Fundamental Research on Sustainable Building Design for the Rural Elderly: A Field Study of Various Subjective Responses to Thermal Environments and Comfort Demands during Summer in Xi’an, China

Abstract

Well-designed residential thermal environments that meet the comfort and health needs of elderly individuals can enhance their well-being and decrease associated health risks, which is one of the foundations of sustainable building development. However, limited evidence on thermal response patterns and thermal comfort needs of rural elderly leads to insufficient support for improving thermal environments. This study was conducted in seven villages in Xi’an, a cold region of China, and both subjective questionnaires of thermal comfort and objective physical environment tests were adopted. Correlations between nine kinds of human subjective responses and thermal environmental parameters were examined. The neutral values of operative temperature, relative humidity, and air velocity were calculated to be 23.9 °C, 63.3%, and 0.28 m/s, respectively. Comfort ranges for temperature, humidity, and air velocity were determined to be ≤29.3 °C (80% acceptability), 43.3–81.0% (80% acceptability), and 0.16–0.41 m/s (90% acceptability). Clothing insulation of rural elderly in summer was relatively higher and less sensitive to temperature shifts. The slope value of the thermal adaptative model was larger, with higher comfort temperatures in warmer environments. These results are the outcome of the prolonged adaptability to the regional climate and the poor indoor thermal environment, as evidenced by lower psychological expectations, higher behavior adjustment, and environmental decisions influenced by household low income. The findings in this study can be used as a basis for the design or improvement of residential thermal environments for rural older adults.

1. Introduction

Fundamental research on sustainable building design should not only focus on the utilization of local climate resources to achieve energy savings but also on the creation of thermal environments that meet the comfort demands of occupants. Since the thermal comfort of occupants is essential for guiding the design and evaluation of thermal environments in thermal standards [1], designing indoor thermal environments that match elderly occupants’ demands for comfort across varying climates is critical to achieving comfortable and healthy living environments. The “World Report on Ageing and Health” states that the interactions each of us has with the environments we inhabit across our lives have an impact on human functioning [2]. Older adults spend a minimum of 80% of their time indoors, exceeding the time spent by other age groups. Individuals with limited mobility could potentially spend up to 95% of their time indoors [3]. As people age, numerous underlying physiological changes occur. For example, as the immune system of older adults becomes weaker [4,5], as well as the sensitivity to the thermal environment [6], they may be more susceptible to dehydration, hypothermia, and hyperthermia [2], and the risk of chronic disease rises. Additionally, older persons have reduced immune function and a heightened risk of developing chronic diseases [4,5]. Research on the thermal comfort of the elderly has attracted more and more attention worldwide, and some field studies have been conducted in different countries or climate zones, such as hot summer and warm winter zones [7,8,9], hot summer and cold winter zones [10,11,12,13,14], and cold zones [3,15,16,17,18,19,20,21] in China; a cold continental climate with dry winter and hot summer in South Korea [5]; a temperate climate with dry and hot summer in Spain [22,23,24]; a temperate climate without dry season in Australia [25]; and a warm and humid climatic zone in India [26]. However, almost all of the above studies have been conducted in urban nursing homes or urban residential buildings. The results of the seventh national census show that the elderly over 60 years of age accounted for 18.7% of China’s population in 2020, with nearly 60% of the elderly living in rural areas [27]. It is projected that by 2050, the percentage of elderly individuals in China will increase to 35.1% [28], which will result in an increase in demand for various types of senior care facilities. For the elderly in rural China, family care is the most common form. Consequently, dwelling houses, in which rural elderly people reside, will be an important type of aged care facility in China, and more attention should be paid to the quality of its indoor thermal environment.

The economic development of rural areas lags behind that of urban areas in China. As a result, in the field of urban and rural construction, there are many differences between rural dwelling houses and urban residential buildings in terms of building layouts, thermal performance of building envelope, mode of building operation, indoor thermal environment, etc. [10,18,29]. According to national statistics, China’s per capita disposable income in 2022 was RMB 20,133 for rural residents and RMB 49,283 for urban residents [30]. Dwelling houses in rural China are primarily financed and constructed by the residents themselves. There exists a significant association between their housing conditions and income levels [31]. To reduce house construction costs, most rural houses are built using brick-concrete construction without thermal insulation, and the windows are typically single-glazed with frames made of aluminum or wood [32,33]; and in order to save operating costs, a considerable proportion of rural houses are naturally ventilated almost throughout the year. Living in a rural environment for a lifetime has resulted in greater thermal tolerance for rural residents compared to their urban counterparts [34], as well as reduced psychological expectations [35]. In addition, there are also differences in the lifestyles, cultural customs, and physiological adaptations between urban and rural residents, all of which may cause the rural elderly’s subjective responses to the thermal environment of the building, as well as the demand for comfort, to be different from that of the urban elderly. However, a literature review revealed that this has not been confirmed in the established studies, because few relevant articles detail the relative research on this topic in rural areas in China.

Summer is the hottest season of the year, and maintaining an appropriate indoor thermal environment is crucial for the comfort and well-being of occupants. High temperatures above the long-term averages during summer months and extreme heat events, such as heatwaves, are linked with increased mortality rates [36], with people 65 years and older, especially those with comorbidities such as cardiovascular disease, being among the most vulnerable [37]. Clarifying the elderly occupants’ thermal responses to the summertime residential environment, as well as their required thermal comfort levels, is crucial for maintaining their health. Some field studies on this issue in different climate zones have been carried out worldwide, and the detailed information has been summarized in

Table 1. Field research on thermal comfort of the elderly in summer in various climatic regions.

