Chronic heat stress in tropical urban informal settlements

Abstract

The health and economic impacts of extreme heat on humans are especially pronounced in populations without the means to adapt. We deployed a sensor network across 12 informal settlements in Makassar, Indonesia to measure the thermal environment that people experience inside and outside their homes. We calculated two metrics to assess the magnitude and frequency of heat stress conditions, wet bulb temperature and wet bulb globe temperature, and compared our in situ data to that collected by weather stations. We found that informal settlement residents experience chronic heat stress conditions, which are underestimated by weather stations. Wet bulb temperatures approached the uppermost limits of human survivability, and wet bulb globe temperatures regularly exceeded recommended physical activity thresholds, both in houses and outdoors. Under a warming climate, a growing number of people living informally will face potentially severe impacts from heat stress that have likely been previously overlooked or underestimated.

Keywords: Environmental health; Environmental issues; Weathering.

Graphical abstract

Figure 1

Map of Makassar, Indonesia, showing locations and elevation of informal settlements where loggers were deployed, and weather stations Elevation data is from the Shuttle Radar Topography Mission (SRTM; Farr et al., 2007; USGS EROS Archive: https:://doi.org/10.5066/F7PR7TFT), means were calculated for each settlement and point values extracted for each weather station. Shapefiles of Indonesia sourced from the GADM database version 3.4 (www.gadm.org). Shapefile of roads sourced from OpenStreetMap (https://www.openstreetmap.org).

Figure 2

Flowchart summarizing logger deployment, data collection and data processing in informal settlements (A and B) 65 loggers were deployed across each of the 12 settlements studied (B) recording either temperature (T) or temperature and relative humidity (RH). (C) Wet bulb temperature (TW) was calculated using adjusted temperature (adj. T) values and relative humidity derived from settlement-level vapor pressure (VP). Wind speed and solar radiation data downscaled from the National Centers for Environmental Prediction (NCEP) were used in addition to logger-derived variables to calculate wet bulb globe temperature (WBGT) for each outdoor logger.

Figure 3

WBGT in houses and outdoors in informal settlements in Makassar, Indonesia (A) Representative photos of informal settlements in Makassar, Indonesia. (B) Frequency of continuous time periods above WBGT activity thresholds. (C) Frequency of records of WBGT in houses (left) and outdoors (right). WBGT activity thresholds are from Parsons (2006) including heavy work (metabolic load >260 Wm⁻²), moderate work (200 < metabolic load <260 Wm⁻²), moderate/light work (130 Wm⁻² < metabolic load <200 Wm⁻²), light work (65 Wm⁻² < metabolic load <130 Wm⁻²) and resting (metabolic load <65 Wm⁻²). Photograph credit L - R: E. E. Ramsay, E. E. Ramsay, RISE Consortium, B. C. Josey.

Figure 4

Frequency of WBGT records under different heat stress scenarios (A) Frequency of records of WBGT calculated from relative humidity (%) and temperature (°C) in houses. (B) Frequency of records of WBGT calculated from relative humidity (%), temperature (°C) and different wind and solar radiation scenarios for day time data collected by outdoor loggers. Contour lines show WBGT activity thresholds from Parsons (2006) including heavy work (metabolic load >260 Wm⁻²), moderate work (200 < metabolic load <260 Wm⁻²), moderate/light work (130 Wm⁻² < metabolic load <200 Wm⁻²), light work (65 Wm⁻² < metabolic load <130 Wm⁻²) and resting (metabolic load <65 Wm⁻²).

Figure 5

TWs in informal settlements and at weather stations (A) Monthly maximum TW (°C) from a full-time series of available weather station data in Makassar, Indonesia showing the time period of this study (note that not all weather stations cover the full time period). (B) Daily maximum TW, from combinations of relative humidity (%) and temperature (°C), in informal settlements and from weather station data, showing the limit of human survival (TW 35°C; Sherwood and Huber, 2010).