in very many different sizes. Big bumblebee to a small, little, highly SP. And that determines what flowers they can visit and what flowers they provide those pollination services to. So bumblebees, for instance, can't land on a like a little chickweed flower that's just not feasible for them. For a diet, some bees can visit many different plants and use pollen from many different types of plant species. Well, others, and those are generalist species. Other species have to use a specific type of plant, even down to a species. So those are specialized bees. So that determines which plants they provide those pollination services to. Sociality, bees vary greatly in their social groupings. Bees can be super social. It's like honeybees and bumblebees where you have a main egg layer queen and a lot of workers helping out. But most bees are actually solitary where they collect their own pollen and provision their own nest themselves. And that can be important. And these social strategies kind of influence their foraging strategies. So that'll affect how they provide their services. There's also kind of traits that fall between the diet and sociality groups. And this is called parasitism. So there are bee parasites. There are kleptoparasites, which enter another bee's nest. The parasitic larvae will kill the host larvae and eat all its food. So it's stealing. So it pens the word klepto. And then there's also other kinds of parasitic bees. So parasitic bumblebees, for instance, kind of take it a step farther. They enter another bumblebee's nest, take out the queen and make all the workers work for them. So it's kind of a pretty cool thing. And then for nesting habit, bees nest in very many different ways. Most bees nest in the ground. However, some bees have to dig their holes. They're programmed to dig their holes. Others are programmed to use existing cavities. They can't dig it themselves. They have to find those cavities. There's also cavity nesting bees, which I separated out because those ones will nest in like pithy stems and other kind of existing cavities. And I also have a hive here as another category. North Dakota is the biggest honeybee producer in the US. And so we have lots of honeybees here, but honeybees don't, they don't really live anywhere like the native bees species that we have. So it's in its own category there. There's other functional traits, but those ones are the four I'll be focusing on. So it is important to keep in mind that we want to maintain this functional diversity on our landscapes. But bee diversity is affected by certain things. So resource loss is a main one, just general grassland reduction, fragmentation of those grasslands. Also resource loss brings great form in the change of resources. So we do have a lot of exotic species and that changes the availability of resources to bee species. And as we know, these are very, very common things in the Northern Great Plains. We have lots of cropland and we also have lots of exotic species. So and because exotic grass invasions are expected to work expected to, but also have been found to influence the plant community, I would expect them to contribute to be resource loss. However, this question, as far as I know, hasn't been really looked at. So this is kind of a first look at the relationship between invasive grasses and associated plant characters and the bee community. There is a related study in Canada, I think, that they used non-foral components of the plant community, but I don't think they included the invasive grass aspect, hopefully. There is, since it is the first look, I kind of just wanted to see a simple literary relationships between invasive grasses, also the related plant community characters on bee communities and forb communities. I also wanted to determine how those different plant community characters influence and shape the structure of bee and forb communities, I should say. Second objective, we'll be looking at the relationships between those plant community characters and different bees based on their functional traits. So the methods, this data comes from a statewide survey of bees, plants and greater plant communities across North Dakota. So this was conducted between 2017 and 2020. In this project, there's a total of 477 grassland sites that we surveyed at, and they are managed by many different agencies down here, public, and we had about 120 something private sites, as well as a couple NGOs. However, for this particular study, I filtered the data set to use data from 67 sites. The reason for that is because I limited the data to 67 sites because I only limited it to the sites that received two different bee sampling methods. The reason I did that is because when you, insect communities are super hard to get good measures of. And so when you use different methods, you kind of, it's a better representation of the present bee community, because they all have their own biases and stuff. So the first method that we used is netting surveys. So that just involves observers netting for bees, basically, within a plot. And our second type of method was beeple surveys. And those were colored cups full of soapy water that were kind of placed out on a transect out on the grassland and left on the site. For us, it was 24 hours. And then we could pick up the samples the next day. So these are attracted to the color and they'll fall in. So each of the 67 sites were visited two times with both bee sampling methods. And this gives us measures of relatively abundance, species richness, and diversity at each site. For the plant site of things, we did vegetation cover surveys, 51 meter square quadrats per site. And we did these surveys after, you know, the growth had started. So like after July 1st had started, we estimated the cover of three different grass species, or three different grass groups. So Kentucky bluegrass, the mountain smooth brome. Also, I have a category called other grasses. And other grasses are because it's not really feasible to train the 30 plus technicians that we had on our team, how to identify grasses. I don't feel very good in that being very reliable. And so we grip those in the other grasses. You can kind of translate that as mostly native grass species. However, it is inclusive of other exotic grass species that just aren't Kentucky bluegrass and smooth brome. So also estimated the amount of bare ground. We measure litter depth and