Urban Micro-climate Research in High Density Cities: Case Study in Nanjing Academic research paper on "Earth and related environmental sciences"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedía Engineering 169 (2016) 88 - 99

Procedía Engineering

www.elsevier.com/locate/procedia

4th International Conference on Countermeasures to Urban Heat Island (UHI) 2016

Urban Micro-climate Research in High Density Cities: Case Study

in Nanjing

Jingjin Li a*, Jianguo Wang a, Nyuk Hien Wong b

aSchool of Architecture, Southeast University, Sipailou 2nd ,Nanjing 210096,China bSchool of Design and Enviroment, Department of Building, National University of Singapore,singapore,117566,Singapore

Abstract

What kind of high density urban form presents a better micro-climate environment in summer days is the main question of this study. The micro-climate of three different urban form were simulated using the software ENVI-met. Parameters including building density, floor area ratio, green plot ratio, pavement area, building height are concerned in the simulation. By comparing and analyzing the air temperature (Ta), solar radiation, mean radiation temperature (Tmrt) and wind speed, the micro-climate performance in three cases are presented. The alternative thermal index physiological equivalent temperature (PET) is also used to evaluate the outdoor thermal comfort in the three cases.

©2016 The Authors.PublishedbyElsevierLtd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the organizing committee of the 4th IC2UHI2016

Keywords: urban micro-climate; urban form; ENVI-met; urban heat island; high density; Nanjing

1. Introduction

The rapid urbanization in China is characterized by the expansion of urban area and the increasing of building density and height, which has significantly changed the sky view factor, solar reflectivity, heat capacity and roughness of the land surfaces. The phenomenon that air temperature in urban areas being significantly higher than that in rural areas is commonly termed as urban heat island (UHI) effect [1]. UHI effect can be observed all over the world, Gedzelman found that the maximum heat island intensity in New York was about 8.0°C [2]. It was reported that the

* Corresponding author. Tel.: +86-136-5516-6606; fax: +86-025-8379-4517. E-mail address: seuljj@163.com

1877-7058 © 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of the organizing committee of the 4th IC2UHI2016 doi: 10.1016/j.proeng.2016.10.011

annual mean air temperature in the centre of Tokyo city increased by 3°C remarkably during the last century. The UHI does not only impact the urban environment but also increases the mortality and the economic burden. A study in Netherland demonstrated that each degree increase in the air temperature above the comfortable range caused an mortality increase of 2.72% [3].The Department of Power in U.S. reported that government spent about 10 billion dollars on mitigating the UHI every year [4]. Considering that over 50% Chinese people are living in the city now, it is easy to understand why this topic is receiving more and more attention.

Many researchers have focused on the factors such as urban population, albedo of the ground surface, urban vegetation which makes contribution to UHI. Asphalt's albedo value is about 0.1 on average and the concrete is 0.3 or 0.4, the low albedo cause the material temperature rise rapidly when exposed to solar radiation [5]. It is reported that proper vegetation can reduce the air temperature by 5°C in Singapore [6]. Dimoudi reported that the wind speed in the pedestrians' level inside the urban canyon is the 1/3-1/4 of the suburban area [7].

Few researches have been conducted to compare the micro-climate conditions in different high density cities. The micro-climate conditions in High density urban are concerned by the urban designers and citizens. The next section will present the progress of the UHI in Nanjing and highlight the location of the high surface temperature areas. Section 3 and Section 4 show the method and results of the case study. After the discussion of the main results, the research reports which kind of high density form is the reasonable choose to mitigate the Urban heat island in Nanjing.

2. Urban Heat Island in Nanjing

Nanjing the capital of Jiangsu province is situated in the Yangtze River Delta with the latitude of E31°14" to 32°37" and longitude of N118°22" to 119°14". Under the influence of the East Asia Monsoon, Nanjing has a humid subtropical climate. The annual mean air temperature in Nanjing is 15.9°C. Nanjing is usually referred to the "Furnacelike City" for the high air temperatures during the summer, even though the greenery coverage of Nanjing is about 44%, which is highest among those of all the cities in Jiangsu Province. Fig. 1 shows that July and August are the hottest months and the average maximum temperature is 32°C [8] while January is the coldest month with the average minimum temperature of -1°C. The average temperature difference between summer and winter is about 10°C. The average UHI intensity of Nanjing is about 0.5°C, reaching approximately 6°C in extreme cases. From 1960 to 2009,with the development of the urban scale, calculated by annual air temperature, the UHI intensity increased by 0.109°C/10a [9].

