2 - Literature Review

This literature review comprises several documents, a number of which were agreed in advanced with Transport Scotland, including: research documents, academic papers, technical guidance and British Standards. Other documents, identified subsequently, have also been added and/or referenced in the literature review, where appropriate.

2.1. Statistical Pass-by method

2.1.1. BS EN ISO 11819-1:2001

BS EN ISO 11819:2001 – “Acoustics – Measurement of the influence of road surface on traffic noise – Part 1: Statistical Pass-By method”. London: British Standards Institution, 2001, provides a test methodology for measuring tyre-road noise using the statistical pass-by method, which requires measuring the noise and speed of individual vehicles passing a fixed point on a live carriageway and processing the data after the testing to determine the acoustic performance of the road surface.

The Statistical Pass-By (SPB) method requires the speed and maximum noise level (LAmax) produced by an individual vehicle to be measured, using a “fast” time response. At least 180 vehicle pass-bys are required to meet the test requirements, consisting of 100 cars and 80 heavy goods vehicles. The standard requires the tests to be carried out on a relatively straight and flat section of road, covering a length of 30-50m, and requires the microphone to be positioned 7.5m horizontally from the centre of the lane of travel (the test lane) at a height of 1.2m. The microphone should be oriented horizontally and perpendicular to the road.

The test requires approximate free-field conditions, noting constraints from roadside furniture such as safety barriers, and requires the sound pressure level prior to the measurement to be 6 dB below the measured maximum noise level when the vehicles passes the microphone. The standard also recommends the use of third-octave band measurements to check that the spectral data from the measurement to ensure that the tested vehicle is representative of its vehicle class (for example, the vehicle is well maintained). Meteorological conditions are recorded during the measurements, including wind speed, wind direction, air temperature and the road surface temperature.

When the measurements are complete, the results are normalised using a regression analysis and sound levels are corrected to a reference air temperature of 20 degrees. The Statistical Pass-By Index is calculated by taking into account the measured noise levels for each vehicle category (cars and HGVs) at the different measured speeds. The Statistical Pass-by Index allows comparison of the acoustic performance of the test surface against a reference surface, which is defined as:

“The reference surface is a dense, smooth-textured, asphaltic concrete surface with a maximum chipping size of 11mm to 16mm. From the acoustical point of view, this is approximately equivalent to a stone-mastic asphaltic surface with the same maximum chipping sizes. The surface shall have been trafficked for at least one year when used as a reference here. Macrotexture depth as measured according to ISO 10844 or ISO 13473-1 shall be within 0.50mm and 1.00mm.”

2.1.2. Highways Authority Product Approval Scheme

Appendix A8 of the “Guidelines Document for the Assessment and Certification of Thin Surfacing Systems for Highways Appendix A8: Noise”, British Board of Agrement (BBA), Watford, 2008 (HAPAS) provides a method for determining the acoustic properties of a road surface. The HAPAS methodology also uses Statistical Pass-by measurements and incorporates aspects of BS EN ISO 11819. It does however introduce some specific requirements and differs from BS EN ISO 11819 in some areas, namely:

  • At least two test locations should be tested for each road speed category (High Speed 110kmph, Medium Speed 85kmph), which may be at the same site provided that they are at least 100m apart
  • The road surface at the test location shall have been open to traffic for a period of not less than 12 months
  • Slightly different reference speeds for high speed roads (90 kmph compared with 85 kmph in BS EN ISO 11819 “Acoustics – Measurement of the influence of road surface on traffic noise – Part 1: Statistical Pass-By method”. London: British Standards Institution, 2001.)

The HAPAS guidelines also provide equations for calculating the Road Surface Influence (RSI), which is a measure of how the test surface performs acoustically compared with a newly laid Hot Rolled Asphalt (HRA) surface. The DMRB H213/11 (Design Manual for Roads and Bridges (DMRB), Volume 11, Section 3, Part 7 - Noise and Vibration (HD 213/11 - Revision 1), 2011.) and recent Smart Motorways tool provides a means for converting the RSI into an acoustic correction that can be applied to road traffic noise level calculations.

