 So our next presenter is Tyler Jenke. Tyler is a natural resource, natural area specialist with the Minnesota Department of Natural Resources. Tyler work involves acquisition, management and monitoring of public scientific and natural areas and privately owned native prairie bank conservation easements throughout Northwest Minnesota. Tyler is a native of Frazi, Minnesota and holds a bachelor's of science degree in the University of Minnesota, Go Gophers and a master's of science degree from University of North Dakota. I won't say anything there. So Tyler's talk is effectiveness of glyphosate herbicide on controlling non-native cool season grasses in a degraded prairie remnants. Let's welcome Tyler. Well, thank you very much for allowing me to be here and I look forward to sharing my project with you. So I'm Tyler from Minnesota DNR and I'm here today to talk about a project that I've been working on from, one of the last three to four years on and off. So in my project, it was looking at the effectiveness of glyphosate herbicide or roundup at controlling non-native cool season grasses in a degraded prairie remnant. So to start with just a little background in project objectives, as we all are abundantly aware invasive cool season grasses pose a major threat to biodiversity, particularly grassland biodiversity and control of widespread entrenched invasive grasses in prairies, at least in the area that I work in Northwest Minnesota, it's largely focused on varying the frequency and the timing of prescribed fire. And ultimately the results on that over the long-term have been mixed. There's been some successes. There's been cases where there's been no success but largely mixed results and no really clear paths forward with that. Also, it's become aware that a prairie managers in this area are often hesitant to use herbicides in controlling native grass in non-native grasses in these remnant prairies because of their concerns of impacts to non-target species. So this project, the study objectives were to quantify the effectiveness of a late fall application of glyphosate at controlling invasive grasses in a degraded remnant, documenting any impacts to non-target vegetation resulting from that herbicide application and then to share that information with other prairie managers in the region so they can make informed decisions on managing their lands. The study site where this project took place was the Crookston prairie unit of the Pembina Trail Scientific and Natural Area. So this is in Polk County, Minnesota, essentially straight east of Grand Forks, North Dakota for those of you not familiar with Minnesota counties. And the, this particular, so I'll give you a little background on scientific and natural areas. This prairie, unlike a lot of the sites we've heard about today, this prairie doesn't have to pay the bills anymore. It is owned by the state and it's managed for biodiversity and the biodiversity that's present on the site. The majority of the site is high quality lowland seepage prairie and saline prairie, but there's a dry ridge that was sand gravel prairie on the site that through some mechanism at some point in time became degraded. We don't know why and we probably never will. And this particular area on the site was the location of my study. So the study plots were located in this remnant dry sand gravel prairie with what we would say is a C to D condition ranking. Minnesota Biological Survey ranks the condition or quality of remnant prairies kind of on an A to D range. So CD isn't that good. It's not the worst, but it's an area this dry sand gravel prairie has problems. And as the manager of the site, I kind of looking at it and scratching my chin wondering what to do, realize that even though this dry gravel prairie has a lot of issues with bromine and bluegrass, it still has a lot of really desirable species. There were the presence of many things that we want to maintain. And just some examples of those are some of the prairie birds, foot violets, ground plums, pecuns, as well as a lot of other species at relatively low abundances were still present. And a couple of unique features of the site in addition to that, were that the abundance of the native gramanoids, especially native cool season gramanoids on this dry ridge was very low. So not a lot of porcupine grass, not a lot of june grass or other species, green needle, not a lot of those species were present. They're there, but they're very scattered. And starting in about 2015, I had been dabbling, just kind of experimenting with some grass-specific herbicides, particularly Cetoxidium, which goes by the trade name of post. Those projects largely failed in that the grass-specific herbicides, really all they did was stunt the brome for about one growing season, kind of hit the pause button. And then the second year following treatment, it's as though you were never there. So we didn't see any durable or lasting effects. That same issue existed with a late application of Clethodium, which is kind of the new grass-specific herbicide goes by the trade name of Intensity. We didn't see any effect of a fall application of that, which kind of goes with what other land managers have seen using those chemicals. But I had heard from some local practitioners that had utilized a really late application of Roundup in restorations to control bromine bluegrass. And they were reporting anecdotally some pretty good results. And I thought, hmm, I wonder if there's any, how would that apply to a remnant site where we have things we don't wanna kill, but we wanna try to