*Article* **Listening to Japanese Gardens: An Autoethnographic Study on the Soundscape Action Design Tool**

#### **Gunnar Cerwén 1,2**


Received: 19 June 2019; Accepted: 3 November 2019; Published: 22 November 2019

**Abstract:** Landscape architecture and urban design disciplines could benefit from soundscape thinking in order to enhance experiential qualities in their projects, though the available tools are not yet fully developed nor tested. The present research aims to substantiate one of the available tools, Soundscape Actions, and thereby increase the understanding of soundscape design. The study focuses on the Japanese garden tradition, which is known for high preference ratings, tranquil qualities and consideration for sound and other sensory experiences. An autoethnographic approach was used to conduct field studies in 88 gardens in Japan, the majority of which are located in urban areas with potential noise disturbance. The studies are based on observations in situ, supported by video documentation, field recordings and readings of sound pressure levels (SPL). A total of 19 Soundscape Actions are described and discussed in the paper. They are structured around three main categories: localisation of functions, reduction of unwanted sounds and introduction of wanted sounds. The study provides concrete examples of how the tool can be used to enhance tranquil qualities, particularly focusing on small green spaces in dense urban settings, involving the (simultaneous) reduction of unwanted sounds and enhancement of wanted sounds/effects. The autoethnographic approach allowed for the phenomenological perspective to be brought forward, which contributed new insights regarding the design tool. The findings are discussed in relation to health and soundscape research, focusing on multisensory experiences, masking strategies and potentials for implementation and future developments of the design tool.

**Keywords:** soundscape design; sonic experience; tranquillity; noise; garden therapy; landscape architecture; Japanese gardens; autoethnography; soundscape actions

#### **1. Introduction**

Japanese gardens are sources of inspiration for gardeners, landscape architects and designers around the world [1]. Originally inspired by gardens in China, the Japanese garden tradition has a long history with many particular styles, including the dry landscaped garden *karesansui*, the stroll garden *kaiyu-shiki-teien ¯* and the tea garden *cha-niwa* [2,3]. It is a diverse tradition held together by its Japanese sense of aesthetics, characterised by asymmetry, symbolism, geomancy, careful detailing and the use of natural materials (c.f. [4,5]).

Japanese gardens tend to receive high preference ratings compared to other types of gardens and landscapes [6–9], which can partly be explained by their informal character and natural expressions [10]. Rocks, water and vegetation constitute some of the most typical materials, meticulously shaped to represent nature "at its best". For instance, the shape of pruned trees in Japanese gardens have been used as motivation for the "Savannah theory" [11]. Studies have shown that spending time in a Japanese garden can lead to a reduction in heart rate [12] and an improved mood [13]. Some of the

most famous gardens are Zen Buddhist gardens, rich with symbolic meaning and spiritual qualities, and sometimes specifically designed to stimulate meditation [3]. Other gardens encourage physical activity and engage the bodily senses [4,14].

The green and natural landscapes of Japanese gardens have constituted an important part of the urban fabric for several hundred years [15]. In contemporary Japan, the gardens still offer a contrast to the restraints of modern life and the highly densified cities that often surround them. Space is a scarce resource in Japan, and gardens are often designed to appear larger than they actually are [16], which makes them interesting as reference objects for pocket parks and other small green spaces.

Today, 55% of the world's population lives in urban areas, with an expected increase to almost 70% in 2050 [17]. It is known that urbanisation and urban lifestyles can lead to stress and negative health effects [18]. Problems relating to stress constitute a major challenge for the global community. While a certain amount of stress is a natural part of human life, extended periods of exposure may lead to negative health effects, such as sleeping problems, depression, cardiovascular disease and chronic fatigue [19].

Research has shown that nature and nature-like environments can alleviate [20] and prevent [21] such negative health effects. Accessibility to green spaces is thus a crucial factor to consider in city planning, albeit one that may be difficult to adhere to, especially as densification proceeds. Gardens and other small green spaces with natural components may prove to be increasingly important in the future, as their spatial requirements are more flexible than parks. Yet, problems with noise from neighbouring urban activities constitute a challenge. This is the case in Japan, as well as in many other countries around the world.

The sound environment influences the quality of gardens and other green spaces [22,23] and has repercussions on health, both positive and negative [24,25]. It is pertinent to ensure that planning and design is optimised to take this into account. However, it has been repeatedly argued that sound is a limited concern within architectural disciplines and that, if sound is considered at all, it is typically with reference to noise issues [26,27].

Soundscape research, on the other hand, adopts a comprehensive understanding of the sound environment, including problems as well as positive experiences. It is a broad and interdisciplinary field, focusing on the contextual and subjective experience of sound environments. Initiated in the late 1960s [28–30], it has gained increased momentum in recent years [31], not least in urban planning and design projects [31–33]. In 2014, soundscape was defined by the International Organization for Standardization (ISO) as an "acoustic environment as perceived or experienced and/or understood by a person or people, in context" [34]. To date, a number of tools and approaches have been developed to aid the design of outdoor environments through use of the soundscape approach [35–39], yet few of these have been tested and validated in situ.

In the present study, one of the available tools called Soundscape Actions [27,36] was applied to study sonic experiences in the Japanese garden tradition. The aim was to substantiate Soundscape Actions as a design tool in landscape architecture and to increase the understanding of the design of tranquil soundscapes. Given their health promoting potential [13] and careful consideration of sonic experiences [4,40–44], Japanese gardens were considered an ideal context to appraise the tool, the idea being that the rich knowledge accumulated in a centuries-old tradition could potentially be used to inform future developments of the tool.

#### **2. Materials and Methods**

The present study is based on empirical material that was collected on site in Japan using an autoethnographic approach [45]. The material was collected during 136 visits to 88 renowned Japanese gardens, the majority of which are located in Kyoto (Figure 1). For further details about the selection of gardens, times of visit and other details, see Appendix A and Section 2.2.

#### *Int. J. Environ. Res. Public Health* **2019**, *16*, 4648

**Figure 1.** This map illustrates the geographical distribution of the 88 studied Japanese gardens.

Most of the studied gardens have existed for several hundred years, during which time the gardens and their soundscapes may have undergone change to various extents (not least the soundscapes surrounding the gardens). Alas, it should be pointed out that the study does not aim to contribute with historical insights as to how gardens were designed in different periods or regions in Japan. Rather, the tradition is considered as a context from which to draw general understandings regarding soundscape design which can then be applied to other (contemporary) gardens and green areas. It may well be that some of the effects that were experienced in the gardens were not the result of intentional acts by their designer. However, this does not make it any less interesting for the purposes of the study.

Field notes from the gardens were collected in a digital document amounting to a total of about 23,000 words, which served as the main research material. The design tool Soundscape Actions was subsequently applied to analyse and structure the material. The findings are supported by extracts from the field notes as well as photographs, videos, field recordings and sound pressure level (SPL) readings taken in the gardens. For an overview of video and sound material referred to in this paper, see Supplementary Material—List of video files. For further information on data collection, see Section 2.2.

#### *2.1. Using Autoethnography to Substantiate Soundscape Actions*

Autoethnography is a qualitative method focusing on the researcher's own experience and reflection of a phenomenon [45]. It has been increasingly applied in architectural disciplines, where it can be used to uncover various aspects of the cultural practice of designing and experiencing landscapes [46–49]. In this study, the autoethnographer has a background as a soundscape researcher and a landscape architect. It is in the intersection of these two fields that sonic experiences in Japanese gardens are notated.

Soundscape research has tended to focus on participants' generalised experiences (e.g., [50–52]), while relatively little attention has been paid to how individual subjects experience a phenomenon in detail. The subjective perspective is emphasised in the ISO definition of soundscape [34] and could potentially lead to a better understanding of how contextual factors influence the experience of sound. Subjectivity could also be an important factor when connecting soundscape research to design disciplines; it has been pointed out that architects often use their own subjective site experiences as a starting point when proposing new design solutions [46]. Autoethnography can be used to bring such experiences to the foreground, allowing for comparisons between practitioners [45]. If experiences are shared as autoethnographic narratives, these may also be studied by other researchers and designers as components of their reference libraries. In the present study autoethnography was used to evaluate a design tool, with the intention of substantiating its usage and making it more accessible to practitioners (c.f. [53]).

The approach is akin to the notion of "skilled listeners" as proposed by Hedfors and Berg [54] which has previously been used to study the relationship between sonic phenomena and landscape architecture [55]. In that study, data from skilled listeners were used to provide intricate understandings regarding the experiential qualities of environmental sounds, which were then formulated as a terminology for practitioners. A related approach has also been used by Amphoux [56], involving trained listeners to map urban soundscapes.

The present study differs from the above by focusing on a single listener's experience, rather than several. As an autoethnographic study it is subjective in nature, and care should be taken when interpreting the results. To ensure validity, efforts have been taken to make subjectivity explicit where it is present [57]. In order to increase generalisability, sonic experiences are discussed in relation to previous research and supported by data collected on site (Section 2.2). In accordance with what has been termed "analytical autoethnography" [58], the findings are used to gain insights on a theoretical framework (Soundscape Actions).

#### *2.2. Data Collection*

Most of the research was undertaken during 2018, preceded by some initial surveys in 2015. Following the autoethnographic approach, criteria for site selection were guided by a general intention to learn from the Japanese garden tradition in terms of how soundscapes could be designed (see Sections 2.2.1 and 2.2.2). In total, 88 gardens in multiple locations around Japan were visited (see Figure 1 and Appendix A). The majority of these (n = 54) were located in Kyoto, which is known for its many gardens of high quality. During the spring and autumn of 2018, a total of three months were devoted to field studies in Kyoto. Additionally, a number of gardens in Tokyo (n = 11) and Kanazawa (n = 10) were included in the study, as well as chosen gardens in other locations. The data collection process can roughly be divided into five phases as discussed below. However, it should be noted that the chronology sometimes overlapped. For instance, when a new city was visited, Phases II and III were repeated.

#### 2.2.1. Phase I: Surveying the Field

In the first phase, a survey of previous literature pertaining to Japanese gardens (e.g., [2–5,14,16]) and their soundscapes (e.g., [40–44]) was undertaken. Contact was also established with key researchers in Japan, leading to some valuable recommendations for gardens to study. Based on the literature review and contact with researchers, prospective gardens were continuously plotted using Google Maps.

#### 2.2.2. Phase II: Initial Study Visits in Gardens

This phase was intended to provide an overview of a wide variety of traditional gardens in Japan. Visits in this phase were relatively short (typically 15–45 min), the intention being to identify gardens that would be able to contribute with insights on soundscape design. Each garden was briefly described in a notebook, including the general impression, the context for the visit, notable features, the date, weather and soundscapes. The notes were brief, though could be elaborated if there were particular sonic features. The notes were taken either directly on site or written down later in the evening of the same day. Each garden was photographed, and in case there were notable sonic features, these were recorded.

#### 2.2.3. Phase III: Extended Visits to Gardens

Gardens that had been noted for further investigation in Phase II were revisited once or several times, with the intention to study selected sonic events more thoroughly. Visits in Phase III were longer, typically lasting more than one hour. The extended time allowed for reflective on-site writing and the capturing of video, SPL readings and field recordings (for technical details, see Table 1). In a few gardens that were difficult to access, Phases II and III were combined.


**Table 1.** Equipment and processes used to notate sonic experiences in the gardens.

<sup>1</sup> The setting had been tested prior to the study in an urban environment sheltered from wind (SPL varying between 50–65 dBA). A Norsonic 140 (Class 1) SPL meter was used as control, and the manual setting in NIOSH SLM was calibrated to −2.1 dBA. After the study, the setting was tested again, this time by playing back different types of noises indoors and using a Brüel & Kjaer 2270 (Class 1) SPL meter as control. This test indicated that the setting had been satisfactory for the situations observed in the study (35–65 dBA), where the discrepancy did not exceed ±1.5 dBA. With this being said, it should be noted that SPL readings taken with smartphone applications do not comply with international standards. Studies on smartphone applications for SPL measurements have shown that, even though some of them may be suitable to measure occupational noise [59], the performance varies and limitations in accuracy can be problematic in the wrong context [60]. The SPL readings should be interpreted with this in mind.

#### 2.2.4. Phase IV: Summary of Research Material

Garden visits were continuously noted in a Microsoft Excel spreadsheet (Appendix A). Field notes were collected in a digital document, amounting to a total of 23,000 words. This document served as the main research material. Images, video and sound were organised as a digital library that was used as a reference during the process of analysis. Extracts from this library were used to support the findings in this paper.

#### 2.2.5. Phase V: Control

This phase was carried out in conjunction with data analysis, and entailed revisits to some of the previously studied gardens. The intention for these visits was either to confirm findings, make further comparisons between gardens and/or to collect audio-visual material and SPL readings. Most of the work in Phase V was completed in the autumn of 2018.

