FNM 2013 - GOV.UK

FNM 2013

Additional Analysis Report

Comparison with previous Noise Attitude, Noise Incidence, and London Noise Surveys.

December 2014

47065784_R2_03

Prepared for: Defra

UNITED KINGDOM & IRELAND

Defra – Focussed Noise Monitoring 2013 – Additional Analysis

47065784_R2_03

August 2015

2

REVISION SCHEDULE

Rev Date Details Prepared by Reviewed by Approved by

1 October 2014

Draft Report David Gerard

Acoustic Technician

Heather Billin

Acoustic Consultant

Chris Skinner

Associate – Acoustics

Paul Shields

Technical Director

2 December 2014

Revised following client comments

David Gerard

Acoustic Technician

Heather Billin

Acoustic Consultant

Chris Skinner


Paul Shields

Technical Director

3 August 2015 Final Report David Gerard

Acoustic Technician

Heather Billin

Acoustic Consultant

Chris Skinner


Paul Shields

Operations Director

URS

12 Regan Way

Chetwynd Business Park

Chilwell

Notts

NG9 6RZ


47065784_R2_03

August 2015

3

Limitations

The conclusions and recommendations contained in this Report are based upon information provided by others and upon the assumption that all relevant information has been provided by those parties from whom it has been requested and that such information is accurate. Information obtained by URS has not been independently verified by URS, unless otherwise stated in the Report.

The methodology adopted and the sources of information used by URS in providing its services are outlined in this Report. The work described in this Report was undertaken between 1

st December 2013 and 31

st March 2014 and is

based on the conditions encountered and the information available during the said period of time. The scope of this Report and the services are accordingly factually limited by these circumstances.

Where assessments of works or costs identified in this Report are made, such assessments are based upon the information available at the time and where appropriate are subject to further investigations or information which may become available.

URS disclaim any undertaking or obligation to advise any person of any change in any matter affecting the Report, which may come or be brought to URS’ attention after the date of the Report.

Certain statements made in the Report that are not historical facts may constitute estimates, projections or other forward-looking statements and even though they are based on reasonable assumptions as of the date of the Report, such forward-looking statements by their nature involve risks and uncertainties that could cause actual results to differ materially from the results predicted. URS specifically does not guarantee or warrant any estimate or projections contained in this Report.

Unless otherwise stated in this Report, the assessments made assume that the sites and facilities will continue to be used for their current purpose without significant changes.

Where field investigations are carried out, these have been restricted to a level of detail required to meet the stated objectives of the services. The results of any measurements taken may vary spatially or with time and further confirmatory measurements should be made after any significant delay in issuing this Report.


47065784_R2_03

August 2015

4

TABLE OF CONTENTS 1 INTRODUCTION ............................................................... 5

Previous Surveys ............................................................. 5 1.1

2 METHODOLOGY .............................................................. 7

Overview ........................................................................... 7 2.1

National Noise Attitude Study 2012 ............................... 7 2.2

National Noise Incidence Survey 1990 and Analysis of 2.3Diurnal Patterns ............................................................... 9

Diurnal Patterns ............................................................... 9 2.4

Strategic Noise Maps..................................................... 11 2.5

Self-Completion Questionnaires .................................. 12 2.6

1962 London Noise Survey ........................................... 13 2.7

3 RESULTS AND DISCUSSION ........................................ 14

National Noise Attitude Study 2012 ............................. 14 3.1

National Noise Incidence Survey 1990 and Analysis of 3.2Diurnal Patterns ............................................................. 37




4 SUMMARY ...................................................................... 75

Overview ......................................................................... 75 4.1

National Noise Attitude Study 2012 ............................. 75 4.2

National Noise Incidence Survey 1990 and The Analysis 4.3Of Diurnal Patterns ........................................................ 76




5 REFERENCES ................................................................ 79

APPENDIX A: NOISE PERCEPTION AND TERMINOLOGY ........... 80

APPENDIX B: DERIVATION OF NOISE INDICATORS FOR ANALYSIS AND OVERALL SITE SUMMARY ............... 82

APPENDIX C: EFFECTS OF 2 MINUTE SAMPING ON LA10 AND LA90 NOISE INDICATORS FOR LONDON NOISE SURVEY . 84

APPENDIX D: LONDON NOISE SURVEY LOCATION DATA AND CURRENT AERIAL VIEW ............................................... 89


47065784_R2_03

August 2015

5

1 INTRODUCTION

This report presents further analysis of the data collected by URS in the Focussed Noise Monitoring 2013 (FNM 2013) project. This project was commissioned by Defra (Department for the Environment, Food and Rural Affairs) via the Government Procurement Service Environment and Sustainability Advice Framework Agreement RM830. Further details of the main monitoring project and the definition of terms used in this report are included in a separate project report

1.

Specifically, this report considers the data collected during FNM 2013 and undertakes further comparisons and analysis in relation to data from the following sources:

(A) National Noise Attitude Study 2012;

(B) National Noise Incidence Survey 1990, (including further analysis of diurnal patterns in noise levels);

(C) Strategic Noise Maps;

(D) Self-completion questionnaires from FNM 2013; and

(E) London Noise Survey 1962.

An introduction to noise perception and relevant terminology is included as Appendix A to this report.

It should be noted that, due to the small sample size, and the sampling methodology chosen, the results of the analyses presented in this report cannot be considered to be nationally representative. However, they do provide a snapshot of noise levels at a number of individual sites.

Previous Surveys 1.1

The FNM 2013 project included measurements of noise levels outside the front of 77 sites in England. These sites were selected from those included in a number of previous noise surveys:

48 sites from the National Noise Incidence Study 2000 (specifically targeting sites also used in the National Noise Incidence Study 1990);

19 sites from the National Noise Attitude Survey 2012;

five sites from the London Noise Survey (1962); and

five sites from the Camden Licensing Survey (2003).

The previous studies are summarised below:

• National Noise Incidence Study 1990 (NNIS 1990)2

In 1990 the then Department of the Environment commissioned a national study of environmental noise levels based on 24 hour measurements outside 1000 dwellings. The study generated objective estimates of the exposure of the population to noise and the pattern of the noise exposure of the population of England and Wales.

• National Noise Incidence Study 2000/2001 (NNIS 2000)3,4 5

During 2000, the Department for the Environment, Food and Rural Affairs (Defra) and the Devolved Administrations commissioned a similar study which involved new measurements and produced new estimates of the overall population exposure and its pattern. NNIS 2000 was based on 24 hour measurements obtained outside 1160 dwellings over the UK, of which 1020 were in England and


47065784_R2_03

August 2015

6

Wales. The majority of these 1020 sites were chosen as repeats or equivalents of those measured in 1990 and used for comparisons against the 1990 study.

• National Noise Attitude Survey 20126

A National Noise Attitude Survey covering the whole UK was undertaken by Defra in 2012 following previous similar studies in 1990 and 2000. A specific question was included in this survey to ask if recipients would be willing to be involved in further noise related research projects.

• London Noise Survey (1962)7

A survey of noise levels throughout central London was undertaken by the Building Research Station in the 1960s.This survey included measurements at locations selected on a grid throughout central London. Due to the available instrumentation at that time, measurements were taken on a sampled basis, with a short sample recorded each hour. The data from this survey are not available electronically, but values of measured LA10 and LA90 noise indicators can be estimated from graphs shown in the project reports that are available for some locations.

• Camden Licensing Survey (2003)8

A targeted noise survey was undertaken by the Building Research Establishment (BRE) for Defra in 2003 within the London Borough of Camden in areas where noise from licensed premises may contribute to measured levels (either directly or due to local activity). This survey was completed

prior to the liberalisation of licensing laws9.

Where it was not possible to repeat the FNM 2013 measurements in exactly the same place as the previous survey (exact repeat sites), monitoring was undertaken at Acoustically Equivalent Sites (AES).

Full details of the site selection procedure can be found in the FNM 2013 project report. However, the following key factors were considered in identifying potential acoustically equivalent sites:

same local road and similar distances to other nearby roads;

orientation, height and kerb to facade distance;

matching of screening (for example because of hedges, fences and walls); and

any other acoustic factors/sources.


47065784_R2_03

August 2015

7

2 METHODOLOGY

Overview 2.1

This report considers relationships between the measured FNM 2013 sound environment and the following data sources:


(B) National Noise Incidence Survey 1990, including an analysis of diurnal patterns;

(C) Strategic Noise Maps

(D) Self-completion questionnaires from FNM 2013

(E) London Noise Survey 1962

Statistical tests have not been carried out on the data as this is not justified by the small sample sizes and sampling methodology used.

The following subsections give details of the approach taken to the analysis for each of these items.

National Noise Attitude Study 2012 2.2

2.2.1 Scope of work

Where possible, correlations between subjective responses to the NNAS 2012 questionnaire and measured sound levels from FNM 2013 have been considered.

The following analysis has been undertaken:

A review of questions included in the NNAS 2012 responses to identify those expected to be related to level of satisfaction with the sound environment.

An assessment of measured sound levels from FNM 2013 versus satisfaction with the sound environment (from NNAS 2012).

Since there were no questions documenting specific noise events at particular times of day (other than the broad periods of day/evening/night), it was considered analysis of the FNM 2013 1-hour data would not provide additional information for either detailed analysis of individual sites or the entire set of sites. Hence this analysis has been limited to longer duration (e.g. 12-hour day, 8-hour night) noise indicators.

2.2.2 Review of NNAS 2012 questions

An initial review identified questions within the NNAS 2012 which were expected to be related to level of satisfaction with the sound environment and likely to be quantifiable from the long term noise descriptors.

The NNAS 2012 survey allowed limited responses to most questions, on a sliding scale or from a range of word cards.

NNAS 2012 asked questions about a variety of noise sources, however in this analysis only questions concerning road traffic were selected. The FNM 2013 monitoring locations were at the front façade of properties, which are usually exposed to road traffic and therefore noise levels are likely to be dominated by road traffic. Road traffic noise was also the most common dominant source of noise noted by survey teams during site visits across the FNM 2013 monitoring locations.


47065784_R2_03

August 2015

8

Similarly NNAS 2012 asked questions about annoyance from specific road traffic noise sources (e.g. brake squeal, engines revving and music from vehicles), road types and vehicle categories. As it is not possible to differentiate between these categories within the noise measurements, these detailed results were not considered in this analysis.

The following are the NNAS 2012 questions identified as expected to be related to level of satisfaction with the sound environment and which in turn are likely to be quantifiable from the available noise level indicators.

The question numbers shown are those used in the NNAS 2012.

A7 In general, how do you feel about the amount of noise around here?

A8 How sensitive would you say you are to noise?

A9 When you are at home, to what extent are you personally bothered, annoyed or disturbed by noise from road traffic? This question is followed by clarification questions about whether road traffic noise is heard or not.

RTN3 Which five of the words that I am about to show you best describe how the noise you hear from road traffic makes you feel? (from a list of 21 emotion words)

RTN4 And does road traffic noise interfere with any of these aspects of your home life? (from a list of 11 activity words/phrases)

RTN5 Does road traffic noise particularly bother, annoy or disturb you, at home, at each of the times listed on the card? (Categories for day, evening and night periods both during the week and at the weekend)

RTN6 How much would you say you are bothered, annoyed or disturbed by the road traffic noise while it is going on?

RTN7 And how much do you feel that road traffic noise spoils your home life, in general, not just when the noise is going on?

O1 Taking all these noises [heard at home and discussed during the survey] together, I would like you to look at the statements on the card and tell me which one best describes the extent to which noise spoils your home life?

O2 Over the past year, while in this home, have you been kept awake, or woken up, or changed the time when you go to bed or get up, because of … noise from outdoors…?

O6 Thinking now about the whole home, not just the bedroom, do you ever close the windows, or keep the windows closed for … Noise coming in through the window…?

2.2.3 Selection of FNM 2013 data

The NNAS 2012 questions identified above included details relating to time of day with responses for day, evening and night. Therefore the data selected for analysis for each site were the long term noise descriptors Lday, Levening, Lnight, LAeq,16hr, LAeq,24hr and Lden.

The sites analysed were the FNM 2013 measurement locations which were chosen from the NNAS data set and are shown in Table 1. The FNM 2013 survey selected Acoustically Equivalent Sites (AES) where the measurements were not possible at the exact NNAS address, therefore the noise measurements can be considered equivalent to those at the exact NNAS addresses. For the purposes of this work package an initial analysis was carried out considering exact repeat addresses and acoustically equivalent sites separately. However no significant differences were observed between these categories, and so the data are presented across all 19 sites.


47065784_R2_03

August 2015

9

TABLE 1. NNAS 2012 SITES ALSO SURVEYED IN FNM 2013

Area Number of Sites

Type of site

Exact Repeat Address Acoustically Equivalent Sites

London (Kingston Upon Thames)

2 1 1

London (Enfield)

2 0 2

Colchester / Ipswich

3 0 3

Nottingham 3 2 1

Manchester 3 0 3

Newcastle / Ashington

3 3 0

Exeter 3 1 2

TOTAL 19 7 12

National Noise Incidence Survey 1990 and Analysis of Diurnal Patterns 2.3

2.3.1 Changes from 1990 to 2013

A detailed analysis of sound level data from 1990 to 2013 using data collected from NNIS 1990, NNIS 2000 and FNM 2013 has been undertaken. This considers data from all the measurement sites in FNM 2013 which were also used in NNIS 2000 and NNIS 1990.

The items considered in this analysis are:

Variation between LAeq, LA10 and LA90 indicators over different periods of day and night (e.g. 12 hour day, 16 hour day, 4 hour evening, 8 hour night, 24 hours) between 1990, 2000 and 2013.

Variation in Lden noise levels between 1990, 2000 and 2013.

Note that the diurnal pattern work described in Section 2.3.2 below includes a detailed review of time histories for 1990, 2000 and 2013, and hence this has not been repeated within this overview of changes from 1990 to 2013.

It is also worth noting that the measurements in 2013 were all undertaken in a short period from January to June 2013, whilst those in 1990 and 2000 were spread over longer periods. Whenever possible the measurements in 2000 were undertaken at the same time of year as those in 1990; however this was not possible to 2013 due to this shortened survey period.

Hence, in addition to changes in noise level with time from 1990 to 2013, the differences seen between the 2013 and other surveys may also include contributions from seasonal variations in noise level.

Diurnal Patterns 2.4

A more detailed analysis of any changes in diurnal patterns in sound level between 1990 and 2013 has been undertaken, using data collected from NNIS 1990, NNIS 2000 and FNM 2013. This considers data from all the measurement sites in FNM 2013 which were also used in NNIS 2000 and NNIS 1990.


47065784_R2_03

August 2015

10

The key focus of this analysis was graphs of average time histories from the relevant sites from each of these surveys. In order to obtain a like-for-like comparison, Monday-Thursday average noise levels from FNM 2013 were used as these match the timescales of measurements undertaken for NNIS 2000 and NNIS 1990.

The analysis of diurnal patterns includes investigation of potential changes to the:

timing of the evening decrease in sound levels and the morning increase;

rate of the evening decrease in sound levels and the morning increase;

difference between daytime and night time sound levels;

timing, magnitude and duration of the period of increased sound levels during the morning rush hour; and

length of the noise “night” (the period of the lowest levels at night)

Data measured over 1-hour time frames have been used from all three surveys. In addition, data over 5-minute time frames are available for the NNIS 2000 and FNM 2013 surveys but were not logged in the NNIS1990 survey, so analysis of these data was restricted to the two more recent surveys only.

