estimation procedure of the descriptor laeq,t from the

Estimation procedure of the descriptorLAeq,T from the stabilization time of the

sound pressure level valueTorija Martinez, AJ, Ruiz, DP and Ramos-Ridao, A

http://dx.doi.org/10.1260/0957-4565.43.1.11

Title Estimation procedure of the descriptor LAeq,T from the stabilization time of the sound pressure level value

Authors Torija Martinez, AJ, Ruiz, DP and Ramos-Ridao, A

Publication title Noise and Vibration Worldwide

Publisher SAGE Publications

Type Article

USIR URL This version is available at: http://usir.salford.ac.uk/id/eprint/53226/

Published Date 2012

USIR is a digital collection of the research output of the University of Salford. Where copyright permits, full text material held in the repository is made freely available online and can be read, downloaded and copied for non-commercial private study or research purposes. Please check the manuscript for any further copyright restrictions.

For more information, including our policy and submission procedure, pleasecontact the Repository Team at: [email protected].

mailto:[email protected]

1. IntroductionA major difficulty encountered in thedescription of an urban area from an acousticperspective is the variability of environmentalnoise in time [1-3]. It is well known that roadtraffic noise levels vary highly over time and

space [4]. Short-term variations due toindividual vehicles’ pass-by or fluctuations setby traffic cycles are quite common [5].Moreover, urban environments often involveshort time events that introduce a largeamount of sound pressure. In this work, suchphenomena are known as impulsive soundevents. They include, in the urban context,ambulances, vehicles moving at high speeds,honking horns, banging or crashing sounds,shouting voices, etc., but also church bells,sirens of a school, etc. The occurrence of theseimpulsive sound events largely affects thetemporal structure of urban sound ambient[6].

It has been widely asserted that noisevariability should be taken into account inenvironmental noise research [4]. Forexample, Botteldooren et al. [7] maintain thatthe temporal structure of the environmentalnoise level should be considered in order tocorrectly characterize an urban soundenvironment. In fact, the timeframe (e.g. howlong the measurement should be) must beconsidered [8], with the aim of ensuring thatthe results are representative of the realconditions [9-11]. Within this same context,Gonzalez et al. [12] establish that urban noisemeasurements must extend over a long enoughperiod of time to be stable and reliable, butnot extremely long in order not to increasefield work expense beyond aceptable levels. Aparameter which can inform about the timerequired to obtain a sound pressure levelrepresentative of the evaluated urban soundambient is the stabilization time of the soundpressure level [6].

Accordingly, this work aims to propose aprocedure to estimate a value of LAeq,T,representative of a given urban localization ina short-term time period, from the utilizationof the value of the stabilization time of thesound pressure level. For it, the behavior ofthe stabilization time against both the roadtraffic intensity and the appearance of

ESTIMATION PROCEDURE OFTHE DESCRIPTOR LAEQ,T FROM

THE STABILIZATION TIME OFTHE SOUND PRESSURE LEVEL

VALUE

11JANUARY 2012

Estimation procedure of the descriptorLAeq,T from the stabilization time of thesound pressure level valueAntonio J. Torijaa,*, Diego P. Ruiza, Ángel Ramos-Ridaob

aDepartment of Applied Physics, University of Granada, Avda. Fuentenueva s/n, 18071 Granada, SpainbDepartment of Civil Engineering, University of Granada, Avda. Fuentenueva s/n, 18071 Granada, Spain

*Corresponding author. Tel.: (+34) 958 241533; fax: (+34) 958 243214. E-mail address: [email protected].

Abstract

Temporal structure of sound pressure level isa key aspect at the time of characterizingurban sound environments. In urbanagglomerations, environmental noise levelsfluctuate over a large range as a result ofthe great complexity of these settings, withconsiderable temporal and spatialheterogeneity. Furthermore, the domain inurban environments of noise sources, suchas road traffic, commercial or leisureactivities, construction works, etc., togetherwith the occurrence of sudden sound-levelmaxima events (bells, sirens, vehicles at hightraffic speed, honking horns...), which arequite frequent in urban agglomerations,generate the appearance of very high valuesof the impulsiveness of sound pressure level.This aspect causes a great influence on thetime necessary for environmental noiselevels to become stabilized, which is a keyaspect for the accurate measurement,interpretation and guarantee of astatistically representative sample of agiven urban sound environment. Therefore,the goal pursued in this work is to put forth aprocedure for the calculation of a value ofLAeq,T, representative of a certain urbanlocation in a short-term time period, from theutilization of the value of the stabilizationtime of the sound pressure level.

