THE EU NOISE POLICY AND ITS RESEARCH NEEDS

Transcription

1 THE EU NOISE POLICY AND ITS RESEARCH NEEDS PACS: x Ten Wolde, Tjeert Formerly: TNO TPD, Delft Graaf Willem de Rijkelaan TK, Leidschendam The Netherlands Tel: Fax: - tjeert.tenwolde@planet.nl ABSTRACT The new EU noise policy and particularly the new Directive relating to the Assessment and Management of Environmental Noise has a significant influence on the research and development in the field of noise control. An important part of this R&D is generated by the need for harmonised noise mapping and for harmonised assessment of annoyance and sleep disturbance. Other R&D needs are related to a greater attention for noise control at the source and to a greater attention for the economic aspects of noise abatement. A special area of research concerns the protection of quiet areas in the open country. INTRODUCTION For noise control in Europe, 2002 is a very important year because in this year the European Union will adopt the Directive relating to the Assessment and Management of Environmental Noise (DAMEN) [1]. That directive covers the field of environmental noise in a similar way as the European directives for the quality of the air and for the quality of water. It provides harmonised information for authorities and citizens, the making of national and local action plans and the extension and improvement of the EU legislation on the reduction of noise emission. The new directive will introduce noise legislation in a number of EU Member States that do not have such legislation at present. In the other Member States it will supplement the existing legislation with new elements. The introduction, execution and further development of the Directive requires a significant amount of research. This paper provides a concise overview of that research. The relation between noise policy and research is not ideal. The communication between scientists and policy makers is strongly handicapped by semantic problems, for example on the meaning of the words health, environmental noise, annoyance and even computation methods. To avoid such problems for the readers of this paper, some definitions are provided and commented in the next chapter. A second disturbing factor is the tendency by some policy makers to take personal experience or assumptions as the basis for judgements, for example on the effects of noise and on the

2 costs of noise control. For scientists it is obvious that personal experience and assumptions are seldom reliable in this area and that scientific results should provide the main guidance. A third source of misunderstandings is the difference in the legal approaches in different countries. In some countries noise control is mainly based on reducing nuisance for individuals or for small communities by a fair local approach. Improvements are beneficial for individuals or small groups. In other countries environmental noise control is mainly a matter of a strategic long term approach on the basis of central legislation. Improvements occur for large groups, but may be negligible for specific individuals. The EU approach is mainly of the second type, but can be very well combined with national or local actions of the first type. A fourth source of misunderstanding and disagreement is related to the uncertainty of data. Authorities have great difficulty to accept that measured or computed data are uncertain. It is a complication for comparisons with limit values, for the definition of noise zones and for other elements of legal processes. For scientists it is obvious that the proper action should be to make the uncertainty as small as possible and to improve the methodology continuously. Authorities however, seldom take that line and are generally more motivated to create a stable legal situation in which methods are frozen and uncertainty is denied. The DAMEN is based on the scientific approach, but elements of the second approach have also been introduced. Thus, the Directive is a compromise between the two attitudes, which will undoubtedly cause some difficulties in the future. In this paper it is assumed that the scientific line will be dominant. DEFINITIONS The definitions of the DAMEN apply. Unfortunately, due to the processes outlined above, some of these definitions are not entirely clear and need some further explanation. The first notion is environmental noise, which means according to article 3 of [1]: unwanted or harmful outdoor sound created by human activities, including noise emitted by means of transport, road traffic, rail traffic, air traffic, and from sites of industrial activity such as those defined in Annex I to Council Directive 96/61/EC of 24 September 1996 concerning integrated pollution prevention and control. This definition is correct concerning the origin of the sound but makes not clear that the directive only concerns unwanted or harmful sound in the domestic environment i.e. in and around dwellings, in parks, in schools, etc. Excluded are noise in the working environment, noise inside means of transport, noise from neighbours and noise made by the exposed persons themselves. The latter points are made clear in article 2.2 of the directive. On the other hand article 2.1 of the directive may confuse the reader by giving the impression that environmental noise is only noise outside buildings, while in reality it mainly concerns noise penetrating buildings and affecting the people inside. As an alternative for environmental noise the term community noise is often used. In this paper community noise is defined as environmental noise plus the noise from neighbours. Human health means a state of complete physical, mental and social well-being, as defined by the constitution of the World Health Organization. With this definition, the harmful effects of noise on humans are clearly part of the public health domain. The EU has adopted this line. Unfortunately, in discussions on noise (and on other environmental problems), health is often defined as the absence of physical illness or even as the absence of mortality. In that approach environmental noise is reduced to a comfort issue or a quality of life issue, i.e. primarily a personal matter. The DAMEN defines L den (day-evening-night noise indicator) as the noise indicator for overall annoyance, as further defined in Annex I. The latter annex provides the general formula and all the necessary details on the quantity. It is a pity for the general harmonisation and for the comparability of results that the directive allows a shift of one or two hours from the evening period to the day period or to the night period. In this paper that option is neglected.

