Introduction

As the proportion of urban population to the total population continues to rise, the rapid increase in population density has exacerbated the phenomenon of noise pollution, posing a significant threat to the physical and mental health of residents. Numerous studies indicate that noise exposure not only triggers anxiety, irritability, sleep disorders, and daytime sleepiness1,2, but is also significantly associated with increased hypertension rates3,4, impaired cognitive function5, and learning disabilities in school-aged children6. Long-term exposure to noise pollution, particularly for those living near major roads, is linked to significantly higher dementia prevalence and cardiovascular disease risk7,8,9. Chronic exposure to noise has been shown to elevate blood pressure, increase the risk of heart disease, and disrupt sleep rhythms4, with its pathological mechanisms involving autonomic nervous system dysregulation and abnormal secretion of stress hormones10. Notably, impulsive and tonal noises tend to provoke more annoyance than steady-state noise11,12, and the time of individual noise perception13, combined with subjective differences14 further complicates the health impacts.

Urban noise sources are notably diverse, primarily encompassing traffic noise15,16,17,18,19,20,21,22, industrial noise23,24, construction noise25, community noise26, and natural noise27. Noise propagation paths are influenced by both the built environment and natural surroundings28, with key factors such as the distribution of green spaces29, spatiotemporal characteristics, transportation network connectivity, and land use/building morphology30 interacting to shape these pathways. Physical control strategies for noise propagation include the installation of sound barriers31,32,33,34, vegetation for noise reduction, and urban spatial separation design33. Socioeconomic factors significantly mediate noise perception: areas with younger populations and a higher proportion of single residents tend to generate more noise complaints15, while variables such as income level, education, and housing type influence residents’ noise sensitivity, indirectly affecting health outcomes15. Additionally, the spatiotemporal variation in noise propagation11,12 and individual psychological response mechanisms35 collectively present a multidimensional challenge for environmental noise management.

Effective noise control requires the implementation of a"source-path-receptor"intervention system31,32,33,34,36,37,38. Source control prioritizes the use of low-noise vehicles16, optimized traffic flow17, and quieter machinery31, while noise reduction in transportation systems can be achieved through flight path planning19,20 and port operation management21,22. Propagation path control focuses on sound energy attenuation, including upgrading building insulation materials33,34, sound barrier projects32, and urban soundscape planning39,40. On the receptor protection front, strengthening building soundproofing standards36, promoting personal protective equipment37, and improving the regulatory-education collaborative mechanism36,37 are essential. Notably, machine learning-assisted noise map optimization17 and strategic noise management39,40 are becoming crucial tools in smart city governance, and the coordinated implementation of multi-scale interventions38 is key to enhancing urban sound environment quality.

Background

Traditional noise management typically begins with acoustic zoning, which involves classifying areas based on their sound environment characteristics. This zoning serves as the foundation for regulating noise levels and implementing appropriate control measures. Noise mapping, which follows zoning, is a key tool for assessing the acoustic environment within designated areas. It is essential for identifying noise patterns and evaluating potential risks in urban environments41.

However, traditional methods such as surveys and on-site sound level measurements using Class I sound level meters are commonly employed for specific noise assessments, such as evaluating individual shops, activities, or smaller-scale areas. These methods are also used to validate noise maps, ensuring their accuracy and refining noise models. Despite their importance, these traditional approaches are constrained by spatial limitations and are time-intensive, making them less suited for large-scale, city-wide noise management. As a result, noise maps have become an indispensable tool in urban planning, allowing for more efficient assessment, management, and control of environmental noise levels1.

In 2002, the European Union introduced the requirement for strategic noise maps to be generated every five years under the European Noise Directive (END 2002/49/EC)42, with the aim of improving the management of traffic noise. However, capturing real-time changes in the acoustic environment through noise maps remains challenging. The subjective nature of noise complaints complicates the accurate representation of noise perception, as it often varies based on individual sensitivities43.

To bridge the gap between monitoring and decision-making, three promising methods have emerged: dynamic noise mapping, the use of smart sensors and smartphones for environmental noise monitoring, and soundscape studies44. Dynamic noise mapping leverages road surveillance videos to identify traffic and environmental elements, converting them into geographic coordinates to create real-time noise maps45. Similarly, smartphone-based noise mapping, as proposed by Zuo et al., enables users to collect noise data while moving, ensuring measurement quality through the use of external microphones46. In contrast, soundscape studies, which consider both the experiential and acoustic environments, employ an interdisciplinary approach incorporating surveys, audio recordings, and visual data47. Despite their potential, soundscape studies are not yet widely used in standard noise assessments due to their complexity.

In contrast, using complaint data can significantly reduce time costs, as it reflects the public’s perception of noise. Unlike traditional methods that assess physical sound levels, complaint data captures the public’s response once noise reaches a certain threshold. This perception-based approach is particularly valuable for urban management, as it enables the rapid identification of areas where noise is perceived as problematic, without the need for extensive field measurements48.

Noise complaints arise from the objective transmission of noise to individuals, which then prompts a subjective feedback process. While numerous studies have focused on objective sound environments, there is a relative scarcity of research on subjective factors. Most studies addressing subjective responses rely on questionnaires49. However, questionnaires are constrained by factors such as time, space, and the number of respondents, making it difficult to analyze them from a global, macroscopic perspective. Furthermore, the dynamic nature of urban environments complicates the task of adapting noise attributes to reflect changing urban characteristics.