2.2. TRL PPR443 – A review of road surface noise reduction techniques, 2010

This TRL report from January 2010 (Published Project Report PPR443 “A review of current research on road surface noise reduction techniques”, Transport Research Laboratory, P G Abbott, P A Morgan and B McKell, January 2010.) is a detailed review of the available research on road surface noise reduction. which includes: discussions of noise generation, surface types; acoustic performance; non-acoustic performance, and cost benefit analysis. The review included a consultation with transport departments and Local Authorities in Scotland. Some of the relevant findings from this review are detailed below:

  • SMA surfacing on the Munich ring road was observed to perform well in terms of durability and noise reduction (although the later was only measured subjectively).
  • The report discusses a trial using Thin Stone Mastic Asphalt (TSMA) on the M8 in Scotland which was showing positive results for early life skid resistance, and that monitoring of the actual noise characteristics is also planned.
  • Based on responses from 27 Local Authorities “Most Departments report that HRA is the most cost-effective surface material. It was believed that, although the initial costs are higher there is a longer life expectancy for HRA that TSMA which means less maintenance and therefore reduced costs in the longer term”. Further that, “noise reduction benefits are not, as yet, considered a major factor when determining road surface material”.
  • Surface noise reduction performances are predicted for TSMA using the Harmonise/Imagine calculation method, relative to HRA 20mm. Initial reductions of 1 to 4dB(A) are predicted at speeds up to 60kmph, independent of aggregate size. At higher speeds, aggregate size does affect noise emissions and at speeds of 110kmph reductions of 4, 5 and 6 dB(A) were estimated as aggregate size decreased from 20mm to 14, 10 and 6mm, respectively. – It is not clear if these reductions are in relation to a new or worn/aged reference HRA surface.
  • Over time the noise performance of TSMA deteriorates at a rate of about 0.5 dB(A)/year compared with about 0.2dB(A)/year for dense surfacing such as HRA.
  • Reference is made to an average measured Road Surface Influence – High Speed (RSIH) of -4.8 dB(A) for 14mm SMA and -0.6 dB(A) for 20mm HRA, both taken from Abbot & Nelson, 2001 [6]. RSIH relates to a reference speed of 110kmph.
  • A simplified assessment of whole life costs (60 years) has been undertaken which concluded that the benefit of SMA, due to lower noise levels, outweighed the costs over the 60-year period. However, a -4.8 dB(A) surface correction (benefit compared to HRA) appears to have been used for all roads (regardless of speed), with no apparent adjustment for a deterioration in noise performance over time. The report acknowledges there are limitations in the assumptions made and the need for further research in this area.
  • A discussion on cost benefit analysis is provided which includes how the benefits are monetised by considering the impact of changes in noise affecting households and the resulting house price or ‘willingness to pay’. – This approach has since been replaced by the current method of monetising impacts on human health and quality of life, discussed in Section 5 below.

2.2.1.1. Abbott & Nelson, 2001

This 2001 TRL document (“Revision CRTN road surface correction for medium and high-speed roads” (PE/SE/289/2001) TRL Limited, P. G. Abbott and P.M Nelson, 2001) is included here only as a referenced document from within PPR443 and has not been reviewed in full. However, the extract below indicates the potential noise benefits, concluded by the authors, by going from an HRA surface to a SMA surface (-4.2 to -5.1 dB(A)). Please note - the levels given in the extract below do not correlate exactly with the -4.8dB(A) offset referred to in PPR443, and further review would be required to fully explain this.

2.3. TRL PP485 – The performance of quieter surfaces over time, 2010 

This TRL report from March 2010 (Published Project Report PPR485 “The performance of quieter road surfaces over time”, Wokingham: Transport Research Laboratory, M Muirhead, R E Stait, and L Morris, March 2010.) is a compressive study collating data on the acoustic performance of low noise surfaces over time in England and Wales. Some of the relevant findings from this study are detailed below:

