Woman at work and other untold stories from 35,000 wearable camera images taken on an informal electronic-waste recovery site

Abstract

GREEN, AIR, LMIC: PECHAKUCHA COMES TO ISEE! Johan Friso Foyer, Floor 1, August 27, 2019, 10:30 AM - 12:00 PM As part of my doctoral research, I have viewed 35,588 time-lapse images taken by a wearable camera worn by electronic-waste (e-waste) recovery workers at an informal worksite in Ghana. Each consecutive image captures an instant of life on the job. The images were collected as an alternative to standard pen and paper time-activity diaries. Despite the effort to process the images, they proved to be a highly precise and transparent source of time-activity data. When combined with contemporaneous and continuous measures of personal exposure to size-specific particulate matter (PM), they allowed us to describe associations between specific job tasks and personal PM exposure. We observed several predicted associations, but it is the unpredicted patterns captured by images that I want to share in this presentation. For example, our study is comprised of 100 male e-waste workers, however, it was the colorfully dressed women on the worksite that kept catching my eye. Not only were they on-site selling food and water, but they were most commonly standing near the male workers while they burned e-waste in open, surface fires. Among all the jobs captured in our data, burning e-waste is associated with the highest overall levels of PM2.5 exposure (Geometric Mean: 90 µg/m3). Why was this vulnerable population lingering near the men putting e-waste in and out of the fire and why were they not included in our study? One image captured a woman pouring water on copper wires that had recently been pulled from the fire. In low-and-middle-income countries, there are more women in informal employment than men. This presentation is, most importantly, to honor these female workers who are not invisible. Note: The images to be used in this Pechakucha presentation were collected with the permission of all study participants. No images with identifiable visual information will be used.