2.4.1 Sites available for analysis

The sites analysed were those from the NNIS 2000 data set which had been chosen for FNM 2013 monitoring. All of these sites had also been monitored during the NNIS 1990. At each stage of repeat, where use of the precise same address was not possible, acoustically equivalent sites were selected in the immediate vicinity and used for measurements.

There were 48 sites in the FNM 2013 data set of which 15 were exact repeat addresses from the NNIS 2000 and 10 of which were exact repeat addresses for all three surveys.

The breakdown of the sites by area is shown in Table 2.

TABLE 2. FNM 2013 SITES PREVIOUSLY MONITORED IN NNIS 2000 AND 1990

Area Number of FNM 2013 sites previously

monitoring in NNIS

Acoustically Equivalent

Sites in 2013

Exact Repeat Address for FNM

2013 and NNIS 2000 only

a

Exact Repeat sites monitored for FNM

2013, NNIS 2000 and NNIS 1990

London (Kingston Upon Thames)

6 4 2 0

London (Enfield)

5 3 0 2

Colchester / Ipswich 7 6 1 0

Nottingham 7 5 0 2

Manchester 7 4 1 2

Newcastle / Ashington 7 5 0 2

Exeter 9 6 1 2

TOTAL 48 33 5 10

a NNIS 2000 Survey used acoustically equivalent site but FNM2013 used exact repeat of NNIS 2000.


47065784_R2_03

August 2015

11

Strategic Noise Maps 2.5

2.5.1 Comparisons with Noise Mapping

This section of analysis compares the measurements from FNM2013 with noise predictions from the strategic noise mapping undertaken by Defra under the requirements of the EU Environmental Noise

Directive10

. The analysis presented in this report covers two distinct data sets:

comparison of Round 1 (2006) noise mapping data with NNIS 2000 measurements; and

comparison of Round 2 (2011) Noise mapping data with FNM 2013 measurements

A number of additional factors relating to the production of the noise maps should be considered. These include:

Strategic noise mapping data were produced for strategic purposes, and are not necessarily considered to be accurate at a very local level.

The levels obtained from the strategic noise maps were generated for specific sources only, and these will differ between different areas of the country. Levels used in this analysis are for road traffic and railways only, and levels from these sources are considered separately.

There are significant differences between the road traffic noise sources modelled for Round 1 and Round 2 noise maps in certain locations. It is understood that the following approach was taken at each round:

o Round 1

Locations with agglomerations of population >250,000 outside of London included all roads, with default traffic flows assigned to roads for which no actual data were available

Locations within London included all roads for which data were available from the London Atmospheric Emissions Inventory

Locations outside of agglomerations included major roads (with a defined minimum flow) only.

o Round 2

Locations within all agglomerations of population >100,000 included all A-roads and motorways.

Locations outside of agglomerations included major roads (with a lower defined minimum flow than used for Round 1) only.

2.5.2 Sites used in Comparisons

The number of sites available for inclusion within the analysis varied between the rounds of mapping, noise source considered and noise indicator, as wherever the predicted noise level was below 50 dB for an indicator, this site was excluded as a precise predicted noise level was not available.

Table 3 shows the number of sites which have been included in the analysis for each noise indicator and noise source. It should be noted that no further analysis has been undertaken for combinations for which only one site was available. Furthermore, not all indicators for a particular site necessarily have values of 50 dB or more.


47065784_R2_03

August 2015

12

TABLE 3. SITES INCLUDED IN ANALYSIS

Noise Level Indicator

Number of Sites – Round 1

Number of Sites – Round 2

Road Traffic Noise Railway Noise Road Traffic Noise Railway Noise

Lden 24 1b 28 2

Lnight 13 1b 9 0

LAeq,16hr 23 1b 17 1

b

Lday 23 1b 19 1

b

LA10,18hr 23 N/Ac 22 N/A

c

LAeq,18hr N/Ac 1

b N/A

c 1

b

Self-Completion Questionnaires 2.6

2.6.1 Scope of Work

This section of analysis considers relationships between the measured sound environment and responses from the FNM 2013 self-completion questionnaires. Where possible, any correlation between subjective responses and measured sound levels were considered. Fifty-two responses (of varying degrees completeness and detail) were received from the FNM 2013 questionnaire.

An initial review was undertaken of the questions included in the FNM 2013 questionnaire, to identify those which relate to the sound environment around the measurement position, or subjective responses which might be expected to relate to sound levels or the sound environment.

For each identified question, the questionnaire responses have been compared to measured sound levels using appropriate sound level indicators (e.g. accounting for any time of day specified in the question).

It should be noted that much of the existing dose-response analysis shown in the literature about similar datasets has been focussed on community response, rather than individual level data. Such analyses have been based on identifying relationships between noise level and the percentage of people highly annoyed (%HA) by noise. However, due to the small sample size available in the dataset being considered in this report, it was not considered appropriate to use this approach, and the analysis included in this report has been undertaken based on individual response data.

2.6.2 Review of FNM 2013 Questions

The FNM 2013 self-completion questionnaire asked about various aspects of the sound environment around each monitoring location.

The following questions were identified as expected to relate to level of satisfaction with the sound environment and are likely to be quantifiable from the long term noise descriptors.

The question numbers shown are those used in the self-completion questionnaire.

Q5 In general, how do you feel about the amount of noise (or absence of noise) around here?

b No further analysis has been undertaken for the combinations for which only one site was available.

c Predictions for 18-hour daytime were using the LA10 indicator for road traffic noise and the LAeq indicator for railway

noise due to the calculation methodologies employed.


47065784_R2_03

August 2015

13

Q6 Would you say that you live in a noisy area?

Q7 How sensitive would you say you are to noise?

Q8 Thinking about the last 12 months or so, when you are at home, to what extent are you personally bothered, annoyed or disturbed by noise coming from outside your home?

Q9 When you are at home does any noise coming from outside your home particularly bother, annoy or disturb you at each of the following times:

a) During the week (Monday to Friday) – Day (0700-1900), Evening (1900-2300), and Night (2300-0700)

b) During the weekend (Saturday to Sunday) - Day (0700-1900), Evening (1900-2300), and Night (2300-0700)

Q10 Do you ever visit outdoor places in order to find somewhere peaceful or quiet, for example, a park, country area or some other open space?

2.6.3 FNM 2013 Sound Level Data

A review of the above selected FNM 2013 questions showed that those relating to time of day involved responses for broad day, evening and night time periods only. Therefore the data selected for analysis for each site were the long term noise descriptors Lday, Levening, Lnight, and LAeq,24h.

1962 London Noise Survey 2.7

2.7.1 Aims of the Analysis

During the initial analysis of data from the FNM project, significant differences were noted between measured levels from 2013 and those from the five sites which were repeated from the 1960s LNS.

The aim of the work described in this report was to investigate these differences further and identify possible causes for these changes in noise level over 50 years. As the data from the London Noise Survey were available only in terms of LA10 and LA90 noise indicators only, the comparison has focussed on these indicators.

The key areas investigated were:

The temporal sampling methodology of the LNS measurements;

The location of LNS and FNM measurements with respect to the existing sound environment; and

Historic noise policy interventions (as reported in Defra research project NO0234: An

investigation into the effect of historic noise policy intervention)11

.

There are 5 sites in the FNM 2013 which were paired to sites from the LNS locations monitored in 1962.

When considering all data in this analysis, it is important to bear in mind that the LNS data have been read from graphs of noise levels against time from the LNS report. As such, there is a degree of uncertainty in the reported values. However, this is expected to be small in comparison to the differences seen between these surveys.


47065784_R2_03

August 2015

14

3 RESULTS AND DISCUSSION


3.1.1 Overall noise levels

In line with the approach taken in the main FNM report, the various noise indicators used in this analysis have been derived from the raw measurement data, the methodology for which is given in Appendix B along with the definition of each time period used. Appendix B also contains the minimum, mean and maximum long term descriptors measured for the overall data set of the 19 sites common to NNAS 2012 and FNM 2013.

3.1.2 Comparison of Questionnaire Responses to Noise Levels

This section sets out the results of comparisons between specific questions from the NNAS 2012 questionnaire and the appropriate noise level indicators.

3.1.3 Question A7

A7 In general, how do you feel about the amount of noise around here? Possible responses:

1 Definitely like 2 2 3 3 4 4 5 5 6 6 7 Definitely don't like 8 Don’t know 9 Not answered

Figure 1 below presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question A7. Figure 2 to Figure 5 show the same information for Levening, Lnight, LAeq,24h and Lden respectively.

Figure 1 A7: NNAS 2012 responses vs measured FNM 2013 Lday levels


47065784_R2_03

August 2015

15

Figure 2 A7: NNAS 2012 vs FNM Levening

Figure 3 A7: NNAS 2012 vs FNM Lnight

Figure 4 A7: NNAS 2012 vs FNM LAeq,24h

Figure 5 A7: NNAS 2012 vs FNM Lden

Figure 1 to Figure 5 show that the relationship between noise level and the NNAS 2012 response is similar for each of the noise level descriptors. This finding was also noted across all other questions analysed for this element of the report. Hence, for clarity graphs in the rest of this element have been shown for Lday only unless the NNAS question is specifically about a different time period, when the noise indicator for the relevant period has been used.

The responses were generally towards the positive end of the scale. All but one of the respondents answered the question on the 1 (definitely like) to 7 (definitely don’t like) scale. Of these, eleven made the more positive responses (response 1 or 2), seven made neutral responses (response 3, 4 or 5) and none selected negative responses (response 6 or 7).

There is not a clear correlation between measured noise levels and reported level of satisfaction. Although the ‘definitely like’ category is dominated by sites with relatively low level noise levels (46 dB to 54 dB Lday with a mean of 50 dB Lday) the spread for the next category (2) is wider (50 dB to 72 dB Lday with a mean of 57 dB Lday). The neutral responses (3 to 5) correspond to noise levels of 48 dB to 65 dB Lday with a mean of 55 dB Lday.

The one respondent who replied ‘don’t know’ lives at one of the quietest sites in the FNM 2013 dataset, with an Lday level of 44 dB.


47065784_R2_03

August 2015

16

3.1.4 Question A8

A8 How sensitive would you say you are to noise? Possible responses:

1 Not at all sensitive 2 2 3 3 4 4 5 5 6 6 7 Very sensitive 8 Don’t know 9 Not answered

Figure 6 below presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question A8.

Figure 6 A8: NNAS 2012 responses vs Measured FNM 2013 Lday levels

All the respondents answered the question on the 1 to 7 scale (not at all sensitive to very sensitive). Of these, seven regarded themselves as not sensitive (response 1 or 2), eight regarded themselves as in the middle of the sensitivity range (response 3, 4 or 5) and four regarded themselves as more sensitive (response 6 or 7).

Figure 6 shows the main body of measured Lday levels (50 dB to 60 dB) are spread across the sensitivity response band. However at those sites with the highest noise levels respondents do regard themselves as less sensitive. The homes of those who regard themselves as sensitive to noise are not subject to high noise levels.

Possible reasons for this pattern include:

people living in noisier areas become less sensitive to noise; or

people who are sensitive to noise choose to live in quieter areas.


47065784_R2_03

August 2015

17

3.1.5 Question A9 Section 3 (road traffic noise)

A9 When you are at home, to what extent are you personally bothered, annoyed or disturbed by noise from...road traffic? Possible responses:

1 Not at all 2 A little 3 Moderately 4 Very 5 Extremely 6 Don't know 7 Don't hear 8 Not answered

Question A9 related to different noise sources including road traffic noise and asks the respondent to consider noise inside the home and in their garden / balcony areas. All respondents reported that they had access to an outdoor area.

An analysis of the responses to road traffic noise is presented below. If people responded ‘not at all’, ‘don’t hear’ or ‘don’t know’ to question A9 further questions were asked to clarify the response to Question A9. The results of this clarification are included in the analysis presented below.

Figure 7 below presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question A9.

Figure 7 A9: NNAS 2012 responses vs Measured FNM 2013 Lday levels

None of the 19 respondents felt they were ‘very’ or ‘extremely’ bothered, annoyed or disturbed by road traffic noise when they were at home. The most negative response was ‘moderately’ for three respondents followed by ‘a little’ with seven responses.

Initially the remaining nine respondents chose “Not at all” bothered but following further clarification questions these were split into two groups – four “sometimes hear” road traffic noise but were still in


47065784_R2_03

August 2015

18

the “not at all” bothered category, whereas the other five stated they “never hear” road traffic noise. These last five were recategorised into the “don’t hear” category in Figure 7.

d

Figure 7 shows the three respondents in homes with the highest measured external noise levels report being at least “a little” bothered by road traffic noise. External noise levels for those “moderately” bothered range from 48 dB to 65 dB Lday with a mean of 55 dB Lday and those “a little” bothered also have a wide range (50 dB to 72 dB Lday with a mean of 58 dB Lday).

Those who are “not at all” bothered experience lower external noise levels (46 dB to 52 dB Lday with a mean of 50 dB Lday) and those who responded “do not hear” experience similar external noise levels (44 dB to 56 dB Lday with a mean of 52 dB Lday).

The finding suggests those experiencing the highest external noise levels report being at least “a little” bothered by road traffic noise, those experiencing moderate and low noise levels have different reactions to road traffic noise. The two respondents experiencing the lowest external noise levels report being “not at all” bothered or “don’t hear”.

3.1.6 Question RTN3

RTN3 The next set of questions is about all kinds of road traffic noise that you hear when you are at home. Which five of the words that I am about to show you best describe how the noise you hear from road traffic makes you feel? Possible choices: Irritated Tense Safe Anxious Startled Relaxed Annoyed Stressed Bothered Comforted Tired Angry Frightened Stimulated Excited Worried Depressed Upset Good Fed up Unhealthy/ill

RTN3 was only asked if:

Road Traffic Noise was indicated in a previous question (A10) where the respondent chose whether Road Traffic Noise (amongst other choices) bothers, annoys or disturbs them at home; and

A previous question (RTN1) indicated any type of road, class of vehicle or type of road traffic noise bothered, annoyed or disturbed them at least ‘a little’ when at home in the last 12 months.

Therefore five respondents were not asked this question at all because road traffic noise did not bother, annoy or disturb them. The distribution of the noise levels at these sites against the sites where the respondents were asked RTN3 are shown in Figure 8.

d Note that two of these points appear in the same place in Figure 7 due to very similar noise levels.


47065784_R2_03

August 2015

19

Figure 8 RTN3 NNAS 2012 Not Asked and Asked vs Measured FNM 2013 Lday Levels

Figure 8 shows that those not asked, and hence not bothered, annoyed or disturbed by road traffic noise experience relatively low external noise levels at home (44 dB to 54 dB Lday with a mean of 50 dB Lday) when compared to those who did (48 dB to 72 dB Lday with a mean of 56 dB Lday).

Not all of the possible choices in RTN3 were selected and Table 4 below shows the breakdown of the responses.

