How Land-Use Planning Deeply Affects the Spatial Distribution of Composite Soundscapes

Abstract

Urban noise pollution poses a significant obstacle to sustainable development by compromising public health and well-being. Within this context, the soundscape emerges as a critical component in creating healthier and more livable cities. To further investigate the relationship between urban land-use planning characteristics and soundscape distribution, this study examines the spatial distribution of urban soundscapes and urban spatial functions. It explores the influence of urban land-use types on both the acoustic environment and soundscape perception and evaluation, aiming to better understand the influencing factors and dynamics of composite soundscapes in urban environments. The results show that (a) acoustic environment characteristics and soundscape perception evaluations are influenced by urban land-use function, exhibit a spatial aggregation effect, and are affected by the surrounding environment. (b) The key acoustic indices affecting the perception and evaluation of urban soundscapes are the equivalent continuous A-weighted sound pressure level (L_Aeq), the background sound level (L₉₀), the difference between C-weighted and A-weighted levels (L_C–L_A), and loudness. People perceive quiet environments more positively and report strong discomfort in noisy environments. (c) Urban land-use planning significantly impacts the urban soundscape, with significant differences observed in both the acoustic environment and soundscape perception evaluations across different land-use types. This study deepens the understanding of the acoustic environment and demonstrates that soundscape-oriented land-use planning can function as an effective tool for fostering inclusive, healthy, and socially sustainable communities.

1. Introduction

Urban noise is increasingly becoming a global urban problem. Governments have invested much effort in acoustic environment optimization and noise management [1]. The World Health Organization (WHO) and other research organizations have conducted studies on noise. Several research institutions, including the World Health Organization, have conducted studies and confirmed that noise can seriously affect the residents’ sleep quality, leading to irritability and anxiety, and severe noise pollution may even cause neurological and cardiovascular diseases [2,3,4]. Current research on urban acoustic environments often focuses on reducing the objective sound pressure level for noise control to improve the acoustic environment. The research technology methods are mainly divided into two kinds: one is to control the source of noise generation; and priority can be given to the use of low-noise pavement, i.e., porous or modified asphalt pavement, which can effectively absorb and reduce the noise generated by the friction between the tires and the road surface, and reduce the road traffic noise from the source [5]. The second one is process-type noise reduction, which mainly reduces the intensity of noise during the propagation of sound waves [6]; there are also studies on green space and vegetation in the urban landscape that explore how green space and vegetation reduce noise [5,7]. However, noise cannot be entirely eliminated, and merely reducing the sound pressure level does not necessarily improve the perceived quality of the acoustic environment. Simply reducing the sound level does not necessarily improve the subjective evaluation of human perception [8]. Individuals’ perception of the acoustic environment is affected by various factors, including the frequency and duration of the sound, the source of the sound, the environmental context, etc. Therefore, “soundscape”, which emphasizes the listener’s perception and evaluation, has gradually become an important focus in optimizing the urban sound environment. [8].

Since the development of the WHO Environmental Noise Guidelines [8], including the publication of international standards for soundscapes in recent years [9], the potential of soundscape research in urban sound environments has been increasingly emphasized. Soundscape research has become a multidisciplinary school of thought that considers landscape, sound, and experience [10]. Soundscape design aims to integrate sound, audience, and environment to create a holistic landscape [11]. In addition, soundscape design focuses not only on the objective properties of sound but also on the subjective perceptions of human beings [12,13,14]. At the same time, soundscape research identifies sound as a resource that can be utilized to enhance human experience based on the space available to them [15,16]. For example, natural sounds have a significant role in masking noise [17,18] and increasing group tolerance [19] as well as making an environment feel pleasurable, relaxing, etc [20,21]. In addition, it has been demonstrated that it affects different urban spaces, including green spaces [22,23] and blue spaces [24]. In addition, studies have confirmed that different urban spaces can have different impacts on soundscape characteristics and their effects. Although this proves that soundscape can be one of the perceptual dimensions of urban design, there is still a disconnect and mismatch between land-use planning and soundscape planning in cities. It is therefore crucial to understand how land-use planning affects the spatial distribution of composite soundscapes.

For this work, we tend to understand the complexity and diversity of the urban acoustic environment and quantitatively study the relationship between urban land-use planning features and the distribution of urban soundscape. We aim to (a) explore the potential relationship between the urban soundscape and urban spatial functions, based on the spatial distribution characteristics of the urban soundscape and urban spatial functions. (b) Combine objective acoustic indicators and subjective perception evaluation indicators of soundscape, explore the influence of urban objective acoustic elements on the subjective perception evaluation of urban soundscape, and identify the key objective acoustic indicators affecting how individuals perceive a soundscape. (c) Combining the urban planning land-use indicators, we explore the influence of urban planning land use on the objective acoustic environment and the evaluation of urban soundscape perception. We explore the influencing factors of the composite soundscape and its changing rules in the urban environment, intending to deepen the current understanding of the acoustic environment, as well as increase the possibilities and methods of regulating urban soundscape by means of urban land-use planning.

2. Methods

2.1. Study Area

Fuzhou City is among the five cities honored with the first UN-HABITAT Global Sustainable City Award. In this study, Gulou District, Fuzhou City, Fujian Province, China, was selected as the study area, as shown in Figure 1. Fuzhou City, the capital city of Fujian Province, is located in the southeast coastal region of China and has a subtropical monsoon climate, with an average annual precipitation of about 900–2100 mm, an average annual temperature of about 20–25 °C, and an annual relative humidity of about 77%. Gulou District is located in the northwestern part of the urban area of Fuzhou City, with 9 streets and 1 town under its jurisdiction, totaling 69 communities and a resident population of about 675,000 individuals. Gulou District has a beautiful ecological environment, with intertwined mountains and water systems, and is the only national ecological zone in the province located in a core urban area. Topographically, Gulou District is relatively high in the northwest and low in the southeast. There are several mountains in this district, such as Wushan, Yushan, Pingshan, Wufengshan, Dabuishan, Jinniushan, etc. [25]. In addition to the unique natural environment, Gulou District has a deep humanistic heritage, along with representing economic prosperity, which makes this district one of the most urbanized areas in Fuzhou City.

2.2. Delineation of Research Space Units

The spatial continuity in the morphology and soundscape characteristics of the urban green spaces is of interest to us in this study, so vector grid cells were selected as the basic spatial units of the study [26]. The quantitative calculation of data and layout speculation can realize the comprehensive coverage of data in the study area, so as to better capture the local changes and features. Regarding the division of the grid cell size, previous studies have shown that the spatial continuity of the acoustic environment can be better observed within a buffer zone with a radius of 300 m [6]. In previous studies, when studying the spatial distribution characteristics of environmental noise in different urban areas, a grid of 600 m × 600 m was selected for field monitoring, which proved the feasibility of this grid cell size in practical research [27]. The feasibility of this grid cell size in practical research is proved. Therefore, considering the actual situation of the study area and the data analysis needs, this study sets the grid cell size to 600 m × 600 m, and realizes the division of the grid cells with the help of ArcGIS 10.8, with a total of 120 grid cells. Five objective acoustic monitoring sessions and five subjective questionnaire surveys were conducted in each grid unit, which satisfy the accuracy, operability, comprehensive coverage of the data of the study, and the results of the division are shown in Figure 1.

