The renew plans of urban thermal environment optimization for traditional districts in Xi'an, China

During the process of the high-speed urbanization in Chinese cities, the social, economic, and political status and the interaction between each factor have been more focused on urban traditional district renewal. However, the effects on urban microclimate and the residential living conditions in traditional districts are not well discussed, which is strongly related to the living comfort and citizens’ well-being. In this study, two typical traditional districts in Xi'an are selected. According to the original situation of building functions and the community characteristics, two renewal plans are proposed by adding vegetation in open spaces (V), and adding vegetation combined with building redevelopment (V&B), in order to balance the living convenience and thermal environment. Via ENVI-met simulation, the effects of the district renewal plans on thermal environment including wind speed, air temperature, and mean radiant temperature are evaluated. This study provides method of environmental evaluation for traditional district renewal, which contributes to sustainable urban planning in historical districts, and provides recommendations for related policy development.


Introduction
Along with the large-scale and high-speed urban development in cities, some of the traditional districts inside the city are falling behind the average quality of living conditions. More and more attention is paid on traditional district renewal in the cities. Meanwhile, urban design is a process involving multiple aspects. The urban physical and functional changes during the urban renewal would affect the urban microclimate condition (Couch 1990). However, in the process of urban development and redevelopment, urban climate issues have constantly been put in an insignificant position behind the important and dominant influencing factors, such as politics, economy, society, transportation, and landscape. Especially in the process of urban design which emphasizes urban function planning, the effects on climate parameters are rarely evaluated (Patra et al. 2018). However, urban design based on climatic evaluation is conducive to create a comfortable and pleasant urban environment, which could reduce environmental effects caused by urban heat island phenomenon (Nakano 2015;Santos et al. 2021;Okumus and Terzi 2021). Environmental evaluation on traditional districts renew should be considered.

Urban renewal
In "Xi'an's Urban Renewal Measures" (2021), urban renewal is an activity to rectify, improve, and optimize the urban spatial form and function according to the city's national economic and social development planning and land space planning. It is a process of slum clearance and redevelopment, which aims to improve the urban functional service, social-economic, and ecological through redevelopment, rehabilitation, and heritage preservation (Couch et al. 2011;Lai and Lorne 2019). In recent decades, it has also become Responsible Editor: Philippe Garrigues * Dian Zhou dianz_lab@163.com 1 a very important research area in the field of urban and rural planning (Pérez et al. 2018). Urban renewal is a comprehensive process that includes various aspects (Guo et al. 2017). In the process, deteriorated buildings and living environment should be improved (Yung et al. 2017). In Berlin, Germany, the tendency for buildings in poor (good) conditions in the target area to decrease (increase) by an average of 1.2-3% (0.6-2.5%) per year is guided by the space renewal policy (Ahlfeldt et al. 2016). Furthermore, historic buildings and spatial structures need to be preserved. Through the investigation of heritage buildings in two urban renewal districts of Hong Kong, the protection and redevelopment measures for these buildings are proposed. In addition, partial urban area can be reshaped by improving building layouts, open spaces, road networks, and other infrastructures (Yung et al. 2017). It has also become a key issue in the traditional period of China (Romano 2020). By applying the historical land utilization model in the Yau Tsim Mong District in Hong Kong, four future land use scenarios-the baseline scenario, the open space scenario, the residential scenario, and the balanced scenario-were simulated for 2018, which provide alternatives for reshaping future urban areas (Zheng et al. 2015). From the perspective of sustainable land use, mitigation, and improvement, a variety of measures and policies are proposed for urban renewal in Shenzhen, China Liu et al. 2017;Lai et al. 2020).
However, in the urban renewal planning and design regulations, improving urban microclimate is less considered (Lin et al 2017;Bartesaghi-Koc et al 2021). Often, the thermal conditions are worsened by poor urban design, and urban regulations are poorly adapted to the climate (Johansson and Yahia 2015). Therefore, the research between urban renewal and climate has gradually been concerned. For local area renewal, in Rome, the impact of a new building was analyzed on the thermal environment of the actual surrounding urban space in the redevelopment area (Battista et al. 2016). In the central areas of Dar es Salaam, Tanzania, a variation in building height would be better to enhance the microclimate in the implementation of urban renewal schemes (Johansson and Yahia 2015). In Guangzhou, China, the changes in population density, building density, and landscape in urban renewal would lead to different increases in land surface temperature (Qiao et al. 2020). However, the local change of buildings and greening in urban renewal areas was found to decrease land surface temperature and improve the thermal comfort (Wu et al. 2022).
For the city, a new urban form of development was proposed based on possible climate-optimal forms, which can be better to cultural, ecological, and economic services and benefits (Ramyar et al. 2019). These provide advice for planners to combine the urban design and weather factors in the process of urban renewal. Therefore, this study will explore local urban renewal based on the improvement of the thermal environment.

