In summer 2021 and summer 2022, we investigated fire effects on Liatris aspera across 30 patches of remnant prairie in Solem Township, MN. We did not collect any new Liatris data in 2023. All harvested samples were processed in the lab during spring 2023. All data has been curated and are analysis-ready.
Had to go back to 2021 to get a Liatris photo, good riddance!
Start year: 2021-2022
Location: 30 patches of remnant prairie in and around Solem Township, MN
Data collected: Spatial location, reproductive effort, and reproductive outcomes data have been curated in the remLa repo. All data are analysis ready.
Samples or specimens collected: All harvested reproductive material have been processed in the lab. Half of the collected seed was sowed in the recently burned southeast hill at the farm. The other half have been reserved for sowing in P2 (provided it burns in spring 2024).
Products: Jared has conducted preliminary analyses. Stay tuned in 2024 for more info!
You can read more about the Liatris fire and flowering experiment, as well as links to prior flog entries about this experiment, on the background page for this experiment.
It has been an absolute pleasure getting to work on this project. After many weeks of hard work, I presented the final poster I made, going over the details and findings of everything we did to learn about how Liatris head count affects predation. We did find some evidence that there is an effect on predation rate based on headcount, but the evidence itself was not strong enough. We did fail to reject the null hypothesis, but that did not discourage me from thinking about other things that could have impacted the results we found. I mentioned that confounding variables such as soil richness, fires, and rainfall may have affected our results, especially given the differences found between 2021 and 2022. These confounding variables are worth looking at, making me curious to learn more about how they influence Liatris.
With that, there is so much to learn still, and it does not only apply to Liatris. Other prairie plants, such as the Big Bluestem (Andropogon gerardii) and Milkweed (Asclepias viridiflora), are being looked at by the Echinacea Project, and there is so much to learn about them. Keep up the excellent work with what you do, and most importantly, keep learning how to protect our precious prairie ecosystems and plant species. You are their only hope!
Lastly, I enjoyed getting to know all of you and learned a lot while I was here. Conservation science has been something I have always been interested in, so having the opportunity to research prairie plants was something special. I want to thank Stuart Wagenius, in particular, for letting me intern here and learn the ways of scientific research in a professional setting. I look forward to hearing what comes next from the project, and I sincerely thank you!
Remember when I talked about only having three batches to work with for the 2022 Liatris data? That is now no longer the case. Thanks to several volunteers continuing to chip away at the remaining Liatris plants in 2022, batch four got officially completed last week. Now with the newly completed batch, I managed to update 2022’s Liatris data in the figure above. It might not be noticeable just by looking at the graph, but the slope and p-values changed quantitatively. The p-value for 2022 now sits at roughly 0.09, which is better than what we had prior (I think it was 0.16 if I remember correctly). However, that is still not as good as 2021’s data, and that data still holds the better slope between the two. Outside of adding the fourth batch, I also tweaked the graphs to be more easily read and understood. I added some color to the data points, reworded the axes, and made the linear regression line bolder for all of you to see better. Overall, these were the updates I have to share with you, but I also wanted to inform you that I have started the process of creating the final poster. Over the course of this week, I will focus primarily on completing the poster, and I hope to have it ready to go by the last week of April. I will be presenting it to you all on April 27th (subject to change), so I hope you look forward to it. It has been quite a journey for this project, but it sure has been an interesting one, to say the least.
We have data! So far, we have seen some fascinating results from the 2022 data we worked on over the past 6-7 weeks and the 2021 data collected last year. The data shown above is still in its preliminary stages, and there is still some additional work to be done, but we can make out some things with what we got so far. First, 2022’s data is on the left, and 2021’s is on the right. Comparing the two, we did find that the predation rate, on average, was higher for the 2021 plants than those in 2022. We also created a linear regression line for both years to examine the correlation between headcount and predation rates. 2021’s data had a higher slope than 2022’s, meaning the data in 2021 had a more substantial representation of the predation rate increasing with a higher headcount than in 2022. Overall, these findings suggest that the plants in 2021 saw much more predation activity than those in 2022. With that in mind, it makes me wonder why predation rates are higher in 2021 than in 2022 and what the results would look like if we were to combine the data. This is only the beginning of our data collecting and interpretation, so I’m looking forward to learning more about the results we got and if we can reject the null hypothesis in the end. Note that the null hypothesis states no relationship between headcount and predation rate.
We are so close! We are now a little over two-thirds of the way done with the third batch, and things are starting to get exciting. I want to thank those who helped with randomizing; we would not have gotten this far without your help. With that said, there are now less than 20 plants left to randomize, and we are just about ready to move on to the data entry and interpretation steps. We should have randomization wrapped up by next week, so we should hopefully have our results in by next week. After that, we will then analyze the final results and begin to construct a final poster. More details will be provided for the poster, but in the meantime, we are quite excited about the next steps once randomization is complete.