Authors	Location	Climate Zone	Building Type	Mode of Operation	Human Responses		Reference
					Type	Results
Hwang et al.	Urban Taiwan, China	HSWW	RB	NV	Thermal	TSV = 0.39 Top − 9.84; Tn = 25.2 °C; 80% acceptable range: 23.2–27.1 °C.	[7]
Tao et al.	Urban Hongkong, China	HSWW	NH	NV and AC	Thermal	TSV and Ta: p = 0.164; Majority acceptable range: 24–27 °C (TSV = ± 1);	[8]
					Humidity	Mean HSV = 1.12; HSV and RH: Non-significant.
					Air movement	Mean ASV = 1.00; ASV and va: p = −0.186.
Wu et al.	Urban Chongqing, China	HSCW	NH	NV	Thermal	TSV = 0.13 Ta − 3.2; Tn = 24.6 °C;	[10]
			RB	NV	Thermal	TSV = 0.10 Ta − 2.3; Tn = 23.0 °C;
				AC	Thermal	TSV = –0.07 Ta + 3.3;
			NH and RB	NV and AC	Humidity Air movement	HSV and RH: rs = −0.276; ASV and va: Non-significant.
Jiao et al.	Urban Shanghai, China	HSCW	NH	NV	Thermal	TSV = 0.128 Top − 3.249; Tn = 25.4 °C; Acceptable range: 28.2–29.1 °C (TSV = ± 1).	[11]
Jiao et al.	Urban Shanghai, China	HSCW	RB	NV	Thermal	TSV = 0.124 Top − 3.145; Tn = 25.4 °C; 80% acceptable range: <26.8 °C; 90% acceptable range: <24.8 °C. Acceptable range: 23.8–27.0 °C (TSV = ± 0.2).	[12]
Zheng et al.	Urban Xi’an, China	C	NH	NV	Thermal	TSV = 0.135 Top − 3.25; Tn = 24.1 °C; 80% acceptable range: <30.3 °C; 90% acceptable range: <28.3 °C.	[3]
Zheng et al.	Urban Baoding, China	C	NH	NV	Thermal	60 < Age < 69: TSV = 0.20 Ta − 5.15; Acceptable range: 23.25–28.25 °C (TSV = ± 0.5); 70 < Age < 89: TSV = 0.06 Ta − 1.53; Acceptable range: 17.17–33.83 °C (TSV = ± 0.5); 90 < Age < 99: TSV = 0.05 Ta − 1.26; Acceptable range: 15.20–35.20 °C (TSV = ± 0.5);	[15]
					Humidity	60 < Age < 69: HSV = − 0.04 RH + 1.99; 70 < Age < 89: HSV = − 0.02 RH + 0.31; 90 < Age < 99: HSV = − 0.01 RH + 0.05;
					Air movement	60 < Age < 69: ASV = 3.57 va − 0.59; Acceptable range: 0.02–0.31 m/s (ASV = ± 0.5); 70 < Age < 89: ASV = 2.46 va − 0.32; Acceptable range: 0–0.33 m/s (ASV = ± 0.5); 90 < Age < 99: ASV = 0.43 va − 0.24; Acceptable range: 0–1.72 m/s (ASV = ± 0.5);
Wang et al.	Urban Baoding, China	C	NH	NV	Thermal	TSV = 0.2903 Top − 7.9097; Tn = 27.25 °C; Acceptable range: 27.52–28.97 °C (TSV = ± 0.5).	[16]
Feng et al.	Urban Tianjin, China	C	NH	NV and AC	Thermal	TSV = 0.2532 Top − 6.752; Tn = 27.6 °C; 90% acceptable range: 26.8–28.3 °C.	[17]
Yang et al.	Urban Seoul, South Korea	Dwa	NH	NV and AC	Thermal	20 < Top < 25: TSV = 0.57 Top − 13.63; 25 < Top < 27: TSV = 0.01 Top − 0.10; 27 < Top < 32: TSV = 0.12 Top − 3.12.	[5]
Forcada et al.	Barcelona, Tarragona and Valencia, Spain	Csa	NH	NV	Thermal	Residents: TSV = 0.2838 Top − 7.1687; Tn = 25.3 °C; 90% acceptability: 23.2–28.4 °C; Non-residents: TSV = 0.5599 Top − 13.395; Tn = 23.9 °C; 90% acceptability: 23.2–26.3 °C.	[22]
Baquero et al.	Las Rozas, Alcobendas, Colmenar Viejo and Madrid, Spain	Csa	NH	AC (VRV and FC)	Thermal	TSV = 0.15569 Top − 4.0227; Tn = 25.6 °C; 90% acceptability: 19.4–32.3 °C (TSV = ± 1).	[23]
Tartarini et al.	South-eastern NSW, Australia	Cfb and Cfa	NH	NV and AC	Thermal	Residents: Tn = 22.9 °C; 90% acceptable range: <26.2 °C; 47% wanted cooler (Top > 26.5 °C); Preferred 23.2 °C; Non-residents: Tn = 22.0 °C; 90% acceptable range: <24.2 °C; 80% wanted cooler (Top > 23.5 °C); Preferred 21.7 °C.	[25]
Sudarsanam et al.	Urban Salem, India	WH	RB	NV	Thermal	TSV = 0.3546 Top − 10.62; Tn = 30.0 °C; 80% acceptable range: 25.7–32.0 °C; 90% acceptable range: 26.9–30.9 °C.	[26]
					Humidity Air movement	HSV and RH: rs = 0.17; ASV and va: rs = 0.39;

Note: Climate zone: HSWW: Hot summer and warm winter zone in China; C: Cold zone in China; Csa: Temperate, dry summer, hot summer (Köppen–Geiger); Dwa: Cold, continental, dry winter, hot summer (Köppen–Geiger); Cfb: Temperate, without dry season, warm summer (Köppen–Geiger); Cfa: Temperate, without dry season, hot summer (Köppen–Geiger); WH: Warm and humid climatic zone in India; Building type: RB: Residencial buildings; NH: Nursing homes; Mode of operation: NV: Naturally ventilated; AC: Air conditioning; VRV: Variable refrigerant flow; FC: Fan coils; Human responses: TSV: Thermal sensation vote; HSV: Humidity sensation vote; ASV: Air movement sensation vote; T_op: indoor operative temperature, °C; T_n: Neutral temperature, °C; T_a: Air temperature, °C; RH: Relative humidity, %; v_a: Air velocity, m/s; r_s: Spearman’s rank correlation coefficient.

. As can be seen from the table, scholars have investigated issues concerning the thermal comfort of elderly individuals in nursing homes or residential buildings for various climate types. The relationship between TSV (thermal sensation vote, a core indicator of subjective thermal response) and temperature, a core parameter of thermal environment, was explored in almost all studies, and neutral temperature or comfortable temperature ranges were then obtained. Furthermore, a small number of studies have also discussed the indoor relative humidity (RH) and air movement (v_a), as well as the relationship between RH and HSV (humidity sensation), v_a, and ASV (air movement sensation). Regrettably, there are relatively few studies on older adults residing in rural areas.

In rural areas of China, most dwelling houses are predominantly naturally ventilated, causing the indoor thermal environment to be significantly affected by the outdoor climate. During most of hot days in summer, rural older people usually adopt opening windows and using fans to improve the indoor thermal environment, while in a few very hot cases (extremely hot in summer night [38]), some older people in well-off homes use air conditioners to keep comfort. This leads to a distinct characterization of the indoor thermal environment in rural dwellings compared to that of urban residential buildings. This is supported by the evidence that indoor air temperatures in rural settings are slightly higher than those in urban areas [39], as are indoor relative humidity [29,40] and indoor air velocity [34]. Theoretically, the thermal neutrality and comfort of temperatures are subject to change with alterations in humidity and wind speed. However, most studies have not established the comfortable range of humidity and wind speeds [34]. These imply that when evaluating the indoor thermal environments in rural dwelling houses for the elderly, it is important to consider not only the subjective response indicators such as thermal sensation vote (TSV), thermal acceptability (TA), and thermal preference (TP) related to indoor temperature but also other subjective response indicators related to the indoor relative humidity and air velocity, which are two other parameters of the thermal environments.

