Kopi fra DBC Webarkiv

(1)

Kopi fra DBC Webarkiv

Kopi af:

Rapport : Community response to noise reducing road pavements

Dette materiale er lagret i henhold til aftale mellem DBC og udgiveren.

www.dbc.dk

e-mail: dbc@dbc.dk

(2)

Community response to noise reducing road pavements

Rapport nr. 502

(3)

Community response to noise reducing road pavements.

Authors:

Torben Holm Pedersen, Guillaume Le Ray, DELTA SenseLab, Hans Bendtsen, Jørgen Kragh, Danish Road Directorate Date:

February 2014 ISBN (NET):

978-87-7060-269-3 Copyright:

Vejdirektoratet, 2014

(4)

Preface

The Danish Road Directorate has conducted a project where questionnaires on noise annoyance have been mailed to residents before and after replacing an old road surface with a new noise reducing wearing course. The results of this project are documented in Danish Road

Directorate report 442 in Danish: “Befolkningsreaktioner på støjreducerende vejbelægninger”. The present report describes the main results in English, based on a pres- entation given at the Internoise 2013 conference.

(6)

5 Fo ro rd

Forord

på dansk i rapport 442 fra Vejdirektoratet. Nærværende rapport resumerer projektets hovedresultater på engelsk.

Rapporten er en let bearbejdet udgave af en præsentation fra Internoise konferencen i 2013.

Vejdirektoratet har gennemført et projekt, hvor der før og efter udlægning af støjreducerende asfalt blev gennem- ført spørgeskemaundersøgelser af beboernes oplevelse af støjgenerne. Projektets resultater er dokumenteret

(7)

6 S am m e n f a t nin g

Sammenfatning

I 2007 blev der udført belægningsarbejder på dele af Kastrupvej på Amager samt på dele af Frederikssundsvej og Mørkhøjvej i Husum. De gamle slidte vejbelægninger af typen tæt asfaltbeton blev udskiftet med nye støjredu- cerende tynde slidlag. Støjmålinger viste, at støjen langs disse vejstrækninger blev reduceret med gennemsnitlig 4 dB. Vejdirektoratet har sammen med Københavns kom- mune gennemført en spørgeskemaundersøgelse før og efter udlægningen. Det primære formål var at klarlægge, hvilken reduktion af støjgener som kan opnås ved brug af støjreducerende tynde slidlag. Svarprocenten i før- og efter-undersøgelserne var henholdsvis 38 % og 44 % hvilket må anses for rimeligt godt for denne type undersø- gelser. Før-undersøgelsen blev gennemført i maj 2007 før vejarbejderne blev påbegyndt. Efter-undersøgelsen blev gennemført maj 2008 efter at beboerne havde haft et år til

at vænne sig til den nye situation. Støjgenerne var blevet mindre efter at der var udlagt ny asfalt. Den del af beboerne der var kraftigt og ekstremt generet faldt fra 21 % til 11 %. Den del der var moderat generet faldt fra 27 % til 24 %. De væsentligste konklusioner er at:

1. Det giver en statistisk signifikant reduktion af støjge- nerne at udlægge støjreducerende slidlag

2. Selv om trafikstøjen stadig er den største gene fra trafikken i de to undersøgte områder, er andelen af kraftigt og ekstremt generede væsentligt reduceret med det støjreducerende slidlag

3. Generelt var ca. 10 % færre generet af trafikstøjen i eftersituationen og 7 % flere syntes at støjniveauet fra trafikken var blevet acceptabelt.

(8)

7 A bs t r a c t

reducing type. Even if the traffic noise still was the major cause of annoyance from the traffic the fraction of very annoyed and extremely annoyed persons was reduced. In general there were 10% fewer persons annoyed from the traffic noise in the after situation. It was also found that the dose-response curves were moderated by respondents’

noise sensitivity, age and concerns about accidents.

The general dose-response curves were the same in the before and the after situation. The curves were compared to the international (”Miedema”) dose response curves. It was found that the Danish respondents were 3-6 dB more sensitive to noise.

Abstract

Noise annoyance was investigated by social and socio- acoustic surveys before and after the replacement of old pavements by new noise reducing wearing courses on primary roads in two areas of Copenhagen. Questionnaires were sent out and the results are based on answers from 2870 respondents (41% of the questionnaires that were sent out).