TABLE 4. RTN3 SUMMARY OF RESPONSES

NNAS Ref Word choice Number of Responses

Not asked Not selected Selected

rtn0301 Irritated 5 7 7

rtn0302 Tense 5 14 0

rtn0303 Safe 5 14 0

rtn0304 Anxious 5 14 0

rtn0305 Startled 5 12 2

rtn0306 Relaxed 5 14 0

rtn0307 Annoyed 5 11 3

rtn0308 Stressed 5 13 1

rtn0309 Bothered 5 10 4

rtn0310 Comforted 5 14 0

rtn0311 Tired 5 13 1

rtn0312 Angry 5 14 0

rtn0313 Frightened 5 13 1

rtn0314 Stimulated 5 14 0

rtn0315 Excited 5 14 0

rtn0316 Worried 5 14 0

rtn0317 Depressed 5 14 0

rtn0318 Upset 5 14 0


47065784_R2_03

August 2015

20


NNAS Ref Word choice Number of Responses

Not asked Not selected Selected

rtn0319 Good 5 13 1

rtn0320 Fed up 5 13 1

rtn0321 Unhealthy/ill 5 14 0

rtn0322 None selected 5 13 1

The words never selected relating to road traffic noise were: tense, safe, anxious, relaxed, comforted, angry, stimulated, excited, worried, depressed, upset and unhealthy/ill.

For those words which were selected by only one respondent no further meaningful analysis is considered possible. Figures for those words selected at least twice are presented below.

“Irritated”

Figure 9 presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site when “irritated” was chosen as a response to Question RTN3 (seven respondents). This was the most commonly selected response for road traffic noise.

Figure 9 RTN3 “Irritated”: NNAS 2012 responses vs Measured FNM 2013 Lday levels

Figure 9 shows that there is a pattern across the categories, those who report being “irritated” experience generally higher external noise levels than those not asked or not selecting “irritated”.

Those selecting the “irritated” response experience noise levels between 50 dB and 72 dB Lday with a mean of 57 dB Lday. Those who did not select “irritated” (including those not asked the question) experience external noise levels between 44 dB and 66 dB Lday with a mean of 52 dB Lday.

“Bothered”

Figure 10 presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site when “bothered” was chosen as a response to Question RTN3 (four respondents). This was the second most commonly selected response for road traffic noise.


47065784_R2_03

August 2015

21

Figure 10 RTN3 “Bothered”: NNAS 2012 responses vs Measured FNM 2013 Lday levels

Figure 10 shows that the external noise levels experienced by those who report being “bothered” are not notably different than those who do not, with a wide range of levels experienced by both categories of respondents.

Those selecting the “bothered” response experience noise levels between 48 dB and 66 dB Lday with a mean of 53 dB Lday. Those who did not select “bothered” (including those not asked the question) experience external noise levels between 44 dB and 72 dB Lday (the full range of measured values) with a mean of 54 dB Lday.

“Annoyed”

Figure 11 presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site when “annoyed” was chosen as a response to Question RTN3 (three respondents). This was the third most commonly selected response for road traffic noise.

Figure 11 RTN3 “Annoyed”: NNAS 2012 responses vs Measured FNM 2013 Lday levels


47065784_R2_03

August 2015

22

Figure 11 shows that the external noise levels experienced by those who report being “annoyed” are no greater than those who do not with those not selecting this option covering a wide range of noise levels.

“Startled”

Figure 12 presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site when “startled” was chosen as a response to Question RTN3 (two respondents). This was the fourth most commonly selected response for road traffic noise.

Figure 12 RTN3 “Startled”: NNAS 2012 responses vs Measured FNM 2013 Lday levels

Figure 12 shows that the external noise levels experienced by those who report being “startled” are no greater than some who do not.

Two respondents identified that road traffic noise “startled” them. With such a small sample no notable patterns can be identified. The spread of noise levels of those selecting “startled” is smaller than and falls within the noise levels experienced by those did not select this response.

The small sample size makes it challenging to determine patterns in response to road traffic noise, particularly where there are few selections of a particular response word. Just three words were chosen by more than two respondents (irritated, annoyed and bothered).

3.1.7 Question RTN4

RTN4 And does road traffic noise interfere with any of these aspects of your home life? Possible choices: Listening to TV, radio, music Having a conversation (including on the telephone) Reading, writing or other quiet activities Concentrating Sleeping Resting Using the whole house Spending time in the garden or balcony/terrace Having the windows or doors open Having guests or visitors


47065784_R2_03

August 2015

23

Spending time in the home

RTN4 was only asked if RTN3 was asked (see above). Therefore the same five respondents as in RTN3 were not asked this question at all because Road Traffic Noise did not bother, annoy or disturb them. In addition a number of respondents did not select any of the activities presented to them. The distribution of the noise levels across the three categories are shown in Figure 13.

Figure 13 RTN4 NNAS 2012 not asked, no selection and something selected vs Measured FNM 2013 Lday levels

Figure 13 shows that those not asked, and hence not bothered, annoyed or disturbed by road traffic noise experience relatively low external noise levels at home (44 dB to 54 dB Lday with a mean of 50 dB Lday) including the two lowest.

Overall those who identified road traffic noise interfered with at least one aspect of their home life experience higher external noise levels at home (48 dB to 72 dB Lday with a mean of 59 dB Lday) including the three highest, but also including some respondents with lower measured external noise levels.

Those who were asked question RTN4 but did not identify any aspect of their home life affected by road traffic noise experienced noise levels concentrated in the middle of the measurement set (50 dB to 57 dB Lday with a mean of 53 dB Lday)

Not all of the possible choices in RTN4 were selected and Table 5 below shows the breakdown of the responses.


47065784_R2_03

August 2015

24


NASS Ref Activity Number of Responses

Not asked No Yes

rtn0401 Listening to TV, radio, music 5 14 0

rtn0402 Having a conversation (including on the telephone) 5 14 0

rtn0403 Reading, writing or other quiet activities 5 13 1

rtn0404 Concentrating 5 14 0

rtn0405 Sleeping 5 11 3

rtn0406 Resting 5 12 2

rtn0407 Using the whole house 5 14 0

rtn0408 Spending time in the garden or balcony/terrace 5 13 1

rtn0409 Having the windows or doors open 5 11 3

rtn0410 Having guests or visitors 5 14 0

rtn0411 Spending time in the home 5 14 0

None of the above selected 5 6 8

The activities never selected as interfered with by road traffic noise were: listening to TV, radio, music, having a conversation (including on the telephone), concentrating, using the whole house, having guests or visitors and spending time in the home.

As for RTN03, no further analysis has been undertaken for words which were selected by only one respondent. Further details for those selected at least twice are presented below,

“Sleeping”

Figure 14 presents the Lnight levels measured during the FNM 2013 survey at each NNAS 2012 site when “Sleeping” was chosen as a response to Question RTN4 (three respondents), and was one of the two most frequently chosen aspects affected by road traffic noise.

Figure 14 RTN4 “Sleeping”: NNAS 2012 responses vs Measured FNM 2013 Lnight levels


47065784_R2_03

August 2015

25

Figure 14 shows that the external noise levels experienced by the three respondents who report sleep being affected by road traffic noise are not notably different to those who do not.

“Having the windows or doors open”

Figure 15 presents the LAeq,24h levels (all day and night) measured during the FNM 2013 survey at each NNAS 2012 site when “Having the windows or doors open” was chosen as a response to Question RTN4 (three respondents), and was one of the two most frequently chosen aspects affected by road traffic noise.

Figure 15 RTN4 “Having the windows or doors open”: NNAS 2012 responses vs Measured FNM 2013 LAeq,24h levels

Figure 15 shows that the external noise levels experienced by two of the three respondents who report having windows or doors open being affected by road traffic noise are slightly higher than those experienced by the majority of those who did not report this.

“Resting”

Figure 16 presents the LAeq,16h levels (day and evening) measured during the FNM 2013 survey at each NNAS 2012 site when “Resting” was chosen as a response to Question RTN4 (two respondents), and was the third most frequently chosen aspect affected by road traffic noise.


47065784_R2_03

August 2015

26

Figure 16 RTN4 “Resting”: NNAS 2012 responses vs Measured FNM 2013 LAeq,16h levels

Examination of Figure 16 shows that the external noise levels experienced by the two respondents who report resting being affected by road traffic noise are widely spread and within the range of those experienced by respondents who did not select this response.

For those who report resting is affected experience external noise levels between 50 dB and 65 dB LAeq,16h with a mean of 58 dB LAeq,16h.

Those who did not select resting (including those not asked the question) experience external noise levels between 44 dB and 71 dB LAeq,16h (the full range of measured values) with a mean of 53 dB LAeq,16h.

Although the mean external noise levels were 5 dB LAeq,16h higher for those whose resting is affected than those whose resting is not, this is a small sample size and the spread of noise levels of those affected is smaller than and falls within the noise levels of those who are not affected.

RTN 4 Question summary

The small sample size makes it difficult to determine patterns in the five aspects of home life chosen as affected by road traffic noise, particularly since there are few selections of each aspect. Just three aspects were chosen by more than one respondent (sleeping, having the windows or doors open and resting). The other two aspects chosen by just one respondent each were reading, writing or other quiet activities and spending time in the garden or balcony/terrace.

At the three sites with the highest external noise levels road traffic noise was identified by the respondents as interfering with aspects of their home life.

3.1.8 Question RTN5a

RTN5a Does road traffic noise particularly bother, annoy or disturb you, at home, at each of the times listed on the card during the week - that is Monday to Friday? Possible choices:

1 Yes 2 No 3 Not generally home at this time 4 No particular time 5 It varies


47065784_R2_03

August 2015

27

6 Don't know 7 Not answered 8 Not applicable

Responses were recorded for each of three time periods:

RTN5a_1 Day 7am - 7pm

RTN5a_2 Evening 7pm - 11pm

RTN5a_3 Night 11pm - 7am

RTN5 was only asked if RTN3 was asked (see above). Therefore the same five respondents as in RTN3 were not asked this question at all because road traffic noise did not bother, annoy or disturb them. The responses “it varies” and “don’t know” were not selected.

A second question (RTN5b) was then asked, identical to RTN5a but asking about the weekend rather than weekdays:

Each set of responses were compared to the measured noise levels for the appropriate time of day and week.

External noise levels for those who selected “yes” have been compared in Table 6 to those who selected “no” or were not asked. Where “not generally home at this time” or “no particular time” were selected the noise level data were not included in the analysis.

TABLE 6. RTN5 SUMMARY OF RESPONSES WITH NOISE LEVELS

Time of week Time /

descriptor

“Yes” “No” or not asked

Respondents LAeq,T (dB)

Respondents LAeq,T (dB)

min mean max min mean max

Monday - Friday (weekday)

07:00 to 19:00 Lday

3 55 65 73 14 45 53 67

19:00 to 23:00 Levening

4 45 53 64 14 41 50 69

23:00 to 07:00 Lnight

5 41 47 54 13 38 46 62

Saturday & Sunday (weekend)

07:00 to 19:00 Lday

3 51 53 56 15 45 53 71

19:00 to 23:00 Levening

3 43 50 55 15 38 50 68

23:00 to 07:00 Lnight

6 41 47 56 12 36 46 62

Figure 17 below presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN5a (Monday to Friday, day).


47065784_R2_03

August 2015

28

Figure 17 RTN5a Monday to Friday daytime NNAS 2012 responses vs Measured FNM 2013 Lday levels

Figure 17 shows that the weekday daytime external noise levels experienced by the three respondents who report being particularly bothered, annoyed or disturbed by road traffic noise during the weekday days are higher than most of those who do not.

Figure 18 below presents the Levening levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN5a (Monday to Friday, evening).

Figure 18 RTN5a Monday to Friday evening NNAS 2012 responses vs Measured FNM 2013 Levening levels

Figure 18 shows that the weekday evening external noise levels experienced by the four respondents who report being particularly bothered, annoyed or disturbed by road traffic noise during the weekday evenings cover a similar range to those experienced by respondents who did not report this.


47065784_R2_03

August 2015

29

Figure 19 below presents the Lnight levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN5a (Monday to Friday, night).

Figure 19 RTN5a Monday to Friday night NNAS 2012 responses vs Measured FNM 2013 Lnight levels

Figure 19 shows that the weekday night external noise levels experienced by the five respondents who report being particularly bothered, annoyed or disturbed by road traffic noise during the weekday night times cover a similar range to those experienced by respondents who did not report this. The respondents experiencing the highest weekday night time noise levels did not report being particularly bothered annoyed or disturbed at this time.

Figure 20 below presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN5b_1 (Saturday and Sunday, day).

Figure 20 RTN5b Saturday and Sunday daytime NNAS 2012 responses vs Measured FNM 2013 Lday levels


47065784_R2_03

August 2015

30

Figure 20 shows that the weekend daytime external noise levels experienced by the three respondents who report being particularly bothered, annoyed or disturbed by road traffic noise during the weekend days are similar to the average for those who do not, but with a much smaller spread of values.

Figure 21 below presents the Levening levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN5b (Saturday and Sunday, evening).

Figure 21 RTN5b Saturday and Sunday evening NNAS 2012 responses vs Measured FNM 2013 Levening levels

Figure 21 shows that the weekend evening external noise levels experienced by the three respondents who report being particularly bothered, annoyed or disturbed by road traffic noise during the weekend evenings are on towards the lower end of range of levels experienced by average those who do not report this, and also cover a smaller range of values.

Figure 22 below presents the Lnight levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN5b_3 (Saturday and Sunday, night).


47065784_R2_03

August 2015

31

Figure 22 RTN5b Saturday and Sunday night NNAS 2012 responses vs Measured FNM 2013 Lnight levels

Figure 22 shows that the external noise levels experienced by the six respondents who report being particularly bothered, annoyed or disturbed by road traffic noise during the weekend nights cover a similar range to those who do not. The respondents experiencing the highest weekend night time noise levels did not report being particularly bothered annoyed or disturbed at this time.

In general, those not asked the various parts of this question experienced noise levels towards the lower end of the range. These were individuals who had reported that road traffic noise did not bother, annoy or disturb them in previous questions.

3.1.9 Question RTN6 and RTN7

RTN6 How much would you say you are bothered, annoyed or disturbed by the road traffic noise while it is going on? Possible responses:

1 Not at all 2 2 3 3 4 4 5 5 6 6 7 Very seriously 8 Don't hear 9 Don't know 10 Not answered 11 Not applicable

A second question RTN7 was then asked, similar to RTN6 but enquiring about the respondents’ emotions not just when road traffic noise is present: RTN7 And how much do you feel that road traffic noise spoils your home life, in general, not just when the noise is going on? The choice of responses were the same as for RTN6.

RTN6 and RTN7 were only asked if RTN3 was asked (see above). Therefore the same five respondents as in RTN3 were not asked these questions because Road Traffic Noise did not bother, annoy or disturb them.


47065784_R2_03

August 2015

32

Figure 23 below presents the LAeq,24h levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN6.

Figure 23 RTN6: NNAS 2012 responses vs Measured FNM 2013 LAeq,24h levels

All the respondents who were asked answered the question on the 1 to 7 scale (not at all bothered to very seriously bothered while road traffic noise is going on). Of these, six regarded themselves as not bothered (response 1 or 2), seven regarded themselves as in the middle of the bothered range (response 3, 4 or 5) and none regarded themselves as more seriously bothered (response 6 or 7).

Figure 23 shows a pattern across the response band with those experiencing the highest external noise levels regarding themselves as moderately bothered, those experiencing lower noise levels report themselves on the not bothered scale or were never asked the question. Some significant outliers are also evident.

Figure 24 below presents the LAeq,24h levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question RTN7.