3 For scientist it is probably not necessary to define computation methods, but it appears that other people very often have another notion of this term. Examples are the following: 1. the computer programme is often confused with the method, 2. a particular database with emission values is sometimes seen as an integral part of the method and not as an input which can be varied and should be updated, 3. the definition of the quantity to be determined is sometimes coupled to a particular prediction method. In this paper the scientific approach is followed, making a clear difference between the quantity to be determined, the input, the computation method and computer programmes. EU NOISE POLICY AND DIRECTIVES The central element of the new EU policy on environmental noise is the DAMEN [1]. It is the basis for a harmonised approach in all EU Member States and the basis for an improved EU approach regarding the reduction of noise emission by certain sources. Article 1 of the DAMEN describes the objectives: a common approach to avoid, prevent or reduce on a prioritised basis the harmful effects, including annoyance due to environmental noise by: Noise mapping. Public information. Action plans by the Member States. Completing the EU measures to reduce the noise emitted by specific sources. The other articles and the annexes of the directive provide more details. Background information can be found in [2] and [3]. Figure 1 shows the principles. HARMONISATION OF NOISE INDICATORS AND ASSESSMENT METHODS LIMIT VALUES (optional; set by the Member States) NOISE MAPS for MAJOR AGGLOMERATIONS, ROADS, RAILWAYS AND AIRPORTS European Commission ACTION PLANS INFORMATION/CONSUL- TATION of the PUBLIC Proposals for: 1. FURTHER EU ACTIONS, particularly on NOISE SOURCES 2. PROTECTION OF QUIET AREAS IN OPEN COUNTRY European Council and European Parliament Fig. 1.- Overview of the European directive on environmental noise (DAMEN). The directive will have a great impact on noise control in the EU Member States which have presently no or little central legislation on environmental noise (Belgium, Greece, Ireland, Portugal, Spain and the United Kingdom). For the other Member States, the directive adds significant elements to the existing legislation. Particularly the following elements of the directive will generate research needs:

4 1. the harmonisation of noise indicators and assessment methods, including the making of noise maps and the prediction of health effects, 2. the (optional) setting of limit values, 3. the action plans, 4. the further EU actions and particularly those to reduce noise emitted by the major noise sources (road and rail vehicles and infrastructure, aircraft, outdoor and indoor industrial equipment and mobile machinery), 5. the development of a policy for the protection of quiet areas in the open country. There are already several EU directives on the control of noise emission and others are being prepared. Table 1 gives an overview. The directives have rather different set-ups, as well legally as technically. Some are very effective for the reduction of environmental noise, others do contribute little - see for example [4]. The DAMEN provides the opportunity to bring the noise emission directives gradually more in line with each other and to make them more effective for the improvement of the environment. RESEARCH NEEDS On The Character Of The Research In this paper the word research is used in a very wide sense. It covers academic research, development of better methods based on existing academic knowledge and existing technology, gathering of data with existing methods, noise control at the source, noise control in the path(s) between source and receiver. Thus, it may include contributions from universities, research institutes, consultants and industry. In the next paragraphs the research needs are explained, starting from the needs of the EU Directives. Measurement Of The Value Of The Noise Indicators In the DAMEN, two quantities to characterise noise have been defined: L den and L night. Both are based on the determination of A-weighted long-term average sound levels. For L night it concerns the L Aeq over all night periods of a year, for L den it also includes the L Aeq values over all the day and evening periods of a year. The determination of the L Aeq values can be done by measurement, by computation or by a mixture of measurement and computation. The fact that the directive allows the different approaches implies that the results shall agree within reasonable margins, which means in practice that the methods shall be as accurate as technically and financially feasible. In principle the measurement of L Aeq is straightforward and shall be carried out in accordance with ISO : 1987 [5] or ISO 3891: 1978, or in the future, in accordance with newer versions of these standards. In reality there are some problems, however. In order to avoid these the European Commission should consider the development of guidelines. There are two major cases for which that is useful: 1. Unmanned long-term measurements ( noise monitoring ). Such measurements are rather common near airports and are sometimes applied near industry, roads and railways. 2. Manned or unmanned short-term measurements (i.e. a combination of short-term measurement and computation of the long-term quantity from the measured data). In both cases there is the problem of the signal-to-noise ratio and particularly the discrimination of the noise under consideration from the noise from other sources. Furthermore, if the measurements are done in relation to annoyance or sleep disturbance of people in buildings, there is the problem to avoid (or correct for) the contribution from the façade reflection. Due to these difficulties, the uncertainty in the value of the long-term noise indicators as determined