For each residence, the traffic noise exposure (L_den) was calculated before and after the repaving. The community responses were analysed together with the noise levels.

Statistical significant reductions of the noise annoyance were found after having replaced the pavement by a noise

(9)

8 01 I n t ro du c t io n

on the noise annoyance for this type of pavement. Further- more the influence of moderators on the noise annoyance such as noise sensitivity, gender, age, children in the household and the type of dwelling was investigated.

A full report (in Danish) on the work can be found in [1].

When the pavement on two major roads (Frederikssunds- vej including some neighbouring roads and Kastrupvej) in Copenhagen was renewed with noise reducing thin layer asphalt, questionnaire surveys on noise annoyance were performed before and after the repaving. The primary pur- pose was to investigate the effect of the noise reduction

1 Introduction

(10)

9

2 M e t ho d of I nve s t i g a t io n

2.1 Areas and noise exposure

The two areas of the investigation were characterized mainly by multi storey buildings in open plan but also with small areas with detached family houses. The road surfaces in the before situation were 8 years old asphalt concrete having 11 mm maximum aggregate size. In the after situation the pavements were new noise reducing thin asphalt layers. At the repaved roads, an average noise reduction of 4 dB was measured. Details on the pavements can be found in [2] and [3].

The noise exposure (the day-evening-night level, L_den) at the facades of the buildings (not including the façade reflection) was calculated for all the respondents’ addresses by means of the road traffic noise prediction method, Nord2000 [4].

2.2 Survey

7000 questionnaires were sent out by mail (with one reminder) to the households in the two areas before and after the repaving. Answers from 41% of these were received. The before investigation was performed in May- June 2007, the after investigation in May-June 2008. The number of respondents is given in Table 1.

Table 1

Number of respondents in the before and after survey in the two areas.

Frederikssundsvej Kastrupvej Total

2007 - Before 501 829 1330

2008 - After 563 977 1540

Total 1064 1806 2870

The questions and answering scales for the assessment of the noise annoyance were in accordance with ISO 15 666 [5]: “Thinking about the last year or so, when you are here at home, how much does noise from road traffic bother, disturb, or annoy you?”. The respondents gave their answers on both semantic and 0-10 point numerical categorical scales. A high linear correlation was found between the answers on the two types of scale.

The annoyance scores from the numerical scale are expressed as:

The percentage of highly annoyed (% HA):

Answers in categories 8, 9 and 10

The percentage of (at least) annoyed (% A):

Answers in categories 5 to 10

The percentage of (at least) little annoyed (% LA):

Answers in categories 3 to 10

Figure 1

Photos from the Kastrupvej area before the pavement was changed.

2 Method of Investigation

(11)

10

2 M e t ho d of I nve s t i g a t io n

where

A is the percentage of annoyed (HA, A, LA) respondents u is the upper limit of A (i.e. u = 100)

s is the slope of the inverse logit function E is the noise Exposure

f is the value of E for a fifty per cent annoyance response

2.3 Dose-response curves

The percentage of annoyed (HA, A and LA) were calculated in 1 dB classes and the resulting dose-response curves were found by logistic regression, see [6]. The dose-response curves are expressed as:

A = u 1+e^–s(E–f) Figure 2

L_den from traffic noise in the Frederikssundsvej area before the pavement was replaced. Noise levels were approximately 4 dB lower in the after situation.

3 Figure 2 - L

den

from traffic noise in the Frederikssundsvej area before the pavement was replaced.

Noise levels were approximately 4 dB lower in the after situation

The noise exposure (the day-evening-night level, L

den

) at the facades of the buildings (not including the façade reflection) was calculated for all the respondents’ addresses by means of the road traffic noise prediction method, Nord2000 [4].

2.2 Survey

7000 questionnaires were sent out by mail (with one reminder) to the households in the two areas before and after the repaving. Answers from 41% of these were received. The before investigation was performed in May-June 2007, the after investigation in May-June 2008. The number of respondents is given in Table 1Table 1.