  • On average, the acoustic performance of low noise road surfaces on high speed roads deteriorates at a rate of 4.5dBA over 10 years compared with a deterioration of 2dBA over 10 years for HRA and EAC (Exposed Aggregate Concrete).
  • On average, low noise surfaces laid on high speed roads outperform the reference HRA surface by 4-6dBA when new, and outperform HRA by 1-3dBA after 10 years (one HRA surface was tested).
  • Low noise surfaces on high speed roads can provide acoustic benefits between 3-8dBA compared with a new HRA surface. The level of the acoustic benefit is dependent on how the surface was constructed and laid, and the stone size of aggregates used (smaller stone size = quieter).
  • RSI derived from SPB measurements as a correction to Calculation of Road Traffic Noise’s (CRTN’s) [26] basic noise level gives reasonably good agreement between measured and predicted noise levels.
  • Close Proximity (CPX) measurements found a 1dBA variation in the CPX index over just a few hundred metres of road surface and highlighted poor performance at the beginning and end of test sections.
  • The -3.5dB correction for low noise road surfaces provided in the DMRB provides a good approximation of the average benefit of an existing low noise surface that has been part of the SRN for 5 years.
  • Research recommended the following equation for a surface correction: surface correction = RSI + 0.45 x (age of surface) dBA.

2.4. Interim Advice Note 154/12 

IAN 154/12 (Interim Advice Note 154/12 (IAN154/12), “Revision of SHW Clause 903, Clause 921 and Clause 942”, September 2012.) provides guidance for the specification of thin surface materials, including where to use different surfaces based on their “Road/Tyre Noise Level” performance. Table NG 9/30 in IAN 154/12 categorises road/tyre noise performance by listing the different road surface influence values, alongside a description. This information is reproduced in Table 2-1 below.

Table 2-1 – Road/Tyre Noise Levels (Source: Table NG9/30)
Level Description RSI (dB(A))
3 Very quiet surfacing material -3.5
2 Quieter than HRA surfacing materials -2.5
1 Equivalent to HRA surfacing materials -0.5
0 Equivalent to Cold Applied Ultra Thin Surfacing +1.2
NR No requirement No requirement

IAN 154/12 states that “Levels 2 and 3 are necessary in noise-sensitive areas. In the interest of sustainability, Level 3 should only be specified in very noise sensitive areas”. Further that “Level 0 must not be specified at sites with existing noise barriers or earth bunds, where the latter have been specifically installed as a noise mitigation measure and must not be used at locations that have been identified as an Important Area, either with or without First Priority Locations, in any of England’s Noise Action Plans published by DEFRA in March 2010”.

IAN 154/12 identifies that the Road/Tyre Noise Levels are demonstrated by the optional value stated on HAPAS Certificates, or equivalent certification.

Guidance in DMRB HD213/11 (Design Manual for Roads and Bridges (DMRB), Volume 11, Section 3, Part 7 - Noise and Vibration (HD 213/11 - Revision 1), 2011.) suggests that in the absence of measured data, an existing thin surface be classified as Level 2 and new thin surfaces to be Level 3.

Below is an extract from an HAPAS Certificate of a Cold Applied Ultra-Thin Surface:

“The road surface influence (RSIH) was recorded as +0.8 dB(A). The system meets the requirements of Level 0 in accordance with Interim Advice Note 154/12, Table NG 9/30.

Road traffic noise levels will be affected by several factors, including location, traffic type and the condition of the road and, therefore, the RSIH value may not be reproduced on other installations”.

2.5. Highways England - Noise Evaluation of Road Surfaces, 2017

This report (“Sub Task 3 – Noise Evaluation”, Collaborative Research Programme, Highways England, Mineral Product Association and Eurobitume UK, November 2017.) was produced by AECOM for Highways England, Mineral Product Association (MPA) and Eurobitume UK, who have come together to fund development of a more durable low noise pavement for use in England. The report provides an overview of the mechanisms involved in tyre/road noise generation, how tyre/road noise is measured, the acoustic properties of different road surfaces and how road surfaces may be classified in terms of their impact on traffic noise.

The report mainly provides a details of a Premium Asphalt Surfacing System (PASS) that is in development. However, it also provides some useful up to date commentary on the issues of assessing low noise performance of road surfaces, albeit not SMA, and the main issues to consider.

