Citizen Science and Soundscape Perception during COVID restriction in the UK

Jim Smith 1 Sustainable Acoustics Ltd., UK Unit 1, The Rural Business Hub, Gander Down Barns, Rodfield Lane, Ovington, SO24 0HS, UK

Peter Rogers 2 Sustainable Acoustics Ltd., UK Unit 1, The Rural Business Hub, Gander Down Barns, Rodfield Lane, Ovington, SO24 0HS, UK

ABSTRACT Winchester Science Centre in Southern England hosts a significant number of informative exhibits pertaining to principles of acoustics; one such example being on the subject of euphony, which seeks to collect evidence from the visiting public by asking questions about their sound environment. In a period of time between the first and third COVID lockdowns, visitors to this exhibit were asked what sounds they noticed more or less of, and what their favourite and least favourite sounds were during this period. The results are shared with two-fold benefit: to assist the public perception of noise and euphony, and to capture the aural response of the public to the 2020 COVID restrictions. This paper explores these results, but also the potential for conducting research from a larger cohort of visitors using a citizen science approach to deliver big data sets, to further explore the perception of sound in the environment with regard to its context.

1. INTRODUCTION

COVID-19 (SARS-CoV-2) was officially declared a pandemic by The World Health Organization (WHO) on 11 March 2020 i . In the UK, A national “lockdown” was introduced on 23 March 2020, replaced in May by specific regional restrictions; additional nationwide lockdowns were introduced in the autumn and winters of 2020 in response to a surge in cases. During these times, where many commercial and transport areas were temporarily shut or partially suspended, the public were also advised/ordered to stay at home where possible. The intention of this regime was, of course, to protect the public from the virus and accelerate its passing, however, there were significant secondary effects; two of which are explored in this paper: the actual change in noise levels generated in the UK, and the perception of sound and soundscape in the UK.

“The Quiet Project” was created in association with the Institute of Acoustics, Association of Noise Consultants and the UK Acoustics Network, with the primary aim is to record environmental sound levels and observations across the UK during the lockdown and to repeat these observations at fixed

1 jsmith@sustainableacoustics.co.uk

2 progers@sustainableacoustics.co.uk

milestones as the UK recovers, and illustrate the impact of lockdown on the external sound environment.

The Natural History Museum has worked with the Quiet Project to provide a piece on this, in which they say “People in cities across the world felt as though the birds were singing louder. In fact, with competing urban noise down by five decibels (60% quieter), the birds were probably actually singing quieter. And this is good news ii .”

One such study from The National history Museum depicts the grand scale of the change in noise using the metric of ground acceleration:

Figure 1 : Change in ground vibrations in the earth during lockdown at different seismic sensor locations for the first three weeks of lockdown. Data collected by British Geological Survey for National History Museum

Winchester Science Centre is a hands-on, interactive, science and technology centre located just outside the city of Winchester in Hampshire. In the second half of 2020, The Winchester Science Centre & Planetarium was open for business, albeit with COVID restriction in place. As part of their recently improved exhibits, there is a significant section dedicated to sound and vibration. A short questionnaire in the form of luggage tags was made available to patrons visiting this major area, which provided four open-ended questions:

• What sounds did you notice more of during the lockdown? • What sounds did you notice less of during the lockdown? • What are your favourite sounds? • What are your least favourite sounds? This paper intends to explore the results and the potential links between objective changes in the environment and subjective response to its soundscape and also explore whether a proper approach in the collection of such data could be useful in design or curation of a soundscape.

This presents an opportunity to communicate the benefits to the public of soundscape designing, to improve their environments.

2. TERMINOLOGY Soundscape - this paper adopts the definition presented in ISO 12913-1 iii as “The acoustic environment as perceived or experienced and/or understood by people, in context iv ” Anthropic is of or pertaining to mankind or humans, or the period of humanity's existence while anthropogenic is pertaining to the origin of man, or anthropogeny.

