Differences in Outdoor Thermal Comfort between Local and Non-Local Tourists in Winter in Tourist Attractions in a City in a Severely Cold Region

Abstract

The unique climate and the landscape of severely cold regions in winter attract many tourists. The outdoor thermal environment affects the space use and the tourist experience, becoming one of the key factors in the design of tourist attractions. The outdoor thermal comfort of tourists from different regions should be considered, but it has been poorly studied in winter in severely cold regions. This paper explores the differences in outdoor thermal comfort in winter between local and non-local tourists through the field measurement of the thermal environment and a questionnaire survey of thermal comfort at tourist attractions in Harbin, China. The results show that the proportion of local tourists who expect the air temperature and solar radiation to rise in winter is higher than that of non-local tourists. The thermal sensation vote of local tourists is generally higher than that of non-local tourists. When the Physiologically Equivalent Temperature (PET) < −6 °C, the thermal satisfaction of non-local tourists is higher than that of local tourists. When the PET value is −10 °C, the thermal comfort of non-local tourists is the highest. The thermal comfort decreases with the rise or fall of the PET value. When −28 °C < PET < −7 °C, the thermal comfort of non-local tourists is generally higher than that of local tourists. This paper provides a reference and evaluation basis for urban tourist attractions’ outdoor thermal environment design in severely cold regions.

1. Introduction

As one of the fastest growing, most diversified and largest industries in the world [1], tourism is the main source of income and employment opportunities for many countries in the world, promoting the development of local social productivity [2,3]. Climatic conditions are one of the main natural factors affecting tourism [4,5,6] The majority of outdoor tourism programs depend on pleasant and attractive climates [7]. First of all, regional climate characteristics have a certain impact on the choice of tourist destinations. Compared with the local climate, people prefer to experience different climate characteristics of other regions [8]. Secondly, the local climate conditions will affect the travel arrangements of tourists. A good climate environment can increase the number of tourists, while bad weather conditions will also reduce the corresponding travel demand [9]. In addition, tourist attractions of the city are an important part of the urban environment and local brand [10,11]. The thermal comfort evaluation of tourist attractions can improve the development potential of tourist attractions [12] and create comfortable outdoor environments [13].

In recent years, outdoor thermal comfort studies on tourist experiences have covered different regional and seasonal climates, as well as different attraction types. In severely cold regions, Jin et al. determined that the preferred temperatures for the transition and hot seasons that made people feel comfortable in urban pedestrian streets were 30.4 °C and 26.8 °C Universal Thermal Climate Index (UTCI), respectively. They also analyzed the effects of thermal adaptation and demographic factors on thermal comfort [14]. Yan et al. studied the influence of street landscape elements and their induced changes of emotion on thermal comfort. In terms of pleasure and arousal, combining trees and waterscape improved thermal comfort scores by 2.09 and 1.34, respectively [15]. In cold regions, Liu et al. compared the outdoor thermal comfort of urban river landscape belts in different seasons, proving the importance of thermal experience and psychological adaptation. In this study, “slightly warm” in spring and winter was considered the most comfortable thermal sensation, while “neutral to cool” was regarded as the most comfortable in summer [16]. Xu et al. showed that solar radiation and landscape components are important factors affecting thermal comfort in urban parks in winter. The neutral PET and UTCI of local residents in Xi’an ranged between 13.3 and 23.6 °C and 14.9 and 23.2 °C respectively [17]. In hot summer and cold winter regions, Xiong et al. revealed the effect of the mechanism of microclimate and human activities on wintertime thermal sensation and comfort in cold humid environments [18]. Xu et al. established a quantitative evaluation model of outdoor dynamic thermal comfort [19]. In hot summer and warm winter regions, the researchers investigated the local variation of outdoor thermal comfort in different urban green spaces [20] and the effect of short-term thermal history on outdoor thermal comfort [21,22]. These studies reveal the correlation between the subjective thermal state and objective thermal environment.

The influence of thermal adaptation on thermal comfort caused by behavioral adjustment, physiological acclimatization, and psychological habituation or preference [23,24] has been paid continuous attention by scholars. Knez and Thorsson studied the differences in heat sensation caused by different cultural backgrounds and attitudes towards the environment. The results showed that in two parks from Japan and Sweden when the air temperature was between 18 and 23 °C, the Swedes were more sensitive to the thermal environment than the Japanese [25,26]. Le et al. found that people’s long-term thermal history affected their outdoor thermal comfort. In winter, when thermal comfort voting was conducted in the cold regions of Xi’an, Beijing, and Hami, the proportion of “neutral” was positively correlated with latitude [27]. De Dear and Brager used the concept of preference in the field of thermal comfort research as a psychological adaptation method to explain why people can accept a wider range of thermal conditions in naturally ventilated built environments [28]. Yang et al. studied the effects of climate differences and dress preferences on thermal preferences. The results showed that Changsha respondents had a more urgent preference to be cooler in hot summer than Singapore respondents in urban outdoor spaces with similar thermal environments [29]. Michelle Rutty et al. found that even at a UTCI of 39 °C, more than 60% of respondents expected the thermal conditions to remain the same, with an additional 10% desiring to feel warmer on Caribbean beaches. The thermal preferences of beach users were up to 18 °C warmer than those of urban park crowds [30]. These studies show that people’s different thermal histories and thermal preferences lead to differences in thermal comfort in urban outdoor spaces with similar thermal environments.

