The Role of Street Elements on the Social Activities of the Elderly in Severe Winter Conditions: A Case Study of Harbin, China

Abstract

The phenomenon of global population aging poses considerable mobility challenges for older adults, particularly in cold climate regions, where the accessibility and configuration of street elements exert a significant impact on social participation and safety during severe winter conditions. Employing a combination of non-participatory observation, behavior mapping, and spatial analysis across different winter periods, this study investigates three residential streets in Harbin, China. The research systematically documents the types, frequencies, and spatial distributions of both social activities and street infrastructure utilized by the elderly. Subsequently, kernel density overlays of elderly social activity and street element distributions enable a nuanced analysis of the influence of environmental features on older adults’ social engagement throughout the three delineated winter phases. The findings reveal the following: (1) There is persistent demand for outdoor social interaction among the elderly, with participation rates inversely proportional to the severity of winter, peaking in early winter and declining through late and harsh winter stages; (2) Variations in activity types and durations are closely associated with spatial configurations: dynamic activities are predominantly observed along linear street segments, whereas passive behaviors cluster at intersections and broader street expanses; (3) There are several key aspects of street design and street furniture provision that help to support the use of streets in winter by the elderly. However, the influence of seating and fitness elements on mobile activities is limited. This study contributes to promoting inclusive urban design for older people in cold climates.

1. Introduction

1.1. Social Health Problems of the Elderly

Population aging constitutes one of the most profound demographic challenges of the twenty-first century. According to United Nations projections, the global population is anticipated to reach between 9 and 10 billion by 2050, with approximately 2 billion individuals aged 60 years or older and 1.5 billion aged 65 years or above [1].

The aging process is commonly accompanied by a progressive decline in physical and cognitive capacities, increased susceptibility to chronic diseases, and a heightened risk of social disengagement [2]. The disengagement theory proposed by Cumming and Henry suggests that older people tend gradually to withdraw from active social participation after retirement, leading to reduced social networks and altered interaction patterns [3]. Research has shown that social isolation significantly impairs the health of older adults, correlating with elevated risks of anxiety, depression, and suicidal ideation [4]. These challenges are further exacerbated in regions with severe cold climates, where harsh winter conditions, including sub-zero temperatures [5], limited daylight, and icy surfaces, significantly restrict outdoor mobility [6]. This restriction is particularly acute for the elderly in severe cold regions of China, where long winters (up to five months) and ground covered with snow pose significant threats to travel safety and drastically shorten the duration of outdoor activities, negatively impacting their quality of life [7]. The confluence of aging and cold-climate living has been extensively explored in international research, providing valuable comparative perspectives. For instance, studies in Canada have examined how winter city planning in areas such as Quebec and Alberta can better accommodate older adults through enhanced snow removal practices, accessible public transportation, and age-friendly architectural design [8] (see

Table 1. Geographic distribution of studies on elderly social activities and built environments.

Climate Zone	Representative Regions/Countries	No. of Studies	Key References
Temperate	Southern China, Italy, UK, Australia	37 (82.2%)	[10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30]
Cold High-Latitude	Canada, Scandinavia, Northeastern China	5 (11.1%)	[8,9,10,13]
Other Climates	Tropical, Arid	3 (6.7%)	[31,32,33]
Total		45

). Scandinavian scholarship, especially from Sweden and Finland, underscores the importance of social policy and community support systems in alleviating seasonal isolation among the elderly [9]. Research focusing on Northeastern China has highlighted the acute vulnerability of older populations in these environments, demonstrating that environmental factors—including the duration of snow cover and inadequate microclimate regulation—substantially influence both health outcomes and patterns of outdoor activity [10]. Furthermore, studies conducted in Harbin have elucidated the relationship between built environment variables (such as sidewalk quality, public transit availability, proximity to parks, outdoor seating, and vegetation types) and subjective satisfaction among the elderly, offering critical guidance for environmental design in cold regions [7]. A robust body of evidence indicates that sustained social engagement markedly enhances quality of life [11], enhances emotional resilience, and promotes holistic health outcomes among aging populations [12]. Consequently, optimizing the design of social spaces in urban environments with cold climates represents a critical intervention strategy to support the well-being of the elderly [13].

1.2. The Advantages and Necessity of Promoting Streets

According to the World Health Organization (WHO), streets should not be regarded merely as outdoor spaces or conduits for vehicular and pedestrian movement within urban environments. Instead, they function as multifunctional urban spaces that extend beyond their traditional role as transportation corridors [13], serving as vital settings for fostering a supportive and inclusive community for older adults [34]. A well-designed street environment can significantly facilitate social engagement among the elderly [35], providing restorative experiences beyond domestic environments [14] and improving physical and mental well-being [15]. Notably, the elderly show a preference for integrated urban spaces, known as living streets [36], which combine residential, commercial, and recreational functions and align with principles of healthy urban design [16]. In order to create an elderly-friendly living street, Yang proposed a structured optimization strategy of built environments from four aspects: space configuration, place shaping, path accessibility, and function induction [17].

A substantial body of research has established that social interactions exert both direct and indirect influences on the physical and mental health of older adults [37]. Nevertheless, street design in cold regions continues to encounter persistent systemic challenges, including inadequate infrastructure for urban functionality [38], insufficient pedestrian-oriented amenities [31], and underdeveloped cultural–aesthetic elements. While existing research underscores the health benefits of social interaction for the elderly, the majority of studies are concentrated in temperate zones, with limited empirical investigation into high-latitude cities. Although prior studies have highlighted mobility challenges faced by the elderly in cold-climate cities such as Montreal, Stockholm, and Sapporo, few have systematically examined the micro-scale role of street elements across different phases of winter severity. Integrating insights from winter urbanism studies [39] could offer valuable comparative perspectives for inclusive street design in extreme cold climates. Addressing this research gap is essential for formulating evidence-based urban design strategies that are specifically responsive to the unique challenges posed by cold climate contexts [39].

