Correlation between spatial characteristics and overflow rate of back alleys in high‐density city of Hong Kong

Abstract

In recent years, scholarly attention has shifted within urban informality research from informal settlements to the appropriation of public spaces in formal urban settings. Overflow, a common research subject within this context, has drawn particular interest. This study investigates overflow dynamics in a representative district of Hong Kong, utilizing the overflow rate as a measure of the extent to which overflow occurs. Three‐dimensional (3D) scanning technology was applied to collect spatial data from back alleys. Multiple linear regression analyses were conducted to identify the spatial characteristics of back alleys that influence overflow rates. The results revealed that back alleys with a flatter spatial interface, more overhead shelter, greater numbers of alley‐shops, and shorter overall length corresponded to a higher number of overflow occurrences. Two novel findings emerged: First, a negative correlation between spatial interface unevenness and overflow rates, and second, differential impacts of the shop density and number of alley‐shops on overflow rates. These findings may serve as a valuable reference for governmental policy formulation regarding overflow management and may offer architects workable insights for designing back alleys and street spaces conducive to public use.

Introduction

Urban informality

This study examines the phenomenon of overflow within the context of urban informality. Initially rooted in the examination of the informal sector during the 1970s,¹ the concept of urban informality primarily centered on studying informal settlements such as slums.² Early conceptualizations surrounding urban informality lacked uniformity but generally underscored its association with a dearth of laws, regulations, and controls.³ Arguably, urban informality poses significant challenges the sustainable development of cities, manifesting in various issues related to planning, management, and environmental concerns.^4,5

In recent years, there has been a paradigm shift in the study of urban informality. Academics have begun to question the rigid dichotomy between informality and formality and to recognize urban informality as an integral part of city planning.⁶ A widely accepted definition currently characterizes urban informality as a flexible coping strategy with diverse political and socioeconomic implications, as well as a spatial discipline involving intricate interactions among the state, localities, capital, and society.⁷

The focus of urban informality research has shifted from informal housing and settlements toward the appropriation of public spaces within formal city blocks.^8–11 Appropriation, in this context, entails temporary activities occurring in public spaces that diverge from their original design purposes and accentuate certain identifiable qualities.¹² The associated concept of “overflow” is a prominent research subject concerning the appropriation of public spaces within the context of urban informality.¹³

Overflows in Hong Kong

Overflows are removable private items placed in public spaces that expand private storage space or convey the personalities of inhabitants.^14,15 Beyond their utilitarian function, overflows offer multidimensional social benefits. They facilitate community participation in shaping urban spaces, imparting complexity and vitality to cities.¹⁶ Notably, occupants displaying overflows in front of their entryways tend to have stronger connections with their neighbors,¹⁷ often influencing others to follow suit by placing similar items.¹⁸ Aesthetically, overflows epitomize the concept of “architecture without architects”¹⁹ and contribute to the diversity and intricacy of urban landscapes. Moreover, they reflect the imagery, symbols, and values associated with specific sites, buildings, and urban elements, offering insights into the ways that individuals interpret their environments.²⁰

Overflow phenomena in Asian cities is strongly correlated with high population density,^11,21 a key trait of the region's urbanization.²² For instance, Hong Kong, a typical megacity with a population density of 6740 persons per square kilometer,²³ exhibits particularly pronounced overflow phenomena.²⁴ Hong Kong's overflows are predominantly concentrated in back alleys, (Figure 1) remnants of the grid-like urban planning of the British colonial period.^24,25 These narrow passages run behind buildings parallel to streets, supplementing traffic flow through the area.²⁶ Similar back alleys exist in other high-density cities with colonial histories, such as Singapore and Yangon. However, unlike in Hong Kong, these alleys either vanished during modernization or remain underutilized as public spaces due to their differing scale.^27,28

[IMAGE OMITTED. SEE PDF]

Possible factors affecting overflows in back alleys

Compared to other urban public spaces such as streets and squares, back alleys exhibit a heightened prevalence of overflows due to factors such as local laws and regulations, ownership, strong public-private sphere connections, and efforts to maintain environmental sanitation.

First, lax enforcement or laws and regulations play a significant role. While governments do prohibit retail shops from encroaching on public space,^29,30 enforcement agencies often prioritize monitoring streets over back alleys, leading shop operators to redirect overflows from sidewalks and roads to back alleys.³¹ Second, ownership ambiguity complicates regulatory efforts. Overflows often occupy areas perceived as part of buildings' common spaces rather than public spaces, (Figure 2) so private interests can complicate formal regulations governing overflows in back alleys.³² Moreover, the interconnectedness of public and private territories within back alleys facilitates the swift relocation of overflow items by users, perpetuating the issue when law enforcement is not present. Furthermore, overflowers actively contribute to maintaining sanitation in back alleys. Shop owners and scavengers alike take proactive measures (eg cleaning, recycling, repairing leaks, moving trash) to effectively manage overflows.

[IMAGE OMITTED. SEE PDF]

Past studies have employed overflow rate as a metric to quantify overflow phenomena,¹⁴ revealing a strong correlation between overflow rates and spatial characteristics.^14,33–36 This study focuses on the appropriation of public space within formal city blocks, specifically examining high-density back alleys in Hong Kong. Each back alley in the case study area was divided into several back-alley units 30 m in size, and the ratio of the planar projected area of overflow within each back-alley unit to the ground area of the back-alley unit was used as the overflow rate.^14,33–36 Spatial characteristics that may affect the overflow rate were referenced from previous studies. Quantitative analysis was conducted to elucidate the relationships between overflow rates and these various spatial characteristics.