Non-local tourists who visit tourist attractions have different thermal environment experiences and seasonal climate preferences from local tourists. Therefore, it is necessary to deeply explore the different thermal comfort states of non-local tourists in the same thermal environment as local tourists. In recent years, a series of studies on the difference in thermal comfort between local and non-local tourists have been carried out. Lu analyzed the thermal comfort of tourists in semi-open hotel lobbies on a tropical island in Hainan, China, pointing out that non-local tourists were more sensitive to changes in air temperature and wind speed than local tourists when wind speed compensation was considered [31]. Lam et al. evaluated the thermal comfort of local and overseas tourists during extreme heat events at the Melbourne Royal Botanic Garden in summer [32]. They proposed that when the thermal sensation of the environment was “hot”, the preference of European tourists to maintain the same temperature (36.8%) was much higher than that of Australian tourists (12.2%) and Chinese tourists (7.5%). Based on a comparison of the differences in thermal comfort caused by different thermal histories, tourists’ travel purposes and preferences for experiencing different climates also have a strong correlation with the data analysis results. Xi et al. pointed out that residents and tourists had different thermal adaptation and thermal regulation mechanisms. The tourists dressed in more clothing in winter than the residents. Their activity levels were not influenced by the outdoor thermal environment, while the residents preferred to do more activities in the late morning and afternoon to avoid the very cold early morning [33].

Yang et al. pointed out that compared with non-local people, residents in Umeå still had a higher preference for the enhancement of solar radiation even under a “slightly warm” thermal sensation because they exposed themselves to a long-term low-temperature climate in winter [34]. Lopes et al. showed that the voluntary nature of the tourism experience made tourists more tolerant of various climatic conditions in Porto. They were more willing to evaluate the thermal environment as comfortable than the locals [35]. From a physiological perspective, tourists’ thermal regulation mechanism can adapt to an unfamiliar climate in a short time [36]. From a psychological perspective, the climatic optimum of tourists tends to expand more because tourists are willing to have greater climatic variability than residents engaged in common daily activities [37,38].

From the perspective of research regions, most of the current studies on the thermal comfort of tourists in tourist attractions are conducted in a summer thermal environment in cold regions [39], hot humid regions [22,40], and hot dry regions [41]. There are relatively few studies conducted in winter in severely cold regions with more serious thermal comfort problems and a low utilization rate of the outdoor space. From the perspective of research methods, previous studies mainly compared the single thermal comfort of tourists from different places, lacking a comprehensive analysis of the influence of the outdoor thermal environment on tourists’ thermal comfort from multiple aspects. The purpose of this paper is to study the difference in outdoor winter thermal comfort between local and non-local tourists in urban tourist attractions in a severely cold region. Through thermal environment measurements and a questionnaire survey, the local and non-local tourists’ subjective evaluations of the thermal environment and the differences between them were analyzed from the dimensions of thermal preference, thermal sensation, thermal satisfaction, and thermal comfort. This paper provides a reference and evaluation basis for the outdoor thermal environment design of urban tourist attractions in severely cold regions.

2. Methods

2.1. Study Location and Climatic Conditions

As Figure 1 shows, the whole process of the method is studied in this study, which is divided into five main steps. After the research plan was determined, data were collected through a subjective questionnaire survey and thermal environment parameter measurement, the thermal comfort indexes were selected for calculation, and finally, data were analyzed and processed.

Harbin, a typical city in a severely cold region, was chosen as the research site [42]. Harbin (45°41′ N 126°37′ E) is the capital city of Heilongjiang Province and the political, economic, and cultural center of the region, located on the Northeast China Plain. By the end of 2022, the permanent population of Harbin reached 9.395 million [43]. With its unique geographical climate and border culture characteristics, Harbin has become one of the most distinctive tourist cities in China, known as “Ice City” and “Eastern Moscow” [44]. In winter, Harbin’s tourism is particularly prosperous to attract a large number of tourists, most of whom come from outside the city [45,46]. The Code for thermal design of civil building (GB50176-2016) divides China into five climate regions: cold, hot in summer and cold in winter, hot in summer and warm in winter, and temperate regions [42,42,47]. Figure 2 displays the geographical location of Harbin and the thermal zone of China.

Table 1. China’s thermal technical division standard.

Region	Main Standard
Severely cold region	The average temperature of the coldest month: ≤−10 °C
Cold region	The average temperature of the coldest month: −10~0 °C
Hot summer and cold winter region	The average temperature of coldest and hottest months: −10~0 °C, 25~30 °C
Hot summer and warm winter region	The average temperature of coldest and hottest months: >10 °C, 25~29 °C
Temperate region	The average temperature of coldest and hottest months: −13~0 °C, 18~25 °C

shows China’s thermal technical division and the location of Harbin.

Harbin has four distinct seasons and a mid-temperate continental monsoon climate with long winters and short summers throughout the year, and meteorological data released by the National Meteorological Center (from 2005 to 2019) showed that the average monthly air temperature in Harbin is −1724 °C [48]. The average monthly relative humidity is 42~75%. The average annual global horizontal irradiance is 5078–5381 MJ/m², and the average annual direct horizontal irradiance is 36644474 MJ/m². Winter tourism peaks in Harbin in December and January. Figure 3 and Figure 4 reveal that the average air temperature in December and January is −15 °C and −17 °C. The average relative humidity is 68.8% and 68.0%. The average global horizontal irradiance is 603 MJ/m² and 746 MJ/m², and the average direct horizontal irradiance is 376 MJ/m² and 449 MJ/m², respectively.

