Keywords

1 The Risk of Urban Heat Hazard

The increase in frequency and duration of excessive heat is a major concern for public health in highly populated cities at the tropics. Urban areas are generally several degrees warmer than rural surroundings due to the replacement of natural surfaces with hard pavements and buildings during the process of urbanisation, which the phenomenon is well-known as the urban heat island effect (Gago et al. 2013; Yu et al. 2017). In recent years, global climate change has increased the magnitute, frequency, and duration of extreme high temperature, which further worsens thermal discomfort in cities (Revi et al. 2014). This has led to hazardous heatwave events in cities across a range of climate types (WMO and WHO 2015). Increasing mortality rates and heat related health illness are reported worldwide regardless of the development status of the country (Norton et al. 2015; WMO and WHO 2015). According to Mora et al. (2017), around 30% of the world’s population is exposed to potentially lethal heat for at least 20 days a year. By 2100, this number is projected to increase to 48% even under a scenario of drastic reduction of greenhouse gas emission (Mora et al. 2017). Together with the urban heat island effect, the economic losses due to global warming could be 2.6 times higher for cities (Estrada et al. 2017). At the worst impacted cities, the cost of extreme high temperature could reach up to 10.9% of GDP by 2100 (ibid.).

2 The Cooling Effect of Green Infrastructure in Cities

Cities play a critical role in countering adverse impacts from climate change and building adaptive capacity in local societies (UN-Habitat 2017). The local-level governments which determine land uses and development of cities play a crucial role in heat risk reduction. There is a growing interest in urban climatological planning, of which green infrastructure strategy manifests an opportunity to reconstruct city-nature relationship by rehabilitating a natural process and regulating services in built environments (Kabisch et al. 2017; European Commission 2013; Shih et al. 2020). Green Infrastructure (GI) is a strategic spatial planning, which has been recognised as a critical nature-based solution for cities to attain climate resilience, health, social inclusion, and sustainability (e.g. IPCC AR5 and AR6). It refers to “an interconnected network of natural and semi-natural areas with other environmental features that strategically planned, designed and managed to sustain health nature process and provide multiple functions” (Benedict and McMahon 2012; EC 2013; Hansen and Pauleit 2014). Despite the disparity in grounded knowledge regarding the synergies and trade-offs of multiple functions associated with the spatial pattern of GI (Hansen and Pauleit 2014; Meerow and Newell 2017), scholars and practitioners increasingly argue integrating GI with cities’ grey infrastructure to foster resilience and sustainability in urban societies and environments (Kabisch et al. 2017; Mell 2016; Shih et al. 2020).

Urban climatological planning aiming at optimising regulating services from green infrastructure may take place at different forms and spatial scales. In Stuttgart (Germany), a comprehensive GI strategy from a city-region level to a site level is used to reduce urban heat island effects and to improve air quality (VRS 2008). At the city-region level, the forested hills around Stuttgart is protected by law as a conservation area to generate cool and clean air. At the city level, urban greenspaces are strategically allocated through re-zoning and and the change of building codes (VRS 2008). This land use scheme enables the creation of wind corridors, which facilitate air producing in the surrounding hills to flow across the city. At the site scale, large trees having diameter above one metre at the breast height, are protected for cooling services (City of Stuttgart 2017). In Melbourne (Australia), the city government has initiated an urban forest strategy and aims to increase canopy cover from 22% to 40% by 2040 (City of Melbourne 2011). A similar strategy has been undertaken in Chicago (USA), where the municipal government aimed to add one million new trees to parks, parkways, and private yards by 2020 for mitigating urban heat (City of Chicago n.d.).

Indeed, vegetation removing latent heat from the surroundings via evapotranspiration and reducing incoming solar radiation on land surfaces through shading is an important mechinism to regulate city’s temperature (Shashua-Bar et al. 2009). Empirical studies found that green infrastructure is generally several degrees cooler than impervious urban surfaces (Shih 2017a, b; Bowler et al. 2010). In the tropics, Giridharan and Emmanuel (2018) indicated that the urban-rural temperature difference in warm climates has a closer magnitude to the intra-urban temperature difference, which is mainly attributable to the distribution of vegetation. This emphasises the vital role of green infrastructure in mitigating the urban heat island effect in warm climates (Shih et al. 2020). Together with wind, the cooling effect from GI can be delivered to built-up areas and lower temperature several metres away from a green space (e.g. Shih 2017a, b; Shih et al. 2020; Narita et al. 2004). As this cooling mechanism provides better thermal comfort for urban dwellers, it also holds a great potential to reduce energy demand from buildings (Ca et al. 1998; Oliveira et al. 2011).