get this bromine bluegrass under control? So that's a list, a little bit of background as to the study site and how we got where we were going. So my study design, this is a very simple study. I didn't spend a lot of time collecting data on this. It was meant to be efficient to where we could get things done. Sampling took about a day and then analysis maybe a few days. So it was meant to be more of an adaptive management project to inform us as we go. But the study design, we utilized two 50 meter transects that were located in areas of abundance bromine bluegrass. Pre-treatment data, plant species data were collected from both the treatment and control transects in early September of 2020. After that, the entire study area was mowed very short, kind of in mid September. After that mowing, we waited until we had at least two hard frosts had impacted the area. And we then sprayed using a UTV sprayer, a 3% glyphosate solution to the treatment transect, following those hard frosts. And that was basically it for the actual management work. But in the years following in 2021, and then again in 2022, we followed up with post-treatment data collection from both treatment and control transects to see how the vegetation changed in both the treatment and the control. Our sampling design likewise was very simple. Within the transects data were collected from 10 permanently marked meter square quadrants. All plants identified were identified to species and cover class estimates were assigned to each plant species in each quadrant. And for analysis, the midpoint of cover values were used. And I just kind of on the right here, kind of show a reference card that we carried with us one looking at each species. So we would ensure consistency in applying these cover values. So I wanna mention a little bit on, so why did we pick ball herbicide application? Seems like an inconvenient time to be spraying things. Some of the research that I've read on other invasive grass species have shown that fall application of chemicals tend to be more effective than this accompanying spring application. So there was some literature supporting that. Also, we felt that post-frost we would be less, there'd be less of the likelihood for us to impact those non-target species that we really wanted to maintain. Also, as we kind of are aware, these cool season grasses, they continue to grow after a frost. They have that cool season bump. I learned a lot about today with the tillering and those aspects. So thank you very much. And we felt that we could even mow the site and essentially force some of the native species that may still be there, force them into maybe an early senescence or an early dormancy and then allow those cool season non-natives, which we know are super abundant on the site to regrow following that mowing and giving us good herbicide leaf area contact. Other more practical considerations for fall where it's pretty easy to track when we get a freeze. It's on the news all over the place. There's freeze warnings and we can track that pretty easily through weather stations. It's much harder to know with certainty when individual species start growing in the springtime. So we felt, maybe this is all right, let's try this fall thing. And then also it's a less busy time of the year for us. In the spring, we're busy with prescribed fire and other activities and interns and student workers and all sorts of activities. Late October, it's easier to fit this in. So everything just kind of aligned into doing it at that time. So we need to have some results that we saw from this project. So mean bromine and bluegrass cover values were similar in our treatment and control plots at the beginning of the study. See here. But in the treatment plots, those that were sprayed, bromine and bluegrass cover values decreased significantly from year one to year two and then increased slightly in year three. But our year three values are still substantially lower than our initial year one values. Now in contrast, the bromine and bluegrass cover values in the control plots, again, they were the same, roughly the same as the treatment when we started, they increased slightly in year two and that increased substantially in year three. So the endpoints for where the bromine and bluegrass cover was in this project, they're vastly different. And I need to point out that the timeframe and the weather when this project spanned kind of represents some extremes. 2021 or year two in the project was a very, very extreme drought year. So it was very dry and seeing reductions or at least species staying the same, we weren't surprised. And then in year three, 2022, the spring was very cool and very wet. So I think we're seeing a little bit of that play out in the bromine and bluegrass here with this major increase in cover from year two to year three. I do think that these results, I think they clearly indicate that this late application, this October post frost application does suggest that it is effective at providing control on these species. But obviously control, this is an earth shattering. We sprayed plants with an herbicide and some of them died, but we're glad to know it was a question mark. Most people were wondering if it was too cold for the herbicide to uptake, but we were able to answer that and when we feel pretty good about it. So next we wanted to understand So next we wanted to understand, all right, how? So we've seen what this did to the bromine and bluegrass, what were our results regarding our individual species? So we started looking at functional group cover