Past research

Previous studies have explored how these spatial characteristics affect overflow rates in four distinct ways: scale, openness, business, and spatial interface. Regarding scale, Masuda et al. (2018) observed that narrower back alleys and shorter distances between these alleys and streets with widths of 6 m or more result in increased plant-life overflows.³⁴ In terms of openness, Aoki et al. (1993) noted a greater occurrence of overflows in open back alleys compared to enclosed ones.¹⁴ Wang et al. (2019) also found a positive correlation between the size of atriums located under eaves and overflow rates in traditional houses in Quanzhou, China.³³ In the context of business, Zhang et al. (2020) pointed out that store density is positively correlated with overflow rate in traditional commercial streets in Zhejiang, China.³⁵

Finally, spatial interfaces, as observed by Muminović (2023), are correlated with overflow rates in places like the Nezu region of Tokyo. However, this connection has not been fully elucidated.³⁷ Architect Herman Hertzberger (2013) states in Lessons for Students in Architecture that spatial interfaces with more “additional surfaces” tend to encourage greater use,³⁸ while Patil et al. (2013) noted that the unevenness of the spatial interface contributes to overflow occurrences.³⁹ Kubota et al. (2018) also found that the number of alcoves and balconies in spatial interfaces on either side of public corridors in apartment buildings positively correlate with overflow rates.³⁶

It is worth noting that most prior research on overflows focused on back alleys in traditional residential areas, public corridors of apartment buildings, or traditional commercial streets, which are relatively homogeneous in terms of overflow types. In contrast, Hong Kong's back alleys are more representative of informality in high-density urban spaces.⁴⁰ These back alleys feature complex spatial characteristics accommodating a wide array of activities and people, resulting in a diverse array of overflow items. Therefore, studying Hong Kong's back alleys is particularly valuable as many countries are transitioning toward higher population densities.

Additionally, most previous studies relied on manual measurements during site surveys, which can be labor-intensive and time-consuming. Given the limited spatial scale of back alleys, employing a ground-fixed laser scanner for data collection could also be excessively time-consuming. Several scholars have validated the feasibility of using 3D scanning technology with consumer handheld devices such as the iPhone Pro equipped with Apple LiDAR Sensors for collecting spatial data.^41–43 In this study, Apple LiDAR Sensor-based 3D scanning technology was utilized to acquire spatial data within back alleys, a novel approach in the field of overflow and street research. With back alleys in Mong Kok District, Hong Kong as the study object, this work began by compiling a list of spatial characteristics that may impact the overflow rate. Quantitative analysis methods were then applied to determine the correlations between these spatial characteristics and the overflow rate in order to rank the influence intensity of each spatial characteristic. The findings of this study may provide valuable insights for planners involved in the design of back alleys and street spaces.

Research Methods

Research materials

Mong Kok, the district selected for this study, is situated in the northern part of the administratively zoned Yau Tsim Mong District, which is renowned as one of Hong Kong's prime tourist and shopping destinations. Mong Kok is extremely densely populated, with an average of 130 000 people per square kilometer. However, it offers only 0.6 square meters of public space per capita, which falls significantly below the government's standard of 2 square meters. This figure is even lower than the range of 5.8–7.6 square meters in other developed Asian cities such as Seoul, Tokyo, and Singapore.⁴⁴ The urban space in Mong Kok is predominantly shaped by mixed-use buildings, combining light industries and offices with ground-floor shops and restaurants while the upper floors are dedicated to residential purposes.⁴⁵

Back alleys can be found throughout Hong Kong, including areas like Hong Kong Island, Tsim Sha Tsui, Jordan, Sham Shui Po, and Kowloon City. However, Mong Kok is distinguishable by its more regular and uniformly oriented block layout. This uniformity minimizes the influence of variables such as the angle of sunlight, making it an ideal choice for this study. Restricting the study area to a single administrative district like Mong Kok also helped eliminate the potential impact of policy variations between different districts on overflow rates.

In this study, the smallest unit of analysis material was a 30-m-long back-alley unit, sequentially labeled from north to south. A total of 72 back-alley units (28 back alleys) were identified within Mong Kok. Figure 3 illustrates the locations of the study objects and the spatial relationships between the back alleys and adjacent buildings on both sides, with both plan and section views for clarity.

[IMAGE OMITTED. SEE PDF]

Site survey and research process

The survey was conducted from November 1 to November 29, 2022. First, the researcher walked through the back alleys with an iPhone 14 Pro in hand and 3D-scanned 72 back-alley units (28 back alleys). The 3D scanning data were imported into Rhino7 3D modeling software, then a 3D model was created with reference to this data to obtain spatial characteristics and overflow rates for the back alleys (Figure 4). Finally, the relationship between back alley spatial characteristics and overflow rate was derived through multiple linear regression in SPSS29.

[IMAGE OMITTED. SEE PDF]

Spatial characteristics of back alleys

The spatial characteristics of the back alley were classified into four categories: scale, openness, business, and spatial interface.

Scale

Scale encompasses the dimensions of the back alley, including its width and length. The width of the back alley can significantly affect the manner in which people behave within it.⁴⁶ To account for variations throughout the study area, the widths of back alleys were obtained by referring to Zhang et al.'s (2020) calculation of street width.³⁵ The length of the back alley plays a role in determining whether pedestrians use it as a throughway,⁴⁷ impacting its utilization by individuals other than pedestrians. In this study, the length of the back alley was measured as the distance between its head and tail in the direction of the longer side of the block (Figure 5).

[IMAGE OMITTED. SEE PDF]

Openness

In this context, openness reflects the proximity to the street and the rate of overhead shelters. Proximity to the street is defined by the spatial distance between different locations within the back alley and the adjacent street. Locations closer to the street are more directly connected to it, while those farther away are less so. To assess this, the shortest distance from the planar center point of the back-alley unit to the street was used as a measure of the proximity to the street.