Rainer Guski’s analysis of noise complaints through the lens of psychology suggests that the primary source of noise complaints stems from psychological factors, proposing that individuals’ subjective perceptions have a greater impact on noise complaints than objective environmental conditions50. Huan Tong conducted extensive analyses on urban noise complaints, exploring their relationship with urban density30,51, urban development patterns29, and socio-economic factors52, while also examining the changes in noise complaints during the COVID-19 lockdown53, vividly revealing the symbiotic relationship between noise complaints and urban planning. Noise complaints can serve as an indicator for urban management, reflecting aspects such as urban density, population characteristics, economic conditions, and unemployment rates, thereby enabling governments to plan more effectively and implement targeted governance strategies to promote urban development. To explore the key factors influencing noise complaints, Yalcin Yildirim introduced socio-demographic variables54 and found that traffic-related noise complaints lead to social inequities, with vulnerable groups less likely to report noise issues, exacerbating the impact of noise complaints55. Yaran Jiao incorporated socio-economic factors, land use, and building environment-driven noise management strategies in his analysis, concluding that population density is the most significant driver of noise complaints, with economic investment, housing prices, building heights, and school density also having a significant impact56.

Building on previous research in related fields, this study aims to further explore the factors influencing noise complaints. By collecting noise complaint data from China’s unique noise complaint platform, the study attempts to investigate the relationship between social perception and noise complaints. Additionally, it introduces variables such as population density, urban acoustic zoning, and the spatiotemporal variation of noise complaints, in order to validate the research of previous scholars and explore potential reasons for the differing conclusions.

The 12,345 government service platform serves as a key feedback platform for Chinese citizens to address their concerns. Through this 24-h online platform, local governments collect and categorize complaints from citizens, which are then forwarded to relevant authorities for resolution. Feedback is provided promptly to address citizens’ needs effectively57,58. The 12,345 platform leverages big data processing and artificial intelligence technologies for the classification, supervision, processing, and feedback of complaints. This data is crucial for supporting city planning and addressing urban issues, offering substantial opportunities for improvement59.

Therefore, studying urban noise complaints from the perspective of social perception is both unique and essential, particularly when compared to previous approaches that relied solely on objective data or incomplete questionnaire surveys. This study aims to: 1) uncover the temporal and spatial distribution patterns of noise complaints in Hangzhou, 2) analyze how human activities influence noise complaints using the spatial distribution characteristics of POI. 3) utilize the noise complaint data from the 12,345 platform to examine the factors influencing urban noise complaints from a social perception perspective, in order to propose solutions and strategies for improving the acoustic environment and to inform future urban noise management.

The initial hypotheses of this study include: 1) In the temporal dimension, the distribution of urban noise complaints exhibits regularity and fluctuation, significantly influenced by factors such as season, weekends, and different time periods throughout the day. Noise complaints are expected to decrease during seasons with comfortable temperatures, and to be more frequent during rest periods (e.g., nighttime and weekends) compared to work periods. 2) In the spatial dimension, social perception significantly affects both the quantity and type of noise complaints, with higher frequencies of complaints observed in areas with higher population and POI densities, and more pronounced noise complaints in wealthier areas.

Overview of the study area

The study focuses on Hangzhou City (Fig. 1), located in Zhejiang Province, China. As the provincial capital, Hangzhou is a key hub for economy, culture, education, and tourism in the Yangtze River Delta region. It is also recognized as one of China’s National Famous Historical and Cultural Cities. Geographically, the city lies in the southern part of the Yangtze River Delta, at the western terminus of Hangzhou Bay. The city is composed of ten districts: Shangcheng, Gongshu, Xihu, Binjiang, Xiaoshan, Yuhang, Linping, Qiantang, Fuyang, and Lin’an, as well as one county-level city, Jiande, and two counties, Tonglu and Chun’an.

Fig. 1
figure 1

Hangzhou City zoning map. (The Map is derived fromAPI version 3.0,https://lbsyun.baidu.com).

Full size image

China’s Technical Specifications for Regionalizing Environmental Noise Function60 categorizes cities into five distinct types (Table 1). According to data from the Hangzhou Ecological Environment Bureau (Figs. 2, 3, 4, and 5), the table uses the following notations: M represents the Main Districts (Shangcheng, Gongshu, Xihu, Binjiang, Xiaoshan, Yuhang, Linping, Qiantang, Fuyang); T denotes Tonglu County; C refers to Chun’an County; J represents Jiande County; and 1, 2, 3, and 4 correspond to sound functional zones of categories 1, 2, 3, and 4 in the region.

Table 1 Environment noise limit. unit: dB(A).
Full size table
Fig. 2
figure 2

Noise Situation of Hangzhou Sound Function Area in The First Quarter.

Full size image
Fig. 3
figure 3

Noise Situation of Hangzhou Sound Function Area in The Second Quarter.

Full size image
Fig. 4
figure 4

Noise Situation of Hangzhou Sound Function Area in The Third Quarter.

Full size image
Fig. 5
figure 5

Noise Situation of Hangzhou Sound Function Area in The Fourth Quarter.

Full size image

Noise in the Main Districts is primarily generated by traffic, including road and rail transport, as well as commercial activities. Due to high population density and significant traffic, these areas experience a mix of office, residential, and commercial noise. In contrast, the counties of Tonglu, Chun’an, and Jiande, with lower population densities and less developed infrastructure, are mainly impacted by noise from agricultural machinery, rural activities, and natural environments, with minimal influence from industrial or heavy traffic noise.

