Whenever William Smith runs a calcium imaging experiment, he spends 40 minutes staring at a screen. “It gives you a lot of reflective and contemplative time to think about what’s going on,” says Smith, a graduate student in Stefan Pulver’s lab at the University of St Andrews.
During these reflections, Smith says he started to wonder how much energy he was using, and in turn how much carbon that released into the atmosphere. So he decided to find out.
Over the course of one year, Smith’s research on Drosophila melanogaster generated 16.11 kilograms of carbon dioxide equivalents (kg CO2e), he estimates. Research-related travel during that time produced another 442 kg CO2e, and emissions linked to equipment procurement ranged from 81.63 to 542.92 kg CO2e.
Calculating the carbon footprint of a research program is a step graduate students and labs can take to become more sustainable, Smith and Pulver argue in a preprint about the calculations, posted on bioRxiv on 22 January. Ph.D. dissertations could include a “carbon appendix” that outlines the calculated emissions, the pair proposes.
Though institutions should also take on emission-reducing actions, “if you want a degree of progress in society, you can’t just point to another individual. You have to take on some responsibility yourself,” Smith says. “One of the barriers to sustainability implementation, particularly within academia, is that most people have this hesitancy to get involved. Which means there’s very low-hanging fruit to be had.”
Smith and Pulver spoke with The Transmitter about how they carried out their calculations and why they think other researchers should follow suit.
This interview has been edited for length and clarity.
The Transmitter: How did you go from musing about the climate impact of your research to landing on the final form of this project?
William Smith: Like the origin of most science projects, it’s a dynamic, iterative process. Part of my motivation was working out, “Is what I’m doing sustainable? And how would I quantify that?” Tracking a single year was the easiest way to imagine coming up with an estimate. Putting a determinant number on it seemed to be a way in which you could scale it up but also keep it defined.
TT: How did you determine which activities to include in your calculation?
WS: I took things that were most proximate to me and then thought about, “How much, as an individual, can I actually estimate?” I started simple, such as the energy used during calcium imaging and data analysis. Then I expanded it to things like procurement and the persistent energy use in the lab.
I modeled the process off life-cycle assessments. It’s an exercise from other fields, where you track a product from the beginning of its life to the end. In the case of a plastic bottle, you get the oil under the ground, and then you refine it into plastic, and then you sell it to a consumer, and then the consumer disposes it in a landfill. For my research, I looked at the provision of the equipment I need, the preparation I do before experiments, the experiments themselves, the analysis, and then what I do with the output.
TT: How did you perform the emissions calculations?
WS: There are two parts to it. The first is the direct energy that we use, which was significantly easier to work out as a translation to emissions. There is software called the Carbon Intensity API, which provides information on carbon emissions from the United Kingdom’s energy grid.
The difficulties arose when we looked at how much energy we use during supply procurement. We tried to find this information by contacting providers and learning about the origin site for each part of the manufacturing cycle, and then made best-guess estimates about how products get from that location to us. So our calculation is a lower- and upper-bound estimate.
TT: Did anything about this exercise surprise you?
WS: In the lab, we release CO2 directly to anesthetize flies. I thought the emissions wouldn’t be that large. But in the grand scheme of things, it’s actually quite big, especially if you scale it up to the number of Drosophila labs that exist around the world—that number is probably humongous.
Stefan Pulver: It’s been wonderful having William and my other Ph.D. students thinking about this problem. They’ve generated a lot of interesting data, but they’ve also kept their own research on track. One of the things that I’ve been impressed by is how if you have a group of folks thinking about this, the core research of a lab can continue. I don’t think it’s been a burden.
Being able to see the different emissions ratios of things we do every day in the lab was really interesting. These numbers are estimates, and we hope there will there be iterative improvement of them. But they also have resulted in real actions. We’ve made decisions about what not to purchase and when to work during the day.
TT: What sorts of changes have you made?
WS: Most electricity grids have fluctuating demands, and during high-demand times the grid will start relying on nonrenewable energy sources. So now we don’t do high-energy research, like an optogenetic experiment, at peak times, and instead do something with a lower carbon cost, like analysis.
TT: In the preprint, you suggest that Ph.D. students take on the role of “carbon accountants” by including the carbon emissions produced by their project in their dissertation. Why that group in particular?
WS: We don’t think that Ph.D. students should be the only people doing this, but they are an untapped stakeholder in this battle. Ph.D. students have massive cross-disciplinary scale. It doesn’t matter what discipline the Ph.D. student is in—they can utilize the carbon-calculating resources we describe in the preprint to make reliable estimates across the board. These individuals are the next generation: It’s not just about making carbon reports, but educating a base of researchers and potential thought leaders about sustainability.
SP: Research theses have a common structure that spreads across national borders. They’re a way for young scientists to communicate with one another in a language that is at least somewhat similar across boundaries.
We’re not trying to make the case that Ph.D. dissertations are the only way that information could be disseminated. But it’s one way that early-career scientists can contribute, and that could complement other approaches happening at the institution or lab level.
TT: Wouldn’t carbon accounting efforts spearheaded by institutions be more effective than putting the onus on individual students?
SP: I don’t think individual and institutional efforts are mutually exclusive. There can be local efforts made by students and individuals to try to come up with emissions estimates; and institution-wide efforts to accredit labs, for example, are useful because they try to bring everyone to common standards. The two ways of generating data and taking action can coexist together and actually be synergistic and helpful to each other. They can make each approach better.
Part of the challenge is trying to find avenues where individuals can take action and feel empowered and that they’re making a difference. Calculating carbon emissions is one way to do that—if we generate real data about research’s carbon footprint, then that’s one way to empower people and solve larger problems.
For example, if we can understand more about how the filters used in florescent microscopes are manufactured, and the carbon costs of shipping them, then those data can be something that everyone who does fluorescence microscopy can use to make sustainability decisions. You can break down this larger sustainability problem into smaller chunks that individuals can grapple with.
