© The Author(s), Published by EDP Sciences, 2026

1 Introduction

According to the World Health Organization, noise is currently one of the main risks to human health and can cause several physical and mental problems, such as cardiovascular diseases, sleep disorders, hearing impairment, stress, irritability and attention problems [1]. Consequently, addressing noise pollution has become a central concern in urban planning and public health management.

For a long time, the management of human exposure to noise has followed an approach that focuses primarily on physical parameters, usually consisting of measuring sound pressure levels and comparing the results obtained with the thresholds defined in national regulations, in line with the framework established by the Environmental Noise Directive [2]. These thresholds correspond to limit values used to assess environmental noise exposure and to identify areas where mitigation measures may be required or recommended. Within this framework, sound is perceived only as a factor to be minimized or mitigated.

However, over the years, several studies have shown that reducing sound pressure levels not always leads to an improvement in quality of life [3–5], and that sound can instead be seen as a valuable resource, used to improve people’s health and well-being [6–8]. Consequently, a new perspective on this problem has emerged, called the “soundscape” approach.

This method differs from the traditional environmental noise management perspective in the sense that the first one perceives sound as a resource and focus on sounds of citizens’ preference, while the later conceives it as a waste to be reduced, and focus mainly on sounds of humans’ discomfort [9, 10]. The two approaches also differ in how sound is measured and managed. Traditional noise management is only dependent on the physical characteristics of sound, such as overall sound pressure level (SPL) and frequency spectrum description, assuming that human response is only related with these factors. It typically focuses on reducing overall noise levels without accounting for differences between sound sources. In contrast, the soundscape methodology recognizes that the human perception of sound involves more than just its physical properties, and therefore its goal is not necessarily to reduce overall SPL [9, 11].

Building on this broader understanding of sound perception, the soundscape approach does not aim to replace traditional environmental noise management but rather to complement it [12], for example, by integrating conventional tools such as noise mapping with soundscape-based design and evaluation [13, 14].

1.1 Soundscape approach

The concept of the soundscape was originated in 1969, when Southworth first applied it in an urban context [15], and was later popularized in 1977 by Canadian musician Schafer [16]. Since then, this concept has been used in various scientific branches, such as civil engineering, architecture, urban planning and design, ecology, social sciences, medicine, among others. Although soundscape research began in the late 1960s, it has gained significant attention over the past twenty-five years, particularly within the fields of community noise and environmental acoustics, first, among researchers and, more recently, among policy makers and practitioners [8]. The growing interest in this field of research has led to the development of standardized approaches for defining and analysing soundscapes, resulting in the ISO/TS 12913 standard series [17–19].

According to the first part of the standard [17], soundscape can be defined as an “acoustic environment as perceived or experienced and/or understood by a person or people, in context”. In this sense, the soundscape of a place differs from its acoustic environment, in that it refers to a perceptual experience, whereas the latter primarily describes the physical phenomenon, without explicitly addressing sound perception, although both are shaped by context of a particular time, place, and activity. This distinction underscores the importance of subjective perception in evaluating a place’s sound environment, pointing out that while physical measurements and parameters are necessary, they alone are insufficient to fully characterize, assess and understand the acoustic experience of a place.

For this reason, in the soundscape analysis, the sound sources play a crucial role, as different types of sounds are often associated with different perceptions of the acoustic environment, affecting the acoustic comfort evaluation, i.e., the subjective judgment of the degree to which an acoustic environment is perceived as pleasant, acceptable and appropriate for users and their activities [3, 10]. For instance, even under controlled loudness conditions, soundscapes dominated by technological and industrial sounds are typically perceived as unpleasant, whereas those dominated by natural sounds are considered pleasant and soundscapes dominated by human sounds to be eventful [20]. However, it is important to note that the contribution of individual sound sources to soundscape quality varies depending on contextual factors, and when the appropriateness of the sound environment is considered, some of these relationships may not be immediately evident. For example, Yang et al. [21] found that human-related sounds, such as commercial sounds, negatively impacted the perceived appropriateness of soundscapes in urban parks, as they conflicted with the parks’ peaceful atmosphere. Moreover, according to Jo and Jeon [22], even noise sources such as traffic can, in certain contexts, positively influence urban soundscape perception, even though their overall impact on soundscape quality remains predominantly negative.

Essentially, this perspective shifts the focus away from mere quietness as the primary factor for acoustic preferences in outdoor environments. Instead, the consistency between the soundscape and the surrounding landscape is a key factor, as research shows that even places characterized by high sound pressure levels, can still have a good soundscape perception [23–25]. In this sense, not all sounds should be considered as a contamination of the urban environment, but rather as a part of its identity, and therefore worthy of preservation and protection from intrusive or unwanted sources.

Regarding the methods to evaluate a place’s soundscape, according to the ISO/TS 12913-2 [18] standard, this process can be conducted either in situ, typically by performing a soundwalk, or off-site, by reproducing the sound using headphones or loudspeakers in controlled conditions, instead. The soundwalk method involves participants walking through a designated area while focusing on its acoustic environment and then answering a questionnaire. This standard recommends a group of at least 20 participants, with each listening silently and attentively to the sound environment for a minimum of three minutes at each stop, before completing the questionnaire. In addition to the human perception collected via the questionnaires, binaural acoustical measurements should also be performed to provide information on physical and psychoacoustic indicators [18]. The recordings should also have a minimum duration of three minutes. The third part of the same standard outlines the guidelines for analysing the data collected using the methods specified in Part 2 [19].