The report states that the “simplest way to measure road traffic noise is to record average noise levels at the side of a road and, where the average traffic speed is not low (>50 km/h say) and tyre/road noise dominates over engine noise”, and that “this can be a reasonable proxy for the acoustic performance of the road surface”. Where conditions can be fixed these measurements are very repeatable at these specific sites. However, the report points out that with this approach there are too many variables which are not accounted for such as “vehicle speed, traffic composition, measurement averaging time and the environment which can all have a significant effect on the results” and therefore, “whilst this method can be useful for directly comparing the relative acoustic performance of different test sections laid next to each other and subject to the exactly the same traffic stream, it is not appropriate for any other application”.

The report discusses the method of Controlled Pass-By (CPB) measurements which limits the number of variables by measuring the pass-by noise level of single test vehicle at specified speeds. Whilst this accounts for vehicle speed it is not representative of the noise emissions from all types of vehicles, such as Heavy Goods Vehicles (HGVs), which vary in terms of propulsion noise and tyres. The report points out that in situations where a CPB measurement may be required it is nearly always preferable to perform a series of SPB measurements instead. 

As discussed in Section 2.1 above, SPB method involves measuring the speed and maximum pass-by level of a number of cars and HGVs and performing a regression analysis to determine a reference noise level at a given speed. The report states that “SPB method is still the most commonly used measurement procedure for classifying the acoustic performance of road surfaces in the UK”.

The CPX method is also discussed, whereby a specialist instrument records the tyre/road noise in the near field through the use of microphones placed close to a designated tyre of a test vehicle or trailer and enclosed in a semi-anechoic chamber. Although this method is arguably the best way to understand the levels of noise generated by the mechanisms outlined in Section 1, they are more difficult to relate to traffic noise levels experienced by residents close to the road.

ROSANNE is an EU programme that puts forward a new classification procedure for road surfaces and a metric called the Road Surface Noise level (RSNL). It is based on solely on CPX measurements, however it is reliant on the statistical relationship between CPX and SPB measurements.

In terms of a classification system the report concludes by saying that “rather than promote another classification procedure it is recommended that the ROSANNE methodology be given consideration for use in the UK and in the meantime existing knowledge on the performance of surfaces within the UK over time and under different traffic conditions (Published Project Report PPR485 “The performance of quieter road surfaces over time”, Wokingham: Transport Research Laboratory, M Muirhead, R E Stait, and L Morris, March 2010.) be exploited in traffic noise modelling and guidance”.

2.6. Low Noise Road Surfacing – A9 Dualling, 2017 

It is understood that TS2010 was the preferred road surface to use on the A9 Dualling Scheme, and discussions were held with Transport Scotland to confirm noise modelling assumptions regarding the acoustic performance of the road surface. The outcomes of these discussions were formalised in a technical memorandum (“A9 Dualling: Low Noise Road Surfacing”. Memorandum, N. Weller, Glasgow: Jacobs, 30 March 2017.), where it is recommended to use the DMRB HD213/11 corrections for low noise road surfaces for TS2010 in lieu of detailed and robust information about the performance of the road surface. This means that the following corrections were used (relative to HRA surfaces) for TS2010 on the A9 Dualling Scheme:

  • -3.5dB for a newly laid TS2010 surface on roads where average speeds exceed 75 kmph (i.e. a newly laid low noise road surface);
  • -2.5dB for an existing TS2010 surface on roads where average speeds exceed 75 kmph (i.e. an existing low noise road surface of uncertain age), and
  • -1dB for TS2010 surfaces on roads where the average speed is below 75 kmph.

The Memorandum acknowledges that comparative Statistical Pass-By measurements were undertaken during 2009 for the TS2010 road surface, from which it is hypothesised that the road surface was classified as having a “Level 2 Performance” according to the MCHW Volume 2 (Highways England (2018). Manual Contract Documents for Highway Works, Volume 2, 900 Series, Table NG 9/30. London: TSO.) (see Table 2-1 above). It is assumed this would have meant a correction of or -2.5 x 0.7 = 1.8 dB(A), in accordance with DMRB HD213/11 guidance. However, it is stated that as some aspects of the SPB measurements did not meet the BS EN ISO 11819 (“Acoustics – Measurement of the influence of road surface on traffic noise – Part 1: Statistical Pass-By method”. London: British Standards Institution, 2001.) requirements, it was considered more “defensible” to use the corrections provided in the DMRB HD213/11 (Design Manual for Roads and Bridges (DMRB), Volume 11, Section 3, Part 7 - Noise and Vibration (HD 213/11 - Revision 1), 2011.). The full details of the SPB measurements would ideally be reviewed by Atkins before undertaking further assessments.

