Urban heat in Johannesburg and Ekurhuleni, South Africa: A meter-scale assessment and vulnerability analysis

Abstract

Heat stress is an important threat for human health and urban areas are affected at higher rates compared to rural environments. Additionally, climate change will increase the vulnerability towards urban heat stress in the future. Current high-resolution urban heat stress assessments are limited in time and space due to the high computational costs. In this paper, the UrbClim numerical model is used to simulate urban heat accurately at a fast rate and high spatial resolution for the cities of Johannesburg and Ekurhuleni, South Africa. Using detailed terrain information, (future) urban heat stress assessments are provided at 30 m resolution for both city agglomerations, while meter-scale simulations are executed for a selection of neighborhoods. These model simulations are evaluated using an extensive monitoring campaign in which the local community was heavily engaged. Distinct spatial differences in the urban heat island effect are observed, with greatest heat stress in areas with high building densities and low vegetation numbers. These areas are often characterized by lower socio-economic living conditions. The meter-scale analysis further shows the importance of shade provided by vegetation to lower heat stress in both present and future climate. These assessments offer assistance in the design of climate-resilient urban planning strategies.

Keywords: Ekurhuleni; Heat monitoring; Heat stress; Johannesburg; UrbClim; Urban heat island.

Fig. 1

Methodological workflow. Red boxes show input datasets or temporary products. A table with details of the different input datasets is provided (Table 1). Blue boxes indicate different methodological steps, while green boxes denote the final products.

Fig. 2

Overview of the modeling domain of UrbClim (yellow) and the six neighborhoods in which meter-scale heat stress modeling is executed. J1: Braamfontein; J2: Lombardy; J3: Soweto/Kliptown; E1: Tembisa; E2: Primrose; E3: Actonville. Locations of meteorological validation sites are denoted by green stars.

Fig. 3

Data logger and typical setup during a campaign with one device in the sun and one in the shade.

Fig. 4

UrbClim evaluation.

Fig. 5

Comparison between modeled and measured Wet Bulb Globe Temperatures for the six quarters defined in Fig. 2.

Fig. 6

Average 2 m temperature for present (2001–2020) and future scenarios for the metropolitan areas of Johannesburg and Ekurhuleni. Inset showing the high spatial detail that is achieved.

Fig. 7

Urban heat island corrected for topography. The green star indicates the rural location used as a baseline in the calculation. 1  = Central Business District; 2  = Alexandra township; 3  = Tembisa township; 4  = Soweto township.

Fig. 8

Spatial and temporal evolution of WBGT for Lombardy/Alexandra during the heatwave of 6 January 2016. The color scheme in the background shows metabolic activities for people unacclimatized to heat (Table A.3).

Fig. 9

Future WBGT projection of the 6 January 2016 heatwave for the locations specified in Fig. 8. The color scheme in the background shows metabolic activities for people unacclimatized to heat (Table A.3).

Fig. A.1

Schematic overview of the field measurements, starting with the initialization and activation of the WBGT data loggers, the hand-over to the community participants, the dual (shade-sun) measurement approach, and the return and read-out of the loggers.

Fig. A.2

Comparison of the modeled and measured individual components of the Wet Bulb Globe Temperature for the six quarters defined in Fig. 2.

Fig. A.3

30 m land cover map for the metropolitan areas of Johannesburg and Ekurhuleni.

Fig. A.4

Meter-scale land cover map for the meter-scale heat modeling domains. Black stars indicate the locations of the monitoring campaign measurements.

Fig. A.5

Wet Bulb Globe Temperatures for all domains specified in Fig. 2 for the heatwave of 6 January 2016 at 15.00 h local time.