1.2 Soundscape of historical places

In 2003, the UNESCO Convention for the Safeguarding of the Intangible Cultural Heritage [26], recognized the existence of a close relationship between tangible and intangible cultural heritage, which together create the unique cultural identity of a place.

In this sense, several studies have been conducted with the aim of characterizing and reproducing the indoor sound environment experienced at some important monuments [27–29], recognizing that the acoustic environment of these buildings is an equally significant cultural element as their architectural and monumental features.

This growing interest in the acoustic characteristics of places with high historical and cultural values is also evident in the field of soundscape analysis. Consequently, more studies have been focusing on understanding how the sound environment of such places is perceived by visitors and locals, and how it affects, or it may have affected, their experience of the space. For instance, in the medieval historic centre of Cáceres, in Spain, it has been found that vehicles, passers-by, and other individuals made major contributions to the area’s sound energy, masking other noise sources such as birds and animals [30]. Similarly, Brambilla and Maffei [31] highlight that even relatively low-level sounds from motorcycles can negatively influence people’s perception and enjoyment of a place, illustrating that human-generated noise can have a significant impact on acoustic experience. Additionally, as shown in Huang and Kang [32], in the historic city of Lhasa, although some traditional sounds still exist, the sound pressure levels in the historic centre were found to be high, potentially making this area acoustically uncomfortable and affecting the perception of positive sound sources.

Moreover, the globalization and homogenization of human activities can lead to similarities in soundscapes of historical places with different cultural backgrounds. In Zhiyong et al. [33], two city centres in the Sinkiang and Hebei provinces of China were selected to assess the soundscape in historical areas and its relationship with the local community. The study found that local people’s assessments of the soundscape strongly depended on whether the sounds preserved traditional ambient characteristics or reflected new sounds resulting from urban development and modernization, rather than on objective sound pressure levels. Furthermore, the study highlights the necessity of considering subjective human perception, especially from the local community, and cultural context when planning and managing historical urban soundscapes, in order to preserve their acoustic identity and heritage and to mitigate soundscape homogenization caused by modern development.

Therefore, soundscape analysis of sites with high historical and cultural value is essential, as it helps to identify the characteristic sounds that contribute to a place’s cultural identity, recognized by UNESCO as intangible cultural heritage, and also helps defining planning actions, to reduce unwanted noise from traffic, tourism, or nearby commercial activity that might interfere with the visitors’ experience. In this way, the soundscape approach contributes to the protection and preservation of the acoustic environment, not only by enhancing how people connect with the space but also by supporting the broader goal of safeguarding both tangible and intangible aspects of cultural heritage.

1.3 Present study’s objective

This study aims to evaluate the soundscape of two areas of high heritage value located in the city of Coimbra, in Portugal, by collecting both perceptual and physical data (sound pressure level measurements), through the soundwalk method, following the guidelines defined in ISO/TS 12913-2 [18].

The selected locations, recognized for their cultural and historical values, are currently under evaluation as part of an initiative to redesign, revalue and enhance public spaces. By analysing the acoustic environment of these historical areas and how it is perceived, the study seeks to understand the impact of existing sound conditions, contributing to more informed and sustainable urban planning decisions. The study was conducted in context, with data collected from a defined set of locations at a specific point in time (with only a single assessment carried out at each site). Therefore, it represents a first attempt to understand the sound environment of these areas and how it is perceived, providing an initial basis for future research and urban planning interventions.

2 Case study: areas of high heritage value in Coimbra

As previously mentioned, this study evaluates the soundscape of two areas of high heritage value in Coimbra, namely the João das Regras Avenue (see Fig. 1), and the Polo I Campus of the University of Coimbra (UC) (see Fig. 2).

Figure 1. Study site: João das Regras Avenue.

Figure 2. Study site: Polo I Campus of the University of Coimbra.

Located along João das Regras Avenue is one of the city’s most significant historical monuments, the Monastery of Santa Clara-a-Velha, which dates back to the 13th century and holds great meaning and cultural importance to the city of Coimbra. Over time, the historic soundscape of this area has gradually been replaced by contemporary urban noises, namely traffic and sounds originating from nearby cafes and restaurants. The most prominent historically rooted sounds that remain today are those associated with the Monastery of Santa Clara-a-Nova, which was built as a replacement for Santa Clara-a-Velha in 1647, due to recurrent flooding [34].

Since 2013, the Polo I Campus (Alta and Sofia) of the University of Coimbra (UC) is classified in the UNESCO World Heritage List, recognized for its architecture and specific urban typology, and for its unique role in the history of Portuguese world-wide influence, education and culture [35]. Regarding the evolution of the area’s soundscape, several elements of the historical sound environment remain present, particularly sounds associated with students and their academic traditions, as well as the iconic chime of the University’s Tower, known as the “Cabra” [36]. At the same time, new auditory elements have emerged, including some traffic noise and sounds associated with touristic activity, such as conversations in multiple languages and accents, as well as the noise of tourist buses [36]. It is worth mentioning, that the presence of tourists has increased significantly since the site’s classification on UNESCO’s World Heritage List.