20% 70 60 50 40 30 20 0 O% 0 120 130 HO 150 160 170 som +” ‘di cena bn itbcation’s Geneciacl nccalentices nas’

Noise – generally understood to be described as unwanted sound, or a particular part of the soundscape as described Euphony – derived from the Greek, meaning “pleasing to the ear”. In this context, “euphonic” is used to describe particular sounds with positive connotations. Whilst “noise” is commonly referred to as unwanted sound, no commonly used word is used for pleasant sound – but it is apparent to everyone that such a concept exists. “ Euphony ” is one example of such a word and may, one day, be used in greater circulation ( Rogers 2019) v . Applying this a euphonic soundscape would be one that is pleasing to the perception of the listener, within its context. Wellbeing is defined as part of health by the WHO in its constitution as “a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity” vi Tranquillity – the concept extends beyond aural stimulus, and is not straightforward to define. It is explored in “The importance of auditory-visual interaction in the construction of “tranquil space”, Journal of Environmental Psychology (Pheasant 2010) vii ”. Paraphrasing the descriptions here, tranquillity can be described as “The quality of restorative environments that enable us to recover our sense of wellbeing, with sufficient sensory stimulation to keep us engaged, whilst at the same time providing an opportunity for reflection of relaxation”. It is usually associated with the lack of anthropogenic sound.

3. SOUND PERCEPTION & DETERMINING POSITIVE AND NEGATIVE SOUNDS

The Evaluation of a soundscape is complex, and the human response to a sound can be dependent on a large number of factors viii . Factors influences sound perception can be thought of as being divided in a number of ways; this paper proposed dividing factors into 4 categories:

• Objective factors relating to the sound/signal • Subjective factors relating to sound/signal • Objective factors relating to context • Subjective factors relating to context It is considered that it may be critical to consider all 4 of these categories; as analysis of a signal alone will only ever yield information about factors in the first of the category, and will not tell the entire story when it comes to sound perception by the listener in context. Those in blue are quite different in terms of context, but are all often put into the umbrella term of “psycho-acoustics” as they can all influence sound perception, but can’t necessarily be discerned from the signal data. Table 1 below indicates examples of factors that may influence perception (Kang viii ISO 12983-2 iv .

Acoustic factors (can be inferred from a signal)

Non-acoustic factors (cannot be inferred from a signal ) Objective These are generally acoustic parameters, or can be described with acoustic parameters used by

Factors in this category are objective, but cannot be inferred from a physical acoustic signal. They may still

acousticians to describe noise and signals. • Absolute Level (e.g. Sound Pressure

influence an individual’s response to a sound. • Time of Day • Lighting • Humidity • Temperature • Environment – visual, landscape, architectural characteristics • Distance from source • Direction of source

Level, in dB) • Spectral Shape (including strength of

harmonics). • Temporal variation in either/both of

the above. Intermittent/ Impulsive. • Duration of sounds or “noises” ; or

SEL • Signal-to-noise ratio (if there is, in

fact, a clear “signal”) • L N values • Combinations of the above (e.g.

occurrences of maximums over a certain threshold in a certain band). • Reverberation / Reflections • Background noise / other types of

sound • Direction (Head Related Transfer

Function)

Subjective Factors in this category relate to the content of the signal itself (not the context), but in a way that cannot be described physically. • Natural sounds – do we have instincts

This category contains factors that cannot be discerned from the signal, and are entirely to do with context in which it is heard and psychology of the individual. • Mood of individual • Cultural background, mindset

that relate these sounds to survival? Does that make them more pleasant, even at higher levels? • Comfort / Threat (masking noise vs

towards noise (upbringing/ social/demographic factors) • Expectation – is the individual

silence?) • Association / connotations – for

“ready” for the noise? • Perceived distance from source – does a source further away feel louder than one at the same level, closer to the receiver? • Perceived direction of source • Understanding of situation

music and speech, the mind may makes associations and experience emotions that are specific to the individual, and significantly unrelated to the brief acoustic signal carrying the information. • Response to background noise, or

other sounds around the space. • Familiarity and sound marks • “Loudness”, “Roughness”, “Sharpness” (subjective qualities made into objective qualities)*