Keywords: Temporal Structure;Environmental Noise; Temporal Variability;Road Traffic; Measurement Time.

deterministic impulsive sound events havebeen studied.

In the following section we focus on theparameter “stabilization time of the soundpressure level” (Section 2). In Section 3.1, thedefinition of an impulsive sound event isprovided, along with brief methodology for itsdetection. Section 3.2 describes themethodology used to obtain the data. Finally,in Section 4 the results of the work arepresented and discussed.

2. Stabilization time of the soundpressure levelThe continuous equivalent sound pressurelevel (LAeq) can be defined as a constant noiselevel whose sound energy value is equal to anaveraged energy of the noise level fluctuationover a total measurement time interval [13].When an environmental noise measurement istaken, the LAeq descriptor changes as timepasses, according to the sound pressure levelprevailing in the ambient. Over time, thesound pressure level accumulates, and thevalues of this sound pressure level fluctuateuntil its stabilization takes place. Due to LAeqbeing a parameter with “many inertia”, therecomes a point at which great variations in theinstantaneous levels are necessary in order thatfluctuations in the value of LAeq appear [6].

The stabilization time of the sound pressurelevel is defined as the required measurementtime (ti) in order that the measured continuousequivalent sound pressure level LAeq,ti differsfrom the overall continuous equivalent soundpressure level LAeq,T in less than ε.

(1)

where ε is the accuracy claimed for themeasurement [6]. The value of ε used in thiswork is ±1 dBA (the value of ε was chosenfrom a deep analysis of the works [14-16], aswell as of a preceding investigation [17]).Therefore, stabilization time is calculated bycomparing the overall LAeq of a given locationwith the LAeq of the first second ofmeasurement, of the first two accumulatedseconds, of the first three accumulatedseconds, etc. (LAeq1, LAeq1+2, LAeq1+2+3, etc.). It isestablished that n is the stabilization timewhen, from the second n of measurementonward, the accumulated LAeq and the overallLAeq of the location differ less than the given ε.

This parameter informs us about the timerequired for measuring in a given localizationin order to obtain a value of LAeq,Trepresentative of the considered urban soundambient. In other words, to attain a reliable

short-term characterization of the soundclimate of a certain urban location, one mustmeasure during at least a period equal to thestabilization time. Nonetheless, it should bementioned that this parameter is very sensitiveto high emerging noise, called impulsive soundevents in this research. Although these soundevents are mostly not representative of theanalyzed environment, in certain cases, appeardeterministic impulsive sound events, whichare representative of a given urbansoundscape, since, these emerging soundevents are an essential part of the soundambient (e.g. church bell, siren of a school,etc.). Because the latter, the behavior of thestabilization time against the occurrence of adeterministic impulsive sound event is a keyissue examined in this research.

3. Methodology3.1. Identification of impulsive soundeventsAn impulsive sound event (hereafter ISE) canbe defined as an acoustic occurrence whichgenerates an increase of 1 dB in theforeground continuous equivalent soundpressure level LAeq (for more detail see [6]).

Following the procedure established inTorija et al. [6], to detect an ISE, the evolutionin time of the parameter LAeq,foreground isanalyzed. To obtain the parameterLAeq,foreground, from each of the instantaneoussound pressure levesl (Lp,i) the overallbackground noise level (LA90) waslogarithmically subtracted, thus obtaining theforeground instantaneous sound pressurelevels (Lp,foreground,i) and then from Lp,foreground,ivalues LAeq,foreground descriptor is calculated.Thus, if a great emerging sound event appearsin time ti, this event will be considered as animpulsive sound event whenever theforeground continuous equivalent soundpressure level in time ti is at least 1 dB higherthan the foreground continuous equivalentsound pressure level in time ti-1.

