Content uploaded by Galen Newman
Author content
All content in this area was uploaded by Galen Newman on Jan 10, 2020
Content may be subject to copyright.
2019 ― Volume 11.02
Research Journal
Perkins and Will is an interdisciplinary design practice
offering services in the areas of Architecture, Interior
Design, Branded Environments, Planning and
Strategies, and Urban Design.
Editors:
Ajla Aksamija, Ph.D., LEED AP® BD+C, CDT
Kalpana Kuttaiah, Associate AIA, LEED AP® BD+C
Journal Design & Layout:
Kalpana Kuttaiah, Associate AIA, LEED AP® BD+C
Acknowledgements:
We would like to extend our appreciation to
everyone who contributed to the research work and
articles published within this journal.
2019 ― Volume 11.02 Research Journal
04
Sound Parks: Invisible Agents of Urban Well-Being
Maria Debije Counts, ASLA, maria.counts@perkinswill.com
Galen Newman, PhD, ASLA, APA, gnewman@arch.tamu.edu
Abstract
Urban areas are often times subject to elevated levels of noise pollution. Urban noise levels exceeding 55 decibels (dB)
can result in negative public health outcomes through chronic/long-term exposure. The design of urban open spaces
and parks has been shown to help decrease noise pollution. What remains unclear is to what extent parks attenuate
noise pollution and what design factors best lower noise levels. This research compares measurements of sampled
noise in four urban parks in New York City: Paley Park, Bryant Park, Washington Square Park, and Brooklyn Bridge Park
Pier 3-4 Uplands. Auditory conditions were measured using a combination of advanced digital measuring devices and
then displayed using 2D and 3D information visualization techniques. Results show that different design strategies
account for a 22 dB reduction of sound, on average. Moreover, the sound is reduced around 1dB per every 5 foot of
linear space within the interior of urban parks. Findings from this research imply that the sound measurements should
be included in the inventory and analysis phase of the design process. Strategies are suggested to be employed into
future designs to best integrate sound into future design concepts and schemes.
Keywords: soundscapes, public health, noise, site design, urban landscape
1.0 Introduction
Noise pollution can be understood as any unwanted
sound. It has been found to contribute to negative human
health impacts and the degradation of occupiable
spaces. In fact, noise pollution is considered one of the
primary sources of pollution in contemporary urban
environments1. In 1972, the World Health Organization
(WHO) declared urban noise as an ofcial pollutant,2
due to its negative effects on human health3. If exposure
to noise is chronic and exceeds certain levels, then
negative health outcomes including annoyance, sleep
disturbance, cardiovascular disease, and impairment of
cognitive performance in children can result4. While there
is a growing number of contemporary electric-powered
mobility and transportation options that are quieter
than traditional automobiles, the largest contributor
to environmental pollution remains noise emanating
from traffic,⁵ the dominant mode of transportation
within the urban context. In contrast, parks can help
counteract urban noise pollution¹ and thus, improve
not only the experience of, but factors contributing to
public health in cities where decibel levels are not the
sole indicator of noise pollution—quality also matters.
Moreover, access to and time-spent sensing natural
sounds have been linked to increased health and well-
being⁶. The design elements of parks and the degree to
which they perform as noise mitigation, however, remain
relatively understudied and misunderstood. This study
asks to what extent can urban parks attenuate noise
pollutants, and what design factors have the ability to
naturally lower decibel levels, and generate healthy and
acoustically comfortable soundscapes?
45
Sound Parks
To answer these questions, we focused on testing decibel
levels and their associations to park design elements in
four exemplary, but typologically different parks. The
sites under investigation are located in New York City:
Washington Square Park, Bryant Park, Paley Park, and
Brooklyn Bridge Park Pier 3-4 Uplands. All case studies
are highly urban parks with signicant unhealthy context
noise levels above the WHO’s threshold for the onset of
negative health effects from environmental noise of 55
decibels⁴. Through our analysis, we evaluated nuisance
noise sources and noise reduction design elements
through comparing decibel levels within and outside of
each park, as well as their degree of change.
At least one major contributor to environmental noise
pollution, such as urban trafc (road, rail, and air) that
has been primarily linked to public health issues related
to urban noise,⁷ was evaluated at each site in terms of
the extent to which it was attenuated by the landscape
design. As green spaces have been proven to have a
positive effect on noise pollution at the local scale8 and
have been shown to contribute to numerous positive
public health outcomes,9 we selected only publicly
accessible spaces that are considered urban parks. In
addition, all selected case studies are located within
densely populated areas, and serve as advantageous
sites for investigating and testing sound uctuations
and how landscape design can play a role in effective
altering these changes10.
1.1 Existing Landscape Methods in
Noise Mitigation
Understanding a particular location through its
soundscapes or acoustic environment—all of the sounds
audible to a person in a given location11—is an under-
represented eld of study in landscape architecture12.
This presents a growing need for landscape architects
to develop methods for evaluating elements in the
landscape that impact experience, well-being and
opportunity to design with and around sound. While
a variety of noise mitigation walls such as outdoor
sound curtains, absorptive panel systems and a variety
of other noise barriers exist, their applicability, scale
and design are not typically suited to the scale of the
urban park. More appropriately scaled for landscape-
based solutions with proactive noise mitigation include
a variety of interventions that range from planting en-
masse to earthworks, such as sound berms and structural
designs barrier congurations and hybrid interventions.
For example, Amsterdam’s landscape at Schiphol
Airport, designed by West 8, included four “layers”,
including runway verges, green route, inll planting and
visual access to mediate the soundscape. To mitigate
noise, the scheme included over 80 acres of park area
with grassy hedges and pyramid-shaped landforms
that trap the soundwaves and signicantly reduce the
airport noise13. In the case of rail, high-performance
materials are able to dampen the concentrated noise at
the source of the infrastructure associated with rail lines.
