P2 and P8 in the Winter

Stopped by P8 this morning and snowshoed out to P2. If you like these plots in the summer, the winter offers a new, unique perspective of the beauty of a prairie. The last few days, all vegetation has been covered in hoarfrost so the prairie lacks all color and seems like a black and white movie scene.

My dog Clyde (official spelling changed to Collyde). If you could meet him, you would understand.

On the south side of the P2, overlooking the plot.


2020 Update: Interremnant Crosses

This summer I started a new experiment to understand how the distance between plants in space and in their timing of flowering influences the fitness of their offspring. This experiment builds on my study of gene flow and pollen movement in the remnants, asking the question of how pollen movement patterns affect offspring establishment and fitness. If plants that are located close together or flower at the same time are closely related, their offspring might be more closely related and inbred, and have lower fitness than plants that are far apart and/or flower more asynchronously. In other words, if distance in space or time is correlated with relatedness, we’d expect mating between more distant or asynchronous individuals to result in more fit offspring.

To test this hypothesis, I performed crosses between plants across a range of spatial isolation (within the same population, in adjacent populations, and in far-apart populations). With the team’s help, I also kept track of the individuals flowering time so that I can assess whether reproductive synchrony is associated with reduced offspring fitness, suggesting that individuals that flower at the same time are more closely related.

I ended up using 42 focal plants (two of which were mowed before I could harvest them) and a total of 167 sires. I planted 359 offspring from these crosses in November. Next spring and summer, I will measure the seedlings to collect data on emergence and growth. Seed set was lower than I wanted it to be (only ~20%, when I would have expected 60-70% based on compatibility rates in the remnants), so I will also likely perform more crosses in summer 2021 to shore up my sample size.

Crossing at scenic On 27

Start year: 2020

Location: On27, SGC, GC, NGC, EELR, KJ, NNWLF, NWLF, LF

Overlaps with: phenology in the remnants, gene flow in the remnants

Data/Materials collected: 40 seedheads, style shriveling and seed set and weight from crosses, start and end date of flowering, coordinates of all individuals in the populations listed above

Products: I planted the seeds from the crosses in a plot adjacent to P1 in November, as detailed in this flog post.

What did we do in the summer of 2020?

Over the next few weeks we will be posting updates on projects from summer 2020. The team was able to accomplish an astonishing amount while enduring a global pandemic!

Team Echinacea 2020 says “wear a mask!”

Put a bookmark on our update page to stay caught up. We’ll post all updates on that page.

2020 Update: Seedling Establishment

This field season the team continued the seedling recruitment experiment begun in 2007. The original goal of the project was to determine rates of establishment and growth of seedlings in remnant populations of Echinacea angustifolia. From 2007 to 2013, plants which had flowered in the preceding year were visited in the spring to find any emerging seedlings. Each fall since then, the team has searched for the seedlings, then juveniles, and measured them.

It is awkward to refer to our group of plants former seedlings (all plants were seedling). It’s too long to say juveniles or seedlings. So, Team Echinacea uses ‘sling’ as shorthand for a plant that was initially found as a seedling (with cotyledons!) in one of our remnant populations. After finding the sling, we censused it annually up to the present, unless it died.

In 2020 Team Echinacea visited 66 focal maternal plants in 12 prairie remnants to determine the survival and growth of their offspring (slings). The team searched for 140 of the original 955 seedlings and found 70 of them and couldn’t find another 69. The searches spanned August 24th – September 18th and took place on seven days. The majority of the slings were searched for on the first day, August 24th, 77 slings to be exact. One of the slings flowered this year, however, no achenes were produced–the head was a dud.  

In 2020 Emma Greenlee investigated if siling survival can be predicted by its surrounding microhabitat. It turns out that something else is most likely affecting seedling survival not microhabitat.

Mia and Amy D. search for that last seedling at East Elk Lake Road

Sites with seedling searches
East Elk Lake Road, East Riley, East of Town Hall, KJ’s, Loeffler’s Corner, Landfill, Nessman, Northwest of Landfill, Riley, Steven’s Approach, South of Golf Course, Staffanson Prairie

Start year: 2007

Location: Douglas County, MN

Overlaps with: Demographic census in the remnants

Data/materials collected: The EchinaceaSeedlings repository holds the data for this experiment. Lea Richardson restructured the repo in December 2019 to facilitate collaboration on the new project.

The master datasheet and stakefile can be found here Dropbox/remData/115_trackSeedlings/slingRefinds2020. None of the 2020 data are online yet.

Team members who refound seedlings in 2020: Lea Richardson, Drake Mullett, Emma Greenlee, Mia Stevens, Anna Meehan, John Vankempen, Amy Dykstra, Stuart Wagenius.