values and the functional groups that we assessed for this project included native perennial forbs, native perennial graminoids, native annuals and biennials and native subsrubs. So things like lead plant, curry rose, wolfberry. And what we found were the general trends in forb cover, we'll start with forbs. The general trends in forb cover tracks similarly in both transects. So here's our control year one, two and three and treatment year one, two and three. The general trend was the same in the control. They started high and they dropped substantially and then increased in the treatment. We saw the same impact. We saw the same effect where forb cover dropped in year two but that drop was much, much less severe in the treatment plot where we sprayed the bromine bluegrass. So where we in the control forb cover values decreased substantially following in year two which is the year of the severe drought. And then they rebounded in year three but never really at least yet haven't caught up to the cover values where they were at in year one. Our treatment, same general trend, the decline in forb is much, much less severe not statistically significant. And in fact, the rebound that took place after the drought we're now seeing forb cover values that are statistically more significantly higher than they were when we started the study. So our thoughts here were that this is indicating that the drought stress that happened was exacerbated by the presence of this high, high amount of bromine bluegrass. They're kind of resource pigs. They start growing early and they use a lot of resources and don't really leave a lot for other things. So what we're thinking this means is by effectively removing the vast majority of those from this area, the Forbes are really responding positively to the increased resource availability. The next cohort I want to talk about are native perennial graminoids. So our native perennial graminoids had very different trajectories. So these are down here. In our control plots, we basically where we didn't do any treatment, our native perennial graminoids decreased almost to the point of non-existence in year three, but in our treatment plots where we did the spraying the native perennial graminoids, they are noticeably increasing. Now none of these changes were statistically significant. The overall cover values of these species are still very low but the change is noticeable and it's promising. And what we were seeing in the treatment plots were some of the lesser known cool season graminoids, dicanthelium species, panic grasses, little sedges, which we believe to be one to be caraxumbalata. It didn't have much of an inflorescence and we couldn't tell. They're noticeably now reestablishing in this area and maybe they were there before and they were just so suppressed we couldn't see them, but we are now seeing an actual increase in some of these cool season graminoids. I just, I didn't graph for this particular presentation the trends in annuals and biennials and subshrubs because they basically track the same in the treatment and control plots. There was no difference. And I just left those out for brevity. But if anybody wants to see those, I'd be happy to share. Next, we decided, well, what about floristic quality? And Dr. DeCazer gave a good explanation of what FQI is and how it works based on species conservatism. We wanted to look at this because the initial thought was if herbicide impacts affected non-target species, maybe we could see that in FQI values because we'd be herding species that are of higher conservatism value and potentially seeing them replaced with more weedy species, which have lower C values. What we found was at the beginning of the study, FQ mean, plot-wise FQI values were about the same. Now there's always gonna be variability in certain different areas. We did see a drop in FQI values for both treatment and control in year one. Again, this is the year where we had the severe drought and this drop is not surprising because a lot of species entered dormancy early. Some of them didn't even start growing because it was really dry and this is a really droughty community anyway. I did find it interesting that FQI values did decrease more in the control where we didn't do anything. And then in year three, 2022, things are kind of back to normal. They've rebounded. So they always tracked similarly. We did not see a specific decline or more of a decline in the treatment area suggesting that at least from an FQI standpoint, the herbicide treatments did not impact any of our non-target species. Lastly, we did look at individual species responses. I apologize for not knowing how to graph individual species in a more visually pleasing way. But a summary here would be that individual species responses with regards to change in cover values. In our treatment, our species that decreased and cover the most were the non-natives that we were targeting, Roman bluegrass, as well as there was a few natives that did decrease, but there was a pretty low reduction in percentage. Whereas our increase of species, those species that increased their cover values in the treatment area over the course of the study, were native species that are pretty typical of this community, stiff sunflower, flax and other species that were already present. In comparison, the control and the control, the species that increased the most over this timeframe were the species we don't wanna see increase, the Roman bluegrass. And I think it makes sense. They have the most initial cover, so they have the most elasticity and