1.3. The Influence of Street Elements on the Social Activities of the Elderly

Street elements, including their density, diversity [31], presence of vegetation [39], and walkability [18], have been shown to significantly influence health-related behaviors among older adults. These elements support physical health, psychological well-being, and social integration [19], particularly when public service allocations align with the psychosocial needs of older adults through demand-driven distribution and quality assurance [20]. Notably, the design and configuration of street elements directly modulate the extent of older adults’ participation in social activities [21], a cornerstone of active aging strategies [22].

However, existing research predominantly focuses on isolated factors, such as transport accessibility [32], medical service coverage [33], or walkability indices [23]. While such studies assess the rationality of element layouts and propose configuration standards [24], they largely overlook the synergistic effects of integrated public service elements on meeting daily social needs [25].

This study seeks to address this research gap by systematically examining the influence of various street elements on the social activities of the elderly during different stages of winter in Harbin, a city located in a cold region of China. The research is guided by the following questions:

• Are seating and fitness elements associated with higher levels of social activities among the elderly?
• In what ways do road traffic elements impact the social participation of the elderly?
• How does street vegetation affect social activities among the elderly in winter?

2. Materials and Methods

2.1. Study Sites

Harbin, the capital of Heilongjiang Province, is a prominent urban center situated within China’s severe cold climate zone. The city is characterized by marked seasonal temperature variation, with average summer temperatures reaching approximately 25 °C and winter temperatures averaging around −15 °C; notably, the winter season can persist for up to five months [26]. The challenge of population aging in Harbin is particularly acute. According to the results of China’s seventh national census, there are 2.2 million permanent residents aged 60 and above in Harbin, accounting for 21.98% of the population. This proportion has increased by 9.21% since the previous census, far exceeding the national average.

Three residential streets in Harbin—Pinggong Street, Fanrong Street, and Anshun Street—were selected as survey sites, each with comparable total lengths, as illustrated in Figure 1. These streets serve as primary residential thoroughfares in the city, featuring similar proportions of elderly residents but marked variation in the types of street elements present. The selection criteria for these residential streets were as follows: (1) Each street primarily comprises urban secondary roads and branches, with comparable total lengths, thereby providing a relatively quiet and safe environment conducive to the social activities of older adults; (2) The street interfaces are densely lined with residential buildings, with similar population densities and proportions of elderly residents, thus representing the typical travel behaviors and spatial patterns of Harbin’s elderly population; (3) The types of street elements are significantly different, carrying diverse social activities for the elderly.

To ensure both the homogeneity and integrity of the sample street sections, each street was subdivided into segments ranging from 150 to 250 m for detailed analysis. Pinggong Street was divided into six segments, with street lights distributed along the south side at 35 m intervals, while the north side predominantly features street trees and traffic-related elements. Fanrong Street was segmented into twelve sections; its north side is characterized by the presence of street trees and lighting (with lights spaced at 35 m intervals), whereas the south side contains more greenery and road traffic elements. Anshun Street was divided into thirteen segments, with both sides of the street containing relatively few elements, primarily planter boxes, street lights, and road signs; street lighting is mainly distributed on the south side at 37 m intervals. The spatial layout of these three streets is depicted in Figure 2, and basic information is depicted in

Table 2. Basic information of sample streets.

Streets	Pinggong Street	Fanrong Street	Anshun Street
Administrative division	Nangang district	Nangang district	Daoli district
Lane characteristics	Single lane	Double lane	Double lane
Street length	750 m (6 street segments)	1000 m (12 street segments)	950 m (13 street segments)
Function distribution	The north side of the street is mainly residential, and the south side of the street is mainly commercial.	Residential areas are mainly distributed among schools and businesses.	Mainly residential.

.

2.2. Study Procedure

In this study, individuals aged 60 years or older were classified as elderly. Street elements were categorized into three principal categories: rest fitness elements, road traffic elements, and green landscape elements (see

Table 3. Definition and examples of three types of street elements.

Street Elements	Definition	Example
Rest fitness elements	They refer to elements such as chair corridors for people to rest, communicate, and watch in densely populated blocks, such as commercial living blocks.	Seats, fitness equipment, corridors, etc.
Road traffic elements	They refer to the furniture on the street used to organize traffic, ranging from parking lots and pedestrian overpasses to bicycle parking spots.	Bus stations, subway stations, overpasses, signal lights, etc.
Green landscape elements	They refer to street landscape elements that can create urban cultural atmosphere for people in addition to meeting basic needs.	Street trees, shrubs, landscape sketches, art paving, flowerbeds, sculptures, fountains, stage venues, etc.

). Based on field observations and established typologies of social behavior, this article emphasizes social activities that enhance interpersonal relationships [27], invigorate neighborhood vitality, and foster a supportive social atmosphere. These behavioral activities were divided into two primary categories: mobile social interaction and stationary social interaction. The classification of social activities is grounded in public space interaction theory.

Specifically, following Mehta’s [10] conceptualization of passive social behavior and Gehl’s [28] framework regarding the intensity of social contact, stationary social activities were further subdivided into three types:

• Passive Social Activities: Non-verbal engagements enabling social learning through environmental co-presence [10].
• Short Social Activities: Brief exchanges during necessary acts (e.g., greetings while shopping), corresponding to Gehl’s ‘transient interactions’.
• Continuous Social Activities: Sustained engagements requiring active participation (e.g., group conversations). Mobile activities (e.g., talking while walking) are analyzed separately due to their distinct spatial dynamics (see

  Table 4. Definition and examples of four types of social activities.