The rate of overhead shelters gauges the level of visual openness of the back alley to the taller buildings on its sides. Various sheltering objects in the back alley, located above the height of windows and doors on the ground floor, include self-built rain-shading canvases, outdoor air-conditioning units, and awnings for windows and doors. In this study, the ratio of the horizontal projection of sheltering objects at least 2 m above the ground floor to the ground area of the back-alley unit was used as an index for evaluating the rate of overhead shelters (Figure 6).

[IMAGE OMITTED. SEE PDF]

Business

Business-related factors encompass the number of alley-shops⁴⁸ and the shop density. Alley-shops are authorized by the government for individual operation and are often located at the ends of back alleys, with their fronts facing both the street and the alley. The number of alley-shops can influence the sense of security perceived within the back alley, as alley-shops provide “informal surveillance.”^47,49–51 Alley-shops are often key vendors, newsstands, fruit shops, and grocery stores,⁵² with shopkeepers who can monitor activities in the back alley as they attend to their businesses.

Based on the location of their territorial boundaries, alley-shops can be roughly divided into four categories: embedded, dependent, independent, and mobile.⁵³ An embedded alley-shop's territory is located in the interior space of a building adjacent to the alley, clearly delineated by the building's boundaries. Owners of these shops extend their merchandise, shelving, or seating into the back alley. A dependent alley-shop does not appropriate the interior building space but instead occupies the space between the walls on either side of the alley. These shop owners reinforce their territory with temporary roofing structures. An independent alley-shop is positioned alongside a building and has distinct metal box-shaped enclosures as territorial boundaries. Owners typically utilize the interior of these enclosures to display goods or as resting areas, while also extending merchandise outside to appropriate public space. Lastly, a mobile alley-shop has less defined boundaries, often employing umbrellas and other overflow items to demarcate territory. In this study, the number of alley-shops in the back-alley unit (regardless of type) was quantified as an evaluation index.

The shop density potentially affects the likelihood of ground-floor shop employees on either side of the back alley placing overflow items in the alley. The ground floors of buildings in the Mong Kok area are mostly commercial, and back alleys frequently link the backyards, restrooms, storage rooms, and other service areas of the ground-floor shops on both sides. Shop employees may use the back alley for temporary storage of work items or for work-related tasks like cleaning and maintenance.

It is important to note that the shop density and the number of alley-shops are two separate and distinct variables. The shop density refers exclusively to shops on both sides of the back alley with their rear facades facing the alley, while the number of alley-shops refers exclusively to shops positioned at the back-alley's end, with their fronts oriented toward it. In this study, the shop density was determined based on the ratio of ground-floor shops on both sides of a back-alley unit to its length (Figure 7).⁵⁴

[IMAGE OMITTED. SEE PDF]

Spatial interface

The primary focus of this analysis was the unevenness of spatial interfaces within back alleys. The spatial interface of the back alley encompasses the ground surface and walls flanking the alley. Elements contributing to unevenness include steps, alcoves, wall protrusions, corners, air conditioners and air conditioning racks, pipes, shelves privately installed by residents, storefronts of alley-shops, and garbage bins installed by the Food and Environmental Hygiene Department. The presence of these uneven elements can provide additional space for overflows beyond ground level, such as chairs positioned in alcoves, carts secured to wall-mounted pipes, or plantings suspended between pipes and walls. Therefore, unevenness may affect the space available for overflow placement in back alleys. In this study, the ratio of the volume of unevenness elements in back-alley units with a floor height of 3 m or less to the volume of the space with a floor height of 3 m or less in the back-alley unit was used as an evaluation index for unevenness (Figure 8).

[IMAGE OMITTED. SEE PDF]

It is important to note that public garbage bins and wall-mounted shelves were considered factors contributing to unevenness in the spatial interface. While the former is “removable” and the latter has “private properties,” they do not fall under the definition of overflow used in this study. Here, an overflow is defined as a removable private object placed in a public space. Accordingly, government-provided garbage bins in the back alley are not private property; individual-fixed shelves on alley walls are not removable. Thus, these elements were not recognized as overflows.

Overflow content in back alleys

Back alleys are rich with a diverse array of activities that appropriate their space through various types of overflows. First and foremost, back alleys are hubs for business-related activities. These spaces offer additional logistical room for the ground-floor shops on both sides or serve as ideal locations for residents to set up their own less formal alley-shops. Overflows encompass a range of commodities from alley-shops, including tables, chairs, cleaning tools, workbenches, ladders, and other items.

Second, back alleys create space for the activities of daily life. The ground-floor spaces on either side of the back alley may house property manager's offices, residences, and other similar spaces. Users of these spaces frequently place personal items in the back alley, such as potted plants, motorbikes, bicycles, clothes for drying, electric fans, and storage lockers. Back alleys also provide space for activities that require discretion, such as smoking and the disposal of waste products. These activities involve overflows such as homemade paper-cushioned chairs for smokers, pushcarts, buckets for catching water, and other sundry items.

In this study, the amount of overflow was quantified in terms of the planar projected area occupied by the overflow. The ratio of the overflow quantity within the back-alley unit to the ground area of the back-alley unit was calculated to express the overflow rate of the back-alley unit (Figure 9 and Appendix A).

[IMAGE OMITTED. SEE PDF]

Multiple linear regression technique

Linear regression is a statistical technique employed to model the linear relationship between two or more variables,⁵⁵ typically aimed at comprehending how the dependent variable behaves under the influence of one or more independent variables.⁵⁶ Common guidelines for determining sample size recommend 10–20 sample sizes for each independent variable.^57,58 Seven independent variables and a sample size of 72 were utilized in this study, which satisfies these criteria.

Several spatial characteristics were utilized as independent variables, including the width and the length of the back alley, proximity to the street, the rate of overhead shelters, the number of alley-shops, shop density, and unevenness. These were correlated with the overflow rate, which served as the dependent variable. SPSS29 software was employed to conduct multiple linear regression analyses using data from 72 samples (Figure 10, Table 1, and Appendix B). The aim was to unveil the relationship between these spatial characteristics of back alleys and overflow rate, ultimately for ranking the influence intensity of the spatial characteristics.