The ground can also attenuate noise, depending on how
it is shaped, its overall size, and the distance it is from
the source of sound. In Western Europe, noise barriers
and earth berms have been used to mitigate noise along
railways and highways since the 1970's, especially when
located near existing residential neighborhoods. Urban
parks have also been found to be effective tools for noise
mitigation14.
These examples reveal environmentally-based noise
reduction strategies; however, they mostly remain
focused on planning-scaled efforts, and due to their size
and approach, are not appropriate for most urban park
designs. Sounds that emanate from the landscape vary
spatially and temporally15. There is a need for further
investigation into how these solutions and others can
be re-appropriated to t the urban pedestrian-scaled
site within the context of the city where environmental
pollution is often well-above the health threshold and
where people seek opportunities for social engagement.
Moreover, noise from car trafc, rail, air trafc and
highways are environmental noise contributors projected
to increase with population growth and urbanization in
the future16. As a profession, the practice of landscape
architecture impacts the “health, safety, and welfare of
the public”17, which can fundamentally make spaces safe
for people, or not. Although sounds are invisible and
often undervalued, they ought to be designed with, for,
and without, in order to truly realize the best possible
environments for people.
46
2019 ― Volume 11.02Research Journal
2.0 Methods
In order to generate a preliminary assessment of sound
sources and how they are mediated through landscape
design in urban parks, recording techniques involving
both direct-point-source and binaural recording
strategies were employed in this research. This approach
involves collecting sonic data within urban park settings
through eld recording devices, followed by investigating
sound frequencies and other variables to assess overall
soundscape quality of each park’s performance as a
sound mitigator. While there are a growing number of
sound capture devices and technologies available to
record and assess outdoor environments, a growing
body of knowledge in landscape research for human
well-being and a growth in the technologies available to
better measure theses associations, a large gap remains
in the current literature. Soundscapes, the human
perception of the acoustic environment,18 are subjective
by nature. To increase objectivity, we used a multi-
combinational model for measuring and visualizing
sound to evaluate each park site’s performance at the
scale of individual station points. For each point, we
captured sounds using technologies that were capable
of recording on-location, within occupiable zones, in 3D,
and at the level that sounds become audible.
Field recordings were recorded at their source as
individual clips as well as lengthier recordings of the
larger acoustic environment. As a meas to establish
a context baseline, areas around the park were also
measured. Interior station points representative of
each major programmatic zone within each park were
then selected as noise sampling areas. We analyzed
each occurrence of audible sound and displayed these
outputs using both 2D and 3D information visualization
techniques. Sounds were processed, visualized and
mapped using digital software to provide the ability to
determine the decibel levels, frequencies and overall
characteristics as they relate to audible areas for
people within case studies. The classication includes
geophonies (comprised of non-biological sounds),
biophonies (biological sounds such as birdsong or
talking), and anthrophonies (made by technological
devices), as seen in Figure 1. A comparative analysis and
calculation of the existing conditions to the change in
decibel levels, type and quality within the case study
sites were evaluated to test the extent of the noise
Figure 1: Visual diagram of the soundscape as perceptible by a
human, and classification of those elements found within each type.
Figure 2: On-site field recording devices illustrating different
equipment types and capabilities used for recording landscape
soundscapes both surrounding parks and within each park.
47
Sound Parks
attenuation and unique design elements performing
this function.
It should be noted that low frequencies scatter
unpredictably and need to be partially absorbed, while
higher frequencies typically reflect off surfaces and
travel greater distances. Landscapes that place forms
or related barriers between sound sources and human-
experience zones help to absorb or reect noise. The
impact depends on whether it is a low or high frequency.
Variability in ability to absorb or reect sound is primarily
based on proximity to the source, overall volume and
material composition19. Our study included testing
volume in terms of decibel (dB) levels and frequencies in
terms of Hertz (Hz) over time.
Plants have been found to reduce noise in myriad ways
depending on factors such as massing, density, foliage
and stems to absorb, reect, refract and scatter sound20.
Soil is also able to attenuate sound based on its design
and its permeability. Therefore, planted areas within the
individual park designs were also examined as design
factors that perform to create comfortable acoustic
zones for people in otherwise undesirable and unhealthy
spaces. Other elements, such as walls (embedded and
stand-alone) and water features, were also studied. All
parks were drafted digitally in AutoCAD from measured
base-plans in plan and section-elevation. All major
surfaces, plantings, material changes, architectural
components and features, and site furniture (permanent
and movable) were included. The drawings were drafted
at 1:1, with the ability to zoom in for detailed analysis and
ability to identify individual elements and features, as
distinguishable from larger earthwork or planted form
site choreography but scaled for comparative analyses.
2.1 Sound Recording
While each park varies in contextual spatial conguration
and design program, the amount of change and design
features were examined across all sites using an identical
methodology. On-site eld recording equipment was
used to capture different soundscapes and individual
sound sources to capture both sounds at their source, as
well as sounds as close the human perception of sound
as possible. Recordings were paired with photographs
and video as a reference for temporal conditions. The
individual sound source recordings were conducted with
Tascam DR-05 and DR-40 stereo recorders and Tascam
headsets. For the omnidirectional soundscape recordings,
an Ambeo Sennheiser headset (mic embedded in ear
buds) with Apple iPhone and Sennheiser Ambeo VR 3D
microphone on an Atlas Sound MS-20E mic stand and
Zoom F4 multi-track eld recorder was paired with the
Tascam headphones. Photography and video included
the use of a DSLR Canon EOS 80D camera on Magnus
VT-300 video tripod with fluid head. Windjammers,
including the Rycote Windjammer and Movo WS-G9
outdoor windshield, were added to the recording
equipment to reduce noise. An overview of the selected
equipment can be seen in Figure 2 and the photographs
of recording on-site can be seen in Figure 3 below.
