Using a blue landscape to mitigate heat stress during a heatwave event: a simulation study in a hot-humid urban environment

Urban areas have grown rapidly since the last century and the urbanization phenomenon is going to accelerate further in the future (Angel et al. 2005). The urban areas develop microclimates that can be significantly affected by the urban heat island (UHI) effect which refers to the phenomenon of possessing a higher temperature in the urban area as compared to the non-urban surroundings (Sarrat et al. 2006; Kovats & Akhtar 2008; Moonen et al. 2012; Klemm et al. 2015). In addition, the growing intensity of climate change is adversely affecting the population and environment of urban areas across the world, including Pakistan, in the form of elevated outdoor temperatures (increased heat index), increased frequency of heatwaves, and urban flooding, among others (Oke 1982; Fischer et al. 2004; Stott et al. 2004; Haines et al. 2006; Mirzaei & Haghighat 2010). These high levels of rising outdoor temperature have a direct impact on the energy systems inside urban areas as these intensify the electricity demand (e.g. air conditioning and chilling), specifically at the time of high solar irradiance, up to 0.45–12.3% (Santamouris et al. 2015). Hence, there is a great interest and increased effort now to adapting the cities to the changing climate, UHI, and the resulting rising outdoor temperatures. This is usually done by changing the aerodynamic and bio-physical attributes of urban areas using some mitigation measures which alter the energy flux between the earth's surface and upper surfaces, resulting in the reduced intensity of the UHI (O'Malley et al. 2015; Shiflett et al. 2017). The common mitigation measures include water bodies (Syafii et al. 2017; Toparlar et al. 2018; Yang et al. 2020; Zeeshan et al. 2022b); vegetation (Toparlar et al. 2018; Zeeshan & Ali 2022a; Zeeshan et al. 2022b); urban geometry Jamei et al. (2016); surface materials Zeeshan & Ali (2022b); and reduction in heat emissions from the anthropogenic sources (Rizwan et al. 2008).

Water bodies, in the context of sustainable urban development, have shown significant potential for alleviating the UHI due to their high thermal and evaporative cooling capacity. Various types of techniques including field measurements and numerical simulations have been used to investigate the cooling effect of water bodies in real urban areas including rivers and ponds (Soultana & Gianniou 2007; Taleghani et al. 2014). Hathway & Sharples (2012) documented an average temperature decrease of 1 °C in Sheffield, UK due to the existence of the river during hot weather conditions. Chen et al. (2006) analyzed the mitigation effectiveness of a small lake in China using remote sensing techniques and observed an air temperature reduction of 1.3 °C. Theeuwes et al. (2013) studied the effectiveness of water bodies by implementing it into multiple small patches rather than one large landscape and witnessed its relatively high cooling effectiveness toward thermal regulation of the urban environment. Syafii et al. (2017) undertook an experimental study and found the same observations. Despite various experimental studies on water bodies in urban microclimates, this approach can be subjected to challenges owing to its inherited cons of providing data at only certain discrete locations and its poor hold on boundary conditions coupled with huge costs/endeavors. In contrast, the studies involving computational approaches such as computational fluid dynamics (CFD) are escalating owing to their ability to simulate the coupled effect of heat transfer, wind flow, and moisture transfer (Lun et al. 2009; Toparlar et al. 2015) with proper boundary conditions control and data gathering at any discrete point (Blocken 2014). However, the reliability of CFD results may be of concern which can be tackled with suitable validation studies such as the replication of the previous experimental study with CFD using the sub-configuration method (Tominaga et al. 2008; Franke et al. 2010; Blocken 2014). Yang et al. (2020) carried out a numerical study and found that the water bodies help modulate the thermal environment of urban areas causing a reduction of up to 2 °C, and promoting the wind flow. Sun & Chen (2012) conducted a numerical study on water bodies by focusing on their landscape parameters such as area, location, geometry, and percentage of built density; and documented their significance for regulating the thermal environment of urban microclimates. Tominaga et al. (2015) performed another numerical study and demonstrated a decrease of 2 °C with the use of water at pedestrian height in a real urban microclimate. In addition to temperature decrease, water bodies also cause the flow velocity to increase (Yang et al. 2020). On the flip side, it results in increased localized humidity, which makes achieving thermal comfort conditions very difficult in hot-humid urban areas as its mitigation potential/effectiveness weakens due to impaired transpiration.