2.3. Data Sources and Processing Procedures

2.3.1. Objective Acoustic Parameters

In terms of the selection of physical acoustic parameters, the equivalent sound level (Leq) reflects the average energy of fluctuating sound pressure levels over a specified measurement period, and commonly used weighting network measurements mainly include A, C, and Z weighting. This study selected the equivalent continuous A sound level (L_Aeq) and the difference between L_Ceq and L_Aeq (L_C–L_A). The statistical sound level (Ln), or cumulative percent sound level, refers to the A sound level that occurs more than n% of the time or within a frequency under the specified measurement time or measurement frequency. We mainly selected the foreground sound (L₁₀) and the background sound (L₉₀), and the difference between the two, i.e., the sound source variability (L₁₀–L₉₀), which are the three metrics used to observe the temporal variability of the acoustic environment.

Psychoacoustic metrics are usually based on the properties of the human auditory system and the subjective perception of sound and are mainly used to describe and evaluate the relationship between the physical properties of sound and hearing [11], considering properties such as loudness [4,27]. Several studies have shown that loudness and sharpness have a greater impact on listeners’ subjective emotions and satisfaction; therefore, loudness and sharpness were selected for this study. Loudness is measured using sone units. This metric quantifies the human ear’s subjective perception of “how loud a sound appears.” Unlike the linear relationship of dB SPL, loudness exhibits an approximately logarithmic relationship with human perception, enabling it to more accurately reflect the total contribution of different frequency components (particularly low-frequency sounds) to the overall sensation of noisiness. For instance, a sound containing significant low-frequency components may be underestimated by its A-weighted sound level, whereas its loudness value can more truthfully reveal its potential to cause annoyance. Sharpness is measured using acum units. This metric describes the proportion and intensity of high-frequency components within a sound’s spectrum, reflecting how “piercing” or “sharp” the sound is perceived to be. Even if two sounds have the same overall loudness, the one with more high-frequency energy (e.g., a braking noise, certain mechanical noises) will be judged as having higher sharpness. Such sounds are typically more strongly associated with unpleasantness and auditory discomfort.

2.3.2. Soundscape Subjective Parameters

In this study, a questionnaire was used to obtain soundscape perception evaluation data. The questionnaire mainly includes three parts: the first investigates the respondents’ basic information, the second investigates the respondents’ sound source perception, and the third investigates the respondents’ soundscape evaluation.

(1)

Basic information of respondents. This includes gender, age (seven options: under 18, 18–24, 25–34, 35–44, 45–54, 55–64, and 65 and above), employment status (seven options: student, incumbent, retiree, and other), and educational background (four options: high school and below, technical vocational school, university graduate, and postgraduate and above).

(2)

Respondents’ perception of sound sources [1]. According to the international standard for soundscape (ISO/TS 12913-2) [28], sound sources are categorized into natural, human, and mechanical sounds, and the specific types of sound sources are shown in Table 1, and a five-level scale is used to quantify the perceived frequency of each type of sound source.

(3)

Respondents’ evaluation of the soundscape. In this study, based on the theoretical framework of multidimensional soundscape perception (e.g., pleasantness, eventfulness, familiarity), standardized ISO soundscape descriptors, and empirical validation across diverse environments, these five indicators—Appropriateness, Quietness, Comfort, Satisfaction, and Matching—were selected for their comprehensive ability to capture the physical, psychological, and contextual dimensions of soundscape quality. Suitability (characterizing the degree of relaxation and Appropriateness of the soundscape), Quietness (characterizing the degree of Quietness of the soundscape as perceived by the respondents), Comfort (characterizing the degree of Comfort of the soundscape as perceived by the respondents), Satisfaction (characterizing the degree of Satisfaction with the soundscape as perceived by the respondents), and Match (characterizing the degree of matching, as perceived by the respondents, of a soundscape with the soundscape in their area). The soundscape evaluation was quantified through a five-level scale divided into five degrees: strongly agree (+2 points), relatively agree (+1 point), moderate (0 points), relatively disagree (−1 point), and strongly disagree (−2 points). At the same time, we categorized the specific sound sources in the study area into three main types: natural sounds: wind, wind blowing leaves, water, chirping of birds and insects, and dogs barking; human sounds: conversation, footsteps, entertainment (excluding conversations), and sports; and mechanical sounds: music (musical instruments), broadcasts (electric broadcasting), transportation, and construction.

Table 1. The composition of the sound source type.

Sound Source Type	Specific Sound Source
Natural sound	wind, leaves blown by the wind, water, birds, insects, dogs
Human sound	conversation, footsteps, entertainment activities fitness exercises
Mechanical sound	music, radio, traffic, construction

Table 1. The composition of the sound source type.

Sound Source Type	Specific Sound Source
Natural sound	wind, leaves blown by the wind, water, birds, insects, dogs
Human sound	conversation, footsteps, entertainment activities fitness exercises
Mechanical sound	music, radio, traffic, construction

2.3.3. Urban Land-Use Planning Data

The urban land-use data used in this study is from the open access data “China Basic Urban Land Use Classification (EULUC-China) (https://data-starcloud.pcl.ac.cn/zh/resource/7), accessed on 30 December 2022. The classification standard of this data is based on the China General Land-Use Classification (GBT 21010-2017) [29]. These data were processed based on big data such as 10 m resolution satellite images from Sentinel-2A/B, OpenStreetMap, nighttime light data from the LuoJia1-01 Satellite, POIs from Gaode Maps, and Tencent Mobile Positioning. After preliminary processing in ArcGIS 10.8 software, we obtained the land-use classification profile of Gulou District, Fuzhou City (Figure 2).

Based on the urban land-use data obtained in the preliminary preparation, the current land-use status of Gulou District in Fuzhou City was verified by combining the central area plan of the Fuzhou Municipal Government and satellite images from the National Geospatial Information Service Platform. According to the “Guidelines for Land Spatial Mapping Planning and Control of Land and Sea Area Division in China”, the original data were reclassified into five categories, and based on the study area-related research spatial unit grids obtained by dividing the study area in the previous work, the land-use data in each spatial unit grid were summarized and visualized to obtain the land-use planning characteristics of Gulou District, Fuzhou City (Figure 3). Based on Figure 3, it can be seen that there are obvious characteristics of land-use planning in Gulou District of Fuzhou City. As the center of the provincial capital city, the area has a rich variety of land-use planning, with more residential land, commercial land, and land for public administration and services, and less industrial land and transportation land. Residential land is mainly distributed in the western and northern parts of the study area, commercial land is mainly distributed in the eastern and southern parts of the study area, and public administration and service land is widely distributed in the eastern, southern, and western parts of the study area, with less distribution in the northern part; industrial land is mainly distributed in the northern part of the study area, and transportation land is mainly distributed in the northeastern part of the study area.