Figure 24 RTN7: NNAS 2012 responses vs Measured FNM 2013 LAeq,24h levels


47065784_R2_03

August 2015

33

All the respondents who were asked answered the question on the 1 to 7 scale (not at all spoiled to very seriously spoiled, not just when road traffic noise is going on). Of these, thirteen regarded road traffic noise as not spoiling their home life not just when it was occurring (response 1 or 2), one regarded themselves as in the middle of the range (response 3, 4 or 5) and none regarded their home life as seriously spoilt by road traffic noise (response 6 or 7).

The one respondent who felt their home life is moderately spoilt by road traffic noise even when it is not occurring experiences noise levels of 49 dB LAeq,24h, which is towards the lower end of the noise levels measured at these sites.

3.1.10 Question O1

O1 Taking all these noises [heard at home and discussed during the survey] together, I would like you to look at the statements on the card and tell me which one best describes the extent to which noise spoils your home life?

This question was posed to all the respondents, but two alternate sets of responses were presented on alternate surveys (as part of a backwards compatibility comparison), so each respondent only saw one set of responses. Both sets of responses were on a sliding scale and had the response ‘a little’ in common. For the purposes of this analysis the responses have been combined as follows:

Response set 1 Response set 2 Combined Coding for this analysis

1 Completely 1 Totally Completely

2 A great deal 2 Quite a lot Quite a lot

3 A fair amount

4 A little 3 A little

A little 4 Not very much

5 Not at all 5 Not at all Not at all

6 Don't know 6 Don't know Don't know

7 Not answered 7 Not answered Not answered

Figure 25 below presents the Lday levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question O1 noise spoiling home life.


47065784_R2_03

August 2015

34

Figure 25 O1: NNAS 2012 responses vs Measured FNM 2013 Lday levels

All the responses were on the scale above – no one answered “Don’t know”.

No one identified noise as spoiling their home life “completely”, the one respondent who identified “quite a lot” experiences external noise levels of 50 dB Lday which is towards the lower end of the range of levels experienced at these sites.

The results indicate that generally those whose home life is “not at all” spoiled by noise experience lower noise levels than those who feel noise does in some part spoil their home life.

3.1.11 Question O2 (response 4)

O2 Over the past year, while in this home, have you been kept awake, or woken up, or changed the time when you go to bed or get up, because of … noise from outdoors…? Possible responses:

0 No 1 Yes

All respondents were asked this question, the responses to which covered a variety of internal noise sources and other personal and comfort factors. The “noise from outdoors” response is the only one likely to be captured by the FNM 2013 measurements. The Lnight descriptor is discussed as the question relates to sleep.

Figure 26 below presents the Lnight levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question O2, noise from outdoors.


47065784_R2_03

August 2015

35

Figure 26 O2: NNAS 2012 responses vs Measured FNM 2013 Lnight levels

The three people who selected noise from outdoors as affecting their sleep patterns experienced the moderate to high external Lnight noise levels, from 46 to 54 dB with a mean of 50 dB Lnight.

The remaining sixteen experience a wider range of noise levels from 37 to 61 dB with a mean of 45 dB Lnight.

The majority of respondents do not identify external noise as affecting their sleep patterns. There is little evidence of a pattern in comparing these results with the external noise levels, but may be affected by a number of factors such as variations in façade sound insulation, whether people choose to keep windows open the type of external noise and whether the bedroom is at the front.

3.1.12 Question O06 (response 1)

O6 Thinking now about the whole home, not just the bedroom, do you ever close the windows, or keep the windows closed for … Noise coming in through the window…? Possible responses:

0 No 1 Yes

All respondents were asked this question, the responses to which covered a variety of reasons affecting whether windows are closed, including air quality, warmth and security. The “noise coming in through the window” response is the only one analysed here. The LAeq,24h descriptor is discussed as the question relates to day, evening and night.

Figure 27 below presents the LAeq,24h levels measured during the FNM 2013 survey at each NNAS 2012 site arranged according to the response to Question O6, noise coming in through the window.


47065784_R2_03

August 2015

36

Figure 27 O6: NNAS 2012 responses vs Measured FNM 2013 LAeq,24h levels

The majority of respondents do not identify noise coming in through the window as influencing whether they close the windows. There is little evidence of a pattern in comparing these results with the external noise levels.


47065784_R2_03

August 2015

37

National Noise Incidence Survey 1990 and Analysis of Diurnal Patterns 3.2

3.2.1 Changes from 1990 to 2013

In order to obtain a like-for-like comparison between the surveys, Monday-Thursday average noise levels from FNM 2013 were used as these match the timescales of measurements undertaken for NNIS 2000 and NNIS 1990.

For the comparison between the surveys averaging has been applied across sites for the noise indicators, Lday, Levening, Lnight, LAeq,16h, LAeq,24h and Lden. For the FNM 2013 where multiple days of measurement exist within each category (e.g. across Monday to Thursday), an average over all relevant days has been used.

Table 7 shows the longer term data averages (Lday, Levening, Lnight, LAeq,16h, LAeq,24 and Lden) over all 48 sites for each of the three surveys.

TABLE 7. LONG TERM INDICATORS FOR NNIS 1990, NNIS 2000 AND FNM 2013

Noise Indicator Time period Longer term noise indicators averaged over 48 sites (dB)

NNIS 1990 NNIS 2000 FNM 2013

Lday (LAeq,12h) 07:00 – 19:00 56.2 55.7 55.0

Levening (LAeq,4h) 19:00 – 23:00 52.1 52.1 50.4

Lnight (LAeq,12h) 23:00 – 07:00 47.6 47.8 46.7

LAeq,16h 07:00 – 23:00 55.5 55.1 54.3

LAeq,24h 00:00 – 00:00 54.3 53.8 53.0

Lden D 07:00 – 19:00 E 19:00 – 23:00 N 23:00 – 07:00

57.2 57.0 56.1

The changes in these indicators between the surveys show that the FNM data is lower for all of the indicators than in either of the NNIS surveys.

The NNIS 2000 levels are the same or lower than the NNIS 1990 levels for all the indicators other than Lnight which is very slightly higher in NNIS 2000. The noise indicators in the two NNIS surveys are closer to each other than to the FNM 2013 levels.

It is necessary to recognise that the FNM sample size is small and covers only seven of the wide range of areas previously covered by the NNIS 1990 and NNIS 2000. The FNM 2013 was not designed to produce data statistically significant at the national population level whereas the NNIS surveys were. However, it is interesting to note that the pattern of noise level changes between 1990 and 2000 calculated over the whole of England and Wales showed a similar pattern of changes to those seen between 1990 and 2000 for this smaller dataset.

Within this FNM sample an analysis of the significance has been undertaken. For each site the differences between the three surveys were calculated for each noise indicator (Lday, Levening, Lnight,

LAeq,16h, LAeq,24 and Lden). Then the means of these differences across the 48 sites were calculated and the confidence interval and the significance results are presented in Table 8.


47065784_R2_03

August 2015

38

TABLE 8. DIFFERENCE IN LONG TERM INDICATORS FOR NNIS 1990, NNIS 2000 AND FNM 2013

Noise Indicator

NNIS 2000 – NNIS 1990 FNM 2013 – NNIS 1990 FNM 2013 – NNIS 2000

Mean difference

(dB)

95% CI (dB)

Signi-ficant?

Mean difference

(dB)

95% CI (dB)

Signi-ficant?

Mean difference

(dB)

95% CI (dB)

Signi-ficant?

Lday (LAeq,12h) -0.5 0.8 No -1.2 1.0 YES -0.7 0.8 No

Levening (LAeq,4h) 0.1 1.0 No -1.6 1.0 YES -1.7 0.8 YES

Lnight (LAeq,12h) 0.2 1.1 No -0.9 1.2 no -1.0 1.0 YES

LAeq,16h -0.4 0.8 No -1.2 0.9 YES -0.8 0.7 YES

LAeq,24h -0.4 0.8 No -1.3 0.9 YES -0.8 0.7 YES

Lden -0.2 0.9 No -1.2 0.9 YES -1.0 0.8 YES

Inspection of Table 8 shows no significant differences between the NNIS 1990 and NNIS 2000 surveys. The FNM 2013 noise indicators show a statistically significant change (decrease) from both NNIS surveys for many different sound level indicators.

Comparison of FNM 2013 to NNIS 1990 shows a statistically significant decrease in all the indicators apart from Lnight. The decrease in the mean Levening levels is the most marked.

Comparison of FNM 2013 to NNIS 2000 shows a statistically significant decrease in all the indicators apart from Lday. Again, the decrease in the mean Levening levels is the most marked.

Extended detail and analysis of these findings can be gained from an analysis of hourly time histories for LAeq, LA10 and LA90 noise levels for the three studies. These results are discussed in the diurnal analysis section of this report.

3.2.2 LA10 and LA90 Noise Indicators

Table 9 and Table 10 below present a similar analysis to that above, but for LA10 and LA90 noise indicators over different periods of the day. In common with the approach taken for previous surveys, these values have been derived as the average of the appropriate 1-hour LA10 and LA90 noise levels.

TABLE 9. LONG TERM LA10 AND LA90 INDICATORS FOR NNIS 1990, NNIS 2000 AND FNM 2013

Noise Indicator Time period Longer term noise indicators averaged over 48 sites (dB)

NNIS 1990 NNIS 2000 FNM 2013

LA10,12hr 07:00 – 19:00 57.2 56.7 55.8

LA10,4hr 19:00 – 23:00 52.9 53.0 51.1

LA10,8hr 23:00 – 07:00 45.2 44.3 43.8

LA10,16hr 07:00 – 23:00 56.1 55.8 54.7

LA10,24hr 00:00 – 00:00 52.5 51.9 51.1

LA90,12hr 07:00 – 19:00 44.3 43.6 43.5

LA90,4hr 19:00 – 23:00 39.5 40.0 38.7

LA90,8hr 23:00 – 07:00 33.8 34.1 33.2

LA90,16hr 07:00 – 23:00 43.1 42.7 42.3

LA90,24hr 00:00 – 00:00 40.0 39.8 39.3


47065784_R2_03

August 2015

39

TABLE 10. DIFFERENCE IN LONG TERM LA10 AND LA90 INDICATORS FOR NNIS 1990, NNIS 2000 AND FNM 2013

Noise Indicator

NNIS 2000 – NNIS 1990 FNM 2013 – NNIS 1990 FNM 2013 – NNIS 2000

Mean difference (dB)

95% CI (dB)

Signi-ficant?


95% CI (dB)

Signi-ficant?


95% CI (dB)

Signi-ficant?

LA10,12hr -0.5 0.9 No -1.3 1.1 YES -0.9 0.6 YES

LA10,4hr 0.1 1.0 No -1.8 0.8 YES -1.9 0.8 YES

LA10,8hr -0.9 1.2 No -1.4 1.3 YES -0.4 1.0 no

LA10,16hr -0.3 0.9 No -1.4 1.0 YES -1.1 0.6 YES

LA10,24hr -0.5 0.9 No -1.4 1.0 YES -0.9 0.6 YES

LA90,12hr -0.7 1.0 No -0.8 1.3 no -0.1 0.9 no

LA90,4hr 0.5 1.1 No -0.8 1.1 no -1.3 1.0 YES

LA90,8hr 0.3 1.6 No -0.6 1.5 no -0.9 1.2 no

LA90,16hr -0.4 1.0 No -0.8 1.2 no -0.4 0.9 no

LA90,24hr -0.2 1.1 No -0.7 1.2 No -0.6 0.9 no

The results reported in these two tables show a generally similar pattern to those seen for the LAeq indicators in Table 7 and Table 8. All noise indicators except those for the 8-hour night time period and the 4-hour evening LA10 show slight decreases from 1990 to 2000, and all indicators show slight decreases between 1990 and 2013 and between 2000 and 2013.

The picture with regards to statistically significant differences is less clear, although the majority of decreases in LA10 levels from 1990 to 2013 and from 2000 to 2013 are statistically significant (with the exception of the night time decrease from 2000 to 2013). Fewer of the decreases in LA90 are seen to be statistically significant (only the decrease in LA90,4hr from 2000 to 2013 does show as significant), and no changes from 1990 to 2000 are statistically significant (as was found for the LAeq indicators).

3.2.3 Diurnal Analysis

Hourly data averaged across the 48 sites has been examined for the three surveys, the LAeq,1hr, LA90,1hr

and LA10,1hr data are shown in Figure 28 and are referred to throughout this section of the report.

Five minute data averaged across the 48 sites have been examined for FNM 2013 and NNIS 2000. These were not recorded for the NNIS 1990. The LAeq,5min, LA90,5min and LA10, 5min data for the two surveys are shown in and are referred to throughout this section of the report.

The data for each of the FNM 2013 sites have been averaged over the Monday to Thursday period, before being averaged across all 48 sites. This is compared to the single day measurements averaged over 48 sites for the NNIS 2000. The effect of the extra averaging in the FNM 2013 dataset is seen as a smoother line in Figure 29 than for the NNIS 2000 data.

Note that these graphs use a different vertical scale to the other graphs in the FNM 2013 reports to allow easier reading of the graph.

The analysis presented in this report covers a number of topics including:

timing of the evening decrease in sound levels and the morning increase in sound levels;

rate of the evening decrease in sound levels and the morning increase in sound levels;

difference between daytime and night time sound levels;


47065784_R2_03

August 2015

40

timing, magnitude and duration of the period of increased sound levels during the morning rush hour; and

length of the noise “night”.

It should be noted that all findings reported from these graphs relate to the 48 sites only, and cannot be considered to be statistically significant, or necessarily representative of patterns which might be seen across the country as a whole.

Defra – Focussed Noise Monitoring 2013

47065784_R2_03

August 2015

41

Figure 28 Comparison of hourly data for all NNIS sites (NNIS 1990, NNIS 2000 and FNM 2013)


47065784_R2_03

August 2015

42

Figure 29 Comparison of 5 minute data for all NNIS sites (NNIS 2000 based on single day and FNM 2013 averaged Mon - Thurs)


47065784_R2_03

August 2015

43

Timing and rate of the evening decrease in sound levels

Examination of the hourly data in Figure 28 shows the following for the evening period 19:00 to 23:00:

LAeq,1hr noise level graphs show FNM 2013 noise levels are consistently lower than both of the NNIS noise levels, specifically, they are:

lower than NNIS 1990 by around 1.7 dB on average but with a similar rate of decrease; and

lower than NNIS 2000 by around 2.0 dB on average but with a steeper rate of evening decrease, typically 1.0 dB lower at the start of the evening period increasing to around 3.0 dB lower at the end.

Hence, the 1990 and 2000 LAeq,1hr levels over this time period are very similar, but with a slower rate of decrease in 2000 than for 1990. In 2013, the levels have reduced slightly overall, and the rate of decrease has returned to be comparable with that measured in 1990.

LA10,1hr noise level graphs show FNM 2013 noise levels are consistently lower than both of the NNIS noise levels:

lower than NNIS 1990 by around 1.1 dB on average and with a slightly shallower rate of evening decrease typically 1.4 dB lower at the start of the evening period decreasing to around 0.9 dB lower at the end; and

lower than NNIS 2000 by around 1.9 dB on average but with a steeper rate of decrease, typically 1.0 dB lower at the start of the evening period increasing to around 2.5 dB lower at the end.

Hence, from 1990 to 2000 there was a slight increase in LA10,1hr noise levels over this period, and this increase was greater later in the evening. From 2000 to 2013 there was a slight decrease in LA10,1hr levels, reducing them to below the 1990 levels. This decrease was also greater later in the evening.