In total, over 20 site visits, over 400 photographs, and
over 100 eld recordings were captured for this research
project. Field recording was conducted on-site, in person,
on days that the parks were being used actively, and on
sunny days where the weather was not inclement.
Direct-point-source recorders were used to capture
individual sounds and their respective properties
and binaural recorders with multiple-channels and
Figure 3: Photographs of field recordings on-site in New York City in Fall of 2018.
48
2019 ― Volume 11.02Research Journal
multi-directional capabilities were used to record
acoustic environments of human occupiable and
programmed areas within each site. This research was,
by design, conducted to capture sounds as closely
to what humans hear with mechanical equipment.
Because decibels are logarithmic, a reduction by 10
decibels is described in the literature equivalent to a 10-
fold decrease experientially21. So, while it may sometimes
seem like a park reduces noise by a few decibel levels,
the impact is perceptually 10 times more powerful for the
human experience when in the soundscape itself.
2.2 Sound Processing
In order to get a baseline for assessing park performance
to mitigate noise through comparative analysis,
captured sound recordings were visualized using a
variety of sonic digital tools. Processing of the eld
recordings make use of different software programs,
and in this study Reaper, Ambeo A-B plugin, ReSample,
Sonic Visualizer, AbletonLive, Praat and Adobe After
Effects were used. Figure 4 illustrates the workflow
from collection through visualization. Sound les were
processed in a way that did not alter the recording, but
instead made visible what is otherwise not observable.
Files were processed to provide a means for playback
compatible with smartphones, laptops and speakers
with both mono, stereo, binaural, and VR capabilities.
The processing was conducted to provide critical data,
needed to assess ndings at each park and within each
sound le.
2.3 Sound Visualization
Clips for processing and assessment were selected based
on their ability to best exhibit at least one instance of
each different type of soundscape within the park; nine
unique locations are illustrated for each park. Sound
Figure 4: Workflow sound processing illustration showing sound capture to processing and data visualization for assessment.
49
Sound Parks
clips were taken and clipped to eight second intervals.
The types of sound images include soundwaves,
formant, intensity, and frequency. For the purposes of
this study, we focused on the ndings for the intensity
and frequency, as they reveal decibel level change
and the visible characteristics of the soundscapes
to assess park performance. Intensity visualizations
and frequencies were used to examine the decibel
levels for each eld test. Intensity diagrams reveal the
range of 0-100 decibels (dB). The frequency analysis of
spectrograms included a range of 0-20,000 Hertz (Hz),
all frequencies that are known to be audible to humans.
2.4 Soundscape Mapping Interpretations
In order to relate loudness and sound-type with physical
site design layout, each sound-test was coded to a
measured and scaled digitized section and plan. Sound
maps were based on grading, planting, and features
using technical drawing information in AutoCAD in
plan, and then converted to section drawings based
on site measurements and 3D information. The overall
site layout, topography, and choreography play an
important role in each park; each are related to the
distance and ability to attenuate noise from their unique
context. Layouts were conducted both in plan (layout)
and section across through each site. The identied
elements were then illustrated in color on top of the
measured black-and-white drawings to indicate the
overall findings of the sound classifications, source
locations (in section) and field recording stations
(in plan). Key plans and a graphic chart were then
generated in order to indicate the overall location of the
elements, as well as the total range of decibel change
from context to inner park. The results are discussed in
the next section.
3.0 Results
3.1. Overview
In each analyzed urban area, the noise pollution at the
street level and overall street experience is well above
what is considered a “normal” or healthy level due to
congestion, construction, and sirens. The parks show
to be spaces that, through their basic design elements
and features, mitigate perceivable contextual noise in
varying degrees of success. Results show that each park
is able to impact the overall park soundscape through
landscape design and features. In all cases, objects
alone are not the primary feature of noise attenuation,
suggesting that, for parks, achieving acoustically
healthy environments is achievable through the shaping
of space, site choreography, and planted form. The
addition of sound elements, such as water features and
healthy ecologies that attract animal species that make
their own sounds, also plays a role in the soundscape of
parks contributing to the overall perceivable sounds and
health within a given area.
3.2 Paley Park
Paley Park, located at 3 E 53rd street in New York,
is a small pocket park designed by Zion and Breen
Richardson Associates, as seen in Figure 5. It was rst
opened to the public in 1967. The park is approximately
4,200 square feet (less than one acre) and situated
between three buildings, with one face open to 53rd
street, a busy one-way (west-bound) street in Midtown.
It is situated between 5th Avenue to its west and Madison
Avenue to its east. The main landscape features of the
park include a quincunx arrangement of 12 Honey Locust
trees (Gleditsia triacanthos), an elevated plaza made of
granite setts, and moveable seating elements including
tables and chairs. The inner-most area within the park
is set roughly sixty feet back from the busy street and
its primary trafc-related noises. However, noise from a
water wall is sampled with the loudest natural acoustic
element in the park, almost matching that of the street
level at 65 dB. While this may seem negative in terms of
sound, the water wall not only serves as a visual amenity
to the park and backdrop to the entire site, but serves as
noise cancellation, mufing street noises, for the interior
of the park. This suggests that the decibel range while
above the healthy zone, does not deter people from
engaging socially and that they type of noise should
be factored to determine its potency in impacting well-
being for people in the outdoor setting.
The decibel level range along the street is in the high
60’s to low 70’s, above the threshold of 55 dB for
contributing negatively to health from noise exposure⁴.