2.7. Lane by lane calculations

It is possible to improve the accuracy of road traffic noise predictions for multi-lane carriageways by taking into account the road surfacing laid on each lane. This is useful for calculating an overall acoustic correction for the carriageway if it contains a mixture of different surfaces of different types and ages. The acoustic correction is based on the Road Surface Influence of each surface type and includes a weighting factor for each lane, with the nearside lane being assigned the highest weighting. This approach is based on the following equation from an unpublished revision to the Calculation of Road Traffic Noise (CRTN revision and update. Institute of Acoustics Sound Transport Modelling, Manchester, M.Muirhead, 14 March 2017. (“Multi-lane correction”)) and the RSI equation in the DMRB:

Correction = 0.7 x RSI = 0.7 x 10

[Eq.1]

Where L is the number of lanes and RSIi is the RSI in the i th lane (the nearside running lane of the carriageway being i = 1). This approach, although not yet industry standard, could further improve accuracy of noise predictions and cost benefit analysis for resurfacing programmes.

It is possible, although cannot be guaranteed, that this approach, or something similar, could be included in the impending revisions to CRTN and/or DMRB HD213/11. If this proves not to be the case, there is still sufficient precedence on smart motorway schemes in England for it to be adopted more widely.

2.8. Literature from Europe and the Rest of the World

Several studies from various other countries outside the UK, looking at the noise reducing performance of SMA, have been reviewed. This is not an exhaustive list of relevant contemporary research on this subject from around the world.

2.8.1. Noisiness of the Surfaces on Low-Speed Roads, 2016 [Poland]

This article was published online, by authors Gardziejczyk et al. (“Noisiness of the Surfaces on Low-Speed Roads”. Authors: Wladyslaw Gardziejczyk, Pawel Gierasimiuk and Marek Motylewicz, Division of Road Engineering, Faculty of Civil and Environmental Engineering, Bialystok University of Technology. © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons by Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).), from the Division of Road Engineering, Faculty of Civil and Environmental Engineering, Bialystok University of Technology.

The aim of the article is to assess noisiness of different types of surface used for roads and streets servicing cars moving at lower speeds. The statistical pass-by measurement method (SPB) was used to evaluate the noise levels and how they varied in relation to the type of road surface and the speed of passing vehicles. The study reviews three types of SMA surface, SMA11, SMA 8 and SMA5, where the number relates to the maximum size (diameter) of the aggregate in millimetres. Specific details of the product name or construction were not provided. Other surfaces are also assessed in the study, including: single layer porous asphalt with maximum aggregate size of 8mm (PAC8); two layers of porous asphalt with maximum aggregate size of 8mm on the top layer and 16mm on the lower layer (DPAC8+16), and paving (PS: paving stones, CBP: concrete block paving).

Some of the key findings relating to the different road surfaces, that may have implications for this review of TS2010, are as follows:

  • Due to increased air void content, porous asphalt surfaces are characterised by significantly lower levels of sound than stone mastic asphalt pavements in the frequency range above 800 Hz;
  • Surfaces with reduced noise (PAC8 and DPAC8+16) are initially quieter by 5.5–6.0 dB than the reference surface SMA11;
  • Spectral analysis confirmed that the SMA11 surface, above frequencies of 630 Hz, is much louder than the surfaces of porous asphalt for up to two years of operation;
  • Damages occurring on the PAC8 surface after four years of operation increase its macrotexture and create higher sound levels as compared with the levels of sound on the SMA11 in the frequency range of 500–1600 Hz. After four years of operation, the PAC8 demonstrated a higher level of tyre/road noise than a SMA11;
  • Noise level measurements conducted by using the SPB method on vehicles passing at 60 kmph have shown that paving stones surfaces are louder by +6.8 dB in comparison to SMA11, and
  • Maximum aggregate size is a very significant feature in the stone mastics asphalt; its reduction from 11 mm to 5 mm decreases the sound level of a car by -2.5 dB and SMA8 was -1.9dB quieter than SMA11.