Traditional urban districts renewal
The traditional district renewal faces comprehensive problems, involving the improvement of economic, social, and environmental benefits, and the comprehensive utilization of traditional features and cultural heritage (Li et al. 2019). For the renewal plan of the traditional district, it usually involves large-scale demolition and redevelopment, which lead to destruction of existing social networks, expulsion of vulnerable groups, and adverse impact on living environments (Wang and Aoki 2019). Therefore, the protection of the traditional district has been proposed. In Hong Kong, in order to protect the historical integrity of the traditional districts, the collective impact of conservation of historical buildings in the two urban renewal areas was investigated and evaluated by using questionnaires, which determined the necessity of protection (Yung et al. 2017). Although it is necessary to adopt appropriate measures to protect the traditional districts, the living environment needs to be improved. This is because for residents in the traditional districts, as the living conditions gradually deteriorate, they may rebuild historical buildings and courtyard spaces spontaneously and arbitrarily to meet the needs of modern life, which can pose a threat to the protection of the historical authenticity and style integrity of the districts (Liwei and Siyuan 2018). Therefore, it is not enough to use only protection way.
Compared to the protection way, the way of demolition and reconstruction is more adopted in modern society. It is because as the city expands, the traditional district gradually becomes a central part of the city, which makes its commercial value grow rapidly. Therefore, such districts are often demolished and rebuilt into commercial areas in some cities on a large scale, which has lost the historical and cultural characteristics (Li 2020). In 1960 San Jose, California, the urban renewal plan was adopted in a district, which completely demolished all the buildings and replaced them with skyscrapers, and led the original residents to go to different communities, local businesses to relocate or close down, and buildings with research value were disappeared (McDaniel 2017). This shows that the complete demolition and reconstruction are also unacceptable. Consequently, in an urban renewal area, the appropriate design of the built environment contributes to economic and social sustainability by enhancing the attractiveness of the district and improves the environment by forming open and green spaces (Yildiz et al. 2018).
In addition, the renewal of traditional districts also affects the local climate of the city. During the process, especially in hot summer, the local outdoor microclimates and the thermal comfort of old city district should be focused (Peng et al. 2015a, b, c). For this aspect, researchers often use microclimate simulation software to conduct the research. By using CFD simulation, the architectural elements that could moderate outdoor thermal conditions were explored, and the role of open spaces and alleys were evaluated in a traditional urban district quantitatively (Shaeri et al. 2018). By using ENVI-met software, the effectiveness of strategies, such as green roofs, cool roofs, cool pavements, green areas, and air passages, was verified for improving the thermal environment in a dense and old city district (Evola et al. 2017;Peng et al. 2015a). Therefore, research on the improvement of thermal environment in the process of traditional district renewal has gradually attracted attention. Based on the local climate, the issue is also explored by using simulation software in this study.

Urban greening
Urban greening is an effective way in mitigating excessive heat and improving the outdoor thermal environment for residents in cities. It can mitigate summer mean, maximum, and minimum temperatures, which can improve the thermal comfort (Song and Wang 2015). Therefore, diverse green or open spaces in urban areas are widely considered to be important for health (Jennings et al. 2016), which can enhance urban micro-climate, protect water quality, and improve air quality (Nice et al. 2018;Sharma et al. 2018;Zhang et al. 2020).
In different urban spaces, such as urban street, open space, and the high-density area, the effectiveness of adding vegetation to improve urban thermal environment has been verified. For urban street, the proper planting pattern and proportion of the tree can improve the thermal comfort (Yang et al. 2018a, b). For urban open space, planting vegetation can decrease air temperature 2 K . For high-density areas, the microclimatic effect of urban greening was verified in Hong Kong (Tan et al. 2017), and 10% increase in the urban vegetation can effectively decrease the precinct temperature by 0.6 °C in Sydney's high-density precincts (Sharifi and Lehmann 2015). With adding greening in different urban spaces, green spaces are formed, which can reduce the temperature in the commercial and single residential areas (Song and Park 2015). Moreover, these spaces provide an important place for public communication that can enhance social interaction, especially in the urban renewal district with a relatively high concentration of elderly people (Yung et al. 2016). Therefore, urban greening can not only add the social function value, but also improve the microclimate (Yang et al. 2019).

