Noise Pollution Workshop
49049 – Air and Noise Pollution
Raissa Gill & Peter Irga
Learning Objectives
1. Calculate and utilise a calibration curve to standardise noise measurements of a smartphone
app to a Sound Level Meter,
2. Perform a noise assessment using a smartphone app and analyse signals to determine level
and nature of exposure,
3. Construct a basic sound pressure model to estimate noise exposure and compare model
estimate to in situ data, and
4. Perform a risk hazard assessment using Australian Government standards.
Background: Noise Assessment
• A noise assessment involves examining the characteristics of noise sources influencing a
location of interest. It involves determining:
• the geographic location of noise source(s) relative to your location of interest,
• its audibility (level, nature, character, and/or quality of sound produced),
• the time of day and duration of effect, and
• the reported effect is has on people/surrounding wildlife.
• Offensive noise is defined as noise that:
• is harmful to a person outside of the premises where the source is located,
• may interfere unreasonably with those persons’ comfortability, and
• exceeds its approved audibility as prescribed by rules and regulations.
State of NSW and the Environmental Protection Authority. (2013). Noise Guide for Local Government.
https://www.epa.nsw.gov.au/-/media/epa/corporate-site/resources/noise/20130127nglg.pdf
Part 1: Practical component
1. Select your sampling site and times
Select a location at your local residence where you can perform noise measurements at 3 distinct times
of day (e.g. 10am, 2pm & 6pm) for at least 15 minutes duration per session.
Your 3 noise samples must be taken on the same day to keep noise influencing factors such as weather
relatively consistent. If the weather suddenly changes on the same day, note this down as it will
influence how you discuss your results in your report.
If possible, select somewhere outside such as in front of your house/building or a backyard. If this isn’t
possible that’s okay – inside will do, just ensure whether you keep doors/windows open or closed is
consistent between samples. Your location must be the same for each sample, so ensure it’s available to
you for your elected sampling times.
Part 1: Practical component
2. Take noise samples
Ensure your mobile device is fully charged and download the phyphox app from iTunes (iPhone) or the
Google Play store (Android). This is a free app developed by the RWTH Aachen University in Germany,
which uses your microphone as a noise sensor and measures sound pressure level received in decibels
(dB, see https://phyphox.org/ for more information).
At your 3 sampling times, open the phyphox app and select Audio Amplitude under Acoustics. Normally
we would begin calibration with a Sound Level Meter, however due to the COVID-19 situation we will do
this part differently in the next section. So for now, just ensure the Amplitude tab is selected and hit the
more options symbol > Timed run. Tick the Do a timed run box, adjust the Start delay (s) to 3.0, change the
Experiment duration (s) to 900 (15 minutes in seconds), and hit OK. This will return you to the recording
screen. When you’re ready, you can select the play symbol and the sound recording will start after a 3
second delay (use this time to put your phone down in a secure spot).
You will know that the experiment has started as the plot begins to fill up and the 900 s timer starts
ticking. Once time is up, the app will stop collecting data and the timer will stop at 3 s. You can then
select the more options symbol > Export data > Excel > OK. Select how you would like to share the file
with yourself (I use Outlook) and follow the prompts to share.
Tip! Ensure your phone has enough
battery in time for each measurement!
Phyphox is a little draining so you might
need to charge in between samples.
1 2 3 4
8 7 6 5
Part 2: Calibrate your noise data
First, a little background
If we were on campus, we would normally calibrate your smartphones directly via the Calibration section
of phyphox using a testo 816-1 Sound Level Meter. This is because smartphone microphones are not
designed to be highly accurate noise sensors, so your data is not necessarily accurate or consistent
across other devices (differences in software, hardware, phone case etc.).
To partially address this limitation, you will be generating a calibration curve using smartphone and
sound level meter data that Dr Peter Irga collected at a range of decibels.
A calibration curve is a regression model used to determine an unknown value of interest based on the
measured, known response of a standard. This is also known as a standard curve, and is typically used in
engineering when cross-calibrating instruments.
So for this workshop, you will use the sound detected by Peter’s smartphone (Samsung Galaxy 8) and by
the testo 816-1 Sound Level Meter to create a standard curve, and use the equation of the line to
calibrate your smartphone noise data.
We are aware that not everyone has the same smartphone device, however this calibration is good
enough for the purposes of this workshop/report. In the real world, we would do this properly!
Part 2: Calibrate your noise data
1. Download the Excel file titled ‘Calibration data’ on CANVAS
This file contains a range of sound pressure levels in dB measured by Peter’s Smartphone (column A) and
the Sound Level Meter (column B).