The studies carried out so far have evaluated the performance of water bodies in regulating the thermal environment only at a horizontal scale, i.e. at pedestrian's height, and mostly in summer (Yang et al. 2020; Bartesaghi-Koc et al. 2021). However, their evaluations in a hot, humid climate such as the currently studied location have not been widely investigated which is necessary for carrying out a complete and meaningful analysis of water effectiveness toward promoting cooling in its surrounding environment (Haddad et al. 2020). Such climatic conditions impaired the transpiration process of water, thereby impacting its mitigation potential. Moreover, an assessment of the cooling potential of water bodies for regulating the outdoor thermal environment in isolated streets has not been completely executed. This is attributed to the availability of huge temporal variability and spatial heterogeneity in microclimatic/boundary conditions (Lai et al. 2019; Haddad et al. 2020), and the effect of urban geometry and the advection phenomenon at locations of isolated spaces (Berardi & Wang 2016; Jamei et al. 2016).

Therefore, this study aims to demonstrate that thermal comfort conditions of the urban environment, having hot, humid climatic conditions in isolated (open spaces) streets, can be improved with water bodies. In this regard, the unmitigated scenario without implementing water bodies is first simulated to model realistic climatic conditions of I. I. Chundrigarh road, a central area of Karachi, a city in Sindh province, Pakistan. Moreover, this case will serve as a reference for the subsequent assessment to identify and prioritize the hotspots for the incorporation of water bodies. To fulfill this study's aim, the finite volume method employing URANS equations was adopted to perform CFD simulations. The model used for the water body is first validated using the sub-configuration method based on its evaporation phenomenon. This study furnishes important information, in terms of scientific recommendation, and facilitates urban architects and policymakers regarding the implementation of the blue landscape for ameliorating the local environment of the studied microclimate and for enhancing the naturally enhanced ecosystem at the street, neighborhood, and city scales.

To evaluate the thermal comfort conditions in the urban climate of I. I. Chundrigarh Road, Karachi Pakistan, the climatic parameters (Supplementary material, Dataset) used in the current study are obtained from the Kiamari station observatory managed by Pakistan Meteorology Department (PMD). The Kiamari station is positioned 10 m above ground level. This observatory is located 4 km away from the studied area. The hourly averaged meteorological data for air temperature (°C), relative humidity (%), wind speed (m/s), and wind direction were collected for the heatwave period (18–22 June 2015). The representative data for carrying out the study in this area are given in Table 1.

The CFD simulations have been performed using a finite volume-based method (FVM) along with a realizable k–ε turbulence model in the commercially available tool ANSYS FLUENT 16.2 (Shih et al. 1995; ANSYS 2016). The governing flow equations for velocity and turbulence are solved through the URANS equations model. Moreover, the CFD simulations for modeling urban flow were performed using best practice guidelines of Tominaga et al. (2008); Franke et al. (2010). The SolidWorks computer-aided design (CAD) tool was adopted to model the buildings with their configurations for the studied microclimate. The standard wall function, as proposed by Launder & Spalding (1974) with modified sand grain-based roughness, was imposed on wall type boundaries to resolve the fluid–wall interaction. Z_o, the aerodynamic roughness length of 0.03, is set for all buildings including ground inside the inner circular sub-domain while it ranges from 0.03 to 1 for ground surface outside of the circular sub-domain (Zeeshan et al. 2022a). Moreover, the equations of turbulence, energy, and mean flow are resolved with the use of a second-order discretization scheme to avoid numerical diffusion. The PISO algorithm, for its stability and application for transient flow, was used for pressure velocity coupling (Ansys 2016; Toparlar et al. 2018). The discrete ordinate (DO) radiation model was used to model the (transmissivity) transparency of water bodies (Yang et al. 2020). The solar irradiation and radiative transfer are handled with a solar-ray tracing model and converges set for study results at 10⁻⁵ for all variables except continuity 10⁻⁴ at the end of each time step.