I know it has been a while since I last gave you all an update, but we have some good news to share! We managed to finish cleaning the third batch and are close to completing randomization for the second batch. I will say our progress has been pretty steady before the last update I shared with you, but we decided to out some ways to accelerate processes, such as randomization. We created an assembly line (shown in the image above). We tasked people to either set up the stations by pouring and spreading out the achenes onto the randomization grids or randomly select grids in which achenes will be inspected for predation. We found this setup to be much more efficient, and hopefully, by the end of next week, we may have three batches fully cleaned and randomized. However, our initial goal of completing five batches may be somewhat too ambitious, so instead, we decided to just focus on three batches so there may be time to collect and analyze the data. In the figure below, you will find our overall progress made on the project, and given that my focus would be on studying predation rates on Liatris plants, the stuff that comes after randomization does not necessarily apply to the findings I will need. With that said, those other sets of data will most likely be of use for other studies, and the Echinacea Project will most certainly take a look into those things. Overall, things are looking good for us with the project, and I cannot wait to see the results soon.
P.S., I did not have a chance to specifically address my project’s intentions in an ABT prompt, so here it is: “We know that Liatris plants get eaten more often than Echinacea plants, and predation rates tend to be higher in the absence of fires, but we do not know which types of which types of Liatris plants may be more likely to be eaten than others. Therefore, we are studying how the number heads of a Liatris plant affects the predation rate on the plant’s achenes.”
We made it to the third batch for the cleaning procedure! After completing the first batch for randomization last week, we needed to pick things up on cleaning, and we hit another promising milestone. Ultimately, we wanted the cleaning process to stay well ahead of the randomization process, and we did manage to get ahead of ourselves some on randomization last week. We made much more progress in that department than I had envisioned, and we caught up a little too much concerning cleaning progress. We focused on cleaning all this week, and now here is where things stand. There is still much more to go, but we are steadily getting there.
Good news! We managed to get through randomizing the first batch! Now we are working our way through the second batch, and so far, things have been going pretty well in terms of pace and efficiency for both cleaning and randomization. Overall, we have seen a considerably wide range of head counts in the Liatris plants, but predation rates have been relatively low compared to what was observed in 2021’s data. Who knows what we will find in the next several batches, but I am pretty excited about the progress made so far and the results we may see. Hopefully, we will complete the next batch within the next few weeks, if not sooner.
So far, things are going well in the randomization step, but there is still quite a ways to go to complete the first batch of Liatris plants. However, we have counted a considerable number of achenes that have been eaten, and we are starting to get a rough estimate of the predation rate per Liatris plant. We are finding that there are, on average, 1 to 3 achenes eaten per plant out of a total of at least 30 randomly selected achenes. This equals approximately a 3 to 10 percent range for what has been found, but we have also found a few that have had either none or many achenes eaten. In light of these findings, I think we are heading in the right direction with the project, but hopefully, we will need to get much more work done to get through all five batches. Hopefully, we will have all the plants cleaned and randomized before the burning season (which starts in April), but it will take a lot to get to that goal within a little over a month’s worth of time. However, the good news is whether or not we meet our goal, there will be more than enough data to work with, and ultimately we can come to conclusions at the end.
In lab today, I started the process of randomization. This step in the project does take some time and patience, but all of it is worthwhile toward reducing bias and collecting achenes that can be x-rayed or identified as eaten. Similar to the randomization of achenes from Echinacea plants, sheets of randomization and counting grids are used. In addition, we also use plastic bags, white envelopes, stickers, tweezers, and a magnifying glass for this procedure. The first step involves pouring the achenes evenly across the randomization grid and ensuring all the achenes are placed in a specific square on the grid without being on the lines. Then using a randomized list of grid coordinates, we go down the list until we hit a coordinate (an example would be “B2”) with achenes present. We then examined each achene in the selected coordinate to see if there was any predation. If less than 30 achenes were counted in the coordinate, we randomly selected another one going down the randomization list. Once we categorized 30 achenes as either eaten or eligible for x-ray, we took our white envelope and plastic bag and put stickers on them. The envelope will store the eaten achenes while the plastic bag will contain the achenes ready to be x-rayed. On the envelope, we write the date, initials, and the quantities of achenes eaten, not eaten, and uninformative. After putting the achenes in their respective containers, we put them in a pile marked “randomized,” and we move on to the next one. As of today, we managed to get through a little over one third of our first batch, so we still have much work to do before proceeding onward.