causing noise, knowledge of if/when it may end

Table 1: Example factors affecting sound perception

*Note – Attempts have been made in the industry to represent the subjective influence of sound objectively (Fastl and Zwicker 2007) ix .Whilst level of sound is related only to the physical property of the sound, Loudness refers to the measurement of subjective sound

strength, often measured in sones, and a dominant metric in psychoacoustics. “Roughness” is the human perception of temporal variations of sound, measured in aspers. “Sharpness” is linked to the spectral characteristics, and is measured in acums The above model is one possible way to visualise the factors affecting sound perception in a different way. With so many different factors influencing how sound, noise and euphony are perceived, objective analysis becomes difficult, yet it is evident that some type of objective analysis is necessary in order to optimise spaces for the highest levels of wellbeing where sound is concerned. With the above in mind, there is no obvious technical objective criteria for what makes a universally “bad” soundscape; however, it is widely documented that noise can have a negative influence on health ; and it is with information like this that subjective terms such as “bad” and “good” demand objective criteria in order to prevent detriments to health. “Listening to Cities” published by Francesco Aletta reports that, In Europe, long-term exposure to environmental noise causes 12,000 premature deaths and contributes to 48,000 new cases of ischemic heart disease yearly, with a colossal 22,000,000 people in Europe suffering from chronic noise annoyance x .

4. NATURAL AND ANTHROPOGENIC SOUND Studies have shown that Soundscapes influence health, experience, and quality of life, both for the better, and the wor se: Unwanted sounds, most typically noise from infrastructure and other technical sources, have been associated with negative effects on health (Basner et al., 2014 xi , WHO, 2011 xii ), including stress, cardiovascular disease, and sleeping disorders. However, sounds may also have a positive effect on health, for example it has been shown that sounds from nature can aid recovery from stress (Alvarsson et al., 2010 xiii , Wallergard et al., 2013 xiv ). ‘Euphony’ relates to pleasantness of sounds. It is noted that an increase in the perception of euphonic sounds, due to lockdown restrictions, does not necessarily suggest new sounds have been added, rather, and may suggest that simply noisier sounds have been taken away. With regard to traffic noise, the following are considered:

• The source of outdoor noise worldwide is mainly caused by machines, transport, and propagation systems. xv xvi • High noise levels can contribute to cardiovascular effects in humans and an increased incidence of coronary artery disease. xvii xviii • Noise poses a serious threat to a child’s physical and psychological health, and may negatively interfere with a child's learning and behaviour. xix • Traffic noise is one of the most commonly experienced sources of noise pollution; xx • Continuous exposure of people to road traffic noise leads to suffering from various kinds of discomfort, thus reducing appreciably the number of their well-being elements xxi And it is considered possible to logically infer that, in an urban environment with significant traffic flow, improved soundscape design has the potential to increase happiness and improve quality of life and health .

Being a significant hypothesis, this should be tested by first identifying a means to measure soundscape quality pertaining to traffic, and then undertaking subjective surveys of affected individuals, before and after works to an area to lower the level of traffic; under known controlled conditions.

With regard to natural sounds, the Natural History Museum generated the following graphic to illustrate an increase in wildlife sightings and activity during the initial lockdown ii.

Figure 2 – Change in Animal activity, UN report

The findings suggest a link between a reduction in human activity and an increase in wildlife activity. Wildlife activity is not inherently linked with euphonic sound, but natural sound is a consideration in the following study.

5. RESULTS FROM QUESTIONNAIRE / AURAL RESPONSE TO COVID RESTRICTIONS

It is assumed from the data provided by the centre that over 300 people returned the questionnaire; fields allowed for multiple answers so it is not understood exactly how many people were in the dataset. The table below summarises the responses received

Question Responses Unique Answers Sounds types that have increased 528 64 Sound types that have decreased: 494 45 Favourite sounds 396 93 Least favourite sounds 384 109 Table 2: Summary of result response numbers Of the 300+ people who took the questionnaire, there were many different responses. For the purposes of meaningful analysis, opinions that are not widely shared have been categorized as “other”. Two lists were generated for favourite and least favourite types of sound. The lists use only types of sound

