Impact of well-watered trees on the microclimate inside a canyon street scale model in outdoor environment

Highlights

• The influence of well-watered trees on urban microclimate was investigated.

• A canyon street scale-model (1/5) in outdoor environment was instrumented.

• Climatic parameters, tree transpiration, and soil water status were measured.

• Trees reduced air and wall temperatures, cancelling out urban overheating.

• Trees reduced the level of heat stress, as measured by UTCI, in the canyon street.

Abstract

Cities experience overheating due to factors such as urban form and materials, concentration of human activities, reduction in the amount of vegetation and water surfaces. Vegetation is one of the ways to reduce temperature peaks in the city during heat waves. The objectives of this paper are twofold: first, to study the impact on the microclimate of a north-south oriented canyon street at reduced-scale (1/5), then to study the impact of well-watered trees on the street microclimate. This study provides a highly integrated view of climatic mechanisms through the measurement of a set variables, including air temperature, relative humidity, wall temperature, conductive and radiation fluxes as well as tree transpiration and thus allows a better understanding of the physical phenomena at stake. It shows that the canyon street created an urban overheating of up to 2.8 °C during the night, and up to 2.4 °C during the day, and that trees reduced the air temperature in the street by up to 2.7 °C during the day. Finally, trees improved human thermal comfort with a reduction of 8 °C of the Universal Thermal Climate Index at midday.

Introduction

Urbanization is the process of migration of people from rural area to urban area for the improvement of their lives (Kuddus and Rahman, 2015). At the beginning of the twentieth century, 15% of the world population lived in cities (Susca et al., 2011). In 2018, 55% of the world's population resided in urban areas, and by 2050, 68% of the world's population is projected to be urban (United Nations, 2018). In Europe, the urbanization rate was already 74.5% in 2018 and a projection indicates that this rate will reach 83.7% in 2050 (United Nations, 2018). The trend towards urbanization leads to the expansion of cities and the replacement of vegetation, natural soil, and water surface by impervious surfaces such as asphalt. The effect of urbanization on the urban thermal environment has attracted increasing research attention for its significant relationship to local climatic change and habitat comfort (Xiong et al., 2012). Among the consequences of urbanization on the urban microclimate is the Urban Heat Island (UHI) phenomenon, which is a heat accumulation process within an urban area due to urban buildings and human activities (Yang et al., 2016). According to Rizwan et al. (2008), UHI is one of the biggest environmental problems of the 21st century caused by the urbanization and industrialization of our society. Extensive studies of the characteristics of UHI effect were carried out in recent decades by authors such as Aboelata (2020); Arnfield (2003); Cantat (2004); Chen et al. (2006); Ridha (2017); Rizwan et al. (2008); Santamouris et al. (2001); Weng et al. (2004) and Yang et al. (2016). They showed that the UHI effects (in terms of intensity in particular) vary from one place to another as they are closely related to urban heat release, surface properties, vegetation coverage, population density. Measured air temperature in UHI can exceed 10 °C compared with the neighboring rural areas for the city of Athens (Santamouris et al., 2001) and 11.4 °C in Paris (Cantat, 2004).

Among the different existing urban topologies, canyon streets are a classical downtown configuration and therefore deserve a particular attention. A canyon street is a straight street continuously bordered by tall buildings. Since the pioneering work of Nakamura and Oke (1988), canyon streets are an emblematic configuration of urban microclimate research on which many data are available. They are generally defined by their aspect ratio which is the ratio between the height of the buildings and the width of the street (or the inverse depending on the authors) and the orientation of their main axis. Canyon streets are ideal places to observe different phenomena at stake in the UHI such as wind sheltering effect (when the wind blows perpendicularly to the street), and radiative trapping (resulting from absorption and reflection of solar radiation during daytime, and retention of long-wave radiation at night). They are also often considered as a geometry of reference in modelling tools. The TEB (Town Energy Budget) model for instance is able to simulate the climate of an entire city from the prediction of the climate inside a set of canyon streets (Lemonsu et al., 2004; Redon, 2017; Redon et al., 2020). Many outdoor experimental studies on canyon street focused on aeraulics and turbulence phenomena (Andreou, 2014; DePaul and Sheih, 1986; Georgakis and Santamouris, 2006; Louka et al., 2002; Najjar et al., 2005; Nakamura and Oke, 1988; Niachou et al., 2008; Rotach, 1995; Rotach et al., 2005) and will not be further commented here. In the following, we will focus our attention on studies dedicated to urban microclimate inside canyon streets, including thermal effects, without (Table 1) or with (Table 2) vegetation.