As shown in Table 1, few studies have focused on subjective response indicators other than TSV, TA, and TP. For the thermal environment evaluation of elderly homes in rural China, it is necessary to conduct a comprehensive study to grasp the relationship between various types of subjective responses and the corresponding objective thermal environment parameters in order to better guide the design of the thermal environment or comfort improvement of dwelling houses in this area. To achieve this objective, the study conducted a survey in rural Xi’an, representing a typical city in the cold climate zone of China. The study aimed to:

• To determine the characteristics of various subjective responses of rural elderly to indoor thermal environments and develop the relationships between them.
• To identify the comfort needs of rural elderly people, establish thermal adaptive model, and perform a comprehensive analysis of humidity and air movement to determine comfort ranges.

2. Methods

2.1. Study Area

Xi’an is a representative of China’s megacities, with an area of 10,108 km² and a population of 13 million. It is situated at latitude 34°15′ and longitude 108°55′ (Figure 1) and serves as the study location in this research. China is separated into five climate zones based on the average temperature of the hottest and coldest months throughout the country [41]. The city undergoes significant alterations in outdoor temperature and humidity across seasons, as illustrated in

Table 2. Main outdoor climatic parameters and general climatic characteristics in Xi’an, China [3].

Season	Period (in 2020, for Example)	Mean Outdoor Temperature/ °C	Mean Relative Humidity/%	General Climate Characteristics
Spring	20 March to 20 June	14.8	66.1	In spring, the climate is warm and dry, causing a sharp rise in temperature. Summer brings a hot and rainy climate, while fall is characterized by a cool and humid atmosphere with a sharp drop in temperature. Winter exhibits cold weather with less frequent snow/rain.
Summer	21 June to 21 September	26.0	68.0
Autumn	22 September to 20 December	14.4	72.8
Winter	21 December to 19 March	1.0	65.0

. The highest annual temperature can rise to around 40 °C, while the lowest temperature can drop to −8 °C [3]. The field survey for this study was conducted during the summer of 2020 and 2023, with a total of 8 days (2020: 21–22 July, 23–24 August, 7 September, 19 September; 2023: 22–23 July). Based on the meteorological data of Xi’an in the summer of 2020 and 2023 [42], the mean daily temperature (T_m), mean relative humidity (RH), maximum temperature (T_max), and minimum temperature (T_min) were counted, and the statistical results are shown in Figure 2. The figure illustrates that the average daily maximum temperature reached 32.6 °C on 10 July 2023, while the minimum temperature on 21 September 2020 was 14.8 °C. The maximum outdoor temperature was 37 °C, which occurred on 7 July 2020 and from 8 July to 11 July 2023, while the minimum temperature was 11 °C, which occurred on 17 September 2020. The average relative humidity reached a maximum value of 99% on 25 July 2020 and a minimum value of 50% on 9 July 2023.

The research work of this study was carried out in 7 villages, including Shangcao Village, Xiacao Village, Yejiazhai Village, Jingjia Village, Jiangnan Village, Nanqiang Village, and Guonan Village, all of which are located in the Chang’an and Huyi Districts of Xi’an, and the specific locations are shown in Figure 3. Information on the basic form of the villages, the appearance of dwelling houses, and spatial layout is shown in Figure 4. It can be seen that the buildings in rural areas of Xi’an are mainly of the “single-family house”, usually including the main building, gate building, and yard, forming a narrow and long north–south plan. Most residential sites for this type of dwelling house typically have a width of 10 m and a length ranging from 20 to 30 m. The main-building serves as the primary living space for young couples and children, usually measuring 10 m in width and 8 to 10 m in length, spanning across 2 floors. The gate-building usually includes a bedroom of the elderly and a kitchen and is generally a one-story building. All the buildings are uninsulated and are naturally ventilated throughout the year. In winter, there is no centralized heating, and “Chinese kang” [43] is usually used to improve the thermal environment of the residence. In summer, electric fans typically improve the human thermal comfort, while some families use air conditioners to enhance the thermal environment during the hottest times.

2.2. Respondents Surveyed

Older persons aged 60 years and above voluntarily participated in the survey, and 230 valid questionnaires were collected, of which 84 were from men and 146 from women, whose physical, mental, and social health status was considered healthy (as per [44]).

Table 3. Background information of respondents.

Sample Size	Gender	Parameters	Min.	Max.	Mean	S.D.
230	M: 84	Height/cm	156.0	183.0	168.0	5.86
		Weight/kg	45.0	98.0	66.6	10.0
		Age/yr	60.0	85.0	69.6	6.84
		Metabolic rate/met	0.7	2.0	1.19	0.42
	F: 146	Height/cm	140.0	172.0	158.4	5.43
		Weight/kg	43.5	80.0	60.7	8.88
		Age/yr	60.0	90.0	68.6	6.67
		Metabolic rate/met	0.7	2.0	1.35	0.46

presents a summary of participants’ background information, showing their height, weight, age, and metabolic rate (determined by the activity level of elderly participants who completed the questionnaire and ASHRAE 55-2020 [45], refer to

Table 4. Metabolic rates for typical activity of the rural elderly [45].

Activity	Metabolic Rate/Met	Activity	Metabolic Rate/Met
Sleeping	0.7	Cooking	1.6–2.0
Reclining	0.8	House cleaning	2.0–3.4
Seated, quiet	1.0	Seated, heavy limb movement	2.2
Standing, relaxed	1.2	Sawing (table saw)	1.8
Walking, 0.9 m/s	2.0	Light electrical industry	2.0–2.4
Walking, 1.2 m/s	2.6	Heavy electrical industry	4.0
Reading, seated	1.0	Handling 50 kg (100 lb) bags	4.0
Writing	1.0	Pick and shovel work	4.0–4.8
Walking about	1.7	Dancing, social	2.4–4.4
Lifting/packing	2.1	Calisthenics/exercise	3.0–4.0

), and the distributions are represented in Figure 5.

2.3. Questionnaire

Thermal comfort is inherently subjective and is defined as the perception of comfort with the thermal environment [46]. The questionnaire determined respondents’ subjective sensations to the thermal environment surrounding them, including thermal sensation vote (TSV), thermal acceptability (TA), thermal preference (TP), humidity sensation vote (HSV), humidity acceptability (HA), humidity preference (HP), air movement sensation vote (ASV), air movement acceptability (AA), and air movement preference (AP). TSV, HSV, and ASV were measured on a 7-point scale. On the TSV scale, values ranged from “cold” (−3) to “hot” (+3), with 0 denoting neutrality. The HSV scale ranged from “too humid” (−3) to “too dry” (+3), with 0 indicating neutrality. Similarly, the ASV scale ranged from “too weak” (−3) to “too strong” (+3), with 0 representing just right. TA, HA, and AA were assessed using a 4-point acceptability scale, ranging from “very unacceptable” (−2) to “very acceptable” (+2), with “just unacceptable” (−1) and “just acceptable” (+1) falling in between. TP, HP, and AP were assessed using a 3-point scale. A +1 indicates a desire for “want warmer”, “want drier”, or “want stronger”, while 0 indicates “no change”, and −1 indicates “want cooler”, “want more humid”, or “want weaker” for TP, HP, and AP. The detailed sensation rating scales are presented in

Table 5. Occupants’ subjective sensation rating scales.