5 Table 1. - Overview of the EU legislation on the control of noise emission in relation to environmental noise. Noise source(s) Equipment for use outdoors Aircraft Document code(s) 2000/14/EC 80/51/EEC, 89/629/EEC, 92/14/EEC Type of measures Type approval with marking and limit values Type approval (certification) with limit values; outphasing (limitations for the use) Aircraft 2002/30/EC Establishment of rules and procedures with regard to the introduction of noise-related operating conditions at Community airports Cars, trucks and motor cycles (motor vehicles) Tyres for motor vehicles Trains Recreational craft (boats) 70/157/EEC, 73/350/EEC, 84/372/EEC, 84/424/EEC, 96/20/EC, 97/24/EC 2001/43/EC Technical Specifications for Interoperability (TSIs) for the Directives 96/48/EC and 2001/16/EC COM(2000)639 final. Revision of Directive 94/25/EC Type approval with limit values Type approval with limit values Limit values and/or recommendations for rolling stock and/or infrastructure Type approval with limit values Formal status Operational since Operational Entered into force on Operational Soon operational Under preparation Proposal in discussion in Council and European Parliament Remarks Ground moving machinery, gardening machines and others; the directive replaces 9 old directives Civil subsonic jet aeroplanes Council regulation No 925/1999 ( hushkits ) is repealed by the entry into force of this new directive Directive 70/157/EEC has many amendments First legislation on tyre noise in the world High speed and conventional railway systems

6 with the present noise monitoring systems can be considerably larger than 1 db [6], and improvements are necessary. In the case of short-term measurements, assumptions must be made on the samples which are necessary to allow the computation of L den or L night from the measured data. Without guidance these assumptions will be very different and generate great variations and uncertainty. Furthermore also these measurements must have a sufficiently large signal-to-noise-ratio, including a good discrimination of the noise under investigation (the signal ) from the noise from specific other sources. An overview of the two research items that are shown in the next table. necessary to solve the above problems are Table 2. - Research items related to the measurement of the noise indicators. 1 Accurate unmanned long-term measurements ( noise monitoring ), particularly near airports and industry. 2 Guidelines for manned or unmanned short-term measurements, particularly near industry, roads and railways. Improvement of the signal-to-noise ratio, ( noise including all unwanted signals), for example with the aid of microphone arrays. Guidelines for the selection of short-term samples, for the measurement of these samples, and for the computation of the long-term L Aeq, L den and/or L night. Computation Methods And The Related Emission Measurement Methods It is certain that in most cases the value of the noise indicators will be determined by computation because it is cheaper, quicker and not necessarily less accurate than determination by measurement. Furthermore computations can also be used for the design of certain measures, like the introduction of a noise barrier. In order to play their role properly, the computation methods should be able to determine the value of the noise indicator with an uncertainty not greater than 1 db. Unfortunately, the present methods are not able to provide this accuracy. All methods that are presently used in practice (for rail, road and aircraft noise and for industrial noise) are empirical or semi-empirical methods which contain many simplifying assumptions and have a very simple description of the source. Such methods can only be accurate for sources and geometries which are similar to the sources and situations of the original data set. Another weakness is that the description of the source and the description of the transmission are interdependent. The basis for most of these methods was laid some 30 or 40 years ago and they are scientifically and technically outdated. The best are certainly able to give acceptable results for simple situations close to the source, but they probably all fail for more complicated geometrical situations, for situations with noise barriers and for somewhat larger distances [7,8,9,10,11,12]. Regarding noise barriers, there are strong indications that the attenuation predicted by the present models is systematically too high. The reason is a greater influence of meteorological factors [16,17,18]. Some aspects of the weaknesses and the differences of the models are illustrated in figure 2. These are very alarming. Differences of 5dB are no exception. In spite of the obvious shortcomings, the methods form the backbone of the present noise policies in most EU Member States. It is obvious that the differences in computation methods seriously undermine the possibilities for comparison of results. In the last 20 years there has been considerable progress in the modelling of the transmission of outdoor sound, using modern computer technology [13,14,15,18,20]. The technical possibilities to characterise the source and to introduce the system geometry have also been improved. Some of these possibilities are introduced in the French NMPB method [1,20], in the transmission models of the newest versions of the Integrated Noise Model (for aircraft noise) [19] and particularly in the Scandinavian Nord2000 method [30]. Several problems have not been completely solved however, and sufficiently accurate, validated, modern models are not yet available. Thus, completely satisfying methods are presently not available and the European Commission had no other option than to start the DAMEN with existing computation methods. The development of better methods has started. These methods shall be introduced in a later