  Types of Social Activities	Definition	Example
  Mobile social activities	Activities on foot for transport, generally with walking as the main focus.	Talking and playing with parents, friends, and neighbors during commuting, walking, jogging, and running.
  Continuous social activities	Communication activity of low contact intensity with others, involving generally no communication through language and action.	Resting alone, stopping to look (at merchandise, billboards, other people, or window displays), playing musical instruments, humming, eating, reading, working, using mobile phones, etc.
  Short social activities	Communication activity of medium contact intensity, involving static and moderate direct communication with others and short duration activities.	Greetings, nodding to people, short conversations caused by accidental encounters with neighbors, and chatting between shopkeepers and customers during shopping.
  Passive social activities	Communication activity with high contact intensity, which occurs in the process of dynamic walking and static presence in the street. Lots of talking and activities with a longer duration	Many people having parallel conversations, neighbors gathered to sit and talk or to play chess and card activities, group business activities, square dancing, exercise, eating, or street performances.

  ).

On-site behavior mapping and systematic counting methods were employed in this study, supplemented by high-definition satellite imagery and Baidu Street View data, to document both the characteristics of elderly social activities and the existing conditions of street elements across the three selected streets. Utilizing ArcGIS kernel density analysis in conjunction with field survey data and questionnaire results, this study explored the spatial and temporal relationships between street elements and social activities across the sampled streets, capturing the daily activities of thousands of older adults. Based on these findings, a street facility update strategy tailored to the social interaction needs of the elderly was proposed.

2.3. Study Measures

Through non-participatory observation and behavior mapping, the number, spatial, and temporal distribution characteristics of the four types of social activities of the elderly were recorded, and the characteristics and differences were analyzed and summarized.

According to the meteorological service, an average temperature continuously below 10 °C for 5 days is defined as winter, and the cold levels are divided into eight grades. Based on winter temperature and duration in Harbin, we categorized the winter season into three distinct stages: early winter, severe winter, and late winter (see

Table 5. The definition and examples of winter stages.

Winter Stage	Cold Levels	Definition
Severe winter (December, January, February)	Grade 1	“extreme cold,” with temperatures below −40 °C
	Grade 2	“severe cold,” with temperatures ranging from −30 °C to −39.9 °C
Late winter (March)	Grade 3	“cold,” with temperatures ranging from −20 °C to −29.9 °C
	Grade 4	“great cold,” with temperatures ranging from −10 °C to −19.9 °C
	Grade 5	“minor cold,” with temperatures ranging from −5 °C to −9.9 °C
Early winter (November)	Grade 6	“light cold,” with temperatures ranging from 0 °C to −4.9 °C
	Grade 7	“slight cold,” with temperatures ranging from 0 °C to 4.9 °C
	Grade 8	“cool,” with temperatures ranging from 5 °C to 9.9 °C

Note: Observed temperatures during data collection aligned with stage definitions. Early winter: −4.2 °C to 8.1 °C (Grades 6–8); severe winter: −28.5 °C to −37.3 °C (Grades 1–3); late winter: −12.4 °C to −18.9 °C (Grades 4–5).

). To reduce the impact of random factors such as sudden, short, extreme weather, two days were randomly selected from each stage for data collection, accounting for behavioral differences between weekdays and weekends, to reflect work and school schedules.

Temporal representativity was ensured through the following methodological strategies:

• Stage-aligned sampling: Observation days were randomly selected within each meteorologically defined winter stage, thereby capturing the typical temperature range for each period (from −5 °C to −40 °C).
• Weekday/weekend balance: For each stage, one weekday and one weekend day were sampled to account for variations in routine and social activities.
• Exclusion criteria: Days with extreme weather (e.g., blizzards and visibility < 100 m) or public holidays (e.g., Spring Festival) were excluded, as these disproportionately alter behavior patterns.

Utilizing the street satellite map as a base, the activities of the elderly were recorded by behavior annotation to form point data. Symbols and colours were used to capture information such as gender and the type of social activities of the elderly. Each data collection stage consisted of a two-hour observation period, conducted within the timeframe of 07:00 to 21:00. Data collection involved the researcher standing at predetermined locations within each street segment, marking observed individuals and their corresponding activities during the designated observation intervals, and documenting activity scenes through photography. In total, 6841 data points were recorded, comprising 2157 from Pinggong Street, 2696 from Fanrong Street, and 1988 from Anshun Street (see

Table 6. Observation summary of Pinggong Street.

	Early Winter Stage	Severe Winter Stage	Late Winter Stage
7:00–9:00	175	90	84
9:00–11:00	168	67	86
11:00–13:00	237	54	120
13:00–15:00	264	65	125
15:00–17:00	221	77	124
17:00–19:00	111	99	78
19:00–21:00	58	43	81
Total	1234	495	698

,

Table 7. Observation summary of Fanrong street.

	Early Winter Stage	Severe Winter Stage	Late Winter Stage
7:00–9:00	64	58	116
9:00–11:00	93	87	148
11:00–13:00	163	122	141
13:00–15:00	168	156	186
15:00–17:00	118	106	258
17:00–19:00	90	70	108
19:00–21:00	72	32	70
Total	768	631	1027

and

Table 8. Observation summary of Anshun Street.

	Early Winter Stage	Severe Winter Stage	Late Winter Stage
7:00–9:00	244	120	119
9:00–11:00	187	59	74
11:00–13:00	120	40	69
13:00–15:00	123	61	99
15:00–17:00	82	90	110
17:00–19:00	87	54	86
19:00–21:00	57	35	66
Total	900	459	623

). The annotated data were subsequently entered into ArcGIS as a series of spatially referenced layers for further analysis.

In order to unobtrusively observe and document the unconscious activity behaviors of the subjects, the research protocol involved informing participants of the anonymous recording and photography after data collection had concluded and obtaining their informed consent retrospectively. Preliminary identification of elderly individuals was conducted based on observable indicators such as walking posture, facial characteristics, and corroborating information from social surveys. After the experiment, when informing the subjects of their right to know, it was confirmed that they were over 60 years old, which falls within the conceptual scope of the research subjects. Their facial information and other identifiable information are not displayed in the research results. The records of gender and age are anonymous. The photos are displayed in the paper, and no further photos are retained.

