This past winter, students in my 1st-year course, Climate Change: A History, transcribed diary entries from the James Barry diaries. Barry made observations on the weather, and its effects, virtually every day and the students entered data from the period 1880-84 into a database. The results were perhaps problematic – more below – but quite spectacular. As a lesson in how climate proxies can help us understand modern climate change, it was in my estimation a major success.
That sounds straightforward enough, but the technical challenges of navigating three hundred and twenty students’ data into a single manageable file while working through the Brightspace LLM required support. Much appreciation to Giulia Forsythe and Melanie Elliot of Brock’s Centre for Pedagogical Innovation for crafting that for me. I could not have made this work without their enthusiastic support.
The students were each given a single page from the diary from the period 1880-84 (roughly 320 pages, with three to six entries per page) and they were to create and then upload a transcription of the page and a completed Excel file. The assignment submission then consisted of three files: the transcription, the completed Excel page and a short report/reflection on what they found. In a second assignment they created a new database for the same dates from the second diary, and wrote a short essay comparing the two and discussing the value and limits of such data and comparisons, as well as their method. They also had a seminar based on readings by Renée Sieber, Victoria Slonosky, and Lisa Piper as guides for thinking through some of the issues.
I’ll later write something about the pedagogical challenges and limits, but it was a good exercise that allowed students to work with proxies, to see how comparisons offer scale possibilities, and particularly for seeing behind the curtain of how climate data is made. Participating in the creation of data, seeing the good-faith and often very thoughtful actions of your fellow researchers, and getting an impressive set of outputs made a strong impression on most of the students. Transcribing cursive handwriting was a challnge for students no longer trained in that practice, but the vast majority pulled it off well.
After the data was assembled into the database, we created a series simple graphs . Below, the first is the raw data. Some students entered a range for their temperature estimates and so this graph uses their higher estimate. Only a few did this, and with 320 students, it really had very little measurable difference. The range sometimes suggests some problematic data . For example, 28 degrees would be a close to a record high even today for mid-October Pictou, and -10 would be equally unlikely for mid-April. Still, the generally lower figures for April and May tend to confirm something I’ve observed: that spring in the second half of the 19th-century was about two weeks slower than today.
We then created averages for the period 1880-84:
To then compare that data with today, we looked for comparable Environment Canada data. Unsurprisingly, there’s no SIx Mile Brook data. The best complete dataset is problematic: it’s from Caribou, which is along the coast of the Gulf of St Lawrence and thus undoubtedly moderated compared with an inland location. There’s also data for Lyon’s Brook – off Pictou harbour, but much further inland – but it only goes to the year 2000, thus missing the period we assume has the sharpest rise. Geographically, Lyon’s Brook is a better comparison, but I suspect it underplays the historic change.
In both cases, but more evidently in the case of Lyon’s Brook, we see a clear rise in average temperatures.
There’s lots to be said here for method – and this coming year I’ll be implementing some changes to attempt to modify the estimating process – but as an exercise for students it was a huge success. That they can see how such historical data is made, that they themselves can participate in creating such data, that the process is transparent (not without flaws, but generally sound and moreover honest), that they can can look at current climate data and see a model of how that world works, that “that world” is not an esoteric place of things beyond the comprehension of some Humanities majors – all of these outcomes are truly worthwhile.