As the sound environment functional zones’ classification increases, a clear trend of rising daytime and nighttime equivalent sound levels is observed. The highest daytime equivalent sound level, 63 dB(A), occurs in Tonglu County’s category 3 functional areas during the fourth quarter, while the lowest level, 51 dB(A), is observed in Chun’an County’s category 1 functional areas during the first quarter. Similarly, the highest nighttime equivalent sound level, 59 dB(A), is found in Tonglu County’s category 3 functional areas in the fourth quarter, while the lowest is 48 dB(A) in Tonglu County’s category 2 functional areas in the third quarter.

The largest variation in the day-night equivalent sound pressure levels occurred in the third quarter in Tonglu County’s category 2 functional area, with a change of 21 dB(A). In contrast, the smallest variation was observed in the second quarter in Chun’an County’s category 1 functional area, with a change of only 0.8 dB(A).

As the sound environment category decreases (from Category 1 to Category 4), equivalent sound pressure levels show an increasing trend between day and night. In terms of temporal variation, spring exhibits the most stable changes in day-to-night sound pressure levels, while winter displays the most pronounced fluctuations. This seasonal variation is mainly attributed to reduced nighttime activity during the colder winter months, which increases the gap between daytime and nighttime sound levels.

From a regional perspective, Tonglu County exhibits the largest variation in day-to-night equivalent sound pressure levels. Although daytime sound levels in this area are comparable to those in other regions, the nighttime changes are notably more pronounced. This can be attributed to the area’s economic underdevelopment, low population and building density, and extensive mountainous terrain. The significant reduction in nocturnal activity in such regions results in a rapid decline in nighttime sound levels, further accentuating the contrast between day and night.

In contrast, functional zones 1 and 2, typically located in the city’s core, experience minimal changes in nighttime sound levels. This is because urban areas in these zones continue to have nighttime city activities and noise, Maintaining a relatively constant sound environment. On the other hand, in category 3 and 4 functional zones—often associated with industrial production and transportation—there is a more pronounced decline in nighttime sound levels due to reduced human activity.

Research has shown that higher equivalent sound levels are associated with poorer recall performance and increased aggravation ratings, suggesting that elevated noise levels in certain functional zones may have a negative impact on residents’ well-being and cognitive function61.

Results

Time-varying characteristics of noise complaints

Noise complaints were analyzed concerning the season, time of day, and weekday status to conduct a comprehensive examination of temporal variations.

Characteristics of seasonality changes in noise complaints

Examination of seasonal noise complaints (Fig. 6) reveals a clear clustering pattern, with complaints showing an upward trend in the first half of the year and a downward trend in the second half. The highest number of complaints occurs in summer (4,867), while winter records the lowest (2,201). The variation across the four seasons indicates that the transition from spring to summer sees the highest increase in complaints, while the shift between summer and autumn exhibits the greatest fluctuation compared to other seasonal transitions.

Fig. 6
figure 6

Seasonal variation in the five noise complaints.

Full size image

This study categorizes noise sources into construction, community, traffic, industrial, and natural sources. Analysis of quarterly data reveals that the increase in complaints during the transition from spring to summer is the highest among all quarters, with summer complaints peaking. Construction noise shows the most significant seasonal fluctuations, with a disparity of up to 2,376 complaints between seasons. The degree of seasonal fluctuation in complaints is ranked from largest to smallest as follows: construction, community, traffic, industrial, and natural noise.

Time-varying characteristics of noise complaints

Analysis of noise complaints over a 24-h period (Fig. 7) reveals distinct temporal patterns. The complaint frequency peaks at three key times: from 8 to 11 a.m., from 2 to 4 p.m., and from 10 p.m. to 1 a.m. Complaints are lower between 2 to 7 a.m., 12 to 1 p.m., and 5 to 9 p.m. In the morning, complaints surge around 8 a.m., leveling off by 11 a.m., followed by a slight increase around 2 p.m. Complaint frequency drops to around 200 by 4 p.m., reaching its lowest point of 178 by 7 p.m., before rising again at 10 p.m., peaking at 1,644 complaints by 11 p.m., and gradually decreasing to around 200 complaints by 2 a.m.

Fig. 7
figure 7

Changes in five noise complaints within a day.

Full size image

Examining the five types of noise complaints shows an increase during the same key periods: 8 to 11 a.m., 2 to 4 p.m., and 10 p.m. to 1 a.m. Construction noise peaks at 11 p.m., while the other types peak at 9 a.m. The lowest point for construction noise occurs at 7 p.m., while the other types reach their lowest between 3 and 5 a.m. Construction noise exhibits the most pronounced temporal fluctuations, with sharp spikes at 8 a.m. and 9 p.m., and a significant rise after 9 p.m., coinciding with the highest overall complaint peak of the day.

Characteristics of weekday and rest day variations in noise complaints

In this study, we differentiate between weekdays and weekends to better understand the daily distribution of noise complaints (Fig. 8 and 9). The analysis reveals that the average number of noise complaints on weekends throughout the year is 1,687, while the average number of complaints on weekdays (Monday to Friday) is 2,025. Noise complaints on weekends are significantly fewer than those on weekdays, and overall, the number of complaints decreases from Monday to Sunday, weekends exhibit greater fluctuations, but with smaller overall changes compared to weekdays. The variation in complaint numbers on weekends follows a smoother pattern, with changes occurring at similar time points, making it more consistent than on weekdays. On weekdays, the fluctuations are more pronounced, indicating a greater variation in complaint frequency throughout the day.