Urban building space renewal
The traditional district is often a high-density and low-rise building area. It is providing a big impact on the urban thermal environment (Wang et al. 2018). Simultaneously, lowrise buildings lead to more stressful urban spaces than the areas with high-rise buildings (Yahia et al. 2018). In Bangkok, the large low-rise building zone and compact low-rise building areas have a higher mean land surface temperature (Khamchiangta and Dhakal 2019). In China, based on the local climate zone in Nanjing, the mean heat island magnitudes from 22:00 to 2:00 and the differences in diurnal temperature range of the compact low-rise with compact mid-rise space are higher than these of the open high-rise space (Yang et al. 2017). Therefore, the high-density and low-rise building space is easy to form a higher temperature environment.
Some researchers have suggested that the increase in building height can improve the outdoor thermal comfort. This is because of the high-rise urban space, although not conducive to internal site ventilation, cast shadows that reduce the mean radiation temperature, relieving the high outdoor thermal sensation at pedestrian height during the day . Combined with field measurements and ENVI-met software simulation, high-rise residential buildings are 4 °C lower than low-rise residential buildings and are thermally comfortable during most of the daytime, while low-rise buildings are subject to long-term thermal stress (Qaid et al. 2016). Moreover, urban low-density building area has more green and empty space, the air ventilation efficiency is high, the temperature is relatively low. In contrast, the temperature is relatively high in the urban highdensity building area (Yang et al. 2018a, b). Therefore, in high-density urban areas, reasonable changing the urban space can decreases air temperature 2.1 K on outdoor thermal comfort ). In addition, reducing building density and adding potential wind channels are very effective for the renewal of the high-density urban areas (Peng et al. 2015b). Therefore, in the urban renewal involving changes in building space, it is necessary to pay attention to the relationship between the building space renewal and the improvement of climate.
Urban renewal is increasingly taking place in cities all over the world, which affects the climate in cities (Wang and Shu 2020;Hestad et al. 2020). Different building layout patterns and vegetation arrangement affect the thermal comfort on pedestrian level (Hong and Lin 2015), which should be evaluated for urban sustainability. Therefore, in this study, the impact of the building layout patterns and vegetation arrangement on outdoor thermal environment were explored in two traditional districts in Xi'an, which is located in a cold climate zone in China. This study aims to providing hints for environmental urban redevelopment in traditional districts.

Methodology
Two traditional districts were selected for this study, and the thermal environmental conditions were evaluated by using ENVI-met simulation.

Two traditional districts
In Xi'an, the traditional districts are occupied by low-income citizens after the high-speed urbanization. In the traditional districts, building functions of residential, commercial, office, and historical heritages are combined together, creating the high-density low-rise building districts with diversity (Wang et al. 2018). As shown in Fig. 1, two typical traditional districts in the central area of the Xi'an City were selected according to the community characteristics. One of the districts is Sanxue District which is located close to the south gate of the old city wall, another one is Huimin District which is located in the center of the old town, and the size of each district is 500 m × 500 m.
Sanxue District is a provincial-level historical and cultural district, which is close to the city wall. There are many cultural relics and traditional dwellings. Due to historical reasons, its development has been stagnant for a long time, and the transformation is faced with the problem of neither protecting the current situation nor pushing down and rebuilding. As shown in Fig. 2 and Table 1, in Sanxue District, the historical preservation areas are located in the east and west side of the district, and the main commercial area is located in the south side of the district. In this district, a big proportion of the area is a residential area, which reaches up to 62.0% of the whole area. The quality of the residential buildings is relatively poor, and the basic service facilities are not complete. Meanwhile, most of the vegetation in this district is only planted along the main roads.
Huimin District, which has a history of thousands of years, is located in the center of the old urban area. There are cultural relics such as the Hanguang gate of the Tang Dynasty, the west gate complex of the Ming Dynasty, and many well-preserved mosques. Because of its national characteristics and large population density, the renewal is also facing some difficulties. It is balanced with a residential area (35.1%) and a commercial area (33.4%). The historic building protection areas are occupying 7.2% of the whole area, which is less than that in Sanxue District. In this district, the