where Z₀, Κ, U*, U_ref, and C_μ refer to the aerodynamic roughness length for building elements and ground, the von-Karman constant has a value of 0.42, the friction-velocity, velocity of the reference height (height of PMD Observatories).

The symmetry boundary conditions are applied at the lateral and top sides of the computational domain having zero gradients of all parameters. The pressure outlet condition is imposed at the outflow. The buildings inside the circular sub-domain and the ground of the computational domain correspond to the wall. The building is modeled implicitly with an equivalent thickness of 0.35 m with enabled conduction equations and is treated as air-conditioned at a temperature of 23 °C. The thermal and radiation conditions, imposed on walls, have a heat transfer coefficient of 0.5 W.m⁻².K⁻¹, free stream temperature, and radiation temperature of 23 °C, in the water zone with heat flux obtained from Equations (4)–(9). The water and ground are modeled with 1 and 10 m thicknesses. At this ground depth, zero heat flux is imposed. The detailed properties of the material for all surface elements are given in Table 2.

The mean radiation of the sun for the studied climate for a given heatwave period, as calculated from equations of radiation, was 164.6 W/m² and is denoted with I. ⁠, ⁠, and represent the water reflectivity of water for solar shortwave radiation, longwave radiation, and the atmosphere; their average values are 0.1, 0.03, and 0.97. are the air saturated vapor pressure and evaporation pressure of air near water, expressed in mmHg. The resulting heat flux for the water zone, computed using Equations (4)–(9), is incorporated into the governing energy equation.

This section is comprised of two sub-sections which are: Validation studies given in Section 3.2 and thermal effects of water bodies, given in Section 3.3. Before simulating the water bodies’ effectiveness, a base case simulating the existing conditions was modeled which served as a reference case for predicting the water effectiveness and for identification of locations to propose representative water bodies. The cooling effect has been given in terms of hourly average data for 19 June 2015; and in terms of cooling intensity evaluated at the pedestrian height at selected horizontal and vertical planes for 1500 LST.

The validation of surface temperature obtained from current study results was made with MODIS (Moderate-resolution Imaging Spectroradiometer) satellite data and has also been described in detail in the study by Zeeshan et al. (2022a, 2022b). MODIS, with its bandwidth of 3.66–3.84 nm and resolution of 1 km, was used as it has been extensively applied to validate CFD results in urban microclimates (Montazeri et al. 2015; Toparlar et al. 2018; Antoniou et al. 2019; Zeeshan et al. 2022b).

The validation study for modeling the thermal effect of water on the surrounding environment was also performed by employing heat flux at the water surface. The wind tunnel experiment data of Kato & Nakane (2009) was taken as a reference. They examined the basic mechanism of evaporation from the water surface in the wind tunnel by maintaining steady-state conditions. For this validation, steady-state simulations were performed and compared with the measured data. The computational domain was modeled as per the actual wind tunnel domain which had a dimension of 3 m × 1 m × 0.98 m along streamwise, lateral, and vertical sides, respectively. The water depth was taken as 20 cm. The domain discretization contained a structured mesh with 50, 20, and 60 cells along x, y, and z dimensions, and meshing was performed in the POINTWISE meshing tool. The wall adjacent cell height was kept at 0.005 m. The computational domain had one inlet, one outlet, and symmetry at the top and lateral sides. The bottom of the ground was assigned as a wall with water at a temperature of 16 °C. At the inlet, the profile of turbulent properties and velocity was assigned according to Equations (1)–(3) with U_ref as 3 m/s and h_ref 0.3 m. The air inlet temperature and humidity of 20 °C and 8 g/kg, respectively, were used. The results were compared in normalized form with these values as reference.