To further explore the potential influencing factors of the urban soundscape, we divided five tangent lines A, B, C, D, and E according to the main roads in the study area and conducted a tangent analysis (Figure 4), in which Line A is the section from Wusi Road to Wuyi Zhong Road, one of the main roads in the study area, Line B is the section from the West North Second Ring Road to Middle North Second Ring Road, one of the main express loops in the city where the study area is located, Line C is the Honggan Road to Yangqiao East Road to Wuyi Middle School section, one of the main roads in the study area, Line D is the section from Baima North Road to Hudong Road, one of the main roads in the study area, and Line E is the section from Wushanxi Road to Fuma Road, one of the main roads in the study area. The urban planning land-use data and the urban soundscape data on each line were normalized. In order to facilitate the comparative analysis, we further explored the influence law of urban land-use planning on the urban soundscape by identifying and analyzing the urban land-use planning features and urban soundscape features on different tangent lines.

2.4. Procedure

2.4.1. Objective Acoustic Data Acquisition and Processing

The measurement instruments employed in this study comprised an AWA6228+ (Hangzhou Aihua Instrument Co., Ltd., Hangzhou, China) multifunctional Class I sound-level meter and a Sony PCM-A10 recorder (Sony Corporation, Tokyo, Japan). Field measurements were conducted on weekdays between 9:00 and 17:00 from April to May 2023. Meteorological conditions during the measurement sessions were consistently cloudy or sunny, with temperatures ranging from 15 to 25 °C and wind speeds maintained below 5 m/s to minimize potential interference with acoustic propagation (referencing, e.g., ISO 1996-2:2017) [30]. In accordance with standard practices for environmental acoustic measurements, all measurement points were positioned at least 3.5 m away from any sound-reflecting surfaces (other than the ground) and at a height of 1.2 m above the ground to reduce reflection-induced errors (referencing, e.g., ANSI S1.13-2005) [31].

Within each grid cell, as defined by the study’s sampling design, five repeated measurements were taken. Each measurement involved the simultaneous collection of objective acoustic data using the sound-level meter and audio recordings obtained via the PCM-A10 recorder, with a minimum duration of 3 min per point. Physical acoustic parameters were directly obtained from the sound-level meter, while psychoacoustic metrics were derived through subsequent analysis using AWA6228+ (Hangzhou Aihua Instrument Co., Ltd., Hangzhou, China), which provides integrated time-domain, frequency-domain, and modulation-domain processing capabilities. A total of 600 objective acoustic measurement points were acquired.

2.4.2. Questionnaire Distribution and Collection

The questionnaire survey was conducted in parallel with the collection of objective acoustic data. The respondents were randomly selected for the questionnaire distribution. The survey was conducted in strict accordance with the data collection requirements specified in the ISO international standard for soundscape: (1) inform the respondents of the purpose of the survey before they fill in the questionnaire and ask for their consent; (2) explain the content of the questionnaire to the respondents in detail, to better assist them in understanding the questions; (3) before filling in the sound source perception and soundscape evaluation sections, the respondents were asked to listen carefully for 1 min at the current location before filling in the questionnaire, to reflect the overall acoustic environment of the surroundings fully. A total of 600 valid questionnaires were collected in this study, and the statistics obtained from respondents are shown in Figure 5. To further confirm the validity of the questionnaire, the results of the questionnaire were analyzed for reliability and validity. The results showed that the Cronbach’s alpha coefficient was 0.905 (>0.7), and the questionnaire reliability was high. The validity analysis was carried out by KMO and Bartlett’s test of sphericity, which showed that the KMO value was 0.869 (>0.6), the significance p-value was 0.000 (<0.01), and the validity of the questionnaire was good.

2.5. Research Process

The study of the impact of urban land-use planning on generating an urban soundscape is divided into three main research steps.

In the first step, we carried out a spatial distribution characterization of the urban acoustic environment, and initially analyzed the urban objective acoustic indicators, including physical acoustic parameters and psychoacoustic parameters, to obtain the spatial distribution characteristics of the urban objective acoustics; at the same time, we carried out a preliminary analysis of the urban subjective acoustic scene perception evaluation, including sound source perception indicators and soundscape evaluation indicators, in order to obtain the spatial distribution characteristics of the urban subjective acoustic scene perception evaluation.

In the second step, to explore the relationship between the objective acoustic environment and the perceptual evaluation of soundscape, we used the objective acoustic characteristics as the independent variable and the subjective soundscape perceptual evaluation as the dependent variable. We used SPSS Statistics 26 software to conduct Spearman correlation analysis. We also included the objective acoustic indicators that showed a significant correlation with the subjective evaluation of soundscape in the multivariate stepwise linear regression model, and we used the perceptual evaluation of the soundscape as the dependent variable, with the existence of the significant correlation with the objective acoustic indicators as the independent variable, to identify the critical objective indicators affecting the subjective evaluation of soundscape.

In the third step, to explore the influence of urban land-use planning on the urban soundscape, we used SPSS software to conduct a Pearson correlation analysis between urban land-use classification data (the area of various types of land in the grid unit) and urban soundscape indicators, including objective acoustic indicators and the soundscape perception evaluation, to explore the influence of urban planning land use on the urban soundscape, and to further identify the key land-use planning indicators affecting the urban soundscape. Meanwhile, we further investigated the different characteristics of the urban soundscape under different urban land uses through tangent analysis and explored the impact of urban land planning on urban soundscape. Finally, we combine the key land-use planning indicators affecting the urban soundscape with the urban land-use planning characteristics and soundscape characteristics on the tangent line to analyze the influence pattern and path of urban land-use planning on the soundscape (Figure 6).

3. Results

3.1. Spatial Distribution of the Urban Soundscape

3.1.1. Characterization of Objective Indicators

We collected and summarized the objective acoustic data of each measurement point obtained from the field monitoring, and carried out inverse distance-weighted interpolation analysis in ArcGIS 10.8 software to visualize the objective acoustic indicators in the study area, and finally obtained the spatial distribution characteristic maps of the physical acoustic indicators (Figure 7) and psychoacoustic indicators (Figure 8) among the objective acoustic indicators.

The spatial distribution characteristics of physical acoustic indicators among objective acoustic indicators are shown in Figure 7. After visualizing the acoustic environment indicators and land-use classification indicators in the study area, we find that the spatial distribution characteristics of the two are more obvious. In terms of physical acoustic characteristics, the spatial distribution of L Aeq in the study area varies greatly, ranging from 39.30 to 76.20 dB, lower in the northwest and higher in the southeast, and presenting an obvious point-axis spatial pattern; L10 has sound-level values ranging from 40.40 to 78.40 dB, with spatial distribution characteristics similar to those of L_Aeq; The sound-level values of L₉₀ range from 30.20 to 70.40 dB, showing spatial distribution characteristics that are lower in the northwest and higher in the southeast, which are similar to that of L_Aeq; the sound-level values of L₁₀–L₉₀ in the northern part of the study area are higher, with more significant fluctuation in sound level, i.e., the sound level is low most of the time but sometimes there are bursts of higher sound levels, while those in the southern part of the study area are lower, with the sound level being relatively stable, and there is not much difference between the background sound and the occasional high level; the sound-level values of L_C–L_A in the periphery of the study area and the center of the study area are higher.