2.4. Data Analysis

2.4.1. Spatial Density Analysis

Field-collected behavioral annotation data, including activity type, time, and coordinates, were digitized into structured Excel tables. Using ArcGIS 10.8, coordinates were converted into point layers with consistent spatial references. Attribute fields (activity type, element type, gender, and observation time) were added to each point, generating social activity and element point layers for three winter stages (early, severe, and late). Kernel density analysis was used to visualize the social activity of the elderly and street element densities through spatial mapping.

Kernel density mapping identifies hotspots by analyzing spatial distribution densities. Activity point layers (categorized by four activity types) and element point layers (categorized by three element types) were processed separately. The Silverman criterion was applied, which optimizes bandwidth to 50 m, balancing resolution and smoothness, with a 1 m × 1 m output grid. Heat maps display density gradients from red (high) to green (low), with red and blue dots representing female and male participants, respectively, as shown in Equation (1).

Density calculation formula:

$$\lambda (s)={\sum }_{i=1}^n{\displaystyle \frac{1}{h^2}}K\left({\displaystyle \frac{d_i}{h}}\right)$$

(1)

λ(s): Kernel density estimate at location s.

K: Kernel function (Gaussian kernel is used in this study).

h: Bandwidth (50 m), optimized via Silverman’s rule of thumb for spatial point patterns.

d_i: Euclidean distance from the observed activity point i to location s.

n: Total number of observed activity points within the bandwidth.

2.4.2. Overlapping Analysis

Overlapping analysis shows the fitting characteristics of street elements and elderly activities. To quantify the spatial fit between the distribution of elements and social activities and to standardize the grid, the activity density (Dactivity) and element density grid (Delement) were unified to the same spatial range and resolution, as shown in Equation (2).

The ArcGIS grid calculator was used to perform pixel-by-pixel difference calculation as follows:

ΔD = ∣D_activity − D_facility∣

(2)

D_activity: Normalized kernel density of social activities (rescaled to 0–1 range).

D_facility: Normalized kernel density of street elements (rescaled to 0–1 range).

ΔD: Absolute difference between normalized densities.

High fit: ΔD ≤ 0.1ΔD ≤ 0.1 (i.e., ≤10% of max density).

Low fit: ΔD ≥ 0.3ΔD ≥ 0.3 (i.e., ≥30% of max density).

The absolute difference between the kernel density values of elderly social activities (mobile, continuous, short, and passive) and the kernel density values of street elements (rest fitness, road traffic, and green landscape) was calculated to obtain ΔD. A smaller ΔD value (ΔD ≤ 10% of the maximum kernel density) indicates a higher degree of spatial fit between the distribution of activities and street elements. Conversely, a larger ΔD value (ΔD ≥ 30% of the maximum kernel density) suggests a low degree of fit, highlighting the need to enhance the provision of street elements in those areas.

Furthermore, with a larger ΔD value (ΔD ≥ 30% of the maximum kernel density), it is possible to assess whether the quantity of street elements in a given area is appropriate by analyzing the kernel density of elderly social activities. For example, a low kernel density of elderly social activities may indicate an oversaturation of street elements. On the contrary, a high kernel density of elderly social activities may indicate that the street elements cannot meet the social activity requirements. However, it still needs to be statistically analyzed based on the usage of street elements by elderly social activities to demonstrate the supporting relationship between the two.

Using the ArcGIS grid calculator, spatial fit difference maps (ΔD maps) were generated to visualize the degree of alignment between the distribution of street elements and elderly social activities under varying conditions. In total, 12 distinct ΔD maps were produced, corresponding to combinations of three winter stages (early, severe, and late), two genders (male and female), and two analysis categories (see the results maps below). On these maps, grid cells are color-coded using a gradient from yellow to blue, where yellow indicates areas of low spatial fit (i.e., high ΔD values, indicating a significant mismatch between the densities of social activities and street elements), and blue indicates areas of high spatial fit (low ΔD values, signifying close alignment).

Kernel density analysis effectively visualizes the spatial clustering and relative density of both the observed elderly social activities and the street elements. This enables the analysis of the spatial–temporal relationships between them [28].

3. Results

The findings reveal how the quantity, distribution, and type of street elements impact elderly social interaction activities and compare changes in these activities across early, severe, and late winter stages.

Field observations reveal that elderly individuals consistently engaged in mobile, passive, and short social activities across all street samples; however, the prevalence of continuous social activities differed among streets (see

Table 9. Social activities of the elderly in the sample streets.

	Pinggong Street	Fanrong Street	Anshun Street
Mobile social activities	Friends talk and walk	Walking hand in hand	Talking with friends and family, taking a walk, shopping, and returning
Continuous social activities	Talking while waiting for the cars to pass, watching the loading and unloading of goods, watching snow shoveling	Waiting for cars to chat, square dancing, chess and card games, fitness	Chatting while waiting for the car, watching loading and unloading goods, watching snow shoveling, outdoor conversation, multiple people chatting in parallel
Short social activities	Shopping, child walking, pet walking, waste collection, express delivery communication	Strolling with children, walking with pets, greeting, and chatting by chance	Shopping transactions, greeting, casual conversations, asking for directions, walking with children, walking with pets, collecting waste, picking up express delivery for communication
Passive social activities	Shop owners observing, observing while waiting, waiting for the bus, sweeping snow	Sitting in the sun, waiting for the bus	Singing alone, sitting and resting to observe, stopping to look at small stalls, answering and making phone calls, stopping to look at mobile phones, waiting for buses

). For example, regular activities like family conversations occur in all streets, while special activities like chess, card games, and square dancing are limited to a few streets. This indicates that continuous social activities are highly dependent on street elements such as steps, flower planters, trees, barriers, fitness equipment, public seats, and trash cans, with public seats being the most frequently used.

The gathering locations and types of social activities reveal different patterns, according to space type:

• Corner spaces with rest and fitness elements support continuous social activities such as square dancing and chess, using seats, cushions (brought by the users), trash cans, and tables.
• Intersections serve as sites for passive social activities, such as waiting at traffic lights.
• Residential entrances/exits see passive social activities like sitting and resting and continuous social activities like walking together.
• Spaces with vegetation, such as trees, pavilions, seats, and flower planters, support short activities like dog walking and passive social activities like sitting and resting.

