The American Psychiatric Association (APA) Annual Meeting was one of many events to be made virtual during the COVID-19 global shutdown. Reduced air travel significantly improved global air and water quality during this time [See Venter et al. (2020) and Yunus, Masago, and Hijioka (2020)]. We sought to estimate the carbon footprint of the past two APA meetings as a proxy for the carbon emissions avoided by holding the 2020 Annual Meeting virtually. Using anonymized attendance data we assess how the location of APA meetings impacts this carbon footprint and develop conservative estimates of carbon dioxide equivalent (CO2e) emissions from each attendee’s travel in 2018 and 2019. We create counterfactual scenarios to quantify carbon footprints had the meetings been held at locations of the past 40 APA meetings instead. We identify likely transport modes and routes each attendee would use to travel to and from the meetings then apply an emissions function to the ground and air legs. The carbon footprint of the New York City 2018 and San Francisco 2019 meetings are 19,819 (1.19 per Capita) metric tons CO2e and 21,456 (1.61 per Capita) metric tons CO2e, respectively. This means the carbon footprint of each conference was equivalent to burning about 500 acres of dense forest or 22 million pounds of coal. Counterfactual scenarios at previous locations in the Western US and Hawaii increased the total carbon footprint by 60% - 164% compared to meetings held in the Northeastern US. These findings were further confirmed by an unconstrained geometric minimization algorithm that sampled 100,000 random coordinates across the globe. This study shows that the CO2e emissions of the APA Annual Meetings are significant and that optimizing conference location or intermittently making the conference virtual can dramatically lower this carbon footprint.
Below is a PDF explorer that visualizes the flight network models used to illustrate each attendee’s commute to the APA Annual Meeting for the 2018 New York City and 2019 San Francisco Annual Meetings. The third page shows a counterfactual scenario had the attendees who registered for the 2020 Philadelphia meeting actually attended; However, the 2020 meeting was canceled with a virtual substitute in the wake of the COVID-19 pandemic. The arcs measure the distance as the crow flies, or the geodesic distance between each attendee’s origin and the Annual Meeting location. The width and intensity of the arcs is proportional to the number of people originating from a given location.
It may take some time for the PDF explorer to load, as these are large vectorized images.
If you are interested in viewing further visualizations or the analysis itself, there is a comprehensive repository containing all the R code used to generate the plots and model the attendee emissions available at my GitHub.
I would like to thank my coauthors, especially Dr. Josh Wortzel and Dr. Elizabeth Haase for the opportunity to participate in this important research.
I would also like to thank GoClimate for granting us access to their Flight Emisions API.
Please feel free to reach out with any questions