[IMAGE OMITTED. SEE PDF]

Table 1 Results of descriptive statistics (n = 72)

Variables	N	Minimum	Maximum	Mean	Std. deviation
Width of back alley (m)	72	2.586	4.578	3.071	0.234
Length of back alley (m)	72	58.931	120.419	89.225	18.964
Proximity to the street (m)	72	14.058	65.951	28.317	14.635
Rate of overhead shelters	72	0.015	0.647	0.224	0.128
Number of alley-shops	72	0.000	2.000	0.431	0.668
Shop density	72	0.070	0.300	0.166	0.054
Unevenness	72	0.003	0.123	0.045	0.031
The overflow rate	72	0.025	0.480	0.145	0.085

Results

Examination of model

As per the data presented in Table 2, the R² value of the model is 0.277, which signifies that the width and the length of the back alley, proximity to the street, the rate of overhead shelters, the number of alley-shops, shop density, and unevenness collectively account for 27.7% of variability observed in the overflow rate. A test for the multiple collinearity indicated that all the variance inflation factor (VIF) values in the model are less than 5, suggesting that there are no significant collinearity issues among the independent variables (Table 3). Additionally, the Durbin–Watson value, which hovers around 2, indicates that the sample data are independent and the model fits well (Table 2). Therefore, this model can be considered robust and reliable.⁵⁹

Table 2 Model summary of multiple linear regression (n = 72)

Model	R	R2	Std. error of the estimate	Durbin–Watson
1	0.527a	0.277	0.076	2.054

Table 3 Results of multiple linear regression (n = 72)

	B	Beta	p-Value	VIF
(Constant)	0.340	–	0.016*	–
Width of back alley (m)	−0.022	−0.060	0.586	1.059
Length of back alley (m)	−0.001	−0.265	0.032*	1.298
Proximity to the street (m)	0.001	0.138	0.334	1.792
Rate of overhead shelters	0.221	0.334	0.008**	1.310
Number of alley-shops	0.052	0.413	0.004**	1.724
Shop density	−0.316	−0.202	0.093	1.243
Unevenness	−1.447	−0.531	0.000**	1.529

Effect of independent variable on dependent variable

In Table 3, a p-value less than 0.05 indicates a statistically significant relationship between the independent variables and the dependent variable. The sign of the unstandardized regression coefficient (B), whether positive or negative, signifies the nature of the influence these independent variables exert on the dependent variable.

The p-values for the width of the back alley, proximity to the street, and shop density, all exceed 0.05, indicating that they have no significant effect on the overflow rate. The B-value of the length of back alley is −0.001 (p = 0.032 < 0.05), implying a substantial negative impact on the overflow rate. The rate of overhead shelters, with a B-value of 0.221 (p = 0.008 < 0.01), also appears to significantly positive influence the overflow rate. The number of alley-shops, with a B-value of 0.052 (p = 0.004 < 0.01), similarly has a significant positive relationship with the overflow rate. Finally, the B-value of unevenness is −1.447 (p = 0.000 < 0.01), suggesting that unevenness has a significant negative effect on the overflow rate.

Ranking of influence intensity

Comparing the absolute values of the standardized regression coefficients Beta allowed for evaluation of the relative influence of each independent variable on the dependent variable.⁶⁰ As indicated in Table 1, the most pronounced effect on the overflow rate is attributed to unevenness (|Beta| = 0.531), followed by the number of alley-shops (|Beta| = 0.413) and the rate of overhead shelters (|Beta| = 0.334). In contrast, the length of the back alley (|Beta| = 0.265) has a relatively weak influence on the overflow rate (Figure 11).

[IMAGE OMITTED. SEE PDF]

Discussion

Multiple linear regression was employed in this study to examine the correlation between seven distinct spatial characteristics of back alleys and their impact on the overflow rate. In summary, the rate of overhead shelters and the number of alley-shops were found to influence the overflow rate positively and significantly. The length of the back alley and unevenness, conversely, exhibited significant negative effects on the overflow rate. Notably, the width of back alley, proximity to the street, and shop density did not appear to substantially affect the overflow rate.

The R² value of this study is 0.277, which suggests that the overflow rate in Hong Kong's back alleys cannot be solely predicted by the seven examined spatial characteristics. Urban informality and overflows as spontaneous appropriation phenomena have intricate contexts and influencing factors, which include but are not limited to high population density, urban planning forms, physical and legal ambiguities of public spaces, individual or group demands for the use of public spaces, negotiable spatial management, and individuals' or groups' perceived cultural identities within urban public spaces.^8,12,13,61 These multifaceted factors resist singular explanations through the lens of physical space.

However, in line with empirical research standards in social sciences, if some of the independent variables are statistically significant (p-value less than 0.05), a model with an R² value between 0.10 and 0.50 is acceptable.⁶² Hence, the R² value in this study is acceptable despite the complexities inherent to the phenomena under analysis.

Unevenness

In this study, a negative correlation was observed between the unevenness of the spatial interface and the overflow rate—a finding that departs from the results of previous studies. Unlike past studies that evaluated unevenness affecting overflow in terms of surface area or quantity, this study adopted a novel perspective in analyzing unevenness based on volume.

For a given volume of the back alley, both convexity and concavity protrusions at the spatial interface were found to impede the potential for appropriation. Larger protrusions compress the available space in the back alley, limiting both pedestrian movement and overflow placement, thus reducing the likelihood of overflow occurrences to ensure that pedestrian flow remains unimpeded. Conversely, concavities and depressions within the spatial interface, often associated with building entrances, pose access challenges thereby discouraging overflow placement in those areas.