Increasing natural coverage in cities is fundamental for mitigating the urban heat island effects, but it is particularly challenging for densely developed cities where lands for greenery is scarce. However, previous studies suggested that urban temperature varies not only by the proportion of greenspaces (Chen et al. 2012; Ren et al. 2013; Tan and Li 2013; Chang et al. 2015), but also by their configuration (Shih 2017b; Xu et al. 2017; Zhou et al. 2011). In other words, increasing green volume is not the only solution—better cooling effect may be achieved by modifying urban morphology. It is understood that urban temeprature are jointly influenced by various factors, including (a) meso-scale climate conditions and topographical characteristics; (b) greenspace composition and configuration (Kong et al. 2014), such as greenspace coherence (Shih 2017a, b, c; Li et al. 2012), surface area (e.g. Chang et al. 2007; Tan and Li 2013; Shashua-Bar and Hoffman 2000), and shape (e.g. Li et al. 2012; Ren et al. 2013); as well as (c) geometry of buildings in three dimensions (e.g. Shih et al. 2020; Xu et al. 2017). Greater greenspace cooling effect may thus be delivered through an integrated consideration of these factors.

This chapter elaborates opportunities and challenges for optimising cooling services from GI in cities by using Taipei Metropolis as a case study area.

3 The Warming Trend in Taipei and the Need for Adaptation

This study takes the urbanised area of Taipei Basin (25°’N, 121°’E), including parts of Taipei City and New Taipei City, as empirical study areas (Fig. 15.1). It covers approximately 2726 km2 and has population estimated at 6.67 million by 2014. The climate in the North Taiwan is humid subtropical type with an annual average temperature about 23 °C (CWB 2014). The summer months, June, July, and August, are the warmest in a year with mean temperature at 28.8 °C (ibid). Due to global warming and rapid urbanisation in the past decades, Taipei has shown a distinct warming trend (Bai et al. 2011; Hsu et al. 2011). In the last 30 years, annual mean temperature has raised at the rate of 0.27°Cper year in Taipei City (CWB 2014) and the decadal mean number of hot days (above 35¯C) has increased from 5–22 days on average to 37 days per year in the 2000s (Liu et al. 2010).

Fig. 15.1
figure 1

The location of Taipei Metropolis and the studied area (Source: Google Maps, adapted from Shih 2017b)

Full size image

The projection from the Taiwan Climate Change Protection and Information Platform (TCCIP 2017) further indicates that the increase of summer temperatures will reach 1.125–1.25 °C between the 2021–2040 period under the IPCC RCP 8.5 scenario (i.e. ‘business as usual’). Even under the RCP 2.6 scenario of radical emissions reduction, average summer temperatures are still projected to increase by between 0.625 and 0.75 °C in Taipei over the same period.

Thermal comfort in the summertime has thus become a crucial issue to be addressed within urban development. The need for heat mitigation and adaptation planning is further emphasised by Taiwan’s ageing demographic characteristics. There were 14.4% and 18.07% of the population older than 65 years old in New Taipei City and Taipei City in 2019 (DBAS-NTCG 2020; DBAS-TCG 2020). The World Population Prospect predicts that by 2050 Taiwan will hold the oldest population in the world, with a median age of 56.2 years old (UN-DESA 2015). Age is an important factor determining the sensitivity of individuals to excessive high temperature, because the physiological mechanism to regulate body temperature is declined with age (Watts et al. 2019). Excess urban heat in summer, which exacerbates air pollution, increases the spread of diseases, and triggers both physiological and mental health problems, is particularly threatening for an anging society (Watts et al. 2019; Huang et al. 2011).

4 Green Infrastructure and the Thermal Distribution of Taipei Metropolis

This section summarises the findings regarding thermal environments and greenspace cooling effect of Taipei basin (Shih 2017a, b, c). The principal data used for the analyses was satellite images acquired from LANDSAT 8, which periodically visited Taiwan at 10:20 am local time with a 14-days interval. Therefore, it is important to note that the findings and discussions in this article are specific to the climate conditions in the daytime. The temperatures mentioned hereafter refer to land surface temperature (LST) derived from the TIR band of LANDSAT 8 unless otherwise specified. The urbanised areas of Taipei metropolis are closely in line with the area of Taipei basin, which is surrounded by mountains and hills. Due to topographical constraints limiting development, most of the area in the mountain region remains forested. By contrast, substantial natural areas in the low-land basin areas have disappeared and been converted into buildings due to urbanisation (Shih 2010).