3.2. Data samplingA number of 80 measurement locations of thecity of Granada (Spain) were selected to be asgenerically representative as possible of thewide range of urban settings. In this samplewere included locations where the main noisesource was road traffic, but also somelocations affected by other sources ofenvironmental noise. It should be mentionedthat 35% of the selected localizations wereaffected directly by noise sources fromcommercial/leisure activities and 17.5% of thelocations were not directly impacted by road

L LAeq T Aeq ti, , ,− ≤ ε



VALUE

12 NOISE & VIBRATION WORLDWIDE

traffic. In this case, the main noise sourceswere those proceeding from natural (waterfountains, birds, etc.) and social (peoplewalking, talking, etc) activities.

In each of 80 selected locations ameasurement of the sound pressure level A-weighted every second (LAeq,1sec) wasperformed, with a total duration of 60minutes. Therefore, the measurementcampaign consisted of a sample of 80 1-hourrecords. Measurements were carried outfollowing the ISO 1996-2:2007 guidelines[18], using a 2260 Brüel & Kjaer type-Isound-level meter, with tripod and windshield,placed at a height of 4 m above local groundlevel as well as 2 m away from the nearestvertical surface. Before carrying out themeasurement a calibration of the sound-levelmeter was performed using a 4231 Brüel &Kjaer calibrator. Moreover, a manually countof road traffic was done simultaneously and,other relevant information, such as noisesources and meteorological conditions, wasalso noted.

It is necessary to indicate that themethodology established in this work refers toa 60-min period. Nevertheless, thismethodology could be extrapolated to anyshort-term time period.

4. Results and discussion4.1. Stabilization time vs. road trafficintensityThe behavior of stabilization time of soundpressure level against different road trafficintensities and the occurrence of an impulsivesound event is investigated in the next twosections.

Observing the results shown in Table 1, wecan assert that urban sound environmentswith direct impact of road traffic have higherstabilization time values than those withoutroad traffic in the proximities. Furthermore,in Torija et al. [6] it is found that the greatestvalues of stabilization time are obtained inurban localizations with low road trafficintensity and intermittent traffic flowdynamics. Because of this, the relationship



VALUE

13JANUARY 2012

Figure 1. Relationship between road traffic intensity and the stabilization time of the sound pressurelevel [6].

Table 1. Average stabilization time value in urbanlocations with/without road traffic and impulsivesound events.

Stabilization

time [min]

Road Traffic With 23.31

Without 4.15

Impulsive Sound Events With 24.84

Without 2.81

between stabilization time and road trafficintensity is analyzed (Figure 1).

According to Figure 1, an inverselyproportional linear relationship betweenstabilization time of sound pressure level androad traffic intensity is observed (R2 = 0.87).Low road traffic intensities correspond tointermittent traffic flow dynamics, so theshort-term temporal fluctuation is higher, andtherefore stabilization time has high values. Incontrast, great road traffic intensities implymoderately stable sound levels, because of thesaturation and accumulation of noise sources.Under these circumstances short-termtemporal fluctuation is smaller, and thestabilization time is quite low. Therefore,under lower road traffic intensity, stabilizationtime value increases [6].

From Figure 1 and, knowing the roadtraffic flow of a given urban localization, theminimum required time to measuring toundertake a short-term sound characterization(calculation of LAeq,T) of this urban locationcan be obtained.

4.2. Stabilization time vs. impulsivesound event occurrenceAs stated previously, in urban environmentsthe occurrence of impulsive sound events isvery common [19]. In Table 1 it can be seenthat in urban locations with impulsive sound

events appearance, stabilization time value isgreater than in the urban settings without thiskind of sound events.

In this section, the behaviour of thestabilization time of the sound pressure levelagainst the occurrence of one impulsive soundevent is examined. The incidence of the timeposition and magnitude of the impulsivesound event on the stabilization time value isthoroughly analyzed.