#### *2.3. Analysis of Data*

The Soundscape Action tool was used as a framework to analyse the research material and present the results. The tool was originally developed in collaboration with practitioners aiming to improve urban areas exposed to noise [27,36,61]. The tool has been described as "a group of acts that can be taken with the intention of designing a soundscape" ([36], p. 509). There are currently 23 Soundscape Actions divided into three main categories: localisation of functions, reduction of unwanted sounds and introduction of wanted sounds (Figure 2).

**Figure 2.** The design tool Soundscape Actions was used as the basis for analysis. In total there are 23 Soundscape Actions divided into three main categories: localisation of functions, reduction of unwanted sounds and introduction of wanted sounds.

The field studies were conducted in 88 gardens around Japan, but the analysis and presentation focuses on gardens located in Kyoto (n = 54). The focus on Kyoto is partly a result of the extensive material collected in this area, and it was also deemed necessary for reasons of clarity. The focus on Kyoto is most evident in Category I (localisation of functions), where other gardens are excluded entirely, so as to allow for comparisons within the city.

Before analysis was initiated, all field notes had been studied and checked for connections to supportive data. In cases where additional data or controls were required, this was noted and carried out (Section 2.2.5). In the analysis, each of the Soundscape Actions was compared with the field notes in order to identify relevant sections in the material; this process was guided by the author's recollection of the visits as well as strategic searches within the notes. Identified passages were copied to a new Microsoft Word document which was used as a base from which to summarise the findings. Supportive data was compared with the Soundscape Actions and, where applicable, links were provided in the new document.

It was found that four of the 23 Soundscape Actions were not applicable to the given context—embrace unwanted sounds, abolish certain functions, low noise screens and atmospheric design (loudspeaker-based)—and have thus been omitted from Section 3, leaving a total of 19 Soundscape Actions. These four are further described and discussed in Section 4.

#### **3. Soundscape Actions in Japanese Gardens: Descriptions and Findings**

The results are structured according to the three main categories illustrated in Figure 2. Each of these three categories is first given a brief *Category introduction* including any particular *Prerequisites and delimitations* of importance to the study. The 19 subcategories (Soundscape Actions) are then introduced using the following three headings: *Description*—A brief definition of each Soundscape Action based on previous studies; *Gardens of particular interest*—A list of gardens that were important for the findings presented and; *Findings*—A description of how the Soundscape Action related to the studied Japanese gardens and the findings that were made. Where applicable, connections to previous research are indicated.

#### *3.1. Localisation of Functions*

*Category introduction*: The first main category in the Soundscape Action tool regards the localisation of functions in relation to their surroundings (in the present study: gardens in relation to their urban surroundings). This category typically entails the consideration of noise and the compatibility between functions.

*Prerequisites and delimitations*: The Soundscape Actions in this category are mainly described based on observations and comparisons between gardens, complemented with the mapping of garden locations, images and SPL readings. For the purposes of clarity, all gardens mentioned in this main category are limited to one city, i.e., Kyoto (Figure 3). While acknowledging the considerable age of many gardens and the complex development of the city over time, the study focuses on how the gardens and their contexts are conceived today.

**Figure 3.** This map illustrates the locations of the 54 gardens that were studied in and around Kyoto. See Appendix A for names and further details of the study visits.

#### 3.1.1. Compensation/Variation

*Description:* Compensation/Variation acknowledges the differences between neighbouring soundscapes as a potential quality. Gardens are frequently referred to as "safe havens" and "oases", thus implying a relative tranquillity and quietness where it is possible to escape from the hustle of the outside (urban) world. Consequently, busy urban settings surrounding a garden may be regarded as an asset, by offering contrast and juxtaposition (provided that the tranquillity inside the garden can be maintained).

#### *Gardens of particular interest:* Ohashi-ke (35), Konchi-in (27), Murin-an (31). ¯

*Findings:* The notion of Compensation/Variation was particularly pronounced in lesser known gardens located close to busy tourist routes. For instance, Ohashi-ke garden in southeast Kyoto is located only ¯ two to three hundred metres from the immensely popular Shinto shrine Fushimi-Inari Taisha. In this intense tourist area, the narrow roads are filled with visitors en route to or returning from the shrine, lingering among market stalls and food shops (Figure 4a). In the study, the overwhelming intensity around Fushimi-Inari Taisha was found to influence the experience inside Ohashi-ke by making ¯ the garden seem calm and quiet in comparison. Hence, the experience of the garden's soundscape depended not only on sounds experienced inside the garden, but also on previous encounters in the surrounding area. This polarity was captured in a comparative recording between the two sites (Video S1, https://vimeo.com/350108144).

**Figure 4.** Intense soundscapes may work as a contrast to enhance tranquillity in nearby gardens. (**a**) A busy tourist street at Fushimi Inari-Taisha in Kyoto. (**b**) The entrance to Konchi-in temple leads the visitor across a loud water stream.

A similar effect of contrast can be experienced about 5 km north, in Konchi-in garden, where the effect is articulated through the sound of water (in addition to the proximity of a tourist route). In order to reach the garden, the visitor needs to cross a loud water stream (Figure 4b). On the bridge across the water, the SPL temporarily reaches over 60 decibels (62 dBA; 21 April 2018), but drops rapidly as the main gate is approached on the other side. Inside the garden itself, the SPL is about 20 decibels lower (38–43 dBA, as observed in different parts of the garden; 21 April 2018). This example illustrates how the addition of sound could be intentionally used to temporarily raise the SPL in order to produce a relative tranquillity at the offset. The effect could arguably be further enhanced if it were combined with screens to increase the contrast. Such a combination can be experienced in another garden in the vicinity, Murin-an, though only occasionally, as the channel that produces the sound is dependent on rainfall to reach sufficient water flow.

#### 3.1.2. Embrace Wanted Sounds

*Description:* To embrace wanted sounds is to make use of pre-existing sounds as a quality when selecting the locations of new functions. For instance, by selecting a location next to a forest, this may add the sound of twittering birds and rustling vegetation, among other experiences.

*Gardens of particular interest:* Nanzen-in (32), Sanbo-in (41), Ents ¯ u-ji (10), Shisen-d ¯ o (45), Sanzen-in (2), ¯ Ruriko-in (37).

*Findings:* Kyoto is surrounded by wooded mountain ranges on three sides: Higashiyama to the east, Kitayama to the north and Nishiyama to the west. A substantial amount of the most well-known gardens in Kyoto are located on the fringes of the city (c.f. Figure 3), often bordering the foothills of the mountain ranges directly. Not only is this a scenic setting which offers good views, but the proximity to nature also creates good prerequisites for a rich soundscape, inviting the sound of rustling leaves, purling water streams and the twittering of birds to be "borrowed" from the surrounding mountains:

*Nanzen-in is an exquisitely designed small pond garden, parts of which date back to the 13th century. Lush mountainous woodlands surround the garden and add to the atmosphere through fragrance and sound. In the southeastern corner of the garden, a waterfall supplies fresh mountain water to the ponds.*

Nanzen-in garden; 14 May 2018 (Figure 5)

**Figure 5.** Nanzen-in, a sub-temple to the Nanzen-ji complex, has a small pond garden that borders mountainous woodlands attractive to birds and other animals.

In the Japanese garden discourse, *shakkei* is an important concept used to describe a technique whereby scenic features from the surrounding landscapes are "borrowed" and incorporated as part of the garden design. A typical example is when a distant mountain is framed to become part of the garden, articulated by carefully pruned vegetation. The notion of *shakkei* has mostly been discussed as a visual concept [62], however it is also fruitful to consider it as referring to sound and the other senses (c.f. [42]). This is exemplified in Nanzen-in above, as well as in other gardens like Sanbo-in (Video S2, ¯ https://vimeo.com/312665806), Entsu-ji, Shisen-d ¯ o, Sanzen-in and Ruriko-in. ¯

#### 3.1.3. Avoid Unwanted Sounds

*Description:* Avoiding unwanted sound is about considering existing sound sources that could potentially be disturbing and to make sure to locate new functions in areas shielded from such sources by providing sufficient distance and/or using shelter from buildings, mounds and other existing features.

*Gardens of particular interest:* Enko-ji (9), Rurikoin (37), Shisen-d ¯ o (45), Konpuku-ji (28), Giou-ji (14), ¯ Saihoji (40), Shosei-en (48). ¯

*Findings:* Studying several gardens in the same city made it possible to make comparisons between different locations. While the majority of studied gardens were located on the fringes of the city, others had more urban locations, and thus provided an important reference. Almost all gardens located on the fringes of the city were perceived as quiet. Enko-ji is an example of one such garden, where ¯ the background ambience was found to be around 40 dBA (12 May 2018). In that example, most of the ambience consisted of subtle natural sounds from the garden and the surrounding forest. It was striking how the low SPL made it possible to hear "further away":

*Int. J. Environ. Res. Public Health* **2019**, *16*, 4648

*I measure the background ambience at the veranda to be around 40 dBA. The ambience consists of a stream, distant birds and the occasional wind in the trees. In a way, it is so quiet that it seems I can hear further away.*

Enko-ji; 12 May 2018 (Figure ¯ 6)

**Figure 6.** Autumn in Enko-ji temple. The sliding doors to the main hall have been opened, thus ¯ allowing the sounds and sights from the garden to enter into the building.

Needless to say, gardens in central locations tended to be more exposed to noise, inevitably leading to reduced dynamics and a shorter "acoustic horizon" [30]. Gardens in urban locations could still be experienced as surprisingly tranquil, partly owing to the extensive use of water streams with masking capabilities (see further in Section 3.3.1. Auditory Masking). Other important factors for reducing noise include distance to major roads, use of vegetation and screens (garden walls and other structures).

The most problematic of the gardens studied in Kyoto was Shosei-en, located close to the central ¯ station. Shosei-en is a large garden, and the disturbances varied quite extensively inside the garden ¯ (45–60 dBA, as observed on different locations in the central parts of the garden; 29 April 2018). Such local variations constitute important prerequisites when functions within a garden are planned, as some features are more sensitive to noise than others.

#### *3.2. Reduction of Unwanted Sounds*

*Category introduction:* Soundscape Actions in this main category concern the reduction of unwanted sounds, most typically sounds from urban settlements and related activities like traffic. In contrast to the previous category, these Soundscape Actions are performed after the relative locations of functions have been decided upon.

*Prerequisites and delimitations:* Descriptions in this category are mainly based on observations. It was not deemed fruitful to use SPL to assess noise reduction in most cases, the reason being that this would have required comparisons that could not be obtained due to practical reasons (such as observing SPL with and without a garden wall). Instead, the studies focused on how various features in the gardens were designed, and discussed this in relation to previous research. The focus remains on gardens in Kyoto, though other examples are included as well.

#### 3.2.1. Vegetation for Noise Reduction

*Description:* Vegetation is commonly used to reduce the impact of noise; one of the most typical applications is vegetation belts along roads. However, the actual effect of vegetation for reducing SPL is debated [63], most likely because the effect varies extensively depending on a number of factors such as vegetation type/s, distribution pattern, and the absorbing qualities of the ground cover [63–65]. Psychological factors are also important, in particular source visibility [66] (see further in Section 3.3.2. Visual Masking). Moreover, the sound of vegetation can have a positive masking effect that shifts focus from noise (see also Section 3.3.1. Auditory Masking and Section 3.3.4. Sounds of Vegetation).

*Gardens of particular interest:* Saihoji (40), Murin-an (31), Shin'en (44), Chishaku-in (5).

*Findings:* Woodlands constitute a recurring feature in Japanese gardens, often found together with moss as in the famous Saihoji: the moss temple (Figure 7). Woodlands' effect on noise reduction is multifaceted. The trees themselves have some effect, depending on species composition, stem thickness, stand density and other factors [63–65], though there are several indirect factors that need to be taken into consideration, such as interference from wind. Wind is known to carry noise, hence woodlands can limit noise propagation in some situations [64]. For gardens, the wind reduction is particularly important along garden walls where there is otherwise a risk of micro-metrological turbulence that can have extensive negative effects.

**Figure 7.** Moss and a garden wall in the woodland of Saihoji in Kyoto.

Almost all Japanese gardens in the study had garden walls, and these were often combined with vegetation. The combined effect of vegetation and wall is beneficial, not only for its enhanced noise reduction [64], but also as a means of articulating the garden space and delimiting it from the outside world (Video S3, https://vimeo.com/311434655, see further in Section 3.3.2. Visual Masking). Moreover, wind causes trees to rustle and this sound (together with birds and sounds from other animals attracted to the forest) can have the effect of masking noise (see further in Section 3.3.1. Auditory Masking and Section 3.3.7. Biotope Design).

Another aspect of woodlands is the effect that trees have on the ground cover. For instance, moss is one of the most characteristic features in Japanese gardens and it seems to thrive particularly well in moist woodlands, where it is shaded by the tree canopy (see further in Section 3.2.7. Absorbing Qualities of Materials). Moss is interesting for present purposes as it has a good ability to absorb ambient noise, an ability it shares with organic soil, grass and other soft ground covers (c.f. [67]).