We have listed in Table 1 the main studies carried out on non-vegetated urban full-scale or reduced-scale canyon streets. The aspect ratio in this table is calculated as the ratio of building height to street width. The scales for the reduced-scale canyon streets are determined from a reference height of 10 m. Table 1 shows that prior to the 2000s, experimental studies on canyon streets were essentially full-scale. Indeed, full-scale outdoor experimental studies make it possible to assess several physical phenomena at the same time with fewer hypotheses. In particular for full-scale studies, Najjar et al. (2005) studied a canyon street in Strasbourg Eastern France oriented north north-east/south south-west with an aspect ratio of 0.9. The experiment was focused on the radiation balance of the street, but also looked at sensible and latent heat flux. Many authors have also been interested in the impact of canyon street orientation on the microclimate (Andreou, 2014; Eliasson, 1996). They have found that for the same aspect ratio, the interception of solar radiation by the solid walls, and thus the wall and air temperatures, depended on both the orientation of the street and the inclination of the sun. Andreou (2014) showed, for example, that the ground of north-south oriented streets, in comparison to east-west oriented streets, is more shaded during summertime, and less shaded during wintertime. As for the wall façades, north-south oriented streets receive more solar radiation than east-west oriented streets in summer. This was observed for aspect ratios within the range [0.6; 3]. The author also showed that the larger the aspect ratio, the more important the shading in the street, especially for North-South oriented streets.

However, more and more outdoor canyon street experimental studies now tend to be conducted at a reduced-scale despite the difficulties associated with the transposition of results to full-scale. To our knowledge, there are for the moment only 4 models of purely mineral canyon streets that have been built outdoors at reduced-scale, with scale reduction factor ranging from 1/2 to 1/8. Athamena et al. (2018) studied a reduced-scale (1/2) canyon street in an outdoor environment in Nantes, looking at phenomena coupling aeraulics and thermics. Idczak et al. (2007) conducted micrometeorological measurements at an experimental site (scale 1/2) located in the industrial area of Guerville (48°56′N, 1°44′W) in France. Idczak et al. (2007) observed that thermal effects were significant only in areas close to walls, and that buoyancy forces were generally negligible in the canyon street. Wang et al. (2017) carried out field experiments in an east-west 1/8 scale canyon street built in Sun Yat-Sen University, Guangzhou, China (23°4′N, 113°23′E) to study the impact of thermal mass of walls on air and wall temperatures. Their measurements show that during the daytime, walls with the lowest thermal mass reach their peaks earlier than walls with the highest thermal mass and that the opposite results occur after sunset. Chen et al. (2020) used the same experimental facility with various aspect ratios of the streets. Their studies showed that thermal storage capacity and aspect ratio are two essential factors that determine the urban microclimate. On the aspect ratio, they found that the wider the street, i.e., the lower the ratio between the height of the buildings and the width of the street, the higher the wall and air temperatures during daytime. On the contrary, during the night, the street with lower aspect ratio cools down faster.