Although both locations present notable historical, cultural, and touristic relevance, they represent two distinct geographical and urban contexts. João das Regras Avenue is characterized by relatively wide streets and a high daily traffic volume (approximately 1700 veh/h during the morning rush hour [37]). The area is predominantly composed of mid-rise residential and commercial buildings, with three to four stories, and includes an urban park that introduces vegetation into the surrounding environment (Fig. 1).

In contrast, the Polo I campus of UC is located in the historic upper part of the city of Coimbra, with a steeper topography, narrower streets and some small plazas, presenting significantly lower daily traffic volumes (around 200 veh/h at the busiest streets during the morning rush hour [38]) and slower-moving vehicles. The built environment is characterized by older and taller buildings, and the presence of vegetation is limited to a few trees along sidewalks and small garden areas (Fig. 2).

3 Materials and Methods

As mentioned above, the study was conducted by applying the soundwalk methodology in compliance with the requirements specified in the ISO/TS 12913-2 standard [18]. However, the binaural recordings were not performed and, therefore, psychoacoustic indicators were not included in the analysis.

The soundwalk paths are illustrated in Figures 3 and 4. The difference in the number of measurement points at each location is justified by the marked differences in the urban context of the two zones, as well as in the area dimensions. In the first location (João das Regras Avenue), the objective was to evaluate the soundscape of a green urban park located near a historical building and mainly used for recreation and leisure. As this area presents relatively low spatial variability, the two selected points were chosen to capture the main contrasts in terms of traffic proximity and visibility, as well as the presence of greenery.

Figure 3. Soundwalk path for João das Regras Avenue, with two measurement points (Source: Google Maps).

Figure 4. Soundwalk path for Polo I Campus of UC, with six measurement points (Source: Google Maps).

At Polo I campus of UC, the objective was to assess the soundscape of a historical urban area characterized by greater spatial and functional diversities. This location exhibits higher variability in traffic presence, urban morphology (including squares and narrow and wide streets), and the presence and activity of people. Therefore, in order to provide participants with a limited yet representative set of locations capable of characterizing the overall sound environment of the site, six measurement points were defined for the soundwalk.

During the walks, the participants were led along a predefined route. They were asked to stop at the predefined locations to listen in silence during a 4 min period of time and to use all their senses to perceive the site. The person leading the soundwalk instructed the participants to pay attention to what sounds were heard, what they liked and disliked and how they believed the acoustic environment could be improved. After the listening period, the participants were then requested to fill out online questionnaires performed using a google form which was made available using personal smartphones, through a QR code. When completing the assessment, the participants were asked to move to the next defined site.

This survey, whose questions are presented in Table 1, was based on Methods A and B provided in the referred standard [18]. Although the questionnaires were made to be mutually exclusive, some literature suggests that the choice of one protocol over the other can lead to some differences in the soundscape data collected and that a combined version of the two methods would constitute a more reliable option [39]. All questionnaire responses were collected anonymously, ensuring full compliance with privacy regulations.

As previously mentioned, in addition to the subjective perception data collected via the questionnaires, data on physical indicators should also be ensured. Therefore, in parallel with the 4-minute listening evaluation, sound pressure level (SPL) measurements of the equivalent sound pressure level, L_Aeq, and of the A weighted exceedance levels L₉₀, L₅₀, and L₁₀, were also performed for the same 4 min time. The standard requires a minimum measurement time interval of 3 min. The period of 4 min was adopted as a compromise that allows the capture of all significant and typical sound sources and events while avoiding an excessive extension of the soundwalk. The equipment used was a Class 2 sound level meter, Rion NL-42 [40], in accordance with IEC 61672-1:2002 standard, adjusted to fast time weighting, which allows a frequency range between 20 Hz and 8 kHz and a sampling frequency of 48 kHz. The time step employed in the measurements logging was 1 s.

By applying the soundwalk methodology that combines questionnaires with sound level measurements, a more complete and comprehensive representation of the acoustic environment is obtained, which constitutes a fundamental factor for its effective analysis and management.

The first soundwalk took place at João das Regras Avenue in the morning, on Friday, November 22nd, 2024, between 10:32 AM and 10:36 AM, in the first measurement point, and between 10:44 AM and 10:48 AM, in the second (Portuguese time) (Fig. 5). During the evaluation, a temperature of 14 °C, relative humidity of 80% and wind speed of 15 km/h were measured. The selected participants were a group of 30 students (12 women, 18 men), Portuguese native speakers, with ages between 18–24 years old, with basic knowledge in environmental acoustics and no particular association with the place (all had been there at least once, but most of them were not frequent visitors).

Figure 5. Picture of the first measurement point at João das Regras Avenue during the soundwalk.

As mentioned above, this location included two evaluation points, both situated near some of the city’s main roadways, even though the first point was clearly more exposed. During the soundwalk, and throughout the four-minute recording period, a total of 96 vehicles was counted at the first evaluation point and 72 vehicles at the second, with a small portion of this traffic (about 4%) consisting of heavy vehicles, such as buses and trucks.

The second walk took place at Polo I of the University of Coimbra, on a Friday afternoon, January 16th, 2025, between 3:30 PM and 5:00 PM (Fig. 6). During the evaluation, a temperature of 13 °C, relative humidity of 65% and wind speed of 12 km/h were registered. The participants were a group of 19 foreign students (living in Coimbra for 1 semester), and one Professor (7 women, 13 men) from a Master’s programme in acoustics, with ages between 20–50 years old and no particular association with the place (all had been there at least once, but most of them were not frequent visitors).