2. Create a scatterplot in Excel
Highlight the two columns, and select Insert > Scatter to plot the smartphone data on the x-axis, and
sound level meter data on the y-axis.
Part 2: Calibrate your noise data
3. Fit a line of best fit
Click the scatterplot, select [+] on the top right > Trendline arrow > More options. This
will bring up the Format Trendline window. Ensure Linear, Display equation on chart, and
Display R-squared value on chart are selected, and close the window. This will place a
line through your data, as well as give the equation of the line (y=mx+b) and R2 value.
4. Use the equation to standardise your noise data
You can do this by opening your noise data Excel files, and in the column adjacent to
your measurements in dB, use the = function to sub in the data values for x in your
equation and apply to the whole column by dragging down. I.e. Standardised dB =
(gradient x smartphone db) + intercept. An example using mock noise data and
equation of line (don’t use these in your report):
Mock equation: y = 1.59x + 132.87
Part 3: Map noise source(s)
You will now construct a sound level model of your local residence (this is the fun part!)
We will do this via: http://noisetools.net/dbmap/
This is a freely available resource that allows you to perform basic sound level modelling.
But first, you require the following:
• List of potential sources near your local residence – you can use Google Maps and Google Street
View (if you cannot leave your residence), “field” notes, and research (primary/government/industry
sources) to determine current land use and deduce the most likely source(s) affecting your area
within a 2-street radius.
• Estimated sound pressure levels (dB) for your list of potential sources – you can use any reputable
online estimates from primary literature, government and/or industry sources. A place to start:
https://www.safeworkaustralia.gov.au/noise. You will need to include a table of your sources and
estimated sound pressure levels in your report with references to your estimates. If you are unable to
find an estimate (we will check ease of search), you can make an informed estimate/guess based on
your data and the following government report (see next slide):
If you use this, don’t forget to reference:
https://www.epa.nsw.gov.au/~/media/EPA/Corporate%20Site/resources/noise/130127NGLGPart2.ashx
For example, a noise source I
located appears roughly in
between a busy office and
busy city street. Therefore, I
will estimate it as 65 dB
Part 3: Map noise source(s)
• Google Maps image of your local
residence and surrounding area – this
can be the default or satellite variant,
but you must be able to identify
pollutant sources near your local
residence and boundaries of key
barriers (buildings etc). Crop the map
image to approximately 2 streets from
your local residence on its North,
South, East and West boundaries as
per this example, and save it to a
known location on your computer.
Part 3: Map noise source(s)
1. Open http://noisetools.net/dbmap/ and explore the tools.
Do a brief read through and then close the introduction message [1] as you open the link, and click on
each of the tools [2] on the right and read what they do. Try to get an idea of what features lie within
each so you know how to use them later.
SETTINGS
SELECT TOOL
PAN TOOL
ZOOM TOOL
EDIT OBJECTS
ADD OBJECTS
1
2
Tip! You can move or delete objects by
selecting the SELECT TOOL and either
dragging or right-clicking the object and
hitting Delete
Part 3: Map noise source(s)
2. Upload your cropped 2-street map image of your local residence.
Do this by dragging your map image file anywhere onto the gridded map screen [2]. Click the SELECT
TOOL from the toolbar on the right and drag the corners of the grey map to the edges of your image [3].
This space now becomes your modelling field.
3 4
Part 3: Map noise source(s)
3. Add the approximate location of your phyphox smartphone samples.
Do this by selecting ADD OBJECTS > RECEIVER, and clicking the location on the map to drop the point.
Add the height of the receiver location and close the pop-up window [5]. This should now be located
on your map [6].
You will use this receiver to estimate the noise pollution level at that location in the model, and as a
comparison to your in situ measurements.
The receiver cannot be inside a building, so if you needed
to do your practical measurements indoors, select a
location just outside the building
5
6
4. Map key buildings.
Do this by selecting ADD OBJECTS > BUILDING, and clicking the location on the map to start your
square. The BUILDING option allows square/rectangle structures only, for more complex shapes, select
the BARRIER option instead and close the polygon [7].
If you know the building, select the reflection coefficient that describes the feature the best (i.e. the wall
descriptions in blue). Next you need the height of the building in metres.
To get this go to: https://earth.google.com/web/
Once you’ve specified the height in the model,
close the window to complete the building.
Use the Search tool to find your location, and use zoom and
shift-click-drag to get a better angle for measuring height
of your building [8]. When you’re ready, use the Measure
distance and area tool, re-adjust your angle with shift-clickdrag, and measure the height of your building. Press Done
on the white box to the right to stop measuring.