Fig. 8
figure 8

Five types of noise complaints on weekdays.

Full size image
Fig. 9
figure 9

Five types of noise complaints at weekends.

Full size image

Characterization of the spatial distribution of noise complaints

Overall spatial distribution of noise complaints

All noise complaint locations were mapped using GIS software (Fig. 10, left), overlaid on Hangzhou’s urban layout, and the Spatial Analyst tool in GIS was used for kernel density analysis of the noise complaints (Fig. 10, right). The analysis reveals that noise complaints are concentrated primarily in the central urban districts of Hangzhou, particularly in Gongshu, Xihu, Binjiang, and Shangcheng Districts. Notably, the Changhe Sub-district in Binjiang District reported the highest number of complaints, with 657. This concentration reflects a higher frequency of complaints in densely populated urban areas, along major transportation routes, and in high-traffic zones.

Fig. 10
figure 10

Spatial distribution (left) and kernel density (right) of overall noise complaints.

Full size image

Although the spatial distribution of the five types of noise complaints—construction, social activities, transportation, industrial, and natural—overlap in many areas, each category exhibits distinct spatial characteristics. These findings align with global trends observed in major urban centers, where noise complaints often correlate with population density, infrastructure development, and transportation networks.

Spatial distribution of different types of noise complaints

Industrial noise complaints are concentrated in Xiaoshan and Lin’an Districts (Fig. 11), with notable clusters in Guali and Xintang Sub-districts in Xiaoshan, and Jincheng, Jinbei, and Qingshan Lake Sub-districts in Lin’an. The high factory density in Guali and Xintang, combined with population growth in Lin’an, contributes to the elevated complaint levels. Sub-districts like Jincheng and Qingshan Lake, associated with high-tech industrial parks, report higher noise levels due to ongoing industrial activities. Additionally, rural noise sources, such as grass mowing, further contribute to the overall increase in complaints.

Fig. 11
figure 11

Spatial distribution (left) and kernel density (right) of Industrial noise complaints.

Full size image

Construction and transportation noise (Figs. 12 and 13) are primarily concentrated in Hangzhou’s older urban areas. In particular, Binjiang District, especially Changhe, Xixing, and Puyan Sub-districts, ranks highly for construction-related complaints, likely due to rapid development and preparations for the Asian Games. These areas, with dense populations and high building densities, also rank among the top for transportation-related noise. Fuchun Sub-district in Fuyang and Jincheng Sub-district in Lin’an are similarly affected by traffic-related noise, driven by increased car ownership, traffic congestion, and nocturnal car racing, which further exacerbate the issue.

Fig. 12
figure 12

Spatial distribution (left) and kernel density (right) of Construction noise complaints.

Full size image
Fig. 13
figure 13

Spatial distribution (left) and kernel density (right) of Traffic noise complaints.

Full size image

Social and natural noise complaintsexhibit similar spatial distribution patterns. These complaints are more prevalent in densely populated urban Sub-districts, where human activities and environmental factors, such as animal noise (e.g., barking dogs), contribute to the noise levels (Fig. 14 and 15).

Fig. 14
figure 14

Spatial distribution (left) and kernel density (right) of Community noise complaints.

Full size image
Fig. 15
figure 15

Spatial distribution (left) and kernel density (right) of Nature noise complaints.

Full size image

Characterization of the spatial distribution of POI

Overall spatial distribution of POI

Human activities and noise generation are closely linked, as demonstrated by the correlation between POI distribution and noise complaints. GIS analysis (Fig. 16) shows a strong relationship between POI density and noise complaints, particularly in Hangzhou’s urban core, where the highest concentration and variety of POI are located. In contrast, the number and diversity of POI decrease as one moves outward to Lin’an District, Chun’an County, Fuyang District, Tonglu County, and Jiande City, reflecting a decline in density.

Fig. 16
figure 16

Spatial distribution (left) and kernel density (right) of POI.

Full size image

The central region, including Gongshu, Xihu, Shangcheng, and Binjiang Districts, houses about 35% of Hangzhou’s population within just 3% of the city’s land area. This high population density, coupled with these areas being the political, economic, and cultural centers of Hangzhou, drives the concentration of POI and, consequently, the higher frequency of noise complaints.

Spatial distribution of different types of POI

To improve the accuracy of the correlation analysis, the POI data is categorized into 14 types, and individual kernel density analyses are conducted (Fig. 17). The results show that most POI types are concentrated in Gongshu, Xihu, Shangcheng, and Binjiang Districts, while other districts exhibit a radial dispersion pattern, with POI clustering around district centers and diminishing towards the outskirts. Notably, POI density is higher in the northwest compared to the southwest, influenced by population density and topography. The northwest features plains, while the southwest is more mountainous, limiting urban development and concentrating POI along roads and in urban areas.

Fig. 17
figure 17

Kernel density of different types of POI.