Fig. 1. Average monthly maximum and minimum air temperature in Nanjing (1951-2008)

The land surface temperature collected from satellite imagery clearly shows the UHI effect in Nanjing (Fig. 2). The main urban area shows the highest land surface temperature from 28 to 32°C. In contrast, the Yangzi river and Zijing mountain shows the lowest surface temperature about 23°C. The mean surface temperature of the rural area is about 26°C.

3 " I

j ijjj T -m/m w. ■ ■ № H 1 • 1 ■ ■

k'Ju"*- J 9 ^ J M

r < B9

16 18 20 22 24 26 28 30 32 34 36 38 40 (°C )

Fig. 2. Land surface temperature of Nanjing by

16 18 20 22 24 26 28 30 32 34 36 38 40 (°C )

data: (a) September, 2002; (b) September, 2010

In the recent ten years, with the fast expansion of urban built-up areas, the structure of the urban form of Nanjing has experienced great changes both in horizontal and vertical. The boundary of built-up areas enlarged especially in the part south of Yangzi River. Large numbers of tall buildings were built in the central urban areas for the growing population in the city, which is not beneficial for ventilation. These factors were answerable for the increasing of the UHI effect. In September of 2013, the mean land surface temperature of the urban area had grown up to 35°C, the form of area with the highest temperature has enlarged from spots to belts. The maximum resolution of the land surface temperature picture is about 1 km2, the data ignore the contribution of the building's surfaces.

Especially the high tower buildings in the city centre which have high impact for the solar heat absorption and reflection. From 2001 to 2010, 63 cases with buildings higher than 50 meters were built inside the old city. The recent research suggested a combination of detailed analysis at the street block scale in order to explore the effects of the urban forms. Hence the motivation to investigate which type of the high density urban form may be implemented to mitigate the UHI effects.

3. Method

Three sites representing different high density forms in the central business area were selected with the size 400 m x 400 m. Indexes including land coverage, building density and average building height are calculated in Table 1. Site a (Xinjiekou city centre) is the typical central business district combined with business and shopping center in china. This site is constituted of massive high tower buildings with podiums, the highest tower is about 260 meters while its average building height 260 m the building coverage ratio is the largest of all the three sites due to the application of the podium. Site b (Daxinggong) as a culture and office center is combined with density high towers in the south part and a square in the north, twelve high tower were built within 8 hectares which is much denser than Site a, also the

average building height of Site b is highest of all the three sites. Site c is another type which is combined with courtyard building and high tower buildings.

The value of land use, building height including average and maximum and floor area in each site were show in Table 2. Building coverage ratio (BCR) is the ratio of the building area divided by the land area. Building area means the floor space of a building when looking down at it from the sky. Floor area ratio is the ratio of a building's total floor area divided by the size of the piece of land upon which it is built.

Fig. 4. (a) Satellite image of site a; (b) Satellite image of site b; (c) Satellite image of site c; (d) Photo of site a; (e) Photo of site b; (f) Photo of site

Table 1. Land use and building height properties in each site

Site a

Site b

Site c

Land Cover Building 0.32 0.24 0.25

Road 0.14 0.15 0.09

asphalt

Pavement 0.54 0.61 0.66

Building Maximum 260m 150m 100m

Height Average 38.7m 49m 34.3m

Floor Area

1,050,000m2

1,150,000 m2

980,000 m2

Tower + Podium Density tower + square Tower + courtyard _buildings_

3.2. Urban environment simulation

ENVI-met was designed for simulating the urban microclimate in research, and has been extensively validated and used worldwide [10]. In this study, the building materials were the same in the three sites, and the anthropogenic heat emissions were not considered. The research mainly compared the different building layout and urban canopy. June

23rd were selected as typical summer day. The simulation time starts from 2:00 for 24 hours. The weather data measured in the city were derived for the simulation [11]. The simulations evaluate the parameters in grids with dimensions 4 m x 4 m x 10 m (Fig. 5). The initial input parameters for ENVI-met are presented in Table 2. To improve the accuracy of the results, the boundary data is not calculated in this research.