Detected trafc-noises from the context include sirens
from emergency vehicles and heavy trafc. The park is
removed both horizontally and vertically from the street:
50
2019 ― Volume 11.02Research Journal
horizontally, there is an acoustic separation of roughly
30 feet and ve trees, and vertically, roughly two feet
above street level. On average, the interior of the park is
in the 50 dB range. The total reduction in decibel levels
from the source (street) is reduced by 5-18 dB levels in
the interior park spaces. The site is also anked by
walls on the east and west side that serve as vertical
lawns covered in English Ivy (Hedera helix), creating an
intimate space. Paving comprises just over 70 percent
of the site in the form of rough-hewn unit pavers and
granite pavers. Shrubs surrounding the site make up
roughly 15 percent of the site in addition to the vertical
green walls.
From the street level, the park’s elevation change (mass)
and trees reduce sound in the inner courtyard by roughly
1 dB / 10 ft. The frequency for the most prominent sound
recorded is of the water and occurs in the 7,000- 12,000
Hz range, remaining relatively constant throughout the
park. Therefore, it can be inferred that the water sound
is not attenuated through this design, but instead, that
it serves as a background “white noise.”
Although the decibel range reaches 65 dB at the source
(10 dB above the healthy threshold), there is evidence
of people engaging social interaction as depicted in the
spectrograms in Figure 6. This suggests that users may
not perceive these sounds as bothersome nuisance, and
that they able to engage with one another depending
on the types of sound and programmatic opportunities,
not necessarily by volume or necessarily the threshold
of what is considered to be harmful or not. This implies
that healthy noise levels may be less about a number
and more about a relationship between context and
interior space, or type and quality not just loudness.
While much of the interior areas within the park are
lled with sound from the water wall (roughly 8 percent
of the entire site), which fall within the nuisance sound
threshold, the sound samples reveal that people were
engaging socially in this area, despite the decibel level.
Figure 5: Park Soundscape Mapping Interpretations—Plan and section elevation site-scaled sound maps illustration of Paley Park.
51
Sound Parks
The spectrogram analysis, as seen in gures 6 and 7,
revealed that the water frequencies, when visualized,
were of consistent texture that served as a background
to people chatting and otherwise engaging socially.
3.3 Bryant Park
Bryant Park, located between 40th and 42nd Streets and
5th and 6th Avenues in New York, is an above-ground
elevated lawn surrounded by formal tree plantings, as
seen in Figure 8. The park was originally designed by
Lusby Simpson in the 1930’s, but was recently updated
and re-opened in 1996 with a design by Laurie Olin
and Dennis McGlad. The size is approximately 9 acres,
with the main study area located in the park area west
of the New York City Public Library. The stacks of the
library are located underneath the park. Major urban
arterial streets surround the park, creating nuisance
noises from heavy trafc. The context decibel levels
range in the 70's around the park. A number of train
stops for the underground subway, which can be heard
from street level, also surround the park, inuencing
the soundscape.
As noted, the decibel level along the street are in
the mid-70's range, and include a high volume of
automobiles, multiple underground subway lines, taxis
and sirens. The park itself is removed vertically above
the main street level with several ramps and staircases,
Figure 6: Acoustic Environment Analysis. Decibel levels and spectrogram analysis studies of Paley Park at nine locations ranging from the exterior
to the interior of the park at all major soundscape areas. Numbers correspond with sample location as indicated on the plan, in Figure 5.
Figure 7: Color spectrograms—spectrograms with color filter adjusted to reveal most about the character of the sounds sampled within three
select recordings on-site. Left: Car horn beeping and water from water wall, Center: People chatting with water in background, Right: Cars and
construction.
52
2019 ― Volume 11.02Research Journal
with the most prominent entry on the west side of the
park along 6th Avenue, the busiest of the adjacent
streets. The occupiable path areas within the park
range between a 10-20 feet horizontal distance from
the street, with roughly ve feet vertically to the main
park elevation along the west and north sides closest
to 6th avenue, a gradual elevational change across the
site. The north and south entries along 42nd and 40th
streets are approximately three feet above the street
level, but are less noisy and are one-way streets, with
the New York Public Library to the east portion of the
main Bryant Park green. The total reduction in decibel
levels from the source (street) is +/- 20 dBs, even in
areas with close horizontal distance to the street—
the primary source of context noise.
From the street level, the park’s design using elevation
changes (mass) and trees reduce the upper circulation
routes and seating area noises by 1 dB / 1 ft with 5 feet
elevation and 0.5 dB / 1 ft, where there is approximately
2-3 feet of grade change. See Figure 9 for decibel level
analysis. This suggests a relationship between the
amount of landscaped grade change and attenuation
of context noise. The frequency for the most prominent
sound recorded is of the water fountain, and is 56 dB
at the source in the 12,000 Hz range, dissipating as
one moves away. Because the park is not enclosed, it
does not remain a “white noise” throughout the park.
Sound clips reveal distinctively different characteristics
throughout the park. This is most likely due to the level
of noise mitigation from the elevated nature of the
park, dissipation of the water fountain, and highly
programmed interior spaces that invite opportunities
for social activities. Context noise is approximately 70
dB outside the park, and in the mid-50's inside the park,
making room for other activities, some of which can
be seen in Figures 9 and 10. For example, a harmonica
plays in the foreground while cars honk on the street,
Figure 8: Plan and section elevation site-scaled sound maps illustration of Bryant Park.
53
Sound Parks
as seen in Figure 9. The park essentially inverts the
sounds of foreground and background through its
design. The prevalence of the individual human-made
sounds in higher frequencies reveals that there is
clearly an audible acoustic zone for people to occupy
within the park. Other examples of additional noise
factors include recreational undertakings such as ping
pong, the carousel, juggling, story times, dining, and a
number of individuals appearing to be working-on-the-
go on business calls, working on their laptops or using
mobile devices in the movable seating along the edges
of the park.