Products: Amy Dykstra used seedling survival data from 2010 and 2011 to model population growth rates as a part of her dissertation. Scott Nordstrom has used some of the sling data in a manuscript that is now in review.

You can read more about this seedling establishment project, as well as links to prior flog entries mentioning the experiment, on the background page for this experiment.

2020 Update: common garden experiments

Since 1996, members of Team Echinacea have walked, crawled, and ~sometimes~ run next to rows of Echinacea angustifolia planted in common garden experiments. Although protocol varies depending on the experimental plot, every year team members record flowering phenology data, measuring data, and harvest the heads of the thousands of plants we have in common garden experiments.

Note that these experiments are not really gardens. “Common garden” refers to the experimental design. Most of our experimental plots are prairie restorations, a few are abandoned ag fields that are manged as grassland habitat. Some plots have multiple experiments within. The Echinacea Project currently has 10 established experimental plots:

exPts01-10. To avoid repetitiveness of reports on yearly phenology, measuring, and harvesting, this project status report will include updates on all experimental plots 1, 2, 4, 5, and 8. Reports for the others will be elsewhere: Amy Dykstra’s plot (exPt03), the hybrid plots (exPt06, exPt07, exPt09), and the West Central Area common garden (exPt10).

exPt01: Experimental plot 1 was first planted in 1996 (cleverly termed the 1996 cohort), and has been planted with nine other experiments in subsequent years, with the most recent planting being Amy Waananen’s inter-remnant crosses. It is the largest of the experimental plots, with over 10,000 planted positions; experiments in the plot include testing fitness differences between remnants (1996, 1997, 1999), quantifying effects of inbreeding (inb1inb2), and assessing quantitative genetic variation (qgen1). There are also a number of smaller experiments in it, including fitness of Hesperostipa sparteaaphid addition and exclusion, and pollen addition and exclusion. In 2020, we visited 4,340 of the original 10,622 planted and found 3,258 alive. Only 484 plants were classified as “flowering” in exPt01 this year. This is a drastic increase from the nearly 70 plants that flowered in summer 2019 – perhaps it is a testament to the benefits of controlled burning (we burned in spring 2020 but not in 2019). In summer 2020, we harvested ~815 total Echinacea heads in exPt01 (I have not finished the reconciliation process). In the fall, we added 484 staples to positions where plants were gone for three straight years, however, we ran out of staples, so 130 of these positions have “flaples” which are bent pin flags.

exPt02: Heritability of flowering time is the name of the game in exPt02. Planted in 2006, exPt02 was planted to assess heritability of flowering start date and duration in Echinacea. In summer 2020, we visited 2,010 positions of the 3,961 positions originally planted. We measured 1,638 living plants, of which 444 were flowering. In the fall, we harvested ~626 heads from exPt02. We do not have an exact number of heads harvested from exPt02 yet, as we have not had time to complete head reconciliation. Location: Hegg Lake WMA

exPt04: Experimental plot 4 was planted to gauge whether Echinacea from small remnant populations could be genetically rescued via an outcross to larger, more genetically diverse populations. Caroline Ridley and other members planted this plot in 2008. We did not measure exPt04 this year. Location: Hegg Lake WMA

exPt05: The only experimental plot planted at Staffanson Prairie Preserve (SPP), exPt05, was planted to compare progeny of maternal plants from burned and unburned sections of SPP. There were 2800 plants planted originally, but high mortality made it impractical to visit the plot row-by-row. Now, we and treat the plot like demography. We use our survey-grade GPS to find plants in exPt05 that have previously flowered and add more plants to the stake file if new plants in the plot flower. In 2020 we found 11 living plants, two of which were flowering! There was only one head to collect, since one of the flowering plants exhibited only vertical development (no head). Location: Staffanson Prairie Preserve

exPt08: Team Echinacea established quantitative genetics experiments to quantify additive genetic variance of fitness in Echinacea, with the idea that we can estimate evolutionary potential of study populations. The maternal parents of qGen2 and qGen3 are plants in the 1996, 1997, and 1999 cohorts. These plants were crossed with pollen from plants in remnants to produce seed for qGen2 and qGen3, which now inhabit exPt08. Originally, 12,813 seeds were sown in the common garden. Seeds from the same cross (shared maternal and paternal plants) were sown in meter-long segments between nails. A total of 3,253 seedlings were originally found, but only 562 plants were found alive in 2020. There were 5 flowering plants in 2020, and 5 heads. Note that there were an additional 2 heads collected from transplant plot.