ability to change to environmental conditions. Now as for the decreases in the control, it was again, just kind of a smattering of native and non-native species, but clear differences between treatment and control and again, clearly showing positive effects on Roman bluegrass in the treatment and then showing surprisingly increase in cover in the control for those species of interest. Now for species that just naturally have low cover values, most of the species you find in a prairie don't have really high cover values. Those are hard to measure using cover change because if a species has 0.75% average cover and it changed to 0.6, is that significant? It's really hard to tell. So we looked at frequency of occurrence, how the frequency of occurrence of species changed, particularly for those of low values. And in our treatment transects, we only had one species, native species or any species that actually decreased in frequency and that was solid ego rigida, which is stiff golden rod. And it was a pretty, from a frequency standpoint, it was pretty significant, being present in five of the quadrants, the start of the study, and then present in zero at the end. I will mention that even though that it was eliminated in the plots, it was still present in the treated area, those individuals that were in the transects seem to have gone. We really don't know if that was the result of drought stress. Maybe it is a herbicide impact. We don't know, but that's the only species we saw decrease. Now, in comparison, in our control, we had three species have significant frequency decreases. Oops, sorry. Three species in the control where we didn't do anything is violet, wood, sorrel, Western ragweed and yellow sweet clover decrease. Not really sure what to make of this. I think if anything, it just reinforces the fact that doing nothing has consequences too and things are gonna change whether or not we're fiddling with the plant community. So nothing stays the same, but we felt pretty confident that this, we didn't see any smoking guns pointing to that the herbicide caused a problem. And I'll also mention if in fact the Soledego rigida were impacted by herbicide, it's not a species I'm concerned about in an establishment consideration. It's easily established through seed and potentially seeding in from surrounding populations. So the species we were worried about the most, the violets, the cocoons, the ground plums, they're all still present and either at the same level or slightly increasing from when we started. So that was really, we're really happy about that. So in summary, we did find that late fog, life-side applications significantly reduced bromine bluegrass cover and bromine bluegrass cover values reductions have remained significant two years after treatment. The native forbs and graminoids appear to be responding positively to bromine bluegrass control and no negative impacts to functional group cover, individual species cover or fluorescent quality index values were clearly attributable to herbicide application. And but only one species in the treatment plots experienced a statistically significant reduction in frequency of occurrence in the treatment area, which was Soledego rigida. This photo was taken in August of 2022 and it shows that you can clearly see the spray area two years after we did it. All of this straw colored stuff is bromine fluorescences and you can see it still looks pretty good. Had to call out blanket flower to Minnesota. I know it's no big deal in North Dakota but blanket flower is a listed species in Minnesota. It's actually quite rare and it showed up in our treatment area. It didn't fall in a quadrant which kind of stunk but it was there so I had to call it out. So that was exciting to see. And lastly, this is the same view the year of the spring following application just to kind of showcase how variable it can be. In summary kind of conclusions the long period of dormant season growth exhibited by invasive cool season grasses is a strength of those species but we can also exploit it as a weakness. And that's what we tried to do here. We did find that glyphosate or Roundup can be used in a selective manner for some species by altering the season of application and implementing some simple straightforward site control measures. I do want to be really careful and say I'm not saying go out and everybody do this on thousands of acres but consider this another tool in the toolbox. There may be places this is appropriate. I encourage everyone to experiment with it and I would love to take people out to see this if they're interested. And we obviously have a lot of questions that came from this project. We don't know how long lasting the control effects are. That's kind of a universal question across all of these management types. How durable is the response? We know at least two years. This project wasn't set up to answer that question because it's a long and narrow transect. I think re-invasion from the sides is going to happen much faster than if it were a block. So I'll probably collect data one more year and then call it good. We also don't know if you could, can you do more than one treatment on the same area to try to go after the residual bromine bluegrass? We just don't know. And can we increase the resiliency and decrease likelihood of re-invasion of an area like this by overseeding some of those species that we know are already missing? So lots of questions. Thanks very much. I have to acknowledge the few troopers that helped me with the project and thanks so much for the opportunity to present.