the cover of each four species in the quadrat. So that gives us measures of the grass cover, amount of bare ground, litter depth, and a couple for bridge or for diversity values. So for the first objective, for in terms of analysis, we just use general is linear mixed models to see the associations between invasive grasses and litter depth and for richness with our be abundance, be species richness, be Shannon diversity for species richness and for Shannon diversity. And after we saw those linear relationships, we looked at which one of those predictor variables kind of significantly contributed to the bee community space and we used ordnation methods for that. Another objective too for the functional part, I did a method called fourth corner analysis, which is a way to associate different environmental variables. So those grass, grass cover and litter depth and for richness to functional grouping. So the bees based on diet group, sociality group, body size and that's seen habit for general results. So I had 20,000 bee individuals from 182 species found from those 67 sites and then found 249 estimated the cover of 240 and four species. For the grass side of things, this is just a map of relative cover of Kentucky bluegrass, mobrome and other grasses found on our sites. So it'd be really cool to see this with like all 477 sites across the state. You kind of like, it's tempting to look at trends and be like, oh, there's a lot of Kentucky bluegrass in the great Pothole region, but I wouldn't probably try that right now with this particular number of sites. But we can say as well, there's a lot of exotic grasses in North Dakota. So for the first objective, when we're just looking at linear relationships, I plotted the estimated coefficients and the 95% confidence intervals. If that confidence interval crosses that zero line, that means it's not significant if it doesn't. And it's on the right side, it's positive association. If it doesn't and it's on the left side, it's a negative association. So bee abundance, we didn't find any relationship with any of the variables that we included. I don't really like using bee abundance, but a reviewer made me do that and so I included that. Bee species richness, the only thing that we found that impacted or was associated with bee species richness and bee diversity was forb species richness. And that kind of follows a lot of things that we know about these is that they like, what's the amount of forbs there? So, but when we looked at the bee community, and this is bee community defined by a genera, and I know it's stupid of me to include a 3D generation, but I like it because of the perspective and it gives you a better look at where things actually are falling. But this is for a nation and oopsie, doopsies. These names are the names of the bee genera. And with this, you can kind of translate that as like things that are closer on the plot are more associated and things that are further or not. So what we're looking at here is these are the axis highlighted in yellow here. And these are plotted environmental vectors here. So those are highlighted in blue. And in terms of the relationship between the bee genera and those vectors, if it's closer to the line, that means it's more associated as if it's in the direction, if it's in the opposite direction, it's associated but in the opposite way. So what we found here is that forb species richness was to significant contributor to the community space defined by bee genera. And litter depth was also, we also found the litter depth was also significant contributing to the space too. So really interesting that another method would pull that out. I think something interesting to pull out is I thought it was interesting that honey bees and bumble bees were really associated with grasslands of really high litter depth and conducted the rest. I have some theories on that. And I think like one of them is bumble bees might benefit from some more cover. And that was from previous, I have a previous experience that kind of saw that a bit. Flattening it on a 2D surface and using the bee community as their species, which are, so the species name here, this kind of gives us more specific look into the bee community. It's just not useful unless you know these names and you know like things about these species to look at. So like here, there's like those bumble bees and honey bees. Really interesting was bare ground. There's lots of ground nesting bees here. There were things that dig holes. So that's interesting opposite direction where those bumble bees. Another interesting thing was like some, there's a couple species that are specialists on pollen. So they were, some of them were really associated with Fort richness. So like I said, unless you know things about those species, it's not as helpful. Which is why we do the functional stuff later on. For the forb side of things, forb species richness and forb diversity were positively associated with the presence of other grass species. So like I said, most of the other grass category is the native grass species. So it kind of makes sense that there would be more, there could be more forb diversity on sites that also have more native grass presence. Litter depth was negatively associated with forb species richness. So the more litter there were, less forb species. Oopsies. And the last, and forb species, Shannon diversity was negatively associated with smooth brome, which is interesting. And I swear, these plots are different. They look very similar. I, you know, that freaked me out the first time I saw it. So, but they are different. So on the forb community side of things, what we found was that litter depths and smooth brome were both significant contributing to this community. So what I thought was interesting here is that sweet clover, alfalfa, and this little triangle's can of the fissile really associated with areas of high litter depth when smooth brome. So that was kind of interesting. The functional results, the functional results, fourth core analysis gives you these like nice little grids that show the association or the correlation between these variables. So if it's blue, it's a positive association. If it's red, it's a negative. And if it's darker, it's more. If it's less dark, then it's less. And so this is the diet-breath category, nesting category, sociality category and body size category. So for the first one, some