Figure 6. Picture of the six measurement points at Polo I of UC during the soundwalk.

At this location, of the six selected evaluation points, one was closed to general traffic (the fifth measurement point), while another was reserved exclusively for public transport (the sixth measurement point). During the four-minute measurement periods, traffic volumes consisted entirely of light-weight vehicles, with 15 recorded at the first location, 4 at the second, 3 at the third, 5 at the fourth, and 1 at the sixth location. Since, in the first case, all students were Portuguese, the questionnaires were translated for easier interpretation. Although efforts are being made to develop official translations of the standard [41], a complete protocol is not yet available for Portuguese language. Therefore, the authors carefully have selected the most appropriate terms to keep the original meaning and ensure clarity. In particular, for protocol regarding Method A, which includes the eight soundscape attributes that define the perceived affective qualities (PAQ) of physical acoustic environment (see English version of the PAQ in Tab. 1), the Portuguese translation provided within the Soundscape Attributes Translation Project [41] was employed.

For the second soundwalk, attended by foreign students, the questionnaires were completed in English, as all participants possessed a minimum B2 English proficiency level, previously certified through IELTS or Cambridge examinations.

In both cases, a briefing session prior to the soundwalk was promoted, in order to introduce the participants to the concept of soundscape, the study’s objectives and methodology and showing the questionnaire to ensure participants’ comprehension of the questions. In particular, regarding PAQ, besides the questions to be answered and possible responses, the two-dimensional graphical representation, with “Pleasantness vs. Unpleasantness” in the X-axis and “Eventfulness vs. Uneventfulness” on the Y-axis, was also displayed. The session also aimed to address the gaps in acoustic knowledge between the two groups, thereby ensuring that such differences would not influence the results.

4 Results and discussion

Once the soundwalks were completed, the data collected was analysed following the guidelines outlined in the third part of the ISO/TS 12913 standard [19]. The measurement results, for the four parameters considered (L_Aeq, L₁₀, L₅₀ and L₉₀) are presented in Tables 2 and 3, for each recorded point in both locations. On the other hand, the questionnaire responses are plotted and summarized in Figures 7–24.

Figure 7. Overall time-weighted and per-second SPL levels in the two measurement points at João das Regras Avenue.

4.1 João das Regras Avenue

At João das Regras Avenue, results show two very distinct scenarios, according to Table 2 and Figures 7–15. The first measurement point exhibited a high L_Aeq of 58.2 dB(A), with the statistic sound levels L₉₀ and L₁₀ fluctuating between 53.7 and 60.4 dB(A), and occasional peaks reaching up to 65 dB(A), as illustrated by Table 2 and Figure 7.

The dominant sound source at this point was traffic noise, as confirmed by both the sound sources subjective analysis chart, from Method A, and the open-ended responses, from Method B (respectively, Figs. 8 and 12), where traffic noise was referred in all the answers. Additional sources identified included sounds from traffic lights (pedestrian warnings), mechanical equipment, as well as people and birds, though these were clearly secondary (sounds of nature were reported in 53% of the responses, while all other sources were mentioned in less than 35% of the answers) (Fig. 12).

Figure 8. Sound source identification (Method A of ISO/TS 12913-2 [18]): mean scores and range – João das Regras Avenue (MP: Measurement Point).

The perceptual experience at the first measurement point was notably negative. As illustrated by the radar plot of median PAQ scores in Figure 9 and by the 2D (Pleasant-Eventful) soundscape model in Figure 10, the first point falls within the eventful and annoying (chaotic) quadrant, indicating a high level of unpleasantness. Participants commonly described the environment using terms such as “noisy”, “annoying”, and “stressful”, as reported in 63%, 23% and 20% of all the answers, respectively, reflecting a generally unfavourable acoustic perception (Fig. 13).

Figure 9. Perceived Affective Qualities (PAQs) median scores according to Method A of ISO/TS 12913-2 [18] (radar plot) – João das Regras Avenue (MP: Measurement Point).

Figure 10. Graphical representation of the 2D soundscape model according to Method A of ISO/TS 12913-2 [18] (scatter plot, individual and median scores) – João das Regras Avenue.

In contrast, experience at the second evaluation point was considerably more pleasant. Participants classified this location as highly pleasant and calm (Fig. 9), placing it within the Pleasant–Uneventful quadrant of the 2D soundscape plot (Fig. 10). As shown in Figure 13, described the environment primarily as “calm”, which was referred in 50% of the answers, followed by terms like “relaxing,” and “neutral”, reported in 30% of the answers. The overall soundscape was rated more positively than at the first point, however, regarding the appropriateness of the sound environment, both locations presented similar results (Fig. 11), suggesting that the differences in the soundscape from the two locations somehow are aligned with the differences in context: at the first measurement point, due to its close proximity to the main road and several cafes and restaurants, participants likely expected a more lively and noisy sound environment, consistent with what was observed during the soundwalk, whereas at the second point, located further inside the park, a quieter sound environment was expected and was also reflected in participants’ perceptions during the soundwalk. Unlike the first measurement point, one dominant sound source was not identified, but rather the sound of nature together with traffic noise was more noticed than other sounds, as evidenced by both the sound source identification chart from Method A (Fig. 8) and the open-ended responses from Method B (Fig. 12), where traffic was referred in 90% of the answers and nature in 87%.