To evaluate the influence of impulsivesound event time-position on the value ofstabilization time, in a sample measurementwith a relatively low short-term temporalfluctuation (stabilization time equal to 11.92min) an impulsive sound event with 30 dBAmore than the sampled measurement (with noimpulsive sound event) LAeq is introduced.When this event is introduced the stabilizationtime increases to 41.33 min (Figures 2b-c).This event is added in minute 20 (Figure 2b)and 41.33 (Figure 2c), obtaining that thestabilization time value is the same in bothcases. Indeed, the value of the stabilizationtime if an impulsive sound event is addedinside the stabilization time period, in thesampled measurement, can be said to beindependent of its position. However, whenthis impulsive sound event is introduced attime above 41.33 min, as in figure 2(d) wherethe impulsive sound event appears in minute



VALUE


Figure 2. Evolution of the stabilization time according to the location of impulsive sound event. (a)Without impulsive sound event, (b) in minute 20, (c) in minute 41.33 and (d) in minute 45 [6].

45, then the stabilization time correspond tothe time in which the high emerging noiseappears [6].

On the other hand, a sound event with 20dBA (Figure 3a), 25 dBA (Figure 3b) and 30dBA (Figure 3c) more than the sampledmeasurement (with no impulsive sound event)LAeq is introduced to evaluate the influence ofmagnitude of the impulsive sound event onstabilization time. As expected, if the soundpressure level added by the sound event ishigher, the stabilization time value (26.16 min,34.42 min and 41.85 min, respectively)increases. From this results can be deducedthat, the required time for the sound pressureincrease to become steady depends on itsmagnitude: the greater the generated soundpressure increase above the mean soundpressure level (without sound event), thehigher the time necessary for stabilization totake place [6].

If the appearance of an impulsive soundevent takes place, stabilization of the soundpressure incorporated by the impulsive soundevent can be produced according to thefollowing function:

(2)

where Leq(t) is the energy-equivalent soundpressure level; T is the total time; TISE,i is thetime corresponding to the appearance of theimpulsive sound event; SPLmean is the meansound pressure level without the appearanceof impulsive sound events; and SPLISE,i is thesound pressure level introduced by theappearance of the impulsive sound event [6].

In turn, when at the beginning of a soundpressure level sample with a constant valueevents with 1 to 50 dBA of sound pressuremore than this constant LAeq (Figure 4) areintroduced, can be checked that thestabilization time increased, according to alogistic function (R2 = 0.99), when the soundevent magnitude was higher. According toFigure 4, for increases below 10 dB and above30 dB, stabilization time increases slightly.However, if the sound augment is comprisedbetween 10 and 30 dB, an increase in thestabilization time value appears close to thelinear function.

4.3. Calculation of LAeq,T from thestabilization time valueA general procedure to LAeq,T descriptorestimation from the value of the stabilizationtime of the sound pressure level is put forth inthis section. The presented procedurecontemplates two ways of calculation.

On the one hand, to obtain the minimumrequired time for measuring in a given urban

L t LogT

T t teq ISE i

SPL

ISE i

mean

( ) , ,= ⋅ −( ) ⋅ + ⋅101

101010 110 10

SPLISE i,



VALUE

15JANUARY 2012

Figure 3. Evolution of the stabilization time according to the sound pressure introduced by the impulsivesound event, 20 dB(A) (a), 25 dB(A) (b) and 30 dB(A) (c) [6].

localization a representative value of LAeq,T ofthe considered short-term time period, thestabilization time of the sound pressure levelcan be estimated from the road trafficintensity value (Figure 1).

On the other hand, if a deterministicimpulsive sound event, which deeply marksthe temporal structure, being a quiterepresentative element of the ambient sound(e.g. siren of a school, church bell, etc.) appearin a given urban location, to take arepresentative environmental noisemeasurement of this localization we mustmeasure up at least the necessary time, asobtained in Figure 4, in order that the soundpressure level introduced by this event bestabilized. However, if the stabilization timeindicated by Figure 1 is higher, this one will bethe minimum time to measuring.

It is suitable indicator that if the leastmeasurement time indicates by the Figure 1 isover and yet the deterministic impulsive soundevent (that we know appears once during themeasurement period in this urban location)has not appeared, then it is necessary toextend the period of measurement at leastuntil this event appears and the soundpressure level introduced become stabilized.In addition, if already the required time to thesound pressured level introduced become

stabilized has passed, only until the eventappears (Figure 2).