LA90,1hr noise level graphs show FNM 2013 noise levels are lower than the NNIS 2000 but similar to the NNIS 1990 levels:

the same as NNIS 1990 on average and with a slightly steeper rate of decrease typically 0.2 dB higher at the start of the evening period decreasing to around 0.4 dB lower by the end; and

lower than NNIS 2000 by around 1.3 dB on average but with a steeper rate of decrease, typically 0.5 dB lower at the start of the evening period increasing to around 1.9 dB lower at the end.

Hence, from 1990 to 2000, LA90,1hr values increased slightly over this period, whilst from 2000 to 2013, a similar reduction was seen, returning to 1990 levels. This reduction was slightly greater later in the evening period.

Examination of the 5 minute data for FNM 2013 and NNIS 2000 in Figure 29 shows patterns very similar to the hourly data, but with a greater variability due to the higher resolution. No additional conclusions about the rate of evening decrease were drawn from the 5 minute data. However the start of the evening reduction in noise levels appears to begin at around 17:50 for the FNM 2013, slightly later than for the NNIS 2000 in which the noise levels start to reduce at around 17:30.

Night time patterns

The Lnight indicator defines the night period between 23:00 and 07:00. As can be seen from Figure 28 and Figure 29, within this time there are a wide range of noise levels, with the lowest noise levels


47065784_R2_03

August 2015

44

generally occurring between midnight and 04:00 followed by a steep increase in levels toward the daytime noise levels by 07:00.

For the purposes of this report the quietest times are referred to as the “noise night” and are defined as the lowest consecutive three hours of hourly noise levels for each of the LAeq,1hr, LA10,1hr and LA90,1hr.

Analysis of both the hourly (all surveys) and 5 minute data (NNIS 2000 and FNM 2013) give further details of the differences in timing and rate of increase of the morning increase in sound levels.

Length of the noise “night”

The consecutive three hours with the lowest averaged noise levels are consistent across the surveys and the noise indicators. All three surveys have the lowest three hourly noise levels starting at 01:00 for each of the noise indicators, LAeq,1hr, LA90,1hr and LA10,1hr .

The minimum overall average hourly noise levels occurred at the following times for each survey. Where two or more hourly averaged noise levels are within 0.5 dB they are regarded as similar and all are reported:

NNIS 1990:

LAeq,1hr for the hour starting at 03:00, minimum of 39.9 dB;

LA10,1hr for the hours starting at 02:00 and 03:00; minimum of 38.9 dB and

LA90,1hr for the hours starting at 02:00 and 03:00, minimum of 30.3 dB.

NNIS 2000:

LAeq,1hr for the hours starting at 02:00 and 03:00, minimum of 40.7 dB;

LA10,1hr for the hours starting at 02:00 and 03:00, minimum of 39.4 dB; and

LA90,1hr for the hours starting at 01:00, 02:00 and 03:00, minimum of 31.3 dB.

FNM 2013:

LAeq,1hr for the hour starting at 02:00, minimum of 38.6 dB;

LA10,1hr for the hour starting at 02:00, minimum of 38.1 dB and

LA90,1hr for the hours starting at 01:00, 02:00 and 03:00, minimum of 30.1 dB.

The minimum levels for each of the noise indicators are the highest in the NNIS 2000 survey and lowest in the FNM 2013.

Further detail is available in the 5 minute data sets for the FNM 2013 and NNIS 2000 surveys as shown in Figure 29. The period at which the five minute data was below the 12.5% percentile (i.e. the quietest 12.5% of the day) are presented below. This corresponds to the 3 hour selection for the hourly data above.

NNIS 2000:

LAeq,5min starting from 00:55 until 04:00, with levels for single 5-minute periods below the 12.5 percentile at 01:25 and 03:00;

LA10,5min starting at 00:55 until 03:55; with levels for the single 5-minute period below the 12.5 percentile at 03:00; and


47065784_R2_03

August 2015

45

LA90,5min starting at 01:00 until 04:00; with levels for the single 5-minute period below the 12.5 percentile at 02:00.

FNM 2013:

LAeq,5min starting at 00:50 until 03:45;

LA10,5min starting at 00:50 until 03:45; and

LA90,5min starting at 00:45 until 03:40.

These are close to the timings of the three lowest hours in the hourly dataset (the lowest three hourly noise levels starting at 01:00 for each of the noise indicators, LAeq,1hr, LA90,1hr and LA10,1hr). The FNM “noise night time” by this measure starts slightly earlier than the NNIS 2000, by 15 minutes for the LA90,5min and 5 minutes for the LAeq, 5min and LA10, 5min.

The minimum overall average 5 minute noise levels occurred at the following times for each survey. All those up to 0.5 dB greater than the minimum are regarded as similar and are also reported:

NNIS 2000:

LAeq,5min minimum level of 37.1 dB at 02:30 and within 0.5 dB of this level from 02:25 to 02:39 and from 03:35 to03:44;

LA10,5min minimum level of 38.0 dB at 03:40 and within 0.5 dB of this level from 02:25 to 02:34 and from 03:35 to 03:49; and

LA90,5min minimum level of 31.6 dB at 02:40 and within 0.5 dB of this level from 02:15 to 03:54 and a single five minute period at 01:30.

FNM 2013:

LAeq,5min minimum level of 35.3 dB at 02:40 and within 0.5 dB of this level from 02:40 to 02:54;

LA10,5min minimum level of 36.7 dB at 02:40 and within 0.5 dB of this level from 02:40 to 02:54, with 1 level within 0.5 dB of this at 02:00; and

LA90,5min minimum level of 30.0 dB at 02:50 and within 0.5 dB of this level of this from 01:15 to 02:04 and 02:15 to 03:09.

This shows for the NNIS 2000 dataset there were two particularly quiet periods at around 02:30 and 03:40. The FNM 2013 dataset indicates just one quietest period at around 02:40 but no second minimum. However, this difference may be due to the additional smoothing which exists in the FNM 2013 dataset due to the longer measurement durations, and the apparent second minimum in the NNIS 2000 data may simply be an artefact of the greater random variation in this graph.

It can also be seen that the minimum LAeq,5min, LA90,5min and LA10,5min levels are lower in the FNM 2013 than the NNIS 2000 across all these indicators. This confirms the pattern seen in the hourly data for the two surveys.

Timing, magnitude and rate of the morning increase in sound levels including rush hour.

Examination of the hourly data in Figure 28 shows the start of the increase in noise levels from the quietest night time periods up to highest morning “rush hour” noise levels starts at 03:00 for all three surveys but increase at different rates to a maximum at between 08:00 and 09:00 depending on the survey and noise indicator.

It is noticeable at the start of the slope just after 03:00 the LAeq,1hr and LA10,1hr noise levels coincide for all three surveys. The LA90,1hr levels for the different surveys are more spread out at this time.


47065784_R2_03

August 2015

46

The detailed pattern of noise levels over this period has been examined separately below for each of the three surveys.

NNIS 1990

LAeq,1hr noise levels rise at a fairly constant rate up to 07:00 and then at a slower rate into the rush hour maximum of 56.6 dB in the hour starting 09:00;

LA10,1hr noise levels rise at a fairly constant rate up to 07:00 (rate of increase is higher than the LAeq,1hr) and then at a slower rate into the rush hour maximum of 57.9 dB in the hour starting at 08:00; and

The LA90,1hr levels experience a more gentle rise up to 06:00 followed by a steeper increase to its maximum of 44.5 dB during the hour starting at 08:00.

NNIS 2000

LAeq,1hr noise levels rise at a fairly constant rate up to 07:00 and then at a slower rate into the rush hour maximum of 55.5 dB which is maintained over the hours starting at 08:00 and 09:00;

LA10,1hr noise levels rise at a fairly constant rate up to 07:00 (rate of increase is higher than the LAeq,1hr) and then at a slower rate into the rush hour maximum of 57.7 dB in the hour starting at 08:00; and

The LA90,1hr levels experience a more gentle rise up to 07:00 followed by a steeper increase to its maximum of 45.0 dB during the hour starting at 08:00.

FNM 2013

LAeq,1hr noise levels rise steeply between the hours starting at 03:00 and 04:00 then at a slower rate up to 07:00 before the rate of increase tails off further into the rush hour maximum of 55.1 dB in the hour starting at 08:00;

LA10,1hr noise levels also rise steeply between the hours starting at 03:00 and 04:00 then at a slower rate up to 07:00 before the rate of increase tails off further into the rush hour maximum of 57.2 dB in the hour starting at 08:00; and

The LA90,1hr levels rise at a fairly constant rate up to 07:00 and then at a slower rate into the rush hour maxima of 44.7 dB.

Comparison of the rate of increase of the hourly data for the three surveys shows that whilst generally the rate of increase is similar in each survey the FNM 2013 shows a sharper increase in noise levels in the early hours of the morning than either of the NNIS surveys. This rate of increase then tails off toward the maximum rush hour levels.

For all surveys the LA10,1hr levels rise more sharply than the LAeq,1hr as rush hour approaches.

Comparison of the absolute noise levels in the lead up to rush hour show:

The FNM 2013 LAeq,1hr and LA10,1hr noise indicators are higher than or similar to both NNIS levels during the early morning increase up to the hour beginning at 07:00.

The FNM 2013 LA90,1hr noise indicators are similar to the NNIS 2000 levels during the early morning increase up to the hour beginning at 07:00. However both FNM 2013 and NNIS 2000 surveys show higher LA90,1hr levels than the NNIS 1990 over much of this period.


47065784_R2_03

August 2015

47

Comparison of the absolute noise levels at rush hour show:

The LAeq,1hr maximum noise levels during rush hour are lower for the FNM 2013 than the NNIS surveys (0.5 dB lower than the NNIS 2000 and 1.5 dB lower than NNIS 1990). The FNM 2013 data show a definite peak in the hour starting at 08:00 whilst the NNIS surveys both show a broader peak stretching into the hour starting at 09:00.

The LA10,1hr maximum noise levels during rush hour are lower for the FNM 2013 than the NNIS surveys (0.4 dB lower than the NNIS 2000 and 0.7 dB lower than NNIS 1990). All the surveys show a definite peak in the hour starting at 08:00.

The LA90,1hr maximum noise levels during rush hour occur in the hour starting at 08:00 for all the surveys, and at this maximum point the noise levels are very similar for all three surveys. The FNM 2013 levels are 0.3 dB lower than the NNIS 2000 and 0.2 dB higher than NNIS 1990.

These changes indicate that for FNM 2013 the morning increase in all three noise indicators starts slightly earlier than in the previous surveys even though the overall tendency is for lower noise levels in the FNM 2013 during the day and night, including the morning rush hour.

The highest LAeq,1hr noise levels associated with the rush hour show up as a peak in one hour only in the FNM 2013 data, whereas the duration of the rush hour peak was longer in both NNIS surveys, stretching later into the morning.

This analysis also indicates that since NNIS 1990 the LA90,1hr levels have begun increasing earlier in the morning. The NNIS 1990 LA90,1hr levels are markedly lower than the other two surveys prior to the 08:00 rush hour peak.

Five minute data

The changes between NNIS 2000 and FNM 2013 have been investigated further by means of the 5-minute noise levels presented in Figure 29. This shows that for both the NNIS 2000 and FNM 2013 the LAeq,5min and LA10,5min levels are similar around 03:35 when they begin to increase toward the morning rush hour. The FNM 2013 increase starts earlier as the noise levels are lower prior to this time.

FNM 2013 noise levels increase at a slightly greater rate than those from NNIS 2000 until around about 06:45 when FNM levels fall below the NNIS 2000 levels and stay below during the rush hour.

The LA90,5min changes are consistent with those of the LA90,1hr showing very similar levels between the surveys until the rush hour when the NNIS 2000 data is slightly higher than the FNM 2013. A small peak occurs in the NNIS 2000 data at 04:15 which is not seen in FNM 2013, although both surveys show flat points with no significant increase in level over a short period after 04:15.

The FNM 2013 rush hour maximum of 54.0 dB LAeq,5min is reached at 08:25, while the smoothed NNIS 2000 data peaks at 08:45 (55.4 dB LAeq,5min). The FNM 2013 LAeq,5min levels drop away from their maximum further than the NNIS 2000 levels.

3.2.4 Daytime patterns

Examination of the hourly data in Figure 28 between the end of the morning rush hour (around 10:00) and the evening rush hour (the hour starting at 17:00) shows noise levels just below those of the morning rush hour peak. It is noticeable that both NNIS surveys show a drop in LAeq,1hr LA10,1hr and

LA90,1hr levels in the early afternoon (around 14:00), this effect is not seen in the FNM 2013 data, where levels remain more constant to a later time, before dropping at a faster rate than that for either of the NNIS surveys.


47065784_R2_03

August 2015

48

3.2.5 Differences between day and night levels

Table 11 shows the longer term data averages for LAeq,T LA10,T and LA90,T over the time periods defined by Lday (07:00 – 19:00) and Lnight (23:00 – 07:00) over all 48 sites for each of the three surveys.

The LAeq,T data are a true energy average over the time periods and the LA90,T and LA10,T data are derived from the averaging of 1-hour indicators across the relevant time periods. For all indicators, the results presented are averages over the 48 sites.

TABLE 11. DAY AND NIGHT INDICATORS FOR NNIS 1990, NNIS 2000 AND FNM 2013

Time period Duration, T

NNIS 1990 NNIS 2000 FNM 2013

LAeq,T

(dB) LA10,T

(dB) LA90,T

(dB) LAeq,T

(dB) LA10,T

(dB) LA90,T

(dB) LAeq,T

(dB) LA10,T

(dB) LA90,T

(dB)

DAY 07:00 – 19:00

12 hours 56.2 56.4 43.1 55.7 56.7 43.6 55.0 55.8 43.5

NIGHT 23:00 – 07:00

8 hours 47.6 44.0 32.9 47.8 44.3 34.1 46.7 43.8 33.2

DAY - NIGHT

8.6 12.4 10.2 7.9 12.4 9.5 8.3 12.0 10.4

The surveys show similar differences between the night and day levels for the LAeq, LA90 and LA10 indicators:

LAeq – the difference varies from 7.9 dB in the NNIS 2000 survey to 8.6 dB in the NNIS 1990 survey, with the FNM 2013 at 8.3 dB.

LA10 – the difference is 12.4 dB in both NNIS surveys and the FNM 2013 difference is 12.0 dB.

LA90 – the difference is 9.5 dB in the NNIS 2000 survey, 10.4 dB for in FNM 2013 and between these is the NNIS 1990 survey (10.2 dB).

Differences in noise levels between day and night have been further investigated using the hourly data in Figure 28. This figure shows the maximum noise levels for all indicators occur during the morning rush hour and the minimum at 02:00 or 03:00 in the morning.

As an alternative approach to evaluate the difference in day and night noise levels the minimum night time noise level has been subtracted from the average noise level between 10:00 and 18:00. This is presented in Table 12.

TABLE 12. 10:00 TO 18:00 VS LOWEST NIGHT FOR NNIS 1990, NNIS 2000 AND FNM 2013

Time period Duration, T

NNIS 1990 NNIS 2000 FNM 2013

LAeq,T

(dB) LA10,T

(dB) LA90,T

(dB) LAeq,T

(dB) LA10,T

(dB) LA90,T

(dB) LAeq,T

(dB) LA10,T

(dB) LA90,T

(dB)

DAY 10:00 – 18:00

8 hours 54.7 56.2 43.0 54.9 56.7 43.5 54.0 55.7 43.4

MINIMUM NIGHT

1 hour 39.9 38.9 30.3 40.7 40.7 31.3 38.6 38.1 30.1

DAY - NIGHT

14.8 17.3 12.7 14.3 16.0 12.2 15.4 17.6 13.3

It is noticeable that the differences between the typical day noise level and the minimum night across all the indicators are lowest for the NNIS 2000 data and highest for the FNM 2013 data.