This study follows two principles when selecting the survey sites. Firstly, the spatial type of the tourist attraction can represent the layout pattern of Harbin’s urban tourist attractions [49]. Secondly, the tourist attractions can attract a large number of tourists to carry out sightseeing activities to obtain sufficient and representative subjective evaluation data.

In this paper, three typical tourist attractions in the urban center of Harbin were selected as the survey sites. The measuring point arrangement and site environment are shown in Figure 5. The No. 1 measuring point was located in the center of the Flood Control Monument Square at the intersection of Harbin Central Avenue and the Songhua River riverside sightseeing belt. The Flood Control Monument won the highest honor award of the Chinese construction industry in 2009. The No. 2 measuring point was located in the leisure plaza in front of the Modern Cold Drink Restaurant on Central Avenue. Central Avenue was built in 1900, known as the “First Street in Asia”. The cold winter does not inhibit the enthusiasm of tourists to buy local special cold drinks. Therefore, the daily number of tourists in front of the Modern Cold Drink Restaurant is high. The No. 3 measuring point was located in the Cultural Landscape Square in Stalin Park. The square is covered with snow in winter, with many kinds of experiential activities such as ice skating and riverside sledding. Many tourists choose to enjoy and entertain themselves there.

The research time of this study was December and January from 2017 to 2019, which is the busy winter tourist season of Harbin. The specific research time of day was from 8:30 to 17:00. The reasons for choosing the research time are as follows. Firstly, this period was in the peak period of winter tourism. A large number of tourists are attracted to Harbin for sightseeing during this period. Secondly, the experiment under a wide range of outdoor thermal environment parameters can reflect the thermal comfort of tourists in different thermal environments. Thirdly, the weather conditions during the field investigation were good, representing the climate characteristics of the daytime period in winter in the severely cold region.

2.2. Thermal Comfort Questionnaire

The thermal comfort of tourists in urban tourist attractions is affected by various subjective judgment factors [50]. For example, in the outdoor survey, tourists often give “It is cold now”, “The weather is so comfortable today”, “I am so satisfied with the climate environment of the tourist attraction”, and so on as comments. People are used to evaluating the thermal comfort of tourist attractions from many subjective angles. In this study, according to the conclusions of previous studies, several subjective indicators with high frequency were selected as the basis for a comprehensive evaluation of thermal environment comfort. They were thermal preference, thermal sensation, thermal satisfaction, and thermal comfort.

Thermal preference can reflect people’s willingness to adjust according to the environment [51]. The thermal sensation is the cold or hot sensation of the human body in the thermal environment, which is the most direct feedback of the receptor when it receives the influence of the outdoor thermal environment. Thermal satisfaction is a comprehensive psychological state of tourists, that is, their attitude towards the tourism thermal environment [52]. Thermal comfort is the condition of mind that expresses satisfaction with the thermal environment [53].

This study adopted the random sampling method, selecting tourists to conduct subjective thermal comfort questionnaire surveys. A total of 692 valid questionnaires were obtained for this survey, of which 297 were from local tourists and 395 were from non-local tourists. The questionnaire included three parts: personal basic information, thermal preference for four environmental parameters, and thermal comfort state details.

Table 2. Questionnaire used in this study.

Date: _________

Time: _________

Tourist attraction: __________

Gender: ☐Male ☐Female

Residence city: _____________

Age: ☐<10 ☐11~20 ☐21~30 ☐31~40 ☐41~50 ☐51~60 ☐61~70 ☐>70

Your preference for thermal environmental parameters:

Air temperature

Increase

No change

Decrease

☐ 1

☐ 0

☐ −1

Relative humidity

Increase

No change

Decrease

☐ 1

☐ 0

☐ −1

Wind speed

Increase

No change

Decrease

☐ 1

☐ 0

☐ −1

Solar radiation

Increase

No change

Decrease

☐ 1

☐ 0

☐ −1

How do you feel at this moment:

Extremely cold

Very cold

Cold

Cool

Slightly cool

Neutral

Slightly warm

Warm

Hot

Very hot

Extremely hot

☐ −5

☐ −4

☐ −3

☐ −2

☐ −1

☐ 0

☐ 1

☐ 2

☐ 3

☐ 4

☐ 5

Your satisfaction degree of the thermal environment:

Very dissatisfied

Dissatisfied

Slightly dissatisfied

Satisfied

Very satisfied

☐ −3

☐ −2

☐ −1

☐ 0

☐ 1

Your thermal comfort level at this moment:

Very uncomfortable

Uncomfortable

Slightly uncomfortable

Comfortable

☐ −3

☐ −2

☐ −1

☐ 0

lists the information from the questionnaire.

The thermal comfort state details include the thermal preference vote, thermal sensation vote (TSV), thermal satisfaction vote (TS_aV), and thermal comfort vote (TCV). The questionnaire counted the tourists’ preferences for four thermal environmental parameters: air temperature, relative humidity, wind speed, and solar radiation. Generally, in the outdoor comfort studies, no standard TSV scale was put forward, and different TSV scales were used according to the authors’ experience and local climate characteristics [54]. To adapt to the winter climate in severely cold regions, the TSV in this study was extended based on the ASHRAE standard seven-point thermal sensation scale. This scale was selected because it is more objective than the PET nine-point scale [55]. However, Harbin is located in a severely cold region and has extremely cold weather with the lowest air temperature at −37.7 °C as recorded; thus, the widely used seven-point TSV scale may be limited in capturing the cold thermal sensation characteristics [54]. Therefore, the four points “extremely hot”, “very hot”, “very cold”, and “extremely cold” were added so the TSV was rated on an eleven-point scale [56]. Referring to the ISO 10551 (2015) standard for assessing the impact of the thermal environment [57], and in combination with previous subjective surveys related to thermal comfort in this climatic region [14,33,58], this study set the thermal satisfaction scale to five points and the thermal comfort scale to four points to make the research results more targeted. In the thermal satisfaction scale, the points from “−3” to “1” were defined as “very dissatisfied”, “dissatisfied”, “slightly dissatisfied”, “satisfied”, and “very satisfied” [59]. In the thermal comfort scale, the points from “−3” to “0” were defined as “very uncomfortable”, “uncomfortable”, “slightly uncomfortable”, and “comfortable” [60].