Besides the elevational change that doubles as
acoustic separation from the street and room for the
library stacks underground, Bryant Park is known for its
formal arrangement of London Plane Trees (Platanus x
acerifolia), which are located on the primary level of the
park, with the central fountain, and then the main green
(roughly 10 percent of the site), roughly 1-2 feet below
the allee. The main green is surrounded by 300 feet
long garden beds and borders comprised of a variety
of woody shrubs and perennials (roughly 15 percent of
the site), which provide additional acoustic attenuation
from the street in planted mass and material form. The
roughly one-acre green is located in the center of the
park, with the quietest area recorded in the center of
the lawn around 50 dB. Movable seating can typically
be found on and around the lawn. The paths are a
combination of granite and decomposed granite. The
paved areas are comprised of bluestone and gravel
paths (roughly 7 percent of the site). The water feature
makes up roughly 1 percent of the overall site.
Figure 9: Decibel levels and spectrogram analysis studies of Bryant Park at nine locations ranging from the exterior to the interior of the park at all
major soundscape areas. Numbers correspond with sample location as indicated on the plan, in Figure 8.
Figure 10: Color spectrograms—spectrograms with color filter adjusted to reveal most about the character of the sounds sampled within three
select recordings on-site. Left: automobile brakes, Center: Harmonica, Right: Cars beeping from street.
54
2019 ― Volume 11.02Research Journal
3.4 Washington Square Park
Washington Square Park is located in Greenwich
Village, New York as seen in Figure 11. The park is located
at Washington Place between Washington Square
North and Washington Square South, surrounded by
buildings that are primarily New York University. Robert
Moses was the rst to renovate the site into an urban
renewal project in 1934, expanding its size and use from
predominantly the arch and fountain. A temporary arch
was originally built to honor the inauguration of George
Washington, and later a permanent Washington Square
Arch designed by Standford White. The park was most
recently renovated by the New York City Department of
Parks and Recreation, and opened in 2009. While the
arch, fountain and sculptures throughout the park serve
as historical and visual keystones of the site, the planting
around the site, gentle grade change and overall design
scheme provide opportunities for acoustic separation
from the context and multiple soundscapes within it.
The park is surrounded by vehicular streets, with an
average decibel range of around 80 dBs. As one enters
the park, one is immediately surrounded by planting
areas of large beds with large canopy trees to the east
and west, or wide paved paths, all directing to the site’s
center. Most of the higher-level noises emanating from
the street are a result of construction or emergency
vehicles, with one-way traffic surrounding the park
on all sides. Washington Square does not have a
noticeably signicant vertical topographic change but
has approximately 14 feet of gradual grade change from
west to east. The average change in dB from exterior to
interior is 1 dB / 9 ft. See Figure 12 for decibel analysis.
A series of micro-plazas recordings reveal this difference.
In addition, programmed outdoor rooms having a variety
of distinct sounds are collected, such as in Figures 12 and
13 within the park ranging from specied uses such as
playgrounds to the northeast and southwest, to more
exible open plazas spaces where students can be found
Figure 11: Plan and section elevation site-scaled sound maps illustration of Washington Square Park.
55
Sound Parks
rehearsing plays, singing, or other public gatherings and
impromptu musical events. This highlights the number
of distinctly different soundscapes within one site.
These programmed spaces were recorded in the high
30's to low 40's decibel range. This is evidence of areas
that were acoustically comfortable for children and
adults to activate and were located within 80-100 feet
horizontally from the street. The site is relatively uid in
its choreography with 42 percent of the site being paved
and roughly 56 percent of the site being planted with
large canopy trees and planted beds and green lawns.
About 2 percent of the site is the water feature and an
architectural element. The series of paths provide an
acoustical separation from the street noise. With the
exception of the edges of the park where the vehicular
trafc is most audible, the central fountain is the loudest
part of the site that was recorded at 66 dB at the source.
As one approaches the fountain, the noise is loudest. The
permanent seating elements surrounding the fountain
are in the high 30 dB range, and the furthest (besides
being in the actual fountain) from the street noises.
General lawn areas located between the street areas
and the fountain at the center are used as areas for
people to relax on the grass, and in some instances as
off-leash dog areas with an average dB level of 30.
Figure 12: Decibel levels and spectrogram analysis studies of Washington Square Park at nine locations ranging from the exterior to the interior of
the park at all major soundscape areas. Numbers correspond with sample location as indicated on the plan, in Figure 11.
Figure 13: Color spectrograms—spectrograms with color filter adjusted to reveal most about the character of the sounds sampled within three
select recordings on-site. Left: birds chirping, swings, children, and an airplane, Center: Sirens, Right: Children singing.
56
2019 ― Volume 11.02Research Journal
3.5 Brooklyn Bridge Park Pier 3-4 Uplands
Brooklyn Bridge Park Pier 3-4 Uplands is a ve-acre area
within the larger 80-plus-acre Brooklyn Bridge Park
along the East River in Brooklyn, New York. The park was
designed by Michael Van Valkenburgh Associates and
was opened in 2014 along the East River on Brooklyn’s
waterfront as seen in Figure 14. The new Brooklyn Bridge
Park establishes a linear park that relates the park to the
waterfront and serves as an inland edge. It is an example
of industrial urban land converted to a recreational and
leisure space. It is a former inoperative cargo shipping
facility, transformed into a civic landscape.