ENVI-met simulation
ENVI-met is a simulation tool to recreate urban 3D models realized with deterministic equations that couple thermal and fluid-dynamics processes, which can stimulate the surface-plant-air interaction in an urban environment with a typical resolution of 0.5 to 10 m in space and 10 s in time (ENVI-met 2020). It is suitable to use with appropriate validation for most urban studies on the thermal environment (Ramyar et al. 2019;Taleghani et al. 2019;Battista et al. 2019). This is because that the relatively simple input scheme of ENVI-met allows most researchers to be able to run this software with minimal training.
The process of creating 3D models of the selected districts is shown in Table 2. Satellite images of the selected districts are obtained from the Baidu map. The computeraided design (CAD) models are constructed based on the satellite images. With importing the CAD models into ENVI-met interface, the 3D models are built using cubic grids of 125 m 3 (5 m × 5 m × 5 m), which contain building information, vegetation information, and ground surface information. For running the simulation, the initial input parameters used in the study are shown in Table 3, which are set as constants with the exception of atmosphere temperature parameter that was changed in order to perform the model validation process. Finally, the environmental results were exported after the simulation. ENVI-met simulation needs the initialization time to provide accurate calculation, the simulations were run for 27 h, and the results from the 4th hour after the start time were used for analysis (López-Cabeza et al. 2018).
In Fig. 3, the correlation between simulated results of the two districts were compared with that from the meteorological data of the day for data validation. By comparing the simulation results of air temperature of two districts with meteorological data, similar temperature change trends proved that the simulation results were reasonable for the analysis.

Urban renew plan of adding vegetation (plan V)
Open spaces such like parking lots and local squares on the sides of the main roads in Sanxue District and the main entrance of the Huimin District are the potential spaces for planting urban vegetation. The detailed plan of input model for simulation, and the image of open spaces are shown in Table 4. According to the field observation, some open spaces, accounting for 0.04% of the total area of each district, can be added with vegetation in two districts.

Urban renew plan of adding vegetation and building redevelopment (plan V&B)
Based on the design in plan V, plan V&B is considering the environmental effects from building redevelopment. According to the characters of the two traditional districts, some of the historical buildings are required to be protected, commercial buildings along the main streets are with great value for attracting people. Meanwhile, most of the residential buildings are self-built with poor construction quality, inadequate infrastructure, and poor maintenance condition. Therefore, the old residential building area is more suitable to renewal. In order to verify the effectiveness of plan V&B, simplified models were designed and are shown in Table 5. The size of the model is 150 × 150 m. For the original plan, the base area of each unit building in the model is 150 m 2 , all of the building units are three-story residential buildings     10 layers of the central building model is based on the "Regulations of Xi'an Historic City Protection": the overall building height within the city wall shall not exceed 36 m (< 36 m). According to the average floor height of 3.3 m, the overall building can be built up to 10 floors. ENVI-met was used for simulation under the same initial conditions. As shown in Fig. 4, compared to the original plan, the average wind speed in plan V&B is 0.16 m/s lower than that of the original; the average air temperature decreased by 0.14 °C, and the highest air temperature decreases from 31.85 to 31.41 °C at 16:00; the average mean radiant temperature decreased by 1.17 °C, and the highest value decreases from 69.17 to 63.29 °C. The results from simplified models demonstrated that the plan V&B is effective in improving the outdoor thermal environment.
Based on the actual situations, plan V&B was designed for the two districts in Xi'an. In Sanxue District, due to the building height near the historical area are limited, two renewal areas (50 m × 50 m for each area) at the edge of the district were selected for redevelopment, which are close to urban secondary roads. In Huimin District, two areas located in the residential area near the commercial street were selected for redevelopment.
In Table 6, after selecting the renewal areas, the residential buildings floor area of the demolishment is 3150 m 2 in Sanxue District and 3888 m 2 in Huimin District. Currently in Xi'an, the average living area per capita is 34.4 m 2 , which is based on no. 3 special edition report of Xi'an Daily on September 18, 2019. Therefore, the number of residents in the renewal areas can be roughly calculated as 92 people in Sanxue District and 113 people in Huimin District. Based on "Xi'an City Building Demolition Management Regulations" (Xi'an Municipal People's Government 2014), while maintaining the original living area, 10 m 2 should be added to each person after the redevelopment for refunding. The area multiplied by the estimated number of people is the Fig. 4 All-day variation of wind speed, air temperature, and mean radiant temperature of validated model total additional floor area required: 4070 m 2 in Sanxue District and 5018 m 2 in Huimin District. Due to the height limit of the building is 36 m in the old town of Xi'an (Xi'an Municipal People's Government 2002), 10-storey apartment building with 33-m height is proposed for redevelopment in two districts, with vegetation and parking spaces around the building. The ENVI-met input models are also shown in Table 6.