To conclude, this procedure will be valid ifit is known that the deterministic impulsivesound event appears only once during theconsidered short-term time period. Moreover,this procedure is independent of the timewhen the measurement begins. Finally, ifother impulsive sound events, which are notrepresentative of the considered urban soundenvironment (e.g. ambulances, honking horns,vehicles circulating at high speeds, etc.) appearwhile the environmental noise measurement istaken, these sounds events have to be detectedand removed to the estimation of thedescriptor LAeq,T (from the value of thestabilization time).

5. ConclusionsIn this paper it is put forth a generalprocedure to estimate, from the value of thestabilization time of the sound pressure level,a representative value of the LAeq,T descriptorin a short-term time period for a givenenvironment. According to the introducedmethodology, to obtain an appropriate short-term characterization of the ambient sound ina given urban localization it is necessary tomeasure at least a minimum period of time,which is provided by the parameter



VALUE


Figure 4. Relationship between the stabilization time of the sound pressure level and the sound pressureintroduced by the impulsive sound events [6].

stabilization time. In this work, thismethodology is applied to 60-min time period,nevertheless, this procedure could beextrapolated to any short-term time period.

Acknowledgments This work was supported by the ‘‘Consejeríade Innovacion, Ciencia y Empresa de la Juntade Andalucía’’ of Spain under project TIC 03269.

References[1] L. Leclercq, J. Lelong, Dynamic

evaluation of urban traffic noise, in:Proceedings of the 17th InternationalCongress of Acoustics, Tome, Italy,2001.

[2] B. De Coensel, T. De Muer, I. Yperman,D. Botteldooren, The influence of trafficflow dynamics on urban soundscapes,Applied Acoustics 66 (2005) 175-194.

[3] J. Alberola, I.H. Flindell, A.J. Bullmore,Variability in road traffic noise levels,Applied Acoustics 66 (2005) 1180-1195.

[4] A.J. Torija, N. Genaro, D.P. Ruiz, A.Ramos-Ridao, M. Zamorano, I.Requena, Priorization of acousticvariables: Environmental decisionsupport for the physical characterizationof urban sound environments, Buildingand Environment 45 (2010) 1477-1489.

[5] A. Can, L. Leclercq, J. Lelong, J.Defrance, Capturing urban traffic noisedynamics through relevant descriptors,Applied Acoustics 69 (2008) 1270-1280.

[6] A.J. Torija, D.P. Ruiz, A. Ramos-Ridao,Required stabilization time, short-termvariability and impulsiveness of thesound pressure level to characterize thetemporal composition of urbansoundscapes, Applied Acoustics 72(2011) 89-99.

[7] D. Botteldooren, B. De Coensel, T. DeMuer, The temporal structure of urbansoundscapes, Journal of Sound andVibration 292 (2006) 105-123.

[8] J. Romeu, M. Genescà, T. Pàmies, S.Jiménez, Street categorization for theestimation of day levels using short-termmeasurements, Applied Acoustics 72(2011) 569-577.

[9] W.A. Utley, Temporal samplingtechniques for the measurement ofenvironmental noise, Applied Acoustics15 (1982) 191-203.

[10] E. Gaja, A. Gimenez, S. Sancho, A. Reig,Sampling techniques for the estimationof the annual equivalent noise levelunder urban traffic conditions, AppliedAcoustics 64 (2003) 43-53.

[11] H.B. Safeer, Community noise levels-astatistical phenomenon, Journal of Soundand Vibration 26 (1973) 489-502.

[12] A.E. González, M. Gavirondo, E. Pérez,A.A. Bracho, Urban noise: measurementduration and modeling of noise levels inthree different cities, Noise ControlEngineering Journal 55 (2007) 367-372.

[13] Y. Mitani, M.A. Ohta, A calculation ofLx and Leq noise evaluation indices byuse of statistical information on the noiselevel fluctuation, and its microcomputer-aided on-line measurement, AppliedAcoustics 25 (1988) 33-47.