Plot management: To ensure that the common garden environment is as similar as possible to the prairie environment we must actively manage it. This management includes removing foreign species and supplementing with natives. One of our main management methods is through fire. We were unable to burn plots this fall however we hope to burn p8 and p1 this spring. We also collected seed to spread after burns including Schizachyrium scoparium, Bouteloua curtipendula, along with multiple species of Solidago and quite a few Asters.

Asclepias viridiflora in p1: In 2019, plugs of an uncommon prairie milkweed, Asclepias viridiflora, were planted in Experimental plot 1. The purpose of this experiment is to assess the survival and fitness of A. viridiflora. Assessing vitality will also provide a frame of reference for species conservation across modern prairies. In 2020 a protocol was developed to identify and measure A. viridiflora. These data are waiting to be entered and analyzed.

Hesperostipa demography:  In 2009 and 2010, porcupine grass (Hesperostipa spartea, a.k.a. “stipa”) was planted in experimental plot 1. In total, 4417 seeds were planted, 1 m apart from each other and all 10 cm north of Echinacea plants. Between 2010 and 2013, each position was checked, and the plant status recorded. Since 2014, we have searched for flowering plants. The data from this summer can be found here cgdata/summer2020/stipaSearch, these data have not been processed yet.

Inb1: The INB1 experiment investigates the relationship between inbreeding level and fitness in Echinacea angustifolia. Each plant in experiment INB1 originates from one of three cross types, depending on the relatedness of the parents: between maternal half siblings; between plants from the same remnant, but not sharing a maternal or paternal parent; and between individuals from different remnants. We continued to measure fitness and flowering phenology in these plants. In 2020, of the original 557 plants in INB1, 111 were still alive. Of the plants that were alive this year, 30 (27%) were flowering. This is up from the 4% that were flowering last year. All individuals were planted in 2001.

qgen: The qGen1 (quantitative genetics) experiment in p1 was designed to quantify the heritability of traits in Echinacea angustifolia. We are especially interested in Darwinian fitness. Could fitness be heritable? During the summer of 2002 we crossed plants from the 1996 & 1997 cohorts of exPt01. We harvested heads, dissected achenes, and germinated seeds over the winter. In the Spring of 2003 we planted the resulting 4468 seedlings (this great number gave rise to this experiment’s nickname “big batch”). In 2020 we assessed survival and fitness measures of the qGen1 plants. 1,642 plants in qGen1 were alive in 2020. Of those, 17% flowered in 2020. All were planted in 2003.

Stuart and John measure Echinacea plants in exPt02 under a gloomy sky

For more information on survival in common garden experiments, see this flog post about survival in common gardens.

Start year: Various, see individual listings above. First ever planting was 1996.

Location: Various, see above

Overlaps with: Pretty much everything we do.

Data/ materials collected: Measure data for all plots. All raw measure data available in cgData repository. Processed data should eventually be available in SQL database; ask GK for status of SQL database. GPS points were shot for the exPt09 flowering plant. Find the GPS jobs containing the exPt09 corners here: Dropbox/geospatialDataBackup2020/convertedASVandCSV2020/EXPT9_20200917_DARW.asv The stake file to find exPt5 plants is here: Dropbox/geospatialDataBackup2020/stakeFiles2020/exPt05stakeFile.csv Products: Many publications and independent projects.

2020 Update: Seedling microhabitat assessment

As an intern with the Echinacea Project in summer 2020, Emma Greenlee conducted fieldwork for an independent project investigating whether microhabitat characteristics differ between 1 m-radius circles where Echinacea angustifolia seedlings have emerged and survived and circles where Echinacea seedlings emerged and died. An existing, long-term Echinacea Project experiment, the seedling establishment project (“Sling” for short) provided the GPS points corresponding to the surviving and dead seedling circles used in this project. Emma collected data on microhabitat characteristics (litter depth, vegetation cover, slope, aspect, distance to roads and fields, and community composition) and the floral neighborhood. Emma visited 69 maternal sling circles containing surviving seedlings and 66 sling circles where all seedlings were dead. In winter 2020, Emma conducted data analysis in R with help from Mia and Stuart, and plans to present findings at an ecology conference in summer 2021. You will have to read the presentation below to learn preliminary results or wait for the poster.

The floral neighborhood at Staffanson Prairie Preserve

Start year: 2020

Location: Remnant prairies in Douglas and Grant County, MN

Sites: East Elk Lake Road, East Riley, East of Town Hall, KJ’s, Landfill, Loeffler’s Corner, Nessman, North of Northwest of Landfill, Northwest of Landfill, Randt, Riley, South of Golf Course, Steven’s Approach, Staffanson Prairie Preserve

Overlaps with: Seedling establishment, EA fire and fitness 

Data collected: GPS files for navigating to sling circles are at Dropbox/geospatialDataBackup2020/stakeFiles2020. Microhabitat/floral neighborhood data and R scripts are available at aiisummer2020/emma2020. 

