2019 Pennsylvania Mountain Research: Emelyn Piotter and Maya Rayle

A Snowy Start!
Emelyn
So far this field season has been quite different from last years. The summer of 2018 was an extremely dry year and this summer is much wetter than usual. At this time last year virtually all the snow on Pennsylvania Mountain had already melted, flowers were in peak bloom, and pollinators were out all over the mountain. Because of all this winter’s snow the plants and pollinators are getting a much slower start. Right now on Pennsylvania Mtn. bumblebee queens have just started emerging from their winter hideaways underground and only a couple of wildflowers have started to bloom. We even need snow shoes to access many areas. A lot of the research we do on Pennsylvania Mtn. involves studies that span multiple years so the extreme difference in the conditions between last year and this year allow us see how plants and pollinators are impacted by different temperature patterns.

Because dandelions, bumblebee workers, and bee flies aren’t out yet I can’t do much work on the research I have planned for the summer other than writing up methods for how we plan to conduct different studies and making sure we have all the materials we need. It will likely be two more weeks before the native dandelion species starts to bloom. The slow start isn’t unwelcome though as it’s giving us ample time to adjust to the altitude. Until the dandelions start blooming I will primarily be helping Zack, one of Dr. Galen’s graduate students, with his work on bumblebee acoustics. I have spent the last couple of days setting up bee boxes (which are essentially bird houses but for bumblebees), monitoring them for activity, and taking note of what flowers are blooming in the area. I also helped analyze some of last year’s bumblebee acoustic data and got to learn the basics of software program called “Audacity” that we use to look at audio files.

Flower Phenology
Maya

Over the past two weeks, we’ve been locating flower plots, which are marked out areas scattered across the mountain where we count selected species of flowers. The flower plots are part of a long term monitoring project on the mountain which shows when each species of plant reach peak flower abundance, when the first flowers from each species occur, when the last flowering occurs and how long it lasts, and it shows differences in flowering across the mountain. This is really important data to monitor the long term impacts of climate change on alpine flowering, as well as yearly differences in weather. We also use the data to see whether peak flower abundance matches peak bee activity. Mismatches in bee activity and peak flower abundance is a huge problem because the bees don’t get as much food and the flowers don’t get pollinated as much.
There are 25 different flower plots all over Penn mountain. We use the same locations year after year. Each flower plot spans roughly 30 meters by 10 meters. In order to locate them, we used GPS coordinates. Once we get to where a plot is supposed to be, we look around for last year’s flags, then lay out the tape measure and remark them with new flags so they’re easier to see. Some plots are missing a lot of flags, so we had to redo sections of these plots by lining up right angles and remeasuring. Many of the plots are still covered in snow, so we can’t get to them yet.

We just started counting flowers in the plots. We set up a grid with a predetermined origin, and use the long end of the plot as the X axis and the short end as the Y axis in order to keep track of how many flowers are in each specific location within the flower plot. I’ve been getting a lot better at identifying my flowers because of the flower phenology. The colors are so vibrant and I’m enjoying being able to recognize species. We’ll continue counting flowers in flower plots once a week all season.

Dandelions and Bumblebees
Emelyn

There are two main species of dandelions in the Rocky Mountains, Taraxacum officinale and Taraxacum ceratophorum. T. officinale is the introduced exotic species found all across the United States. It is very common and can be identified by the down-turned bracts on its flower and leaves with deep backwards pointed serrations. T. ceratophorum is only found in alpine habitats and has smaller, more rounded leaves and upward-pointing bracts. Perhaps the most interesting difference between these two species is that T. officinale is apomictic, meaning it is able to set seed without being pollinated. In spite of this it still produces pollen and nectar and is an important resource for bees and other insects. T. ceratophorum does not readily self and requires pollination to set seed. Because my project is looking at how different insects compare as pollinators most of my experiments will use T. ceratophorum.

This week Mackenzie, one of Dr. Galen’s research assistants, and I scouted out Pennsylvania mountain for the first signs of bumblebee workers and bee flies. We visited the “bowl” located near the Pika Trail, which is one of the warmest places on the mountain and one of the first places dandelions start to bloom. This area only has T. officinale so it’s role in my research is limited. However, the area does attract a lot of pollinators so we used it as a gauge to see what species have emerged so far this season and to practice our insect catching technique. We found a lot of solitary bees including some megachilidae, also know as leaf cutter bees. I’ve noticed that megachilidae are one of the more aggressive groups of bees and will try to sting you through the net once they have been caught. We also found several syrphid (hover) flies visiting the dandelions. The flies are usually harder to catch than bees because they are more warry and easily spooked from flowers. This is probably because, unlike bees, they can’t sting so they have less of a defense mechanism if they get caught. We didn’t find any worker bees or bee flies in the bowl so it will still be a few more days before I can get started on my research