Which animals did we see more of during lockdown? os Change in cnima sightings compared to lost year ees aa ps eo 66% B% I elt! Hedges, son x 433% | Gc ®. +13% | +18% Sy Rot NE Do edo Rr Te podis38-117 28 as 270 Nts ect hososnc pation Swen gp oct OOD econ ts Oneson seen ed “Gccniomosite

that have been identified by 1% or more of the sample of the public, and are 20 items long and 15 items long respectively The top 10 most popular ‘favourite’ and ‘least favourite’ sound types were found to be as shown in table 3 below:

Favourite Least Favourite 1 Birds Screaming 2 Music Shouting 3 Waves Crying 4 Laughing Nails on chalkboard 5 Rainfall Flatulence 6 Cat’s purr ‘Anything loud’ 7 Flowing water People eating 8 Silence Traffic 9 Singing Thunder 10 Whalesong Fork scraping on china Table 3 – 2020 to 2021 results from Euphony exhibit The following graphic shows the top 20 most brought up sounds in terms of number of times mentioned when the lists are combined. • Where the bars on the chart have been coloured green, these sounds appear in the list of

favourite sounds. • Where the bars on the chart have been coloured red, these sounds appear on the list of least

favourite sounds • The single yellow bar represents a type of sound that appears on both lists. NB – 3 sound

types appeared on both lists – dog, whale, car (note car can be interpreted as traffic)

| Phinna. eeu I. Uys liiiiiitiitBy) Gani _— a Pribarags

Figure 3: Key types of sound perceived as changing during lockdown, in terms of percentage of people surveyed who mentioned it; with positive values representing a perceived increase, and negative values representing a perceived decrease. Green bars represent sounds identified as favourable, red bars represent sound identified as unfavourable.

The following table provides insight into what proportion of the types of sound that mention in each question pertained to natural or unnatural sounds. For the purposes of this paper, any sounds generated by electronic or transport related sources are considered unnatural and anthropogenic.

Question Natural non-human sounds

Natural – sounds (including human ) Sounds types that have increased 56.8% 82.8% Sound types that have decreased: 3.9% 34.4% Favourite sounds 55.3% 84.1% Least favourite sounds 18.6% 63.9% Table 4 : Sounds identified as natural or man-made from survey results For the purposes of the above – music is considered a natural human source when considering favourite sounds; but an unnatural sound in terms of being noticed during lockdown (assuming the majority notice it via media). When reviewing the data from the surveys, several judgement calls had to be made as to what constitutes “natural” or sound and “natural human” sound. For the entry “anything loud” – this was ambiguous, so it was considered to be 50% loud natural sounds and 50% loud unnatural sounds

6. OBSERVATIONS FROM RESULTS – BEYOND THE WAVEFORM

From table 3 it is noted that the fork on plate, nails on chalkboard are examples of a human annoyance (specific frequency amplified due to human ear xxii . Observing figure 3 it is quick to see the effect on soundscapes was largely positive. The perception was that there were more of the “favourite sounds” (indicated by the green colour) during lockdown, and less of the “least favourite sound” (indicated by the red colour). The most dramatic of these two examples are bird noise, and vehicle noise respectively. There were three notable examples where this worked the other way – a favourable sound which was noticed less during lockdown (laughter) and two unfavourable sound that was noticed more (family members, crying/moaning).

Interestingly, there are a few examples (indicated by simultaneous upward and downward bar) of sound types that the sample of people identified as being both more present and less present during the lockdown period.

From table 3, it can also be seen that, when considering people’s favourite sounds from the sample group, a majority of them are sounds that occur in nature, and this is the case whether or not human sounds are included. Of the sounds that were perceived to have increased during the lockdown, a majority were also natural sounds, and whilst this alone doesn’t conclude that it is the favourite natural sounds that have increased, it provides a reasonably strong evidence for this, when observed in conjunction with Figure 3.

On the contrary, when considering least favourite sounds, 81.4% are human sounds (36.1% unnatural sounds, and 45.3% are human-based natural sounds). It is evident from Table 4 that the vast majority of sounds that decreased were either unnatural sounds, or natural human sounds (96.1%).