Responses		Scale
		+3	+2	+1	0	−1	−2	−3
Thermal	Sensation	Hot	Warm	Slightly warm	Neutral	Slightly cool	Cool	Cold
	Acceptability	-	Very acceptable	Just acceptable	-	Just unacceptable	Very unacceptable	-
	Preference	-	-	Want warmer	No change	Want cooler	-	-
Humidity	Sensation	Too dry	Dry	Slightly dry	Neutral	Slightly humid	Humid	Too humid
	Acceptability	-	Very acceptable	Just acceptable	-	Just unacceptable	Very unacceptable	-
	Preference	-	-	Want drier	No change	Want more humid	-	-
Air movement	Sensation	Too strong	Strong	Slightly strong	Just right	Slightly weak	Weak	Too weak
	Acceptability	-	Very acceptable	Just acceptable	-	Just unacceptable	Very unacceptable	-
	Preference	-	-	Want stronger	No change	Want weaker	-	-

, and the distributions of respondents’ subjective sensation votes are shown in Figure 6. Some common adaptive behaviors were recorded, particularly related to opening and closing doors and windows, using air conditioners or fans, and concluding with clothing insulation checklists based on ASHRAE 55 [45]. The sum of individual clothing values (I_cl,i) listed in

Table 6. Clothing insulation values for elderly occupants in summer [45].

Garment Description	Clothing Insulation/Clo	Garment Description	Clothing Insulation/Clo
Underwear		Shirts and Blouses
Bra	0.01	Sleeveless/scoop-neck blouse	0.12
Panties	0.03	Short-sleeve knit sport shirt	0.17
Men’s briefs	0.04	Short-sleeve dress shirt	0.19
T-shirt	0.08	Long-sleeve dress shirt	0.25
Half slip	0.14	Long-sleeve flannel shirt	0.34
Long underwear bottoms	0.15	Long-sleeve sweatshirt	0.34
Full slip	0.16	Dress and Skirts
Long underwear top	0.20	Skirt (thin) mm	0.14
Foot wear		Sleeveless, scoop neck (thin)	0.23
Ankle-length athletic socks	0.02	Short-sleeve shirtdress (thin)	0.29
stockings	0.02	Long-sleeve shirtdress (thin)	0.33
Sandals/thongs	0.02	Suit Jackets and Vests
Shoes	0.02	Sleeveless vest (thin)	0.10
Slippers (quilted, pile lined)	0.03	Single-breasted (thin)	0.36
Trousers and Coveralls		Double-breasted (thin)	0.42
Short shorts	0.06	Sleepwear and Robes
Walking shorts	0.08	Sleeveless short gown (thin)	0.18
Straight trousers (thin)	0.15	Sleeveless long gown (thin)	0.20
Sweatpants	0.28	Short-sleeve hospital gown	0.31
Overalls	0.30	Short-sleeve short robe (thin)	0.34
Coveralls	0.49	Short-sleeve pajamas (thin)	0.42

was used to determine a full clothing ensemble (I_cl) using Equation (1) [47], and insulation was added to the chair when the respondent was seated. For net chairs, metal chairs, and wooden side-arm chairs, insulation was 0 clo, while for wooden stools and standard office chairs, it was 0.01 clo. When it came to executive chairs, insulation was 0.15 clo [45].

$$I_{\mathrm{cl}}=0.161+0.835\sum I_{\mathrm{cl},\mathrm{i}}$$

(1)

Indoor thermal environmental parameters, such as air temperature, relative humidity, globe temperature, and air velocity, were measured with a portable Delta HD32.3 instrument. The instrument included a thermo-hygrometer, a black globe thermometer, and an anemometer (refer to Figure 7a). The instruments met the measurement range and accuracy standards specified in ISO 7726 [48] (see

Table 7. Technical parameters of sensors of the instrument.

Sensors	Type	Measured Range	Accuracy
Thermos-hygrometer	HP3217.2R	Air temperature: −40 °C–100 °C Relative humidity: 0–100%	1/3 DIN ±1.5% (0–90%), ±2.0% (90–100%)
Black globe thermometer	TP3276.2	−10–100 °C	1/3 DIN
Anemometer	AP3203.2	0–5 m/s	±0.2 m/s (0–1 m/s) ±0.3 m/s (1–5 m/s)

). The instrument was positioned at a height of 1.1 m and was within 1 m of the respondent. Simultaneous recordings of thermal environmental parameters were taken during the questionnaire survey (see Figure 7b). The indoor operative temperature index (t_op) was utilized to examine the connection between the thermal environment and human thermal responses. The value of this index can be computed using the methods outlined in ASHRAE 55 [45] and ISO 7726 [48].

3. Results

3.1. Thermal Environment

During the summer period surveyed in this study, indoor air temperature (t_a), relative humidity (RH), globe temperature (t_g), and air velocity (v_a) were measured and recorded, and 230 valid data sets were obtained.

Table 8. Summary of indoor thermal environmental parameters in survey period.

Parameters	Min.	Max.	Mean	S.D.
ta/°C	23.0	35.8	27.57	3.99
RH/%	31.4	100.0	66.88	20.3
tg/°C	22.9	38.3	28.20	4.55
va/m/s	0.01	1.28	0.2	0.22

presents the statistical analysis of each parameter’s minimum, maximum, mean, and standard deviation, and their distribution is illustrated in Figure 8. The study findings suggest that the indoor air temperatures varied between 23.0–35.8 °C, with relative humidity ranging from 31.4–100%. The globe temperature ranged from 22.9–38.3 °C, and indoor air velocity ranged from 0.01–1.28 m/s.

3.2. Clothing Adaptation

Figure 9a shows the distribution of clothing insulation (I_cl) during the surveyed period, with a mean value of 0.50 clo ranging from 0.33 to 0.95 clo. The indoor operative temperature (t_op) was binned every 1 °C, and the mean value of I_cl was calculated for each 1 °C bin. Figure 9b illustrates the variations of I_cl with t_op, and the linear regression equations of I_cl and t_op are presented in Equation (2). Note that sets of data with less than five votes in each 1 °C bin are excluded from the linear fit as they may impact the accuracy of the results. The slope of linear regression equation is −0.01 with R² = 0.82, indicating that there is a strong linear correlation between clothing insulation and operative temperature, and the clothing insulation decreases by 0.01 clo when the indoor operative temperature increases by 1 °C.