#### 3.2.2. High noise Screens

*Description:* A screen or wall with the approximate height of 1.8 metres or above is a high noise screen. Like all screens, they should be located as close to the source as possible [68]. High noise screens make it difficult for most people to see to the other side of the screen (unless the screen is transparent).

*Gardens of particular interest:* Murin-an (31), Shin'en (44), Saihoji (40).

*Findings:* Garden walls constitute a characteristic feature in the Japanese garden. Traditionally, the walls are made from clay, resting on a foundation of rocks and with a tile roofing (Figure 7), yet there are also examples of constructions based on other materials such as wood and stone. The height of garden walls vary; most are tall enough to offer visual seclusion, and some are three metres or higher, thus suggesting that screening of noise might have been an important consideration in addition to

visual/spatial seclusion. Walls can be used to define the edges of a garden as well as having an effect on noise reduction from the outside. The actual noise reducing effect varies depending on several factors, and the reduction is most effective when the screen is located close to the source (or alternatively close to the receiver) [68]. As previously discussed, vegetation can improve the effect [64].

#### 3.2.3. Buildings as Screens

*Description:* This Soundscape Action acknowledges the potential of strategically located buildings to reduce unwanted sound. Buildings can be effective noise screens, especially if they are tall and made of solid material [38]. They may also be combined with traditional noise screens.

#### *Gardens of particular interest:* Enko-ji (9), Shisen-d ¯ o (45), Rurikoin (37). ¯

*Findings*: Most buildings in and around the studied gardens were built in traditional style. One characteristic feature of traditional Japanese buildings is that they are made of light materials. In fact, some parts are even made of paper, including the *Shoji ¯* , sliding doors. Moreover, the buildings are regularly opened up to the garden (c.f. Figure 6). In this respect, traditional buildings are not ideal when it comes to noise reduction, and the lower frequencies particularly are able to pass through buildings easily. However, these limitations with traditional buildings were seldom experienced as problematic in the study. One reason could be that the buildings tended to be located inside the boundaries of the garden walls and/or that they were located with sufficient distance from the noise.

There is an opposite and positive effect associated with the traditional buildings' ability to carry sound. The light construction can be said to "invite" sounds from the garden such that these may be experienced from inside the building, making it possible to listen to water, birds, the chanting of monks, the occasional bell, the murmur of other visitors or any other sound in the vicinity. Japanese buildings are known for their ability to blur the borders between inside and outside [14], and sound plays an important part in this.

#### 3.2.4. Change Topography

*Description:* Land masses can be shaped to form strategic topographical patterns like mounds and/or valleys, which can be used to screen noise. The effect varies depending on factors such as height, width, detailed shape as well as planted vegetation (if any) [64].

*Gardens of particular interest:* Katsura Rikyu (21), K ¯ orakuen (76), Koishikawa-K ¯ orakuen (59) Suizen-ji ¯ Joju-en (77), Rikugi-en (64), Adachi Museum garden (79), Goten (Ninna-ji) (16).

*Findings:* Topographical features constitute a recurring design feature in the Japanese garden. In Katsura Rikyu in Kyoto, for instance, a pond was dug out and the soil was used to create an artificial mound, ¯ *tsukiyama* [4], where a tea house is located. Topographical features offer good views over the landscape, but they can also be visually striking in their own right, such as the characteristically shaped hills in Korakuen in Okayama and Koishikawa-K ¯ orakuen in Tokyo. Mounds are often designed to represent ¯ famous and/or sacred mountains, in which case they have a symbolic value [2,3]. The almost perfectly shaped grass cone in Suizen-ji Joju-en, Kumamoto, made to look like Mount Fuji, is a good example.

Topographical features constitute a useful means to reduce noise, if strategically located [64]. However, neither of the features mentioned above seem to have been built with the deliberate purpose of screening noise from the outside, as they are typically located in the central parts of the gardens where the effect should be limited, given the distance to the source [68].

On the other hand, it has been suggested that topographical features in some Japanese gardens are used with the intention of reducing the sonic impact of waterfalls inside the gardens [4]. The idea is that by "muffling" the sound of a waterfall in this way, it may seem to be further away than it actually is, hence making the garden appear larger (c.f. *shakkei* in Section 3.1.2). This effect was encountered in three gardens: Rikugi-en in Tokyo (Video S4, https://vimeo.com/312519101), Adachi Museum garden outside Matsue and Goten (Ninna-ji) in Kyoto.

#### 3.2.5. Reduce Source Activity

*Description:* This Soundscape Action is concerned with everyday activities in the environment. It focuses on how unwanted sound from such activities can be reduced by decisions made in planning and design, such as enforcement of speed limits on roads or restrictions on social behaviour.

#### *Gardens of particular interest:* Tenryu-ji (52), Daisen-in (7), Shisen-d ¯ o (45), Goten (16). ¯

*Findings:* Many Japanese gardens are popular attractions, and it is common to experience sounds from other visitors. These sounds can be quite intense, particularly in famous gardens with high accessibility, like the UNESCO World Heritage Site Tenryu-ji, which attracts many tourists. Taking photographs is a ¯ popular activity inside the gardens, and one which generates a lot of sound including discussions, shutter sounds and so on. Some gardens, like Daisen-in, do not allow photography, and this seems to have a positive effect on the soundscape as well as corresponding better with the "mindful" spirit of zen in that temple. In some gardens, like Shisen-do and Goten, there are signs encouraging people to ¯ be respectful in their visits (Figure 8).

**Figure 8.** A sign in Goten, Kyoto, encouraging people to be quiet during their visit: "From here. Please be quiet and enjoy your tour.".

#### 3.2.6. Maintenance

*Description:* Maintenance can generate a lot of noise, particularly when combustion engines are used. On the other hand, the use of hand-driven equipment and/or animals can potentially add a quality to the soundscape. Maintenance can be controlled in maintenance plans, but it is also influenced by design solutions. For instance, a lawn requires constant mowing, while a free-growing meadow is less intensive.

#### *Gardens of particular interest:* Ginkaku-ji (13), Rurikoin (37), Konchi-in (27), Shugaku-in Riky ¯ u (49). ¯

*Findings:* Generally speaking, Japanese gardens are intensive when it comes to maintenance. Fallen leaves and other litter are removed meticulously, as are weeds. Moreover, the characteristic shape of trees and shrubs requires extensive pruning to be maintained and the gravelled areas of the dry gardens, *karesansui*, are raked every morning. On the other hand, the widespread use of moss rather than grass reduces the need for lawn mowing. Despite the intensive care required, as well as the fact that a lot of maintenance is carried out during opening hours, the work was rarely perceived as disturbing during the study. To a great extent, this might be ascribed to the fact that most tasks are performed by hand. The soft clipping of secateurs or the sweeping of a broom are sounds that add a layer to the experience, rather than disturb it (Figure 9a,b and Video S2, https://vimeo.com/312665806).

**Figure 9.** (**a**) A garden worker sweeping the moss in Rurikoin, Kyoto. (**b**) A garden worker pruning a pine tree in Ginkaku-ji, Kyoto.

Modern equipment is unusual, but was noted on a few occasions when it became quite disturbing. For example, in a garden normally perceived as a quiet gem, Konchi-in, a leaf blower was encountered on one visit, raising the SPL from 41 dBA to 65 dBA (15–20 m from the machine, 14 May 2018). Leaf blowers were also noted on two visits to Shugaku-in Riky ¯ u, though the garden workers there ¯ waited until the visitors had passed before using the machines (visits to Shugaku-in Riky ¯ u are organised ¯ in groups with a guide).

#### 3.2.7. Absorbing Qualities of Materials

*Description:* This Soundscape Action regards the strategic application of materials with absorbing qualities, such as vegetation soil and moss, to reduce noise impact.

*Gardens of particular interest:* Saihoji (40), Rurikoin (37), Enko-ji (9), Sanzen-in (2). ¯

*Findings:* The prerequisites for moss are good in Japan, as the proximity to the ocean creates a humid climate which helps the moss to thrive. The timeless expression and subtle atmospheric appeal of moss corresponds with the Japanese aesthetics known as *wabi-sabi* [69] and many gardens cultivate moss intentionally. Moss can have a positive effect on the sound environment, as it grows to create a soft and absorbing surface on the ground (c.f. [64,67]). The famous moss temple Saihoji, for instance (Figure 7), was found to be one of the most quiet gardens in the field studies. The ambience at the entrance to the pond section of the garden was noted at 35 dBA (18 November 2018). As speculated during an earlier visit to the garden, moss could be one of several explanations for the quietness:

*I attribute the silence partly to the fact that the garden is situated in the outskirts of Kyoto, making it less likely to be a*ff*ected by city noise. However, the silence can also in part be attributed to the extensive coverage of moss, which is known to absorb sound. Moreover, the garden is surrounded by walls on the one side, and hills on the other, to o*ff*er a shelter from the surroundings.*

Saihoji; 15 May 2018 (Figure 7)

#### *3.3. Introduction of Wanted Sounds*

*Category introduction:* Soundscape Actions in this category can be used to introduce, stimulate or enhance sounds that are considered wanted in a particular situation. Needless to say, wanted sounds vary depending on context and individual preferences along with a number of other cues [51,52,70,71], yet sounds from nature, like water, birds and rustling vegetation, are typically considered to be wanted sources [50].

*Prerequisites and delimitations:* Descriptions of Soundscape Actions in this category are based on subjective observations supported by field recordings, images and SPL readings. The focus remains on gardens in Kyoto, but other examples are included as well.

#### 3.3.1. Auditory Masking

*Description:* Auditory masking is an effect where one sound (masker) is used to reduce the impact of another sound (target). Auditory masking has been used in a number of urban design projects, typically involving water features to reduce the impact of traffic noise [72]. Essentially, there are two types of auditory masking: energetic masking and informational masking [73]. In energetic masking the masker sound literally covers the target sound, rendering it inaudible. It has been found that, in the case of urban water features, an increase of about 8–10 dBA is required to achieve this effect [74]. In informational masking on the other hand, both sounds are audible, yet the focus shifts from the target to the masker.

*Gardens of particular interest:* Eikan-do (8), Murin-an (31), Sh ¯ osei-en (48), Koishikawa-K ¯ orakuen (59), ¯ Nezu museum garden (63), Kenroku-en (68).

*Findings:* The ample use of water features in Japanese gardens constitutes an ideal setting to study auditory masking. The notion of energetic masking was conceived to be relatively straightforward. It would typically be experienced in the proximity of loud water features, where all surrounding sounds were effectively covered. Informational masking, on the other hand, was found to be much more complex, involving several cues including the physical characteristics of the inherent sound sources and their relative locations in space. The studies also indicate that informational masking could be affected by visual information (c.f. Section 3.3.2. Visual Masking), as well as the notion of gestalt psychology (c.f. [30,75]), as the following experience from Eikan-do illustrates: ¯

*A loud stream and two waterfalls border the garden on its southern side. In a way, the powerful sound seems to be reinforcing the edge of the garden, which is also marked by a fence. Behind the fence, there is a high school and a kindergarten. This creates an interesting and potentially disturbing situation, as the sound of the children playing can collide with the activities in the temple, activities that presumably are best performed in a tranquil and quiet environment. The sound of children playing is loud, but it is partly masked by the powerful sound of the water. It can still be heard through the water, but in a way, it is as though it becomes located on a di*ff*erent spatial layer, or to use gestalt psychology, in the background. Without the water, the sound of children would probably have been much more intrusive.*

Eikan-do; 18 April 2018 (Figure ¯ 10a)

Sound level is an important factor to consider in relation to masking strategies. It has been argued that if the SPL reaches above 65–70 dBA, all sounds start to become disturbing, in which case masking strategies are not fruitful [39]. To achieve tranquil qualities, much lower levels than this are needed. It has been shown that good soundscape quality in suburban green areas and city parks typically require a level below 50 dBA [76] and that informational masking is more efficient at levels below 52.5 dBA (traffic noise) [77]. In Eikan-do, the SPL was 58 dBA (29 November 2018) at the point where ¯ the above description was noted, a few metres from the stream that borders the garden's edge.

**Figure 10.** Loud water features can be used to mask outside noise. (**a**) A waterfall in Eikan-do, Kyoto. ¯ (**b**) A stream in Murin-an garden, Kyoto.

#### 3.3.2. Visual Masking

*Description:* The idea behind visual masking is that if the source of a noise is hidden from view it is less likely to attract attention and cause disturbance. It has been argued that visual masking can be a fruitful strategy as long as the noise that is being hidden is not too prominent [66,78]. This strategy has also been referred to as source (in)visibility [66] and is related to the notion of acousmatics [79].