As far as urban vegetation is concerned, Bowler et al. (2010) and Jamei et al. (2016) conducted a state-of-the-art review on studies focusing on the effect of green spaces on temperature and thermal comfort. Bowler et al. (2010) found that authors were more interested in the effect of parks and trees than in the effect of green roofs. All studies with urban green spaces show that they have an impact on the environment, at least locally. In canyon streets, trees can have cooling or warming effects on the air depending on their position on the street, prevailing wind conditions and time of day. During the day, the effects of trees are more noticeable in shallow streets because, in deep streets, their effects are masked by buildings (Bowler et al., 2010; Jamei et al., 2016). During the night, the air temperature beneath trees in deep canyons is slightly higher than the surrounding air temperature due to the low sky view factor that blocks infrared cooling (Jamei et al., 2016). In Table 2, we have listed studies dedicated to vegetation inside canyon streets, both at full-scale and reduced-scale. As it can be seen, most studies were conducted in temperate climate, as already pointed out by Bowler et al. (2010). In spite of the diversity of tree species used, it can be seen that trees organized in two lateral rows is the most frequently considered configuration. Table 2 also shows that information about vegetation such as LAI (Leaf Area Index, defined as the ratio of the cumulated leaf surfaces to the projected area of the crown to the ground), LAD (Leaf Area Density, defined as the ratio of cumulated leaf surfaces to crown volume) and part of ground covered vegetation vary from one study to another. Therefore, configurations are often not easy to compare to one another, and a very broad range of air temperature reduction by trees (from 0.4 °C to 6 °C) has been reported in the literature, depending on the authors. To date, most of the studies on the impact of vegetation on the microclimate in a canyon street have been carried out at full-scale. In Dresden, Gillner et al. (2015) showed that, thanks to the shading effects, vegetation can reduce surface temperatures by 5.5 to 15.2 °C and air temperature by 0.7 to 2.2 K depending on the tree species used. The authors also showed that trees can reduce asphalt temperatures by up to 4.6 °C per unit of LAD Gebert et al. (2019) are some of the few authors to have measured the water content by volume in soil. Their results showed that stomatal conductance (and thus tree transpiration) was closely related to soil water availability. On the impact of trees on human thermal comfort, Coutts et al. (2016) found that, during heat events, trees in East-West oriented Canyons with average aspect ratio comprised between 0.27 and 0.76 had a low impact on air temperature (0.9 °C at mid-morning) but a significant impact on diurnal index in summer, largely due to a reduction in the mean radiant temperature, reducing heat stress from a very high level (UTCI >38 °C) to a high level (38 °C > UTCI >32 °C). We found only three studies in the literature (Djedjig et al., 2015; Ouldboukhitine et al., 2011; Park et al., 2012) that were conducted in reduced-scale street. The first two studies focused only on green walls and roofs and Park et al. (2012) were the only ones to consider trees in their street. Park et al. (2012) conducted a study in Japan (39°04′, 139°07′E) on a small-scale (1/10) urban model on the effect of urban vegetation on the outdoor thermal environment. They found that trees, in comparison with a non-vegetated modality, could reduce the globe temperature (which incorporates the effects of radiation, so this magnitude is different from the air temperature) by 0.6 to 2.2 °C depending on the street orientation and sensors position.

In this paper, we propose to study the impact of a canyon street on the microclimate, and then to study the impact of trees on the microclimate within the canyon street. To reach this goal, we first present the 1/5 scale canyon used for experiments, the instrumentation, and the method of calculation of several quantities of interest. Second, in addition to the variables classically studied in the literature such as air temperatures, wall temperatures, relative and absolute humidities, we analyze other variables such as tree transpiration, radiation and its interception by buildings and trees both in areas without or with trees. An originality of the present study also relies on the fact that the soil was instrumented, which gave access to the soil water status, whereas this information is very rarely available in urban studies. We also investigate the thermal comfort inside the street in the vegetated and non-vegetated modalities and compare it to a control rural-like environment outside the street. A discussion including considerations on the representativeness of the reduced-scale street and on the limitations and perspectives of the present work is also provided.