47065784_R2_03

August 2015

49


For each of the comparisons included within this section, graphs are presented showing the measured noise levels from the appropriate measurement survey plotted against the modelled noise levels from the strategic noise maps. Each such graph includes a blue line indicating where data would be expected to lie if the measured and modelled values matched exactly. Hence deviations above this line represent locations where the measured noise levels are greater than the modelled values, and points below this line represent locations where the measured noise levels are lower than the predicted values.

3.3.1 Round 1 (2006) Modelled v NNIS 2000 Measurements (Road)

Figure 30 below presents the Lden levels measured during the NNIS 2000 survey for which there is a corresponding FNM 2013 site, plotted against to the modelled Lden levels from the Round 1 (2006) modelled data. Figure 31 to Figure 34 illustrate the same for the Lday, Lnight, LAeq,16hr and LA10,18hr NNIS levels, respectively.

Figure 30: Round 1 Modelled Lden vs NNIS Lden levels for road traffic noise


47065784_R2_03

August 2015

50

Figure 31: Round 1 Modelled vs NNIS Lday levels for road traffic noise

Figure 32: Round 1 Modelled Lnight vs NNIS Lnight levels for road traffic noise


47065784_R2_03

August 2015

51

Figure 33: Round 1 Modelled LAeq,16hr vs NNIS LAeq,16hr levels for road traffic noise

Figure 34: Round 1 Modelled LA10,18hr vs NNIS LA10,18hr levels for road traffic noise

Figure 35 compares the distance to the nearest modelled source with the difference between the modelled and measured Lden levels (dB).


47065784_R2_03

August 2015

52

Figure 35: Difference between Round 1 Modelled and NNIS Lden vs Distance to Nearest Modelled Road Traffic Source

From Figure 30 to Figure 34 it can be seen that correlations between modelled noise level and NNIS 2011 measured levels are similar for each noise level indicator with a wide spread of results including sites where the predicted level is higher and lower than the measured values. From this small comparison it can be concluded that when considered in their strategic context, the overall results of the strategic noise mapping appear to give a reasonable representation of the road traffic noise environment.

Figure 35 can be used to identify any relationship between “under-prediction” / “over-prediction” from the strategic noise maps and distance from the closest modelled source. It might be expected that a greater underprediction would be seen with increasing distance from the closest modelled source, as other (non-modelled) noise sources are likely to become more significant. From the results in this figure, a slight tendency for over-prediction at closer distances and under-prediction at greater distances can be seen, but there are also a significant number of sites for which this assumption does not hold.

3.3.2 Round 1 (2006) Modelled v NNIS 2000 Measurements (Rail):

There was only one site within the FNM 2013 measurement sites for which Round 1 modelled railway noise data were available. No further analysis has been undertaken.

3.3.3 Round 2 (2011) Modelled v FNM 2013 Measurements (Road):

Figure 36 below presents the Lden levels measured during the FNM 2013 survey for which there is a corresponding NNIS 2000 site, arranged according to the modelled data from the Round 2 (2011). Figure 37 to Figure 40 illustrate the same for the Lday, Lnight, LAeq,16hr and LA10,18hr NNIS levels, respectively.


47065784_R2_03

August 2015

53

Figure 36: Round 2 Modelled Lden vs FNM Lden levels for road traffic noise

Figure 37: Round 2 Modelled vs FNM Lday levels for road traffic noise


47065784_R2_03

August 2015

54

Figure 38: Round 2 Modelled vs FNM Lnight levels for road traffic noise

Figure 39: Round 2 Modelled LAeq,16hr vs FNM LAeq,16hr levels for road traffic noise


47065784_R2_03

August 2015

55

Figure 40: Round 2 Modelled LA10,18hr vs FNM LA10,18hr levels for road traffic noise

Figure 41 compares the distance to the nearest modelled source with the difference between the modelled and measured Lden levels (dB).

Figure 41: Difference between Round 2 Modelled and FNM LAeq,18hr vs Distance to Nearest Modelled Road Traffic Noise Source

As with the results provided by the comparison between Round 1 modelled and the NNIS measured results, the graphs plotting the Round 2 modelled against FNM measured levels (Figure 36 to Figure 40) show a consistent pattern across all indicators. However, there appears to be a stronger correlation between the measured and predicted levels for this dataset, particularly for the Lnight noise indicator. Possible explanations for this difference include:

The reduced number of sources in Round 2 modelling (where only motorways and A-roads were generally included) will mean a greater number of sites experiencing lower noise levels will be excluded from this analysis due to the modelled noise levels being below 50 dB.


47065784_R2_03

August 2015

56

The absence of default traffic flows on smaller roads within agglomerations for the Round 2 maps will also mean a greater number of sites experiencing lower noise levels will be excluded from this analysis due to the modelled noise levels being below 50 dB

Data from FNM match the timing of the Round 2 mapping data more closely than NNIS matches that of Round 1.

The improved correlation during the night time may also relate to the presence of fewer non-transportation (and hence non-mapped) noise sources during the night time.

In reviewing the data shown in Figure 41, there is a wide range of under and over-predicted values at close distances, but where the closest modelled source is at a greater distance, there is a greater tendency for the modelled values to be lower than those measured. This is likely to be due to the increasing importance of other, non-modelled noise sources at greater distance from any modelled road.

3.3.4 Round 2 (2011) Modelled v FNM 2013 Measurements (Rail):

For Round 2 railway noise maps, data were only available for two FNM sites. Figure 42 shows a comparison of Round 2 modelled data against FNM 2013 measurements for these. This is only presented for the Lden noise indicator as this is the only indicator for which prediction data was available for both sites.

Figure 42: Round 2 Modelled vs FNM Lden levels for railway noise

Figure 43 shows the difference between the predicted and measured Lden values for these same two sites plotted against the distance to the nearest modelled source.


47065784_R2_03

August 2015

57

Figure 43: Difference between Round 2 Modelled and FNM LAeq,18hr vs Distance to Nearest Modelled Railway Noise Source

Due to the low number of data points in Figure 42 and Figure 43, it is not possible to make conclusive observations. It should be noted, however that the two points plotted in Figure 42 show that the measured levels for the two sites are both in excess of the equivalent modelled figures. At both of these measurement sites, railway noise was not noted as being a dominant noise source, so this apparent under-prediction is likely to be due to significant contributions from other sources.


47065784_R2_03

August 2015

58


3.4.1 Question 5

A5 In general, how do you feel about the amount of noise (or absence of noise) around here?

Definitely don’t like 1 2 3 4 5 6 7 Definitely like 8 Don’t know 9 Not answered

Figure 44 below presents the calculated Lday values from the FNM 2013 survey at each site arranged according to the response to Question A5. Figure 45 to Figure 47 show the same information for Levening, Lnight, LAeq,24h respectively.

Figure 44 A5: FNM 2013 responses vs Measured FNM 2013 Lday levels

Figure 45 A5: FNM 2013 vs FNM Levening Figure 46 A5: FNM 2013 vs FNM Lnight


47065784_R2_03

August 2015

59

Figure 47 A5: FNM 2013 vs FNM LAeq,24h

Figure 44 to Figure 47 show that the relationship between noise level and the FMN 2013 response is very similar for each of the noise level indicators considered. As this was seen for all questions considered, for the other questions analysed in this report, the information displayed has been limited to one noise indicator. This is generally the Lday noise indicator unless the FNM question was specifically about a different time period, in which case the indicator for the relevant time period has been used.

The responses were predominantly positive. One response was negative (response 1 or 2), fifteen made neutral responses (response 3, 4 or 5) and thirty-six selected positive responses (response 6 or 7).

There is broad correlation between measured noise levels and reported level of satisfaction. The ‘definitely like’ category is dominated by sites with relatively low noise levels (46 dB to 59 dB Lday) and the only respondent selecting ‘definitely don’t like’ corresponds to the highest level of 72 dB Lday. Between these two extremes on the response scale, there is less clarity, with a wide range of noise levels corresponding to each category on the response scale.

3.4.2 Question 6

A6 Would you say that you live in a noisy area?

Definitely yes 1 2 3 4 5 6 7 Definitely no 8 Not answered

Figure 48 below presents the Lday levels measured during the FNM 2013 survey at each FNM 2013 site arranged according to the response to Question A6.


47065784_R2_03

August 2015

60


The responses were predominantly positive. All respondents answered the question on the 1 (definitely no) to 7 (definitely yes) scale. Of these, thirty-six made the more positive responses (response 1 or 2), fourteen made neutral responses (response 3, 4 or 5), and two selected negative responses (response 6).

There is a clear correlation between measured noise levels and opinion of what constitutes a noisy area. Generally levels at locations where the residents identified that they lived in a noisier area tend to be higher. The wide range within each category indicate that the other factors than a simple single number noise level also influence people’s opinion as to whether an area is noisy.

3.4.3 Question 7

A7 How sensitive would you say you are to noise?

Not at all sensitive 1 2 3 4 5 6 7 Very sensitive 8 Not answered



47065784_R2_03

August 2015

61

Figure 49 A6 A7: FNM 2013 responses vs Measured FNM 2013 Lday levels

The responses were spread broadly across the range of answers, with a marginal lean towards the respondents being more sensitive than not.

There is no clear correlation between measured noise levels and reported sensitivity. An interesting observation, however, is that those responding ‘very sensitive’ generally experience noise levels towards the lower end of the range, with one response corresponding to the lowest measured level at any of the FNM 2013 sites (46 dB Lday).

3.4.4 Question 8

A8 Thinking about the last 12 months or so, when you are home, to what extent are you personally bothered, annoyed or disturbed by noise coming from outside your home?

1 Not at all 2 A little 3 Moderately 4 Very 5 Extremely 6 Not Answered

Figure 50 below presents the Lday levels measured during the FNM 2013 survey at each FNM2013 site arranged according to the response to Question A8.


47065784_R2_03

August 2015

62


The responses to this question are heavily concentrated within the two most positive answers (“Not at all” and “A little”), illustrating that the majority of respondents do not report being particularly bothered, annoyed or disturbed by the noise outside their property.

Figure 50 shows the majority of measured Lday levels (45 dB to 67 dB) are concentrated within the three least sensitive response bands. It is also interesting to note that the only respondent to state that they were “extremely” bothered, annoyed or disturbed corresponded with the highest measured external noise level (72 dB Lday) , but those responding “very” on the same scale were at properties with amongst the lowest measured external noise levels.

3.4.5 Question 9

Weekdays

A9(a) When you are at home does any noise coming from outside your home particularly bother, annoy or disturb you during the week (Monday to Friday)?

Not at all 1 2 3 4 5 6 7 Extremely 8 Not at home

This question included three separate responses on the above scale, for daytime (07:00-19:00), evening (19:00-23:00) and night-time (23:00-07:00) periods. The responses for each of these questions, plotted against the average noise level for the same time of day are shown in Figure 51 (daytime), Figure 52 (evening) and Figure 53 (night time). As this question specifically referred to weekdays, the noise levels used for this analysis have been taken as average values over the weekday measurement periods only.


47065784_R2_03

August 2015

63

Figure 51 A9(a): FNM 2013 responses vs Measured FNM 2013 weekday Lday levels

Figure 52 A9(a): FNM 2013 responses vs Measured FNM 2013 weekday Levening levels


47065784_R2_03

August 2015

64

Figure 53 A9(a): FNM 2013 responses vs Measured FNM 2013 weekday Lnight levels

Most of the responses to these question are concentrated within the three most positive answers (1 to 3), illustrating that the majority of respondents do not feel particularly bothered, annoyed or disturbed during weekdays by outdoor noise.

These figures do not show any notable correlation between the measured noise levels and the responses to the question, although there is a slight tendency towards higher measured noise levels with the responses reporting a higher degree of bother, annoyance or disturbance, particularly for the evening and night time periods.

Weekends

A9(b) When you are at home does any noise coming from outside your home particularly bother, annoy or disturb you during the weekend (Saturday and Sunday)?

Not at all 1 2 3 4 5 6 7 Extremely 8 Not at home

This question included three separate responses on the above scale, for daytime (07:00-19:00), evening (19:00-23:00) and night-time (23:00-07:00) periods. The responses for each of these questions, plotted against the average noise level for the same time of day are shown in Figure 54 (daytime), Figure 55 (evening) and Figure 56 (night time). As this question specifically referred to weekdays, the noise levels used for this analysis have been taken as average values over the weekday measurement periods only.


47065784_R2_03

August 2015

65

Figure 54 A9(b): FNM 2013 responses vs Measured FNM 2013 Lday levels

Figure 55 A9(b): FNM 2013 responses vs Measured FNM 2013 Levening levels


47065784_R2_03

August 2015

66

Figure 56 A9(b): FNM 2013 responses vs Measured FNM 2013 Lnight levels

As with Q9(a), the majority of responses to this question are concentrated within most positive responses,

These figures shows that there is not a strong correlation between the measured noise levels and the responses to the question, but there is a slight majority of lower measured noise levels within the positive responses of 1 and 2, although the measured levels at those sites where residents chose the responses towards the “Extremely” end of the scale are also relatively low and the site with the highest measured noise level also reported “not at all”.

3.4.6 Question 10

A10 Do you ever visit other outdoor place in order to find somewhere peaceful or quiet, for example, a park, country area or some other open space? Possible responses: 1 Yes (Park, country area, other open/outdoor space) 2 No



47065784_R2_03

August 2015

67


In comparing the two possible responses of ‘yes’ (1) and ‘no’ (2), the spread of results between the two is relatively even with similar numbers selecting each response. No particular correlation is seen between the measured noise levels and the responses to the question.


47065784_R2_03

August 2015

68


3.5.1 Differences between FNM and LNS measurement results

The FNM 2013 survey included measurements at five sites which were repeats of those used for the 1962 London Noise Survey (LNS). The precise locations of the measurements from the London Noise Survey were generally not suitable for long-term unattended noise measurements. Sites for the FNM were chosen to be as representative as possible of the locations used during the LNS, whilst standardising the measurement position to 1 m from a building façade.

Figure 58 below shows a comparison between the data from the LNS 1962 survey and the FNM. This is the average across these five sites and can be used to identify patterns within this cluster of sites.

Figure 58 Comparison of sites in London (LNS 1962 & FNM 2013)

Despite the limitations of the data available from the 1962 survey, these are interesting results. Both LA10 and LA90 noise levels are significantly lower in the current survey than those from the LNS over most time periods.

Figure 59 to Figure 63 show the data for the five individual sites in this cluster.

Figure 59. Site FL2F (LNS 1962 & FNM 2013)

Figure 60. FL3F (LNS 1962 & FNM 2013)


47065784_R2_03

August 2015

69

Figure 61. Site FL4D (LNS 1962 & FNM 2013)

Figure 62. Site FL5B (LNS 1962 & FNM 2013)

Figure 63. Site FL14A (LNS 1962 & FNM 2013)

The following sections of this report address the differences across these sites to identify whether other factors could be responsible for any of the apparent changes in noise level that are observed.