#### *Gardens of particular interest:* Murin-an (31), Shin'en (44), Chishaku-in (5), Eikan-do (8), Sh ¯ osei-en (48). ¯

*Findings:* As previously discussed, Japanese gardens make extensive use of garden walls combined with vegetation to mark the edges of the premises. The clearly defined visual spaces keep the visitor's attention within the garden, presumably making noise from the outside world less likely to be noticed (Video S3, https://vimeo.com/311434655). In a similar manner as auditory masking discussed above, visual masking seems to entail a kind of layering of experiences in terms of background and foreground. In fact, it was found that these two masking effects may work together, as the following example from Murin-an garden in Kyoto illustrates:

*It seems the garden wall manages to shield the sound from the street well; the combined e*ff*ect of screening and lack of visual contact makes me feel as though the cars belong to another world. Moreover, I think it helps that the sound of cars mixes nicely with the rustling of vegetation above the wall, as it is a windy day when I visit. In fact, the vegetation follows the exact same stretch as the road, but because only the vegetation is visible from where I sit, it is easy to partly ascribe the sound of the tra*ffi*c to the vegetation. Suddenly, a bus passes by, I can clearly see its white roof above the wall, and the illusion is broken for an instance. The sound seems disturbing now, partly I think because I can see the bus, and partly because the sound is louder; I can even feel the vibration from where I am sitting. As the bus passes away, it becomes quiet again.*

#### Main building, Murin-an; 6 May 2018 (Figure 10b)

The ambience in the example (without passing cars) was noted at around 46 dBA (6 May 2018). The rather high wall surrounding the main building shields noise effectively, and the SPL was only raised by a couple of decibels when cars were passing by, thus helping to keep the noise "not

prominent" [66]. As a reference, outside the wall on the sidewalk, the ambience was noted at around 54 dBA, which was increased by more than 10 dBA as a car passed by (14 May 2018).

#### 3.3.3. Sounds of Water

*Description:* Water has the potential to create a vast amount of sonic effects, ranging from the powerful broadband noise of a waterfall through to the subtle tickling of a single drop. Water can be made to vary in tone, rhythm and strength [80,81]. It is generally perceived as a pleasant sound, though can be annoying depending on the character [82]. Water features have repeatedly been used to produce masking effects from urban noise in strategic locations [32,72,74].

*Gardens of particular interest:* Tsujike teien (75), Kenroku-en (68), Murin-an (31), Eikan-do (8), Shin'en ¯ (44), Ginkaku-ji (13), Funda-in (12), Rurikoin (37), Chishaku-in (5), Tenryu-ji (52). ¯

*Findings:* The ample rainfall and mountainous geography of Japan supplies many gardens with direct access to natural water with a high flow rate. As a general trait, water features in Japanese gardens tend to echo expressions that can be found in nature (Figure 10a,b and Figure 11a,b). For instance, Japanese gardens make frequent use of ponds, natural streams and waterfalls, while formalistic expressions like fountains are less frequent.

**Figure 11.** Two strategies to articulate water sound in Japanese gardens. (**a**) Stones are strategically laid out to enhance the sound of a water stream in Shin'en garden (Heian Jingu Shrine), Kyoto (Video S5). (**b**) A barely audible miniature waterfall in Ginkaku-ji, Kyoto (Video S6b).

Streams and waterfalls with loud and roaring sounds can be found in Tsujike Teien and Kenroku-en in Kanazawa, as well as Murin-an and Eikan-do in Kyoto. Such loud water features are particularly ¯ useful to mask out urban noise through energetic masking (c.f. Section 3.3.1. Auditory Masking). In some gardens, stones are used to deliberately enhance the sound of streams. In Shin'en garden in Kyoto, a stream that connects the eastern and northern ponds is narrowed by the use of boulders, which increases the speed and directs the flow (Figure 11a and Video S5, https://vimeo.com/311170428). As the water descends towards the next level, rocks are strategically located to break the flow. This makes the stream full of life and produces a stronger and more interesting sound than would otherwise have been the case. This technique can be traced back to at least the 11th century in Japan, where it is mentioned in the classic novel *The Tale of Genji*: "The stream above the waterfall was cleared out and deepened to a considerable distance; and that the noise of the cascade might carry further, he set great boulders in mid-stream, against which the current crashed and broke" ([83], p. 428).

If some water features are created in a way that enhances their sound, it is also common to find opposite and more delicate approaches where extremely low sound levels are emphasised. In some cases, such kinds of subtle water features are in fact just above the audibility threshold. They are often found in zen gardens, which could suggest that they are used to support meditation practices (c.f. [3]). The expressions vary from the tranquil dripping of water in a small washing basin, *chozubachi,* as experienced in Funda-in in Kyoto, through to what could essentially be described as miniature waterfalls, found in gardens like Rurikoin, Chishaku-in (Video S6a, https://vimeo.com/311182675) and Ginkaku-ji (Figure 11b and Video S6b, https://vimeo.com/311079936) in Kyoto. Another expression of this kind of aesthetics can be found in the "dry waterfall", *karetaki*, (e.g., Tenryu-ji in Kyoto), ¯ where stones are arranged to look like a waterfall but intentionally kept dry, thus only suggesting the sound [43].

#### 3.3.4. Sounds of Vegetation

*Description:* Vegetation sound is perhaps most commonly associated with leaves that rustle in the wind. The character of the rustle varies depending on several factors, including species composition, spatial layout and wind [27]. Tree branches can produce crackling sounds if they are swaying in the wind, and large leaves can enhance the sound of rain as raindrops fall on them [84]. Some species are known as good rustlers, including the poplar genus (aspen) and bamboo [84,85].

*Gardens of particular interest:* Funda-in (12), Chishaku-in (5), Rikugi-en (64), Rentaroh Taki's Memorial garden (81).

*Findings:* The sound of rustling leaves could be experienced in most gardens, the character varying from a tranquil and subtle atmosphere to strong gusts. On windy days, when the effect was most pronounced, the leaves could take over and drown out other sounds (c.f. Section 3.3.1. Auditory Masking and Section 3.3.2. Visual Masking). Observations in this study indicate that the sound could be enhanced if the trees are surrounded by open land or planted as an avenue to allow the wind better access to the leaves, however no SPL readings could be taken to confirm this. In some gardens, like Funda-in, Chishaku-in, and Rikugi-en, it was apparent that trees planted on mounds had a strategic position to catch the wind. This was particularly evident in Funda-in and Chishaku-in, where the bamboo was planted on mounds:

*A few moments of tense silence before a gust of wind starts to build up in the distance. At first, I can hear how the tall bamboo trees in the left back side of the garden start to move. They are strategically located on a high position which exposes them to more wind. The wind gust gradually approaches the garden, and I can now hear the rustle of the vegetation on the top of the hill, where a variety of species adds to the composition. Eventually, I can feel the wind passing the veranda and hear how the shoji doors behind me shake slightly as a response, before it quiets down again.*

Chishaku-in; 21 May 2018 (Video S6a)

Bamboo has previously been acknowledged for its ability to invite sound into a garden, in China [84] as well as in Japan [44]. When describing the soundscape design in Rentaroh Taki's Memorial garden in Taketa, Torigoe mentions how bamboo makes "sound when blowing in the wind, rustling its leaves and squeaking its stems" [44] (p. 112).

#### 3.3.5. Walking Materials

*Description:* Materials like gravel and wood are known to generate ample sonic responses when walked upon, while other materials like asphalt and stone are quieter. Furthermore, different materials can contribute with different connotations which have bearing on preference [86]. The sound of other people walking may be experienced as a pleasant atmosphere, or in some cases, as a warning [85].

*Gardens of particular interest:* Rentaroh Taki's memorial garden (81), Nomura-ke samurai teien (70), Ryoan-ji (38), Shoren-in (47), Nij ¯ o-j ¯ o (34). ¯

*Findings:* One of the most typical materials used to produce walking paths in Japanese gardens is no doubt stone. Stones are used in different ways: sometimes to create wide, cohesive and formal paths and sometimes as informal stepping stones carefully laid out in the ground, *tobi-ishi* or in water, *sawatari*. Stone paths do not produce so much sound today, though when they were originally conceived, traditional wooden shoes known as *geta* would have created a characteristic clonking sound [43,44]. This effect can still be experienced in Rentaroh Taki's Memorial garden where such shoes are offered to visitors.

Another noteworthy feature is the extensive use of wooden floors. They are typically found in the verandas known as *engawa* (Figure 12). Some gardens, including the famous Ryoan-ji, are designed ¯ to be experienced from the *engawa*, where the thumping sound of wood becomes a characteristic *soundmark* [30]. It is quite common for the wooden verandas to extend into the garden as a system of roofed walkways that connect various buildings on the premises. Such walkways offer shade on hot days and shelter from precipitation, while the sound of thumping and squeaking floorboards constitute a potentially pleasant accompaniment, as the following example from Shoren-in illustrates:

*The wooden walkways make for a nice stroll through the garden, as the roof provides shade in the sunshine. The floor is made from thick wooden floorboards; it has a nice, raw tactility to it. Each floorboard is slightly di*ff*erent in character; the thickness varies. Some floorboards make a heavy thumping sound when walked upon.*

Shoren-in; 28 April 2018

**Figure 12.** Nomura-ke Samurai teien in Kanazawa. View from the east overlooking the upper pond.

The sound of the verandas' wooden floors differs between gardens depending on the construction. One particularly noteworthy effect is the squeaking sound which has come to be called *uguisubari*, or nightingale floor, after the Japanese bush warbler, *uguisu* (*Horornis diphone)*, whose sound it supposedly resembles (Video S7, https://vimeo.com/311380259). Legend has it that the sound of the *uguisubari* was used to warn inhabitants of assassins and other unwelcome guests. This particular function may have played out its role today, yet there may be related applications in other contexts. For instance, in a study reporting on the role of soundscape in a rehabilitation garden, it was found that a patient who wanted to experience "social quietness" used the sound of a gravel path as a warning system from the potential approach of other patients in the garden [85]. Gravel paths can also be found in Japanese gardens, though the extensive use of stone and wood makes it less of a characteristic feature.

#### 3.3.6. Sound Sculptures

*Description:* A sound sculpture is a multisensory embellishment where sound plays an important part. There are many types of sound sculptures, and they can be driven by sources such as wind, water or electricity (speakers). The sound is usually accompanied by some sort of visual container that is typically designed to be a part of the expression.

*Gardens of particular interest:* Ohashi-ke (35), Enk ¯ o-ji (9), Taiz ¯ o-in (51), Giou-ji (14), Shisen-d ¯ o (45). ¯

*Findings:* Two kinds of garden features that qualify as sound sculptures were studied: *suikinkutsu* and *sozu (shishi-odoshi) ¯* .

*Suikinkutsu* is a simple but elegant design element that can be found in conjunction with some washing places, *tsukubai* (Figure 13a). The sound is created as excess water from the water basin, *chozubashi*, is led to drop into a hidden cavity below ground. The cavity essentially consists of a pot which is turned upside down and buried in the ground. The sound is generated at the bottom of the cavity where the drops break a water surface. The sound can be described as metallic and vibrant, fresh and melodic. The sound of *suikinkutsu* was experienced and documented in four gardens (Video S8a–e, https://vimeo.com/showcase/5683754). As a sound sculpture, *suikinkutsu* is unusual, as part of it is hidden beneath ground.

**Figure 13.** Two typical sound sculptures found in Japanese gardens. (**a**) A suikinkutsu in Ohashi-ke, ¯ Kyoto (Video S8b,). (**b**) A *sozu ¯* , also known as *shishi-odoshi* in Shisen-do, Kyoto (Video S9a). ¯

A *sozu ¯* is a contraption originally developed by farmers to keep deer, boars and other wild animals away from their fields (Figure 13b). It was subsequently incorporated as a design element in gardens [43]. It consists of a bamboo tube, which is opened up on one side to allow water to flow inside. The tube is mounted on a rack that works as a hinge so that the tube can move freely around the axes. The open end of the tube is located beneath a water supply. As the tube fills up, its center of gravity gradually shifts until it passes the hinge, causing the contraption to fall over, emptying its water and then making a distinct sound as it falls back to its resting position (Video S9a,b, https://vimeo.com/showcase/5683758). The *sozu ¯* is often called a *shishi-odoshi,* in which case a more overarching reference is inferred; both terms are generally translated as deer scarer.

#### 3.3.7. Biotope Design

*Description:* Biotopes and natural ecosystems can have positive effects on the soundscape, including the presence of song birds and other animals [87]. Song birds are generally attracted by basic features like access to water, shelter and food (most typically insects, seeds and berries) [88,89]. A general guideline is that birds prefer vegetation that is multi-layered, dense and variated.

*Gardens of particular interest:* Rurikoin (37), Shisen-do (45), Sanb ¯ o-in (41), Shin'en (44) Rurikoin (37), ¯ Nanzen-in (32), Shisen-do (45), Rentaroh Taki's memorial garden (81). ¯

*Findings:* Many of the studies of Japanese gardens were conducted during April and May, which provided ample opportunities to experience birds and other animals, particularly in gardens that bordered forests, such as Rurikoin, Shisen-do and Sanb ¯ o-in, or gardens that included woodlands, such ¯ as Shin'en (Heian Jingu Shrine). Some species, like heron, seemed to enjoy sitting on stones laid out in ponds. It has previously been suggested that such stones were deliberately laid out in shallow water with the purpose of attracting birds [43]. Prerequisites for birds may also be increased by the use of vegetation, particularly fruit-bearing plants [44].