Figure 11. Assessment and appropriateness of the surrounding sound environment (Method A of ISO/TS 12913-2 [18]): mean scores and range – João das Regras Avenue (MP: Measurement Point).

Figure 12. Sound source recognition (Method B of ISO/TS 12913-2 [18]): word clouds – João das Regras Avenue.

Figure 13. Subsequent comments (Method B of ISO/TS 12913-2 [18]): word clouds – João das Regras Avenue.

The SPL measurements support these subjective impressions as the second point had a much lower L_Aeq value of 49.7 dB(A), when compared to 58.2 dB(A), indicating a quieter environment (Tab. 2). The plot of the per second-SPL, illustrated in Figure 7, also suggests fewer and less intense noise events.

Figure 14 presents the frequency spectra of the acoustic environment at both locations, expressed in dB(A). The spectra in dB were also included to investigate the presence of tonal noise components that may be masked by A-weighting the responses. The adopted 1/3 octave frequency range (from 50 to 6300 Hz) was selected in accordance with the recommendations of ISO 1996-2 [42] (which define a frequency range between 63 to 8000 Hz in octave bands) and taking into consideration the upper frequency limit of the equipment which is 8000 Hz. The data indicate no major spectral variation between the two sites other than the overall sound amplitude. Additionally, an energy concentration around 1000 Hz (within the 1/3 octave band corresponding to the frequency range of highest human auditory sensitivity) is observed in the plot given in dB(A), which is characteristic of road traffic related noise, one of the most audible sound sources in this location.

Figure 14. Frequency spectra of the two measurement points at João das Regras Avenue, represented in dB(A) and dB, respectively on the left and on the right.

Overall, the results from the soundwalk performed at João das Regras Avenue suggest that there is a relationship between subjective loudness and unpleasantness, as the louder environment (the first measurement point) was identified as the less pleasant (Fig. 15) and overall participants’ experiences are consistent with the physical measurements collected. This relationship was further confirmed by the Spearman correlation analysis performed between the variables “subjective loudness” and “unpleasantness” (Method B), showing a considerable positive correlation at both measurement points (1st MP: ρ = 0.661,  p <  0.01; and 2nd MP: ρ = 0.582,  p <  0.01). The sound sources were also slightly different at the two places and support the idea that sounds of nature (in the second measurement point) are usually associated with a more pleasant experience of the sound environment.

However, a particularly interesting and unexpected result emerged from the participants’ responses to the question “How often would you like to visit this place again?”. Despite the first evaluation point being clearly rated as less acoustically pleasant than the second, participants showed a slightly higher preference for returning to the first one (Fig. 15). This suggests that the acoustic pleasantness evaluated through Method B question and revisiting intention are not always directly correlated, and that other contextual or environmental factors, such as architectural character, urban involvement, cultural relevance or visual appeal, among others, may partially compensate the poorer sound environment. This interpretation is further supported by the absence of a significant correlation between “Pleasantness (PAQ)” and “revisiting intention” at both measurement points (1st MP: ρ = 0.164, p >  0.05; 2nd MP: ρ = 0.135, p >  0.05).

Figure 15. Assessment of the sound environment (Method B of ISO/TS 12913-2 [18]): mean scores and range – João das Regras Avenue (MP: Measurement Point).

4.2 Polo I Campus of UC

On the other hand, the results obtained for Polo I Campus of UC reveal a different scenario, as the values for the pleasantness dimension in the 2D soundscape model (Fig. 19) are not fully aligned with the sound pressure levels (Tab. 3 and Fig. 16).

Figure 16. Overall time-weighted and per-second SPL levels in the six points selected at Polo I Campus of UC.

At the first measurement point, the overall SPL level was evaluated as 53.4 dB(A), with a few significant peaks exceeding 60 dB(A) and even reaching above 70 dB(A) at a certain moment (Fig. 16). Regarding the sound sources, according to the answers from Method B, illustrated in Figure 22, participants identified a combination of sound sources, including traffic noise, human voices, and music, which were referred in 90%, 80%, and 70% of the answers, respectively. This is also confirmed by the ratings of each sound source from Method A, illustrated in Figure 17. As for the perceived affective quality, as illustrated in Figures 18 and 19, this location was classified as highly pleasant and calm, falling within the Pleasant–Uneventful quadrant of the 2D soundscape model (Fig. 19). This perception is confirmed by the open-ended question from Method B, with adjectives like “calm” (35%), “neutral” (20%) and “relaxing” (20%) being the most used by participants to describe the sound environment at this place, even though some also considered it to be “noisy” (10%) and “uncomfortable” (10%) (Fig. 23). It is interesting to note that, even though this measurement point did not receive one of the highest classifications for overall assessment of the sound environment (Fig. 20), it did, however, received one of the lowest unpleasantness scores (Fig. 21), reinforcing a generally positive auditory experience.

Figure 17. Sound source identification (Method A of ISO/TS 12913-2 [18]): mean scores and range – Polo I Campus of UC (MP: Measurement Point).

Figure 18. Perceived Affective Qualities (PAQs) median scores according to Method A of ISO/TS 12913-2 [18] (radar plot) – Polo I Campus of UC (MP: Measurement Point).