2.3. Thermal Environment Measurement

In this study, air temperature, relative humidity, wind speed, and global temperature were measured during the questionnaire survey.

Table 3. Meteorological properties of the measuring instruments used.

Model	Parameter	Range	Accuracy
BES-02B	Air temperature	−30~50 °C	±0.01 °C
BES-02B	Relative humidity	0%~100%	±0.1% RH
Kestrel5500	Wind speed	0.4~40 m/s	±0.1 m/s

shows the models and technical parameters of the thermal environment measuring instrument, which met the requirements of the ISO 7726 standard [61]. The measuring instrument was supported by a tripod, fixed at a height of 1.2 m above the ground [39]. A small handheld weather station recorded every 15 s to calculate the average wind speed data per minute. The air temperature and relative humidity sensors were placed in a highly reflective aluminum-coated box that resisted solar radiation and ensured smooth self-ventilation on both sides. In addition, air temperature, relative humidity, and global temperature were all collected at 1 min intervals [62].

2.4. Thermal Comfort Evaluation Indices

The mean radiant temperature ($T_{mrt}$) is one of the important environmental parameters to evaluate outdoor thermal comfort. The radiant heat exchange between the shell surface and the human body is equal [63]. The indirect measurement method of global temperature based on the ISO7726 standard is one of the commonly used methods to calculate the $T_{mrt}$ [61]. In this paper, a black sphere thermometer was used to measure the global temperature and calculate it indirectly [64]. The black sphere was suspended 1.2 m above the ground to achieve thermal balance with the surrounding environment. The formula for calculating $T_{mrt}$ is

$$T_{mrt}={\left[{\left(T_g+273.15\right)}^4+\frac{1.10\times {10}^8{V_a}^{0.6}}{\epsilon D^{0.4}}\times \left(T_g-T_a\right)\right]}^{0.25}-273.15$$

(1)

where $T_{mrt}$ is mean radiant temperature (°C), $T_g$ is global temperature (°C), $V_a$ is wind speed ($\mathrm{m}/\mathrm{s}$), $T_a$ is air temperature (°C), D is the globe diameter (D = 0.07 m), and $\epsilon$ is the emissivity of the black sphere ($\epsilon$ = 0.95).

As a basis to characterize the comfortable state of the outdoor thermal environment, thermal comfort evaluation indices have been widely used in different regions. They integrate many factors such as thermal environment parameters, activity amount, and the thermal resistance of clothing so that the comprehensive results can be classified into a single variable [65,66,67].

At present, the commonly used outdoor thermal comfort evaluation indices mainly include the Predicted Mean Vote (PMV) [66], Physiologically Equivalent Temperature (PET) [68,69], Universal Thermal Climate Index (UTCI) [70], and Standard Effective Temperature (SET*) [71,72,73]. The above general indices are based on the human heat balance equation, comprehensive consideration of thermal environment elements, and individual parameters, which can be easily calculated by software. SET* is mainly applicable to warm and hot climate regions [71,74,75]. Although PMV can be used in all climate regions, it is mainly suitable for indoor thermal comfort evaluation [66]. In addition, previous studies have found that UTCI indicators may have invalid intervals in the low-temperature places of cold and severely cold regions [76]. Therefore, based on the application scope and considerations of thermal comfort evaluation indices, this paper selected PET to study the outdoor thermal comfort of local and non-local tourists.

PET is a thermal comfort evaluation index derived from the MEMI (Munich Energy Balance Model for Individuals) model [68,77]. It comprehensively and deeply considers the main climatic factors affecting human thermal comfort, including air temperature, relative humidity, wind speed, and solar radiation. It also fully considers the characteristics of human physiology, basically covering all basic thermoregulatory processes [78]. In addition, its accuracy in predicting outdoor thermal comfort has been confirmed in a large number of studies [78,79,80,81].

In this study, PET was calculated by the thermal comfort index calculation software RayMan [82,83]. The physiological parameters were assumed to be a male with a height of 180 cm, weight of 75 kg, clothing thermal resistance of 0.9 clo, and a metabolic rate of 80 W [84]. The values of the relevant thermal environment parameters ($T_a$, RH, $V_a$, $T_g$, $T_{mrt}$) were input into the software to calculate PET.

2.5. Statistic Analysis of Data

This study employed SPSS software to conduct correlation analysis and regression analysis. Pearson correlation analyses were performed between TSV, PET, and thermal preferences for environmental parameters, and the degree of linear correlation was determined. In addition, values of TSV, values of TCV, and values of TS_aV corresponding to each 1 °C PET interval were averaged. Regression analysis was performed to examine the relationship between these averaged values and their corresponding PET values. The curve models with the highest goodness of fit were selected to construct the prediction models of TSV, TS_aV, and TCV based on PET.