Table 1 – Number of respondents in the before and after survey in the two areas

Frederikssundsvej Kastrupvej Total

2007 - Before 501 829 1330

2008 - After 563 977 1540

Total 1064 1806 2870

The questions and answering scales for the assessment of the noise annoyance were in accordance with ISO 15 666 [5]: “Thinking about the last year or so, when you are here at home, how much does noise from road traffic bother, disturb, or annoy you?”. The respondents gave their answers on both semantic and 0-10 point numerical categorical scales. A high linear correlation was found between the answers on the two types of scale.

The annoyance scores from the numerical scale are expressed as:

The percentage of highly annoyed (%HA): Answers in categories 8, 9 and 10 The percentage of (at least) annoyed (%A): Answers in categories 5 to 10 The percentage of (at least) little annoyed (%LA): Answers in categories 3 to 10 2.3 Dose-response curves

The percentage of annoyed (HA, A and LA) were calculated in 1 dB classes and the resulting dose-response curves were found by logistic regression, see [6]. The dose-response curves are expressed as:

) f E (

e

s

1 A

_

u

_

  (1)

where

A is the percentage of annoyed (HA, A, LA) respondents

50-55 dB 55-60 dB 60-65 dB 65-70 dB 70-75 dB Over 75 dB

(12)

11 3 Re sul t s

3.2 The before and after situation

The results from the two areas are merged in Figure 4 which shows the dose-response curves from the before and after situation. In the general picture there are no significant differences between the before and after situation.

There is a small tendency towards a lower percentage of Highly Annoyed in the 50-60 dB range in the after situation, but except for that, people actually reacted to the noise they were exposed to, independent of whether it was before or after the repaving. Those who are exposed to less noise were on the average less annoyed. If the respondents gave the same answers in the after situation as in they did in the before situation a displacement of the curves of 4 dB corresponding to the noise reduction would have been seen. This is not the case. So actually the curves can be used to predict the annoyance reduction for this type of roads when the noise reduction is known.

3.1 Comparison of the two areas

There were demographic differences between the two investigated areas. The main differences are listed in Table 2.

Table 2

Percentage of respondents in the before and after survey in the two areas.

Frederikssundsvej Kastrupvej

Residence owned 56% 36%

Own garden 28% 6%

Years in residence > 10 49% 35%

Below 40 years of age 22% 45%

The dose-response curves for each of the two areas are shown in Figure 3. By comparison of the curves from the two areas it is seen that the confidence intervals for the corresponding curves in the upper and lower graphs are overlapping. This means that there are no significant differences between the dose-response curves from the two areas.

3 Results

(13)

12 3 Re sul t s

Little annoyed 2007 Annoyed 2007 Highly annoyed 2007 Little annoyed 2008 Annoyed 2008 Highly annoyed 2008

Figure 3

Dose-response curves for each of the two areas in the before and after situation. The dotted lines indicate the 95% confidence intervals for the curves.

(14)

13 3 Re sul t s

Little annoyed Annoyed Highly annoyed

Figure 5

General dose-response curves for Danish urban roads. The dose-response curves are based on data merged for both areas and for the before and after situation. The dotted lines indicate the 95% confidence intervals. The curves are based on a total of 2870 answers. The curve for Highly Annoyed is based on 563 answers, the curve for Annoyed is based on 1217 answers and the last curve for Little Annoyed is based on 1758 answers. The fraction of explained variance R² is approx. 90% for all curves.

Figure 4

Dose-response curves in the before (2007) and after (2008) situation. The results from the two areas have been merged. The dotted lines indicate the 95% confidence intervals for the curves.

Little annoyed 2007 Annoyed 2007 Highly annoyed 2007 Little annoyed 2008 Annoyed 2008 Highly annoyed 2008

(15)

14 3 Re sul t s

3.4.1 Noise sensitivity

There was a direct question about the respondent’s noise sensitivity. Figure 6 shows the moderating effect of the noise sensitivity on the dose-response curves for the percentage of Highly Annoyed. The more noise sensitive the respondents are, the more annoyed they are at a certain noise exposure. It should be noted that the curve for Extremely sensitive has a very large confidence interval and that the fraction of explained variance for this curve is as low as 3%.

From the curves it can e.g. be seen that at an exposure of L_den = 65 dB, 14% of the Little sensitive respondents are Highly Annoyed while 40% of the very sensitive respondents are Highly Annoyed.