Table 2. Details of initialization input parameters in Envi-met

23rd June, 2015

Starting time 2:00

Wind speed 2.4 m/s

Wind direction SSE (157.5°)

Initial Temperature 295K

Relative humidity at 2m 50%

Building temperature 293K

Fig. 5. The model in Envi-met (a) site a; (b) site b; (c) site c

3.3. Assessing parameters

The following six parameters including sky view factor, solar radiation, mean radiant temperature, air temperature, wind speed, physiological equivalent temperature (PET) were used for the analysis of the relationship between different urban form and micro-climate.

Sky view factor is the percentage of sky area been seen when you looking up to the sky, which is used to evaluate the spaciousness. ENVI-met is used to calculate the value the present of both building and the vegetation.

Solar radiation is the power per unit area produced by the Sun in the form of electromagnetic radiation, the unit of irradiance is watt per square meter (W/m2) [12].

Mean radiant temperature (TWO is defined as the uniform temperature of an imaginary enclosure in which the radiant heat transfer from the human body is equal to the radiant heat transfer in the actual non-uniform enclosure[13].If relatively small temperature differences exist between the surfaces of the enclosure, the equation can be simplified to the following linear form:

Tmt = T Fp + T2 Fp_2 + ... + TnFp _ n (1)

Tn is the temperature of surface "n", in Kelvins; Fp-n is the angle factor between a person and surface "n".

The Tmrt can be measured using a black-globe thermometer. For the standard globe (D=0.15m, = 0.95): Tmrt=[(GT+273)4 +2,5-108-va°,6 (GT-Ta) ]1/4-273 (2)

where GT is the globe temperature, va is the air velocity at the level of the globe (m/s), £ is the emissivity of the globe, Ta is air temperature (°C).

Compared to predicted mean vote (PMV), perceived temperature (PT), outdoor standard effective temperature (OUT_SET), physiological equivalent temperature (PET) has been widely used for outdoor thermal comfort

assessment, and it has been adopted as an important index by the German guidelines for urban planners. Therefore, PET is applied in this study to evaluate the outdoor thermal comfort. According to Mayer and Hoppe's research, PET is linear correlation with the Tmrt: [14] PET=0.713 Tmrt-2.274

Image processing in Matlab7 is applied to calculate and imaging the PET data. Table 3 shows the thermal sensations and PET classes for the Nanjing (Based on Lin and Matzarakis 2008). The standard is applied in this study to evaluate the outdoor thermal comfort for the three sites.

Table 3. Thermal sensation for Nanjing, Taiwan and West Europe

Thermal sensation PET range in Nanjing (°C PET) PET range in Taiwan (°C PET) PET range for West Europe (°C PET)

Very cold <-4 <14 <4

Cold -4-8 14-18 4-8

Cool 8-16 18-22 8-13

Slightly cool 16-22 22-26 13-18

Neutral 22-28 26-30 18-23

Slightly warm 28-32 30-34 23-29

Warm 32-38 34-38 29-35

Hot 38-44 38-42 35-41

Very hot >44 >42 >41

4. Results

4.1. Sky view factor

Fig. 6 shows that the sky view factor is variety in the three city centres. Site a has the highest average SVF about 0.61 due to the layout of the tower buildings and the least vegetation. The SVF on the N-S street is about over 0.65 while it ranges from 0.2 to 0.6 inside the blocks. The average SVF of Site b is 0.47, this site is divided into two parts by the E-W road, the south part has the lowest SVF ranges from 0.2 to 0.4 while the north part urban square shows higher SVF from 0.3 to 0.7. The average SVF of Site c is 0.54, which is between site a and site b.

Fig. 6. Sky view field of the three sites (a) site a; (b) site b; (c) site c

Normally a higher SVF induces a higher Ta, but in this research due to the complex urban form, the relationship between Ta and SVF is very weak both during day and night. The correlation between SVF and Tmrt is different between day and night. At 2:00 night, the R2 between SVF and Tmrt is about 0.68. Higher SVF means more emission of energy from urban geometry which leads to lower Tmrt. But during Day time the relationship between SVF and Tmrt simulated in is not observed in this simulation. High SVF do not means high solar radiation which is a key effect on Tmrt during daytime (Fig. 7).