Comparative analysis parameters of thermal environment
This study selected the parameters-wind speed, air temperature, mean radiant temperature, and physiological equivalent temperature (PET)-related to the thermal environment to analyze the effect of the renew plans. Based on the combined effects of urban microclimate on pedestrians, thermal parameters typically include meteorological elements such as wind speed, air temperature, and mean radiation temperature (Liu et al. 2016;Santos Nouri et al. 2018;Sharmin et al. 2019).
In order to understand the thermal perception and improve the applicability of the thermal indices, PET is used as one of the analysis parameters based on the Munich energybalance model for individuals (Pantavou et al. 2018

Analysis and results
According to the simulation results of outdoor environment, the parameters of wind speed, air temperature, mean radiant temperature, and PET were used to evaluate the environmental effects from plan V and plan V&B in the two traditional districts.

The effects from plan V
As shown in Fig. 5, with the implementation of plan V, the average wind speed was increased for 0.02 m/s in Sanxue District and decreased for 0.14 m/s in Huimin District. For air temperature, the average value of plan V is 0.02 °C lower than that in the original plan in Sanxue District. Although the temperature increased by 0.1 °C at 24:00, the highest value decreased by 0.2 °C at 16:00. In Huimin District, the average air temperature of the whole day is 27.25 °C, which is the same in the original plan and the plan V. Moreover, the highest temperature at 16:00 and the temperature at 24:00 are the same as the original plan (O). The mean radiant temperature decreased by 0.48 °C, and the highest value decreased by 1.96 °C at 16:00 by implementing plan V in Sanxue District. In Huimin District, the average value of mean radiant temperature decreased by 0.29 °C, and the highest value decreased by 0.28 °C at 16:00. Figure 6 shows the environmental effects from the implementing plan V in the mid-day of a summer day. For wind speed, after plan V, the changes of the two districts were not obvious. The upper and lower quartiles changed slightly: in Sanxue District, the increase of upper and lower quartile was less than 0.1 m/s, compared to the original plan (O); in Huimin District, the upper quartile decreased by 0.01 m/s and the lower quartile was the same, compared to the original plan (O). For air temperature, in Sanxue District, the overall decrease of plan V: the upper quartile decreased by 0.32 ℃ and the lower quartile decreased by 0.12 ℃, compared to the original plan (O). In Huimin District, the overall change of plan V was not significant, with  Fig. 7, the wind speed, air temperature, and mean radiant temperature distribution at 14:00 are shown. After implementing plan V, the wind speed changed locally in the two districts. In Sanxue District, the area of wind speed from 1 s to 1.2 m/s increased in the north boundary, and the area below 1 m/s increased in some open spaces and adding vegetation spaces. In Huimin District, the area of wind speed below 1 m/s increased in adding vegetation spaces. For air temperature, after implementing plan V, in Sanxue District, the area below 30 °C increased in some open spaces and adding vegetation spaces. In Huimin District, although the overall temperature distribution has not changed much, the area of air temperature below 29.8 °C increased in adding vegetation spaces. For mean radiant temperature, the area below 50 °C increased in the adding vegetation spaces, compared to the original plan (O) of Sanxue District. In Huimin District, the area with values below 52.5 °C increased in the adding vegetation spaces. Therefore, plan V had an effect on the increase of average wind speed in Sanxue District, but the opposite effect on Huimin District. The air temperature decreased of Sanxue District was more than that of Huimin District after implementing plan V. For mean radiant temperature, the cooling effect of plan V for Sanxue District was also more obvious than that of Huimin District. Figure 8 shows all-day variation after further implementing plan V&B. In Sanxue District, the average wind speed of the whole day increased by 0.12 m/s, compared to the original plan (O). In Huimin District, the wind speed change was basically the same as the original plan between 8:00 and 12:00, while the average wind speed at other times decreased by 0.1 m/s. For air temperature, in Sanxue District, the average value decreased by 0.23 °C, and the highest value also decreased by 0.22 °C, compared to the original plan (O). Simultaneously, from 16:00, the temperature decreased more obviously than the original plan (O) by the hour. In Huimin District, the average value decreased by 0.04 °C between 0:00 and 8:00, compared to the original plan (O). After 8:00, there was little difference from that of the original plan (O). However, the highest air temperature decreased by 0.1 °C at 16:00. For mean radiant temperature, the average value decreased by 0.5 °C after implementing plan V&B, compared to the original plan (O) in Sanxue District. In Huimin District, compared with the original plan (O), the average value decreased by 1.15 °C.