This last finding from the Poland study, listed above, is of relevance to the review of TS2010’s noise reduction performance, which has three aggregate size variants of 14mm, 10mm and 8mm. 

2.8.2. Quiet Road Surfaces, Euro Cities, 2011 [Netherlands]

This leaflet is produced by M+P consulting engineers, Netherlands for Working Group Noise EURO CITES, issued 9 February 2011 (“Quiet Road Surfaces”. M+P consulting engineers, Netherlands for Working Group Noise EUROCITES, issued 9 February 2011).

This study gives an introduction those who want to “explore the benefits and common pitfalls of silent road surfaces in urban situations”. It identifies that in many action plans low noise surfaces are the most effective and cost-efficient measure in the abatement of environmental noise. However, for heavy goods vehicles, rolling noise only starts to become the dominant vehicle noise source at speeds 60 kmph and above.

The benefits of a noise reduction performance monitoring scheme are discussed in relation to giving managers and decision makers a synoptic overview of the acoustical effects of low noise surfacing policies/programmes. 

Various issues with porous asphalt surfaces are discussed, however, regarding Stone Mastic Asphalt (SMA) the leaflet states it is popular because of its durability and its resistance against rutting. It states SMA “with an aggregate size of 5 to 6 mm has an optimal texture of the surface. The noise reduction for this type of SMA is a maximum of 2 dB in urban situations relative to DAC 0/16 [dense asphalt concrete]. Although considered noise reducing road surfaces, the versions with larger stone sizes can be quite noisy. The effect of a 40 % increase in aggregate size is 1.5 dB for passenger cars”.

The durability of the acoustical effects is also discussed, stating that “when conducting a study into acoustic properties over time, it is important to evaluate the road surface depending on the expected life. A conventional road surface such as dense asphalt concrete (DAC) or stone mastic asphalt (SMA), will show less acoustic degradation, but these surfaces have a significantly longer life”.

One of the main takeaway messages of this leaflet, produced in the Netherlands, is the importance of long-term monitoring programmes to provide insight, allowing well-founded choices to be made in terms of durable, cost effective solutions to noise problems. 

2.8.3. Road Surface Labelling, 2017 [Netherlands]

This ‘informal document’ (“Road surface labelling”, Informal document to UNECE GRB, to be submitted for the September 2017 GRB meeting, Dutch Working group – J. Sliggers et al, Netherlands, 23 August 2017. ) prepared by a Dutch working group on road surface labels, was prepared prior to a meeting in September 2017. The document sets out to propose guidelines for making a road surface label which comprises four performance indicators:

  • Traffic noise reduction
  • Wet skid resistance
  • Rolling resistance
  • Lifespan

As tyres, traffic and environmental conditions also affect these indicators, a ‘standard’ set of tyres and environmental conditions are assumed. The document provides an example label for a Porous Asphalt (16mm nominal grain size) product, which is reproduced below:

Extract from Road Surface Labelling - Netherlands 2017
Figure 2-2 - Extract from Road Surface Labelling - Netherlands 2017

It is intended that the road surface labels would primarily be the responsibility of the producer and they would not replace tyre labelling, rather they would be complementary.

In terms of determining the noise reduction performance, the working group advocated the use of a harmonised European method such as that being developed in the ROSANNE project. However, such a method was not in existence at the time of writing, therefore, specific noise reduction values and a reference benchmark were selected to suit the present Dutch context. 

The method used for determining a road surface correction is a theoretical one taken from the Sections 2.2.3 “Rolling noise” and 2.2.6 “Effects of the type of road surface” of the environmental noise directive 2015/996/EC (2015/996/EC European Noise Directive), for m=1 (light motor vehicles) and A-weighted over all octave bands. A Dutch ‘reference road surface’ is used based on measurements, of in-situ asphalt concrete, undertaken in accordance with ISO 11819-1:1997 Statistical Pass-By method (SPB), but with a microphone height of 3 m, to avoid in-situ measuring problems caused by guard rails. The noise reduction values on the label are based on the performance at 80 kmph.

The document states that Stone Matrix Asphalt (EN 13108-5) with nominal grain sizes of 5 to 16 mm typically will have label class E, +2 dB to -1dB, where a positive number denotes a reduction in noise relative to the reference surface. It is also worth noting that as the label scale is based on “initial” values, the noise reduction at the end of road surface lifespan may be lower than the label class value. 