Spring is also the mating season for frogs, and the sometimes intense croaking could be experienced in pond gardens like Rurikoin, Nanzen-in and Shisen-do. The e ¯ ffect was particularly breathtaking on May 15th 2018 in Nanzen-in, where the whole system of ponds seemed to be "illuminated" by croaking frogs, creating an intense spatial presence (Video S10, https://vimeo.com/270885125). Frogs may in fact have been actively placed in Nanzen-in when the garden was created in the late Kamakura period, as indicated by 15th century sources accessible to Ueyakato Landscape, that manages the garden today [90].

Two other animals are commonly placed in Japanese gardens: turtles and carp fish (Figure 14). Almost all ponds encountered in the study had carp fish, while turtles were less common. In some gardens, for example Konchi-in, carp and turtles live side by side, creating spectacular sightings and the occasional splash. Both carp fish and turtles can make rather loud splashes, an effect that amuses visitors and possibly stimulates a heightened and active "listening" to the environment (as compared to a passive "hearing") [91].

**Figure 14.** A turtle and a carp fish in Konchi-in temple garden, Kyoto.

#### 3.3.8. Attract Activities

*Description:* This Soundscape Action acknowledges the relationship between landscape functions and sound by emphasising the fact that the introduction of new functions may generate activities that benefit the soundscape. A typical example of such a function is a café, the atmosphere of which can also be a quality for people passing by.

*Gardens of particular interest:* Chishaku-in (5), Daisen-in (7), Konchi-in (27), Nanzen-ji (33), Rurikoin (37), Tenryu-ji (52), Suizen-ji Joju-en (77). ¯

*Findings:* Ticket sales, souvenir shops and cafés constitute some examples of recurring activities noted in the study. Religious activities could be experienced in gardens located in (or in the vicinity of) Buddhist temples and Shinto shrines. Even if secluded, the sound of religious activities would easily travel through the thin walls of the traditional buildings, making it customary to hear monks' chanting or the distant ringing of a gong as a faint, mystical atmosphere. For the visitor, these atmospheric tints may constitute a reminder of the sacred context in which the gardens are located. The act of praying is ritualised in Shinto shrines as well as in Buddhist temples. In both cases, praying involves throwing a coin into an offering box, *saisen-bako*: an event which can create a rather loud sound if the box has resonating qualities.

#### *Int. J. Environ. Res. Public Health* **2019**, *16*, 4648

A *temizuya* is a kind of water basin, where visitors go to purify themselves before a visit. The purification is carried out with the aid of a wooden ladle which is used to pour water on one's hands and mouth. The *temizuya* generates a rather specific soundscape typically involving the murmur of people, the sound of water splashing on the ground and the distinct wooden clonks as the ladles are put down. The *temizuya* are typically associated with Shinto shrines though can also be found in Buddhist temples.

To sum up, the sound of activities signals the presence of other people and may be charged with symbolic meaning and atmospheric qualities. In design situations, such aspects should be given consideration in relation to the intended purpose. Social activities may be regarded as a quality in many contexts, yet may also be problematic in some situations. This is the case in gardens intended for recovery from stress-related illnesses, where seclusion from some forms of social activities is required [85,92].

#### 3.3.9. Resonance/Reflection

*Description:* Resonance/Reflection is concerned with how acoustically hard materials can be used to enhance wanted sound sources. For instance, if a concrete wall is strategically located behind a water feature, the sound of water can be enhanced by the reflections the wall creates.

#### *Gardens of particular interest:* Sanbo-in (41), Konchi-in (27), Shoren-in (47), Murin-an (31). ¯

*Findings:* These kinds of acoustical effects were mainly found in and around water features, typically enforced by stones enhancing and directing the sound of water in gardens like Sanbo-in, Konchi-in ¯ and Shoren-in. The design intentions underlying these cavities could be several, and resonance may not have been the main one. Some of these cavities also worked to direct the sound, as the following notation from Konchi-in in Kyoto indicates:

*The use of the grotto has other advantages as well. Not only is the sound amplified through the resonance it creates, but the grotto also seems to have a parabolic shape leading the sound across the pond towards the entrance.*

#### Konchi-in; 21 April 2018

This way of directing sound opens up for a wide range of creative possibilities and experiential effects. In the above example, the sound of the waterfall is best experienced from the opposite (northern) side of the pond, yet because it cannot be seen from that point, an anticipation is created that encourages further exploration of the garden in search for the source. It has previously been described how the sound of hidden water features can be used to create a "spatial appeal" that invites a visitor to go from a "here" to search for a "there" [41]. Such kind of explorations seem to be related to the notion of "soft fascination" as described by Kaplan and Kaplan, as part of their attention restoration theory (c.f. [48,85,93]).

#### **4. Discussion**

#### *4.1. Soundscape Actions as a Design Tool*

The Soundscape Actions presented in the study constitute a collection of examples, illustrating how landscape architects and other professionals could think about soundscape design in their work [27,36]. Covering a total of 23 Soundscape Actions divided in three categories, it is an extensive framework. This emerged as beneficial in the study, as it provided a variety of entry points from which to understand the soundscapes of Japanese gardens. In this way, aspects that would not necessarily have been noted otherwise could be highlighted.

In the study, 19 of the 23 Soundscape Actions were applicable. The other four (Embrace unwanted sounds, Abolish certain functions, Low noise screens and Atmospheric design) were excluded because no concrete examples were encountered. This is partly owing to the particular character of Japanese gardens, where certain expressions are not likely to be found, such as urban vibrancy (Embrace unwanted sounds) or artificial speaker installations (Atmospheric design). In future developments of the tool, it might be possible to take this into consideration by designing a system in which Soundscape Actions are tagged based on their appropriate application (for instance urban vibrancy, traffic planning and parks/gardens). The other two excluded actions seem relevant for Japanese gardens, yet could not be registered in the study. "Abolish certain functions" (such as roads, factories or air conditioners) should be a fruitful strategy to enhance garden experiences, however no clear examples were identified (in part, owing to language difficulties). "Low noise screens" is a novel solution for traffic noise reduction and only a limited number of practical applications are known worldwide.

Some sonic experiences that were noted in the study did not correspond directly to any of the 23 Soundscape Actions. These experiences could generally be described to be of a phenomenological nature, including effects related to movement, orientation, behaviour and subjective contemplation. This discrepancy may partly be regarded as a consequence of the autoethnographic approach, in which such experiences should be expected to be brought forward; it might also be taken to indicate specificities associated with the context being studied, i.e., Japanese gardens. The encountered effects were referred to in the study in conjunction with related Soundscape Actions (for example subtle water features in Section 3.3.3 and interactive walking materials in Section 3.3.5). There seems to be scope to formulate new Soundscape Actions. This potential will be further examined in upcoming publications.

#### *4.2. Implementation and Usage*

The Soundscape Actions constitute a set of strategies with the potential to improve experiential qualities in gardens and other urban (green) spaces. However, it should be noted that, because each situation is unique, the effect of Soundscape Actions will vary, and all of them will not necessarily lead to an improvement every time. One of the challenges of soundscape design is to understand complex relationships and the degree to which various factors effect a situation. The tool brings together a collection of central aspects as a framework that can be used to build an experience base. In future work, this base could be further developed by applying the tool in other contexts and with other participants. The material thus generated could be collected and presented online for accessibility. A digital platform suitable for this purpose already exists (https://soundscapedesign.info). Practitioners should be involved in future developments so as to safeguard compatibility with established working patterns. One way forward could be to evaluate the tool in design workshops and/or use it to build an intervention (which could subsequently be evaluated). Future work could also explore the possibility of integrating the tool with computer aided calculations, known as auralisation. This would potentially allow for real time comparisons between Soundscape Actions in different scenarios.

The study focused on tranquil soundscapes in Japanese gardens, taking into consideration the potential for negative as well as positive health effects [24,25]. The study did not assess health effects per se, but rather used existing research to envisage the effects. As discussed above, there is a strong body of research showing that exposure to noise increases stress and leads to a number of severe negative health effects [24]. This should be taken in account when designing outdoor spaces. Problems with noise have primarily been addressed by engineers and acousticians, yet much can also be achieved by urban planners and designers [27,28,32,94,95]. Increased collaboration across disciplines could further enhance prerequisites for health promoting environments. In terms of positive health effects, there is an increasing number of studies indicating that exposure to nature sounds have beneficial effects on health [25]. This should also be given consideration in design situations so that such sounds are afforded the best possible prerequisites. The present study provided several examples of how this could be accomplished.

#### **5. Conclusions**

This study has focused on the Japanese garden tradition in order to gain insights on the design tool Soundscape Actions, and how it can be used to improve soundscapes in gardens and other green spaces. Field studies were conducted in 88 Japanese gardens, and notated experiences from (in total) 136 visits to those gardens were used as empirical material. Most of the analysis and presentation centred on gardens located in Kyoto (n = 54), as this allowed for comparisons within the city.

The research was designed using an autoethnographic approach [45]. This made it possible to obtain detailed understandings on the intersection between landscape architecture and soundscape research. Autoethnography emphasises subjectivity and therefore allows for embodied experiences to be brought forward. This enables the researcher to go into detail and consider the interplay between senses and various site-specific factors. It should be pointed out that the subjective nature of the method could limit the applicability of the findings in other contexts. To increase generalisability, supportive data was collected in situ, including SPL readings, images, videos and field recordings. This material has been made available in the paper, and should mitigate interpretation. Where applicable, findings were compared and discussed in relation to previous research. Yet, caution should be taken when interpreting the results, and future studies involving more participants are recommended to validate the findings.

Sound constitutes one of several aspects of the environmental experience, and it should be stressed that the interplay between the senses is of importance, even when focusing on a particular sense [85]. The study highlights the role of such relationships when discussing masking and other complex mechanisms. Masking is based on the assumption that sounds experienced as positive can be used to reduce the annoyance caused by traffic and other (unwanted) technological sources [72,74,82]. The benefits of masking have been debated, partly because the underlying mechanisms are not fully understood. This study indicated that spatial context and multisensory impressions play an important role, and that such factors should be given increased consideration in future research. The subjective experience of individuals seems to be essential if these effects are to be uncovered further. Subjectivity is emphasised in the ISO definition of soundscape [34], yet has been given surprisingly little attention in research thus far.

Much effort has been invested to analyse and uncover the knowledge existing in the Japanese garden tradition, though most previous publications have tended to prioritise visual aspects. The present study illustrates how Japanese gardens can also be informative regarding sound. The findings presented here follow the Soundscape Action design tool and cover a wide array of perspectives. The study illustrates multiple strategies to avoid noise, such as the use of remote locations, garden walls and absorbent moss. Reduction of noise allows for other and more delicate experiences to come forward, like sounds of nature. Sounds of nature constitute an essential element in Japanese gardens and they can be articulated and enhanced in design solutions. This is highlighted in the study through the way that rustling trees are used, how animals are attracted and how the sound of water is orchestrated to produce loud masking sounds. Taken together, the soundscapes of Japanese gardens are surprisingly tranquil considering the modern and densified cities that often surround them.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1660-4601/16/23/4648/s1, Video S1: Compensation/Variation in Ohashi-ke; Video S2: Spring in Sanb ¯ o-in; Video S3: Screening and Masking ¯ in Murin-an; Video S4: Muffled Waterfall in Rikgui-en; Video S5: Enhanced Water Stream in Shin'en; Video S6a: Water and Wind in Chishaku-in; Video S6b: Subtle Water Drops in Ginkaku-ji; Video S7: Nightingale Floors; Video S8a: Suikinkutsu in Ohashi-ke, 1 ¯ /2; Video S8b: Suikinkutsu in Ohashi-ke, 2 ¯ /2; Video S8c: Suikinkutsu in Enko-ji; Video S8d: Suikinkutsu in Taiz ¯ o-in; Video S8e: Suikinkutsu in Giou-ji; Video S9a: Shishi-odoshi in ¯ Shisen-do; Video S9b: Shishi-odoshi in Taiz ¯ o-in; Video S10: Frogs in Nanzen-in. ¯

**Funding:** This research was funded by the Japanese Society for the Promotion of Science, JSPS, through an International Research Fellow programme (Postdoctoral short term).

**Acknowledgments:** The author wishes to thank Keiko Torigoe, Kozo Hiramatsu, Ann Bergsjö, Yoshio Tsuchida, Haruyoshi Sowa, Masafumi Komatsu, Jitka Svensson, Catherine Szanto, Sanae Kumakura and Isami Kinoshita for recommendations on soundscapes and gardens in Japan. Moreover, the author wishes to thank Martin Tunbjörk and Gábor Felcsuti at SWECO Acoustics in Malmö, Sweden, for evaluating the equipment used for sound pressure level measurements. Finally, the author wishes to thank three anonymous reviewers as well as Ida Isaksson for valuable comments on the manuscript.

**Conflicts of Interest:** The author declares no conflict of interest.

#### **Appendix A**

This appendix of Japanese gardens provides detailed information about all study visits that were carried out (Table A1).