The spatial distribution characteristics of psychoacoustic indicators are shown in Figure 8. Loudness in the study area ranges from 3.05 to 37.21, with areas of higher loudness mainly located near roads, especially in the more heavily trafficked areas, while areas of lower loudness are mainly located near natural mountains, water bodies, and various types of parks, such as the Fushan Country Park, Fudao Road, and Xihu Park, which suggests that compared with areas with intensive and frequent human activities, areas with more vegetation and water bodies are more quiet, which is also in line with the characteristics of the objective acoustic indicators in the study. Sharpness ranges from 1 to 2.9, with higher values located near Fushan Country Park, Fudao, and West Second Ring Road Central, and lower values located near Minjiang Park, Xihu Park, and Yu Shan.

Overall, the acoustic environment in the Gulou District of Fuzhou City varies greatly in spatial distribution. The southeastern part of the study area is noisier, the northwestern part is quieter, and the source variability in the northern part is significantly higher than that in other areas, which also indicates that the acoustic environment in the northern part of the study area is quieter most of the time, with occasional sudden high sound-level events. The higher L_C–L_A in the periphery and the center of the study area indicates that the low-frequency component of the noise in these areas is more obvious. Meanwhile, the high source variability (L₁₀–L₉₀) is mainly found on some roads and in historic districts, which may be due to the unique traditional cultural activities in the historic districts, leading the sound sources to be various. The higher sharpness in quiet areas may be due to the increase in vegetation and water bodies, which leads to the increase in animal sounds such as birdsong in these areas, resulting in the increase in sharpness in the acoustic environment.

3.1.2. Characterization of Subjective Indicators

(1)

Frequency of soundscape perception

Regarding subjective soundscape perception, we obtained the frequencies regarding respondents’ perception of various types of sound sources, and the statistical results are shown in Figure 9. This study combines the actual situation of the study area, selecting the more representative sound sources. Through the questionnaire, we obtained the frequencies regarding respondents’ perception of various types of sound sources, and the statistical results are shown in Figure 9. It can be found that the most easily perceived sound sources in the study area are mainly transportation, the chirping of birds, and the sound of conversation. The least perceived sound sources in the study area were mainly dogs barking, the sound of sports, and broadcasts.

To further analyze the spatial distribution characteristics of different sound source categories, the sound sources were categorized as natural sound, human sound, and mechanical sound, and the share in the perception of the three sound source categories was counted separately according to the perceived frequency of each sound source the residents heard. Spatial visualization was carried out using the inverse distance-weighted interpolation analysis tool in the ArcGIS 10.8 software, and the results are shown in Figure 10. In the study area, natural sound perception is higher mainly in various types of parks, and there are more chances to perceive natural sound in parks, mountains, and forests, as well as in areas close to water bodies. The areas with higher human sound perception are mainly located in public activity spaces such as parks and squares. In addition, in areas with high traffic flow, such as near viaducts, where sounds of transportation are constant, the percentage of mechanical sound perception is very high.

At the same time, we conducted a global spatial autocorrelation analysis as well as a local spatial autocorrelation analysis (Figure 11) on the data of the three sound source perception ratios, to further analyze their spatial distribution characteristics (

Table 2. Results of global autocorrelation analysis of sound source perception.

Sound Source Perception		Moran’s I	Z-Score
Sound source category	Natural sound	0.326	4.872
	Human sound	0.547	8.030
	Mechanical sound	0.322	4.794

). The results of the global autocorrelation analysis of the percentage of sound source perception are shown in Table 2. The global Moran’s I of the three sound source perception ratios are all greater than 0, and the z-scores are all greater than 2.58, indicating that there is a significant spatial positive autocorrelation and an extremely significant aggregation pattern. This indicates that the distribution of sound source perception is not random; different types of sound sources have significant spatial autocorrelation values, and the region with a high proportion of a specific type of sound source perception tends to have similar perceptions to that of neighboring regions, and there is a spatial aggregation effect.

The results of the local spatial autocorrelation analysis of sound source perception occupancy are shown in Figure 11. The clustering and outlier types can be distinguished from high-value clustering (HH), low-value clustering (LL), high-value but surrounding low-value outliers (HL), and low-value but surrounding high-value outliers (LH). HH indicates that the region has a high percentage of sound source perception and a high percentage of the surrounding source perception; LL indicates that the region has a low percentage of sound source perception and a low percentage of the surrounding source perception; HL indicates that the region has a high percentage of sound source perception but a low percentage of the surrounding source perception; LH indicates that the region has a low percentage of the surrounding source perception but a high percentage of the surrounding source perception; and HL indicates that the region has a low percentage of the surrounding source perception but a low percentage of the surrounding source perception. LL indicates that the proportion of sound source perception in the area is low, while the proportion of surrounding sound source perception is also low. HL indicates that the proportion of sound source perception in the area is high but the proportion of surrounding sound source perception is low, and LH indicates that the proportion of sound source perception in the area is low but the proportion of surrounding sound source perception is high.

As shown in Figure 11, the aggregation patterns of HH and LL appear in natural sound perception, human sound perception, and mechanical sound perception, indicating a relatively strong spatial dependence on sound source perception. In addition, the natural sound perception showed an LH clustering pattern, indicating the existence of areas with low natural sound perception. The emergence of HL clustering patterns in anthropogenic sound perception suggests localized areas of high anthropogenic activity clustering. The areas presenting HH aggregation for natural sound perception are mainly located around Xihu Park, Yushan, and Wushan. The areas presenting LL aggregation are mainly located around Fushan Country Park, Fudao, and Hot Spring Park. The areas presenting LH aggregation are located around Minjiang Park and Jin’an River Park. The areas with HH aggregation of human acoustic perception are mainly located in the northwestern part of the study area, which is located in the vicinity of Fudao and Fushan Country Park; the areas with LL aggregation are mainly located in the southeastern part of the study area, which is located in the more densely populated residential areas; and the areas with HL aggregation are located in the vicinity of the Dongjiekou. The areas where mechanical sound perception shows HH aggregation are located in the vicinity of North Third Ring Road and North Ring Middle Road, and the areas where LL aggregation is shown are located in the vicinity of Fushan Country Park, Yu Shan, and Wuyi Square.