The main preconditions for road surface labels are:

  • Compatible with the existing tyre label;
  • Compatible with existing international standards and measuring methods;
  • Suitable for current and future vehicle fleet;
  • Includes (only) essential road surface features - for both new and existing roads;
  • Allows for (meaningful) innovation (product and process).

The road surface label is intended to be used in multiple ways within the road construction
contracting process:

  • Specification for road construction tenders: Enabling road authorities, or other legal
    entities, commissioning pavement works, to specify functional performance of pavement
    surfacing, by specifying performance classes of important functional characteristics.
  • Corroboration (“creating trust”) for contractor bids for tenders;
  • Works approval, by providing the framework against which the delivered characteristics
    can be compared;
  • Threshold value during warranty period or maintenance period.

2.8.4. Measurements on Low Noise Road Surfaces, 2016 [Germany]

This technical paper was authored by Wolfram Bartolomaeus, on behalf of the Federal Highway Research Institute, Germany, and published in the Inter-Noise 2016 proceedings (“Measurements on Low Noise Road Surfaces” Author: Wolfram Bartolomaeus, Federal Highway Research Institute, Germany, published in the Inter-Noise 2016 proceedings.).

This paper provides an overview summary of the different road surfaces used in Germany, their noise reduction performances, and the various methodologies for quantifying the noise reduction provided by the different road surfaces. The paper states that “a systematic classification method for the quantity of noise reduction is under development”, including different direct measurement methods like SPB, CPX and indirect methods to assess surface texture / acoustic absorption / air flow resistivity.

In discussing SPB measurements, it is mentioned that roadside measurements are taken at heights of 2.4m, 3.6m and 4.8m above carriageway, rather than 1.2m, due to the presence of guardrails and/or low concrete walls. However, SPB measurements are often not possible on busy motorways in Germany, as single vehicle pass-bys are required without significant contributions from other traffic. This can be overcome if measurements are taken at quieter times, notwithstanding the potential health and safety concerns this may raise for night-time working. A method using a backing board is also mentioned, for urban situations, to mitigate interfering noise from other sources.

CPX measurements were undertaken, however, they are not discussed in detail and it is stated that insufficient data was collected to provide meaningful analysis.

The SPB method was used at five sites, all of which had a low noise surfaces (porous asphalt rather than SMA). These surfaces were all different ages and the analysis of the measurements clearly show that noise levels increase with the age of the low noise surface, and that the frequency spectra are different at different speeds. Although SMA do not feature in this study, it is a good example of SPB method being used effectively to demonstrate the performance of low noise road surfaces over time.

In Germany, surface noise correction can be determined using either the SPB or CPX method according to GEStrO-92 (“Verfahren zur Messung der Geräuschemission von Strassenoberflächen (GEStrO92)”, Anlage zum ARS 16/1992, Der Bundesminister für Verkehr, 1992), and in noise prediction models, a correction of -2.0 dB(A) is applied to stone mastic asphalt 0/8 and 0/11 without loose chippings, compared to a smooth concrete reference surface.

2.8.5. Stone Mastic Asphalt – Three Studies from Australia

2.8.5.1. SMA - A review of its noise reducing and early life skid resistance properties, 2006

This technical paper was authored by Gayle Greer of Bassett Acoustics, Sydney, Australia, and published in the Proceedings of Acoustics 20-22 November 2006 (“SMA – A review of its noise reducing and early life skid resistance properties” by Gayle Greer of Bassett Acoustics, Sydney, Australia, Proceedings of Acoustics, 20-22 November 2006.). This paper provides a review of SMA surfaces in terms of noise reductions and the issues around early life skid resistance.

It is stated that tyre/road noise is the dominant factor at speeds above 40-50kmph. In the conclusions it is stated that “SMA and OGA [open grade asphalt] surfacings provide a significant reduction (approximately 3 dB(A)) in tyre/road interfacial noise levels compared with DGA [dense grade asphalt]”. However, no specific reference to measurements is provided. 