**Table A1.** An overview indicating the numbers, names and locations of the studied gardens, as well as number of visits, dates for visits and total time spent in each garden.



**Table A1.** *Cont.*

#### **References**


© 2019 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

International Journal of *Environmental Research and Public Health*

### *Article* **Implementation of Quiet Areas in Sweden**

#### **Gunnar Cerwén 1,\*,† and Frans Mossberg <sup>2</sup>**


Received: 26 November 2018; Accepted: 20 December 2018; Published: 7 January 2019

**Abstract:** The notion of quiet areas has received increasing attention within the EU in recent years. The EU Environmental Noise Directive (END) of 2002 stipulates that member states should map existing quiet areas and formulate strategies to keep these quiet. Quiet areas could play an important role in balancing densified urban development by ensuring access to relative quietness and associated health benefits. This paper reports on a recent study investigating how the notion of quiet areas has been implemented in Sweden. The study, initiated by the Sound Environment Center in 2017, was carried out in two phases. In phase one, an overview of the current situation was obtained by scrutinizing regional and municipal mapping initiatives, aided by a short digital questionnaire sent out to all 290 municipalities in Sweden. This provided a general understanding and highlighted initiatives for further study in phase two. The results revealed that 41% (*n* = 118) of Sweden's municipalities include quiet areas in their general plans, but that significantly fewer of these have sophisticated strategies for implementation (*n* = 16; 6%). Moreover, the interest in quiet areas in municipalities does not seem to be directly related to the END, but is instead inspired by previous regional initiatives in Sweden. The study highlights a number of considerations and examples of how quiet areas are approached in Sweden today. In general, Sweden has come a long way in terms of identifying and mapping quiet areas, but more progress is needed in developing strategies to protect, maintain, and publicize quiet areas.

**Keywords:** quiet areas; environmental noise; noise abatement; soundscape design; landscape planning; urban planning; general plan; sustainability

#### **1. Introduction**

In the EU's Environmental Noise Directive (END) of 2002 (2002/49/EC) [1], member states are asked to make an inventory of existing quiet areas and devise strategies for their protection. The whole concept of quiet areas could be regarded as a way to turn the question of sound environment around, to focus on potential positive qualities of the sound environment and not only on noise and disturbances. The concept connects to tendencies in cultural and general environmental movements of the time with roots back in the soundscape movement [2] and acknowledges the fact that sound may have positive health effects [3].

Sweden is known as one of the countries in Europe working more actively with quiet areas, including several initiatives on regional and municipal level. This could be related to Sweden's early implementation of the soundscape perspective in research on environmental noise, not least through the Mistra project "Soundscape for better health" [4], carried out between 2000 and 2007, in which the notion of "quiet façade" was introduced.

Recent exposure data from the European Environment Agency (EEA) demonstrate that more than 100 million European citizens are negatively affected by high noise levels, impacting human health [5]. The World Health Organization estimates that one million healthy life-years in Western Europe are lost annually as a result of exposure to traffic noise [6]. Noise has been shown to have severe negative effects on health, including hearing damage, sleep disturbance, hypertension, and cardiovascular disease [7,8].

The situation in Sweden concerning exposure to traffic noise, and to airborne particles, in the immediate vicinity of permanent dwellings has become increasingly problematic since the introduction of more relaxed noise regulations by the Swedish government in 2015 [9], in response to political pressure [10]. In sharp contrast to the scientific advice from the research community on health and the environment, more noise is now permitted close to people's dwellings than was the case before the updates to the legislation. This proved to be politically acceptable because of an urgent need for housing, but also because of building industry and business interests. The higher threshold levels now in place mean that, in principle, unlimited noise is permitted on the noisiest side of dwellings in noise-exposed urban areas [9,11].

The recently published Environmental Noise Guidelines for the European Region [7] define the aim of these guiding principles as reducing noise while conserving quiet areas. Based on previous research [12–14], the report states that "people appreciate quiet areas as beneficial for their health and well-being, especially in urban areas" [7]. A growing number of studies indicate that exposure to natural sounds may have positive health effects by reducing stress [3]. Thus quiet areas not only protect against noise, but also reveal positive sounds that would otherwise be masked by noise.

Urban areas around the world are currently facing new challenges, as there is increasing demand for densification related to sustainability. In order to safeguard the quality of parks and other recreational spaces in the future, quiet areas are likely to become even more important, not only in cities, but also in the vicinity of cities and in open country.

There are good reasons to build on the notion of quiet areas, as postulated in the END [1]. The present study investigated how the concept has been implemented in regional and municipal contexts in Sweden. The results may be useful as reference for further implementation of the concept in Sweden and other countries, as well as informing future strategies and guidelines on EU level.

#### **2. Materials and Methods**

The study comprised two phases: a quantitative phase providing an overview of the situation in Sweden and a qualitative phase in which chosen examples were studied further.

In phase one, an overview was made and background information on quiet areas in Sweden was obtained by studying government documents and reports dealing with quiet areas. The internet was searched extensively and key individuals were contacted. The outcomes are described in Section 3 of this paper.

Based on the findings, a subdivision was made into regional and municipal initiatives when selecting examples for further studies. The regional initiatives were studied manually, by going through reports and other documents describing these. A total of 10 initiatives were identified and are described in Section 4. In order to study municipal initiatives, an email was sent out to the registrar of all 290 municipal authorities in Sweden. The email included a short description of the project and a digital questionnaire containing three introductory questions on how the municipality works with quiet areas, which the registrar was asked to redirect to the appropriate department. These questions were intended to provide an overview of the extent to which the concept of quiet areas appears in the municipalities' general plans (Question 2) and to identify initiatives for further study (Questions 1 and 2). Question 3 was included to obtain contact details. For Questions 1 and 2, a set of pre-given options was provided, plus a box for comments:


Within one month, during which two reminders were sent out, 208 of the municipalities had answered (response rate 71.7%).

In order to provide a full overview for question 2, the remaining 82 municipalities were investigated manually by means of a digital search in their respective general plans. The search words used were "quiet" (Swedish: tyst) and "noise-free" (bullerfri). In this way, it was determined whether each of these municipalities deals with quiet areas or not. Moreover, if it was found that the municipality mentioned quiet areas, it was determined whether this mention was brief or comprehensive (corresponding to the alternatives in Question 2). Any other trends or noteworthy issues were entered in an Excel spreadsheet.

A total of 47 municipalities were identified in phase one as interesting for further study. This selection was based on answers given to the questions and information obtained in the manual search of general plans. Moreover, some initiatives were added to this group based on searches on the internet and tips from the public (the project generated media attention that encouraged some people to make contact).

In phase two of the study, the 47 chosen initiatives were investigated further to obtain more information about how quiet areas are treated in Sweden. Furthermore, efforts were made to identify municipalities that had taken action and to explain why others had not taken any action at all. We also sought to identify and discuss general tendencies, challenges, and future possibilities.

Data were collected in phase two through different methods. Initially, a more detailed questionnaire was sent out to municipalities selected for further study. This questionnaire was sent out digitally to email addresses obtained in response to Question 3 in the first digital questionnaire. The detailed questionnaire consisted of 23 questions that dealt with how the municipality had gained inspiration for its work on quiet areas, if there were any measures to protect the quiet areas, and if these areas were used in marketing and tourism, as well as general questions about the level of knowledge and potential for improvement. After one month and one reminder, 25 out of 47 questionnaires were returned (53.2% response rate). Further correspondence followed in some cases.

Results from both research phases are reported below, starting with an overview of how the notion of quiet areas has developed in Sweden and the EU, and important initiatives and definitions (Section 3). We then go on to describe how the concept is used in Sweden, on regional level (Section 4) and municipal level (Section 5). This is followed by a discussion of general trends and challenges identified (Section 6) and conclusions (Section 7).

#### **3. Quiet Areas: Background and Definitions**

In Sweden, the notion of quiet areas can be traced to the late 1990s, when the Swedish Road Administration initiated a pilot project in which two municipalities in Jönköping County were mapped for relative quietness [15]. Within a few years, similar mapping initiatives had been undertaken in two other counties [16,17] and in major urban regions [18,19]. Several of these initiatives were connected to a collaborative project officially launched in 2002 with the purpose of developing a method for inventory of quiet areas [20]. The collaboration involved influential stakeholders, such as the Road Administration, the Rail Administration, the National Board of Housing, the Civil Aviation Administration, the Environmental Protection Agency, and two major county councils. The project finished in 2005 [21], and the outcomes were made accessible in an influential report entitled "A Good Sound Environment—More than Merely Absence of Noise" published by the Swedish Environmental Protection Agency [22].

The method used in the collaborative project was based on the assumption that disturbances vary depending on context, and it considers areas divided into five different noise classes (A–E). Noise class A has the highest demands; it corresponds to areas completely free from community noise, such as remote areas in the mountains, forests, and national parks. The benchmark for noise class A is set at 25 dBA (A-weighted instantaneous sound pressure level), whereas the class with the next-highest demand, noise-class B, has a benchmark of 35 dBA. Noise classes C and D are intended for forests and recreation areas in proximity to urban developments and are both based on a benchmark of 45 dBA (instantaneous), but what is called the "exceeding time" differs. The lowest demands are found in noise class E, where the benchmark is set to 45–50 dBA (equivalent), or 10–20 dBA lower than the surrounding sound pressure level. Noise class E is intended to be applied in urban areas such as parks.

In the EU, the development of quiet areas can be traced back to 1996 and the document "Green Paper on Future Noise Policy" [23], where it is mentioned how noise mapping could be used to identify quiet areas. The thoughts raised in this green paper are enforced in the END [1]. However, the instructions in the END are relatively vague, which has resulted in different interpretations and implementations.

In 2014, "The Good Practice Guide on Quiet Areas" [24] followed up on the END by mapping how the question had been dealt with, through examining existing examples and initiatives. The report provided a good overview of existing initiatives in Europe. No specific recommendations were given, but it was suggested that the review could be used as inspiration and reference was made to "competent authorities" for further guidance.

In 2016, the report "Quiet Areas in Europe: The Environment Unaffected by Noise Pollution" [25] introduced a method called Quietness Suitability Index (QSI) for identifying quiet areas in rural contexts. The QSI approach makes use of existing noise mapping data combined with land use data to identify areas that are potentially quiet. No applications of the method have been made yet in Sweden.

The END [1] distinguishes between two different kinds of quiet areas, "open country" and "agglomerations", which are defined as follows:

"A quiet area in open country' shall mean an area, delimited by the competent authority, that is undisturbed by noise from traffic, industry or recreational activities."

"A quiet area in an agglomeration' shall mean an area, delimited by the competent authority, for instance which is not exposed to a value of Lden or of another appropriate noise indicator greater than a certain value set by the Member State, from any noise source."

In both definitions, the END refers to the "competent authority" of the member state. In Sweden, the competent authority can be said to correspond to the collaboration group mentioned above and the noise classes it proposed [22]. Thus in Sweden, the END definition for rural settings can be said to correspond to noise classes A and B, while the END definition for urban settings corresponds to noise class E. Noise classes C and D, are intermediate, suggesting that there is scope for a third type of quiet area, i.e., those in 'proximity to urban areas'.

The collaboration group's noise classes have been influential for assessing and mapping quiet areas in Sweden. Many of the municipalities in Sweden refer to the benchmarks suggested in the method. However, there are other values that do not relate to this framework. For instance, two benchmarks, 30 dBA and 40dBA, which are still used relatively frequently, derive from early regional initiatives in Sweden undertaken by the Road Administration [16,17] and an earlier version of the collaboration group's method [20].

Moreover, it is common for the benchmarks to be adjusted, depending on the context in which the noise classes are applied and the data available. In most of the initiatives studied here, only one or two of the classes were used. Non-acoustic factors were influential when quiet areas were defined, including accessibility, natural qualities, and cultural qualities.

More recently, two new methods have been proposed in Sweden [26,27], as well as the QSI on EU level [25], but it is too soon to assess how they are being applied in Sweden.

#### **4. Regional Initiatives in Sweden**

On regional level, a total of 10 initiatives, together covering 121 municipalities, were recorded (see Figure 1a). As can be seen from the map, some of these regional initiatives overlap each other. These areas, indicated with darker color on the map, correspond to some of the most densely populated areas in Sweden, which could be a result of more extensive noise problems in these areas. Moreover, it is interesting to note that, of the 15 municipalities found to have carried out more far-reaching work on quiet areas (see Section 5.1 and Figure 1b), all except one had been preceded by a regional initiative. This suggests that regional initiatives are an important catalyst for implementation of quiet areas.

**Figure 1.** (**a**) Location of regional initiatives for quiet areas in Sweden. Overlapping initiatives are marked with darker color. (**b**) Location of municipalities that describe quiet areas in their general plans, with these descriptions divided into three categories: brief, unclear, and elaborate.

A short summary of the regional initiatives is presented below, following the chronological order of introduction within the respective region (see Table 1 for an overview).