Full size image

This study examines the distribution of 14 types of POI and their influence on noise complaints. POI related to basic needs, such as catering, food, transportation, residential services, and living facilities, are distributed in a Manner that reflects the overall urban layout of Hangzhou. Catering and food outlets, essential for daily life, are spread widely across the city, while transportation facilities and business residences are more concentrated, especially in urban centers due to government planning. Catering establishments Make up the largest proportion of POI, and parking lots, which account for 64.6% of transportation facilities, significantly shape the distribution of transportation-related POI. Residential service POI, serving densely populated areas, are most prevalent in the Upper City, Xihu, and Gongshu Districts. Public toilets, a crucial component of living services, are strategically located in high-density areas, similar to business residences (Fig. 17-1a, 1b, 1c, 1 d).

The distribution of other POI, including those related to shopping, education, entertainment, sports, and health services, is also concentrated in urban areas. Shopping POI, which account for nearly one-third of all POI, are primarily located in affluent districts, reflecting higher consumption rates. Educational and cultural institutions are clustered around urban centers, with a notable concentration of training institutions, comprising 45.4% of educational POI. Similarly, automobile services, leisure and entertainment venues, and fitness centers (24.7% of sports-related POI), along with healthcare facilities (with medical sales comprising 35.7% of health-related POI), reflect population demand, collectively constituting a significant portion of the city’s POI (Fig. 17-2a, 2b, 2c, 2 d, 2e, 2f).

Employment-related POI, such as those from companies and financial institutions, show a balanced distribution across the city, with higher concentrations in densely populated urban areas to meet local employment needs. Corporate POI are more prominent within the city, while areas outside the city host a higher concentration of industrial facilities. This pattern reflects the relocation of factories from Hangzhou’s urban core, which primarily engage in tertiary sector activities (Fig. 17-3a, 3b).

Tourist-related POI, such as hotel accommodations and attractions, are concentrated in downtown Hangzhou, known for its historical and cultural sites. The city center hosts a wide array of tourist attractions, while some sites take advantage of the surrounding mountainous landscapes, clustering in these areas. The distribution of hotel accommodations aligns closely with tourist demand, particularly in the downtown and surrounding areas (Fig. 17-4a, 4b).

Coupled analysis of noise complaints with the POI and population density

Spatial correlation analysis was conducted in GIS software using multivariate tools to examine the relationships among various types of noise (construction, community, traffic, industrial, and natural noise), overall noise complaints, the distribution of different POI, and population density.

As shown in Table 2, the correlation analysis reveals a positive relationship between all categories of POI and noise complaints, with the number and density of POI significantly influencing the spatial distribution of noise complaints. Specifically, exercise and fitness, commercial residence, and transportation POI have the most substantial impact on overall noise complaints, while scenic spots exhibit the weakest correlation. Construction noise complaints show the weakest correlation with POI, with exercise and fitness POI having the highest correlation, and tourist attractions the lowest among all noise types and POI correlations. Community and natural noise complaints, on the other hand, show stronger correlations with POI categories than other types of noise. Leisure and entertainment POI demonstrate the highest correlation, while scenic spots exhibit the weakest. Exercise and fitness POI have the strongest correlation with traffic noise, while scenic spots show the weakest correlation. Hotel accommodations exhibit the highest correlation with industrial noise, whereas shopping POI have the lowest correlation. Overall, the distribution of POI shows the highest correlation with community noise and the lowest with construction noise.

Table 2 Correlation analysis based on GIS.
Full size table

According to the population density map of Hangzhou (Fig. 18) and the correlation analysis of various noise complaints (Table 2), while the correlation between population density and noise complaints is significantly lower than that with POI, it is evident that natural and community noise complaints exhibit a stronger correlation with population density than other types of noise. In contrast, construction and industrial noise complaints show relatively lower correlations. The order of correlation strength between noise complaints and population density, from weakest to strongest, is as follows: construction noise, overall noise complaints, industrial noise, traffic noise, community noise, and natural noise.

Fig. 18
figure 18

Population density map(The Map is derived fromAPI version 3.0,https://lbsyun.baidu.com).

Full size image

Discussion

Spatial and temporal patterns of noise complaints

The cyclical nature of human activities results in fluctuations in noise complaints over time62. Notably, there is a marked increase in noise complaints during certain months, such as March–May and August–September, compared to other periods, with an average of approximately 800 complaints during these months.

At this time, the favorable climate and increased human activity lead to a rise in noise levels. These findings are consistent with existing studies, which similarly suggest that seasonal variations influence human activities, thereby affecting the occurrence of noise complaints. Furthermore, factors such as humidity, air pollution, and level of wind contribute to social unrest, further exacerbating noise complaints63.

Analysis of 24-h noise complaint data revealing strong periodicity64, with peaks occurring from 8–11 a.m., 2–4 p.m., and 10 p.m.−1 a.m.. These peaks correspond to human activity cycles: the morning peak aligns with early work-related activities, the afternoon peak follows post-lunch activities, and the late-night peak occurs as people seek rest after a busy day. The rise in complaints during these periods is closely linked to individuals’ rest times65, with nighttime disturbances particularly affecting sleep quality and leading to health issues such as hypertension and cognitive impairment66. A comparison of weekday and weekend data reveals similar trends, though weekend complaints are generally lower and less volatile, except between 8 p.m. and 11 p.m., when fluctuations are more pronounced. The overall pattern shows increased sensitivity to noise during rest periods, especially at night67. Consistent with other studies, weekends experience more moderate increases in complaints, with longer intervals between fluctuations68.