7. FURTHER CONSIDERATIONS All the sounds mentioned and considered in the analysis of the questionnaires are done so with no distinction between short-term events that draw focus for a passing moment, or continuous background noise that is ongoing; it may be that these types of sound are “apples and oranges” when it comes to how our brains process them, and consider them as favourite of least-favourite sounds.

Further considerations could consider this temporal quality in some way, and also what is necessary to consider what sound ma ke s us happy . These do not necessarily neatly overlap with sounds we want and need to hear for our greater good (e.g. fire alarms, wake-up calls, noises that assist us with survival, noises that enhance media) .

For example, the short-term event “opening a beer can” appears as a response to favourite sounds, but may not be as favourable if were repeated over and over. This idea could be explored by crafting new questions.

Some of the sounds discussed (not least of all music) have potential to generate an emotional response by connecting us, the observer to:

• Friends/Family • Nature • A memory/emotion • As a prompt/alarm • To “home” or familiarity • To discovery This idea extends its reach, in terms of sound perception, far beyond what may be inferred by acoustic qualities of the sound that can be analysed with a sound level meter of computational software. With personal, anecdotal experience examples of all of these types of connection when observing the questionnaire responses to people’s favourite sounds. One possible inference from this is that it is desirable to feel connected through sound.

8. IMPLICATIONS ON POTENTIAL SOUNDSCAPE DESIGN

The results of the questionnaire are not so conclusive that a one-size-fits-all approach is considered possible for designing and creating desirable soundscapes as there are some tailored, subjective responses, but there are certainly some generally trends that can be taken into account regarding sounds from nature.

Clearly, it is not possible to return to the COVID soundscapes by force of will in order to reap the positive benefits – so it is necessary to consider ways of recreating aural environments where possible. An assumption is that high levels of wellbeing are considered to be a universally desirable end goal with soundscaping design. A well designed soundscape will, importantly, manage

and reduce stressful sounds, or those that the majority would subjectively describe as “noise”, or be objectively be linked to stress; however, it will equally importantly consider and manage “euphonic” sounds where the context is appropriate – especially those which can be considered to generated tranquillity. According to the Environmental Protection Agency xxiii , humans spend >90% of their life indoors. It is easy to see that how important it is to have a pleasant soundscape indoors; but to look at it another way, with only 10% of our time outdoors, it is worrying to believe that many of us are not capitalising on that small amount of time with an intimate, aural connection to natural sounds as would have been the case for our immediate ancestors. Wellbeing could be in the form of :

• Security / Familiarity • Vibrancy • Calmness • Respite • Connection to the world in a positive way • Entertainment/music – enjoyment without damage • Privacy • Stimulation Building new homes continues to be a priority for the Government, and it is important to consider how planning will ensure that people are not only adequately protected from noise, but also situated in placed with improved environments in and around homes, in a sustainable way. Soundscapes are part of the Noise and Soundscapes action plan (2018-2004) xxiv , and in response to the question why noise and soundscapes matter the response is that “tackling noise and improving soundscapes contribute to achieving all seven of the wellbeing goals”.

Noise mapping is being explored in soundscape design to guide the public to higher quality places. Diverse, multi-dimensional, subjective data would not work well on a map, and so it may be necessary to not only gather appropriate, accurate, and concise data, but to funnel wide-ranging subjective criteria through appropriate grading and categorisation to achieve a few “bite-sized” parameters (and possible an overall level) pertaining to “soundscape quality” that can be visualized on a map

9. CONCLUSIONS

The trends captured from the survey data indicate that people tend to prefer sounds of nature over human sound and unnatural sound by a small margin; whereas they dislike human sound over natural sound by a significant margin. The observations show that natural sounds were perceived more during the lockdown period, and unnatural sounds were perceived less. The specific type of natural sound perceived as having the biggest increase was birdsong (perceived as a favourable sound) and the unnatural sounds that were perceived less were largely travel sounds (considered an unfavourable sound). From the previous sections, it is seen that there are a plethora of different factors to consider when looking to understand the perception of a soundscape, and it is therefore important to simplify the design process when considering possible soundscapes. ISO 12913-2 iv implements a descriptive process for analysing soundscapes, which assists in simplifying this process. It uses the following 8 different adjectives as part of its proposed questionnaire to describe soundscapes:

• Pleasant • Annoying • Calm • Chaotic • Vibrant • Monotonous • Eventful • Uneventful

It should be noted that this is 4 pairs of antonyms. To assist in understanding their relationship, one could plot annoying – pleasant , on a x-axis with ‘pleasant’ being the design goal. On another dimension, entirely unrelated, would be eventful - uneventful . The last 4 words then fall in between, in the 4 quadrants constructed, intuitively. As a reference, this can forms the basis of future questionnaires, as an important distinction is made between pleasant eventful sound (vibrant) and pleasant uneventful sound (calm) without being overly complex. Multiple-choice questions can aid in stream-lining analysis, including consideration of the context and psycho-acoustic factors as mentioned in Table 1 of this report and ISO 12913-2; forming a starting point for future surveys pertaining to the subjective response to sound.

10. REFERENCES

i WHO-convened global study of origins of SARS-CoV-2: China Part". World Health Organization . 30 March 2021 ii Nature Liberated by Lockdown, Anthropocine (published September 2020), https://www.nhm.ac.uk/discover/nature-liberated-by-lockdown.html iii ISO 12913 – 1 Acoustics — Soundscape — Part 1: Definition and conceptual framework - 2014 iv ISO 12913 – 2 Acoustics — Soundscape — Part 2: Data collection and reporting requirements- 2014 v Euphony – a case for evolving acoustics language to include positive or valued sound – Rogers and Simpsons 2019 vi WHO Constitution , 1946 vii Pheasant R. J. et al, The importance of auditory-visual interaction in the construction of “tranquil space”, Journal of Environmental Psychology (2010) viii Kang - Urban sound planning – A soundscape approach – (2019) ix Psychoacoustics – Facts and Models, Fastl and Zwicker (2007) x Listening to cities UN Environment Programme Frontiers – Francesco Aletta – (2022) xi Basner et Al - Auditory and non-auditory effects of noise on health – (2014) xii WHO - Burden of Disease from Environmental Noise - Quantification of Healthy Life Years Lost in Europe (2011) xiii Alversson et Al - Stress Recovery during Exposure to Nature Sound and Environmental Noise – (2010) xiv Wallergard et Al - Inducing physiological stress recovery with sounds of nature in a virtual reality forest--results from a pilot study – (2013) xv Senate Public Works Committee. Noise Pollution and Abatement Act of 1972. S. Rep. No. 1160, 92nd Congress. 2nd session xvi C. Michael Hogan and Gary L. Latshaw, "The relationship between highway planning and urban noise", The Proceedings of the ASCE . Urban Transportation. May 21–23, (1973), Chicago, Illinois xvii Münzel, Thomas; Schmidt, Frank P.; Steven, Sebastian; Herzog, Johannes; Daiber, Andreas; Sørensen, Mette (February 2018). "Environmental Noise and the Cardiovascular System". xviii Hoffmann, Barbara; Moebus, Susanne; Stang, Andreas; Beck, Eva-Maria; Dragano, Nico; Möhlenkamp, Stephan; Schmermund, Axel; Memmesheimer, Michael; Mann, Klaus (2006- 11-01). "Residence close to high traffic and prevalence of coronary heart disease“ xix Noise and Its Effects on Children, EPA, https://www.epa.gov/sites/production/files/2015- 07/documents/ochp_noise_fs_rev1.pdf xx verywellmind.com – stress and noise pollution xxi Annoyance from Road Traffic Traffic Noise : A Rreview – D. Ouis, (2001) xxii Fingernails on a chalkboard: Why this sound gives you the shivers- Laura Geggel – (2016)

xxiii EPA - W hat are the trends in indoor air quality and their effects on human health? xxiv Well-being of Future Generations (Wales) Act (2015)

https://www.futuregenerations.wales/wp content/uploads/2017/02/150623-guide-to-the- fg-act-en.pdf