3. Results

3.1. Thermal Preference Vote

The outdoor thermal comfort of individuals is significantly influenced by their psychological state at the time of assessment. Even under the same thermal conditions, subjective evaluations of the thermal environment may vary among different groups based on their specific psychological adaptability [85,86]. Nikolopoulou’s study demonstrated that thermal preferences play a crucial role in shaping people’s thermal comfort experience [50]. In this study, a comparison was made between local and non-local tourists regarding their thermal preferences for environmental parameters under different TSV points and PET conditions.

3.1.1. Differentiation of Thermal Preference between Local and Non-Local Tourists under Different TSV Points

This study employed a partial correlation analysis method to analyze the impact of thermal sensation variations on the local and non-local tourists’ thermal preference for thermal environmental parameters. The TSV was used as the independent variable. Dependent variables were people’s preference voting for air temperature, humidity, wind speed, and solar radiation, which are four thermal environmental parameters. The significance value (Sig.) and the correlation coefficient (Cor.) were used to assess the significance and degree of correlation between TSV and preferences for each thermal environmental parameter.

Table 4. Correlations between thermal preference and TSV.

	Air Temperature Preference	Relative Humidity Preference	Wind Speed Preference	Solar Radiation Preference
TSV for local tourists	−0.152 **	−0.098	0.193 **	−0.239 **
TSV for non-local tourists	−0.241 **	−0.057	0.099 *	−0.193 **

** Correlation is significant at the 0.01 level. * Correlation is significant at the 0.05 level.

presents the results of the correlation analysis between TSV and preferences for thermal environmental parameters for local and non-local tourists. TSV for both local and non-local tourists showed a significant negative correlation with air temperature and solar radiation preferences as well as a significant positive correlation with wind speed preference. However, there was no significant correlation between TSV and relative humidity preference. Additionally, the correlations between local tourists’ TSV and preferences for wind speed and solar radiation were higher than those of non-local tourists, while the correlation degree of temperature preference was slightly lower than that of non-local tourists.

Figure 6 displays the air temperature preference voting distribution among local and non-local tourists under different thermal sensations. Under “extremely cold” conditions (TSV = −5), more than three-quarters of local tourists expressed a desire for an air temperature increase, while approximately 22% preferred it to remain the same. In contrast, only 56% of non-local tourists desired the air temperature to rise, while 26% expected it to stay the same and 17% wanted it to decrease. As the value of TSV increased from −5 to 0, the proportion of both local and non-local tourists desiring higher air temperature gradually decreased. Conversely, the proportion of non-local tourists desiring a lower air temperature exhibited an upward trend and consistently remained higher than that of local tourists. In the neutral thermal sensation (TSV = 0), the proportion of non-local tourists desiring a lower air temperature was significantly higher than that of local tourists, at approximately 25%.

Figure 7 illustrates the wind speed preference voting distribution among local and non-local tourists under different thermal sensations. Under “extremely cold” conditions (TSV = −5), approximately 65% of local tourists preferred a decrease in wind speed, compared to only 50% of non-local tourists. As the value of TSV increased, the proportion of tourists desiring a decrease in wind speed generally showed a downward trend, with local tourists’ thermal preference exhibiting a more pronounced decrease. When the thermal sensation was “neutral” (TSV = 0), approximately 40% of non-local tourists preferred a decrease in wind speed, while the proportion among local tourists was 30%.

Figure 8 represents the solar radiation preference voting distribution among local and non-local tourists under different thermal sensations. Non-local tourists exhibited a relatively lower preference for increased solar radiation under various thermal sensation states. Under “extremely cold” conditions (TSV = −5), 100% of local tourists preferred an increase in solar radiation, while the proportion of non-local tourists desiring increased solar radiation was approximately 60%. As the value of TSV increased, the proportion of tourists desiring decreased solar radiation generally showed a downward trend, while the proportion preferring unchanged solar radiation increased. Under neutral thermal sensations (TSV = 0), 33% of non-local tourists preferred increased solar radiation compared with 52% of local tourists.

3.1.2. Differentiation of Thermal Preference between Local and Non-Local Tourists under Different PET Intervals

Table 5. Correlations between thermal preference and PET.

	Air Temperature Preference	Relative Humidity Preference	Wind Speed Preference	Solar Radiation Preference
TSV of local tourists	−0.156 **	0.023	−0.195	−0.161 **
TSV of non-local tourists	−0.155 **	0.016	−0.073 *	−0.222 **

** Correlation is significant at the 0.01 level. * Correlation is significant at the 0.05 level.

presents the results of the correlation analysis between the preferences for thermal environmental parameters and PET for local and non-local tourists. Both the air temperature preference and solar radiation preference for local and non-local tourists showed a significant negative correlation with PET. However, the relative humidity preference and wind speed preference were not significantly correlated with PET.

Figure 9 displays the air temperature preference voting distribution among local and non-local tourists under different PET conditions. The proportion of local tourists desiring the air temperature to rise was significantly higher than that of non-local tourists. When −28 °C < PET < −22 °C, the proportion of local tourists preferring the air temperature increase ranged from approximately 85% to 100%, while the corresponding proportion among non-local tourists was only around 50%. Approximately 10% of non-local tourists preferred a lower air temperature. This indicates that local tourists exhibit a stronger preference for an air temperature increase compared to non-local tourists when the thermal environment is colder. Additionally, when −10 °C < PET < 2 °C, the proportion of non-local tourists desiring an air temperature reduction was significantly higher than that of local tourists by about 20%.