2.8.5.2. SMA – Engineering Road Note 10, 2016

Engineering Road Noise 10, Stone Mastic Asphalt, May 2016 (Engineering Road Noise 10, Stone Mastic Asphalt, May 2016, Australia.) provides a good overview of SMA, its physical properties, and the various ways it can be constructed / laid. In terms of surface texture, the document states “10mm SMA will have a surface texture at construction of the order of 1.2 to 1.4mm on average. This texture reduces marginally in the first 4 years or so and then increases slightly as the binder wears from the surface voids of the SMA. The 7mm SMA has a surface texture at construction of the order of 0.7 to 0.9mm and does not change much in its life cycle.”

In term of road/tyre noise the document states that “Resurfacing an existing DGA surfacing with SMA will not achieve significant reduction in road/tyre noise. The new SMA will generate less noise than the old DGA however any comparison of noise should be made against new surfacings of each type of asphalt. In any speed environment 10mm SMA is likely to generate a comparable level of road/tyre noise as new DGA whilst for high speed environments 7mm SMA should generate less road/tyre noise than DGA”.

2.8.5.3. Tyre Road Noise of Open Graded Stone Mastic Asphalt in Perth, 2014

  1. Halligan’s “Tyre Road Noise of Open Graded Stone Mastic Asphalt in Perth – 2014 Review”, S. Halligan, Australia, 2014, shows “Where SMA is to be used and noise reduction is required 7mm SMA will provide a better outcome than 10mm SMA. Where noise reduction is not a factor 10mm SMA will provide a greater surface texture than 7mm SMA”.

The study is based on SPB measurements which compared 7mm and 10mm SMA to OGA. At high speeds for ‘typical’ traffic conditions (15% HGV), the relative difference to OGA was +2.1 dBA and +2.4 dBA for 7mm and 10mm aggregate, respectively. 

The performance over time is also discussed and the performance of OGA did not deteriorate as much as might have been expected and surface voids did not become blocked. The report advises that “SMA can be used where the need for noise reduction is not as critical and where it may not be practical or affordable to install OGA as a wearing course supported by a layer of 10mm dense graded asphalt. In addition, SMA only has surface voids and does not have the same free draining characteristics of OGA and as a result SMA generates surface spray similar to a surface with DGA. Where SMA is to be used and noise reduction is required 7mm SMA will provide a better outcome than 10mm SMA. Where noise reduction is not a factor 10mm SMA will provide a greater surface texture than 7mm SMA”.

2.9. Other References

2.9.1. Inter Noise conference, 2016

A paper presented at the 2016 Inter Noise conference (“Compatibility of the ROSANNE noise characterization procedure for road surfaces with CNOSSOS-EU model”, Inter Noise Conference, 2016.), provides an update on the ROSANNE European project which is preparing a procedure for standardising how road surface noise properties are characterised. This document has not been reviewed in detail, however, the CPX measurement method is used as the basis for characterising noise reduction. The document describes the standard ‘reference’ surface and conditions against which all other surfaces would be compared, including:

  • “an air temperature of 20 °C
  • a traffic with constant vehicle speed and no vehicles with studded tires
  • a flat and dry road, with a virtual reference road surface defined according to average noise emission properties on:
    • a mix of Dense Asphalt Concrete (DAC) 0/11 and Stone Mastic Asphalt (SMA) 0/11
    • all between 2 and 7 years old and
    • in a representative maintenance condition”

2.9.2. CNOSSOS-EU, 2012

The CNOSSOS-EU (2012) prediction methodology has not been reviewed in detail, however, the method implements the European Directive 2015/996/EC which is mandatory, as of 31st December 2018, for EU Member States, in relation to strategic noise maps.

2.9.3. The global experience in using low-noise road surfaces, 2009

A report (“The global experience in using low-noise road surfaces: A benchmarking report”, by Ulf Sandberg, Feb 2009.), has been reviewed in limited detail. This extract below, based on data from the HARMONIOSE project, provides a clear visualisation of the effects of stone size on noise output. The virtual reference (green line) is the same as described, and underlined, in Section 2.9.1 above.

Very rough illustration of how traffic noise level (for free flowing traffic at medium or high speeds) changes with maximum aggregate size and type (DAC versus SMA)
Figure 2-4 - Extract from Ulf Sandberg, Feb 2009