Challenging Urban Heat Island Paradigm: Integrating Idiosyncratic Movement into Neighborhood Scale Effects

Urban heat island research suggests that, while cities are hotter than their surrounding environments, the distribution of heat within cities is uneven and reflects differences in the state of the built and natural environment, as well as urban policies. Recent research has found that urban heat is often concentrated in low-income neighborhoods and that during extreme heat events social or linguistic isolation can be a major determinant of heat-related mortality. Thus place-based risk and neighborhood social context can determine the mortality outcome on hot days.

However, urban residents are generally not confined to the neighborhoods where they live. Residents visit and work and shop in other areas within or outside the city. When it is hot outside, some individuals may take advantage of air conditioning, visit a park, or go to a pool. Access to these cooling sites and residents' mobility across the city likely varies by population group and individual, perhaps even within a single city neighborhood. Based on this premise, this research tests the possibility that individual idiosyncratic movements and lifestyle may be an important factor in temperature experiences and extreme-heat vulnerability. Do people living in the same neighborhood actually experience the same temperatures? If not, what factors might explain intra-neighborhood variation (location within a neighborhood, demographic traits, lifestyle choices, for example)?

I recruited 23 residents living in the same Boston neighborhood, the South End, a diverse central city neighborhood in terms of income, age, race, and occupation. Over the past four decades, this neighborhood has undergone waves of gentrification yet maintained an extensive public housing system. Efforts were made to engage participants representing the range of the income and age spectrum found in the South End. Each research participant was equipped with a Thermochron iButton, a mobile temperature sensor that measured air temperature every 5 minutes. Participants were asked to carry their iButton for 1 week (July 17 to July 24; the first four days included a major heat wave) and were clipped to a belt loop or bag. Thus, the iButton recorded the temperatures experienced by the participant at all times, both indoors and outdoors. Participants filled out daily surveys concerning their subjective heat experiences, participated in exit interviews, and drew “heat maps” that illustrated their perception of temperature distribution in their neighborhood as well as their own idiosyncratic movements. Throughout the summer months, I made ethnographic field observations of the South End to understand the social setting and how people use space to stay cool.

During the study week, average nighttime temperatures (8PM – 8AM) experienced by participants ranged from 21.0˚C to 30.6˚C. Average daytime temperatures experienced (8AM – 8PM) ranged from 22.2˚C to 33.9˚C. Temperature variability experienced during the day (as expressed by interquartile range) ranged from 0˚C to 9.1˚C. Thus, I found that a great disparity in temperature experiences exists within the study group both in terms of average nighttime and daytime temperature and ranges of temperatures experienced during the day.

Factors other than neighborhood residence, therefore, may determine the temperatures that an individual experiences. For example, among participants awareness of local pools, feelings of comfort at some of the neighborhood parks, and feelings of safety along some of the streets, varied dramatically, and thus affected how those individuals used public space and neighborhood resources. While neighborhood residence can set the stage for how individuals experience temperatures, it is each individual's experience of the neighborhood that may play a role in extreme-heat vulnerability. The use of iButtons to measure an individual's temperature experience can further the path to discovering exactly how the urban heat island is manifested on the individual and small group scale. This lens will help us better propose and implement site and group specific heat mitigation strategies and warning systems.