Number of alley-shops

The results of this study highlight a notable correlation between the presence of alley-shops and an increased incidence of overflows in back alleys. This phenomenon can be attributed to the fact that alley-shops, and their associated overflows, increase the sense of enclosure within back alleys in the horizontal direction. While it may be expected that stronger enclosures in relatively closed public spaces like back alleys could potentially induce feelings of reduced security,⁶³ it is essential to recognize that the sense of concealment resulting from this enclosure can provide a distinct form of security. This unique sense of security emanates from the “informal surveillance” exercised by alley-shop owners, which effectively deters criminal activities by outsiders. As a result, the enclosure formed by this informal surveillance engenders feelings of security and privacy among individuals, ultimately fostering greater utilization of back alleys.

While there exists a positive correlation between the number of alley-shops and the overflow rate, the shop density along both sides of the back alley exhibits a very weak correlation with the overflow rate. Alley-shops with storefronts facing the back alley typically require more space for displaying merchandise to maximize profitability. Conversely, shops situated on both sides of the back alley have their backs facing it and primarily utilize this space for auxiliary functions. Consequently, the overflows from these shops tend to consist of cleaning tools, carts, and other items facilitating logistical services. In instances where shops share the back-alley space, informal negotiations^27,53,64 often lead to reorganization of overflows. Specifically, a higher shop density diminishes the amount of public space available for each shop to accommodate overflows, assuming the back-alley's dimensions remain constant. In these cases, shop owners may resort to borrowing overflow items (eg cleaning tools) from one another rather than placing their own,⁶⁵ thereby mitigating the impact of shop density on overflow rates.

Rate of overhead shelters

This study revealed a higher prevalence of overflows in back alleys characterized by a greater presence of overhead shelters. This phenomenon can be attributed to several factors. First, overhead shelters offer a comfortable microclimate, shielding back-alley activities from heavy rainfall⁶⁶ and direct sunlight.⁶⁷ Second, they contribute to alleviating the sense of oppression often associated with confined spaces.

Overhead shelters can obstruct views emanating from the taller buildings on both sides of the back alley. From a psychological perspective, individuals tend to metaphorically associate the vertical position (high or low) in physical space with social attributes such as dominance or submission.⁶⁸ Views from higher vantage points thus tend to induce a heightened sense of oppression compared to views from the same horizontal level. This effect is further compounded by the narrowness of high-density urban environments.^69,70

The presence of overhead shelters in back alleys not only obstructs views from taller buildings but also ensures a more balanced height and width of the back-alley space, mitigating the sense of confinement. Overhead shelters thus create a microenvironment that is more psychologically and physically comforting for individuals within the back alley. Additionally, they provide essential protection for overflows against adverse weather conditions, including rain and intense sunlight, thereby increasing occurrences of overflow.

Though overflows themselves can serve as similar purpose as overhead shelters—offering protection from rain, sunlight, and views from above—they cannot entirely supplant the role of overhead shelters, primarily due to their limited capacity to shelter a projected area.

Length of back alley

This analysis also revealed a higher occurrence of overflows in shorter back alleys. The rationale for this is that shorter back alleys necessitate a greater number of overflows to intensify the sense of enclosure within these spaces. Shorter back alleys inherently possess a more inviting quality for pedestrians to traverse.⁴⁷ Consequently, individuals in back alleys strategically position more overflows to enhance seclusion and cultivate a heightened sense of enclosure within specific areas of the alley. This, in turn, makes their transiently occupied territories less susceptible to disruption by passing pedestrians.

In contrast, the width of the back alley exhibited a weaker effect on the overflow rate, likely because widths throughout the study areas back alleys are very similar (Table 1 and Appendix B). Future research should include back alleys with a wider range of widths to strengthen the findings of this study.

Conclusion

In this study, the overflow rate served as a metric to assess overflow prevalence and intensity in back alleys of Mong Kok, a densely populated area in Hong Kong. The potential spatial characteristics affecting back-alley overflow rates were extracted by reviewing the existing literature. Spatial data of back alleys were collected with 3D scanning technology and then subjected to multiple linear regression analysis. The results revealed the spatial characteristics influencing overflow rate, with a specific emphasis on the impact of the spatial interface's flatness and the prevalence of overhead shelters, alley-shops, and shorter back-alley scales.

This study yielded two findings inconsistent with previous research. First, analysis from a volumetric perspective revealed a negative correlation between unevenness in the spatial interface and the overflow rate. Larger concavities in the spatial interface necessitate access, while larger convexities limit available space, thereby constraining overflows. This underscores the complexity of spatial-interface unevenness and emphasizes the need for comprehensive evaluation across different dimensions (eg volume, surface area, quantity, and type) in future research on public-space appropriation.

Second, while the number of alley-shops is positively correlated with the overflow rate, shop density along back-alley sides exhibited a weaker correlation. This disparity can be attributed to the primary function of the back alley for alley-shops as a profit space, contrasting with its secondary role as a logistics space for shops on either side. This distinction in function influences informal negotiations, with alley-shop owners negotiating over space, while the owners of shops on either side focusing on specific overflow items. Recognizing when public space serves as an extension of profit-making space versus auxiliary space is crucial for analyzing the impact of shop density on public-space appropriation, offering insights into the adaptability of micro spaces within the context of urban informality.

The resilience and dynamism of overflows have prompted significant governmental investment in regulating back-alley overflows, requiring considerable manpower and resources. At present, the criterion for removing overflows is “whether it obstructs traffic or sanitation.”³⁰ However, this criterion's ambiguity may lead to subjective judgment and potentially corrupt practices, such as bribery, among law enforcement officers. Moreover, the government's narrow focus on back-alley functions, like traffic and sanitation, reflects an outdated functionalist mindset that overlooks the social benefits of overflows. Changes in overflow intensity under different spatial characteristics underscore the influence of physical space on users' demands. Therefore, it is recommended that the government tailor permissible overflow limits according to varying spatial characteristics when addressing back-alley overflows. For instance, treating overflows from alley-shops and shops on the sides of the back alley with the same standard may be unjust given the divergent uses the alley facilitates for these entities.