Figure 19. Graphical representation of the 2D soundscape model according to Method A of ISO/TS 12913-2 [18] (scatter plot, individual and median scores) – Polo I Campus of UC (MP: Measurement Point).

Figure 20. Assessment and appropriateness of the surrounding sound environment (Method A of ISO/TS 12913-2 [18]): mean values and range – Polo I Campus of UC (MP: Measurement Point).

Figure 21. Assessment of the sound environment (Method B of ISO/TS 12913-2 [18]): mean scores and range – Polo I Campus of UC (MP: Measurement Point).

Figure 22. Sound source recognition (Method B of ISO/TS 12913-2 [18]): word clouds – Polo I Campus of UC.

Figure 23. Subsequent comments (Method B of ISO/TS 12913-2 [18]): word clouds – Polo I Campus of UC.

The second measurement point, despite having a slightly lower SPL level of 52 dB(A) (Tab. 3), was perceived as less pleasant than the first one, as shown by its classification in the 2D soundscape plot illustrated in Figure 19. This is likely due to the strong presence of noise from construction works and traffic, even though “people talking” was identified as the dominant sound source, being reported in 80% of the answers (Fig. 22). The soundscape at this location was described as “neutral” in 45% of the answers, with participants reporting mixed impressions such as “calm” (25%), “busy” (10%), and “chaotic” (10%) (Fig. 23). These responses reflect an environment where contrasting sound elements disrupt the perception of tranquillity.

The third measurement point, from all the six places evaluated, was the one with the lowest overall SPL level, with an L_Aeq value of 47.8 dB(A) (Tab. 3). This location presented the quietest environment, mostly described as “relaxing”, “calm” and “serene”, which was referred in 35%, 40% and 20% of the answers, respectively (Fig. 23). Although it did not receive the highest score for the pleasantness dimension in the perceived affective quality 2D model (Fig. 19), it did, however, received the highest rating on the overall description of the sound environment (Fig. 20). The dominant sound source was traffic noise, which was referred in 70% of the open-ended answers from Method B (Fig. 22), although it was distant and identified as sporadic passages, as confirmed by the low SPL values (Fig. 16). These results suggest that, because traffic noise is the main sound source and other sources were not dominant, the sound environment may have been rated as slightly less pleasant. Additionally, church bells, audible due to the site’s proximity to the Cathedral of Coimbra, contributed positively to preserve the cultural identity of the space (Fig. 22). Regarding the perceived loudness, this location had the lowest rating of all the points evaluated (Fig. 21).

The fourth measurement point exhibited the highest overall sound pressure level, L_Aeq = 56.5 dB(A) (Tab. 3), even though, as evidenced by Figure 16, there was a significant variation through the 4 min of recording, with moments of low SPL followed by periods of SPL values above 60 dB(A). Traffic noise was undoubtedly the main sound source, as evidenced by Figure 17 and by Figure 22, with all participants referring traffic-noise in their answers. Although traffic counts indicate that the number of vehicles passing close to this measurement point is comparable to those at other locations, traffic was perceived as significantly more dominant in the sound environment. This is likely due to the proximity of the participants to the roadway, the acoustic properties of the road pavement, and the strong contrast between traffic noise and an otherwise relatively uneventful environment. Regarding the perceived affective quality, this location was the only one that resulted in a negative value for the pleasantness dimension in the 2D soundscape plot (Fig. 19), being described by participants as “noisy” and “uncomfortable”, in 15% and 20% of the answers, respectively, but also “neutral” (25% of the answers), possibly reflecting the contrast between quiet and noisy periods (Fig. 23).

At the fifth measurement point, the overall SPL value was 51.3 dB(A), with 90th to 10th percentile values ranging from 48.0 and 53.3 dB(A) (Tab. 3), showing minor variation over time, as can be confirmed in Figure 16. Sounds from human beings, such as conversations, footsteps, and music, were clearly the most audible sound sources, as confirmed by both the results from Method B, where these sources were reported in 85%, 35%, and 30% of answers, respectively, and the sound source analysis chart from Method A (Figs. 17 and 22). As illustrated by Figure 19, this location was the only one whose answers fell exclusively on the Pleasant side of the 2D soundscape plot, being described mostly as “calm” and “relaxing”, each being referred in 30% of the answers. It was also the one with the lowest unpleasantness score and the highest rating for revisit preference (Fig. 21).

Finally, the sixth measurement point showed interesting results. Despite having one of the highest overall SPL values, 56.2 dB(A) (Tab. 3), it was classified on the pleasant-eventful (vibrant) quarter (Fig. 19), even though its position along the pleasantness axis was very close to a neutral value. This indicates that the sound environment was perceived as lively rather than distinctly pleasant or unpleasant. This interpretation is further supported by the results shown in Figure 16, which indicate that this location received one of the lowest scores in the overall description of the sound environment, but a good appropriateness rate. Regarding the participants’ comments, descriptors such as “busy” (25%), “crowed” (30%), “neutral” (25%) and “nice” (25%) were the most frequently used by participants to characterize the sound environment at this location (Fig. 23). These adjectives suggest a high level of activity and a lively atmosphere, but not necessarily in a negative and unpleasant way. As shown in Figures 17 and 22, this relatively positive perception of the sound environment appears to be influenced by the dominance of human-related sounds, such as conversations and footsteps, which were referred in 90% of the open-ended answers from Method B, despite the overall loudness of the environment and the noticeable presence of traffic noise from the nearby streets.