I_cl = −0.01 t_op + 0.77, R² = 0.82

(2)

3.3. TSV, TA, and TP

The statistics show that thermal sensation of “neutral” was the most common response at 51.3%, followed by “hot” (21.3%), “slightly warm” (13.0%), “slightly cool” (7.4%), and “warm” (7.0%). A thermal acceptability of “very acceptable” was the most common response, accounting for 38.7%, followed by “just acceptable” (35.2%), “just unacceptable” (19.6%), and “very unacceptable” (6.5%). A thermal preference of “want cooler” was the most common response at 48.3%, followed by “no change” (43.9%) and “want warmer” (7.8%). Indoor operative temperature (t_op) was binned every 1 °C; the mean value of t_op and TSV in each 1 °C were calculated (bins of data with less than five votes in each 1 °C are excluded as they may impact the accuracy of the results). Regression of TSV and t_op shows a significant linear relationship presented in Figure 10a; the linear regression equation was developed as Equation (3). The equation’s slope indicates the sensitivity of thermal sensation to temperature shifts, which increases by 0.16 when the indoor operative temperature increases by 1 °C. The temperature at which the majority of individuals feel neither hot nor cold is defined as neutral temperature, the value of which will be calculated based on the regression equation (let TSV = 0 in the equation). Thus, the neutral temperature for elderly rural residents in Xi’an during summer is estimated to be 23.9 °C.

TSV = 0.16 t_op − 3.83, R² = 0.65

(3)

The “just unacceptable (−1)” and “very unacceptable (−2)” votes were considered unacceptable, the percentage of which indicates people’s dissatisfaction level with the thermal environment. The percentage of dissatisfied (PD) and the mean value of t_op in every 1 °C were calculated, and then PD was plotted against t_op as illustrated in Figure 10b; the regression equation was developed as Equation (4). When PD = 20%, the upper limit of the acceptable temperature range of 80% can be obtained as 29.3 °C.

PD = 0.0038 t_op² − 0.19t_op + 2.58, R² = 0.75

(4)

Greater discrepancies between indoor and neutral temperatures suggest a stronger preference for a more comfortable temperature among individuals [3]. Similarly, the indoor operative temperature is binned by 1 °C; the percentages of “want warmer (+1)”, “no change (0)”, and “want cooler (−1)” votes were calculated. The statistical result of the thermal preference ratio is illustrated in Figure 10c. During the survey period, the indoor operative temperature ranging from 23 °C to 30 °C showed a mean value of 61.0% of elderly rural respondents voting for the “no change (0)” option. This indicates a strong preference to maintain the current thermal environment. However, as indoor temperatures increased beyond 30 °C, the percentage of elderly respondents who desired a cooler thermal environment rose from 71.4% to 100%, with a mean value of 86.5%. Between 23 °C and 27 °C, a small percentage of participants voted “Want warmer (+1)”, ranging from 12.0% to 14.0%.

3.4. HSV, HA, and HP

The data shows that the most frequent response to humidity sensation was “neutral” at a rate of 49.1%, followed by “slightly humid” (19.1%), “slightly dry” (18.7%), “humid” (6.2%), “dry” (5.2%), and “very humid” (1.7%). A humidity acceptability of “just acceptable” was the most common response, accounting for 42.6%, followed by “very acceptable”, “just unacceptable”, and “very unacceptable” of 33.9%, 21.8%, and 1.7%, respectively. A humidity preference of “want drier” was the most common response at 40.0%, followed by “no change” (37.8%) and “want more humid” (22.2%). In view of the high linear correlation between indoor relative humidity and air temperature (Figure 11) and the measurable impact on occupant thermal sensation [49], indoor air temperature (t_a) was binned every 1 °C, the mean value of RH and HSV in each 1 °C were calculated (bins of data with less than five votes in each 1 °C are excluded). Regression of HSV and RH shows a significant linear relationship illustrated in Figure 12a; the linear regression equation was obtained as Equation (5). The equation’s slope indicates the sensitivity of humidity sensation to relative humidity shifts, which decreases by 0.03 when the indoor relative humidity increases by 1%. The relative humidity at which the majority of occupants feel neither dry nor humid is defined as neutral relative humidity, the value of which will be calculated based on the regression equation (let HSV = 0 in the equation). In conclusion, the neutral relative humidity for rural elderly individuals during summer in Xi’an is calculated to be 63.3%.

HSV = −0.03 RH + 1.77, R² = 0.69

(5)

The votes of “just unacceptable (−1)” and “very unacceptable (−2)” were considered as opinions indicating an unacceptable level of environmental humidity. The percentage of these votes reflects the respondents’ dissatisfaction with the humidity. The dissatisfied percentage (PD) was calculated along with the mean value of relative humidity (RH) for each 1 °C. PD was then plotted against RH, as shown in Figure 12b, and a regression equation (Equation (6)) was developed. When PD is equal to 20%, the lower and upper limits of the acceptable range for a relative humidity of 80% can be calculated as 44.3% and 81.0%, respectively.

PD = 0.0005 RH² − 0.0588 RH + 1.8846, R² = 0.90

(6)

The larger the disparity between indoor relative humidity and the neutral humidity, the stronger the occupants’ desire for the ambient humidity to change. The indoor relative humidity was binned by 5%, and the percentages of votes for “want drier (+1)”, “no change (0)”, and “want more humid (−1)” were calculated within each humidity interval. The statistical outcome of the preference ratio is depicted in Figure 12c. During the surveyed period, when the indoor relative humidity is between 45% and 50%, the ratio of the “no change (0)” option is the highest with a value of 71.4%. The percentage of votes for “want drier (+1)” increases rapidly when the relative humidity exceeds 60%, with an average of 57.3%. When the indoor relative humidity is between 30% and 70%, the percentage of respondents who vote “want more humid” reaches an average of 45.6%.

3.5. ASV, AA, and AP

The statistics show that an ASV of “just right” was the most common response at 51.3%, followed by “slightly weak” (33.5%), “slightly strong” (9.1%) and, “very weak” (2.6%). An AA of “just acceptable” was the most common response, accounting for 47.4%, followed by “very acceptable” (39.6%), “just unacceptable” (10.4%), and “very unacceptable” (2.6%). An AP of “no change” was the most common response at 53.5%, followed by “want stronger” (41.3%) and “want weaker” (5.2%). According to the data, rural elderly people in Xi’an expressed a high level of satisfaction with the indoor air movement during the surveyed summer period. The air velocity (v_a) was used to describe the state of indoor air movement. It was binned every 0.05 m/s, and the mean v_a and ASV values for each 0.05 m/s were calculated. Data bins with less than five votes per 0.05 m/s were excluded as they could affect the accuracy of the results. The regression analysis of ASV and v_a showed a significant linear relationship, demonstrated in Figure 13a. The linear regression equation was developed as Equation (7). The slope of the equation indicates the sensitivity of ASV to air velocity, which increases by 2.54 when the indoor air velocity increases by 1 m/s. The neutral air velocity is the point at which the majority of individuals feel neither strong nor weak airflow. This value is calculated based on the regression equation (by letting ASV = 0 in the equation). As a result, it can be determined that the neutral air velocity for rural elderly individuals in summer in Xi’an is 0.28 m/s.