things that were interesting was that oligolectic bees, which are specialist bees, negatively associated with litter depth and Kentucky bluegrass, which I thought was interesting. Weirdly positive relations, weekly positive relationship with smooth brome. I don't, I can't, I don't, I'm still trying to think why that could have happened. One thing is like these kind of all, you have to think about results from other groupings and see whether that can explain things, which I'll kind of point out later. General estates, which are the polyleptic species, they were positively associated with forbidgeness or the only ones positively associated with forbidgeness. And that was, I mean, my best explanation for that, my best guess for that is that if you eat anything, maybe you benefit a lot from being in a place with lots of forbidgeness. I don't know if that's correct, but that's where I was thinking. Now the nesting category is really interesting because it's kind of nice when something just like works out the way you expect it to. And so ground nesting bees were negatively associated with litter depth. So they need access to the ground. That makes sense that they, in places where there's lots of litter, they can't access the ground much. Maybe there's less ground nesting bees. Also, we're less in places with higher Kentucky bluegrass. Opposite of that were cavity nesting bees. So bees that rely on like the presence of pithy stems and standing litter positively associated with litter depth. So that was kind of cool that that worked out that way. What was really interesting here in the social groupings is that the primitively used social bees, which are somewhat sweat bee species, negatively associated with litter depth, but you have to think that they're kind of all related. So you have to think of multiple categories. So those sweat bees are ground nesting bees too. So I don't know, yeah, that probably has something to do with those results. Also used social bees, which are the social ones like bumble bees and honey bees positively associated with litter depth and Kentucky bluegrass. And so like I said before, I'm like, well, maybe those species, well, maybe bumble bees, for instance, need more cover present for some benefit. For body size, small bees were positively associated with them out of bare ground. And like I said, it's all kind of related. So you have to think about, a lot of the small bees are, you know, those lazy to gossip species. So those primitively social species were ground nesting bees. And so that could make sense there. Medium sized bees, negatively associated with forbidgenose and pretty strongly so. But in this category are honey bees. And we have like, we had very many honey bees in our samples because this is more Dakota. And so honey bees are actually known to not really go so much for places with lots of diversity. They're more so influenced by the quantity of resources. And so that could be an explanation for that. So what we found is that really only forb species richness mattered in terms of those simple linear relationships, but there's more ways to analyze this. What we showed when we looked at communities that litter depths were also important here. And then what we saw is like, it really depends on the species. So like invasive grasses and other plant community characters like litter depth, just how they affect bees just determines on the type of bee that you're talking about. Litter depth had negative associations with forb species richness and smooth roam with overall diversity. So in other grasses, communities with other grasses were positively associated with forb species richness and diversity. Smooth roam and litter depth were important to shaping the forb community. So I think litter depth would be, requires a lot of attention from these results. So basically what we learned here is that, yeah, maybe floral resources are probably more important in explaining bee community, but we don't know that for sure. However, what we found was that plant community characters are associated with invasive grasses and just non floral aspects are important in shaping both bee and forb species. Kind of the star of the show, at least in my opinion is litter depth and how they affected both the bee and forb communities and have a strong association with them. And because both bees need different varying sources or varying types of litter depths, it's important that we maintain structural heterogeneity of the cross grasslands and that's comes from the heterogeneous application of disturbance processes such as this fire grazing Hain would probably disturb some things. But it's not unrelated to invasive grasses because if left idle, Kentucky blue grass and smooth brome are really good at making those thick litter layers and they're really good at making those thick litter layers like a logistic layer across the landscape. So especially if you do have invaded grasslands, you do really require disturbances. So this project was a first look at this topic for the Northern Great Plains at least. And so it provides a gateway for other research to perform. So since I'm looking at indirect relationships, an obvious kind of a logical way forward is to look at path analysis and stroke for equation modeling to investigate these indirect relationships, which I've thought about and I'm probably going to do, but I think it just requires more thought. I'm really scared of making bad models. And because I know how important floral community is, I need to add these floor measures into this mix because that could explain plant diversity maybe more in terms of how it relates to bees. Another thing is I need to make sure that my measures are sufficient for some variables. Also, I wanna just see how important are these non floral community characters to the floral community characters? Do we need to be even spending time thinking too much on it or is floral aspects more important? So that'll kind of be my next direction with this topic here. So this is a huge project and I had a lot of people involved with it. So I'd like to thank all the agencies that have served my land as well as all the private landlords who volunteered us to come chase things on their land. Also our field crew here, also lots of different lab help and other help as well as my current department for letting me come here today. So that's all I've got. I'd be happy to answer questions.