It can be seen from Figure 6(a) that air temperature decreased with the incorporation of water bodies inside the urban microclimate at all time intervals of the day. The mitigation potential was more abundant/prominent at the time of high solar irradiance, owing to the availability of larger energy for evaporation generating more latent heat (Syafii et al. 2017). The maximum temperature reduction that occurred with water bodies was around 0.9 °C at 2 pm. This reduction was mainly attributed to its water thermal capacity, causing an increase in flow velocity, facilitating heat convection between air and its surroundings, thereby resulting in thermal evaporative dissipation and atmospheric turbulence. These phenomena then caused temperature gradients to develop between the water top zone and its surrounding air, discouraging positively sensible heat flux for generating more latent heat to cause cooling. The reduction of 0.9 °C in air temperature, with the use of water bodies, is in line with previous study results (Syafii et al. 2017; Jacobs et al. 2020; Yang et al. 2020). This lessening of temperature has a direct impact on the urban energy systems such as air conditioning applications since lesser temperature put forth lower electricity demand, particularly at times of high solar irradiance (Santamouris et al. 2015). The promoted cooling effect also contributes to energy savings inside urban microclimates (Akbari et al. 2016; Morakinyo et al. 2018). Similiar to air temperature, water bodies also contributed to lowering the surface temperature of urban surfaces by up to 3.5 °C (Figure 6(b)) as it absorbed more solar radiation due to their thermal capacity, causing a decrease in the surface temperature of surrounding elements/surfaces.

The present study aimed to perform the CFD simulations for evaluating the cooling potential of water bodies for the heatwave period 18–22 June 2015 in the hot-humid climate of I. I. Chundrigarh road, Karachi, Pakistan. The CFD simulations were performed using an FVM employing URANS equations. The water bodies model adopted in this study was evaluated and validated through the sub-configuration method. Surrounding temperatures and wind velocity were appraised to evaluate the thermal influence of water. Following are the major conclusions implied from this study's results:

• (1)

  With respect to the water bodies simulation case, the thermal evaporative model with no inclusion of anthropogenic heat sources, as adapted in this study, had proven to have good accuracy, as validated through the sub-configuration method.

• (2)

  Under heatwave and summer conditions, incorporating water bodies inside urban areas could effectively contribute to its thermal environment regulation, i.e. improving the ventilation and decreasing the thermal stress of pedestrians.

  • (a)

    For airflow velocity, it was significantly increased by 0.2 m/s in water areas compared to its surroundings which is attributed to temperature difference between water bodies and around hard ground situations.

  • (b)

    For air temperatures and surface temperatures, these were reduced by 0.9 and 3.5 °C as increased airflow velocity expedites the convected heat to/from air to surroundings causing evaporative thermal dissipation. This facilitates temperature gradients at the water surface to air to develop coupled with atmospheric turbulence, affecting negative sensible heat flux, resulting in lower ambient temperatures. It is thus inferred that water heat capacity could play a part in decreasing surrounding temperatures.

• (3)

  The effect of temperature reduction and improved wind velocity could also influence its surroundings with greater effect in its immediate vicinity.

Nevertheless, water effectiveness in regulating the microclimate and assisting in air cooling varies with the availability of thermal energy, the intensity of climatic parameters (air temperature, relative humidity, velocity), water body properties (quality, depth, area), and surrounding area around water as they significantly influence the microclimate around it so additional simulations are required to gain knowledge of how these parameters affect cooling which would be helpful for designers/planners in designing natural ecosystem services. Moreover, the missing aspects, i.e. the relevant inlet profiles for different atmospheric conditions and anthropogenic heat sources, are proposed to be studied for future research work since their application would ensure and improve the result integrity.

The authors are highly thankful to PMD for providing environmental data for the heatwave period of 2015.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.