(2)

Multidimensional evaluation of soundscape

We summarized the results of the soundscape perception evaluation questionnaire to obtain the statistical results for the soundscape evaluation score in the study area (Figure 12). We visualized the soundscape evaluation through ArcGIS 10.8 software to further analyze its spatial distribution characteristics, and the results are shown in Figure 13. At the same time, we also performed a global spatial autocorrelation analysis (

Table 3. Results of global autocorrelation analysis of soundscape evaluation.

Soundscape Evaluation	Moran’s I	z-Score
Suitability	0.290	9.055
Quietness	0.303	9.469
Comfort	0.307	9.586
Satisfaction	0.354	11.025
Match	0.141	4.441

) and a local spatial autocorrelation analysis (Figure 14) on the soundscape evaluation data in order to further analyze its spatial distribution characteristics.

Figure 12 shows the statistical results of the soundscape evaluation scores for the study area. The scores are at a moderate level overall, reflecting the smooth quality of the soundscape in the study area, with fewer overly prominent or serious problems.

The soundscape evaluation was visualized by ArcGIS inverse distance-weighted interpolation analysis to further analyze its spatial distribution characteristics, and the results are shown in Figure 13. The area with the highest Suitability is located in Fushan Country Park, the area with the highest Quietness is located in Fushan Country Park and Fudao, the area with the highest Comfort is located in Fushan Country Park and Westwood Park, the area with the highest Satisfaction is located in Fushan Country Park, Fudao and Westwood Park, and the area with the highest Match is located in Fushan Country Park and Fudao.

The results of the global spatial autocorrelation analysis of the soundscape evaluations are shown in Table 3. The global Moran’s I for all five soundscape evaluations is greater than 0, and the z-scores are all greater than 2.58, indicating significant spatial positive autocorrelation for all of them, showing an extremely significant clustering pattern. Similar soundscape evaluation scores tended to cluster together spatially, with regions with high soundscape evaluations tending to be adjacent to regions with equally high soundscape evaluations and regions with low soundscape evaluations tending to be adjacent to regions with low soundscape evaluations.

The results of the local spatial autocorrelation analysis for soundscape evaluation are shown in Figure 14. The clustering and outlier types can be distinguished from high-value clustering (HH), low-value clustering (LL), high-value but low surrounding outliers (HL), and low-value but high surrounding outliers (LH). HH indicates that the region has a high soundscape evaluation score and a high surrounding score, LL indicates that the region has a low soundscape evaluation score and a low surrounding score, and HL indicates that the region has a high soundscape evaluation score but a low surrounding score. LH indicates that the region has a low soundscape evaluation score but a high surrounding score.

The areas presenting HH clustering in the soundscape evaluation are mainly located around Fushan Country Park and Fudao Road; the areas presenting LL clustering are mainly located around North Third Ring Road and Wusi Road; and the areas presenting LH anomalies are mainly located around Software Park.

The area presenting HH clustering for Suitability evaluation is mainly located in Fushan Country Park; the area presenting LL clustering is mainly located around North Third Ring Road; the area presenting HL anomalies is mainly located in the vicinity of Fushan Country Park and Qintinghu Park; and the area presenting LH anomalies is mainly located in Software Park. The areas presenting HH clustering for Quietness evaluation are mainly located in the vicinity of Fushan Country Park and Fudao Road; the areas presenting LL clustering are mainly located in the vicinity of Wusi Road and Dongjiekou; and the areas presenting LH anomalies are mainly located in the vicinity of Software Park and Meifeng Road. The areas presenting HH clustering for Comfort evaluation are mainly located in the vicinity of Fushan Country Park and Fudao; the areas presenting LL clustering are mainly located in the vicinity of the North Third Ring Road, Wusi Road and Wushanxi Road Viaduct; the areas presenting HL anomalies are mainly located in the vicinity of Fushan Country Park; and the areas presenting LH anomalies are mainly located in the vicinity of Software Park and Meifeng Road. The areas presenting HH clustering for Satisfaction evaluation are mainly located in Fushan Country Park and Fudao Road; the areas presenting LL clustering are mainly located around North Third Ring Road and Wusi Road; and the areas presenting LH anomalies are mainly located in Software Park. The areas presenting HH clustering for the Match evaluation are mainly located around Fudao and Xihu Park; and the areas presenting LL clustering and LH anomalies are mainly located around Software Park. Among the various subjective evaluation indicators of soundscape, the results of Suitability, Comfort, and Satisfaction ratings are relatively similar, which indicates that individuals’ feelings are relatively consistent. Quietness scores are relatively low overall, which implies that the study area may have noise disturbances at some locations, which have a particular impact on individuals’ need for Quietness. Match scores were relatively high, indicating a high degree of fit between the soundscape and individuals’ expectations in the study area.

3.2. Study of the Relationship Between Urban Land-Use Planning and Subjective and Objective Indicators of Soundscape

The results of the multivariate stepwise linear regression analysis are presented in

Table 4. Results of multiple stepwise linear regression analysis of soundscape perception evaluation and objective acoustic indicators.

Implicit Variable		Independent Variable (Objective Acoustics)	Standardized Coefficient	t	Covariance Statistics		R-Square	F
					Tolerances	VIF
Sound source perception	Natural sounds	/	/	/	/	/	/	/
	Human sounds	L90	−0.296	−3.372	1.000	1.000	0.088	11.373 **
	Mechanical sounds	L90	0.216	2.402	1.000	1.000	0.047	5.772 *
Soundscape Evaluation	Suitability	LAeq	−1.077	−5.297	0.108	9.278	0.483	36.098 ***
		LC–LA	−0.247	−3.552	0.924	1.082
		L90	0.411	2.046	0.110	9.071
	Quietness	LAeq	−0.836	−16.517	1.000	1.000	0.698	272.809 ***
	Comfort	LAeq	−1.247	−6.379	0.108	9.278	0.522	42.241 ***
		LC–LA	−0.236	−3.537	0.924	1.082
		L90	0.569	2.942	0.110	9.071
	Satisfaction	LAeq	−1.234	−5.739	0.108	9.278	0.422	28.261 ***
		LC–LA	−0.250	−3.401	0.924	1.082
		L90	0.648	3.051	0.110	9.071
	Match	LAeq	−1.119	−3.746	0.079	12.702	0.179	12.730 ***
		Loudness	0.791	2.647	0.079	12.702

Table source: authors’ drawing. Note: * indicates p < 0.05, ** indicates p < 0.01, and *** indicates p < 0.001.

. The model retained the significant variables that the F-test tested, while the explanatory variables with strong covariance were excluded. The tolerance of all the selected variables was more significant than 0.2, and the variance inflation factor (VIF) value was less than 10. The indicators retained in the regression model were the key objective acoustic indicators affecting the subjective evaluation of soundscape. The R-square of the regression model was the coefficient of determination describing the degree of fit between the variables. The closer the value was to 1, the better the degree of fit was.