Fig. 7. Regression analysis of SVF between Tmrt and Ta (a) SVF and Ta 2:00; (b) SVF and Ta 14:00; (c) SVF and Tmrt 2:00; (d) SVF and Tmrt

4.2. Air temperature (Ta)

Fig. 8 indicates the trend of the Ta change in June 23rd of the three sites. The temperature of the three sites slightly decrease from 4:00 to 7:00, then increase sharply during 7:00 to 13:00, after 13:00 it decreased gradually. Among the three sites, site c owns the higher temperature than the other two sites, the biggest Ta difference between is about 3°C which occurs at 20:00. The average temperature of site c remains steady at 25°C from 13:00 to 17:00 while the other two decreased by 1°C. Site a keeps the lowest average air temperature from 19°C to 25°C which is about 2°C lower than the other two sites after 8:00. Site c shows the highest Max Ta about 29°C in the three sites. The max Ta of site a and b occurs in 13:00. Max Taof Site b is about 0.8°C higher than site a. As the most important Business Centre of Nanjing, site a has the largest building area while its Ta is the lowest in the three sites, which means that the increasing of the building area is not the main cause of the higher outdoor air temperature.

Fig. 8. Comparison of three site's air temperature chart: (a) average air temperature; (b) max

Fig. 8 shows the largest difference between the max Ta and the min Ta in one site is about 2°C. The high temperature area of Site c is distributed in the south and east of the research site mainly due to the asphalt road's low albedo. Compare to site a, the main high temperature area of Site b is concentrate on the southeast and southwest of the site out of the main road. The main reason may be the high towers shading significantly reduce the environment

temperature. Site a owns a small area about 3200 m2 with high temperature which is much smaller than the other two sites.

From the perspective of air temperature, in summer days the micro-climate condition of site a performs better than the other two sites.

Fig. 9. Air temperature distribution at14:00 (a) site a; (b) site b; (c) site c

4.3. Solar radiation

Fig. 10 shows the trends of the solar radiation of the three sites, which presents obvious linear relationship with time. There is no solar radiation in the three sites before 6:00 and after 18:00. From 6:00 to 12:00, the average direct solar radiation increased sharply from 70 W/m2to almost 600W/m2, after 12:00 it drops gradually to 320 W/m2 at 16:00. Site a gets much more solar radiation than other two sites before 12:00, after that site a and site c shows almost the same average direct solar radiation. The high towers concentrated on the east part of the site c which supply better shadow before 12:00 may be the main reason cause the difference. Among the three sites, site b receives the least solar radiation of about 50 W/m2 less than the other two sites.

1200 ; 1000 :

■ AAvgSR

■ B AvgSR C AvgSR

■ A MaxSR

■ 8 MaxSR C MaxSR

soqqqooSqSqooo

Fig. 10.Comparison of solar radiation chart

Fig. 10 shows the max solar radiation is divided into two parts by 12:00 showing an obvious parabola feature. The max solar radiation doubled from 350 W/m2 to 700 W/m2 during 6:00 to 7:00, from 8:00 to 12:00 the fluctuation of solar radiation is much smaller. The max solar radiation of site c is 1050 W/m2 about 30 W/m2 higher than the other two sites, which can be observed from 10:00 to 14:00. From the perspective of solar radiation, in summer days the micro-climate condition of site b performs better than the other two sites.

4.4. Mean radiant temperature (Tmrt)

Average Tmrt of outside environment is calculated to compare the changing trends of Tmrt. Compare to the Ta variation which varies only 9°C between the day-time and night-time, the Tmrt variation is over 40°C. Tmrt rises up from 6:00 with the growing solar radiation, the average Tmrt increased from 20°C to 52°C during 6:00 to 9:00, from 9:00 to 17:00 the Tmrt keep stable between 50°C to 60°C, after 17:00 the Tmrt drop dramatically to 20°C in two hours. Though the average Ta of Site a is much lower than site c,but its average Tmrt is larger than site c from 7:00 am to 13:00 pm, the largest difference between the two sites is about 4°C. Site b keeps the lowest Tmrt in the three sites which is the same as the solar radiation.

The max Tmrt is caculated to find the discomfortable areas in the three cases, Fig. 11 shows the fluctuate of the Tmrt, two high points of Tmrt during day-time, Site c has the max Tmrt about 80°C in this simulation which occurs at 14:00 ,while another high Tmrt occurs at 9:00.

I50 " 140

■ AAvgTmrt

■ 8 Avg Tmrt

■ B MAX Tmrt

■ C MAX Tmrt

<r <*■ «¿v

Fig. 11. Comparison of mean radiant temperature chart (a) average Tmrt ; (b) max Tmrt

Fig. 12 presents that Tmrt in the three sites was classified to four main grades, the highest Tmrt concentrated on the main roads of the research area, the Tmrt in the shadow area is about 25 to 30°C lower than the area exposured to the direct solar radiation nearby,the trees also have the effect of reducing the Tmrt of about 20°C. Site a and c show larger area with high Tmrt than site b, the main reason is that the size of the open area in site a and c are much larger than site b. From the perspective of mean radiant temperature, in summer days the micro-climate condition of site b performs better than the other two sites.