Part 3: Map noise source(s)
7
8
Part 3: Map noise source(s)
5. Add noise pollution sources.
Do this by selecting ADD OBJECTS > POINT and selecting a location.
Change the Sound Power Level to your estimate, estimate the Height of the source, change the
Frequency to 2000 Hz as urban noise is ~1–4 kHz (Potvin et al., 2011), and close the pop-up window [9].
This will apply the noise source to your map and show its effect across the region (read the key) and
the receiver in decibels. Disable the point source by clicking on it with the SELECT TOOL to make the
map visible and repeat this step for your other sources. At the end, enable all the points again by
clicking on them with the SELECT TOOL.
For something like motor vehicles, create a single point source on the road of interest rather than
several cars, and use a dB estimate that reflects a level of traffic.
Potvin, D. A., Parris, K. M., & Mulder, R. A. (2011). Geographically pervasive effects of urban noise on frequency and syllable rate of
songs and calls in silvereyes (Zosterops lateralis). Proceedings. Biological sciences, 278(1717), 2464–2469.
https://doi.org/10.1098/rspb.2010.2296
9
Part 3: Map noise source(s)
6. Add weather variables.
Specify the temperature and humidity of the day you sampled by selecting the SETTINGS tool
and changing the Temperature (°) and Humidity (%) values under CALCULATION SETTINGS.
You can find weather history for your location via the Bureau of Meteorology website:
Go to http://www.bom.gov.au/climate/data/, select the tab
Select using map, click a weather station closest to your
location, Select Recent observations [10], and select the
appropriate month under Other times and other places [11].
10
11
Part 3: Map noise source(s)
7. Save your map
Save your map under the SETTINGS tool > Export / Import > Grid image (PNG).
This will go in your final report. An example [12]:
*Yours will require multiple building features and sources
12
| Plot the classificatio |
noise signals (the calibrated sound pressure level (dB) through time) and classify the 3 signals according to the n system in the Block 1 Noise Pollution lecture (slides 26-28) – you can do this for the complete 15 m signal or break it up if it changes. Ensure you plot each of the 3 signals as a time series – see plotting do’s and don’ts: https://www.clips.edu.au/displaying-data/. |
Report: Structure (1500 word limit)
| Section | Information to include |
| Abstract ~200 words |
*Not included in word limit Brief summary of all sections of the report – just the key points re-worded |
| Introduction ~200 words |
What is noise pollution and why is it important to study? Describe the land use around your study site relevant to noise production State the aim of this study – to perform a noise assessment and compare in-situ data with sound pressure modelling estimates… |
| Methodology ~350 words |
Describe the noise assessment including: • Sampling times and relative location (describe where and when you collected samples) • Brief Phyphox app and calibration methods: “This study utilised the Phyphox app (insert reference) to measure sound pressure levels in dB from a X-model mobile device for 15 minutes. Data was calibrated post sampling against a regression model of sound pressure levels de tected by a Samsung Galaxy 8 Smartphone and the testo 816-1 Sound Level Meter.” * You must paraphrase this statement, but this is the minimum detail needed for this point. • Description of metric calculations used to report results (see results section requirements below for more information). • Brief sound pressure modelling methods, including: • A table which comprises a list of potential sources near the site, associated estimated sound pressure levels (dB) and references used to obtain them; • Weather inputs for the model from Bureau of Meteorology; • Brief description of model methods – what did you do to create the model in dbmap? I don’t need to know the specifics of how each tool functions, just which tool(s)/method you used and for what purpose. |
| Potential pollutant source | Estimated sound pressure level (dB) | Reference |
| Busy street at kerbside | 75 | (insert APA reference) |
| Section | Information to include |
| Results ~350 words |
Your results must report the following: • The extent of noise pollution exposure at your residence: • Describe the level and nature of sound pressure for your 3 samples, complement your text descriptions with figures showing your noise signal as a time seri es for each sample. • What metrics will you use to report and compare this information – the mean? median? maximum? cumulative sum (recommended for at least 1 metric)? These will all tell you something different about the data. If you’re unfamiliar with what they mean, do a Google Search, read about it and select which metric(s) best reflect sound pressure levels at your local residence. • Metrics can be easily calculated in Excel (see the Low-cost Sensing Workshop for methods). • The output of a sound pressure model, including: • The estimate in dB at your receiver; • The GoogleMap of your local residence (used as model input); • The final sound pressure model map of your local residence (would recommend placing these maps side-by-side). |
| Discussion ~600 words |