Figure 10 illustrates the solar radiation preference voting distribution among local and non-local tourists under different PET conditions. Compared with local tourists, the percentage of non-local tourists who desired increased solar radiation was generally lower, especially when PET was relatively high. When −16 °C < PET < 2 °C, the proportion of non-local tourists desiring increased solar radiation was approximately 10% to 20% lower than that of local tourists. Additionally, there were virtually no local tourists expressing a preference for decreased solar radiation. However, 5% to 10% of non-local tourists desired decreased solar radiation.

3.2. Thermal Sensation Vote

In this study, the scale of thermal sensation was set at 11 points based on the regional climate characteristics of the severely cold region. However, the original scale of PET includes nine points: “very cold, cold, cool, slightly cool, neutral, slightly warm, warm, hot, very hot,” without specifically including “extremely cold.” Therefore, following this approach, this study combined “extremely cold” and “very cold” in the subjective questionnaire into a single Point (“very cold”) when analyzing the PET-based TSV prediction model for local and non-local tourists to facilitate statistics and calculation.

This study referred to the research method of human thermal comfort proposed by Lin et al. to derive the prediction model for TSV [87]. The TSV corresponding to each 1 °C PET interval was averaged, and the resulting values of average thermal sensation were then regressed against their respective PET values to obtain the prediction model. Furthermore, to minimize errors caused by individual differences, this study excluded groups with fewer than three samples in each interval. The PET-based TSV prediction models for both local and non-local tourists are as follows:

$$\mathrm{Local} \mathrm{tourists}:\mathrm{T}\mathrm{S}\mathrm{V}=0.072\mathrm{P}\mathrm{E}\mathrm{T}-1.315 ({\mathrm{R}}^2=0.920)$$

(2)

$$\mathrm{Non}-\mathrm{local} \mathrm{tourists}:\mathrm{T}\mathrm{S}\mathrm{V}=0.058\mathrm{P}\mathrm{E}\mathrm{T}-2.109 ({\mathrm{R}}^2=0.921)$$

(3)

The corresponding relationship and fitted lines between TSV and PET for local and non-local tourists are depicted in Figure 11. The fluctuation range of TSV for local tourists (−3.17 to −1.43) was significantly higher than that of non-local tourists (−3.67 to −1.83). Furthermore, both local and non-local tourists exhibited a positive correlation between TSV and PET, indicating that the TSV values increase with the increase in the PET value. The magnitude of the TSV increase for local tourists was noticeably greater than that for non-local tourists. The slope values corresponding to the fitted curves were 0.072 and 0.058, respectively.

3.3. Thermal Satisfaction Vote

To derive the prediction model of TS_aV based on PET, the average TS_aV was calculated for each 1 °C PET interval [87]. The resulting values of average thermal satisfaction were then regressed against their respective PET values to obtain the prediction model. To minimize errors caused by individual differences, the groups with fewer than three samples in each interval were excluded from this study. The PET-based TS_aV prediction models for both local and non-local tourists are as follows:

$$\mathrm{Local} \mathrm{tourists}:\mathrm{T}{\mathrm{S}}_{\mathrm{a}}\mathrm{V}=0.021\mathrm{P}\mathrm{E}\mathrm{T}-0.534 ({\mathrm{R}}^2=0.788)$$

(4)

$$\mathrm{Non}-\mathrm{local} \mathrm{tourists}: \mathrm{T}{\mathrm{S}}_{\mathrm{a}}\mathrm{V}=-0.00137{\mathrm{P}\mathrm{E}\mathrm{T}}^2-0.04\mathrm{P}\mathrm{E}\mathrm{T}-0.86 ({\mathrm{R}}^2=0.731)$$

(5)

Figure 12 illustrates the corresponding relationship and fitted lines between TS_aV and PET for local and non-local tourists. For local tourists, there was a strong linear relationship between TS_aV and PET. With the increase in PET values, the TS_aV of local tourists gradually increased. The PET value corresponding to the state of “slightly dissatisfied” (TS_aV = −1) for local tourists was −22.19 °C. On the other hand, the TS_aV of non-local tourists showed a good quadratic relationship with PET, reaching its maximum value at a PET of −16 °C. When PET < −16 °C, the TS_aV showed a decreasing trend as PET values decreased. When PET > −16 °C, the satisfaction level also decreased significantly as PET values increased.

3.4. Thermal Comfort Vote

The TCV corresponding to each 1 °C PET interval was averaged, and the prediction model derived from the regression analysis of the obtained average TCV and the corresponding PET value was as follows:

$$\mathrm{Local} \mathrm{tourists}:\mathrm{T}\mathrm{C}\mathrm{V}=0.029\mathrm{P}\mathrm{E}\mathrm{T}-0.704 ({\mathrm{R}}^2=0.755)$$

(6)

$$\mathrm{Non}-\mathrm{local} \mathrm{tourists}:\mathrm{T}\mathrm{C}\mathrm{V}=-0.00164{\mathrm{P}\mathrm{E}\mathrm{T}}^2-0.03\mathrm{P}\mathrm{E}\mathrm{T}-1.06 ({\mathrm{R}}^2=0.778)$$

(7)

Figure 13 depicts the corresponding relationship and fitted lines between TCV and PET for local and non-local tourists. The TCV of local tourists showed a positive linear correlation with PET, indicating that the comfort level of local tourists increased proportionally as PET values increased. In the cold winter thermal environment, the research findings of this study did not reach the comfort state (TCV = 0) for local tourists. The PET value corresponding to the state of “slightly uncomfortable” (TCV = −1) for local tourists was −10 °C, which was close to the PET value when local tourists perceived the thermal sensation as “cool” (TSV = −2). On the other hand, the TCV of non-local tourists exhibited a good quadratic relationship with PET. The PET range for the states of “slightly uncomfortable” (TCV = −1) and above was from −18.4 °C to −1.8 °C. When the PET value was −10 °C, the thermal comfort level of non-local tourists reached its peak. Subsequently, the TCV of non-local tourists showed a downward trend whether it increased or decreased with the PET value.