Life of a Bumblebee
Maya

Since our work focuses so much on bumblebees, I thought it would be worthwhile to take the time to explain the bumblebee life cycle and a little about bumblebee behavior.
Bumblebee queens are born the previous summer, and spend the winter underground. They generally burrow six inches or so into the dirt, and live off of fat reserves. Alpine bumblebee queens face extreme winter conditions, and are often buried deep under the snow. Queens are much larger than workers, which helps them survive the cold. There’s a rule in ecology called Bergmann’s Rule, which states that across geographic regions and across different species, colder climates tend to have larger animals, and warmer climates tend to have smaller animals. This is because a larger body means a lower surface to volume ratio, which helps an animal to conserve heat. Bumblebees are generally large and queens are especially large which allows them to survive in colder environments.

In the spring, when the temperature warms sufficiently, the queens emerge to feed off of nectar. Then, they begin to hunt for a nest site. We’ve seen a lot of queen bees hunting — they fly low to the ground, moving along methodically as they search for holes in the ground. We’ve been placing wooden boxes with an entrance and exit hole and a thick layer of cotton inside to provide a good artificial nest site similar to a natural site.

Once a queen finds a nest, she makes frequent trips all day, each trip lasting 30 minutes or so to collect pollen and nectar. She’ll bring this back to the nest to either save it or consume it. She builds a wax mound (she can secrete wax from her body), and then lays her first round of eggs. It takes about three weeks for them to mature into larvae and then into worker bees. During this time, the queen sits on the wax mound to keep the larvae warm by shivering her body. Bumblebees are excellent thermoregulators.

After the first workers are matured, the queen will hardly leave the nest. She’ll spend most of her time laying and incubating eggs. Worker bees either forage or guard or clean the nest. Worker bumblebees that forage split into three working groups: nectar gathering, pollen gathering, and a group that does both and fills in where needed.
Later on in the summer, the colony produces male bees, which go out to attempt to mate and do not return to the nest, and new queens. When the new queens mature, they leave the nest to find males to mate with, then they spend the rest of the summer feeding on nectar and pollen to store up for their winter underground. Queens only live one year, so the old queen and the rest of the colony dies, with only the new queen surviving. Sometime in the fall, the new queens will burrow, and the cycle starts again.

Fishing for Pollinators!
Emelyn

The season is now in full swing! The native dandelion species, Taraxacum ceratophorum, is starting to bloom all over the mountain and I saw a bee fly for the first time this season a couple of days ago. Bumblebee queens have established their homes for the summer and most have raised their first brood of worker bees. Several days ago we tested some methods that we will start using for actual data collection this week now that bee flies and worker bees are out. We call the method we tried out “fishing for pollinators”. This method consists of using an old fishing pole with two wire loops at the end designed to hold water picks with flowers in them. This device can be used for a variety of experiments but the fishing pole contraption is most often used as a choice-test for pollinators. We put flowers that are different in some way (in this case one dandelion has a black pin in it to mimic a spider on the flower) in the water picks at the end of the fishing pole. We then slowly present the fishing pole with flowers to an insect (we tried it on a syrphid fly but it can be used on ants, bees and other flies) in hopes that it leaves the flower it is on and moves to on of the flowers attached to the fishing pole. If the insect likes one flower option over the other it will most likely move to the flower it prefers. Some examples of preferences you could test with this method are preferences for flower size, flower species, presence of a predator (or pin that looks like a predator), etc. This method requires a lot of patience and a bit of luck. You have to move slowly so you don’t scare the bees or flies off and once you are in position you have to wait until the insect is ready to leave the flower its on and find a new one.

Acoustics of Flower Visits
Maya

Every Friday, we set out microphones to do acoustic monitoring of bee activity on flowers. In the flower plots that have all of the flowers we monitor (there are eight or so flower species that we’re interested in), we visit each plot to set up microphones. Right now, we to to three flower plots.

For the past few Fridays, Zack and I have gotten up early then hiked up Penn Mountain. We first go to a flower plot above the second slope, then one near the false summit, then one near the true summit. In each flower plot, we check which flowers are blooming. Even though these flower plots have all of the flower species at some point in the season, not all of the species bloom at once. Last week, the plots had trifolium nanum, polemonium viscosum, and mertenzia.