Regarding sound source perception, the critical objective acoustic metrics affecting the perception of both human and mechanical sounds are L₉₀. L₉₀ is usually considered to be the background sound level of the environment, and there is a strong association between the background sound level and the perception of the sound source. Human sounds are mainly emitted by individuals’ daily activities, such as conversations, entertainment, etc., which are usually not very strong and intermittent, so when L₉₀ is high, it means that the proportion of human sound perception may be relatively low. Mechanical sounds, on the other hand, are mainly emitted by various mechanical devices, such as traffic sounds and construction sounds, which are usually of higher intensity and more persistent, so when L₉₀ is higher, it means that the proportion of mechanical sound perception may be relatively more significant. In terms of soundscape evaluation, the critical objective acoustic indicators affecting the evaluation of Suitability, Comfort, and Satisfaction are L_Aeq, L_C–L_A, and L₉₀; the critical objective acoustic indicator affecting the evaluation of Quietness is L_Aeq; and the key objective acoustic indicators affecting the evaluation of Match are L_Aeq and loudness.

The results of the Pearson correlation analysis between urban land-use planning indicators and urban soundscape indicators are shown in

Table 5. The results of the Pearson correlation analysis between urban land-use planning indicators and urban soundscape indicators.

	Residential Land	Commercial Land	Public Administration and Service Land	Industrial Land	Transport Land
LAeq	−0.200 *	0.239 **	0.076	0.017	0.063
L10	−0.180	0.235 *	0.080	0.030	0.058
L90	−0.102	0.218	0.020	−0.015	0.069
L10–L90	0.421 **	−0.213 *	−0.042	0.042	−0.141
LC–LA	−0.305 **	0.048	0.097	0.250 **	0.012
Loudness	−0.332 **	0.301 **	0.090	0.011	0.109
Sharpness	0.452 **	−0.082	−0.209 *	−0.235 *	−0.033
Suitability	0.164	−0.075	0.056	−0.353 **	−0.079
Quietness	0.205 *	−0.016	−0.016	−0.155	−0.135
Comfort	0.130	−0.082	0.043	−0.312 **	−0.040
Satisfaction	0.090	0.074	0.151	−0.418 **	−0.120
Match	0.011	−0.008	0.099	−0.081	−0.067
Natural sounds	−0.055	0.270 **	0.100	−0.108	−0.074
Human sounds	0.267 **	−0.288 **	−0.360 **	0.193	−0.130
Mechanical sounds	−0.165	−0.045	0.195 *	−0.046	0.187 *

**. Significant correlation at the 0.01 level (two-tailed). *. Significant at the 0.05 level (two-tailed).

; in terms of objective acoustic environment, L_Aeq is significantly positively correlated with commercial land, and significantly negatively correlated with residential land; L₁₀ is significantly positively correlated with commercial land; L₉₀ is significantly positively correlated with residential land, and significantly negatively correlated with commercial land; L₁₀–L₉₀ is significantly negatively correlated with residential land, and significantly positively correlated with transportation land; L_C–L_A is significantly negatively correlated with residential land, and significantly positively correlated with industrial land; loudness is significantly negatively correlated with residential land, and significantly positively correlated with commercial land; sharpness is significantly positively correlated with residential land, significantly negatively correlated with public administration and service land, and significantly negatively correlated with industrial land. In terms of subjective sound perception evaluation, Suitability is significantly negatively correlated with industrial land; Quietness is significantly positively correlated with residential land; Comfort is significantly negatively correlated with industrial land; Satisfaction is significantly negatively correlated with industrial land; natural sound is significantly positively correlated with commercial land; human sound is significantly positively correlated with residential land, and significantly negatively correlated with commercial land and land for public management and services; and mechanical sound is significantly positively correlated with land for public management and services. There is a significant positive correlation between the sound of machinery and public administration and service land and a significant positive correlation with transportation land.

3.3. Characteristics of Typical Urban Road Soundscape and Land-Use Interaction

The urban land use and urban soundscape characteristics of the five major road sections in the study area are shown in Figure 15. In the Line-A section, when passing through the road section with more residential land, sharpness and Quietness all show an upward trend, and L_Aeq, L₁₀, L₉₀, loudness, natural sounds, and human sounds all show a downward trend. When passing through the section with more commercial land, L_Aeq, L₁₀, L₉₀, loudness, and natural sounds all show an upward trend, and Quietness shows a downward trend; when passing through the section with more transportation land, mechanical sounds shows an upward trend. In Line-B, L₉₀, Quietness, and human sounds show an upward trend when passing through sections with more residential land; L₉₀ shows a downward trend when passing through sections with more commercial land; and mechanical sounds shows an upward trend when passing through sections with more transportation land. In Line-C, sharpness, Suitability, Quietness, Comfort, Satisfaction, Match, and natural sounds show an increasing trend when passing through roads with more residential land; human sounds show an increasing trend when passing through roads with more commercial land. When passing through the road sections with more commercial land, human sounds showed an increasing trend, and Suitability, Quietness, Comfort, Satisfaction, Match showed a decreasing trend; when passing through the road sections with more public administration and service land, mechanical sounds showed an increasing trend; when passing through the road sections with more industrial land, L_Aeq, L₁₀, L₉₀, and loudness all show an increasing trend. In Line-D, Quietness and human sounds show an increasing trend when passing through sections with more residential land; L₁₀–L₉₀, sharpness, human sounds, and mechanical sounds show an increasing trend when passing through sections with more public administration and service land, and L₉₀ shows a decreasing trend. In Line-E, when passing through roads with more residential land, L₉₀, sharpness, Quietness, and human sounds all show an increasing trend, and L_Aeq shows a decreasing trend; when passing through roads with more commercial land, L_Aeq and natural sounds all show an increasing trend, and L₉₀ shows a decreasing trend; and when passing through roads with more public management and service land, L₁₀–L₉₀, sharpness, natural sounds, and mechanical sounds all show a decreasing trend. When passing through roads with more commercial land, L_Aeq and natural sounds show an increasing trend, and L₉₀ shows a decreasing trend.

Overall, an increase in the proportion of residential land area is associated with increased levels of L₉₀, sharpness, Quietness, and perception of human sounds within that space, and it also leads to a decrease in the levels of L_Aeq, L₁₀–L₉₀, L_C–L_A, and loudness. The increase in the area of commercial land leads to an increase in the levels of L_Aeq, L₁₀, loudness, and natural sounds in its space and also a decrease in the levels of L₉₀ and human sounds. The increase in public administration and service land leads to an increase in mechanical sounds in the space and a decrease in sharpness and human sounds. The increase in industrial land leads to an increase in the level of L_C–L_A within the space as well as a decrease in sharpness, Suitability, Comfort, and Satisfaction. The increase in transportation land leads to an increase in the levels of L₁₀–L₉₀ and mechanical sounds within the space.