Based on the relationship between TCV and PET for local and non-local tourists, when PET < −7 °C, the TCV of non-local tourists was generally higher than that of local tourists. However, when −7 °C< PET < −28 °C, local tourists had higher TCV values than non-local tourists.

4. Discussion

4.1. Differentiation of Thermal Preference between Local and Non-Local Tourists

Compared with local tourists, the preference percentage of non-local tourists for the increase in air temperature, the increase in solar radiation, and the decrease in wind speed was relatively lower under various thermal sensation conditions. This is because the long-term living experiences in severely cold regions make local tourists have the stereotype of improving the outdoor thermal environment in winter using rising air temperatures, increasing solar radiation, and decreasing wind speed. Even though the local tourists rated the thermal sensation as “neutral” (TSV = 0), they still preferred the warmer environment compared with non-local tourists. One of the travel purposes of non-local tourists is to experience the low-temperature climate characteristics of the winter in severely cold regions. Therefore, when the thermal sensation was “neutral” or below, the percentage of votes for their preference for thermal environmental quality improvement was significantly lower than that of local tourists.

When the thermal environment was relatively cold, local tourists were significantly more likely to prefer the increased air temperature and increased solar radiation than non-local tourists. With the warming of the thermal environment, the preference of local tourists for increasing air temperature and solar radiation decreased significantly. When PET was increased to the upper limit value during the measurement period (−1 °C < PET < 2 °C), the percentage of votes for local and non-local tourists who preferred higher air temperature and solar radiation was similar, but the percentage of votes for non-local tourists who preferred lower temperature and solar radiation was significantly higher than that of local tourists. This is because local tourists have a strong preference for a warmer environment, influenced by the long and cold winter. Yang et al. also came to a similar research conclusion. Compared with outsiders, residents in Umeå were affected by the long-term low-temperature climate in winter, still having a higher preference for solar radiation enhancement even in the “slightly warm” thermal state [34]. One of the travel purposes of non-local tourists was to experience the cold weather conditions in winter in severely cold regions. When the thermal environment was cold, the psychological expectations of non-local tourists were satisfied. That was why their preference for air temperature rise and solar radiation enhancement was much lower than that of local tourists. This conclusion aligns with the findings of Michelle Rutty et al. on Caribbean beaches during summer high-temperature conditions. They found that even in hot weather, over 60% of beachgoers expressed a preference for a stable temperature, with 10% even desiring an increase in temperature. The thermal preference for air temperature of beach tourists was 18 °C UTCI higher than that of urban park crowds [30]. Our research findings are similar to those of the referenced studies, indicating that tourists’ psychological preferences significantly influence their thermal comfort experiences. The expectations and psychological anticipations of non-local tourists regarding specific climatic conditions can influence their preferences for thermal environmental parameters.

4.2. Differentiation of Thermal Sensation between Local and Non-Local Tourists

The TSVs of both local and non-local tourists were positively correlated with PET. The TSV of local tourists was always maintained at a higher level than that of non-local tourists, generally differing by approximately one unit. This was due to the long-term thermal experience of local tourists in severely cold regions, which led to their strong adaptability to the outdoor thermal environment in winter. In addition, the slope of the fitting curve of TSV and PET for local tourists was higher than that for non-local tourists, which indicates that the thermal sensation of local tourists changed more than that of non-local tourists as the PET varied. Tian et al. pointed out in their study that non-local tourists were slightly more sensitive to PET than residents in the outdoor thermal environment in cold regions in winter [39]. The conclusions of this study are different because Tian et al. performed the research in Xi’an China, which is located at the junction of the cold regions and the hot summer, cold winter regions in the thermal zones of China. The average temperature of the coldest month in winter there is 0.49 °C [88]. Harbin is located in a severely cold region, where the average temperature of the coldest month in winter is about −17 °C. The long-term thermal history of non-local tourists in other climate regions leads to their poor adaptability to the outdoor thermal environment in severely cold regions in winter, which results in the thermal sensation of non-local tourists being basically “cold” or below. As a result, their thermal sensation has a relatively small change range.

4.3. Differentiation of Thermal Satisfaction between Local and Non-Local Tourists

As an evaluation of the comprehensive state of tourists, thermal satisfaction is greatly affected by psychological factors. The TS_aV of local tourists showed a strong linear relationship with PET, which gradually increased with the warming of the environment. The TS_aV of non-local tourists showed a quadratic relationship with PET. Affected by the purpose of travel and psychological expectations, the TS_aV of non-local tourists reached the peak when PET was −16 °C. When the thermal environment got colder (PET < −16 °C), non-local tourists’ thermal comfort worsened, resulting in a downward trend in their thermal satisfaction. When the thermal environment got warmer (PET > −16 °C), the psychological expectations of non-local tourists to experience cold weather could not be satisfied, so their thermal satisfaction also showed a downward trend. In addition, the differences between the travel purposes of local and non-local tourists also affected the results. When the thermal environment was colder (PET < −6 °C), the thermal satisfaction of non-local tourists remained consistently high, while that of local tourists decreased. With the increase in the PET value (PET > −6 °C), the warming of the thermal environment made the TS_aV of local tourists higher than that of non-local tourists.