Street linear spaces are influenced by factors such as the early night market, which increases pedestrian density during certain periods. After early winter snowfall, ice accumulation on pipes attached to street-facing buildings creates slip hazards, prompting elderly individuals to use sidewalks more frequently. Additionally, the absence of streetlights on some streets results in reliance on shop lighting at night, leading to poor visibility [40]. Wooden seats remain usable even in severe cold due to their low thermal conductivity, which helps prevent rapid heat loss and allows for longer periods of use. In contrast, metal seats experience a sharp drop in surface temperature at −20 °C, discouraging their use by elderly individuals. Gender differences are also apparent: women typically prefer windproof wooden chairs for continuous social activities, whereas men are more likely to utilize temporary facilities for mobile activities (see

Table 10. The definition and examples of the winter stage.

Gathering Location	Street Elements	Social Activities
Residential entrances and exits	Stairs, flower containers	Say hello, chat with relatives, friends, and neighbors, wait and see, purchase goods
In front of shops	Steps, flower beds, roadside trees, roadblock stones, signs	Say hello, purchase goods, chat, wait for the bus to observe, inquire, promote products, smoke and communicate
Mobile vendors	Street trees, flower containers	Purchase products, observe and wait
Square	Fitness and amusement elements, flower beds, public seats, trash cans	Square dancing, playing musical instruments, practicing Tai Chi, watching, sitting or standing chatting, walking and chatting, greeting, walking dogs for communication, chess and card games
Street corner space	Fitness and amusement elements, public seats, trash cans	Sitting and observing, greeting, inquiring, waiting for others to observe, waiting for red lights to chat, hairdressing services, purchasing goods, chatting with friends and neighbors
Intersection	Traffic lights, roadblocks, signs	Stop and wait
Bus stop	Public seating	Observe, chat, and inquire

).

3.1. Characteristics of Social Activities of the Elderly

3.1.1. Quantitative Characteristics

Overall, early winter exhibits the highest number of social activity participants, followed by late winter and then severe winter, as illustrated in Figure 3a. Among activity types, mobile social activities are the most prevalent, significantly exceeding other categories, while passive social activities are the least frequent (see Figure 4). In early winter, women participate in social activities more frequently than men; however, this trend reverses during severe and late winter. Notably, mobile social activities demonstrate the greatest fluctuations across stages, whereas short and passive social activities remain relatively stable, as shown in Figure 3.

During early winter, mobile social activities predominate, while short and passive social activities are minimal (see Figure 3b). Fanrong Street records the highest activity levels, likely attributable to milder weather conditions that facilitate outdoor activities such as chess, card games, and square dancing in corner squares without excessive cold discomfort. In severe winter, similar types of activities are observed, but their frequency is reduced by approximately half (see Figure 3c). In late winter, there is a marked increase in the total number of activities (see Figure 3d), with mobile social activities continuing to dominate, while short and passive social activities reach nearly equivalent levels.

3.1.2. Spatial Characteristics

The results indicate distinct spatial patterns for different types of social activities among the elderly. Mobile social activities primarily occur in linear street spaces and are associated with the presence of street trees and lighting. Continuous social activities are concentrated in aggregated spaces such as corner squares, where public tables, chairs, and fitness equipment are available. Short social activities are focused near store entrances and residential areas, related to steps and tree pools. Passive social activities are concentrated in gathering spaces, like corners and intersections, related to bus stops, tree containers, and seats.

On Pinggong Street, mobile and passive social activities are mainly concentrated in segments 2 and 5, with mobile activities particularly common near intersections. Continuous social activities are observed primarily in segments 2, 3, 5, and 6, likely due to the proximity of catering establishments. Short social activities are widespread across segments 1–6, indicating low spatial dependency. Gender differences are also evident: men are more frequently observed in the auto repair area of segment 1 and the catering area of segment 3, while women are primarily found in the beauty and catering areas of segments 2–3. However, gender differences in mobile social activities are minimal, as illustrated in Figure 5.

On Fanrong Street, mobile, short, and passive social activities are widely distributed. Mobile social activities are the most prevalent and are mainly concentrated in the middle sections of the street. Continuous and short social activities are concentrated in the corner square on the west side. Passive social activities are primarily located on the east side, likely because this area receives more sunlight and contains supportive elements such as seating and bus stops, which facilitate activities like waiting for buses and sunbathing. Men are more scattered on the streets, while women are mainly distributed in the clustered spaces, such as street segments 4–5, as shown in Figure 6.

On Anshun Street, mobile social activities are widely and evenly distributed throughout the street. Continuous and short social activities are concentrated in segment 1, likely due to its openness, presence of vendors, and continuous communication among the elderly.

Continuous social activities are dense at intersections, possibly due to waiting for traffic lights. Passive social activities are dense at the intersection of segments 1, 3, and 4, possibly due to the corner green space providing spots for sitting and watching. Men are more distributed in the residential area of street segments 4–7, and women are more distributed in the catering and shop areas of segments 2–3, as well as the corner square, as shown in Figure 7.

3.2. Fitting of Activities According to Rest Fitness Elements, Spatial Pattern, and Stages of Winter

Kernel density analysis of street elements reveals that rest fitness elements are primarily concentrated in corner squares, parks, and restaurant areas. On Pinggong Street, elements are linearly distributed in front of restaurants and shops. In Fanrong Street, they are mainly concentrated in segment 8′s corner square. In Anshun Street, elements are planarly concentrated in segment 1′s corner park and segment 2′s restaurant area. The suitability of these elements for elderly activities varies across different winter stages.