ASV = 2.54 v_a − 0.72, R² = 0.83

(7)

The votes marked as “just unacceptable (−1)” and “very unacceptable (−2)” were deemed not acceptable. The percentage of these votes indicates the level of dissatisfaction people have with the air movement. The percentage of people who are dissatisfied (PD) and the mean value of air velocity (v_a) in increments of 0.05 m/s were calculated. These values were then plotted against each other, as shown in Figure 13b, and a regression equation was developed as Equation (8). When the percentage of dissatisfied (PD) is at 10%, the lower limit of air velocity range that is considered acceptable by 90% of people is 0.16 m/s.

PD = 1.0968 v_a ² − 1.0333 v_a + 0.2389, R² = 0.86

(8)

Similarly to the previous statement, when there is a greater difference between indoor air velocity and neutral air velocity, occupants will have a stronger desire for a change. The indoor air velocity was divided into intervals of 0.05 m/s. Afterward, the percentage of votes for “want stronger (+1)”, “no change (0)”, and “want weaker (−1)” were calculated for each interval. The statistical result of the preference ratio is shown in Figure 13c. During the survey period, the ratio for the option “no change” was highest at 100% when the indoor air velocity was between 0.3 and 0.35 m/s. The average proportion of “no change” options across all intervals was 56.4%. The elderly participants had a preference for stronger air speed across almost all intervals, with a mean ratio of 39.8%. Conversely, only about 3.9% of participants preferred weaker air movement.

4. Discussion

4.1. Indoor Thermal Environment Evaluation

The neutral temperatures (t_n) of all participants were determined using the Griffiths method [50] in this study. The method involves calculating t_n = t_op – TSV/G, where G is Griffiths constant, which is suggested to be 0.5 by Nicol and Humphreys [51]. According to the international standard ASHRAE 55 [45] and the Chinese standard GB 50785 [52], the residential thermal environment level of the elderly in rural Xi’an was evaluated. Figure 14 depicts the distribution of neutral temperatures of elderly individuals residing in rural Xi’an during the study period, and the comparison with the comfort ranges in ASHRAE 55 and GB 50785. As depicted in Figure 14, the temperature range that provides comfort to rural elderly people in Xi’an during summer months exceeds the upper and lower boundaries of the 80% comfort zone set out by the ASHRAE 55 standard. Similarly, the comfortable temperature range also exceeds the upper and lower limit of the 80% comfort zone specified in the GB 50785 standard. This means that older people in rural Xi’an have a higher upper limit of the temperature range in which they feel comfortable during the summer. Therefore, the use of both standards may be biased when evaluating the indoor thermal environment of dwelling houses inhabited by elderly people in rural Xi’an, as they do not accurately reflect the thermal adaptation level of this population.

4.2. Adaptive Model and Comparison with Other Results

The findings from the previous section demonstrate the necessity of establishing appropriate thermal comfort evaluation criteria, particularly for older adults residing in rural Xi’an. A crucial step in achieving this is to first create an adaptive thermal comfort model. The mean neutral temperatures obtained for each surveyed day was calculated and correlated with the mean outdoor temperatures for that day to derive an adaptive thermal comfort model, as illustrated in Figure 15. Its equation, along with those of other representative models, is summarized in

Table 9. Comparisons with some representative adaptive models.

Adaptive Models		Sites and Climate Zones	References
ASHRAE 55	tn = 0.31 tout + 17.8	9 countries in Europe, Asia, North America, Australia, etc.	[45]
EN 15251	tn = 0.33 trm + 18.8	5 countries in Europe	[53]
GB 50785	tn = 0.82 trm + 7.40	Harbin, Beijing and Xi’an in China, SC and C	[52]
Spanish elderly in nursing homes (NV)	tn = 0.26 trm + 18.83	Spain, Csa	[24]
Shanghai urban elderly in summer	tn = 0.418 trm + 15.96	Shanghai, China, HSCW	[12]
Xi’an urban elderly in summer	tn = 0.41 tout + 15.41	Xi’an, China, C	[3]
Xi’an rural elderly in summer (Current study)	tn = 0.638 trm + 11.15	Xi’an, China, C

Note: t_n: neutral temperature/°C; t_out: mean outdoor temperature/°C; t_rm: outdoor running mean air temperature/°C; Csa: Temperate, dry summer, hot summer (Köppen–Geiger); SC: severe cold climate zone; C: cold climate zone in China; HSCW: hot summer and cold winter in China; NV: natural ventilation.

. The slope of the adaptive model equation, which indicates how sensitive human comfort is to outdoor climate shifts, is 0.638 for rural elderly individuals in Xi’an. This implies that a 1 °C change in outdoor temperature results in a 0.638 °C change in neutral temperature, indicating a greater sensitivity compared to the other findings. From Figure 15, it is evident that the comfort temperature of elderly individuals residing in rural Xi’an is very close to some other results during low average outdoor temperatures in summer, e.g., compared to the results of ASHRAE 55, GB 50785, Urban Shanghai, and Spanish models; whereas the value during high average outdoor temperatures is higher compared to the results of EN 15251, Urban Shanghai, ASHRAE 55, and Spanish models.

4.3. Reasons for Differences in Thermal Adaptability

First, there are variances in the thermal adaptability among disparate age cohorts. Due to the elderly’s lower metabolic rate and sedentary lifestyle, the TSV of the elderly is slightly lower than that of young people [54,55], and they are more tolerant to the thermal environment [7]. Therefore, most elderly individuals prefer higher temperatures than the young by a range of 0.7–2.4 °C in summer [9,22,23,24,25,35,56]. However, in the standards of ASHRAE 55, EN 15251, and GB 50785, the thermal adaptive models were developed by including a large sample size of the youth population. No age-specific results lead to a notable difference between the thermal comfort demands of the elderly, calculated specifically for the elderly, and the standards mentioned above, and this difference is also reflected in the results of the comparison between the elderly in urban nursing homes in Xi’an [3] and Shanghai [12] and the standards.

Second, diverse climates lead to various adaptations to thermal environments for elderly individuals. As an illustration, Xi’an and Shanghai are typical representative cities of two climate zones, according to China’s climate classification. Although both cities experience hot summers, Shanghai is relatively more humid than Xi’an. This high humidity presents a challenge for elderly individuals who have lower tolerance to high temperature and humidity in Shanghai compared to those in Xi’an. This results in a higher upper limit of the comfort temperature range for older adults in rural Xi’an than those in the Shanghai area for the same high temperature environment [11,12], which is similar to the results from the literature [3].