4.1 The Characteristics of Green Infrastructure

To date, the greenspace in the basin area is small, fragmented, and artificial because of radical urban development in the past decades prioritising construction and economic growth (Shih 2010). Other than parks designated in the earlier stage of the urban plan and farmlands/wetlands remaining in the Guandu areas, large greenspaces are scare in the basin areas (Fig. 15.2). In Taipei City, more than 80% of parks are smaller than 1 hectare and many of them have low coverage of vegetations (Shih 2010). Official planning policy, which has focused on the creation of small but accessible neighbourhood parks since the 1960s (Shih 2010), is one of the reasons for this green pattern. For New Taipei City, considerable amounts of agricultural lands have been developed in the last decade due to the rising housing demand for people commuting to work in Taipei City. Agricultural lands also face a constant encroachment of illegal factories due to the lack of effective control on land use. Although there are greenspaces persisting in some areas, most of them are facing intensive development pressure.

Fig. 15.2
figure 2

Green infrastructure (vegetated grounds) in Greater Taipei area (The map is generated by calculating the NDVI value on 25 April 2015/LANDSAT 8 imagery)

Full size image

4.2 Thermal Distribution in Summer

Figure 15.3c, d shows the heterogeneity of land surface temperature across the basin areas of Taipei on 12 August 2015 and 29 July 2016. The mean air temperature recorded on each day of observation at 10 am from Taipei and Banqiao meteorological stations was 32.8 °C and 32.7 °C, respectively, in 2015, and 35 °C and 35.1 °C in 2016. The difference in land surface temperature was significant, reaching 10 °C in the urbanised basin areas. Comparing the thermal distribution on each date, a similar thermal pattern is revealed in that New Taipei City on the west of the Danshui and Xindian Rivers is constantly warmer than Taipei City on the east.

Fig. 15.3
figure 3

Land cover and thermal distribution in Greater Taipei area on (a) natural colour of LANDSAT image; (b) land cover types; (c) LST on 12 August, 2015, (d) LST on 29 July 2016

Full size image

Clearly, the thermal pattern is closely related to the types of land cover. Buildings and hard pavements have higher temperature than vegetated grounds and water bodies (Fig. 15.3). On average, water has the strongest cooling intensity, followed by trees and grasses. In New Taipei City, the hot spots are associated with the location of manufacturing and densely developed areas, where vegetation is particularly scarce. This association may be influenced by the kinds of building materials used, as most factories in New Taipei City are low-rise steel buildings, which tend to absorb and store large amounts of heat (Shih 2017c). In Taipei City, hot spots are distributed in the areas of Song-san Airport, Wan-hua District, Da-tong District, and Wu-xing Street. This thermal pattern may be attributable to large and continuous hard pavement, higher building density, and the lack of greenery.

On the contrary, the coolest areas within the basin are associated with the location of major rivers, such as Danshui, Keelung, Dahan, and Xindian. However, it is noteworthy that little or insignificant cooling effect is observed alongside small creeks and ditches, particularly those surrounded by intensive development. For example, the waterfront areas alongside the Wugukeng Creek and Taliaokeng Creek are mostly steel-built factories, hence the cooling intensity from the creeks can barely be detected (Fig. 15.3). Other than rivers, the locations of cool islands are associated with open-air water bodies as well as extensive and dominant green areas, such as riverside greenspaces, wetlands, farmlands, derelict lands, large urban parks, and forested hills (Shih 2017b).

Amongst densely developed areas, lower temperatures tend to be found in large parks, gardens, and institutional grounds. Examples of such lower temperature areas include Daan Forest Park, Youth Park, Songshan Cultural and Creative Park, and Taipei Botanical Garden, as well as universities such as National Taiwan University of Arts, National Taiwan University, and Fu-Jen University. In New Taipei City, however, many cool areas are influenced by remaining agricultural lands and temporary greenspaces, which are designated for further development in the urban plan. It is expected that future development of these areas will worsen the thermal conditions in New Taipei City. In addition to large greenspaces, linear street greenery could also deliver cooling services to the city. Although street greenery is common in Taipei Metropolis, it is important to note that some street greenery demonstrates better cooling performance than the others. As showed in Fig. 15.3c, d, the cooling performance of Ren-ai and Dun-hua South boulevards is more constant and distinct. The difference of cooling magnitude of street greenery is likely related to types of street trees, width of greenery, and the width street canyons (Li et al. 2011).