3.5.2 Sampling methodology assessment

The measurement methodology of the FNM 2013 utilized 1 hour long repeated sampling for the duration of the measurement at a given location. As the data from the 1962 London Noise Survey were based on a single short term (understood to be approximately 2 minutes) sample within each hour period, it has been suggested that the LNS sampling approach may result in significant differences between reported hourly LA10 and LA90 noise levels and hence be the cause of some of the differences seen between the 1962 and 2013 measurement data.

To examine the magnitude of this inherent difference, FNM data were used to extract a sequence of thirty consecutive two minute samples per hour of measurement. The 100ms time history for each of these samples was then used to calculate LA10 and LA90 noise indictors to replicate the sampling approach used for LNS. These values were then analysed in terms of the level differences that could occur among the samples within each hour.

Appendix C presents the data for the five sites, also showing the standard deviation among the 2 minute samples within in each hour. Based on this standard deviation, the magnitude of the possible discrepancies is assessed, giving an estimate of the likely differences errors which could occur simply by relying on a random 2 min sample to represent the levels of a given hour. These results are summarised in Table 13.


47065784_R2_03

August 2015

70

TABLE 13. RANGE OF POTENTIAL LA10 AND LA90 DISCREPANCIES AT THE FIVE LOCATIONS FOR THE DURATION OF THE MEASUREMENT BASED ON FNM 2013 DATA

Measurement Location Reference Difference between LAN,2min and LAN,1hr

LA10 (dB) LA90 (dB)

Average Range Average range

FL2F -1.0 +0.3 to -4.7 +0.9 +6.5 to +0.1

FL3F -1.0 +0.6 to -11.0 +0.7 +4.3 to +0.1

FL4D -1.1 +0.9 to -8.1 +0.9 +4.0 to -0.1

FL5B -0.5 +0.5 to -4.0 +0.7 +2.6 to -0.1

FL14A -1.0 +0.8 to -15.7 +0.8 +9.1 to +0.1

It can be seen from these values that there is a large range of differences in individual hours between the values of LA10 and LA90 which could be obtained from a 2-minute sample, when compared to the true 1-hour statistical noise levels. However, on average, the LA10 values are typically 1 dB below the true 1-hour value, whilst LA90 values are <1 dB above the true 1-hour value.

A further review of the figures shown in Appendix C identified the following patterns:

On average, daytime LA10,1hr values tend to be underestimated by use of a two minute sample.

On average, night time LA90,1hr values tend to be underestimated by use of a two minute sample.

The average differences for night time LA10,1hr and daytime LA90,1hr values are smaller.

A large spread of results is seen for the majority of hours, with possible results from a two minute sample being either significantly greater or significantly lower than the true one our values.

The variations across two minute samples within a one hour period are generally greater during the night time than those seen during the daytime.

From the analysis above, it is possible that some of the variation between the FNM and LNS measurement results could be explained by this difference in sampling methodology. However, such differences would be expected to be random in nature, accounting for no more than a 1 dB increase in LA10 values and a 1 dB reduction in LA90 values. Given that the variations between the FNM and LNS data follow a consistent pattern (i.e. FNM values are consistently lower than LNS values), it is likely that this would not result in changes of the pattern and magnitude seen in the measured data.

3.5.3 Site location

The measurement locations were examined to determine if there are any significant differences in the precise site location, landscape and surrounding built environment, as recorded in the LNS and FNM. It should be noted that the measurement approach was matched to the extent possible between the two surveys to facilitate comparison of results; however, the precise measurement location of the LNS is not always known.

Appendix D provides a summary of each measurement position, including a copy of the site summary from the LNS report, and a current aerial photograph indicating the locations of the LNS and FNM monitoring positions.

The following key findings identified in a review of these site summaries, together with the measured noise levels from both surveys (as set out in Section 3.5.1) are given below:


47065784_R2_03

August 2015

71

FL2F

Both LA10,1hr and LA90,1hr are some 10 to 15 dB lower in FNM than LNS, with the difference being smallest for LA10 indicators during the early hours of the morning.

Both monitoring locations are at the front of properties on Mornington Terrace and likely to be exposed to similar noise sources.

It is not known whether the LNS location was free-field or façade, while the FNM location was 1 m from the building façade. If the LNS position were free field, this would result in levels at the LNS position being some 2.5 to 3 dB lower than those which would have been recorded in the FNM position.

Mornington Terrace is now a one way road. This may indicate that there would be less traffic on this street than was the case during the LNS measurements.

Railway noise was noted as being particularly significant during the the night-time.

FL3F

Both LA10 and LA90 are some 10 to 15 dB lower in FNM than LNS during the daytime, with minimal difference during the night time and early hours of the morning. This is particularly evident for LA90 where the measurements from the two surveys are within 1 to 2 dB).

There was significant variation in LA10 during the daytime LNS measurements.

Noise from road drills was noted during the daytime measurements from LNS, which may be the cause of the variation noted. However evening noise levels, and the lowest daytime values from LNS (which were likely to exclude these works) remain significantly above the FNM levels.

The LNS monitoring location was just off one corner of Vincent Square, whilst the FNM location was on the square itself. From observations during the FNM site work, road traffic noise around all sides of Vincent Square would be expected to be similar, with more distance sources, and activities in the playing fields within the square contributing significantly to the sound environment.

The LNS site may have been slightly more exposed to noise from Rochester Row, to the West. However, the FNM site would be less screened from noise from the busy Vauxhall Bridge Road to the South.

It is not known whether the LNS location was free-field or façade (though given the location, a façade position is considered likely), while the LNS location was 1 m from the building façade. If the LNS position was free field, this would result in levels logged at the LNS position being some 2.5 to 3 dB below those which would have been logged at the FNM position.

FL4D

LA90 noise levels from FNM are typically some 10 dB below those from LNS throughout the day. There is much less difference between LA10 levels, although these are also lower than the LNS, particularly during the night time.

Both monitoring locations are on the same street, though on opposite sides, so would be exposed to similar noise sources.


47065784_R2_03

August 2015

72

It is not known whether the LNS location was free-field or façade, while the LNS location was 1 m from the building façade. If the LNS position were free field, this would result in levels at the LNS position being some 2.5 to 3 dB below those at the FNM position.

The LNS data showed distinct peaks in level during morning and afternoon rush hours, which were noted as likely to be due to increased through traffic. These same peaks were not apparent in the FNM data.

FL5B

Daytime LA10 noise levels from FNM are some 15 dB below those from LNS, with daytime LA90 levels some 7 to 8 dB below the LNS values.

Night time LA10 noise levels from LNS are some 15 to 20 dB below those from FNM, with night time LA90 levels comparable for both surveys. This is due to the large drop in noise levels between day and night from LNS not being so apparent in the FNM data.

Both monitoring locations are dominated by noise from the same main road (Grove Road / Lauriston Road, A1205). The LNS monitoring location was slightly closer to this road than the FNM location, as it was not practical to undertake the longer duration FNM measurements at the precise LNS location. It is estimated that the distance to this road from the FNM location may be two to three times that from the LNS location. Based on simple noise propagation from a line source, this would be expected to result in a difference of 3 to 5 dB between measured noise levels. This could go some way to explain the differences measured, but does not provide sufficient explanation for all of the observed differences.

Given the location of the LNS measurements, it is considered likely that these were free-field, whilst the position for the FNM was 1m from the building façade. This would result in levels at the FNM position being some 3 dB greater than would be experienced for an equivalent free-field position, counteracting the difference due to distance from the traffic noise source identified above to some extent.

FL14A

LA10 noise levels from the two surveys are similar, whilst LA90 levels show a decrease of 2 to 3 dB from LNS to FNM across all periods of the day.

Both monitoring locations are on the same street, so would be exposed to similar noise sources.

Road traffic, and many other sources, including people in the street were noted during both surveys. Railway works at a distance of 200 yards was reported in the LNS data, and is believed to relate to the peaks in LA10 noise levels seen in this survey, and unlikely to effect the LA90 levels.

Overall, some significant factors have been identified which may explain a proportion of the changes seen at some of these five sites. However, it is not considered that any of these causes are sufficient to explain the full extent of the changes in noise levels between the LNS and FNM surveys. These factors and the remaining differences are summarised in Table 14 below.


47065784_R2_03

August 2015

73

TABLE 14. SUMMARY OF FACTORS INFLUENCING CHANGES IN NOISE LEVEL

Measurement Location Reference Factors with potential to influence changes

in noise level Changes in noise values

FL2F Mornington Terrace change to one way traffic 15 dB reduction in LA10 and LA90 unlikely to all

be due to reduced traffic.

FL3F

LNS and FNM locations may be exposed differently to more distant major road traffic

noise sources. Noise from roadworks potentially affecting

daytime LNS LA10 noise levels.

Significant reduction in daytime levels. Local road traffic sources likely to be similar. Levels

in evening unlikely to be influenced by roadworks.

FL4D No significant differences identified LA90 noise levels reduced by some 10 dB.

Peaks in level for rush hour periods reduced.

FL5B Potential reduction of 3 to 5 dB explained by increased distance from major road source.

Daytime LA10 noise levels from FNM are some 15 dB below those from LNS, with daytime LA90

levels showing a reduction of 7 to 8 dB.

FL14A Railway works during LNS measurements may have influenced LA10 levels for some

periods.

LA90 levels show a decrease of 2 to 3 dB from LNS to FNM

3.5.4 Historic data

Historic noise policy interventions, as reported in the Defra research project NO023411

, were reviewed to determine any relevance to explaining the sound level differences between the two noise surveys examined. In view of the subjective descriptions of the sound environment noted at the time of monitoring, interventions relating to road traffic and air traffic noise were deemed most relevant.

Road traffic noise

The noise emission for the different classes of road vehicles have been significantly reduced over the years owing to EC directives on vehicle noise limits. This Defra research project concluded , given that the limits are not applied retrospectively ‘the net effect of the policy (after being partially offset by traffic growth) is estimated to be a 2 dB fall in traffic noise for motorways, a 5 dB reduction in traffic noise for

A-roads and minor roads’ 12

. However, no differentiation is given in the expected changes for different

noise indicators.

The factors which have influenced road traffic noise levels identified by this report were:

road surface characteristics;

regulation of tyre noise;

speed limits;

lorry bans or restrictions; and

congestion charging zones

The overall potential for road traffic reduction from newly manufactured tyres given the available technology is 2-4 dBA, however it is generally acknowledged that while this is already influencing noise levels ‘it could be a decade or more for the benefits of quieter tyres to become apparent’

12.

Differences in the road surface texture can further alter road traffic noise levels significantly. For ‘low noise surfaces’, a typical correction is 3.5 dBA

12.

While the effect of road surface type cannot be quantified for the FNM measurement sites, it would be feasible for this to have contributed to the changes in noise level identified.


47065784_R2_03

August 2015

74

In addition to the sources considered in this review, urban environment reductions in vehicle engine, powertrain and exhaust noise may also contribute significantly to the reductions in measured noise level from LNS to FNM.

It is also likely that changes in traffic patterns over the period will have had a significant effect on measured noise levels, and this is considered the most likely caused for the reduced difference between daytime and night time noise levels seen at several sites

Air traffic noise

The introduction of the regulations contained in Annexe 16 to the International Convention on Civil

Aviation had the effect of beginning the process of substantial reduction in aircraft noise levels13

.

In effect, there was a substantial reduction of areas within the 57 dB LAeq,16hr contour, even when accounting for increased air traffic movements. Indicatively, the areas falling within this noise contour were reduced by 87% for Heathrow Airport (1972-2009), 78% for Gatwick Airport (1972-2009), 21% for Stansted Airport (1988-2009) and 81% for Luton Airport (1976-2010). Given these differences, a reduced noise level could be expected for any area where air traffic is a dominant noise source. However, as aircraft noise was not noted to be dominant at any sites, and only a contributing factor, it is considered unlikely that changes in aircraft noise level would be the cause of the reduced noise levels seen.

Any reduction in noise levels from individual aircraft overflights would also be expected to influence measured LA10 noise levels, but would not be expected to have a significant effect on the measured LA90 noise levels. This indicates that changes in aircraft noise alone cannot be responsible for the changes seen in measured noise levels.


47065784_R2_03

August 2015

75

4 SUMMARY

Overview 4.1

This report has presented details of further analysis of the noise measurement data from the FNM 2013 project. The report consists of five elements, including comparisons with data from the following previous projects:


(B) National Noise Incidence Survey 1990, (including further analysis of diurnal patterns in noise levels);

(C) Strategic Noise Maps

(D) Self-completion questionnaires from FNM 2013

(E) London Noise Survey 1962

The sample sizes for all of these analyses are relatively small, and due to the nature of the sample, results cannot be considered to be representative of a regional or national picture. However, the results provide an interesting and useful snapshot across a number of measurement sites.


This element presented a comparison of subjective responses to the National Noise Attitude Survey 2012 with measured noise levels from the Focussed Noise Monitoring 2013 project at sites where both surveys overlapped. Due to the small sample size, no differentiation has been made between sites where the FNM 2013 measurements were undertaken at exactly the same address as the NNAS 2012 questionnaire, and those where the measurements were at an address considered to be acoustically equivalent to that used for NNAS 2012.

For the majority of questions analysed, no clear relationships were seen between the question response and measured noise levels (using LAeq indicators for the most appropriate time of day in relationship to the specific question). A few interesting findings were identified:

A slight tendency for those who reported definitely liking the amount of noise around their home to experience lower external noise levels, whilst those reporting more neutral responses to this question experienced higher levels. Of the sample available, no respondents reported definitely disliking the amount of noise around their home.

A notable pattern in self-reported sensitivity to noise against measured external noise levels, with those reporting themselves less sensitive to noise generally experiencing higher noise levels than those reporting themselves to be more sensitive. This does not confirm the presence of a causal link between these factors, as it is possible either that sensitivity to noise influences the choice of where people live or that acclimatisation to the external noise environment influences reported noise sensitivity.

When considering respondents who were not asked the detailed questions relating to road traffic noise as they did not report being at all bothered, annoyed or disturbed by any form of road traffic noise, it has been noted that these properties were generally exposed to lower measured external noise levels than those who did report some degree of bother, annoyance or disturbance from road traffic noise.

When looking at more detailed responses to road traffic noise, very few notable patterns have been seen. In general, all responses are associated with a wide range of external noise levels, and given the small sample size it has not been possible to identify and clear patterns.


47065784_R2_03

August 2015

76

National Noise Incidence Survey 1990 and The Analysis Of Diurnal Patterns 4.3

This element presented an analysis of long term changes in noise levels between 1990 and 2013, drawing on measurement data from NNIS 1990, NNIS 2000 and FNM 2013.

It should be noted that this analysis is based on a relatively small number of sites. The sampling strategy for FNM 2013 was not designed to provide a nationally representative sample. Hence, caution should be applied in extrapolation to provide any indication of any national patterns, and the results should simply be considered to provide a picture of changes in noise levels at the sites visited.

Changes in noise level between 1990 and 2000 were seen to generally match the patterns reported for the whole of England and Wales in NNIS 2000, with slight reductions in daytime noise level, and slight increases in night time levels (particularly for LA90 noise indicators). However, none of these changes were statistically significant within the smaller sample size considered in this report.

The FNM 2013 noise levels were found to be lower than those for the same sites in either 1990 or 2000, with the differences in the majority of LAeq noise indicators being statistically significant. When considering LA10 and LA90 noise levels, the same pattern was seen, with FNM 2013 being lower than those from 1990 or 2000, although fewer of these changes were found to be statistically significant.

