It was about this time last week that I stared at a blank Word document, wanting to write more about my research experience but knowing that nothing I said would be genuine. Because honestly, I had hit a wall, and I had spent several days feeling anxious about the research itself and how little I had done to make any progress.
I reached out to a scientist who works at Wood Buffalo National Park to inquire about any location data they might have that they would be willing to share with me, and she responded quickly, sending me a trove of published data in the form of excel spreadsheets.
In the first spreadsheet I opened, there were 18 columns of information and 165,542 rows. At its most basic, I was looking at tens of thousands of data points, and most of the information was coded in language that made absolutely no sense to me.
So, I did the most logical thing. I quickly exited out of the spreadsheet and went about my day, trying my best not to think about it.
But this internship is the work I have always wanted to do, and not only that, but it’s also fulfilling a graduation requirement that is exceptionally difficult for me to fulfill otherwise. So when the date for my weekly check in with my research advisor was suddenly looming far too close for comfort, I finally buckled down and opened the spreadsheet back up, determined to pull something useful from it.
So I started with the things I knew. Okay – these are definitely dates and times, that’s useful and easy to understand. Let’s see how far this data goes back. I had date from April 2010 to October 2018. And this label, BirdID, had to mean the individual from which the data point was taken. There are 62 different IDs listed, so all of these data points only belong to 62 birds. And – oh – there’s a document here that actually clarifies what all these other things mean!
It wasn’t long before I was able to put together that I was looking at the location data over a period of several years from a total of 62 birds, and not only that, but this data spanned the entirety of their year, so much of it was pulled from the Texas Gulf Coast site as well as their migrations to and fro. But as I loaded this information into QGIS, just to visualize what I was looking at, I had to have a bit of a laugh:
This felt pretty useless to me. It didn’t reveal much of anything about factors for habitat selection. There was way too much data, and it was incredibly overwhelming. So I took a step back, and had to think about the data in a different way.
What information can I pull out of this by isolating different data points?
I started with the obvious. I filtered out all of the data points from the Gulf Coast and the migratory paths so that I could just look at what the researchers had deemed the “summering grounds”, AKA the site we were trying to specifically study. The problem here is that the data is still incredibly all over the place, but I remembered that there is a way to estimate these locations based on the use of kernel density measurements. We’ll leave that for another day; what else?
The biggest problem was that even after all of this filtering, I was still looking at over 64,000 data points. It was a huge glob of data on the map that made any reasonable conclusions impossible to make. But luckily, this data includes times and dates for each point. So perhaps the answer is to animate the data; make it move so that it follows individual birds over time, not just plotting each data point as if it was a separate individual bird.
My research advisor was excited at the prospect of this idea, and even told me that she had never used that functionality of ArcGIS. To be fair, I haven’t either, so I’ll be spending my upcoming week in research learning exactly how to do this. My advisor also suggested that I take a look at the migratory corridor to see if it changes with any significance. With my research growing ever closer to the dreamt-up ideal of research I did as a Freshman, I can feel myself becoming a better student, a more passionate academic, and hopefully a soon-to-be world class researcher.
As I regaled friends of my woes during the tough parts, a good friend assured me, “If you knew what you were doing, it wouldn’t be research.” I suppose that’s true. And especially as an undergrad, I don’t think there’s a whole lot of pressure for me to be doing anything groundbreaking right now. I’m learning the tools of the trade, and dipping my toes into actual research before I take the plunge in grad school. So maybe in the upcoming weeks I won’t be so hard on myself if I haven’t figured out everything by first glance.
erica kriner 