The first regional initiative was that in Jönköping County by the Swedish Road Administration, in 1998 [15]. Starting from a 30 dBA level equivalent for the noise sources mapped (road, industry, rail, and recreational activities), four quiet areas were identified. Seventeen years later, Jönköping County Council conducted mapping for the whole county on the methodological basis of GIS data [26]. The method was presented as potentially useful for other areas in Sweden as a cost-effective method for modeling larger areas, including vegetation data. By including vegetation data, it was found that

it was possible to increase the number of quiet areas. The mapping took account not only of human wellbeing, but also of wildlife. Five categories were introduced, described with words rather than decibel levels. The mapping revealed that 22% of the area of Jönköping County fell within the strictest category, i.e., free from noise.


**Table 1.** Overview of regional 'quiet area' initiatives in Sweden.

Västra Götaland County in Western Sweden mapped its quiet areas in 2001 [17]. A benchmark of 30 dBA (equivalent, year) was used and, despite this relatively strict value, it was found that more than half of the area in the county qualified. In that case, the mapping of noise data was coordinated with land use, and quiet areas with experiential potential for nature and recreation were illustrated in maps in a final report. In another initiative in the same county, the municipalities around the Gothenburg region conducted mapping that was published in 2014 [28]. This mapping was based on noise emanating from outside the agglomeration, i.e., roads, rail, air traffic, wind power, industry, and noisy recreational activities. It concerned noise class C (45 dBA instantaneous), as proposed by the collaboration project [22] described in Section 3, but the demands set were higher for some noise sources (large industry, harbor activity, wind power, quarries, dumps, and crushing plants) and equivalent levels were combined with maximum levels as indicators. In all, nine major quiet areas were identified in the region and presented on an illustrated map.

In Scania County, at the southern tip of Sweden, broad mapping was conducted in 2003 based on two benchmarks, 30 dBA and 40 dBA (equivalent levels) [16]. The outcomes were correlated with areas judged to have natural, cultural, and recreational qualities, and these were illustrated in maps. The mapping showed that noise-free areas were rare in Scania County, especially in western parts. Based on this finding, the final report suggested that quiet areas should be valued more highly, particularly if they lie in proximity to urban areas, have connections to other quiet areas, or have high accessibility. For instance, it was found that areas below 30 dBA were generally far away from urban agglomerations, whereas access to areas between 30 dBA and 40 dBA was better. The regional mapping in Scania county was followed up in the next year, when the outcomes were applied in two municipalities, one characterized by an urban setting (Helsingborg) and one characterized by a rural setting (Hässleholm) [29]. In the urban setting, the benchmark 40 dBA was used to identify four areas. In the rural setting, the benchmarks 30 dBA and 40 dBA (equivalent levels) were overlapped with other factors to identify three areas. The municipality found that it was possible to co-locate noisy activities and to use quiet areas to promote the region. In both municipalities, topographical variations were identified as an important factor to consider.

In the Stockholm area, a number of mappings have been conducted on various regional levels since the turn of the millennium [19,27,30,31]. In the first of these carried out in 2000 [19], the benchmark 45 dBA (equivalent) was used to identify quiet areas in relation to green structures. A few years later, another initiative was taken in which five areas were chosen on the basis that they should both be quiet and have other recreational qualities [30]. Data were extracted through a combination of calculations, measurements, and user interviews. The benchmark 45 dBA (instantaneous) was identified as a threshold for experienced quietness, with a certain exceeding time that varied depending on context. This initiative was linked to the collaboration project mentioned in Section 3 [21]. Another initiative taken at county level proposes a new method for mapping of quiet areas in the region [27]. In addition to these initiatives, the region has a shared development plan in which the notion of quiet areas is emphasized, particularly in relation to green structure [31], and Stockholm County Board has presented a map of quiet areas in the region.

#### **5. Initiatives on the Municipality Level**

The following presentation of municipal initiatives is divided into three parts. The first part provides a quantitative overview of how the notion of quiet areas has been dealt with in Swedish municipalities' general plans. In the second part, experiences from six municipalities that have progressed further in this work are described. The third part deals with municipalities that have not been working with the concept of quiet areas and looks at the different reasons for this.

#### *5.1. Quiet Areas in the General Plans*

It was found that 118 of Sweden's 290 municipalities (41%) mention the concept as part of their strategies in their general plans. Based on how extensive the initiatives were, the municipalities were divided into three categories; brief, unclear, and elaborate descriptions (see Figure 1b and Table 2). These categories are further described below. As indicated in Table 2, the majority of these municipalities focus on rural rather than urban settings. Even though several of these rural initiatives are located in proximity to agglomerations, the lack of focus on urban settings suggest that there is scope to elaborate on quiet areas within such contexts in Sweden.


**Table 2.** Overview of municipal initiatives in Sweden, highlighting the context focused upon in the initiative (urban and/or rural).

"Brief descriptions" (*n* = 70; 24%) generally contained a short description and a positive attitude to the notion of quiet areas. Reference was sometimes made to a regional initiative and it was sometimes stated that this issue will be dealt with in future work. Brief descriptions sometimes lacked a definition of how to classify quiet areas and, in most cases, no particular areas were set aside or mapped.

The descriptions of quiet areas provided by 32 of the municipalities (11%) were categorized as "unclear". In these cases, the term 'quiet areas' was confused with other established planning concepts, such as "large unaffected areas" (stora opåverkade områden), "nature reserves" (naturreservat), and "radiation-free areas" (strålningsfria områden).

The descriptions of quiet areas provided by 16 municipalities (6%) in their general plans were categorized as "elaborate". These descriptions generally included definitions of what a quiet area is, as well as maps of designated areas. In some cases, designated areas were protected and descriptions were provided of how they should be maintained. The 16 municipalities with elaborate descriptions

were: Botkyrka, Habo, Helsingborg, Huddinge, Hässleholm, Hörby, Malmö, Munkedal, Nybro, Skövde, Sollentuna, Stockholm, Sundbyberg, Tanum, Tibro, and Östra Göinge. As previously mentioned, 15 of these 16 initiatives (94%) were preceded by a regional initiative (see Figure 1b). In contrast, for the other two categories, brief and unclear descriptions, the corresponding values were 43% (*n* = 30) and 31% (*n* = 10) respectively (Table 3).


**Table 3.** Number and percentage of municipal initiatives that were preceded by a regional initiative.

This suggests that the regional initiatives were important in inspiring initiatives on municipal level. This was confirmed to some extent by responses to the more detailed questionnaire, where municipalities that had progressed in their work were asked to explain their sources of inspiration. The responses revealed that inspiration from regional mappings accounted for 28% (*n* = 7), while inspiration from neighboring municipalities accounted for 12% (*n* = 3). Local motiving forces emerged as the most important source of inspiration, for 60% (*n* = 15) of the initiatives. Interestingly, only 4% of responding municipalities (*n* = 1) said that the END had been an important factor, suggesting that there is scope for further promotion and/or enforcement of the directive.

#### *5.2. Examples of Municipal Initiatives*

Initiatives on municipal level were selected for further study with the ambition to focus on examples that could be inspiring for future work, including initiatives that applied more unusual ways of tackling the subject.

Starting in rural locations, the municipality of Munkedal, north of Gothenburg, was covered by regional mapping in 2001 [17]. In the municipality's general plan, Munkedal ÖP 14 [32], the question is treated on a more detailed level and two quiet areas are designated, both of which border on a wildlife area designated "large unaffected area" (Figure 2). In the plan, the municipality distinguishes between "unaffected" and "quiet areas", but acknowledges overlapping so that certain parts of larger unaffected areas may include quiet areas. This was not the case in some other municipalities, where instead it was assumed that "large unaffected areas" were also quiet areas by definition. The same problem was found in relation to "nature reserves" and other recreational areas.

Nybro, another small municipality in a rural setting, included an addition about quiet areas to its general plan in 2015 [33]. All known sources for noise were mapped. Based on the noise classes proposed by the collaboration group [22], two benchmarks were introduced (35 dBA and 45 dBA, instantaneous). The idea was that by using two different definitions, the various preconditions in the region could be accounted for. After taking account of non-acoustic factors, four areas were proclaimed as free from noise and it was stated that they should remain so. According to the municipality, the designation of quiet areas has already prevented exploitation of wind power in some of these areas. The municipality also says that the areas are maintained through reviewing permits and by supervision, but that more could be done in terms of publicizing them and making them accessible.

In the municipality of Hörby, the initiative to designate quiet areas in the region arose out of activities by the citizens themselves, which were met with positive attitudes by the town council's planning department. In the current general plan adopted in 2016 [34], parts in the east and north of the municipality are designated free from noise and protected (Figure 3). The definition is based on a previous regional mapping that utilized the benchmark 30 dBA [16]. The general plan includes a detailed account of how the quiet areas are to be protected. For instance, wind farms, shooting ranges, sawmills, and similar activities are to be avoided.

**Figure 2.** Map of Munkedal municipality, where "quiet areas" overlap with "large unaffected areas". Adapted and translated from Munkedal's general plan ÖP14 [32].

**Figure 3.** Map of quiet areas in the municipality of Hörby and neighboring municipalities. Adjusted and translated from [34].

In Stockholm, as previously mentioned, several noise mappings on regional level have been made from the turn of the millennium onwards. In 2013, the concept "Guide to Silence" was introduced in the region, and so far it has been tested in the three municipalities: Sollentuna, Sundbyberg, and Stockholm [35]. The "Guide to Silence" distinguishes itself as the most outgoing of the initiatives found in Sweden. Accessibility is an important aspect and the concept includes publicizing quiet areas

on the internet and in illustrated brochures. Signs with maps and symbols have been erected at the sites. According to questionnaire responses from one of the municipalities, Sundbyberg, the concept has been used in marketing the municipality and has received a good response from residents.

The City of Malmö was early to acknowledge quiet areas. Noise measurements of designated areas have been made four times since 1998 [36] and the development can be followed by residents on the City's website [37]. The results have been quite disappointing since, even though these areas were chosen for their relative quietness, there are extensive problems with noise. An alternative way of working with quiet areas was tested in 2011, when an artificial quiet space was constructed in an urban square (Figure 4). Noise screens covered with ivy were used to form a small and secluded arbor next to a noise-exposed road. It was found that the arbor improved the soundscape compared with outside the arbor, and that this effect was further enhanced when sounds of nature were played inside the space to mask traffic noise [38]. The results illustrated that it is fruitful to combine noise reduction with introduction of masking sounds, thus confirming the fact that it is not only sound pressure levels which are important when discussing quiet areas, but also the quality of the sound.

**Figure 4.** City of Malmö. A small arbor constructed from noise screens covered with ivy. Sounds of nature were added using concealed speakers and visitors' experiences were evaluated [38]. Photo: Gunnar Cerwén.

Helsingborg was one of the municipalities that participated in the early initiative undertaken by the Swedish Road Administration in the region [29]. In the municipality's current general plan, the notion of quiet areas is mentioned briefly, but it is further discussed in the action plan for noise [39]. A benchmark of 40 dBA (equivalent level) is used for rural areas. For parks inside the city, another benchmark is used, incorporating accessibility as a factor. It is thus stated that residents should have a maximum of 300 m to a green area where at least half the area is below 50 dBA. Based on noise data from the City's parks, accessibility is illustrated in a map (Figure 5). This approach was included here as it illustrates how the notion of quiet areas could be used to ensure variation in a city soundscape. While some urban sounds can be a quality, it is also important to ensure relative quietness [38,40], as this allows residents to choose an environment based on preference, mood, and other factors.

#### *5.3. Reasons Why Some Municipalities Are Not Working with Quiet Areas*

This section discusses the main reason why some Swedish municipalities do not include quiet areas in their planning. Municipalities that answered no in the questionnaire were given the chance to explain why by ticking boxes. A total of 117 such answers were collected. Among the reasons for lack of activity listed in the questionnaire, "No, our municipality does not have a need for quiet areas" (option 2c) was ticked by 22 municipalities. These municipalities were typically located in remote areas with relatively little human activity and hence good access to quiet areas already. Another 59 municipalities ticked the option "No, and it has not been up for discussion" (option 2e). It was

#### *Int. J. Environ. Res. Public Health* **2019**, *16*, 134

found that the reasons varied, as some municipalities stated that their knowledge was insufficient or that it was an ambiguous term. There were also examples of municipalities, like Salem, which thought that it was useless to work with quiet areas due to the high noise exposure in the region. Another 36 municipalities stated that they had started working on the concept, ticking the option "No, but we have discussed implementation in our future plan" (option 2d), but added that it was not yet implemented in the general plan or that they were about to start work on it soon.

**Figure 5.** Map showing accessibility of parks and quiet parks in the City of Helsingborg. Adjusted and translated from the City's action plan against noise [39].

#### **6. General Trends, Challenges, and Future Prospects**

The final part of the study focused on tendencies and challenges, especially in light of future development, legislation, and implementation.