This study focuses on urban informality and appropriation, with a particular emphasis on the back alleys of high-density cities, thereby broadening the scope of this research topic. The examination of back alleys, a well-defined type of urban space, offers valuable insights for designers when conceptualizing public spaces at comparable scales. The indicators identified in this study can inform the development of more vibrant public spaces. Methodologically, this study innovatively used 3D scanning technology to capture and model back alleys, enabling data extraction for analyzing the relationship between spatial characteristics and overflow rates. This investigative and analytical approach can be adapted to more complex environments.

Nonetheless, this study has certain limitations, including a lack of discussion on the effects of different types of unevenness on the extents and types of overflows, as well as the influence of specific building functions on either side of the back alley. These aspects warrant exploration as potential future research directions.

Acknowledgments

The authors are grateful to Han Ban for her contributions to the data collection.

Disclosure

No potential conflict of interest was reported by the author.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Appendix

Appendix A - Overflow of 72 back-alley units

Appendix

Appendix B - Results of descriptive statistics (n = 72)

References

Hart K. Informal income opportunities and urban employment in Ghana. J Mod Afr Stud. 1973;11(1):61‐89. doi:10.1017/s0022278x00008089

Hindson D, McCarthy JJ. Here to Stay: Informal Settlements in KwaZulu‐Natal. Indicator Press; 1994.

Perlman JE. The metamorphosis of marginality: four generations in the favelas of Rio de Janeiro. Ann Am Acad Pol Soc Sci. 2006;606(1):154‐177. doi:10.1177/0002716206288826

Mohanty M. Challenges of informal urbanization. In: Encyclopedia of the UN Sustainable Development Goals. Springer; 2019:1‐11. doi:10.1007/978‐3‐319‐71061‐7_53‐1

Porter L, Lombard M, Huxley M, et al. Informality, the commons and the paradoxes for planning: concepts and debates for informality and planning. Plan Theory Pract. 2011;12(1):115‐153. doi:10.1080/14649357.2011.545626

el Ghmari S, Zabadi A. Urban informality: its genesis and future beyond stigma and abolition. Afr Mediterr J Archit Urban. 2021;2(5):53‐59. doi:10.48399/IMIST.PRSM/amjau‐v3i1.26746

Roy A, AlSayyad N. Urban Informality: Transnational Perspectives from the Middle East, Latin America, and South Asia. Lexington; 2004.

Dovey K, Polakit K. Urban slippage smooth and striated streetscapes in Bangkok. In: Franck KA, Stevens Q, eds. Loose Space: Possibility and Diversity in Urban Life. 1st ed. Routledge; 2006:19.

Gomez‐Torres A. Temporary appropriation and public space: assessing the CPTED principle of activity support. In: Melis A, Lara‐Hernandez J, Thompson J, eds. Temporary Appropriation in Cities. Springer; 2020:95‐116. doi:10.1007/978‐3‐030‐32120‐8_6

Dovey K, Raharjo W. Becoming prosperous: informal urbanism in Yogyakarta. In: Dovey K, ed. Becoming Places: Urbanism/Architecture/Identity/Power. Routledge; 2009:79‐102. doi:10.4324/9780203875001

Lai A. Temporary appropriation in Shanghai and Hong Kong: two study cases assessing the resilience of women faced with the lack of affordable housing. In: Melis A, Lara‐Hernandez J, Thompson J, eds. Temporary Appropriation in Cities. Springer; 2020:157‐170. doi:10.1007/978‐3‐030‐32120‐8_10

Lara‐Hernandez JA, Melis A. Understanding temporary appropriation and social sustainability. In: Melis A, Lara‐Hernandez J, Thompson J, eds. Temporary Appropriation in Cities. Springer; 2020:11‐26. doi:10.1007/978‐3‐030‐32120‐8_2

Van Oostrum M. Appropriating public space: transformations of public life and loose parts in urban villages. J Urban. 2021;15(1):84‐105. doi:10.1080/17549175.2021.1886973

Aoki Y, Yuasa Y. Private use and territory in alley‐space: hypotheses and tests of planning concepts through the field surveys on alley‐space part 1. J Archit Plan Environ Eng (Transactions of AIJ). 1993;449:47‐55. doi:10.3130/aijax.449.0_47. (in Japanese)

Kobayashi H, Suzuki S. A study on “territory of grouped inhabitants” in small housing groups: part 1 process of territorialization. Trans Archit Inst Jpn. 1981;307:102‐111. doi:10.3130/aijsaxx.307.0_102. (in Japanese)

Alexander C. A City is Not a Tree. Architectural Forum; 1965.

Fujitani H, Kobayashi H. Territoriality and openness of housing space in alley‐space. J Archit Plan (Transactions of AIJ). 2017;82(732):311‐319. doi:10.3130/aija.82.311. (in Japanese)

Kubota A, Honma K, Imai K. Analysis on common passages used and decorated by residents. J Archit Plan (Transactions of AIJ). 2019;84(758):819‐828. doi:10.3130/aija.84.819. (in Japanese)

Rudolfsky B. Architecture without Architects: An Introduction to Non‐Pedigreed Architecture. Doubleday & Company; 1964.

Rapoport A. The Meaning of the Built Environment: A Nonverbal Communication Approach. University of Arizona Press; 1990.

Miao P, ed. Public Places in Asia Pacific CIties: Current Issues and Strategies. Si L, Si R, trans. China Architecture & Building Press; 2007.