3.4.2 Fear of accidents

Figure 7 shows that the moderating effect of respondents having plans to move away from the area due to an unsafe feeling caused by to the traffic, when going about in the area. Persons who would like to move for this reason are more annoyed than other persons.

Extremely sensitive Very sensitive Moderately sensitive Little sensitive Not sensitive

Figure 6

Dose-response curves of Highly Annoyed for various degrees of noise sensitivity. The dotted lines indicate the 95% confidence intervals for the curves. The fraction of explained variance R² for Extremely sensitive is very low (3%).

3.3 General dose-response curve

When there are no significant differences in the dose-response curves in the before and after situation it is

meaningful to merge the data. The result of this is shown in Figure 5. The parameters for these curves can be found in Table 3 Section 3.5.

It has been calculated that the average noise annoyance on the 0-10 point scale among the residents on the primary roads has decreased from 5.3 in the before situation to 4.5 in the after situation. According to a one sided Welch t-test this result is significant on a three star level (p

= 6.9·10^-6).

3.4 The influence of demographic factors

The questionnaire consisted of a total of 40 questions. The answers to some of these were seen as moderators for the noise annoyance as illustrated in the following sections. All curves in these sections are based on the merged answers from both areas in the before and after situation and they show the effect on the percentage of Highly Annoyed.

(16)

15 3 Re sul t s

Yes No

Figure 7

The moderating effect of the question: “If you want to move from this area, what would be the main reason?” The upper curve applies if the option: “It is unsafe to go about in the area because of the traffic” was ticked. The lower curve applies if not. The dotted lines indicate the 95%

confidence intervals for the curves.

Detached house Apartment

Figure 8

The Highly Annoyed dose-response curves for people living in their own houses and in apartments. The dotted lines indicate the 95% confidence intervals.

(17)

16 3 Re sul t s

• Years in residence: Persons who have lived 7-10 years in their residence seem to be more annoyed than other respondents

• Children: If there are children below 10 years of age the annoyance from the traffic is higher

• Gender: There are no significant difference in the percentage of Highly annoyed between men and women

• Hearing problems: Do not have significant effect

• There are no obvious changes in annoyance whether respondents sleep with open windows or not

3.5 Comparison with international results

Figure 9 shows the dose response curves for urban roads in Copenhagen, the same data as in Figure 5, with 2870 observations compared to the international curves from [7], based on 19.172 observations from 26 investigations 3.4.3 Type of residence

Figure 8 shows the Highly Annoyed dose-response curves for people living in their own houses and for people living in apartments. The percentage of Highly annoyed is larger for respondents living in their own house than for people living in an apartment.

3.4.4 Other moderating factors

Besides the previously mentioned variables, other conditi- ons were found to have moderating effects on the dose- response curves:

• Age: The percentage of Highly annoyed is larger in the age group 40-60 years

• Number of persons in the residence: The percentage of Highly annoyed is larger when more than two persons are living in the same household

Figure 9

Dose response curves for urban roads in Copenhagen (thin lines, same as in Figure 5) with 2870 observations compared to the international curves (thick lines, Miedema) from [7], based on 19.172 observations from 26 investigations in various countries. The dotted lines indicate the 95% confidence intervals.

(18)

17

Based on answers from 2870 respondents living near primary urban roads in Copenhagen where the old pavement was replaced by new noise reducing thin asphalt layers, the following main conclusions can be drawn:

The reduction in noise exposure, L_den, was 4 dB on the average.

Repaving with noise reducing asphalt gave a statistical significant reduction of the noise annoyance.

There was an essential drop in the fraction of very and extremely annoyed persons even though the noise from the traffic still was the most prominent discomfort caused by the traffic.

In general 10% fewer persons were annoyed by the traffic noise after the repaving with noise reducing asphalt and 7% more found the traffic noise at their home acceptable.

Respondents living near urban roads in Copenhagen were found to be more annoyed by the traffic noise than the international average.

A number of variables moderate the dose-response curves for the annoyance. The main factors influencing the annoyance are: noise sensitivity, age, number of persons in household, fear of accidents, type of dwelling (detached house or block of flats) and small children in the household.