[14] A.E. Gonzalez, Monitoreo de ruidourbano en la ciudad de Montevideo:determinación del tiempo óptimo demuestreo y desarrollo de un modelopredictivo en un entorno atípico,Doctoral Thesis, Universidad de laRepública, Montevideo, Uruguay, 2000.

[15] A.E. Conzalez, E. Gaja Diaz, A. Jorysz,G. Torres, Monitoreo de ruido urbano:determinación del tiempo mínimo demuestreo en la ciudad de Montevideo,Uruguay, in: Proceedings of the XXXICongreso Nacional de Acústica(Tecniacústica 2000), Madrid, Spain,2000.

[16] E. Gaja Diaz, E. Gonzalez, A. ReigFabado, Influencia de los anómalosacústicos en el tiempo de estabilizacióndel nivel equivalente horario, in:Proceedings of the XXXI CongresoNacional de Acústica (Tecniacústica2000), Madrid, Spain, 2000.

[17] A.J. Torija, D.P. Ruiz, A. Ramos, Amethod for prediction of the stabilizationtime in traffic noise measurements, in:Proceedings of the 19th InternationalCongress on Acoustics, Madrid, Spain,2007.



VALUE

17JANUARY 2012

[18] ISO 1996-2 Acoustics – Description,assessment and measurement ofenvironmental noise – part 2:Determination of environmental noiselevels, 2007.

[19] A.J. Torija, D.P. Ruiz, A. Ramos,Obtaining of a factor to describe theanomalous sound events in traffic noisemeasurements, in: Proceedings of the19th International Congress onAcoustics, Madrid, Spain, 2007.



VALUE


Smart, healthy travel in a single European transport area by 2020

To complete a safe, efficient, low-pollution single transport area the EU must set itselfrealistic but binding environmental and safety goals for 2020, says Parliament in aresolution in December. Pollution, noise and congestion costs should be included in theprices paid by users, MEPs said. The “roadmap to a single European transport area” withfigures, as approved by Parliament, supplements the long-run view of a Commission whitepaper on the future of EU transport policy. “Co-modal efficiency encompasses economic,environmental and safety aspects and it includes decent working conditions”, saidrapporteur Mathieu Grosch (EPP, BE), in the debate. “Parliament wishes to send a strongsignal to Member States to accelerate the transposition of directives into national lawsand abolish remaining ‘invisible frontiers’ so as to complete a single, co-modal andefficient transport area” he added. MEPs ask the Commission to table a proposal by2014 for the “internalisation of external costs” for all modes of transport, with a view toincluding noise, pollution and congestion costs in the price paid by the user. MemberStates should use the revenue that this generates to fund sustainable mobility andtransport infrastructure costs, says the resolution.

Collected Papers inBuilding Acoustics: Room Acoustics andEnvironmental NoiseEdited by: Barry Gibbs, John Goodchild, Carl Hopkins

and David Oldham

ISBN 978-1-907132-14-8 · vi + 419pp · 165 x 240mm · £69 · Publication date: May 2010

This book contains a collection of papers on room acoustics and environmental noise.Room acoustics is split into four sections: auditorium acoustics, acoustics of religiousbuildings, acoustics in schools, and absorption, diffusion and reverberation time; five paperson the technically related topic of environmental noise complete the book. The range ofissues covered by the book cannot be exhaustive, but the editors believe that currentconcerns of researchers and practitioners are well addressed. The book includes papers bydistinguished figures such as Leo Beranek, R. J. Orlowski, Fergus Fricke and MurrayHodgson, among others. Although topicality has been considered in selecting these papersfor publication in book form, having originally been published in the journal BuildingAcoustics, the emphasis has been on quality and importance. Papers were selectedbecause of their rigor, citation history and contribution to the science and practice ofacoustics. In a convenient form, this book constitutes a significant corpus of knowledge onroom acoustics and environmental noise and thus will be of interest to architects and civilengineers, as well as to academic researchers in these fields. The editors are members ofthe Acoustics Research Group at the University of Liverpool.

estimation procedure of the descriptor laeq,t from the

Documents