The effects from plan V&B
Further analysis of the environmental effects in the midday of a summer day is shown in Fig. 9. After implementing plan V&B, in Sanxue District, the wind speed increased: the upper quartile increased by 0.18 m/s and the lower quartile increased by 0.07 m/s, compared to the original plan (O). In Huimin District, the wind speed decreased slightly: the upper and the lower quartile increased by 0.02 m/s on average. For air temperature, the overall temperature decreased in the two districts. In Sanxue District, the air temperature decreased more obvious than that in Huimin District. In Sanxue District, the upper quartile decreased by 0.11 °C and the lower quartile decreased by 0.25 °C, compared to the original plan (O). However, in Huimin District, the upper and lower quartile decreased by 0.05 °C on average. For mean radiant temperature, after implementing plan V&B, the overall temperature of the two blocks decreased, compared to their original plans (O). In Sanxue District, the upper quartile decreased by 1.64 °C and the lower quartile decreased by 2.77 °C. In Huimin District, the upper quartile decreased by 1.05 °C and the lower quartile decreased by 0.91 °C. In Fig. 10, the wind speed, air temperature, and mean radiant temperature distribution at 14:0 are shown. Compared with the original plan (O), the changes were mainly distributed in the building renewal space and the open space of the district. For wind speed, in Sanxue District, after implementing plan V&B, the area above 1 m/s increased in the building renewal spaces and some open spaces. In Huimin District, the area above 0.6 m/s increased in the building renewal spaces. For air temperature, the area below 30 °C increased in the building renewal spaces and some open spaces. However, the area above 30.2 °C decreased in the center space of the district. In Huimin District, the area below 30 °C increased in the building renewal spaces. For mean radiant temperature, the area below 50 °C increased in the building renewal spaces, and the area above 67.5 °C decreased in other spaces of the district obviously. In Huimin District, the area below 50 °C also increased in the building renewal spaces.
Therefore, after further implementing plan V&B, the wind speed of Sanxue District increased, while the wind speed of Huimin District decreased slightly. In terms of air temperature, compared to Huimin District, the cooling effect of plan V&B on Sanxue District is more obvious. For mean radiant temperature, although between 8:00 and 16:00, the value of two districts decreased after the implementation of plan V&B. However, from the comparison of the data at 14:00, the cooling effect in Sanxue District is more obvious.