3.2.1. Early Winter Stage

The spatial distribution of areas with a high degree of fit between street elements and elderly social activities varies across the three streets: Pinggong Street’s high-fit areas are predominantly central, Fanrong Street’s are located at both ends, and Anshun Street’s are concentrated in the western segments. Overall, elderly social activities are more compatible with linear street spaces and less so with aggregated spaces. Mobile social activities are particularly well-suited to linear spaces, whereas continuous and passive social activities are more common along the sides of streets. Short social activities tend to occur at storefronts but exhibit poor fit at intersections.

On Pinggong Street, high fitting is observed in segments 2–5, whereas segments 1–3 show low fitting due to there being insufficient elements. Mobile activities display an east–west variation, while other activity types diverge from the center. On Fanrong Street, it has high fitting at the ends but low fitting at the intersections of segments 2–3 and 4–5. Segment 8 has low fitting, high passive social activities, and insufficient elements. On Anshun Street, high fitting is observed in the parks of segments 1–2, while residential segments 5–7 show low fitting. Activities in segment 8 are concentrated in segment 1 due to safety hazards, highlighting the need for additional elements, as shown in Figure 8.

3.2.2. Severe Winter Stage

The degree of fit between elderly social activities and rest fitness elements varies across different streets and spatial contexts. Intersections and shop areas generally exhibit a higher degree of fit, whereas residential entrances tend to perform poorly in this regard. On Pinggong Street, the highest degree of fit is observed at the shops in segment 1, while the lowest occurs at the intersection of segments 2–3. Mobile social activities show improved fit in segments 1–3, potentially influenced by climatic conditions, while continuous and passive social activities correspond with early winter distribution patterns. Short social activities have shifted from segment 2 to segment 1.

On Fanrong Street, segment 8 demonstrates a good fit, whereas the intersections of segments 4–5 lack sufficient elements to support social activities. Segments 5–12 collectively exhibit poor fit, despite continuous social activities peaking in segment 8. On Anshun Street, the park in segment 2 provides a high degree of fit, while segments 4–7 lack adequate facilities and require improvements, particularly for wind and cold resistance. Mobile, short, and passive social activities are well supported in segments 1–8, with a noticeable eastward shift—likely attributable to climatic factors—underscoring the significant impact of temperature on both mobile and continuous social activities, as illustrated in Figure 9.

3.2.3. Late Winter Stage

The degree of fit between social activities and rest fitness elements is generally very similar in both the late and early winter stages. Linear spaces show higher fitting than early and severe winter. On Fanrong Street, the degree of fit fluctuates at intersections, with continuous social activities reaching their highest levels in segment 8’s square. Anshun Street exhibits low fitting with east–west differences, concentrated in linear spaces. For both stages, patterns of mobile, short, and continuous social activities are largely similar, while passive social activities remain limited due to insufficient supporting elements. On Pinggong Street, mobile social activities are both high and evenly distributed, whereas short activities are notably lower in segments 3–6. The junction of segments 2–3 demonstrates the lowest degree of fit, attributable to a scarcity of supporting elements and high pedestrian flow, as illustrated in Figure 10.

3.3. Road Traffic Elements and the Social Activities of the Elderly

Kernel density analysis of activities observed near road traffic elements indicates that these activities are primarily concentrated at intersections of main and secondary roads, as might be expected. On Anshun Street, segments 3 and 4 exhibit particularly high activity density, likely due to complex traffic conditions in these areas. The presence of anti-slip pavement (with a friction coefficient greater than 0.6) and extended signal light durations (exceeding 30 s) significantly enhance elderly individuals’ sense of safety when navigating icy and snowy roads. Notably, men tend to rely more on anti-skid paving, whereas women place greater emphasis on the presence of snow-proof roofs at bus stops.

3.3.1. Early Winter Stage

Road traffic elements and elderly social activities exhibit high fitting, concentrated at main intersections, with Anshun Street’s residential area showing homogeneous fitting. On Pinggong Street, segments 1–2 beneath the elevated bridge demonstrate a high degree of fit, whereas the intersection of segments 2–3 exhibits lower fitting, primarily due to catering-related traffic. Similarly, segments 3–6 show reduced fitting as a result of high levels of catering activity. Mobile and short social activities predominate on Pinggong Street, while continuous and passive social activities are best supported at the segment 2–3 intersection. Fanrong Street presents an overall low degree of fit, with a peak at the Haicheng Street intersection and the lowest fitting observed at segment 8’s square, where the number of supporting elements is limited. Short and passive social activities show particularly low fitting in segments 1–4, attributable to high pedestrian flow and a lack of adequate facilities.

On Anshun Street, the degree of fit is generally high, except at the intersection of segments 2–9, where dense traffic conditions reduce the suitability of spaces for elderly activities. Notably, mobile social activities demonstrate a lower degree of fit compared to other activity types, likely due to their higher frequency, as illustrated in Figure 11.

3.3.2. Severe Winter Stage

Social activities in relation to road traffic elements show fitting that is concentrated at intersections. On Pinggong Street, the fitting pattern remains consistent with early winter, with an increased presence of mobile social activities at the intersection of segments 2–3. However, mobile activities exhibit low fitting in this area, likely due to the high density of vendors and pedestrian traffic. Fanrong Street displays a fitting pattern similar to that of early winter, suggesting minimal impact from temperature changes. Mobile, continuous, and passive social activities mirror early winter distributions, while short activities increase, potentially as a response to climatic factors.

In contrast, Anshun Street demonstrates high fitting in segments 1–4, which differs from early winter patterns as activities shift to segments 5–7 in response to colder weather conditions. Here, mobile activities correspond closely with short social activities, while continuous activities align with passive activities. Notably, segments 1–2 show low fitting, as illustrated in Figure 12.

3.3.3. Late Winter Stage

The degree of fitting between elderly social activities and road traffic elements remains stable in the late winter and is minimally affected by climate conditions. On Pinggong Street, the fitting pattern is consistent with early and severe winter stages, with the highest levels observed at the intersection of segments 2–3, likely due to reduced outdoor activity. Mobile social activities increase in late winter, while others resemble early winter. On Fanrong Street, the degree of fit peaks at the intersections of segments 5–6, with a notable increase in continuous social activities, which are concentrated in segment 8’s square. On Anshun Street, fitting is highest in residential areas but low at the intersection of segments 1–2 and the park, attributed to concentrated activities and reduced pedestrian flow. The street exhibits similar fitting for mobile and passive social activities, while continuous activities return to early winter levels, and short activities peak but may decline sharply due to climatic factors, as shown in Figure 13.