Finally, rural elderly individuals exhibit better thermal adaptation compared to their urban counterparts. The living habits of older individuals [8], as well as their clothing insulation [13,57] and use of cooling or heating devices [18], were observed to differ depending on the local climate [58,59] and building types [10,29]. According to the theory of thermal adaptation, these factors result in the rural elderly demonstrating distinctive characteristics in psychological adaptation and behavioral adaptation. The rural household’s low income and older people’s frugal money habits in China influence environmental decisions with thermal conditions, e.g., they are less likely to use air conditioning than urban elderly in summer because of expensive running costs. This leads to lower psychological expectations of their environment and results in few expressions of discomfort and dissatisfaction [35]. In addition, differences in behavioral adaptation between rural and urban elderly are also an important reason. Clothing acts as a barrier between the human body and the environment, impacting the exchange of heat and moisture that influences the body’s thermal balance [13]. A clothing adjustment model of the rural elderly in Xi’an was compared with other models in different climate zones [3,7,13,22,25,26], shown in Figure 16 and

Table 10. Some clothing adjustment models in summer of the elderly in different regions.

Clothing Adjustment Models in Summer		Sites/Climate Zones	References
Taiwan urban elderly in residential buildings	Yearly: Icl = –0.07 top + 2.37 (13.8 °C ≤ top ≤ 31.0 °C)	Taiwan, China/HSWW	[7]
Australian elderly in nursing homes (residents not in bed)	Yearly: Icl = –0.0623 top + 2.20 (17 °C ≤ top ≤ 31.5 °C)	South-eastern NSW, Australia/Cfb and Cfa	[25]
Indian urban elderly in residential buildings	Female: Icl = –0.0046 top + 0.9658 Male: Icl = –0.0046 top + 0.5921 (28 °C ≤ top ≤ 35 °C)	Urban Salem, India/WH	[26]
Spanish elderly in nursing homes	Icl = –0.01 top + 0.85 (23.17 °C ≤ top ≤ 28.45 °C)	Spain/Csa	[22]
Shanghai urban elderly in nursing homes	Female: Icl = –0.017 top + 0.969 Male: Icl = –0.025 top + 1.181 (25.6 °C ≤ top ≤ 32.3 °C)	Shanghai, China/HSCW	[13]
Xi’an urban elderly in nursing homes	Icl = –0.003 top + 0.44 (24.7 °C ≤ top ≤ 36.4 °C)	Xi’an, China/C	[3]
Xi’an rural elderly in dwelling houses (current study)	Icl = –0.01 top + 0.77 (23.5 °C ≤ top ≤ 37.5 °C)	Xi’an, China/C

Note: I_cl: clothing insulation/clo; t_op: indoor operative temperature/°C; HSWW: Hot summer and warm winter zone in China; Cfb: Temperate, without dry season, warm summer (Köppen–Geiger); Cfa: Temperate, without dry season, hot summer (Köppen–Geiger); Csa: Temperate, dry summer, hot summer (Köppen–Geiger); HSCW: hot summer and cold winter in China; C: cold climate zone in China.

. From the comparison results, it is evident that the summer clothing insulation levels for elderly individuals in rural Xi’an are higher than those for the elderly in urban Xi’an, male and female elderly in Shanghai, and male elderly in India. The rural elderly individuals in Xi’an also exhibit higher thermal resistance levels than Taiwanese and Australian elderly individuals at temperatures of 27 °C and above. This higher level of clothing insulation and adjustment patterns may also be one of the main reasons why older people in rural Xi’an have lower comfort temperatures during cooler summers, as shown in Figure 15.

4.4. Influence of Humidity and Implications for Thermal Environment Design

With age, the elderly become less sensitive to changes in environment temperature, and the slow response of the body can prevent timely adjustment measures, leading to hypothermia or hyperthermia and specific health risks [3]. In addition to environment temperature, indoor humidity and air movement are two crucial parameters in the thermal environment of residential buildings that have an impact on occupants’ satisfaction with the living environment.

In summer, high humidity levels will not only cause thermal discomfort but also allergic diseases. Exposure to a high humidity environment decreases the rate of sweat evaporation, resulting in less heat dissipation through vaporization and increasing skin discomfort. This phenomenon is known as sweat-induced evaporative cooling, which is the primary physiological process by which the body dispels heat, especially in warm conditions [60,61]. Studies have shown that low relative humidity increases insensible evaporation and leads to decreased skin temperature and a cooler sensation, even at the same temperature [60,62]. However, older participants exhibited lower sensitivity to changes in humidity when compared to younger participants [63], which indicates that the elderly may be at risk of exposure in either high or low humidity environments due to their inability to regulate humidity effectively. In the summertime, high humidity levels raises the potential for heat stroke and certain allergic ailments [63,64].

Therefore, it is crucial to ensure proper indoor humidity control for buildings housing older adults. The current standards of ASHRAE 55 [45] and GB 50785 [52] propose the PMV-PPD model and an adaptive model. Comfort zones are utilized for assessing the indoor thermal environment in HVAC or naturally ventilated buildings. However, in the standards, humidity is simply prescribed as a lower or upper limit, and there is no coupling between humidity and thermal comfort in the adaptive thermal comfort model [65]. For the elderly, it is even more important to figure out how they respond to humidity and what their comfortable humidity needs are. Tao et al. [8], Wu et al. [10], and Sudarsanam et al. [26] conducted studies on this topic separately in Hong Kong, Chongqing, and Salem, where the summers are characterized as extended periods of high humidity and heat. However, their findings revealed no correlation between HSV and relative humidity. It is due to their long-term adaptation to hot and humid climates, rendering them more tolerant to fluctuations in humidity levels [66].

However, a study conducted by Zheng in urban Baoding, a cold climate zone in China, found a significant correlation between HSV and humidity among older adults [20]. In addition, the study revealed that individuals in the 60–74 age group had a neutral humidity level of 65.6%, while those in the 75–88 and 88 and above age groups had levels of 67.2% and 67.7%, respectively. Moreover, the acceptable humidity range was calculated to be 56.5–66.1%, 53.6–74.3%, and 52.0–78.3%, respectively, for the corresponding age groups. In this study, we examined the correlation between indoor relative humidity and subjective responses, including HSV, HA, and HP (Figure 12). Our findings reveal that the acceptable humidity range for elderly individuals in rural Xi’an is 44.34–81.04%, with a neutral relative humidity level of 63.3%. This result is consistent with Zheng’s research [20]. Therefore, when designing the indoor thermal environment for elderly individuals in rural Xi’an, these results can serve as a reliable basis for controlling indoor relative humidity.

4.5. Effects of Air Movement and Implications for Thermal Environment Design

Air movement affects the body’s temperature by influencing evaporative and convective heat exchange [67]. In the summer, air movement creates a cooling effect. Elevated air movement is equivalent to decreased air temperature in terms of thermal comfort [68]. To maintain thermal comfort in the summer, elders commonly adopt the behavioral adaptation strategies of opening windows and using fans. Studies in Australia [25], Hong Kong [8], and Shanghai [11] have demonstrated that elders actively react to environments in this manner. Some studies have been conducted focused on the effects of air speed on thermal comfort and have concluded that the increase in air speed significantly affects the comfortable temperature. For example, a wind speed of 1.0 m/s can raise the comfortable temperature by 2.7 K in Hong Kong [69], while in India, the use of ceiling fans has been found to increase the comfort temperature by 1.8 °C [70]. ASHRAE 55 [45] has established acceptable ranges for operative temperature and average air velocities, wherein 0.2 m/s is designated as the draft perception threshold and 0.8 m/s is identified as the maximum limit when local air speed cannot be regulated.