Comparison of effects of two renew plans on two districts
In Fig. 11, for the wind speed change compared to the original plan (O), two plans have the opposite effect in two districts. For Sanxue District, after implementing plan V, the average wind speed increased by 0.02 m/s, compared to the original plan (O). After implementing plan V&B, the average wind speed in Sanxue District increased by 0.11 m/s. Therefore, plan V&B increased wind speed more than plan V. In Huimin District, the average wind speed decreased by 0.01 m/s after implementing plan V, and decreased less than 0.01 m/s after implementing plan V&B. Therefore, two plans have caused the wind speed to decrease in Huimin District. In addition, at the time of the implementation of Fig. 10 Wind speed, air temperature, and mean radiant temperature distribution of the original plan (O) and plan V&B in two districts at 1.5-m height in 14:00 of a summer day (22 July, 2016) plan V, the adding vegetation area in Sanxue District was concentrated on both sides of the road, while the area in Huimin District was close to the downwind and scattered. Further implementing plan V&B, the building renew areas in Sanxue District were located in the upwind direction of the district to increase the air volume, while the areas were located inside in Huimin District. Therefore, it shows that the renewal area of plan V&B is located in the upwind direction of the district, which are more conducive to the air circulation in the district.
In Fig. 12, the slope of the linear equation fitted by the data show the rates at which the air temperature in two districts decrease from the highest value with the two plans. In Sanxue District, after implementing two plans, the slopes of the decreasing value are the same, which shows that the two plans have the same rate of cooling in Sanxue District. In Huimin District, the absolute value of the slope of the decreasing value of plan V&B is 0.07 more than that of plan V. This shows that the implementation of plan V&B will make the air temperature decrease is faster than the implementation of plan V in Huimin District. Therefore, there is no difference between the two plans for the air temperature decrease in Huimin District, but for Sanxue District, the plan V&B is better. In addition, after implementing plan V, the slope of the decreasing value of the two districts differs only 0.01. This shows that plan V has the similar cooling effect on two districts. However, after further implementing plan V&B, the absolute value of the slope of Sanxue District is 0.06 more than that of Huimin District. It shows that plan V&B is better for the air temperature decrease in Sanxue District.
For mean radiant temperature, in Sanxue District, the absolute value of the slope of the decreasing value of plan V&B is 0.15 more than that of plan V. This shows that the implementation of plan V&B make the mean radiant temperature decreased faster than that of plan V in Sanxue District. In Huimin District, the absolute value of the slope of the decreasing value of plan V&B is 0.08 less than that of plan V. This shows that the implementation of plan V&B make the mean radiant temperature decrease slower than that of plan V in Huimin District. Therefore, plan V is better for the mean radiant temperature decrease in Huimin District than plan V&B, but for Sanxue District, plan V&B is better. In addition, after implementing plan V, the absolute slope of the decreasing value of Sanxue Districts is 0.91 lower than that of Huimin District. After further implementing plan V&B, the absolute value of the slope of Sanxue District is 1.12 higher than that of Huimin District. It shows that the two plans is better for the mean radiant temperature decrease in Sanxue District.
Moreover, looking at the simulation results of physiological equivalent temperature (PET) in Fig. 13, in Sanxue District, the average value of plan V is 1.03 °C lower between 9:00 and 17:00 and is 0.13 °C lower at other times of the day, compared to the original plan (O). After further implementing plan V&B, there is no significant change of PET during the daytime. The PET became lower than both of the original plan (O) and plan V after 20:00, averagely decreased by 0.54 °C between 20:00 and 24:00. From the results of the Huimin District, after implementing plan V, compared to the original plan (O), PET is decreased between 10:00 and 16:00, and the average value of PET in this period decreased by 0.4 °C. During the time before 10:00 and after 16:00, the PET value is slightly lower than that of the original model, which is averagely decreased by 0.02 °C. If change plan V&B, from 10:00 to 16:00, the average PET value become 0.75 °C lower than that in the original plan (O), and 0.35 °C lower than that in plan V. This means that for Huimin District, plan V&B is more effective than plan V.
Comparing the PET in the two districts, by implementing plan V, the average PET decreased by 0.12 °C in Huimin District, and decreased by 0.35 °C in Sanxue District. By   Fig. 11 Wind speed change of plan V and plan V&B in two districts

Fig. 12
Air temperature and mean radiant temperature of plan V and plan V&B from 16:00 to 24:00 in two districts providing plan V&B in the districts, the PET of the whole day became averagely 0.24 °C lower in Huimin District, and 0.67 °C lower in the Sanxue District compared to that in original plan (O). This is to say, adding more vegetation could contribute to the human thermal comfort, especially in the daytime, and providing plan V&B to make more open spaces in the district could further promote the thermal comfort. Compared to the PET results in Huimin District, the environmental urban modification is more effective in Sanxue District.