Our findings align with the noise regulations outlined in the Hangzhou Municipal Environmental Noise Management Regulations. Specifically, the use of high-noise equipment is restricted in areas with noise-sensitive buildings, and work is prohibited during designated quiet periods, including statutory holidays, weekends, and weekdays from 12 a.m. to 2 p.m. and 8 p.m. to 8 a.m. Additionally, noise-generating renovation activities are prohibited in occupied buildings during these timeframes. Notably, the peak noise complaint periods observed in this study, specifically from 8 a.m. to 11 a.m. and 2 p.m. to 4 p.m., are significant and should be considered for inclusion in future regulatory frameworks. These results could inform refinements to existing noise management regulations.

In Hangzhou, noise complaints are most frequent in urban districts such as Gongshu, Xihu, Binjiang, and Shangcheng. Industrial noise is concentrated in secondary industry zones, such as Guali Town and Xintang Sub-districts in Xiaoshan, and Jincheng and Qingshan Lake Sub-districts in Lin’an. Construction noise predominates in rapidly developing areas, including Changhe and Xixing Sub-districts in Binjiang, and Sandun and Cuiyuan Sub-districts in Xihu, typically associated with economic growth, high-rise buildings, and transportation noise, particularly in Fuyang and Lin’an Districts. Densely populated areas, such as Jincheng Sub-district, also report noise from events like drag racing. Social and natural noise complaints, particularly from barking dogs, are concentrated in downtown areas with higher population densities and affluence69,70.

Our study also reveals that districts with larger, uneven layouts and higher population clustering tend to experience more noise complaints. Consistent with existing research, areas with a higher proportion of residents employed in primary and secondary industries report fewer noise complaints, while those with a larger proportion of tertiary sector workers experience more.

Impact of population activity on noise complaints

An analysis of urban POI and noise complaints reveals that ambient noise levels are significantly higher in areas with mixed land uses71. These areas also experience a greater frequency of noise complaints due to the diverse activities associated with different land uses, resulting in various types of noise72. In Hangzhou, numerous POI are concentrated in the urban core and extend outward in a radial pattern.

When coupled with population density maps, the analysis shows that natural and social noise complaints are more strongly correlated with population density, while industrial and construction noise exhibit weaker correlations. Denser populations, often associated with a higher number of pets, correlate with increased natural noise complaints. In contrast, industrial and construction noise are concentrated in low-density suburban or undeveloped areas. These findings suggest that noise complaints are more prevalent in service-oriented regions with high population densities73. Population density emerges as a key predictor of high-risk areas for various types of urban noise, particularly community and natural noise, offering valuable insights for targeted noise mitigation strategies.

Noise complaints show a positive correlation with all types of POI, with the overall POI distribution exhibiting the strongest correlation with social life noise and the weakest with construction noise. This suggests that higher POI density is associated with increased noise complaints74, although construction noise has the least influence on total complaint numbers. Among various types of noise complaints, scenic locations demonstrate the weakest correlation with overall noise, while shopping consumption shows the weakest correlation with industrial noise.

Exercise and fitness activities exhibit the strongest correlation with overall noise, construction noise, and traffic noise. This is due to the concentration of fitness centers in densely populated areas—such as commercial districts, high-density residential zones, and emerging urban areas—typically characterized by high population density and economic development. Increased population mobility in these areas contributes to heightened noise levels. In contrast, leisure and entertainment activities show the strongest correlation with community and natural noise. A notable correlation is also observed between hotel accommodations and industrial noise, as residential areas and hotels located near industrial zones tend to report higher levels of noise complaints.

The proximity of residential spaces to noisy environments significantly affects residents’ tolerance levels. Noise in residential areas, disrupting the comfort and tranquility of home, is associated with non-auditory health effects such as hypertension, cardiovascular diseases, irritability, and sleep disturbances. These findings align with previous studies, which highlight that neighborhoods offering more comprehensive services tend to generate more noise. This trend is particularly evident in residential areas, where noise complaint densities are highest.

Social perception data and urban noise management

To analyze the correlation between social perception and the spatial–temporal distribution of noise, and to derive actionable noise management strategies, we collected noise complaint data through government channels. We then selected POI from the Geospatial Service and performed spatial correlation analysis using the GIS software.

Spatio-temporal dimension

From both spatial and temporal perspectives: Implementing stricter noise control policies is crucial to address the significant rise in noise complaints. Consistent with previous studies, we confirm that noise complaints peak during autumn and spring. However, contrary to earlier findings, complaints do not decline after 8 a.m.; instead, they show a peak between 8 a.m. and 11 a.m16.. The fastest increase in noise complaints occurs at 8 a.m. and 9 p.m., with the rapid rise lasting approximately one hour, after which it decreases and tends to stabilize. A full fluctuation cycle lasts about three hours. Therefore, noise Management should focus on the key time points of 8 a.m. and 9 p.m., adopting segmented time-based control measures. Policies should be introduced to limit noise levels during the periods of 8 a.m. to 11 a.m. and 9 p.m. to 2 a.m., such as restricting construction activities, noise from commercial areas, and traffic honking. Extensive studies have shown that segmented noise control can significantly reduce noise complaints. Furthermore, to address the sudden surge in noise complaints, increasing personnel to manage complaints and enforce noise regulations can accelerate processing times and reduce complaint rates.