3.4. Green Landscape Elements Impact on the Social Activities of the Elderly

Kernel density analysis of green landscape elements (e.g., street trees and plant containers) in the three sample streets shows a linear distribution concentrated on both sides of the street. On Pinggong and Fanrong Streets, these elements provide spaces for conducting leisure activities for the elderly. In contrast, on Anshun Street, green landscape elements are primarily concentrated at street corners and along the sides, thereby enhancing the street’s aesthetic appeal. Evergreen vegetation provides a more effective windproof barrier than deciduous trees and can reduce wind speed at temperatures as low as −15 °C, thereby creating a “microclimate shelter zone.” Gender differences are also observed: women more frequently use pavilion and corridor spaces for passive social interactions, while men are more likely to utilize the shading provided by roadside trees.

3.4.1. Early Winter Stage

The degree of fitting of social activities and green landscape elements shows significant east–west differences. On Pinggong Street, segment 1 is oversaturated due to auto repair activities and low social engagement, attributed to dense catering and beauty-related traffic. Mobile social activities are homogeneously distributed, with the lowest fitting observed near supermarkets and kindergartens in segment 2. In Fanrong Street, segment 8’s square shows high fitting but suffers from a low number of elements, while segments 5–6 are oversaturated due to low pedestrian flow, and segments 4–10 require additional tree benches. In Anshun Street, the intersections of segments 2–9 and 3–4 demonstrate high fitting, supported by dense traffic and complete elements, but segments 5–7 lack adequate green elements. Both Fanrong and Anshun Streets show homogeneous fitting for all activity types, peaking at intersections, as shown in Figure 14.

3.4.2. Severe Winter Stage

The degree of fitting of social activities and green landscape elements is similar to early winter. On Pinggong Street, the fitting is generally homogeneous, except for a decrease in continuous activities in segments 2–3, which may be attributable to climatic factors. Mobile social activities exhibit the lowest degree of fit at the intersection of segments 2–3, while other activity types follow patterns observed during severe winter.

Fanrong Street displays a similarly homogeneous distribution of fitting, with a slight increase in mobile activity at segment 8’s square, despite the continued insufficiency of green landscape elements. Anshun Street’s fitting matches early and severe winter. Both Fanrong and Anshun Streets exhibit homogeneous fitting for all activity types compared to early winter, with high fitting areas at intersections, as shown in Figure 15.

3.4.3. Late Winter Stage

The degree of fit between social activities and green landscape elements across the three sample streets remains consistent with early and severe winter, indicating minimal influence from temperature changes. On Pinggong Street, the degree of fit is higher at the eastern and western ends but lower in the central sections, with particularly low fitting observed at intersections. Fanrong Street shows local aggregation and a high degree of fitting in corner squares and bus stops. Anshun Street shows significant east–west differences, with a high degree of fitting at intersections and parks. Mobile, passive, and short social activities show homogeneity across winter stages, while continuous social activities increase in late winter, adapting to passive social activities. On Fanrong Street, this increase may relate to reduced severe winter activities. In Anshun Street, short social activities increase or sharply decrease due to the climate, as shown in Figure 16.

4. Discussion

This study employed non-participatory observation and behavioral mapping to investigate the impact of winter climate and street elements on elderly social activities in Harbin, a severe cold-region city. The findings confirm a persistent demand for outdoor social engagement among older adults, despite harsh winter conditions—a trend consistent with previous research on aging populations in cold climates. For example, studies in Canadian and Scandinavian cities [7,8,9] have similarly observed that older adults continue to prioritize social interaction, although the frequency of activities declines with increasing cold. Aligning with the existing literature, our results demonstrate that participation in social activities is inversely correlated with winter severity, with the highest levels occurring in early winter and declining during severe and late winter stages. This pattern reinforces the global challenge of seasonal isolation among aging populations [1].

Furthermore, this study affirms that street elements serve as essential mediators for facilitating “health participation” [1]. Even under severe cold conditions, aging-friendly seats and anti-slip pathways continue to support outdoor socialization among the elderly, helping to offset the risk of “winter isolation.” These findings contrast with Cumming’s disengagement theory, suggesting that environmental interventions can delay or mitigate social withdrawal among older adults.

4.1. Rest Fitness Elements: Microclimate Adaptation and Material Selection

The study confirms that the support effect of recreational and fitness elements on continuous and passive social activities was significant (H1 part was established). During the early and late winter stages, the degree of fit between these elements and social activities is high; however, this fit declines in the severe winter stage, primarily due to reduced activity caused by lower temperatures.

Further analysis reveals that the material composition of street elements directly influences the willingness of older adults to use them. Gender differences are evident in the use of windproof wooden seats: women are more likely to choose these seats for continuous social activities, while men tend to use temporary elements, reflecting their preference for more mobile activities. On Pinggong Street, the strong thermal conductivity of metal seats leads the elderly to avoid them during winter, opting instead for personal cushions or utilizing low walls as seating. In contrast, the wooden seats in the corner square of Fanrong Street demonstrate high rates of use, attributable to their lower thermal conductivity. These findings suggest that street elements should prioritize materials with superior thermal insulation properties, thereby extending previous research on microclimate design [5]. The integration of windproof features can further enhance the usability of these elements during winter. It is recommended to install thermally insulated seating in areas frequented by female groups and provide simple rest points along pathways with high male activity frequency.

4.2. Road Traffic Elements: Safety Orientation and Functional Composite

The supporting effect of road traffic elements on mobile social activities is significantly affected by the climate (H2 established). Elements such as signal lights and anti-skid pavement at the intersection are very important to ensure the safety of the elderly. The research shows that due to the use of anti-skid floor tiles and the extension of signal lamp duration, the fitting degree of mobile social activities at intersections 3 and 4 of Anshun Street in severe winter remains stable.