Discussion
In this study, through comparative analysis by using ENVImet simulation, two renew plans-plan V and plan V&B-have positive effects on thermal environment of the selected two traditional districts. In Sanxue District, plan V&B is more pronounced for the increase in wind speed, the decrease in air temperature, and the decrease in mean radiant temperature, compared to plan V. In Huimin District, the two plans have a lowering effect on the wind speed, and plan V makes the wind speed decrease more obvious. For air temperature, the effect of the two plans on the cooling rate is the same, but for the temperature decrease, the plan V&B is more obvious. For mean radiant temperature, although the overall value decreased, the value of plan V decreased faster than that of plan V&B. In general, compared to plan V, plan V&B is more effective in improving the thermal environment in traditional districts, especially in Sanxue District. However, similar renewal methods will also show different effects in different countries and cities. In a historical district of Lodz, Poland, it shows that the introduction of 10% of green area contributed to a maximum air temperature reduction in the typical street canyon (Bochenek and Klemm 2021). In the Priolo neighborhood of Avola city, green park can decrease the local temperature by over 3 ℃ during urban renewal (Evola et al. 2017). In addition, in the historical district of Lar City, the lack of green coverage and public space in urban renewal is the main reason why residents are dissatisfied with the traditional neighborhood environment (Biqaraz et al. 2019). Therefore, the renewal methods that are appropriate to the local climate, social environment, and policies can effectively improve the urban thermal environment and residents' thermal comfort. Although the renew plans are beneficial to the environmental reshaping of traditional districts, there are still some actual limits in the implementation process.
Firstly, urban renewal planning for large-scale demolition of traditional districts with gradual commercial value often neglects the protection of traditional culture. Secondly, in the process of urban renewal demolition, the interests of the original residents may be difficult to be guaranteed because of the neglect of the developers. Thirdly, although the renewal of traditional districts can improve the living environment, it needs to provide more public open spaces, which also affects the private interest in land use. In order to solve these limitations, some methods have been proposed. Firstly, the small-scale demolition, such as the plan V&B in the Sanxue District and Huimin District, can be used to reduce the problems caused by large-scale relocation. In "Xi'an's Urban Renewal Measures" (2022), the urban renewal should adopt the method of "retaining, replacing, and dismantling" and focusing on retention, utilization, and improvement. It not only optimizes urban functions and improves the living environment of urban residents but also raises the threshold of demolition and reconstruction areas, limits the profitability of urban renewal projects, and prevents urban renewal from excessively using the real estate development model. Simultaneously, it can be seen from the analysis results of the two districts that small-scale renovation has a positive effect on improving the thermal environment in this study. In addition, in the traditional districts, the appropriate areas should be selected for the renewal plan, which will not affect the protection of the historical and cultural areas. Secondly, in the process of demolition, the compensation area for demolition should be calculated in strict accordance with government policies. It is also in line with the "Xi'an's Urban Renewal Measures" (2022), the urban renewal work should follow the principle of government leadership. As in the plan V&B of this study, the total floor area of new buildings in renew areas is calculated and reconstructed in accordance with relevant policy regulations. Thirdly, the renew plan needs to increase the public open spaces and optimize the environment of the district while ensuring the interests of the original residents. As the plan V in this study, the adding vegetation spaces can make use of existing idle spaces, or in the plan V&B, plan parking lots and public green spaces around the new building. Finally, before implementing the renew plan, it is necessary to simulate and evaluate the environment before and after the renewal of the district to ensure the scientific rationality of the plan.
In the future, in-depth research will be carried out in two directions. For the plan V, greening with different proportions will be added at different locations in the districts to get a more detailed effect on thermal environment. For the plan V&B, based on the improvement of thermal environment, local building rebuilt and greening layout will be further refined to provide a more detailed layout scheme for urban renewal.

Conclusion
In this study, the thermal environment of the renewal methods for traditional districts in Xi'an were evaluated. Adding vegetation (plan V) can improve the thermal environment of the districts, but the potential for adding vegetation is limited in most of the traditional districts. After adding vegetation according to the field observation, in Sanxue District, the wind speed increased by 0.02 m/s, the air temperature decreased by 0.02 °C, and the mean radiant temperature decreased by 0.48 °C; in Huimin District, the wind speed decreased by 0.14 m/s, the air temperature did not change, and the mean radiant decreased by 0.29 °C. In addition, the PET of Sanxue District decreased by 1.03 °C and decreased by 0.02 °C in Huimin District. Therefore, the effect of adding vegetation in the traditional district is limited, and the contribution is depended on the amount of vegetation added.
Based on plan V and the related regulation for urban renewal, the building floor area for renewal plan in the specific district were calculated and analyzed, and the plan of adding vegetation and building renewal (plan V&B) is proposed and investigated. In Sanxue District, the average wind speed increased by 0.12 m/s, the average air temperature decreased 0.23 °C, the mean radiant temperature decreased by 0.5 °C, and the average PET decreased by 0.24 °C; in Huimin District, the average wind speed decreased by 0.1 m/s, the average air temperature decreased by 0.03 °C, the average mean radiant temperature decreased by 1.15 °C, and the average PET decreased by 0.67 °C. Building renewal in Sanxue District is providing more environmental optimization compare to that in the Huimin District, because the upwind location renewal is more effective to promote the local ventilation and thermal environment.
The two renewal plans for the traditional districts are compared and demonstrated to improve the outdoor thermal comfort. The method of the combination of adding vegetation and building renewal provides possibilities to improve the thermal environment, and the partial urban renewal in traditional districts could also contributes to the community environmental promotion for sustainable urban development.
Author contribution Dixuan Ma carried out field measurements and ran simulations, analyzed the results, and wrote the manuscript; Dian Zhou and Yupeng Wang designed the experiments and guided the writing; Zongzhou Zhu participated in field measurements and revised the manuscript. Data availability The data and materials will be sent based on request.

Declarations
Ethical approval We followed the accepted principles of ethical and professional conduct.

Consent to participate
All the authors agreed to participate in the paper.
Consent for publication All the authors have read and approved the manuscript for publication.

Competing interests
The authors declare no competing interests.