Data analysis reveals a total of 8,166 complaints related to nighttime construction noise, while only 261 of these complaints are associated with authorized nighttime construction permits. Therefore, further restrictions on construction hours and procedures are necessary. Specifically, construction activities generating environmental noise pollution should be prohibited from 12:00 a.m. to 7:00 a.m. the following day. The study noted complaints regarding nighttime construction with valid permits, underscoring the need for stricter requirements and oversight of night construction permits. Continuous nighttime construction should only be approved by the environmental protection department, with established noise limits, operational restrictions to minimize noise pollution, specified continuous operation hours, and public notifications to nearby residents regarding construction activities75.

Socio-perceptual dimension

Analysis of POI correlations reveals a significant relationship between human activity density76, particularly at POI, and the frequency of noise complaints77.

1) Among various POI, sports and fitness facilities show the weakest correlation with noise complaints. This correlation is strongly influenced by factors such as population density, economic development, and the presence of high-density, high-rise areas, which are key contributors to noise pollution72. To mitigate this, high-density residential areas should be located further from major traffic routes, and more landscaped green spaces should be incorporated. This approach would reduce noise exposure, increase distance from sources of noise, and help alleviate the psychological impact of noise. Moreover, positioning public and commercial buildings along roadways can act as effective sound barriers.

2) The findings of this study are consistent with other research, which shows a marked reduction in noise levels near scenic spots. Furthermore, some studies suggest that green vegetation is particularly effective in mitigating noise disturbances. These areas often feature substantial green coverage, which acts as a natural sound barrier while also enhancing residents’ sense of security78. Forested areas, in particular, are highly effective in reducing noise, highlighting the importance of managing urban green space indicators such as green coverage79. To optimize this, it is crucial to expand the use of appropriate tree varieties, including low-branching and broad-leaved species, based on the principle of"appropriate trees for appropriate locations."A comprehensive noise reduction strategy should combine trees, shrubs, and herbs, incorporating the stepped forest belt concept and promoting three-dimensional green planting as a form of noise prevention wall.

Government dimension

Research indicates that high-density urban areas, characterized by concentrated buildings, transportation infrastructure, and industrial or commercial enterprises, exhibit heightened sensitivity to noise. To mitigate this impact, the installation of sound barriers is recommended to limit noise propagation80.

Consistent with other studies, we argue that refining acoustic environment functional zones is crucial for improved noise control81. This refinement should be based on acoustic testing and noise complaint data. The current concentration of complaints in urban areas suggests the effectiveness of the existing zoning framework. Urban master plans must integrate sound environmental protection measures, including mandatory noise control regulations, to maintain tranquility. Moreover, regional development and urban transformation should consider acoustic factors during construction planning. Achieving a high-quality acoustic environment requires a thorough understanding of noise dynamics and a coordinated approach to functional zoning82.

During the process of collecting noise complaint data, we encountered difficulties in data collection due to overlapping responsibilities between different departments. Therefore, our research highlights that addressing noise pollution effectively requires clear delineation of responsibilities among government departments. Overlapping authority in sectors such as transportation, urban Management, and environmental protection can cause inefficiencies. Strengthening noise control enforcement agencies and clarifying authority divisions are critical for effective Management. Additionally, the 12,345 complaint platform, which played a pivotal role in data collection for this study, demonstrates significant value in government operations due to its widespread use and established role in handling public grievances.

Our findings align with existing literature, demonstrating that noise complaints are closely related to factors such as population density, industrial structure, transportation networks, and socioeconomic conditions. In light of this, we propose a shift from the traditional"government-people"model of noise management, which predominantly relies on centralized government regulations, to a more inclusive"people-Sub-districts-government"framework. In this new model, Sub-districts play a crucial role as key urban units where residents interact and engage with their environment. This bottom-up approach emphasizes the importance of involving local communities and Sub-district-level governance in the noise management process. By implementing targeted noise control measures at the Sub-district or community level, urban noise mitigation efforts can be significantly strengthened, thereby fostering more responsive and context-sensitive policies that reflect the lived experiences and concerns of local populations.

Strengths of the study

The government primarily utilizes noise maps for noise control, depicting the sound level distribution in the study area to differentiate between noisy and quiet areas. However, the high production cost and limited time frame of noise maps make long-term monitoring difficult2. Furthermore, objective noise statistics may not fully capture the varied public perspectives on noise issues, and involving the public in noise assessment can enhance the accuracy and relevance of the findings48. Data on social perception collected by the government ensures data accuracy, providing insights into the problem’s scope and people’s subjective noise evaluations. This approach, compared to conventional objective data measurement, saves resources and reflects real-time, dynamic noise trends.

Limitation

The limitation of this study is that: 1) Research suggests that COVID-19 home quarantine measures significantly impacted noise complaints, and the spatial–temporal distribution of these complaints May have been influenced by pandemic-related government interventions. It is important to note that the study period for noise complaints encompasses the year 2022. 2) The observation period of this study is limited to a single year (2022–2023), which May have been influenced by specific climatic or policy conditions of that year. It is recommended that future research incorporate multi-year data to better predict the relationship between social perception data and urban noise complaints, thereby enhancing the generalizability of the conclusions. 3) Analyzing urban issues solely from the perspective of noise is insufficient. The study of noise complaints does not allow for the identification of areas with positive sound qualities that should be protected and enhanced. A multisensory approach to urban studies, integrating visual68, olfactory69, and auditory analyses, will foster deeper research into the creativity, relevance, and social aspects of soundscapes, ultimately contributing to more effective urban noise management.