3 Re sul t s

in various countries. Table 3 summarises the values of the parameters s and f for the curves.

Except from the Highly annoyed in the 40-60 dB level range, people living near urban roads in Copenhagen are more annoyed than the international average. For example the percentage of Annoyed is 15% higher and the percentage of Little annoyed is 10% higher than the international average at L_den = 65 dB.

The slopes of the international curves are in general smaller than the slopes of the Danish curves. This is to be expected when data from many investigations with differ- ent road types, methods for surveys and noise calculations are merged. The fraction of motorways, urban roads and central city streets represented in the international curves are unknown to the present authors.

From the f-values in Table 3 it is seen that for 50% annoyance (HA, A and LA) the residents at the urban roads in Copenhagen express the same noise annoyance as the international average at a 3-6 dB lower noise exposure L_den. Table 3

Parameter values for the dose-response curves in Figure 9. f is the L_den value for 50% annoyed (HA, A or LA) and s is the slope of the curves, see Equation (1) in Section 2.3.

Copenhagen International f_diff

f, dB s f, dB s dB

Little annoyed 56.6 0.102 60.7 0.101 4.1

Annoyed 65.2 0.111 70.7 0.103 5.5

Highly annoyed 76.4 0.130 79.4 0.115 3.0

(19)

18 4 C o n c lusio ns

Based on answers from 2870 respondents living near primary urban roads in Copenhagen where the old pavement was replaced by new noise reducing thin asphalt layers, the following main conclusions can be drawn:

The reduction in noise exposure, Lden, was 4 dB on the average.

Repaving with noise reducing asphalt gave a statistical significant reduction of the noise annoyance.

There was an essential drop in the fraction of very and extremely annoyed persons even though the noise from the traffic still was the most prominent discomfort caused by the traffic.

In general 10% fewer persons were annoyed by the traffic noise after the repaving with noise reducing asphalt and 7% more found the traffic noise at their home acceptable.

Respondents living near urban roads in Copenhagen were found to be more annoyed by the traffic noise than the international average.

A number of variables moderate the dose-response curves for the annoyance. The main factors influencing the annoyance are: noise sensitivity, age, number of persons in household, fear of accidents, type of dwelling (detached house or block of flats) and small children in the household.

4 Conclusions

(20)

19 Re fe re n c e s

[5] ISO/TS 15666. Technical Specification: Acoustics – Assessment of noise annoyance by means of social and socio-acoustic surveys, 2003-02-01

[6] Pedersen, T. H.: The “Genlyd” Noise Annoyance Model. Dose-response relationships modelled by logistic functions, DELTA report AV 1102/07, 2007, www.madebydelta.com (search for the title) [7] European Communities: EU Position paper on

Dose-response relationships between transportation noise and annoyance, ISBN 92-894-3894-0, Euro- pean Communities, 2002

References

[1] Pedersen, T. H., Le Ray, G.: Befolkningsreaktioner på støjreducerende vejbelægninger (Community response to noise reducing road pavements– in Danish with an English summary), Danish Road Directorate, Report 442, 2013, see: www.vejdirektoratet.dk [2] Bendtsen, H., Andersen, B. Thomsen, S.: Noise

reducing thin pavements. Urban roads. Danish Road Institute/Road Directorate. Report 149, 2006 , see:

www.vejdirektoratet.dk

[3] Bendtsen, H., Andersen, B. Thomsen, S.: Optimized thin layers – urban roads – the Kastrupvej experiment, Danish Road Institute/Road Directorate, Technical Note 66, 2008, see: www.vejdirektoratet.dk [4] Jørgen Kragh: Nord2000 - New Nordic method for

predicting road traffic noise, Danish Road Directorate Report 335 – 2011, see: www.vejdirektoratet.dk

(21)

The Danish Road Directorate’s headquarter is situated in Copen- hagen and local offices are situated in Aalborg, Skanderborg, Middelfart, Næstved and Fløng.

You will find more information on www.vejdirektoratet.dk.

Vejdirektoratet Niels Juels Gade 13 Postboks 9018 1022 Copenhagen K Tel.: 0045 7244 3333 vd@vd.dk

vejdirektoratet.dk

Kopi fra DBC Webarkiv