This research focused on the section of the park that
includes a soundberm, a longitudinal area within the
larger park that was specically designed to mitigate
noise coming from the trafc of the Brooklyn Queens
Expressway and park edge along the newly redened
Furman Street. This multi-level highway runs both north
and south to the east of the park, and its decibel ranges
in the 70-80’s, well above the healthy threshold, as seen
in Figure 15. The embedded soundberm was designed
with a maximum slope of 1:1 and reaches over 30 feet
above Furman Street. It ranges in width from roughly 70
feet to approximately 120 feet. The decibel reduction
from Furman street to the inside of the soundberm is
Figure 14: Plan and section elevation site-scaled sound maps illustration of Brooklyn Bridge Park Pier 3-4 Uplands.
*Raw construction drawing set provided and adapted from Michael Van Valkenburgh, Inc.
57
Sound Parks
approximately 7 dB/ 1 foot. The decibel range along
the path was in the low 50's to mid-60's. The landscape
focuses on attenuating the noise from the context
using the soundberm typology, which can absorb low
frequencies and disrupt high frequencies (which cannot
travel as easily when the sight line is no longer open).
The soundberm not only provides noise attenuation, but
also embeds a healthy plant ecology that is enlivened
by bird songs and people as seen in Figures 15 and 16.
In terms of materiality, the sound berm includes over
80,000 cubic yards of ll, and is made of structural ll,
drainage aggregate, horticultural soil, planting soil,
GeoGrids and jute mesh. It is planted with sedges, forbs
and grasses. A variety of tree species surround the base
of the sound berm on both sides.
To the west of the main path is a landscaped area
that functions as a exible terrace space with movable
seating facing the East River. Between the path and the
interior is yet another reduction in sound by roughly 3 dB
/ 1 ft. This intimate space is dened by a granite stone
wall and shrubby planting; this space has the greatest
evidence of social interaction. The decibel range within
this area is in the low- to mid-40's. Children were found
jumping on the stone walls and in and around the
narrow paths that circulate throughout, while adults
could be heard chatting at individual tables and chairs
placed on the granite terrace. Birds were easily visible.
Figure 15: Decibel levels and spectrogram analysis studies of Brooklyn Bridge Park Pier-3-4 Uplands at nine locations ranging from the exterior to
the interior of the park at all major soundscape areas. Numbers correspond with sample location as indicated on the plan, in Figure 14.
Figure 16: Color spectrograms—spectrograms with color filter adjusted to reveal most about the character of the sounds sampled within three
select recordings on-site. Left: Kids playing and saying “Mom, I got to the top” (of the stone wall), Center: Birds chirping, Right: Kids chasing each
other on the upper terrace.
58
2019 ― Volume 11.02Research Journal
4.0 Discussion and Conclusion
This research seeks to determine the extent to which
urban landscape parks can perform as a noise mitigator
to reduce environmental pollutants and decrease
negative health impacts. The purpose was to determine
how cities can provide healthy parks as an amenity
within the urban context. Through this evaluation,
we exposed the environmental landscape conditions,
components, and design elements that contribute
towards noise mitigation and lowering decibel levels
as well as ndings related to their performance. The
ndings of this study expose a number of site-scaled
design elements that can be used to reduce decibel
levels for urban parks, as well as typological landscapes
promoting social interactivity as evidence of physical
and mental engagement. Table 1 describes the major
ndings of this research. Overall, there is an average
of around 75 dB readings in the noises for the context
for each site. Within each park, there is an average of
around 53 dB of noise. This suggests that the green
space designs account for around a 22 dB reduction
of sound on average across all four parks. However
more important than the aggregate of the reduction of
decibels across each park, is the nding that each park
can reduce decibel levels relative to its surroundings and
through its design. The evidence suggests that through
a wholistic design approach centered around form-
making, site layout, materiality and program, parks are
able to not only attenuate urban noise pollution, but
to an extent that renders the sites usable for leisure,
recreation and as cultural venues that promote well-
being and interactivity. The average decibel change
from exterior to interior of each park is around 1 dB per
every 5 foot of linear space. There appears to be no
clear relationship seen between park size and decibel
change, especially from exterior to interior. This implies
that design elements may be crucial in attenuating
nuisance noises than green space alone.
Findings from this research show that parks that
mitigate noise pollution and introduce healthy natural
sounds through healthy ecologies that not only reduce
decibel levels by design, but also provide areas that
motivate people to engage socially within outdoor
spaces. They achieve this through a variety of general
soundscape design principles. Our data collection
methods involved the use of sound capture devices and
technology that were useful in identifying measurable
impacts of landscape elements. Decibel range and
frequency were recorded, visualized and analyzed to
illustrate the perceivable aspects of a landscape and
Table 1: Summary of research findings for noise reduction capabilities in urban park soundscapes.
PARK NAME
DATE
BU ILT DESIGNER SIZE
PRIMARY
NUISANCE NOISE
SOURCES
PRIMARY NOISE
REDUCTION DESIGN
ELEMENTS
AVG DB
RAN GE
OF CONTEXT
AVG DB OF
PARK
AVG DB
CHANGE FROM
EXTERIOR TO
INTERIOR
Paley Park 1967
Zio n and Bre en
Richardson
Associates
420 0 sq. ft . Vehicular trafc,
sirens
Water wall * in rear
of park, green walls
and trees
60’s-70’s 65 1 d B / 10 ft
Bryant Park 1992; 1996 Lau rie Oli n and
Dennis McGlad 9 acres Vehicular trafc,
trains, sirens
Elevation change
from str eet level,
water fountain*
adds positive sound,
planting, walls
Mid 70’s Mid 50’ s 1 dB / 1 f t
Washington
Square Park 2009
New Yo rk City
Department
of Parks an d
Recreation
9.75 acres Vehicular trafc,
construction, sirens
Water fountain*,
earth, landscape
surrounding park
Hig h 70’s –
Low 8 0’s
Hig h 30’s-
Low 4 0’s 1 dB / 9 ft
Brooklyn Bridge
Park Pie r 3-4
Uplands
2014
Michael Van
Valkenburgh
Associates
5 acres
Multi-levels of
vehicular trafc,
sirens, airplanes
Soundberm,
planting, stone
Hig h 70’s-
low 8 0’s
Low 50’s-
Mid 60’ s 7 dB/ 1 f t
*Water features served to provide white noise in parks, therefore having a positive impact on the site and overall soundscape quality.