Fig. 12. Mean radiant temperature distribution at 14:00 (a) site a; (b) site b; (c) site c

4.5. Wind speed

Wind speed as an important factor in reducing the urban heat island effect is also been analyzed. Fig. 13 shows that site a owns the highest average wind speed from 1.2 m/s to 1.4 m/s, the reason is that the main road is relevant to the main wind direction in summer. Though site c has the same road like site a, but the dense trees along the road reduce the wind speed obviously. Site b contains the least wind speed from 0.9 m to 1.1 m due to the different angle between the main street and prevailing wind direction.

■AAveWs

■ B Ave Ws

■ CAveWs

.<$> .<$> <$> .<$> .<$> .o0 .<$> .cP .<£> .1

Fig. 13. Comparison of wind speed (a) average wind speed; (b) max wind speed

Comparing to site a and site c, site b's density high tower buildings facing the wind direction may be another reason of the low wind speed. The fillet design for the podiums in the street corner in site a is beneficial for increasing the

wind speed in the street. In site c, the rectangle corner of the high tower buildings significantly increased local wind speed, which may cause uncomfortable experience at pedestrian level.

Wind speed in this simulation is lower than 5 m/s. In summer days, according to the wind comfort criteria [15], higher speed means better micro-climate condition. From the perspective of wind speed, site a performs better than the other two sites.

Fig. 14. Wind speed distribution at 0:00 (a) site a; (b) site b; (c) site c

4.6. Physiological equivalent temperature (PET)

Fig. 15(a) shows that the max PET of site c is the largest of the three sites which is about 36°C, 3°C higher than Site a and Site b in 15:00. Site b has the lowest PET, even though its windspeed is lower than the other two sites.

According to Fig.15(b), the PET of three sites is divided into five groups from 'cool' to 'warm', Although in this simulation, the PET all belong to the warm class, but in hotter summer days this difference may cause uncomfort feeling in site c while the other two still in the range of comfortable.

---Site AI b

2 1500

Fig. 15 (a) PET Fluctuation; (b) distribution of average PET

Fig. 16. shows that high PET area occurs on road in east-west direction compare to the south-north direction in all the three sites. Site a and Site c also contains some hot spots in the district with more than 30°C . From the perspective of PET, site b owns better micro-climate condition than the other two sites.

a BilWUM' i1 illWBlflil1 b ■¡¡■■■■I C

Fig. 16. Distribution of PET: (a) site a; (b) site b; (c) site c

5. Discussion

5.1. Integrate the different indices

From different perspective of indices, the answer of which site performs better is different. Table 4 is the combination of the five different indices, Site b performs well in three indexes including the physiological equivalent temperature and mean radiant temperature.

Table 4. Combine the different comparing results

Air Temperature Solar Radiation Mean Radiant Temperature Wind Speed Physiological Equivalent Temperature

Site a Good Poor Moderate Good Moderate

Site b Moderate Good Good Poor Good

Site c Poor Moderate Poor Moderate Poor

5.2. Relation between different indices

According to the simulation results of site a, high solar radiation does not directly correlate with high air temperature, but it effects the Tmrt and PET. The relationship between building and open space needs to be carefully designed to avoid unnecessary solar radiation.

The Tmrt of site b is much lower than the other two sites mainly due to two reasons: firstly, the high density of the towers in the south part of the site provides more shading than the other two sites. Secondly, the trees on the open square afford proper shading which avoid the direct solar radiation.

Though wind speed in site b is the lowest of the three sites, it still has the lowest average Tmrt , Compare to solar radiation, wind speed is not the main factor effect the Tmrt. Site a own the largest wind speed mainly due to the chimney effect of the street. Though Site a and Site c both have a south-north street, probably the different width of the street.

5.3. Relation between urban geometry and urban micro-climate

The interest phenomenon is that from the satellite image, we find that high temperature area are mainly concentrated in the high density area, where congested with high towers. But on the view of simulation, more building area or building height do not means higher temperature. For example, site a with the highest building area has the lowest air temperature.