You will need to discuss the following: • Determine the risk hazard to people at your local residence (also see next slide): Use pages 37-38 in this document: https://www.safeworkaustralia.gov.au/system/files/documents/1702/managing_noise_preventing_hearing_loss_work.pdf, to compare your in-situ data to work safety standards and assess the level and length of exposure to classify sound pressure levels at your location as safe, potentiallyhazardous, or hazardous. You can do this as a complete 15m time series (determine cumulative dB exposure in 15m, compare to page 37) but also break your signal up into different time periods if the signal changes to determine periods of potentially hazardous/hazardous exposure (determine cumulative dB exposure in a time interval(s), compare to page 38). • Discuss whether time of day influences sound pressure levels (from your results), and why/why not (this requires primary literature and government research to determine temporal factors influencing your location). • Discuss the accuracy of your sound pressure model compared to in-situ measurements, ensure to describe possible causes of any discrepancies (this requires primary literature research) and assert which may be relevant to your study and why. Using this information, determine whether the model has likely over-, under-, or accurately estimated sound pressure levels at your local residence (ensure you justify this choice with literature and logical reasoning). • Discuss the potential/existing effect(s) that the detected sound pressure levels (or model estimate if deemed more/less appropriate) has on local resident(s). You should use primary literature, your knowledge of noise pollution concepts from lectures, and/or surveys of local resident(s) to back your points. If you opt to do a survey, ensure you briefly describe methods in the methodology and include the survey in an appendix and the outcomes in a table/figure in the results section. |
| References | *Not included in word limit Mostly primary literature (20+ is good) using APA referencing style |
Report: Marking Rubric
| Weight (%) | Criteria |
| 40 | 1. Disciplinary knowledge & its appropriate application Introduction • Demonstrated understanding of research area and study site, including discerning potential source(s) of noise pollution as a function of land use • Ability to synthesise information and present concise aim of study Methodology • Evidence of practical component completed, including sampling via the Phyphox app (n=3) and performing calibration procedure to ensure quality of data • Logical process of investigating the extent of noise pollution exposure evident across samples (n=3), which is supported by t he correct use of key metrics, tables, and/or figures to describe differences and the level and nature of exposure • Evidence of sound pressure modelling completed, including research of potential sources and associated sound pressure estimat es, ability to obtain and use relevant weather data, ability to obtain height data of structures from GoogleEarth, and correct use of dbmap model components for their intended purpose |
| 45 | 2. An enquiry orientated approach Results • Data is plotted and presented to a high standard, acceptable to the level of a designated engineering publication • Data presentation is relevant to the aim of this research and intended purpose of the figure • Written descriptions are concise, and clearly describe the level and nature of sound pressure levels in the data • Use of quantifiable and comparative language is used to describe difference(s) between samples (n=3) Discussion • Risk hazard of study site adequately assessed using relevant government guidelines • Temporal differences discussed and backed by peer reviewed literature to support/explain findings • Insightful and critical comparison of in-situ data to sound pressure model estimate, which places results in context of peer reviewed literature and assesses accuracy and quality of both data sources, and makes an informed judgement backed by literature and logical reasoning • Describes impact(s) of identified sound pressure levels on humans, which is backed by primary literature and demonstrates understanding of noise pollution concepts References • References cited correctly using APA Referencing style and formatted consistently in reference list • References used are of high quality, from relevant engineering/scientific journals • Every relevant statement or use of evidence is referenced throughout report, with minimum 20 references to peer reviewed literature |
| 15 | 3. Communication skills • The report is insightful, logical and clear, with use of disciplinary language that is engaging and comprehensive • Writing is above all clear and concise • Formatting guidelines are adhered to, with minimal spelling/grammatical errors |
Report: Supplementary information…
“Describe nature of sound pressure levels”
• Plot the noise signals: x-axis = time (minutes), y-axis = calibrated sound pressure level (dB)
• Classify the 3 signals according to the classification system in the Block 1 Noise Pollution lecture
(slides 26-28) – you can do this for the complete 15 m signal or break it up if it changes.
• See plotting do’s and don’ts for presentation of figures: https://www.clips.edu.au/displaying-data/
Sign in with your UTS Student account for access:
• General noise pollution concepts: https://link-springer-com.ezproxy.lib.uts.edu.au/book/10.1007%2F978-1-59259-779-6
• Smartphone-based noise mapping: https://www-sciencedirect-com.ezproxy.lib.uts.edu.au/science/article/pii/S0048969716307501?via%3Dihub
• Data sources relevant for noise calculations (see next slide): Murphy, E. and King, E., 2014.
Environmental noise pollution: Noise mapping, public health, and policy. Newnes.
Reference:
Murphy, E. and King, E., 2014.
Environmental noise pollution: Noise
mapping, public health, and policy. Newnes.
The post 49049 – Air and Noise Pollution appeared first on My Assignment Online.