Regarding the frequency spectra of the six measurement points, illustrated in Figure 24, in both dB(A) and dB, the results exhibit similar spectral profiles, despite some variations in amplitude, which suggests a characteristic spectrum for the sound environment at this historical location. The sixth measurement point, however, deviates slightly from this trend, as it lacks the energy peak typically observed in the 800–1000 Hz range with a shift towards lower frequencies, likely due to the reduced presence of traffic related noise at this specific location and the increase of public transport noise contribution.

Figure 24. Frequency spectra of the six measurement points at Polo I Campus of UC, represented in dB(A) and dB, respectively on the left and on the right.

These findings suggest that the type of sound source plays a crucial role in evaluating the pleasantness of a place, since even though the fourth and the sixth measurement points had very similar overall SPL values, and number six had the highest rating on perceived loudness, this location was rated slightly more positively on the Perceived Affective Quality 2D soundscape model (Fig. 19) and received a higher classification for revisiting preference (Fig. 21). Such outcomes demonstrate that human activity related sounds, even when loud, are often associated with positive or engaging experiences, while traffic-dominated environments, even at similar levels, are more likely to be experienced as negative.

Moreover, although the third measurement point was the one with the lowest overall SPL, it did not receive the highest classification for pleasantness on the Perceived Affective Quality 2D model, appearing very close to the first and fifth locations. This again reinforces the conclusion that the qualitative nature and the context of sounds, rather than their intensity alone, may influence how people evaluate and engage with sound environments. It is important to note that the third measurement point received the highest rating for overall description of the sound environment (Fig. 20), and got the highest score regarding calmness and not far from the highest one regarding pleasantness, in the 2D soundscape model (Fig. 19), indicating that the points classification is sometimes not so tangible.

Another interesting result emerges from the analysis of Figure 21. Although some locations received a positive evaluation regarding their sound environment, none received a very high rating for intention to revisit the place. This suggests that the results may be influenced by non-acoustic factors. Although the participants were students from the University of Coimbra, they do not attend classes at Polo I. Therefore, this area is perceived primarily as a touristic site, rather than a functional academic space. As a result, once visited, participants may have little motivation to return, which could explain the low revisiting intention rates despite the good sound environment evaluations.

5 Conclusions

The aim of the present study was to analyse the soundscape of two historical areas of high heritage value in the city of Coimbra, in Portugal, according to the soundwalk methodology outlined in the ISO/TS 12913 international standard series. This approach, which combined subjective questionnaire responses and objective sound pressure level measurements, allowed for a more complete evaluation of the sound environment of these places, by considering the three key components of soundscapes: acoustic environment, people and context.

The results obtained corroborate the idea that the perception of acoustic environments is shaped by far more than sound pressure levels alone. While SPL values provide a necessary baseline for understanding environmental noise, the findings demonstrate that the type, variability, and contextual meaning of sound sources play decisive role in shaping user experience and perception.

As demonstrated by the fourth and sixth measurement points, at Polo I Campus of UC, two locations with similar overall SPLs can be perceived differently, depending on the nature of the dominant sound sources. In particular, environments in cultural and historical contexts, where human and nature sounds dominate are more likely to be rated positively. This is illustrated by the positive evaluations of the second location on João das Regras Avenue, where sounds from nature significantly contributed to the soundscape, and the fifth measurement point at Polo I, dominated by human sounds. On the other hand, spaces dominated by mechanical or traffic noises tend to be perceived as annoying or uncomfortable, as seen at the first measurement point on João das Regras Avenue and the fourth location at Polo I, where traffic noise clearly dominated over other sound sources.

The data from the third measurement point also reveal that quietness is not necessarily synonymous of more pleasantness, since although this location had the lowest overall SPL, it did not receive the highest ratings for pleasantness on the Perceived Affective Quality 2D model. At this point, traffic noise dominated the sound environment but as the source was distant and other noise sources were not relevant, it may have resulted in a slightly less pleasant evaluation.

Regarding the methodology used, this study confirmed the advantages of implementing a combination of Methods A and B from ISO/TS 12913 international standard. Although some responses, such as overall assessment and unpleasantness scores, may be related, as shown before, the evaluation of the sound environment is not always in line with revisiting preference, as evidenced by the low correlation of the two variables (p >  0.05). Furthermore, the open-ended questions provided insights into the sound environment that are not always captured by the predefined categories of Method A.

As mentioned above, one of the main limitations of the present study is that, although recommended by the standard, no binaural recordings were conducted and, therefore, psychoacoustic parameters could not be assessed. Additional limitations include the fact that none of the participants were regular users of the places evaluated, as well as the absence of an analysis of contextual and environmental factors, such as architectural character, urban integration, cultural relevance, or visual appeal. These aspects will be addressed and incorporated in future research.

In conclusion, this research emphasizes the need of considering not only physical indicators, but also contextual factors and sound sources characterization when assessing the acoustic environment of a place. Adopting this comprehensive approach, that combines the traditional environmental noise management methods with the soundscape approach, is particularly important in areas of high historical value, where it is critical to ensure management and planning actions that both preserve the characteristic sounds contributing to a place’s cultural identity, recognized as intangible heritage, and also mitigate unwanted sounds that affect the visitor experience.