Murakami A, Medrial Zain A, Takeuchi K, Tsunekawa A, Yokota S. Trends in urbanization and patterns of land use in the Asian Mega Cities Jakarta, Bangkok, and Metro Manila. Landsc Urban Plan. 2005;70(3–4):251‐259. doi:10.1016/j.landurbplan.2003.10.021

(www.gov.hk). G. Govhk: Hong Kong—the facts. GovHK. January 16, 2024. Accessed February 28, 2024. https://www.gov.hk/en/about/abouthk/facts.htm

DiStefano LD, Lee HY. Hong Kong's back alleys: the ugly, the bad and the good. In: Informal Solutions: Observations in Hong Kong Back Alleys. WE Press Co. Ltd.; 2016:78‐87.

Public Health and Buildings Ordinance. Omeka RSS. 1903. Accessed February 28, 2024. https://oelawhk.lib.hku.hk/items/show/1209

Machado‐León JL, del Girón‐Valderrama G, Goodchild A. Bringing alleys to light: an urban freight infrastructure viewpoint. Cities. 2020;105: [eLocator: 102847]. doi:10.1016/j.cities.2020.102847

Matsushita T, Meguro K, Kubota A. Comparative study of planning history, spatial development and sociological significance of the back alley in Yangon and Singapore. The 18th International Planning History Society Conference, Yokohama, Japan; 2018.

Matsushita T, Kubota A. Study on the revitalization of back drainage space in Yangon City, Myanmar. J Archit Plan (Transactions of AIJ). 2020;85(769):567‐577. doi:10.3130/aija.85.567. (in Japanese)

Public consultation report on strengthening the handling of street obstruction by shops—Legislative Council. Accessed February 28, 2024. https://www.legco.gov.hk/yr14‐15/chinese/panels/ha/papers/ha20150109cb2‐543‐5‐c.pdf. (in Chinese)

Cap. 132 Public Health and Municipal Services Ordinance. Legislation. Accessed February 29, 2024. https://www.elegislation.gov.hk/hk/cap132!en?INDEX_CS=N&xpid=ID_1438402659984_003&SEARCH_WITHIN_CAP_TXT=%E5%90%8E%E5%B7%B7

Zeng F. Kwai Chung ground floor shops are piled with goods in back alleys and roads overnight. District councilors are worried about exacerbating the rodent infestation and urge tougher penalties. HK01. May 26, 2022. Accessed February 29, 2024. https://www.hk01.com/18%E5%8D%80%E6%96%B0%E8%81%9E/722588/%E8%91%B5%E6%B6%8C%E5%9C%B0%E8%88%96%E8%B2%A8%E7%89%A9%E5%A0%86%E6%BB%BF%E5%BE%8C%E5%B7%B7%E5%8F%8A%E9%A6%AC%E8%B7%AF%E9%81%8E%E5%A4%9C-%E5%8D%80%E8%AD%B0%E5%93%A1%E6%86%82%E5%8A%A0%E5%8A%87%E9%BC%A0%E6%82%A3%E4%BF%83%E5%8A%A0%E5%BC%B7%E7%BD%B0%E5%89%87?utm_source=01articlecopy&utm_medium=referral. (in Chinese)

Martin M. Back‐alley as community landscape. Landsc J. 1996;15(2):138‐153. doi:10.3368/lj.15.2.138

Wang X, Fang K, Chen L, Furuya N. The influence of atrium types on the consciousness of shared space in amalgamated traditional dwellings—a case study on traditional dwellings in Quanzhou City, Fujian Province, China. J Asian Archit Build Eng. 2019;18(4):335‐350. doi:10.1080/13467581.2019.1660662

Masuda K, Sadahiro Y, Hino K, Usui H. Physical environment of alleys for the promotion of potted plants. J Archit Plan (Transactions of AIJ). 2018;83(745):447‐453. doi:10.3130/aija.83.447

Zhang Z, Fang K, Wang X, Chen L, Zhang W, Furuya N. Characteristics, correlations of traditional street space elements and tourist density following spontaneous renovation: a case study on Suzhou's Shantang street. J Asian Archit Build Eng. 2020;20(1):29‐43. doi:10.1080/13467581.2020.1781647

Kubota A, Honma K, Imai K. Analysis on common passages used and decorated by residents. J Archit Plan (Transactions of AIJ). 2018;83(748):1015‐1023. doi:10.3130/aija.83.1015

Muminović M. The role of the public‐private interface and persistence of historic character in Nezu, Tokyo. Urban Planning. 2023;8(1):19‐29. doi:10.17645/up.v8i1.6016

Hertzberger H. Lessons for Students in Architecture. Nai010 Publishers; 2016.

Patil AP, Dongre AR. An approach for understanding encroachments in the urban environment based on complexity science. Urban Des Int. 2013;19(1):50‐65. doi:10.1057/udi.2013.12

Feustel M. Everyday monuments. In: Informal Solutions: Observations in Hong Kong Back Alleys. WE Press Co. Ltd.; 2016:6‐9.

Gollob C, Ritter T, Kraßnitzer R, Tockner A, Nothdurft A. Measurement of forest inventory parameters with Apple iPad Pro and Integrated Lidar Technology. Remote Sens. 2021;13(16):3129. doi:10.3390/rs13163129

Luetzenburg G, Kroon A, Bjørk AA. Evaluation of the apple iphone 12 pro lidar for an application in geosciences. Sci Rep. 2021;11(1): [eLocator: 22221]. doi:10.1038/s41598‐021‐01763‐9

Teppati Losè L, Spreafico A, Chiabrando F, Giulio TF. Apple LIDAR sensor for 3D surveying: tests and results in the cultural heritage domain. Remote Sens. 2022;14(17): [eLocator: 4157]. doi:10.3390/rs14174157

Lai C. Unopened Space: Mapping Equitable Availability of Open Space in Hong Kong. Civic Exchange. November 29, 2018. Accessed February 29, 2024. https://civic‐exchange.org/report/unopened‐space‐mapping‐equitable‐availability‐of‐open‐space‐in‐hong‐kong/

Kinoshita H, Nishimura Y. A study on retail markets as determinants of space constitution in Hong Kong. J City Plan Inst Jpn. 1994;29:193‐198. doi:10.11361/journalcpij.29.193

Rodríguez DA, Brisson EM, Estupiñán N. The relationship between segment‐level built environment attributes and pedestrian activity around Bogota's BRT stations. Transp Res D Transp Environ. 2009;14(7):470‐478. doi:10.1016/j.trd.2009.06.001

Jacobs J. The Death and Life of Great American Cities. Pelican; 1965.