4. Discussion: The Urban Soundscape Is Influenced by Many Factors

4.1. Intervention in the Acoustic Environment by the Complex Urban Environment

The complexity of the urban environment shapes the acoustic environment, as the different functional zones of the city determine, to some extent, the activities of individuals within them, leading to spatial heterogeneity in the urban acoustic environment [16]. Combined with Figure 7 and Figure 8, the high traffic and pedestrian flow due to the concentration of all types of services on both sides of the urban roads leads to higher ambient sound [10]. The equivalent continuous A sound level (L_Aeq) is mainly related to the function of urban land use and crowd behavioral activities [16]. The parks and residential areas have relatively singular functions and are at a certain distance from the noisy urban roads and commercial areas, so the L_Aeq is lower.

The traffic flow around major city roads is large, and there are more construction activities such as municipal engineering and store renovations, which are prone to sudden high-noise events, so the levels of foreground sound (L₁₀) and background sound (L₉₀) are high. Excessive sound levels can cause damage to individuals’ hearing, and sudden noise disturbances can also negatively affect individuals’ daily lives, affecting the efficiency of work and study, and even adversely affecting physical and mental health [32]. At the same time, long-term exposure to high noise levels can lead to anxiety, irritability, and other negative emotions, which may also cause a series of health problems [33]. In densely populated areas, sound sources are generally more continuous and uniform, with more minor differences. In historical neighborhoods, however, there are unique sounds from traditional cultural activities, and their sound sources are diverse and varied [19], which may lead to an increase in the amount of sound source variability.

The relationship between L₉₀ and sound source perception is not contradictory but rather reveals a key masking mechanism governed by the acoustic properties of different sources. A high L₉₀, representing an elevated background noise floor, acts as a masking layer that suppresses the audibility of intermittent and lower-level sounds like human voices, leading to their lower perceived presence. Conversely, mechanical sounds (e.g., traffic), which are often continuous and high-energy, are not only audible above this elevated baseline but are frequently the primary contributors to it. Thus, the positive correlation between L₉₀ and mechanical sound perception signifies the dominance of these sources in shaping the acoustic background, wherein they simultaneously raise the noise floor and remain the most prominent audible features.

In the vicinity of industrial areas, there are more low-frequency sound sources, such as the rotation of heavy machinery or equipment motors, resulting in a higher difference between L_Ceq and L_Aeq (L_C–L_A). Significant low-frequency noise can also affect human health without exceeding the sound decibel value, so it is important to pay attention to the low-frequency noise problem [34]. Therefore, the problem of low-frequency noise must be emphasized [35].

In areas with dense traffic and pedestrian flow, the sounds generated by various urban activities are intertwined, resulting in high loudness levels and high-intensity sound environment, while loudness levels are relatively low in parks, natural mountains, and water bodies, which are relatively quiet spaces. In contrast, sharpness levels were higher in spaces with denser road networks and closer to animals, which could be attributed to more animals and high-frequency sounds generated by car braking [10].

4.2. Synergy of Site Functions on Soundscape Resources

There is a specific pattern between the sound source perception characteristics and the urban land-use function of the space where it is located. Areas with the same distribution pattern of sound source perception tend to show similar spatial functional characteristics, and there are some differences in the distribution patterns of different categories of sound source perception.

The areas with a high perception of natural sounds are mainly parks, and Fuzhou, as a city called the “City of a Thousand Gardens”, has a large number of parks; various types of parks usually have a large amount of vegetation and trees, which provide habitats for a variety of animals, thus generating a richer variety of insect sounds, bird calls, and the sound of wind blowing on the leaves. The study area also has many natural water bodies such as rivers, lakes, and artificial water bodies such as ponds and fountains, and the sound of water flowing and splashing is also an important part of natural sound. At the same time, human and mechanical sounds have a particular suppression effect on natural sounds, so individuals’ perception of natural sounds may also be weakened in the green spaces of parks where natural sound sources are more abundant. At the same time, the most natural sound perceived by individuals is birdsong, which means that parks increase the percentage of birdsong in the space by providing pathways for birds to congregate in the area [36], which also allows these areas to produce more natural sounds. Additionally, vegetation has a dampening effect on noise, resulting in a decrease in noise levels within the space, which also allows individuals to perceive natural sounds more clearly [10]. However, in parks and green spaces where natural sound sources are more abundant, individuals’ perception of natural sounds may also be weakened because human and mechanical sounds have a specific suppression effect on natural sounds. Compared to residential areas, parks, plazas, and other areas where individuals talk and stay and engage in various leisure and recreational activities will emit more sound. So, individuals will perceive more human sound in parks, plazas, and other areas. In areas with high traffic flow, the reason for the higher percentage of perceived mechanical sounds may be that residential areas are denser in terms of buildings and roads, and there are more sources of various equipment sounds and traffic sounds, so individuals may be more sensitive to these mechanical sounds [37].

Individuals’ feelings of Appropriateness, Comfort, and Satisfaction are relatively consistent, and noise disturbance affects individuals’ attitudes towards soundscape evaluation to a greater extent [38]. The noise disturbance affects individuals’ attitude towards soundscape evaluation to a greater extent. At the same time, there are also spatial distribution differences between different soundscape evaluation indicators, indicating that different indicators themselves are also affected by the functional environment of urban land use, so corresponding soundscape optimization strategies should be proposed for different urban spatial functions.

4.3. Relationship Between the Public’s Need for Tranquility and Planning

Residential land, as an urban space for individuals to rest and recover, tends to have a higher need for tranquility [33]. By analyzing the Pearson correlation between urban land-use planning indicators and urban soundscape indicators (Table 5) and further analyzing the characteristics of urban land use and the urban soundscapes of major roads in the study area (Figure 15), we find that in the urban space, residential land has a more peaceful sound environment, with fewer types of sound sources, less significant low-frequency components, and less impact on human health. Although the overall noise level in the residential site is lower, and individuals can experience more tranquility and perceive more human sounds, there is also a certain degree of noise generated by vehicles, crowds, and urban activities in the residential site [16], which is present for an extended period and may lead to negative emotions such as anxiety and irritability, as well as a series of health problems caused by prolonged exposure to a noisy environment.

The transect analysis (Lines A–E) provides crucial spatial interpretation and anchors the macroscopic Pearson correlation results presented in Table 5 with empirical evidence. For instance, the increase in natural sounds recorded along Transect A as it traverses commercial areas serves as a spatial manifestation of the positive correlation between this land-use type and natural sound perception. This phenomenon is directly linked to the commercial green landscapes intersected by the transect path, where visual cues and mechanisms that attract biological sounds enhance the perceptibility of natural sounds. This methodology, which mutually verifies statistical patterns with specific spatial trajectories, not only validates the macroscopic correlations but also elucidates the underlying driving mechanism: specific land-use patterns directly govern the spatial heterogeneity of soundscape characteristics by shaping the physical environment, such as the configuration of green infrastructure.