4.3 Wind Path

River corridors and mountain valleys are important to natural ventilation of cities, particularly for those located in basins (Hebbert and Webb 2012). For Taipei Basin, land-sea breeze has a significant impact on local circulation, whilst it is further complicated by terrain roughness. During daytime summer, sea-breeze enters the basin mostly through the major river valleys of Danshui, Keelung, and Dahan; whereas at night time, land-breeze developed on the surrounding mountains flows along with river valleys and enter the city and then the sea (Lin et al. 2008; Kagiya and Ashie 2009). Whilst there has to date lacked direct observation, this circulation mechanism, together with unique topography and green infrastructure, is likely to play a critical role in regulating temperature at the urbanised area of Taipei basin.

However, urban development has been undertaken without understanding and considering this natural mechanism. It is likely that this meso-scale natural ventilation system has been weakened or even destroyed due to construction blocking the way of wind. For example, sea-land breeze should theoretically be able to reach the basin area of New Taipei City not only through the Danshui River, but also through mountain valleys on the west of Taipei Basin. Yet the temperature of these mountain valleys has increased significantly in the past decades to the extent that little cooling pattern can be observed both above and on its extended surroundings. For example, the valleys of Wugukeng Creek and Taliaokeng Creek are intensively built and some high-rise buildings are oriented against the direction of the creeks and prevailing wind. This development pattern will likely heat up cool air flowing between the mountains and the basin and also reduces wind volume and speed.

Another type of construction that might undermine the cooling function from the rivers is the existence of flooding walls, and intensive development at the waterfront areas. Although water surface within the watercourse of large rivers is significantly cooler, the cooling distance tends to be limited from the bank (Shih and Mabon 2018; Shih 2017b). Areas close to the rivers are not necessarily cooler, whereas in areas where open spaces or greenspaces are adjacent to rivers, the cooling patterns show a better extension from the watercourse despite the existence of flooding walls (Shih 2017b). Similar findings have emerged from previous studies, both in Sheffield, UK (Hathway and Sharples 2012) and in Taipei (Chen et al. 2014). This suggests that increasing greenery and openness at water front areas might improve the levels of cooling perceived from the river bank. However, due to a lack of local data regarding how urban form at the water front might influence wind and the cooling extension from the rivers, further research is needed to clarify whether and how the cooling is contributed by wind against the change of land use and land cover.

4.4 Greenspace Cooling Effect on Surrounding Built Environments

The temperature of greenspaces is on average cooler than built environments. The cooling intensity from greenspaces can be particularly significant in summer compared to Spring and Autumn (Shih 2017c). Yet not every kind of vegetation has a parallel cooling effect. In general, areas covered with trees are cooler than areas covered with grass, because trees have higher evapotranspiration rates that absorb heat from their surroundings, and provide shade to reduce the incoming solar radiation.

Greenspaces not only lower temperature from above, but also moderate heat from their surroundings. The cooling extension from greenspaces is more likely to be felt within 100 m when wind is relatively calm (Shih 2017a, b). The temperature of built environments is highly affected by nearby greenspaces and vice versa due to heat exchange between vegetated and not vegetated areas. A reverse heat effect from built environments might occur when the cooling magnitude of greenspaces is weak and the heat from outside the greenspaces is strong. Greenspaces below 2 hectares tend to have higher temperature fluctuation, as heat intrusion from the surroundings can penetrate further (Shih 2015, 2017a).

In this sense, temperature reduction from larger greenspaces can be more stable (Shih 2017a). Large greenspaces possess greater distance from the centre to the surrounding built environments (sources of heat), so cool spots within larger greenspaces might be better preserved if the shape of greenspaces is relatively compact. It has been found that the minimum temperature both within and around larger greenspaces tends to be lower (Shih 2017a). Although a threshold size for greenspace cooling has not yet been identified, more significant temperature reductions are found with a size interval of four hectares (Shih 2017b). In addition, a visual analysis suggested that the coolest spot within a greenspace is mostly associated with the location of water bodies (Shih and Mabon 2018). Increasing ponds, fountains, or creeks together with greenspaces is likely to amplify cooling magnitude.