Moreover, this study highlights soundwalks, and complementary sound pressure level measurements, as a valuable methodology for exploring and evaluating the soundscape of a given area.

Acknowledgments

The authors would like to acknowledge the volunteers that participated in the soundwalks, and the Coimbra City Council for providing information and collaborating in this study. This work was funded by FCT/MCTES under the R&D Unit Institute for Sustainability and Innovation in Structural Engineering (ISISE), under the references UID/4029/2025 (https://doi.org/10.54499/UID/04029/2025) and UID/PRR/04029/2025 (https://doi.org/10.54499/UID/PRR/04029/2025), and under the Associated Laboratory for Advanced Production and Intelligent Systems (ARISE) under reference LA/P/0112/2020. Funding was also obtained under the NEXUS project “Agenda Mobilizadora Sines Nexus Pacto de Inovação – Transição Verde e Digital para Transportes, Logística e Mobilidade” (under reference: IT0251.17).

Funding

This work was funded by FCT/MCTES under the R&D Unit Institute for Sustainability and Innovation in Structural Engineering (ISISE), under the references UID/4029/2025 (https://doi.org/10.54499/UID/04029/2025) and UID/PRR/04029/2025 (https://doi.org/10.54499/UID/PRR/04029/2025), and under the Associated Laboratory for Advanced Production and Intelligent Systems (ARISE) under reference LA/P/0112/2020. Funding was also obtained under the NEXUS project “Agenda Mobilizadora Sines Nexus Pacto de Inovação – Transição Verde e Digital para Transportes, Logística e Mobilidade” (under reference: IT0251.17).

Conflict of interests

No potential conflict of interest was reported by the authors.

Data availability

Data are available on request from the authors.

References

All Tables

All Figures

	Figure 1. Study site: João das Regras Avenue.
In the text

	Figure 2. Study site: Polo I Campus of the University of Coimbra.
In the text

	Figure 3. Soundwalk path for João das Regras Avenue, with two measurement points (Source: Google Maps).
In the text

	Figure 4. Soundwalk path for Polo I Campus of UC, with six measurement points (Source: Google Maps).
In the text

	Figure 5. Picture of the first measurement point at João das Regras Avenue during the soundwalk.
In the text

	Figure 6. Picture of the six measurement points at Polo I of UC during the soundwalk.
In the text

	Figure 7. Overall time-weighted and per-second SPL levels in the two measurement points at João das Regras Avenue.
In the text

	Figure 8. Sound source identification (Method A of ISO/TS 12913-2 [18]): mean scores and range – João das Regras Avenue (MP: Measurement Point).
In the text

	Figure 9. Perceived Affective Qualities (PAQs) median scores according to Method A of ISO/TS 12913-2 [18] (radar plot) – João das Regras Avenue (MP: Measurement Point).
In the text

	Figure 10. Graphical representation of the 2D soundscape model according to Method A of ISO/TS 12913-2 [18] (scatter plot, individual and median scores) – João das Regras Avenue.
In the text

	Figure 11. Assessment and appropriateness of the surrounding sound environment (Method A of ISO/TS 12913-2 [18]): mean scores and range – João das Regras Avenue (MP: Measurement Point).
In the text

	Figure 12. Sound source recognition (Method B of ISO/TS 12913-2 [18]): word clouds – João das Regras Avenue.
In the text

	Figure 13. Subsequent comments (Method B of ISO/TS 12913-2 [18]): word clouds – João das Regras Avenue.
In the text

	Figure 14. Frequency spectra of the two measurement points at João das Regras Avenue, represented in dB(A) and dB, respectively on the left and on the right.
In the text

	Figure 15. Assessment of the sound environment (Method B of ISO/TS 12913-2 [18]): mean scores and range – João das Regras Avenue (MP: Measurement Point).
In the text

	Figure 16. Overall time-weighted and per-second SPL levels in the six points selected at Polo I Campus of UC.
In the text

	Figure 17. Sound source identification (Method A of ISO/TS 12913-2 [18]): mean scores and range – Polo I Campus of UC (MP: Measurement Point).
In the text

	Figure 18. Perceived Affective Qualities (PAQs) median scores according to Method A of ISO/TS 12913-2 [18] (radar plot) – Polo I Campus of UC (MP: Measurement Point).
In the text

	Figure 19. Graphical representation of the 2D soundscape model according to Method A of ISO/TS 12913-2 [18] (scatter plot, individual and median scores) – Polo I Campus of UC (MP: Measurement Point).
In the text

	Figure 20. Assessment and appropriateness of the surrounding sound environment (Method A of ISO/TS 12913-2 [18]): mean values and range – Polo I Campus of UC (MP: Measurement Point).
In the text

	Figure 21. Assessment of the sound environment (Method B of ISO/TS 12913-2 [18]): mean scores and range – Polo I Campus of UC (MP: Measurement Point).
In the text

	Figure 22. Sound source recognition (Method B of ISO/TS 12913-2 [18]): word clouds – Polo I Campus of UC.
In the text

	Figure 23. Subsequent comments (Method B of ISO/TS 12913-2 [18]): word clouds – Polo I Campus of UC.
In the text

	Figure 24. Frequency spectra of the six measurement points at Polo I Campus of UC, represented in dB(A) and dB, respectively on the left and on the right.
In the text