Once we check which species are blooming, we pick out the patch of flowers from each species with the most flowers blooming to set up a microphone. This involves bending a wire into a twist to hold the microphone, leaving the microphone suspended above the plants. It’s best to place the microphone a little ways away from the flowers so it doesn’t physically interfere with bee activity. We also place a muffler on the end of the microphone to reduce wind sounds. You can see the setup in the photograph.
We then leave the microphones out for 6 hours or longer before Emelyn and Mackenzie go to pick them up. After that, we’ll send the audio to a collaborator who’s come up with an algorithm that subtracts the audio from a nearby control microphone where there are no flowers from the microphones at each flower species. This gets rid of background noise such as airplanes, wind, and any other odd sound that isn’t a localized buzz.

The purpose of this microphone experiment is to listen to bees foraging on flowers. Our microphones will pick up bee buzzes within a range of one meter, so we can tell when the bees are foraging on the chosen species of flower. We can learn about bee preferences, what sounds they make when they forage, and we can hear arrivals, departures, and visitation buzzes.

One of the most important pieces of information we get out of this is to measure if peak bee activity matches peak flowering. We get our flower data from flower phenology (counting flowers in plots each week), and bee activity comes partially from this. Since last year was so dry, there was a large mismatch between peak flowering and peak bee activity, such that the flowers peaked long before the bees did. This can be really hard on colonies because they rely on flowers for food. Studies suggest climate change can exacerbate mismatches because plants and bees use slightly different biological indicators for when to begin flowering or to come out from their dormancy period. Our work with microphones will help to measure mismatches in the short term, season to season. A mismatch caused by last year’s drought is not necessarily climate change related, but since climate change is predicted to cause more frequent droughts and warmer temperatures on average, a comparison between a drought year and a normal or even snowier-than-average year like this year is useful for studying the effects of climate change on bee flower mutualisms.

Pollen Swabbing
Maya

At each of the three bee box sites at different altitudes, we pollen-swab workers. The purpose is to find out which flowers they’re visiting throughout the season. To catch the bumblebee workers, we sit a meter or two from the bee box with a butterfly net placed underneath the entryhole. When a worker bee flies into the colony, we catch it in the net, then transfer it to a falcon tube, then place it in a cold lunch box. Once the bee has cooled off, it won’t be able to fly away so it’s much easier to handle. Then, we use a small cube of jelly (made out of gelatin) that’s just a few millimeters wide in order to swipe the bee’s body. We’ll then store the gelatin cube in a labelled microcentrifuge tube to make slides with back at the house.

We’ve also been collecting pollen from different families of flowers in the area so we can identify pollen from the bees — a pollen library of sorts. We’ve been collecting pollen off of workers once every two weeks at each site. This will provide useful information about what flowers workers are foraging from, and how this changes throughout the season.

Who’s the best pollinator?
Maya

For the past few weeks, Emelyn and I have been working on a few different experiments testing the pollination efficiency of bees compared with beeflies. This is a continuation of Emelyn’s work from last year. The question she’s asking is whether bees are better pollinators than beeflies. Based on her work from the past year and field season, it appears that bees are better pollinators than beeflies. The next question is why — is it body shape? The morphology of the bees compared to beeflies, or the hairs on both insects? Last summer, Emelyn and Austin (the PhD student who was a MALT intern last year and has been mentoring Emelyn on her project) observed differences in foraging behavior between bees and beeflies as they were conducting their experiments on bee versus beefly hairs. Part of our job this year is to prove that there are differences in foraging behavior between the two insects.

One thing we’ve been doing to accomplish this is setting up cameras above patches of dandelions and leaving them running for a few hours, so they observe bees and beeflies foraging on dandelions. We also measure the size of the dandelions to keep this in account.

Another thing we’ve been doing is catching wild bees and beeflies that are foraging on dandelions then swab them for pollen. This way, we can count pollen grains and compare the numbers for bees and beeflies. It’s not perfect, since we don’t know how many flowers the bee or beefly has foraged on before we catch it, but it gives a good idea of differences in pollination efficiency between the two insects. Furthermore, we’re collecting a number of replications of each insect, which should minimize any effect of natural variation on the data. More replications strengthen the findings.

The last, and most time consuming part of the experiments we’ve been working on is tent experiments. We release a bee or beefly inside of a mesh tent, then wait for it to forage on two dandelions consecutively. We keep the dandelions from being pollinated by any other insects by keeping them in the tent or covered with a mesh cloth. We usually collect buds that haven’t yet opened two days before the experiment, so we know they haven’t been pollinated by anything else. Once the bee or beefly has foraged on one dandelion (the donor dandelion), we wait for the insect to choose a second flower (out of the four that we place in the tent) and then forage. We film the entire thing so we can keep track of information such as forage time and duration between the donor flower visit and the recipient flower visit.

It’s been really interesting helping to conduct these experiments and I’m certainly learning a lot! We’ve had to tweak certain things to make it work better, and keep track of data and take good notes.