Although greenspaces and water are cool islands in the city, the cooling effect extending from them to surrounding built environments is even more critical to mitigating the urban heat island effects and enhancing thermal comfort for urban dwellers. However, factors contributing to cooler greenspaces, such as larger vegetated areas, greater shape compactness, and higher degree of greenery within greenspaces, do not necessarily have an explicit effect on temperature reduction in the surrounding built environment (Shih 2017b). To extend cooling effects beyond cool islands, planning strategies should consider both greenspace features and the development characteristics of their surroundings (Shih et al. 2020; Shih and Mabon 2018; Shih 2017a, b). Lower development intensity at the adjacent areas of greenspaces can be an effective approach. This includes preserving large undeveloped lands, enhancing greenery at the immediate surrounding of greenspaces, allocating greenspaces with higher coherence rather than dispersal form, and increasing water elements in the city (Shih 2017a, b).

4.5 Thermal Inequity and Socio-Ecological Heterogeneity

Just as thermal environments and green cover vary across a city, differences may also exist in the extent of heat risk to which the population is exposed, and in the presence and/or accessibility of greenspace in their surroundings (Shih and Mabon 2021). These differences are not ‘natural’—rather they are the result of social processes which mean that some people may have higher adaptive capacity than others, or live in a ‘greener’ environment. Those with lower incomes, lower education levels, and less access to mitigating technologies like air conditioning may face higher exposure to heat if they end up living in denser, lower-quality accommodation with limited surrounding greenery (e.g. Harlan et al. 2006; Byrne et al. 2016). When it comes to greenspace provision too, processes such as the ‘luxury effect’ (Hope et al. 2003) or ‘ecological gentrification’ (Dooling 2009) may result in greenery accruing to wealthier areas as cities develop. This ‘double inequality’—where already marginalised and vulnerable groups have the least access to greenspace (Apparicio et al. 2016)—risks not only reducing the availability of cooling service for groups who may already be more vulnerable, but may also reduce the potential for social capital-building activities (Dinnie et al. 2013) and also physical (WHO 2016) and psychological (Fuller et al. 2007) well-being that accessible greenspace provides.

There are indications that such differences in the distribution of vulnerable populations hold true for Taipei as well. Figure 15.4 shows the distribution of greenspaces against heat intensity of Taipei Basin by neighbourhoods. Even on visual inspection, the heterogeneity of heat exposure in association with the location of greenspaces across space is clear. An initial assessment of this spatial pattern with socioeconomic factors of Taipei neighourhoods found that richer neighourhoods tend to have lower heat intensity and greater greenery (Shih and Ahmed 2018). Because of the urban-rural dynamic in the past decades, lower income neighbourhoods were mostly located in New Taipei City, whilst more aging population was found in Taipei City. This may make it challenging to know which areas to prioritise for planning interventions aimed specifically at reducing heat risk. Moreover, it is also vital to remember that heat vulnerability is a product of social and cultural driver which is closely related to urban development decisions. Further research thus needs to evaluate the extent to which factors such as community participation, inequality and social cohesion (e.g. Chang et al. 2015; Cutter et al. 2003; Klinenberg 2002) affect neighbourhood vulnerability in the Taipei context specifically.

Fig. 15.4
figure 4

Distribution of ageing population within Taipei Metropolis

Full size image

It is also vital to consider not only the distribution of vulnerable populations, but also which sections of society benefit from the outcomes of planning processes. Over the last decade, some greenspace-related planning initiatives in Taipei have been criticised by environmental groups for prioritising short-term developer profit over longer-term environmental sustainability (e.g. Taipei Times 2011a). When equity and justice are raised, these tend to be in the context of their absence within greenspace discussions in Taipei (e.g. Taipei Times 2010; Taipei Times 2011b). Such criticism of the prioritisation of economic gain over community well-being in urban development has likewise emerged in other Taipei-focused research (e.g. Raco et al. 2011; Jou et al. 2016). There is hence an indication in Taipei that the benefits of urban greening may accrue to those who are already more privileged or empowered.

More in-depth research into issues of spatial inequity in heat risk and greenspace provision is required for Taipei to verify the points made above. But there is evidence from both socio-economic data and recent greenspace planning controversies in the city to suggest there is a need in Taipei to develop processes and safeguards—perhaps through planning policy—to ensure that climate change adaptation via land use and green infrastructure benefits the most vulnerable people.

