Does sound absorbing pavement really works? Below is a very well conducted experiment over a period of 4 years. I am indebted to J.Kragh and H. Bendtsen for allowing me to render their results on my web pages.
The details of this experiment are published in the EURONOISE 2003 (Naples) proceedings. IIt should be noted that all this was made possible by contributions from Danish Ministry of Transport, the Ministry of Environment , the Danish Road Directorate/Danish Road Institute, the Danish Environmental Protection Agency, NCC Contractors, the association of Danish asphalt contractors, the Municipality of Copenhagen, Atkins Denmark and DELTA, Acoustics & Vibration".


Noise and permeability measurements were made at test sections of a two-lane city street with various twin-lay drainage asphalt pavements. The average traffic was 7,000 vehicles/24 hours, 8% heavy vehicles, and the general speed limit 50 km/h. The test sections were:
No. Designation Description Top Layer Bottom Layer
Thickness [mm] Aggregate size [mm] Thickness [mm] Aggregate size [mm]
I DA8-70 Drainage asphalt 25 5/8 45 11/16
II DA5-55 Drainage asphalt 20 2/5 35 11/16
III DA5-90 Drainage asphalt 25 2/5 65 16/22
IV AB8t (reference.) Dense asphalt concrete 30 0/8 - -
Measurements were made in 1999 just after the test sections had been built, and have been repeated each year to investigate the effect of ageing and wear and the clogging of air voids in the drainage asphalt. Three positions were chosen at each section to ensure spatial averaging. Twice a year the drainage asphalt has been cleaned by high-pressure water / air suction. Permeability was measured prior to and after flushing the surfaces.

For each category, the 'Vehicle Noise Level' Lveh was determined by linear regression analysis of the noise level on the logarithm of the speed. Lveh-values were determined for the speeds given below, which differ from the reference speeds given in ISO 11819-1 in order to increase accuracy in comparison between test sections. Based on the Lveh-values a modified 'Statistical Pass-By Index', SPBI', was calculated.
Category Light Heavy
Two-axle Multi-axle
Speed [km/h] 50 45 40
Wx [ISO11819-1], [-] 0.90.075 0.025
Passenger Van
Wx, This paper,[-] 0.8 0.1 0.075 0.025


Measurements have been made during 4 consecutive summers at twin-lay drainage asphalt in a city street to investigate the noise reducing abilities of such asphalt as a function of time. Generally the noise levels have increased with time. At the drainage asphalts with 2/5 mm grains significant change was seen already after 1 year, while at the drainage asphalt with 5/8 mm stones the increase seems more “gradual”. The noise reduction – the difference between SPBI' or Lveh,p for the passenger cars at the drainage asphalt and the noise level at the reference section with dense asphalt – is shown in the table below. These noise reductions are averages from two or three positions, and especially along Section III there was significant variation.

Noise Reduction in dB(A) I (DA8-70) II (DA5-55) III (DA5-90)
SPBI Lveh,p SPBI Lveh,p SPBI Lveh,p
YEAR 0 4.5 4.6 4.9 6.0 6.0 6.5
YEAR 1 4.6 5.3 3.6 4.8 3.8 4.1
YEAR 2 2.7 3.9 2.7 4.1 2.7 3.2
YEAR 3 2.4 4.1 2.2 3.9 1.0 1.4

The noise level at the reference section has not increased during 3 years, while one would expect a 1-2 dB increase. The content of bituminous binder in the reference asphalt concrete turned out to be larger – and the surface smoother – than intended. If the noise level had increased at the reference section, the noise reduction would have increased by the same amount. The decrease in noise reduction seems to be caused by clogging of the voids, especially at the surfaces with 2/5 mm grains.
The average reduction of passenger car noise levels at test section I and II was 4 dB after 3 years.