Conclusion

Objective noise data collection often fails to fully meet public needs, underscoring the growing importance of subjective noise evaluation in today’s rapidly changing urban environments. This study explores the relationship between the spatial–temporal distribution of noise complaints and social perception, aiming to enhance urban noise pollution management and establish a theoretical foundation for noise planning and regulation.

The data analysis shows that Hypothesis 1 is partially supported. It was found that noise complaints do indeed exhibit significant fluctuations at different times of day and seasons, with a noticeable increase in complaints during nighttime. However, the hypothesis regarding seasons was not supported, as noise complaint levels were neither highest nor lowest during favorable weather conditions. Additionally, the hypothesis of more noise complaints on weekends was refuted. Hypothesis 2 was confirmed, as the study shows a significant increase in noise complaints in areas with higher POI and population densities, especially in wealthier regions, where the frequency and variety of complaints are more diverse.

By incorporating social perception, this study bridges the gap between residents’ subjective experiences of noise and actual noise levels, moving beyond the sole reliance on decibel measurements. This approach provides a fresh perspective on urban noise management. Using residents’ complaints as input for urban planning allows for the integration of their perceptions into noise control strategies, offering a more intuitive reflection of the true acoustic environment and fostering innovative ideas for community involvement in urban development. This method also presents a comprehensive way to analyze the acoustic environment, addressing challenges faced by the city in previous assessments.

Methods

Data sources

This study utilized noise complaint data from the 12,345 government service platform, which gathers resident complaints on various issues, including noise. The dataset spans the entire year of 2022 and contains a total of 13,812 noise complaints. After excluding 308 invalid entries, 13,504 valid complaints were retained for analysis.

Additionally, POI data from the Geospatial Service was used to identify specific locations such as stores, restaurants, parks, and other attractions in Hangzhou. These locations serve multiple purposes, including user searches and route planning, and are typically labeled with relevant information such as text, images, and geographic coordinates. The 2022 POI data for Hangzhou includes four key attributes: name, category, classification, and geographic coordinates. The dataset covers 14 major categories, including exercise and fitness, healthcare, leisure and entertainment, services for daily life, commercial residences, automotive services, scenic spots, science and education, hotels, financial institutions, transportation, shopping, incorporated businesses, and food and beverage.

Processes and steps

Noise complaint data processing

The noise complaint data is categorized based on claim content, location, letter format, issue purpose, and registration time. Since noise complaints are primarily in text format, they exhibit varied expressions due to differences in language styles.

To process this data, a series of preprocessing steps were applied, including semantic disassembly and keyword sorting, to extract essential information such as time, location, noise type, and source. Initially, due to privacy restrictions, the location data in the original dataset is only accurate to the community level. For mapping purposes, the noise complaint locations were geocoded based on the community centers. Additionally, given the colloquial nature of the descriptions, noise types were classified using keyword filtering in Spreadsheet Software. A word cloud analysis was conducted on complaints of the same noise type, identifying high-frequency keywords such as “construction,” “site,” “building,” and “excavation” for construction noise.

Time-related information, including specific weekdays and months, was also extracted using Spreadsheet Software, with the formula = WEEKDAY(XX,1) employed to identify weekends.

This study categorized the 13,504 valid noise complaints into the following types based on their sources: 9,437 related to construction noise, 2,732 related to social life noise, 741 related to transportation noise, 362 related to industrial production noise, and 232 related to natural sounds. Among the 13,504 valid noise complaints, we screened and distinguished the repeat complainants. A total of 13,136 individuals submitted only one complaint, 356 individuals submitted two complaints, and 12 individuals submitted three or more complaints. Since these complaints were caused by ongoing noise issues, all of them have been retained.

Regionalization of noise complaints

This study aims to transition from the traditional ‘government-people’ model, where noise problems are primarily governed by centralized government regulations, to a more inclusive ‘people-Sub-districts-government’ framework. This shift emphasizes a bottom-up approach to urban noise management, driven by social perception, ensuring that noise management policies reflect the concerns and experiences of local communities at the street level. This is achieved by extracting information from the complaint texts, using Sub-districts as the primary spatial unit for noise complaints, and recording the complaints for each Sub-district in Hangzhou. This adjustment enhances the efficiency of spatial analysis in the subsequent stages.

Visualizing spatially and temporally distributed noise complaints

The dataset was used to define Sub-district boundaries, incorporating details of noise complaints, such as time, source, and frequency for each area. Statistical analysis was conducted using Spreadsheet Software, while GIS software was employed for visualizing the noise data. Additionally, the spatial distribution of noise complaints was analyzed in relation to POI to assess the patterns of spatial variation.

Methods of analysis

Kernel Density Analysis in GIS software is a spatial technique used to evaluate event density within a defined area. It identifies high-density regions around specified coordinates, generating hotspot maps that reveal underlying patterns and trends. These maps are valuable for decision-making in resource management, transportation, and urban planning.

In this study, Kernel Density Analysis was employed to Map the distribution of noise complaints in Hangzhou. Using the natural interval classification method, the density was categorized into 10 levels, producing a distribution map that highlights the spatial characteristics of noise complaints and their relationship to various POI. This visualization provides an objective representation of spatial patterns.

Spatial correlation analysis was conducted using a multivariate approach. The correlation between noise complaints and POI distributions was examined using the band ensemble statistical tool. Pearson correlation coefficients were calculated to derive covariance and correlation matrices for each raster layer, all processed with GIS software.