In spaces with dense commercial land, the overall sound levels are lower and the levels at which intermittent noise events occur are higher, probably due to the higher traffic flow in these areas and the presence of more construction activities, which tend to create sudden high-noise events [39]. At the same time, individuals can perceive more natural sounds, which may be due to the existence of commercial green landscapes and the fact that individuals subjectively pay more attention to the natural sounds in commercial green landscapes in crowded environments, which is also related to visual elements [32,40]. At the same time, the sound of human activities has a masking effect on traffic noise and improves acoustic Comfort. There are fewer high-frequency sound sources in the public administration and service landscapes, fewer individuals are moving around in the spaces, and individuals are restrained in their vocal behavior. In addition, the significant low-frequency noise in industrial land, due to the presence of more low-frequency sound sources such as heavy machinery or rotating equipment motors, results in a decrease in the Suitability, Comfort, and Satisfaction of individuals. There are large fluctuations in sound-level levels within the transportation sites, and even though the sound-level levels are low most of the time, sometimes there are still bursts of higher sound levels, which may be caused by occasional noise events such as cars honking, city traffic sounds, etc.

This study aims to explore the complex reasons that affect the composite urban soundscape and supplement the existing research on urban soundscapes through a study of the elements that affect the urban soundscape, combined with the urban land-use planning on the changing laws of the urban soundscape, and explore the complex factors that affect the composite urban soundscape. At the same time, this study also has some limitations, as the urban soundscape itself is a multi-factor environment, complicating results. There are a large number of sudden events in the city that will be on the urban soundscape of the objective acoustic environment as well as part of the subjective perceptual evaluation of the impact of the factors in the analysis of the city’s soundscape. There is a certain degree of error and contingency; it will be necessary in future studies to increase the variety of interviewee backgrounds. In future studies, the different backgrounds of the respondents (such as education level, economic status, family situation, etc.) should be added and incorporated into the factors affecting the evaluation of soundscape perception, which can make the study of soundscape more comprehensive; in addition, further improving the research accuracy of objective acoustic indicators and subjective soundscape perception evaluation would allow us to study the changing laws affecting the city’s soundscape in a small-scale space.

Residential land exhibits a higher demand for tranquility; therefore, planning should establish “Acoustic Quiet Zones” by restricting through traffic and optimizing road network structures to mitigate the long-term penetration and persistence of vehicular noise. For commercial areas, it is advisable to proactively leverage the masking effect of human activity sounds on traffic noise and focus public attention towards natural sounds. Through carefully designed commercial green landscapes—such as the incorporation of natural “islands”—visual perception can be guided to enhance psychoacoustic experiences, thereby improving soundscape comfort. In industrial and transport hub zones, integrated strategies combining source control and transmission path intervention should be adopted. Emphasis should be placed on addressing low-frequency noise and intermittent noise events through physical noise barriers and spatial buffer zones.

Furthermore, urban planning should not rely solely on equivalent sound levels but should incorporate findings on sound source typology, frequency composition, and the intermittent characteristics of noise events into evaluation criteria. It is also essential to recognize that soundscapes result from the coupling of subjective perception and the objective environment. Future planning practices should introduce public participation and integrate socio-demographic factors—such as residents’ educational background and life experiences—into soundscape perception assessments. This approach enables more inclusive and targeted soundscape design. The ultimate goal is to effectively enhance urban acoustic environmental quality through interdisciplinary collaborative planning, thereby effectively safeguarding residents’ physical and mental health and overall well-being.

5. Conclusions

This study examined the influence of the urban objective sound environment on subjective soundscape perception and evaluation, and explored the potential pathways through which urban land-use types influence both the subjective and objective urban soundscape. The results show that:

(1)

Objective acoustic characteristics and soundscape perception evaluation are affected by the function of urban land use. Areas with the same urban soundscape distribution pattern have similar spatial functional characteristics, and soundscape evaluation will be affected by the surrounding environment. This is because the urban land-use function directly affects the natural environmental elements and crowd activities in the space, leading to changes in the objective acoustic environment, and affecting the evaluation of the subjective soundscape. Areas with the same distribution pattern of sound source perception tend to show similar spatial functional characteristics, and there are some differences in the distribution pattern of different categories of sound source perception. The overall spatial distribution of soundscape evaluation is relatively similar, but there are some differences in the urban spatial functions corresponding to high- and low-value areas. The distribution of soundscape evaluation is not random, and different soundscape evaluation indicators have significant spatial autocorrelation and spatial aggregation effects. Therefore, corresponding soundscape optimization strategies should be proposed for different urban spatial functions.

(2)

The critical objective acoustic indicators affecting the evaluation of urban soundscape perception are equivalent continuous A sound level L_Aeq, background sound L₉₀, L_C–L_A, and loudness. Sound level remains the most critical factor influencing individuals’ perception and evaluation of the sound environment, and L_Aeq has a key influence on all soundscape evaluation indicators. Individuals rate the quiet environment highly and experience a strong sense of discomfort in the noisy environment. However, it should be noted that soundscape evaluation results may also vary due to individual differences and other physical environment factors.

(3)

Urban land-use planning has a significant impact on the urban soundscape. Urban land-use planning, to a certain extent, delineates the behavior of individuals in its space and therefore produces soundscape characteristics corresponding to the behavioral activities of the crowd. There are significant differences in the objective sound environment in different urban land-use plans, and urban land-use plans also influence people’s perception of the sound environment.

Therefore, in future soundscape research and soundscape creation, deepening the research of urban land-use planning on urban soundscapes, introducing urban land-use planning as a method of regulating urban soundscapes, and guiding individuals’ activities in different spaces through urban land-use planning to change the urban soundscape may produce better results.

A key limitation of this study lies in its reliance on objective acoustic parameters (Table 1) to characterize the soundscape. While our methodology aligns with established practices in the field, it does not fully account for the semantic and functional dimensions of sound as emphasized in the foundational work of scholars like Schafer, Krause, and Columbiin. Our analysis primarily treats sounds based on their physical properties (e.g., frequency, level) and source type, but it does not differentiate based on their social or communicative functions. For instance, sounds designed to attract attention and convey critical information—such as public address announcements, sirens, or car horns—serve an essential warning function, even though they may induce a transient stress response. The inability of our current parametric framework to distinguish such functionally necessary, albeit potentially aversive, sounds from gratuitous noise represents a significant simplification. Future research should integrate perceptual evaluations of sound meaning and function to achieve a more holistic and societally relevant assessment of the urban acoustic environment. In addition, another limitation of this study lies in its reliance on objective acoustic parameters (Table 1) to describe soundscapes. Although our approach is consistent with established practices in the field, it does not adequately consider the semantic and functional dimensions of sounds, as emphasized in foundational studies by scholars such as Schafer, Krause, and Columbiin. Our analysis primarily processes sounds based on their physical properties (e.g., frequency, intensity) and source types, but does not distinguish them according to their social or communicative functions. For example, sounds intended to attract attention and convey critical information—such as public broadcast announcements, alarms, or car horns—may trigger brief stress responses yet serve an important alerting function. Our current parameterized framework cannot differentiate this functionally essential but potentially aversive noise from meaningless noise, which represents a significant simplification. Future research should incorporate perceptual assessments of sound meaning and function to achieve a more comprehensive and socially relevant evaluation of urban sound environments.