59
Sound Parks
lessen subjectivity22. This approach aided in increasing
the understanding of the acoustic elements that either
contribute positively or negatively within a landscape.
Through studying the parks this way, we were able to
isolate the following recommendations for future park
designs and renovations to design with sound:
njEarthworks with volume such as a soundberm
njIntegrating landscape architectural features such as
walls
njPlanted forms such as large trees, shrubs, and plants
massed to provide buffering
njIncreased distance from sound source and occupiable
zones
njStrategic programming of occupiable zones
njIntroduction of water features and other nature sounds
njDevelopment of healthy rich ecologies that invite
birdsong
njElevation shift from sound source
njInclusion of sound elements that can function as
“white-noise”.
As environmental pollutants increase, the complexities
underlaying such an undertaking can be daunting as
social, ecological and technological equity all play a
role in generating soundscapes, especially in publicly
accessible landscapes. As noise increases, populations
grow, and access to nature in the urban context
becomes scarcer, there is a need to design healthy parks
in cities to become ever more important as key players
in providing citizens to healthy soundscapes to promote
well-being. Unlike noise-cancelling windows, for
example, landscape-based solutions change the noise
level for the entire perceivable area, as opposed to only
blocking the noise where the device is installed creating
a healthy acoustic zone for people to occupy.
Landscape-based solutions for noise mitigation are
possible in cities and can be much less costly than other
structural or engineered devices. More economically
conscientious environments can be employed through
landscape design if appropriate noise mitigation
strategies are applied, when compared to purely
engineered approaches. An approach using open
space is often less costly than traditional engineering
or architectural elements, and simultaneously provides
environmental and social benets that are tied to human
well-being and improved public health outcomes.
The extent to which landscapes can mitigate noise
depends on a variety of factors including context, size,
proximity to sound sources, and unique characteristics
and landscape elements. New technologies will most
likely emerge for assessing park noise performance,
further increasing the capabilities to understand and
react to this phenomenon. Landscape architects can
create outdoor spaces where people can hear one
another and are encouraged to pause and experience
(rather than keep moving as one does on the street).
Design work should, then, perform the function of
mitigating nuisance noises for the people experiencing
the space and reducing unwanted and unpleasant
noises. Visualizing and interpreting soundscapes, and in
particular those frequencies audible to humans, allows
us to evaluate, be critical of, and more adequately
describe the architecture and landscape architecture of
our public outdoor domains. However, it should be noted
that one limitation to this research is that the studied
frequencies were only within a range of 0-20,000 for
human perception and did not include a greater range
(although they exist) that could have implications on
ecological and animal health.
Moving forward, a series of strategies should be
employed for future designs to best integrate sound into
future design schemes. First, designers should develop
and integrate a variety of sound recording devices
into the site analysis phase of the design process. As
shown, proximity to sound sources plays a role in the
overall average decibel level for urban sites. The most
accurate devices for the human experience seemed to
be the binaural headset and omnidirectional recorder
set at the level of human condition. Whatever the
method, sound evaluations should be included in future
inventory and analysis phases in design. Second, it is
important to remember that environmental acoustics
can overlap. Therefore, it is most helpful to measure
sound in areas that capture the primary sources found
within that specic portion of the site, not threshold
zones where the onset of other noises significantly
overlap. Third, dB level is not necessarily always an
accurate measure. For example, although Paley Park
was louder than the healthy threshold, the white noise
blocked nuisance noses and provided a type of positive
soundscape background. People were still socializing
and the water (although technically “loud”) was used
60
2019 ― Volume 11.02Research Journal
to reduce noises from the street. This suggests that a
more thorough investigation into not only the dB range
that is comfortable to humans in landscapes is need,
but within different types of parks to include different
program, types of sounds and overall context. Decibel
level, therefore, is not the only factor contributing to
human well-being. In Paley Park, the dB levels fall within
that threshold, but they are of a noise cancelling type,
so in some cases the threshold of 55 can be inaccurate.
Finally, the acoustic environment varies by day and
night, across cases and contexts, from season to season,
and from region to region. Therefore, temporal elements
may play heavier or lighter roles. Further investigations
into these issues is necessary to test to what extent what
decibel range and type of sound would be a factor
across different geospatial climates.
More cases also need to be studied with a greater
range of typologies across parks to get a better sense
of averages across the different types of spaces and
seasons for generating a complete picture of the noise,
sounds, and catalogue of types of sounds that are
applicable to each. Future research should examine what
do within the context of each typology and the inherent
constraints to creating a healthy soundscape. Landscape
designers can use this information to determine design
impacts and how they shape soundscape noise levels
through program locations, planting typologies,
planting mass, landform mass, relationship to context,
and other related aspects with more complexity to meet
the growing challenges of future projects and sites.
Acknowledgements
Funding in part for this project was made possible
through support from the Independent Projects category
of the Architecture + Design Program at the New York
State Council on the Arts with the support of Governor
Andrew M. Cuomo and the New York State Legislature.
Van Alen Institute served as the scal sponsor.
Aly Martori, City College of New York MLA ’19, Graduate
Research Assistant
Yuanyuan Wang, Clemson University, MLA ’19, Graduate
Research Assistant
References
[1] de Paiva, V., Karina, M., Cardoso, M., and Rodrigues,
R., (2015). "Noise Pollution and Annoyance: An Urban
Soundscapes Study," Noise & Health, Vol. 17, No.