Products: Emma Greenlee’s Powerpoint presentation (below); poster to come!

Read more about seedling establishment on the experiment’s background page, or read more about the microhabitat project in Emma’s flog posts.

2020 update: demographic census in remnants

This summer Team Echinacea did demo and surv in 42 prairie remnants and other sites with Echinacea angustifolia populations. Demo involved measuring traits of individual plants: flowering status, number of flowering heads, and near neighbors. This summer we took 5119 demo records on our handheld data collectors (visors). Surv involved tagging individual plants and recording their location with our super-precise GPS (Darwin). This summer we shot 1494 points for surv. For ‘total demo’, we navigated to adult Echinacea plants that have been previously visited and took demo to generate detailed, long-term records of individual fitness in these fragmented Echinacea populations. At smaller sites we collected data on all adult plants and at larger sites we visited a subset of the adult plants. The demo and survey datasets are in the process of being combined with previous years’ records of flowering plants in “demap,” the spatial dataset of remnant reproductive fitness that the Echinacea Project maintains.

Start year: 1995

Location: Remnant prairie populations of the purple coneflower, Echinacea angustifolia, in Douglas County, MN. Sites are located between roadsides and fields, in railroad margins, on private land, and in protected natural areas.

Total demo: Bill Thom’s Gate, Common Garden, Dog, East of Town Hall, Golf Course, Hegg Lake, Martinson’s Approach, Nessman, North of Golf Course, REL, RHE, RHP, RHS, RHX, RKE, RKW, Randt, Railroad Crossing Douglas County, South of Golf Course, Sign, Town Hall, Tower, Transplant Plot, West of Aanenson, Woody’s, Yellow Orchid Hill

Annual sample: Aanenson, Around Landfill, East Elk Lake Road, East Riley, KJ’s, Krusemarks, Loeffler’s Corner, Landfill, North of Railroad Crossing, Norwest of Landfill and North of Northwest of Landfill (lumped), On 27, Riley, Railroad Crossing, Steven’s Approach, Staffanson Prairie

Overlaps with: Flowering phenology in remnantsreproductive fitness in remnantsEA fire and fitnessfire and flowering at SPP

Data: Dropbox/geospatialDataBackup2020 contains the experiment’s GPS files and the aiisummer2020 repo contains its demo records. The most recent copies of allDemoDemo.RData and allSurv.RData are accessed at Dropbox/demapSupplements/demapInputFiles.

Products: Amy Dykstra’s dissertation included matrix projection modeling using demographic data. The “demap” project merges phenological, spatial and demographic data for remnant plants.

For more information on demographic census in the remnants, visit the experiment’s background page, or explore flog entries that mention the experiment.

Emma using Darwin, our survey-grade GPS, in late August near Hegg Lake

Seedling microhabitat project findings

Hi again, it’s Emma––it’s been three weeks already and I’ve finished the majority of data analysis for my independent project! I presented about it at our lab meeting this morning and it was good to show what I’ve learned to the team and to get some helpful feedback.

To summarize my experiment’s goal, I was investigating whether there are differences in microhabitat between areas with surviving Echinacea seedlings and areas where Echinacea seedlings established but have died. This involved collecting data on site characteristics like litter depth, vegetation cover, slope, aspect, distance to roads and fields, plant community composition, and floral neighborhood at circles where seedlings monitored in the Sling project sprouted between 2007-2013. After analyzing my data, I can report that I found no differences in microhabitat between living and dead seedlings, and that I did not find differences in survival by prairie remnant, either. This suggests that the microhabitat variables I collected data on are likely not the most important factors driving seedling survival and mortality in this long-lived prairie perennial plant. Instead I propose that other factors, like climate, soil moisture & nutrients, pesticide drift, light limitation, herbivory, and genetics, may have greater impact on whether seedlings establish or die. Luckily the Sling project is ongoing and members of Team Echinacea are working to find out what drives seedling fitness in fragmented Echinacea populations!

I learned a LOT about doing data analysis in R during this project. I’m super grateful to Mia and Stuart for all the help they gave me when I had questions about R during the internship! The highlights probably are learning about, and doing, some multivariate analysis and using the R package vegan. It was so cool getting to create my own NMDS and species accumulation graphs after seeing them in many ecology papers I’ve read. From here I plan to do a few final analyses and edits with the intention of presenting my project findings at an ecology conference next summer.

