Improvising in Field Ecology

Research is the “diligent and systematic inquiry or investigation into a subject in order to discover or revise facts, theories, applications, etc.”. This definition is certainly reflected in published papers, with method sections that report orderly data collection to answer a specific scientific question. However, what those papers don’t show is the (oftentimes) unruly process and improvisation that takes place between coming up with a question and publishing a paper.

Harvey Mudd College’s summer research session is the perfect place for undergraduates like me to get a first taste of real research. Last year, Cassie wrote about issues developing bee foraging maps. Although the lab is one year older, and we, as a team, have gained more experience, the research process is far from trouble-free. For example, the series of honey bee experiments we planned and attempted during this summer research session demonstrate the amount of troubleshooting field ecology requires.

In the past, Prof Donaldson-Matasci has conducted similar experiments in different environments (please see Further Readings). However, true to the nature of research results, they have led to more questions.  This summer, we wanted to add to the results of her previous experiments and investigate the benefits of the honey bee’s waggle dance in a Southern Californian summer drought season. Furthermore, we aimed to expand on her previous research by identifying the types of resources the bees use during the barren summer months and implementing some of the tools being developed in this lab: the bee foraging maps, and the smart hive. 

The Bernard Field Station in summer: There aren't many flowers available for the bees.

Our original experiment plan looked something like this:

Question: Given two locations -- one with a homogenous covering of native plants, and one with a heterogenous distribution of native and non-native plants -- is there a difference in the value of communication? Furthermore, is there a difference in the resources the bee colonies choose to exploit?

Data to collect:

-       Daily hive weight change: tracks how well each hive is doing; collected with smart hive scales

-       Pollen collected by colony: reveals what resources the bees use as their pollen sources; collected once every week with a pollen trap

-       Bees’ waggle dances: shows us which resources the honeybees deem as valuable enough to dance for; videotape dances at the hives for 5 minutes at the start of every hour for one day, and then watch the videos to find out the distance and direction of the resource

-       Resources available throughout the experimentation period: areas will be photographed using Cassie’s drones, and supported by ground data

I learnt two things soon after we started the experiment: (1) our experimental plan was way too ambitious, and (2) running an experiment requires different “levels” of improvisation.

            What I mean by levels of improvisation is the experimental scale at which we need to think on our feet. For example, filming waggle dances turned out to be less straightforward than we would have liked. The glare from sunlight on the hives’ windows makes it impossible to capture any images of activity inside the hive. Therefore, while our last-minute orders of proper shelter were being delivered, we had to improvise anti-glare shelter with sheets Prof. Donaldson had lying around. 

Improvising shade with an extra sheet: This way, we can film honey bees in a hive with less of the sun's glare on the window looking into the hive.

When considering the whole experiment, improvising shelter for one task seems like improvisation at the smallest level. It is important to help us with data collection; however, building one instance of sun-protection does not affect the rest of the experiment.

            
When seven of ten weeks were over, we had one too many mishaps with our original experiment plan. The filming and analyzing waggle dances was too time-consuming, shuttling people between two research locations was time inefficient, and pack rats decided to eat the wires of some of our automated scales (because of its biodegradable coating?)! At this point, we decided it was time to scale down the experiment. Our final experimental plan looked like this:

Question: Given a location with a certain resource distribution (which we had to measure), does communication offer a benefit? Can we use the results from Prof. Donaldson’s previous experiments to predict the results of this summer’s experiments?

Data to collect:

-       Daily hive weight change, assuming two of our four remaining scales continue working

-       Pollen collected by colony as planned originally

-       Resources available throughout the experimentation period, supported by ground data

Although I feel a little sad that we couldn’t conduct our whole original experiment, I understand that we did not have the technology or manpower for such a large experiment. It wasn’t quite the right time, as the technology we wished to implement was still premature. Hopefully in the future the Bee Lab will have a chance to conduct something like the original experiment plan. Nevertheless, we finished collecting data, learned a lot about the lab’s limits, and still have something to show for this summer research session. And most importantly for me, I learnt that research requires much more than knowledge and expertise in a field. It requires the ability to keep calm and improvise, and the optimism and perseverance to keep going even if everything goes wrong. 

Further reading:
Donaldson-Matasci, M. C. & Dornhaus, A. 2012. How habitat affects the benefits of communication in collectively foraging honey bees. Behavioral Ecology and Sociobiology, 66, 583-592.  doi: 10.1007/s00265-011-1306-z.