76, p. 125-133.
[2] Noise Control Act (NCA), (1972). Noise Control Act of
1972. Public Law 92-574. Identication of Major Noise
Sources, Noise Criteria and Control Technology.
[3] Basner, M., Babisch, W., Davis, A., Brink, M., Clark, C.,
Janssen, S., and Stansfeld, S., (2014). "Auditory and Non-
Auditory Effects of Noise on Health", The Lancet, Vol. 383,
No. 9925, pp. 1325-1332.
[4] World Health Organization (WHO), (1997). World
Health Report: Prevention of Noise-Induced Hearing
Loss, Retrieved on 11/2019 from https://apps.who.int/iris/
handle/10665/65390.
[5] Agarwal, Pradeep Kumar, Shah, R., Siddiqui, M.,
and Agarwal, A., (2017). "Some Basic Concepts for
Mitigating Trafc Noise," Journal of Advanced Research
in Automotive Technology and Transportation System,
Vol. 2, No. 3&4, pp. 26-30.
[6] Wilson D., Pettit C., Wayant N., Nykaza E., and
Armstrong C., (2017). “Multilevel Modeling and
Regression of Community Annoyance to Transportation
Noise”, The Journal of the Acoustical Society of America,
Vol. 142, No. 5, pp. 2905-2918.
[7] Calixto, A., Diniz, F., and Zannin, P., (2003). "The
Statistical Modeling of Road Trafc Noise in an Urban
Setting", Cities, Vol. 20, No. 1, pp. 23-29.
[8] Margaritis, E., and Kang, J., (2017). "Relationship
between Green Space-Related Morphology and Noise
Pollution", Ecological Indicators, Vol. 72, pp. 921-933.
[9] Newman, G., Smith, A., and Brody, S., (2017).
"Repurposing Vacant Land through Landscape
Connectivity", Landscape Journal, Vol. 36, No. 1, pp. 37-57.
[10] Seeman, T., (1996). "Social Ties and Health: The
Benets of Social Integration", Annals of Epidemiology,
Vol. 6, No. 5, pp. 442-451.
[11] Brown, L., , Gjestland, T., and Dubois, D., (2016).
"Acoustic Environments and Soundscapes", in
Soundscape and the Built Environment, Kang, J.,
and Schulte-Fortkamp, B., eds, Boca Raton, FL: CRC
Press, pp. 1-16.
61
Sound Parks
[12] Payne, T., Southam A., Arvanitis T., and Viant
M., (2009). “A Signal Filtering Method for Improved
Quantification and Noise Discrimination in Fourier
Transform Ion Cyclotron Resonance Mass Spectrometry-
Based Metabolomics Data”, Journal of the American
Society for Mass Spectrometry, Vol. 20, No. 6,
pp. 1087-1095.
[13] GeuzeA., and Buijs, M., (2014). “West 8 Airport
Landscape: Schiphol, Scenario 04: Building the
Urban Forest”, Retrieved on 10/2019 from https://
scenariojournal.com/article/airport-landscape/.
[14] Cohen, B., Bronzaft, A., Heikkinen, M., Goodman,
J., and Nádas, A., (2007). “Airport-Related Air Pollution
and Noise”, Journal of Occupational and Environmental
Hygiene, Vol. 5, No. 2, pp. 119-129.
[15] Cohen, P., Potchter, O., and Schnell I., (2014). “The
Impact of Urban Park on Air Pollution and Noise Levels in
the Mediterranean City of Tel-Aviv, Israel”, Environmental
Pollution, Vol. 195, pp. 73-83.
[16] Pijanowski, B., Farina, A., Gage, S., Dumyahn, S.,
and Krause, B., (2011). "What is Soundscape Ecology?
An Introduction and Overview of an Emerging New
Science", Landscape Ecology, Vol. 26, No. 9, pp.1213-1232.
[17] American Society of Landscape Architects (ASLA),
(2018). Professional Licensure: Denition of Landscape
Architectural Practice. Retrieved on 12/19 from https://
www.asla.org/uploadedFiles/CMS/Government_
Affairs/Public_Policies/Licensure_Definition_of_
Practice.pdf.
[18] Aletta, F., Kang, J., and Axelsson, O., (2016).
"Soundscape Descriptors and a Conceptual Framework
for Developing Predictive Soundscape Models",
Landscape and Urban Planning, Vol. 149, pp. 65-74.
[19] Rice, J., Steele, D., Dumoulin, R., and Guastavino,
C., (2017). "A Review of Transport Noise Management
Plans in Large North American and European Cities", The
Journal of the Acoustical Society of America, Vol. 141, No.
5, pp. 3803-3803.
[20] Yang Z., Dai H., Chan N., Ma G., and Sheng P., (2010).
“Acoustic Metamaterial Panels for Sound Attenuation in
the 50–1000 Hz Regime”, Applied Physics Letters, Vol. 96,
No. 4, 041906.
[21] Kim, K., Park, S., and Kweon, Y., (2007). "Highway
Trafc Noise Effects on Land Price in an Urban Area",
Transportation Research Part D: Transport and
Environment, Vol. 12, No. 4, pp. 275-280.
[22] Aumond, P., Lavandier, K., Ribeiro, C., Gonzalez,
E., Kambona, K., D’ Hondt, E., and Delaitre, P., (2017).
"A Study of The Accuracy of Mobile Technology for
Measuring Urban Noise Pollution in Large Scale
Participatory Sensing Campaigns", Applied Acoustics,
Vol. 117, pp. 219-226.
62
2019 ― Volume 11.02 Research Journal
9781734 488104
90000>
ISBN 978-1-7344881-0-4
© 2019 Perkins and Will. All Rights Reserved.
For more information, please send an email to pwresearch@perkinswill.com