That’s all from me for now! Stay tuned for a groovy poster…


2020 Update: reproductive fitness in remnants

Monitoring reproductive fitness in the remnant populations is a staple Team Echinacea summer activity. Understanding the reproductive success of plants in remnant populations provides insight to a vital demographic rate contributing to the persistence (or decline) of remnant populations in fragmented environments.

In the summer of 2020, we harvested 304 seeds heads from 29 populations (AAN, AAS, ALF-E, ALF-W, BTG, DOG, EELR, ERI, ETH, GC, LCE, LCW, NESS, NNWLF, NRRX, NWLF, ON27, RIN, RIS, RRX, SAP, SGC, TH, TOWER, WAA, YOH). These are the same populations where we measured flowering phenology. We randomly selected 15 heads from each population, if a population did not have 15 heads, we harvested all of the heads. We harvested heads from the following populations.

These heads are currently in the CBG lab and soon we will start the process of removing the achenes and assessing seed set. We are unsure how exactly we will assess seed set because the x-ray at the Chicago Botanic Garden isn’t working now. We may weigh the seeds.

Mia Stevens heading out to harvest
A harvested head

In the spring, we plan on burning some of these remnants and also collecting heads next fall. Estimates of seed set from these heads will serve as a baseline for comparing seed set before and after a burn. We will learn how fire affects reproductive success in small prairie remnants.

Start year: 1996

Location: Roadsides, railroad rights of way, and nature preserves in and around Solem Township, MN

Overlaps with: Phenology in the Remnants

Data/Materials collected:  304 seed heads were collected, these are currently at The Chicago Botanic Garden along with the paper data sheets. These data sheets need to be scanned, double-entered, and checked.

Products: We will compile seed set data from 2020 into a dataset with seed set data from previous years.

You can read more about reproductive fitness in remnants, as well as links to previous flog entries mentioning the experiment, on the background page for this experiment.

2020 Update: Flowering phenology in the remnants

In 2020, we collected data on the timing of flowering for 855 flowering plants (1071 flowering heads) in 31 remnant populations. The plants ranged from having 1 to 8 flowering heads. The earliest bloomers initiated flowering on June 22nd . Plant 22195 at NWLF was the latest bloomer, only beginning to shed pollen on September 14th, nearly a month after the second-latest flowering plant had ceased producing pollen (August 18th). As is typical for the latest bloomer of a season, township mowers had mowed over this plant earlier in the season, which is perhaps why it took longer for it to sprout a new flowering stem. Peak flowering was on July 9th, when 886 heads were flowering.

A major part of the motivation behind this year’s effort in monitoring phenology was to collect baseline data on flowering rates and timing. Team Echinacea recently received funding to perform prescribed burns in these populations. Next summer, we will compare flowering patterns in populations before and after fires to understand how burns drive the effects of timing of flowering on mating patterns and fitness of individuals in natural populations.

Start year: 1996

Location: Roadsides, railroad rights of way, and nature preserves in and around Solem Township, MN

Overlaps with: phenology in experimental plots, demography in the remnants, gene flow in remnants, reproductive fitness in remnants

Data/materials collected: We identify each plant with a numbered tag affixed to the base and give each head a colored twist tie, so that each head has a unique tag/twist-tie combination, or “head ID”, under which we store all phenology data.We monitor the flowering status of all flowering plants in the remnants, visiting at least once every three days (usually every two days) until all heads were done flowering to obtain start and end dates of flowering. We managed the data in the R project ‘aiisummer2020′ and will add the records to the database of previous years’ remnant phenology records, which is located here: The dataset is ready to be updated, but I don’t believe it has been at the time of writing.

A flowering schedule for individuals from all remnants. Notice the gap between when second-to-last flower ceased pollen production and when the latest bloomer began on September 14th!

A flowering curve (created here using the R package mateable) summarizes the flowering phenology data that we collected in 2020, indicating the number of individuals flowering on a given day and the flowering period for all individuals over the course of the season.

We shot GPS points at all of the plants we monitored. Soon, we will align the locations of plants this year with previously recorded locations and given a unique identifier (‘AKA’). We will link this year’s phenology and survey records via the headID to AKA table.

You can find more information about phenology in the remnants and links to previous flog posts regarding this experiment at the background page for the experiment.

Products: A dataset of flowering phenology is ready to be posted on the website. It is currently located in Dropbox\remData\105_assessPhenology\phenology2020\phen2020_out and is available upon request. The headIds in this dataset have not yet been merged with the akas (long-term identifiers) in the demography dataset.