Gender differences are evident: men rely more heavily on anti-skid pavements and signal lights for mobile activities in severe winter, whereas women exhibit a greater need for snow-covered shelters and adequate lighting at bus stops. In certain areas, elderly activities are restricted by icy road surfaces or insufficient provision of supportive elements.

To address these challenges, it is recommended to use high-friction paving materials at key traffic nodes and install snow-melting equipment. Additionally, vehicle separation elements should avoid sharp edges and employ flexible materials to minimize collision risks. Optimization of road traffic element design should consider men’s reliance on road safety features and women’s preferences for comfortable and sheltered waiting environments.

These findings highlight that in severe winter conditions, road traffic elements play a critical role in sustaining mobile social activities among the elderly, in contrast to studies from temperate climates, which primarily emphasize walkability alone [14].

4.3. Green Landscape Elements: Seasonal Adaptive Design and Ecological Function

The supporting effect of greening landscape elements on short and passive social activities showed seasonal fluctuations (H3 was partially established). During early and late winter, tree pool benches along the streets and pavilions in corner parks serve as primary venues for social interaction due to their effective wind-shielding properties. However, in severe winter, areas with only windbreak structures still attract a small amount of passive activity. Women have a higher utilization rate of evergreen vegetation and windproof pavilion corridors in corner parks, while men are more concerned about the sheltering effect of street trees in linear spaces. The type of green vegetation affects the winter microclimate; evergreen trees, such as pine trees, are better able to block the cold wind and improve space comfort than deciduous trees. This finding adds nuance to the results from more vegetation-focused research in milder climates [39]. In future designs, priority should be given to the selection and arrangement of cold-resistant evergreen plants, integrating seat layouts to form effective “green barriers”. Additionally, design strategies should aim to strengthen the thermal insulation function of vegetation in female-dominated gathering areas and improve the practicality of greening strategies in male-intensive spaces.

4.4. Comprehensive Optimization Strategy of Elements

Street elements in cold areas should balance functionality and climate adaptability. Elements should enhance the appeal of gathering spaces, with gender-specific needs in mind. Women-led spaces can add more social equipment, while male-frequented areas should improve pavements and snow-melting systems. Linear spaces should ensure continuous walking paths and minimize cold exposure.

Material selection is critical. Avoid materials that are unsuitable for winter conditions, such as metal and smooth tiles, and instead prioritize anti-skid and insulating materials, such as wood and rubber, to improve safety and comfort.

4.5. International Comparative Perspective

Microclimate adaptation is a critical consideration worldwide, yet the solutions adopted vary significantly by context. For example, Harbin’s use of ground-level wooden benches contrasts sharply with Montreal’s extensive underground pedestrian corridors [7] and Sapporo’s snow barriers, each reflecting strategies shaped by urban density and local conditions. Safety interventions also demonstrate technical disparities: Scandinavian cities often employ heated pavements, while East Asian cities tend to utilize anti-skid materials.

Social patterns differ regionally: in Harbin, passive social activities cluster near commercial spaces, a pattern distinct from Sapporo’s festival-centered winter model. This highlights the integration of commercial environments as a uniquely East Asian approach to fostering winter vitality. While universal principles such as thermal comfort and safety underpin microclimate design, their implementation must be calibrated to urban density and cultural norms. “Linear gathering spaces” in Harbin—such as bus stops and pavilions—contrast with the three-dimensional pedestrian infrastructure of high-latitude cities like Montreal and Sapporo, where underground networks and snow wall ventilation systems extend outdoor activity time. Ultimately, while all these approaches aim to support year-round social participation, Harbin’s reliance on ground-level facilities reflects both its urban form and economic adaptability, distinguishing it from the multi-layered transport solutions seen in other high-latitude cities.

4.6. Study Limitations and Future Directions

While this study offers critical insights into the influence of street elements on elderly social activities in severe winters, certain limitations should be acknowledged. First, the reliance on observational methods, while effective for mapping spatial behaviors, may overlook subjective factors such as personal preferences, health status, or social motivations influencing activity patterns. Additionally, the study did not fully account for microclimatic variations (e.g., wind speed and sunlight exposure) or socioeconomic disparities among participants. Future research should expand to comparative case studies across diverse cold-climate cities to validate findings. Mixed-method approaches integrating qualitative interviews could elucidate the elderly’s lived experiences and unmet needs.

5. Conclusions

In terms of the characteristics of social activities, the total number of social activities of the elderly is the highest in the early winter and lowest in the severe winter. Mobile social activities constitute the predominant form, primarily occurring within the linear spaces of the street. Continuous and passive social activities are significantly affected by climate and depend on the spaces to gather, such as corner squares and bus stations. Women’s social activity in early winter prefers continuous activities, while the proportion of male activities in severe winter and late winter increases, preferring mobile and short activities. It is necessary to design the aggregation space and traffic nodes differently.

With respect to the influence of street elements, rest fitness amenities provide significant support for continuous and passive social activities; however, their utilization decreases markedly at low temperatures. Wood materials are loved for their low thermal conductivity. Recommendations include adding an elderly activity room or chess tables with low thermal conductivity materials like wood. Road traffic elements play a crucial role in supporting mobile social activities, particularly in severe winter, when the implementation of anti-skid and snow-melting designs becomes essential. Green landscape elements demonstrate strong seasonal adaptability, facilitating short and passive social interactions; evergreen vegetation, in particular, is more effective at improving the winter microclimate than deciduous species. Furthermore, the spatial influence of plant layouts observed in summer persists into the winter months.

The study emphasizes that street design in cold regions should take into account both functionality and climate adaptability. Through material innovation, spatial layout optimization, and seasonal adaptive design, the social participation of the elderly in winter should be improved, and the construction of inclusive cities should be promoted.