A further analysis of diurnal patterns of noise levels has been undertaken, with the following key findings:

LAeq noise levels over the evening period are very similar in 1990 and 2000, but with a slower rate of decrease from the daytime levels in 2000 than for 1990. In 2013, the levels have reduced slightly overall, and the rate of decrease has returned to be comparable with that measured in 1990.

The quietest night time noise levels were found to be highest in the NNIS 2000 survey and lowest in the FNM 2013.

The rate of increase in noise levels through the morning (up to the “rush hour” period) is generally similar in each survey, although the FNM 2013 shows a sharper increase in noise levels in the early hours of the morning than either of the NNIS surveys. This rate of increase then tails off toward the maximum rush hour levels.

For FNM 2013 the morning increase in noise levels starts slightly earlier than in the previous surveys even though the overall pattern is for lower noise levels in the FNM 2013 during the day and night, including the morning rush hour.

The highest LAeq,1hr noise levels associated with the rush hour show as a peak in one hour only in the FNM 2013 data, whereas the duration of the rush hour peak was longer in both NNIS surveys, stretching later into the morning.

The difference between the typical day and the minimum night noise level across all the indicators are lowest for the NNIS 2000 data and highest for the FNM 2013 data.


This element presented a comparison of measured noise levels from FNM 2013 and NNIS 2000 with strategic noise mapping data from Round 1 and Round 2 at FNM 2013 sites for which these data are available.

When considering the differences between these two data sources, it is important that the purpose of the noise maps, and the methods by which they have been generated are considered. Of particular importance are the following factors:


47065784_R2_03

August 2015

77

These noise mapping data were produced for strategic purposes, and is not necessarily considered to be accurate at a very local level

The levels obtained from the strategic noise maps are generated for specific sources only, and these will vary between different areas of the country. Predicted levels used in this analysis are also for road traffic and railways only, and levels from these sources are considered separately.

There are significant differences between the road traffic noise sources modelled for Round 1 and Round 2 noise maps in certain locations.

A general finding is noted that apparent “under-prediction”, where the measured noise levels are greater than those from the noise maps occurs increasingly with greater distance from modelled noise sources. It is likely that this is due to the increasing importance of other non-modelled noise sources in these locations.

It has also been noted that the correlation between measured and predicted noise levels is generally closer for the Round 2 data than for Round 1. This is most notable during the night time. Possible explanations for this difference include:

The reduced number of sources in Round 2 modelling (where only motorways and A-roads were generally included) will mean a greater number of sites experiencing lower noise levels will be excluded from the analysis included in this report due to the modelled noise levels being below 50 dB.

Data from FNM match the timing of the Round 2 mapping data more closely than NNIS matches that of Round 1.

The improved correlation during the night time may also relate to the presence of fewer non-transportation (and hence non-mapped) noise sources during the night time.


This element presented an analysis of the responses to key questions from the FNM 2013 self-completion questionnaire, comparing these responses to measured noise levels at each site.

For all questions considered, there is considerable variability between individual responses and measured noise levels. However, the following were identified:

A slight tendency towards respondents at sites experiencing lower noise levels being more likely to report that they liked the amount of noise in the area.

A stronger correlation between measured noise levels and reporting of respondents as to whether they thought they lived in a noisy area

When responses to questions regarding noise annoyance were considered, no clear correlation between subjective responses and measured noise levels was apparent.


This element presented a detailed review of a number of factors which potentially influence changes in noise level between noise measurements undertaken for the 1960s London Noise Survey (LNS) and those completed as part of the recent Defra Focussed Noise Monitoring (FNM) project undertaken by URS in 2013.

A total of five sites were included in the FNM project which had previously been used for the LNS measurements.

Key findings were:


47065784_R2_03

August 2015

78

1) Measurement approach

The LNS measurements were based on a sample of approximately 2 minutes within each 1-hour period. This sampling approach has been demonstrated to be associated with a large random variation in noise levels from the “true” noise levels based on a full 1-hour measurement.

However, on average the difference between the two approaches tends to be small, so this factor would not be expected to result in significant systematic difference between the two surveys, but simply result in greater hour-to-hour variation in the LNS data

Small systematic differences were identified in the results, with average daytime LA10,1hr values tending to be underestimated by use of a two minute sample, and average night time LA90,1hr values tending to be underestimated by use of a two minute sample. However these difference were small (typically around 1 dB), in comparison to the changes in noise level between the two surveys.

2) Measurement locations and significant noise sources

Subjective descriptions and site details from the LNS data set showed a high degree of similarity to the 2013 FNM survey.

Changes in the noise sources that could be partly responsible for the difference in noise levels between the surveys were found for two locations (one where temporary works may have influenced the LNS measurements, and one where the local road had become one way).

Changes in location were considered unlikely to provide explanations of the differences between measured noise levels at the majority of sites. At one site, a reduction of between 3 and 5 dB could be explained by an increased distance from the closest major road. However, it is considered likely that half of this difference would be counteracted by the fact that the FNM project used a façade measurement position, whilst the LNS is considered likely to have been a free-field location.

3) Historic noise policy interventions

Historic noise policy interventions affecting road traffic source noise level were established as a significant contributor in the reduction of noise levels that could account partly for the differences between the two noise surveys.

It was not possible to quantify the cumulative effect of the different factors that could be responsible for the noise reduction throughout the period 1962-2013. All of the factors identified would most likely have an impact, with the most prevalent being the historic noise policy interventions which can directly account for a 5dB+ reduction in the locations of interest, adding to the undefined impact of location specific characteristics (such as temporary construction noise during the LNS monitoring).

It is also likely that changes in traffic patterns over the period will have had a significant effect on measured noise levels, and this is considered the most likely caused for the reduced difference between daytime and night time noise levels seen at several sites.

The influence of the sampling methodology remains a random characteristic which could have a significant effect on an individual 1-hour measurement period, but is unlikely to have resulted in a significant effect when averaged over all individual measurements


47065784_R2_03

August 2015

79

5 REFERENCES

1

Billin H, Skinner C, Tickner C, Focussed Noise Monitoring 2013, URS Report Reference

47065784.R1/02, October 2013

2 Sargent JW, Fothergill LC, The noise climate around our homes, BRE Information Paper IP21/93,

1993.

3 Wright P, Skinner CJ, Grimwood CJ, The National Noise Incidence Study 2000 (England and

Wales), BRE client report number 203938f

4 Skinner CJ, Grimwood CJ, The National Noise Incidence Study 2000/2001 (United Kingdom):

Volume 1 – Noise Levels, BRE Client report number 206344f

5 Skinner CJ, Jackson T, Grimwood CJ, Raw G, The National Noise Incidence Study 2000/2001

(United Kingdom): Volume 2 – Self Completion Questionnaires, BRE client report number 206345f

6 National Noise Attitude Survey 2012/2013, Currently unpublished Defra noise attitude survey

7 London Noise Survey 1961-1962, Extracts from report provided by Defra

8 Walls C, Skinner C, Review of ambient noise levels and the impact for noise maps as a result of

licensing law changes - baseline noise survey Nov/Dec 2004 BRE client report number 220170

9 Licensing Act 2003, The Stationery Office, 2003

10 Directive 2002/49/Ec Of The European Parliament And Of The Council, of 25 June 2002 relating to

the assessment and management of environmental noise;

11 Taylor, R, Defra project NO 0234, An investigation into the effect of historic noise policy

interventions, 2012

12 Defra project NO 0234, An investigation into the effect of historic noise policy interventions, Annex 3:

Road traffic noise, 2012

13 Defra project NO 0234, An investigation into the effect of historic noise policy interventions, Annex 2:

Aircraft noise, 2012


47065784_R2_03

August 2015

80

APPENDIX A: NOISE PERCEPTION AND TERMINOLOGY

Between the quietest audible sound and the loudest tolerable sound there is a million to one ratio in sound pressure (measured in Pascals, Pa). Because of this wide range, a noise level scale based on logarithms is used in noise measurement called the decibel (dB) scale. Audibility of sound covers a range of approximately 0 to 140 dB. The human ear system does not respond uniformly to sound across the detectable frequency range and consequently instrumentation used to measure noise is weighted to represent the performance of the ear. This is known as the 'A weighting' and annotated as dB(A). Table A.1 lists the sound pressure level in dB(A) for common situations.

TABLE A.1: NOISE LEVELS FOR COMMON SITUATIONS

Typical noise level, dB(A) Example

0 Threshold of hearing

30 Rural area at night, still air

40 Public library, refrigerator humming at 2m

50 Quiet office, no machinery

60 Normal conversation

70 Telephone ringing at 2m

80 General factory noise level

90 Heavy goods vehicle from pavement

100 Pneumatic Drill at 5m

120 Discotheque – 1m in front of loud speaker

140 Threshold of pain

The noise level at a measurement point is rarely steady, even in rural areas, and varies over a range dependent upon the effects of local noise sources. Close to a busy motorway, the noise level may vary over a range of 5 dB(A), whereas in a suburban area this variation may be up to 40 dB(A) and more due to the multitude of noise sources in such areas (cars, dogs, aircraft etc.) and their variable operation. Furthermore, the range of night-time noise levels will often be smaller and the levels significantly reduced compared to daytime levels. When considering environmental noise, it is necessary to consider how to quantify the existing noise (the ambient noise) to account for these second to second variations.

A.1 Ambient or Activity Noise Levels

The equivalent continuous A-weighted sound pressure level, LAeq (or Leq dB(A)) is the single number that represents the total sound energy measured over that period. LAeq is the sound level of a notionally steady sound having the same energy as a fluctuating sound over a specified measurement period. It is commonly used to express the average energy level from individual sources that vary in level over their operational cycle.

A.2 Lden, Lday, Levening, and Lnight The day, evening, night level, Lden, is an annual average noise level based on the LAeq, but with a 5 dB weighting for evening (19:00 to 23:00) and a 10 dB weighting for night (23:00 to 07:00).

𝐿𝑑𝑒𝑛 = 10 × log10 {12 × 10

𝐿𝑑𝑎𝑦10⁄ + 4 × 10

(𝐿𝑒𝑣𝑒𝑛𝑖𝑛𝑔+5)10

⁄+ 8 × 10

(𝐿𝑛𝑖𝑔ℎ𝑡+10)10

⁄

24}

Where:

Lday is the 12-hour LAeq noise level over the period 07:00 to 19:00

Levening is the 4-hour LAeq noise level over the period 19:00 to 23:00


47065784_R2_03

August 2015

81

Lnight is the 8-hour LAeq noise level over the period 23:00 to 07:00

A.3 Noise Changes

Human subjects are generally only capable of noticing changes in noise levels of no less than 3 dB(A). It is generally accepted that a change of 10 dB(A) in an overall, steady noise level is perceived to the human ear as a doubling (or halving) of the noise level. (These findings do not necessarily apply to transient or non-steady noise sources such as changes in noise due to changes in road traffic flow, or intermittent noise sources).


47065784_R2_03

August 2015

82

APPENDIX B: DERIVATION OF NOISE INDICATORS FOR ANALYSIS AND OVERALL SITE SUMMARY

In line with the approach taken in the main FNM report, the various noise indicators used in this analysis have been derived from the raw measurement data for each site as follows:

Average 24-hour time histories have been calculated by means of averaging of 1-hour data for each of Monday-Thursday, Friday, Saturday, Sunday and for the complete measurement period,

Lday, Levening, LAeq,16hr, Lnight, LAeq,24hr, and Lden have been calculated separately from these average 24-hour time histories for each of these periods.

Data for weekdays have been calculated as a weighted average of the Monday to Thursday and Friday values in order to remove any influence due to the measurement period not being exactly one week in duration.

Likewise, data for weekends have been calculated as an average of the Saturday and Sunday values.

In defining data for the different times of the week, the following definitions have been taken:

Thursday night (i.e. through to 07:00 on Friday morning) has been included in the Monday to Thursday data;

Friday night (i.e. through to 07:00 on Saturday morning) has been included in the Friday data;

Saturday night (i.e. through to 07:00 on Sunday morning) has been included in the Saturday data; and

Sunday night (i.e. through to 07:00 on Monday morning) has been included in the Sunday data;

This approach has been taken to ensure consistency with the values used within the analysis contained in the main FNM report.

Table B.1 shows minimum, mean and maximum external noise levels measured for the 19 FNM 2013 sites relating to NNAS 2012 for Lday, Levening, Lnight , LAeq,16h and LAeq,24h. The Lday, Levening, Lnight have been split into weekday and weekend periods These results present the minimum and maximum values when averaged across the relevant week/weekend days, rather than the minimum and maximum levels on a single day/evening/night time period. They are provided as aid to compare the noise indicators for all 19 sites to those of the subsets of sites discussed in the analysis of the individual questions.


47065784_R2_03

August 2015

83

TABLE B.1. FNM 2013 EXTERNAL NOISE LEVELS MEASURED AT NNAS 2012 SITES

Noise Descriptor Days

Time period (T) LAeq,T (dB)

Minimum Mean Maximum

Lday

Complete FNM measurement

period

07:00 – 19:00 44 54 72

Levening 19:00 – 23:00 39 51 69

Lnight 23:00 – 07:00 37 46 61

LAeq,16h 07:00 – 23:00 44 54 71

LAeq,24h 07:00 – 07:00 43 52 70

Lday Weekday

(Monday to Friday)

07:00 – 19:00 45 55 73

Levening 19:00 – 23:00 41 51 69

Lnight 23:00 – 07:00 38 47 62

Lday Weekend

(Saturday and Sunday)

07:00 – 19:00 45 53 71

Levening 19:00 – 23:00 38 50 68

Lnight 23:00 – 07:00 36 47 62


47065784_R2_03

August 2015

84

APPENDIX C: EFFECTS OF 2 MINUTE SAMPING ON LA10 AND LA90 NOISE INDICATORS FOR LONDON NOISE SURVEY

This appendix presents graphs showing the effects of 2 minute sampling on LA10 and LA90 noise indicators. For each site the graphs present the “true” value of LA10,1hr or LA90,1hr for each hour of the FNM monitoring period, compared against the average of sampled LA10,2min or LA90,2min from all 2 minutes samples within the 1-hour period. For this second line, the error bars displayed show the standard deviation of the individual 2-minute sample values calculated.

L2F


47065784_R2_03

August 2015

85

FL3F


47065784_R2_03

August 2015

86

FL4D


47065784_R2_03

August 2015

87

FL5B


47065784_R2_03

August 2015

88

FL14A


47065784_R2_03

August 2015

89

APPENDIX D: LONDON NOISE SURVEY LOCATION DATA AND CURRENT AERIAL VIEW FL2F

Contains Ordnance Survey Data. Crown Copyright and database right 2014. Reproduced from Ordnance Survey digital map data. Crown copyright 2014. All rights reserved. Licence number 0100031673. Copyright Natural England 2014. Material is reproduced with the permission of Natural England 2014. Copyright English Heritage 2014. Reproduced under the terms of the Click-Use Licence. © URS 2014.

FNM

LNS


47065784_R2_03

August 2015

90

FL3F


FNM

LNS


47065784_R2_03

August 2015

91

FL4D


FL5B

FNM

LNS


47065784_R2_03

August 2015

92


FNM LNS


47065784_R2_03

August 2015

93

FL14A


FNM

LNS

FNM 2013 - GOV.UK

Documents

Transcript of FNM 2013 - GOV.UK