#### *6.1. Confusion Surrounding the Concepts*

There is some confusion about the concept of quiet areas as such, as a few different terms are used interchangeably in the Swedish planning community. The terms "quiet areas" (tysta områden) and "noise-free areas" (bullerfria områden) are the most well-established, but there are also other terms, such as "undisturbed areas" (ostörda områden), "carefulness areas" (varsamhetsområden), and "tranquil areas" (lugna områden), that are used more or less synonymously.

There are several other closely related terms in the planning discourse, including "large unaffected areas" (stora opåverkade områden), "radiation-free areas" (strålningsfria områden), "consideration areas" (hänsynsområden), "nature reserves" (naturreservat), "green structure" (grönstruktur) and "areas for recreation" (rekreationsområden). As described previously (see Sections 5.1 and 5.2), some of these concepts were confused with quiet areas in the survey responses. This was particularly noteworthy for "large unaffected areas", where it was assumed that these areas were also quiet. For instance, the municipality of Halmstad reported that it had not focused specifically on quiet areas, but that "in the concept of large unaffected areas, the sound environment is naturally included". However, as the municipality went on to state, in two of the areas identified as "unaffected", it was evident that only one would qualify as "quiet". The municipality of Åre, similarly, argues that because a large part of the municipality is protected as a nature reserve, it is probably quiet.

Assumptions such as these could be problematic and confirm the need for a term like "quiet areas" or "noise-free areas" that focuses specifically on the sonic aspects. Some of the other terms, such as "undisturbed areas" and "tranquil areas", do not specifically focus on sound and the use of such concepts may increase the confusion further.

It was found that the notion of "quiet areas" can be overlapped with other planning concepts to provide a more nuanced understanding, as illustrated by the municipality Munkedal [32] and its use of the concept (see Section 5.2).

#### *6.2. Definitions and Identification of Quiet Areas*

Definitions of quiet areas vary widely in Sweden. Most definitions include a reference to a benchmarked sound pressure level, which is derived either from a regional initiative and/or from five different classes of noise exposure developed by a number of influential stakeholders [22]. In Sweden, the noise classes can be said to correspond to what the EU END [1] describes as a "competent authority".

The END distinguishes between two main types of quiet areas, i.e., in "open country" and in "agglomerations". This division was used as a starting point in the present study. It was found that it was significantly more common for Swedish municipalities to work with the concept in open country, suggesting that there is scope to further develop strategies for agglomerations. Such focused initiatives have been undertaken in other parts of Europe [41–44]. Moreover, the findings suggest that it would be feasible to extend the division to three types of quiet areas (urban, urban proximity, and rural), in order to account for areas that are in close proximity to agglomerations. Such a division would allow for a more nuanced and contextual use of quiet areas, as exemplified in the five noise classes discussed in the study [22].

This study confirms what has already been noted in Europe [24], i.e., that approaches and definitions vary extensively. This is not necessarily a problem given that various contexts set different demands. However, the ambiguities surrounding the concept were reported as being problematic by some municipalities, resulting in a need for clarification. An overview of approaches has previously been published on EU level [24], and the present paper contributes a Swedish perspective. It provides an overview of current methods and approaches in use in Sweden, and this can hopefully provide some guidance for future work. However, an independent evaluation of the many methods that are currently being used would be useful for future research.

The overview on EU level [24] describes four main types of methods to identify quiet areas; noise calculations, noise measurements, evaluations by experts, and evaluations by users (e.g., interviews and surveys). No specific recommendations are presented, except that a combination of these four methods should be used. All four methods were found in Sweden and at least two of the methods were simultaneously put into practice in most cases. Most typically, benchmarks of sound pressure levels were combined with expert evaluations. The experts typically contributed knowledge about aspects such as land use, accessibility, and natural and recreational qualities. Of the four methods, user participation emerged as least commonly applied, although there were examples of its use, e.g., in Upplands Väsby, where interviews were used to identify quiet areas. In future implementations, it is possible that users could be involved in a more cost-effective manner through use of digital tools like smart phone applications (see Section 6.4).

#### *6.3. Maintenance and Enforcement*

While there are many approaches for identifying and mapping quiet areas, there are fewer examples in Sweden of active maintenance and enforcement of quiet areas. In many cases where mapping of quiet areas had been performed, it has been used mainly as an inventory, possibly with a recommendation to avoid further exploitation in the areas identified, if possible. For instance, the municipality of Växjö reported that it had been 15 years since it performed its inventory and that it had not made any follow-up since then. Sävsjö reported similar experiences, stating that "quiet areas have been mapped, but other than that, it is not something that has come to concrete usage when we make our detailed planning".

There were some examples in which quiet areas are explicitly protected in the municipality's general plans, sometimes with clear instructions on the types of activities and exploitation that should be avoided. One such municipality is Nybro [33], which reported that the designation of quiet areas had been used to limit wind power installations.

The conflict between economic interests, on the one hand, and quietness, on the other, was a recurring theme. In particular, wind power was a question that seemed to upset people, as several such conflicts were noted in the study. The conflicts sometimes extended across regional borders, e.g., the municipality of Grums reported that it had designated an area as quiet, yet experienced disturbance from the neighboring municipality: "One area has been designated (as quiet), but on the other side of the municipal border, that municipality and the county have decided to locate a wind power park (12 mills), because it is in the outskirts and few people are disturbed there."

Another recurring theme, particularly in Northern Sweden, was regulation of snow scooter traffic. Such enforcements were noted in three northern municipalities: Arjeplog, Dorotea, and Kiruna. The municipality of Umeå reported that it had a ban on motor boats but, apart from this, there were few reports of these kinds of restrictions for private citizens in the study.

A related concept to quiet areas, called "consideration areas", was developed rather recently in a collaboration between three Swedish counties [45]. Consideration areas are used in designated archipelagos, where visitors are encouraged to be careful and not make noise. Boats are encouraged to drive more slowly and to avoid noisy activities. The concept does not have legal support, but is based on common will and collaboration. The experiences to date are reported to be good [45].

Another related initiative should be mentioned in this context; quiet sections in trains. Originally introduced in the early 2000s, quiet sections are now established and used extensively in public transportation around Sweden. A quiet section is usually a separate compartment, secluded from the rest of the train. Potentially disturbing activities such as conversations and mobile phones are prohibited inside the compartment, which allows travelers to work, read, or relax.

To conclude, even though there are some interesting examples in Sweden in terms of maintenance and enforcement of quiet areas in Sweden, there is much scope for further improvement as most of the examples are from other contexts. Maintenance was noted in the questionnaire by 20% of the municipalities as being one of the greatest future challenges. Some related ideas and concepts from other contexts that are described here could possibly be used to promote future development.

#### *6.4. Accessibility and Marketing*

This section is related to the Section 6.3 (Maintenance and Enforcement), as both deal with outward activities aimed at the end users. As mentioned, it was found that there was scope for improving accessibility and marketing quiet areas. Mappings of quiet areas described in general plans and other planning documents may be difficult for the end user to access, and it is therefore noteworthy that few municipalities have focused on outward activities. Even municipalities that have come far in their mapping of quiet areas, such as Nybro, reported that there is more to be done in terms of publicizing these areas and making them accessible to the public.

There was a demand for good examples as inspiration in the municipalities' own work. Relatively few such good examples were identified in the study, with the exception of the "Guide to Silence" in the Stockholm region (see Section 5.2). A related initiative has been developed in the UK and is called "Tranquility Trails" [44].

The internet can be an effective way to reach residents, via channels already established by the municipalities and via marketing on independent websites. No dedicated apps for smartphones were reported to have been developed in Sweden, but an application, Hush City, in the EU allows users to measure sound pressure levels, document favorite places, and inform other users about their favorite quiet areas [41]. It is also possible to plot quiet trails in Google maps.

Signs are another way to raise awareness about quiet areas. This is used in the "Guide to Silence" (Figure 6), where each trail's starting point is marked with a large sign containing a map and a short

description of the concept [35]. Moreover, a dedicated symbol is placed at strategic locations. Signs are also used in the related concept of "consideration areas" [45].

**Figure 6.** In the "Guide to Silence" concept, quiet areas are marked by signs. This one reads: "You are now in one of the places chosen by the City of Stockholm for the project "Guide to Silence". The sounds of the city are always present, but relative quietness can be found." Photo and illustration: Ulf Bohman.

The spatial distribution of quiet areas is an important factor for accessibility and one that can be controlled in planning. In Sweden, it is generally suggested that urban dwellers should be at most 300 m (about 5 min' walking distance) from the nearest green area or park [46]. This recommendation has been discussed as a possible benchmark for quiet areas in urban regions [40,47] and has been applied in the City of Helsingborg (see Section 5.2) [39]. Spatial distribution is a factor that could be given more attention, in particular as it lends itself to quantification. However, for rural regions, no corresponding recommendations were encountered. It can be assumed that the quieter an area is, the farther people are willing to travel to get there. Thus, quiet areas in urban regions compensate for the relative noise exposure by proximity, while quiet areas in open country do the opposite. In formulation of future recommendations, it is possible that multifractal modeling could be used as inspiration [48], i.e., that the distribution could be varied depending on type of quiet area and relationships between different types of quiet areas.

Designation of quiet areas is sometimes associated with economic conflicts, as activities like wind power and other industries cannot exist in the proximity of quiet areas. However, there is also the possibility of positive repercussions for the economy if quiet areas are used in marketing. For instance, in the City of Malmö it is clear that the income levels of residents in different neighborhoods correlate with freedom from noise disturbance [49]. Quietness thus appears to be an experiential quality that can be valued economically. There is potential to use quiet areas to create added value by attracting inhabitants and tourists.

A survey conducted at EU level [25] using the Quietness Suitability Index (QSI) has shown that the Nordic region has relatively good access to quiet environments (see Figure 7). The access to quietness could thus be used as a potential tool in tourism-related marketing. However, only a few examples were found where quietness was marketed in this way, suggesting that there is scope for development in this respect. This could result in economic benefits and also serve as a way to highlight and value quiet areas further, thus providing a means by which to protect them in the future.

**Figure 7.** Map of Europe created using the Quietness Suitability Index (QSI), where the potential for quiet areas is based on noise data and land use. Map previously published by the European Environmental Agency [25].

#### **7. Conclusions**

Although quiet areas are becoming rare in some urban regions in Sweden, it can be concluded that there is growing interest in preserving existing quiet areas. It was found that the concept is well-known in the Swedish planning community, but that few municipalities have been dealing with it in a comprehensive manner. This seems to be due to ambiguities surrounding the concept and lack of good examples. Continuous revision and development of strategies for working with quiet areas is suggested. The study identified a total of 16 municipalities in Sweden that have come further in their work than others and experiences from some of these are described.

The results do not indicate that the EU's Environmental Noise Directive (END) has had a significant impact on the development of quiet areas on municipal level. A more important factor in Sweden seems to be various regional initiatives undertaken in recent decades. The municipality's own driving force and inspiration from neighboring municipalities also appear to be important factors.

A number of challenges to future work on quiet areas, in Sweden and at EU level, were identified. In Sweden, many initiatives have focused on mapping and identification of quiet areas, but less has been done regarding maintenance and enforcement. If the principles stated in the recent World Health Organization guidelines on reducing noise while conserving quiet areas are to be followed, more attention definitely needs to be given to such aspects. Some activities are more problematic than others as regards noise emissions, such as snow scooter use in open country, motorsports, and wind farms. In a few cases, activities like these were specifically described in the environmental strategies of some Swedish municipalities as unwanted and undesirable. However, examples of limitations such as these affecting individual activities were fairly rare in the survey data.

There were significant differences in how municipalities work with quiet areas. Such differences are not necessarily problematic per se, as the prerequisites for quiet areas vary depending on context, but it was found that further clarification and guidelines could benefit applicability. In particular, there is a need for an independent evaluation of the methods and definitions available. In the END, a distinction between two types of quiet areas is proposed (agglomerations and open country), but our results indicate that this could be extended to three types: urban areas (agglomerations), areas in urban proximity, and rural areas (open country).

There were very few examples of public communications and marketing of quiet areas by Swedish municipalities, although quiet areas can be beneficial for residents and for tourism, i.e., they are beneficial for health, quality of life, and the local economy. Many parts of Northern Europe, including Sweden, provide rich access to quiet areas and this could play a more prominent role in marketing the country. In addition to economic benefits from tourism, this could also be a way to raise the value and awareness of quiet recreational areas so that they are protected and maintained in the future.

**Author Contributions:** G.C. and F.M. collaborated throughout the project, where G.C. had a more active role. G.C. collected the data and analyzed the results, guided by continuous discussions with F.M. Both authors contributed to writing the paper, G.C. provided the majority of the writing.

**Funding:** This research was funded by The Sound Environment Center at Lund University, through the project "Quiet Areas in Sweden" (Tysta områden i Sverige).

**Acknowledgments:** The authors wish to thank Eja Pedersen and Magnus Lindqvist for valuable comments during the project. Moreover, the authors thank all municipalities that participated in the study, particularly those that took part in the follow-up study and those that granted the use of images.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

International Journal of *Environmental Research and Public Health*
