 So Drew is a research agronomist and agroecologist with USDA, ARS, and Mandan. He has his PhD in plant biology from Southern Illinois University and has research interests in biological diversity and ecosystem services in grassland and cropline systems. So let's welcome here Drew here to go give us a talk today about the long-term evaluation of planting treatment. So here you go Drew. So I'm going to be talking to you about a CRP restoration project in eastern Montana today and I have a very generic title because when I accepted to speak here today I had not analyzed any of this data. So CRP stands for the Conservation Reserve Program and the goal of this program is to give farmers a financial incentive to convert some of their marginal cropland to perennial vegetation with the goal of reducing soil erosion, enhancing water quality, and improving wildlife habitat. However, many CRP plantings are unsuccessful. There's difficulty in establishing stable communities of native plants because of things like drought conditions, large-scale erosion, and weedy plants. And so if we get these eroded landscapes with weedy plants that defeats the conservation objectives of this CRP program. So the Farm Service Agency approached some ARS scientists in 2016 and they asked us to do some research in low precipitation areas and how can we better establish CRP land. A couple things we're looking at in this first phase of research, the effect of the seeding year. And it's known that the conditions of when you seed can have a long-lasting effect on your plant communities. So here's an example from tallgrass prairie restorations. This older sequence, sequence one. Okay, there's a sequence one. Oh, no, this one here was started in a wet year and sequence two was in a dry year. And you can see that even after a decade, each point is a year, these plant communities remained distinct. We also know that seeding, how we seed, might have an effect on our establishment success. So we were looking at two treatments, green manure. There's been little research on this in CRP restorations specifically, but in crop lands we know this can increase organic matter and suppress weeds. And the other seed treatment approach we're looking at is having alternate grass and forb rows. Again, little research on this, but it's known that high grass density seeding can lead to lower diversity. So hopefully by arranging the planting where we have grass for grass, we can reduce that competition of those forb species and ultimately result in higher diversity plantings. So our research questions, does planted species diversity as measured by the effective number of species vary among seeding year and seeding treatments? And then if so, do richness and evenness, the two components of diversity differ? And then we also investigated, are there any differences in results depending on the evaluation method? Frequency counting, how frequent a plant occurs in these subframes, first density counting the number of individuals within a species. So today I'm focusing on these three sites in Montana at Sydney, Freud and Haver. And here's Sydney, Freud and Haver. So our treatments, we had two different seeding years, either seeded in 2017 or 2018. And we had four seeding treatments, and they all build on each other. So they increase in complexity from one to four. Seeding treatment one is just a standard seed mix. Treatment two is the standard seed mix plus the green manure, which was made up of pea and barley. And then treatment three is the standard seed mix plus the green manure plus the alternate for grass rows. And then treatment four is an enhanced seed mix with the green manure and the alternating grass for grows. So this is precipitation data from the two seeding years. The blue is the average of both years. The orange is the 2017 seeding year. And the gray is the 2018 seeding year. And I want to focus in on the May precip, because that's when the vegetation is first going to be establishing. These were all frost seeded. And you can see that at Haver, it was fairly similar between those two years. But at the other two sites, there's greater precip in 2017 compared to 2018. So this is our standard seed mix. It has four grasses and three forbs. We have blanket flower, purple prairie clover, scarlet glow mallow, western wheat grass, green needle grass, blue grama, and little blue stem. And then the enhanced seed mix can includes all of these standard seed mix plus these species here, wild bergamot, Canada milk vetch, slender wheat grass, Rocky Mountain penstemon, prairie June grass, Psydotes grama, prairie golden aster, prairie cone flower, purple cone flower, and switchgrass. So this one has eight grasses and nine forbs in the mix. So our sampling method, we used a frame like this to measure plant composition from 2018 through 2022. We recorded the frequency and the density of those planted species. And frequency is just a number of times it was present and one of those 25 squares, whereas density was counting the number of individuals in that entire frame. We didn't measure species composition during that first year of planting. And we had four frames set up diagonally across each plot. And then we aggregated that data to the whole plot level before analysis. And this frame is three quarter meters squared. And that's five by five subframe, so 25 total. All right, our statistical methods, we evaluated diversity using effective number of species, which is the exponential version of Shannon's diversity. So I like to think of this as the richness weighted by how uneven the community is. So that was our response. And then we also looked at richness and payloads, evenness as responses. And then we had fixed effects of seeding treatment, seeding year, and the interaction of those two. And the reason I went with this model is that I saw no directional trend in diversity with sampling here. So that didn't need to be included as an effect in the model. I used the LME function in R to do this with a random effect of site and autoregressive order one repeated measures and interpreted marginal p values. And in the future, we plan to do some multivariate analyses as well. All right, so for the diversity as measured by the effective number of species with the frequency method, we saw significant main effects, but no interaction. We saw that the enhanced mix with the green manure and the alternate rows, which is treatment four, had higher diversity than all the other seeding treatments. And that the 2017 seeding year had a higher diversity than the 2018 seeding year. And so this is that in figure form. So you can see treatment four, higher diversity than all the other seeding treatments, 2017 higher diversity than seeding in 2018. All right, and then we get the same interpretation using the density method, significant main effects where treatment four had higher diversity than all other seeding treatments. And the 2017 seeding year had higher plant diversity than 2018. So you get some very similar looking figures here. Again, treatment four is higher than all the others. And 2017 higher than 2018. All right, and then richness, the nice thing here is that both the frequency and the density methods will give you the same answer here. Again, we saw significant main effects, so very similar pattern as we saw with total diversity. And very similar looking figures, treatment four is higher than all the other seeding treatments, 2017 higher richness than 2018. All right, now we see something a little bit different in the evenness using the frequency method. Here we see a significant interaction. So the 2018 seed year had a higher evenness than 2017, but only in seeding treatment two. That's the one with the standard mix and the green manure. And then treatments four and three had a higher evenness than treatment two in 2017. And then in 2018, treatment four had higher evenness than treatment one in 2018. So here's that in a figure, you can see three and four in 2017 have higher evenness than treatment two. And then if you look within the seeding treatments, there's only a difference in treatment two where the evenness was higher in 2018. And then if you look at that same measurement but using the density method, there are no significant differences. So we got slightly lower diversity values using that density method compared to the frequency method. It's likely that the grasses were counted more often using that density method counting the number of individuals. We saw a higher effect size using that frequency method. So this might be a more sensitive method. So this is good news for me because I don't want to have to count all the individuals within every species. And we also had similar conclusions regardless of which method was used. So that gives us greater confidence that those diversity effects that we saw are a real effect. We also saw that richness showed the same pattern as diversity. So that suggests that richness is the stronger driver of diversity rather than evenness. And we saw that treatment three, the green manure and alternate grass for bros can promote evenness in some seed year conditions. So this might be a viable option if you're doing a restoration in an area where it's surrounded by other grassland. But it does seem like the best way to increase diversity of these CRP restorations is to seed more species. So this work was funded in part by the Farm Service Agency. And they've asked us to keep researching on this topic. So some of our future work, we're going to try to integrate these findings across all three of these regions. And then in a second phase of funding, we're looking at how weed pressure interacts with soil moisture to affect establishment of CRP. And in the third phase of funding, we're trying to answer, can we use precipitation forecasting and alternative seeding methods to improve restoration success. And with that, I'll take any questions. I know this is phase three, but do you know what kind of alternative seeding methods you might consider? Yeah, so we're looking at drilling all the seed versus broadcasting all the seed versus drilling the grasses and broadcasting the forbs.