Faan S, Jck, Daanngaazai. Pattern, Language, Setting—A Glossary of City Spaces in Hong Kong. Enlighten Fish; 2021.

Cozens PM, Saville G, Hillier D. Crime prevention through environmental design (CPTED): a review and modern bibliography. Prop Manag. 2005;23(5):328‐356. doi:10.1108/02637470510631483

Scarborough BK, Like‐Haislip TZ, Novak KJ, Lucas WL, Alarid LF. Assessing the relationship between individual characteristics, neighborhood context, and fear of crime. J Crim Just. 2010;38(4):819‐826. doi:10.1016/j.jcrimjus.2010.05.010

Sreetheran M, van den Bosch CC. A socio‐ecological exploration of fear of crime in urban green spaces—a systematic review. Urban For Urban Green. 2014;13(1):1‐18. doi:10.1016/j.ufug.2013.11.006

Poon L. Two architects explore the microeconomy of Hong Kong's alleyways. Bloomberg.com. July 15, 2016. Accessed March 1, 2024. https://www.bloomberg.com/news/articles/2016‐07‐15/two‐architects‐explore‐the‐microeconomy‐of‐hong‐kong‐s‐alleyways

Nakamura K, Furuya N. Hawker's behavior on the spontaneous order as “local code” in Bangkok's Chinatown. J Archit Plan (Transactions of AIJ). 2010;75(649):595‐602. doi:10.3130/aija.75.595

Fang K, Wang X, Chen L, Zhang Z, Furuya N. Research on the correlation between pedestrian density and street spatial characteristics of commercial blocks in downtown area: a case study on Shanghai tianzifang. J Asian Archit Build Eng. 2019;18(3):233‐246. doi:10.1080/13467581.2019.1627215

Miles J, Shevlin M. Applying Regression and Correlation: A Guide for Students and Researchers. Sage; 2014.

Blanck M, Ribeiro JL. Smart cities financing system: an empirical modelling from the European context. Cities. 2021;116: [eLocator: 103268]. doi:10.1016/j.cities.2021.103268

Field AP. Discovering Statistics Using IBM SPSS Statistics: And Sex and Drugs and Rock “n” Roll. Sage; 2017.

Hill T, Lewicki P. Statistics: Methods and Applications: A Comprehensive Reference for Science, Industry, and Data Mining. StatSoft; 2006.

Berry WD. Understanding Regression Assumptions. Sage; 1993.

Lewis‐Beck MS. The Sage Encyclopedia of Social Science Research Methods. Sage; 2004.

Lara‐Hernandez JA, Melis A, Coulter CM. Temporary appropriation and informality. In: Temporary Appropriation in Cities. Springer; 2019:59‐77. doi:10.1007/978‐3‐030‐32120‐8_4

Ozili PK. The acceptable R‐square in empirical modelling for social science research. In: Social Research Methodology and Publishing Results. Information Science Reference; 2023:134‐143. doi:10.4018/978‐1‐6684‐6859‐3.ch009

Jiang B, Mak CN, Larsen L, Zhong H. Minimizing the gender difference in perceived safety: comparing the effects of urban back alley interventions. J Environ Psychol. 2017;51:117‐131. doi:10.1016/j.jenvp.2017.03.012

Abe T, Shimizu I. Illegal construction practices on alley space. J Archit Plan (Transactions of AIJ). 2019;84(758):753‐761. doi:10.3130/aija.84.753

Wang Y. [Human Humanity] The owner of Die Tou Rice sells human kindness. HK01. February 22, 2022. Accessed March 1, 2024. https://www.hk01.com/%E7%A4%BE%E6%9C%83%E6%96%B0%E8%81%9E/737366/%E4%BA%BA%E6%83%85%E5%91%B3%E5%95%86%E5%BA%97-%E7%A2%9F%E9%A0%AD%E9%A3%AF%E8%B3%A3%E4%BA%BA%E6%83%85%E5%91%B3-%E8%80%81%E9%97%86-%E6%83%B3%E6%9C%89%E5%80%8B%E5%9C%B0%E6%96%B9%E6%AF%94%E4%BA%BA%E6%8A%96%E6%B0%A3. (in Chinese)

Climate of Hong Kong. Hong Kong Observatory (HKO): Climate. Accessed March 1, 2024. https://www.hko.gov.hk/en/cis/climahk.htm

Hahm Y, Yoon H, Jung D, Kwon H. Do built environments affect pedestrians' choices of walking routes in retail districts? A study with GPS experiments in Hongdae Retail District in Seoul, South Korea. Habitat Int. 2017;70:50‐60. doi:10.1016/j.habitatint.2017.10.002

Robinson MD, Zabelina DL, Ode S, Moeller SK. The vertical nature of dominance‐submission: individual differences in vertical attention. J Res Pers. 2008;42(4):933‐948. doi:10.1016/j.jrp.2007.12.002

Ashihara Y. The Aesthetic Townscape. MIT Press; 1983.

Zhou T, Shen Q. Study on the D/H ratio of city road's environmental effect based on BIM. J Theor Appl Inf Technol. 2013;49(1):448‐455.