However, it is unclear whether the limit in ASHRAE 55 is applicable to the elderly in Xi’an, given the absence of information regarding the sample population’s age or the climate zone in which they reside. This study examined the correlation between indoor air movement and three subjective perception indicators—ASH, AA, and AP—and found that the 90% acceptable lower limit of air velocity for elderly individuals in rural Xi’an was 0.16 m/s, with a neutral air velocity of 0.28 m/s. During the investigation period of this research, situations with higher air speeds occurred less frequently (as shown in Figure 8d), which resulted in the inability to calculate the 90% acceptable upper limit of air velocity, as illustrated in Figure 13b. As the physical functions of the elderly decline, they may experience a weakened ability to adapt to the thermal environment and reduced body immunity. Excessive air flow speeds can potentially lead to discomforting symptoms such as colds, respiratory diseases, headaches, and dry eyes. Therefore, the upper limit of the comfortable wind environment needs to be calculated.

Tao [8] found that a majority of respondents reported sufficient ventilation when the indoor air velocity ranged between 0.2 m/s and 0.6 m/s during the summer in Hong Kong. Zheng [15] determined the comprehensive comfort zone of air velocity ranges, which were 0.02–0.31 m/s, 0–0.27 m/s, and 0–0.29 m/s for individuals aged 60–69, 70–89, and 90–99 in Baoding, respectively. For the elderly people in rural Xi’an, the lower limit value of 90% acceptable air velocity was calculated as 0.16 m/s in the previous section, and the corresponding ASV was calculated as −0.31 according to the fitting curve (Figure 13a) and its equations (Equation (7)). According to the principle of symmetry of the 7-degree scale of subjective sensation, letting ASV equal 0.31, the upper limit value of 90% acceptable air velocity can be calculated to be 0.41 m/s. This upper limit value is lower than that of Hong Kong, which reaches 0.6 m/s. The difference could be attributed to Hong Kong’s hot and humid summers, making the elderly there prefer higher air velocity to increase convective heat transfer to maintain comfort. In comparison, the upper limit value among the elderly in Baoding of different age groups reaches 0.29–0.31 m/s, potentially due to differences in behavioral regulation between rural and urban elderly groups. The observed distinctions in the thermal environment regulation strategies adopted by rural and urban elderly may stem from their differential behavioral regulation and psychological adaptation patterns. The rural elderly cohort exhibit a propensity for using energy-efficient electric fans and hand-cranked fans, with an increased tolerance for higher air velocities.

4.6. Limitations and Future Work

The use of electric fans is a very effective and energy-efficient measure to improve occupants’ thermal comfort and is a favored behavioral adaptation by elderly individuals residing in rural areas during the summer. Nonetheless, due to the limited number of questionnaires observed in this study, there was a lack of modeling of the behavioral adaptations of the elderly in rural Xi’an. Further research is needed to address this concern. In addition, the study was exclusively carried out during daytime, and the older adults were alert and could easily adopt adaptive adjustment to maintain their comfort. However, at night, the thermal environment characteristics differ significantly from those during the day while occupants are sleeping. Thus, further research is necessary to understand the subjective responses of older adults to nighttime thermal environments and their comfort needs.

5. Conclusions

This paper reports on an in-depth study into thermal sensations, preferences, and acceptability of rural elderly occupants, as well as humidity sensations, preferences, and acceptability, and air movement sensations, preferences, and acceptability. Based on this, an indoor thermal environment evaluation, adaptive thermal comfort model, reasons for differences in thermal adaptability, along with the comfort zones for temperature, humidity, and air velocity for elderly in rural Xi’an, were analyzed and discussed. The following conclusions were drawn:

• Buildings occupied by elderly people in rural Xi’an are not insulated and are naturally ventilated throughout the year, resulting in a poor indoor thermal environment in summer, with a wide range of fluctuations in each thermal environment parameter. The clothing level of rural elderly in Xi’an during summer is notably high, with an average of 0.50 clo ranging from 0.33–0.95 clo. A clear linear correlation exists between clothing insulation and indoor operative temperature, with the value decreasing by 0.01 clo when indoor operative temperature increases by 1 °C. The overall level of clothing is higher than that of urban elderly groups, and the sensitivity of clothing insulation to temperature change is lower in rural Xi’an. These characteristics are related to the characteristics of the indoor thermal environment of rural buildings, the adaptive capacity and living habits of rural elderly people.
• Significant dynamic correlations exist between nine subjective thermal response indicators and three types of thermal environment parameters for rural elderly people in Xi’an. Linear relationships were observed between sensation indicators (TSV, HSV, ASV) and thermal environment parameters (indoor operative temperature, relative humidity, and air velocity), while quadratic relationships were observed between the acceptability indicators (TA, HA, AA) and thermal environment parameters. The neutral values of operative temperature, relative humidity, and air velocity were calculated to be 23.9 °C, 63.3%, and 0.28 m/s, respectively. Additionally, 80% of the elderly felt comfortable with a maximum temperature of 29.3 °C, while the suitable humidity range was between 44.3% to 81.0%. Correspondingly, the appropriate air velocity range that was comfortable for 90% of the population was between 0.16 m/s to 0.41 m/s. When the ambient temperature exceeds 30 °C, over 86.5% of older adults anticipate cooler environmental conditions. Similarly, when humidity surpasses 60%, an average of 57.3% of older adults expect drier surroundings. The elderly expect the air velocity to remain unchanged and expect it to become higher in almost all conditions, with an average percentage of 56.4% and 39.8%, respectively.
• The comfort temperature of the elderly in summer in rural Xi’an exceed the comfort range boundaries in both the international standard ASHRAE 55 and the Chinese standard GB 50785. The adaptive thermal comfort model shows a significant deviation from other models, manifested in a lower comfort temperature in cooler environments and a greater sensitivity to outdoor air temperature changes, with a sensitivity coefficient of 0.638. This is the outcome of the prolonged adaptability to the regional climate and the poor indoor thermal environment, as evidenced by lower psychological expectations, higher clothing levels, and environmental decisions influenced by household low income.

In conclusion, it is essential to objectively determine the comfort requirements and preferences for indoor air temperature, humidity, and air movement of elderly individuals living in rural areas with low household income. This information will help in developing residential thermal environments that cater to their comfort and health needs, which will improve well-being and mitigate health risks. The findings of this paper can be used as a reference for the design of residential environments for the elderly in other rural areas of China’s cold climate zones. Furthermore, the research methodology and thermal environment evaluation index system are also applicable to other climate zones.