4.4. Differentiation of Thermal Comfort between Local and Non-Local Tourists

The TCV of non-local tourists showed a good quadratic relationship with PET. When the PET value was −10 °C, the thermal comfort of non-local tourists reached a peak. When the thermal environment got warmer (PET > −10 °C), the expectations of non-local tourists to experience cold climates could not be satisfied, resulting in poor thermal comfort. When the thermal environment got colder (PET < −10 °C), non-local tourists were not adapted to the extremely cold climate, so their thermal comfort also showed a declining trend.

In addition, when −28 °C < PET < −7 °C, the TCV of non-local tourists was higher than that of local tourists. This is because when non-local tourists travel to Harbin, they have certain psychological expectations for the severely cold climate. Moreover, the voluntariness of the travel experience will affect people’s evaluation of thermal comfort. This is consistent with the conclusion of Lopes et al., i.e., that the voluntariness of the travel experience makes tourists more tolerant of various climatic conditions [35]. However, when PET < −28 °C, the TCV of local tourists was higher than that of non-local tourists, which indicates that local tourists’ local long-term thermal history makes them more adaptable to severely cold climates. The study by Mina et al. showed that when exposed to the same thermal environment, participants with a warmer thermal history felt cooler compared to their counterparts in the colder thermal history groups [89]. Xue also concluded that a longer residence was associated with higher overall comfort [90]. When PET > −7 °C, the TCV of local tourists was higher than that of non-local tourists. This is because the thermal environment at this time was inconsistent with the tourism purpose of non-local tourists to experience the severely cold climate, but the situation was in line with the expectations of local tourists for the warmer environment. Therefore, the TCV of local tourists increased significantly higher than that of non-local tourists.

In addition, the PET value at the peak of non-local tourists’ TS_aV (−16 °C) was lower than that at the peak of non-local tourists’ TCV (−10 °C). It can be seen that when non-local tourists experienced the severely cold climate in Harbin, although the thermal environment did not reach the best level of their thermal comfort, they still showed the highest satisfaction with the current thermal environment.

4.5. Limitations and Research Recommendations and Directions

This study has certain limitations. This paper analyzes the causes of these limitations and discusses future research directions. Firstly, when defining local and non-local tourists, to make the results more relevant, this study categorized tourists from severely cold regions as local tourists, while tourists from other regions were considered non-local tourists. The geographical climatic characteristics of each climatic region have significant differences. Therefore, there are differences in thermal comfort among tourists from different geographical regions due to their long-term thermal histories [22,88,90]. Future research could focus on conducting a more detailed and comprehensive survey of non-local tourists from different climatic regions to explore the differences in thermal comfort experiences between them and local tourists from severely cold regions. Additionally, the thermal comfort experience of tourists in urban is influenced not only by their original locations but also by other factors such as their cultural background, activity status, health status, age, gender, mood, occupation, and clothing [25,26,65,91,92]. Therefore, in future research, the relationship between various factors concerning tourists and thermal comfort should also be explored in depth, considering the original places tourists come from. Furthermore, different scenic characteristics of various tourist destinations, such as shaded areas in waterfront locations, landscape components in mountainous areas, water bodies, and waterfront green spaces, will also influence tourists’ thermal comfort [13,16,17,93].

5. Conclusions

Through on-site measurements and questionnaire surveys, this study reveals the differences in thermal comfort between local and non-local tourists at outdoor tourist attractions in winter in severely cold regions of China. The conclusions are as follows:

• The TSV showed a significant negative correlation with air temperature and solar radiation preference voting and a significant positive correlation with wind speed preference voting. Under the same thermal sensation conditions, local tourists exhibited a significantly larger proportion of votes for increased air temperature, enhanced solar radiation, and reduced wind speed compared to non-local tourists. The PET also exhibited a significant negative correlation with air temperature and solar radiation preferences. Local tourists expressed a larger proportion of votes for increased air temperature and enhanced solar radiation compared to non-local tourists, especially when the environment turned cold;
• The TSV of both local and non-local tourists showed a positive correlation with the PET, with local tourists generally exhibiting higher TSV values compared to non-local tourists. Furthermore, as the PET values increased, the TSV of local tourists showed a more pronounced upward trend compared to non-local tourists;
• There was a linear positive correlation between the TS_aV and PET of local tourists and a quadratic relationship between the TS_aV and PET of non-local tourists. Influenced by the purpose of travelling and psychological expectation, the TS_aV of non-local tourists was highest when PET = −16 °C. Their thermal satisfaction decreased when the PET value increased or decreased. The TS_aV of non-local tourists was higher than that of local tourists when PET < −6 °C. When PET > −6 °C, the TS_aV of foreign tourists was lower than that of local tourists;
• There was a positive linear correlation between TCV and PET for local tourists and a quadratic relationship between TCV and PET for non-local tourists. The TCV of non-local tourists peaked at PET of −10 °C and decreased significantly with the increase or decrease in PET. The TCV of non-local tourists was generally higher than that of local tourists at −28 °C < PET< −7 °C. When PET > −7 °C or PET < −28 °C, local tourists had better thermal comfort than non-local tourists.