5 Planning Implications and Conclusions

Green infrastructure plays an important role in delivering cooling services in cities. For extending these cooling benefits to urban dwellers, planning policy however ought to consider strategies beyond greenspaces (Shih et al. 2020; Prieur-Richard et al. 2019). This study intents to draw attention to the complicated interrelationship of topography, wind circulation, and urban development patterns including green-blue spaces and built-up areas from a meso-scale to a site scale. In Taipei Metropolis, the thermal environment is significantly influenced by the unique basin topography, which forms a natural barrier to reduce sea-land breeze but at the same time enriches the area with mountain winds from the surroundings. Rapid urban development without prior awareness of this natural ventilation mechanism may have destroyed or damaged areas where clean and cool air are developed, and channels that facilitate air exchange.

Although it could be challenging to alter existing land use in such densely built areas, some modification on green infrastructure and built-up areas might be feasible in a long term to optimise cooling services and bring better thermal comfort. At a meso-scale, natural ventilation can be gradually reclaimed through the protection of mountain valleys from intensive development; cataloguing and modifying high-rise buildings which block the way of wind at mountain valleys and waterfront areas (Fig. 15.5); and preserving greenspaces along with wind path, particularly those located in the wider river corridors. In addition to air circulation, a baseline should be established to protect major cool islands, such as farmlands, wetlands, and ponds, from further fragmentation due to either uncontrolled urban sprawl or a poor consideration in official development plans. This can be achieved by strengthening the control of land zoning ordinance. Rezoning might also be used in strategic locations to protect existing cool islands that are at present undeveloped but marked for future development. In this case, transfer for development rights might be applied.

Fig. 15.5
figure 5

Model showing buildings located in mountain valleys that may block wind (Model from the Taipei Vision Plan Studio)

Full size image

At a neighbourhood to site scale, optimising urban green infrastructure for enhancing cooling benefits requires comprehensive consideration of both greenspaces and surrounding development conditions (Prieur-Richard et al. 2019; Shih et al. 2020; Shih 2017a). As creating large greenspaces is less feasible for a densely developed city, mitigation strategies might focus on extending cooling services from existing cool islands and reducing heat at hot spots (Shih and Mabon 2018). In this sense, placing small greenery around large greenspaces or water bodies can help with the extension of cooling services into a wider area. Also, even though increasing small greenery at hot spots might not form a stable cool island (Shih and Mabon 2018), it can offset the accumulation of heat. This modification can be attained by greening marginal areas within different land use types, such as school grounds, institutional grounds, and streets; planting derelict lands; and/or implementing vertical greenery on roofs and walls of buildings. To this end, local governments might provide formal guidelines or incentives through building codes and urban design principles.

In addition to increasing vegetation volume and density, reclaiming water bodies can be another strategy to lower temperature in cities. In Taipei, the ‘green campus’ movement over the past decades has increased ecological ponds in schools for the enhancement of urban biodiversity. This is expected to have brought cooling co-benefits to the area and should be further encouraged. Also, recent public debates around and interest in recovering streams and ditches that have been covered for houses, roads, and parking lots may help to initiate action at the community level towards increasing running water within the city.

6 Conclusions

Urban climate is the complicated and dynamic result of the interaction between weather conditions, topography, development patterns, building geometry, and anthropogenic heat across multiple scales. Although green infrastructure, including greenspaces and water, plays an important role in mitigating urban heat island effects, an effective cooling strategy relies on planning grounded in understanding of climatological processes within the city. It is important to not just focus on greenspace itself, but to consider a much wider range of factors in an integrated manner. These include meso-scale topography, wind environments (e.g. sea-land breeze or mountain valley breeze), and development geometry in three dimensions; and also neighbourhood- and/or site-scale consideration of joint effects from not only greenspaces but also development patterns (e.g. building types, height, density). It is also vital that such ‘evidence-based’ planning encompasses not only the physical principles of urban climate, but also robust scholarly consideration of societal vulnerability and spatial justice in order to ensure that strategies and interventions can be developed in a way that benefits those that require them the most.

It is also important to bear in mind that greenspaces are living entities, hence there is a limitation to the cooling service they can provide. Greening is not a ‘silver bullet’ to the problem of excess urban heat. Greenspace has multiple functions which can be changed due to planning, and care ought to be exercised to ensure that planning greenspace to maximise cooling services does not weaken other functions. For example, a specific greenspace structure might facilitate the delivery of cooling services, but it might not provide a favourable structure for protecting urban biodiversity. Planning actions thus ought to take various ecosystem services into account and show awareness of the possible trade-offs between functions. Finally, in humid subtropical and tropical climates, increasing vegetation could enhance humidity in the air, which may have the effect of reducing thermal comfort. Effective planning actions must therefore pay cognisance not only to the cooling potential of green infrastructure, but also to its potential complexities and limitations.