Compare to site a, site b and c have much more trees but higher Ta, which means trees are not the main factor effect on the air temperature in this simulation.

5.4. Area of research

As Michael batty said about space syntax analysis, the scope of the research area effects the accurate of the results directly, the range of the ENVI-met research area also impacts the simulation results greatly. Because the surrounding environment's effect on the modelling area has been ignored, the climate date on the edge of the model area may be not accurate enough for analysis. The safety buffering area of modelling is also need to be researched.

6. Conclusions

Results of this study show that different urban form has significant effect on the urban microclimate and outdoor thermal effect.

It is important to note that compare to urban geometry, massive building area is not the main reason to affect the urban micro-climate. Site a got the lowest average Ta which is lower than those at the other two sites.

The average building height offered enough urban shading which is an important factor effect the value of Tmrt. Site b with the highest average building height got the lowest average Tmrt of 54°C at 14:00, which is about 6°C lower than those at the other two sites.

Normally courtyard building model can provide better outdoor thermal comfort, but in this simulation the distance between the buildings plays more important role than the building type. Site c with the courtyard building but still owns the highest PET value in the three cases.

Finally, without considering the thermal emission of the buildings in this simulation, this study also confirmed that among the three sites, site b is the suitable choose for a better urban micro-climate.

Acknowledgement

The research was funded by the provincial advantage Discipline Fund (No.1101007002), the Architecture School of Southeast University.

Reference

1 Oke, T.R.: 'City size and the urban heat island', Atmospheric Environment (1967), 1973, 7, (8), pp. 769-779

2 Gedzelman, S.D.: 'OPTICS, ATMOSPHERIC | Optical Phenomena', in Holton, J.R. (Ed.): 'Encyclopedia of Atmospheric Sciences' (Academic Press, 2003), pp. 1583-1594

3 M.M.Huynen.P.Martens, D.S.: 'The impact of heat waves and cold spells on mortality rates in the Dutch population', Health Perspect, 2001, 109(5)

4 Rosenfeld.A.H, R.J., Akabari.H,Pomerantz.M,Taha.H,: 'Policies to reduce heat island: Magnitudes of benefits and incentives to achieve them', Proceedings of the 1996 ACEEE Summer Study on Energy Efficiency in Buildings, 1996, pp. 177-178

5 Akbari, H., Pomerantz, M., and Taha, H.: 'Cool surfaces and shade trees to reduce energy use and improve air quality in urban areas', Solar Energy, 2001, 70, (3), pp. 295-310

6 Wong, N.H., Tan, A.Y.K., Chen, Y., Sekar, K., Tan, P.Y., Chan, D., Chiang, K., and Wong, N.C.: 'Thermal evaluation of vertical greenery systems for building walls', Building and Environment, 2010, 45, (3), pp. 663-672

7 Dimoudi, A., Kantzioura, A., Zoras, S., Pallas, C., and Kosmopoulos, P.: 'Investigation of urban microclimate parameters in an urban center', Energy and Buildings, 2013, 64, pp. 1-9

8 Tang, C.-S., Shi, B., Gao, L., Daniels, J.L., Jiang, H.-T., and Liu, C.: 'Urbanization effect on soil temperature in Nanjing, China', Energy and Buildings, 2011, 43, (11), pp. 3090-3098

9 Zeng, Y., Qiu, X.F., Gu, L.H., He, Y.J., and Wang, K.F.: 'The urban heat island in Nanjing', Quaternary International, 2009, 208, (1-2), pp. 38-43

10 Taleghani, M., Kleerekoper, L., Tenpierik, M., and van den Dobbelsteen, A.: 'Outdoor thermal comfort within five different urban forms in the Netherlands', Building and Environment, 2015, 83, pp. 65-78

11 CMA. http://www.weather.com.cn/weather/101190101.shtml

12 Wikipedia. (2015). https://en.wikipedia.org/wiki/Solar_irradiance.

13 ISO7726: 'Ergonomics of the thermal environment-Instrument for measuring physical quantities ', Geneva, Switzerland: International Organization for Standardization. November, 1998

14 Zheng, W.: 'Street in Green City: Research on Street Design based on Microclimate of City' (Southeast University Press, 2010. 2010)

15 Janssen, W.D., Blocken, B., and van Hooff, T.: 'Pedestrian wind comfort around buildings: Comparison of wind comfort criteria based on whole-flow field data for a complex case study', Building and Environment, 2013, 59, pp. 547-562