7 (A) (B) (C) (D) (E) (F) Fig. 2.- Comparisons of computation models and elements of computation models: A: Results of a round robin test for road traffic noise models; differences between L Aeq daytime and average (after Van den Berg and Gerretsen [8, 9]) B: Emission data for passenger cars in five models (after Van den Berg and Gerretsen [8, 9, 10]) C: Excess attenuation for transmission of road traffic noise over an absorbing ground with downwind conditions in four national models (after Van den Berg and Gerretsen [8, 9, 10]) D: Source position in 14 prediction models for railway noise; the differences result in significant differences in predicted barrier insertion losses (after Van Leeuwen [11]) E: Lateral attenuation for aircraft noise with distance, at two elevation angles (0 o and 10 o ) for three models (after Van den Berg and Gerretsen [8, 10]) F: Ground effect in ISO , measured and computed by ray theory for a specific geometry and absorbing ground (after Kragh, [16]).

8 stage. The DAMEN gives four recommended interim methods : a French method for road traffic noise, a Dutch method for railway noise, an ISO method for industrial noise and the so-called ECAC Doc. 29 method for aircraft noise. All these methods require some adaptation before they are ready for application in the scope of the DAMEN, for example regarding the emission data base. Furthermore, appropriate emission measurement methods shall be explicitly given. According to Article 6.2 of the DAMEN, Member States are allowed to continue the use of their own national methods, provided that those methods give equivalent results to the results obtained with the recommended interim methods. This means in practice that those Member States that continue to use their present national models will violate the directive, unless a very lenient definition of equivalent results is adopted. Apart from the aspects that are directly related to the modelling of the source and the transmission, guidance shall also be given on some other aspects like the modelling of the geometry, the gathering of traffic data, the application of dose-effect relations and the presentation of results. Taking the above into account, the following research projects can be identified see table 3. Table 3.- Research items related to computation. 3 Making the recommended interim computation methods operational for use throughout the EU software. 4 Guidelines for noise mapping 5 Modern, more accurate methods for road traffic noise and railway noise 6 More accurate methods for aircraft noise and industrial noise 7 Operational more accurate methods, applicable throughout the EU Adaptation to the definitions of the EU noise indicators; emission measurement method for some sources; data base for sources; small improvements; Organising the input (traffic data etc.); modelling of the geometry; application of dose-effect relations; presentation of results. Independent source and transmission characterisation; better source descriptions and emission measurement methods; better transmission models, particularly for cases with screening, reflections, different ground conditions, larger distances. Better modelling of the transmission and the source, independent source and transmission characterisation. Emission data bases, including data for different road surfaces, rail tracks, meteorological influences and other aspects; software; guidelines. A project covering most of subject 3 is already underway and results will become available in Some additional actions, like the making of commercial software, will follow soon after. It should be expected that subject 4 will start in A European project covering subject 5, called HARMONOISE [31], has started at the end of Work is also underway for subject 6. For aircraft noise, the subject is scientifically and technically probably less difficult than subject 5, but political influences are much more dominant and may limit the possibilities for fundamental improvements. Subject 7 shall wait till the results of subjects 5 and 6 are available.

9 Exposure-Response Relations ( Dose-Effect Relations) According to annex VI of the DAMEN, final results of noise mapping are presented in terms of the number of affected people. It is believed that such data form a much better basis for communication with the public and for well balanced action plans than the value of the (physical) noise indicators. For that purpose it is necessary to have suitable exposure-response relations. The two effects that are distinguished in the DAMEN are annoyance and sleep disturbance. Both must be defined more precisely. For annoyance that is not too difficult because a reasonably well established procedure for noise surveys has been developed from which three quantities can be derived: the percentage Highly Annoyed people (%HA), the percentage Annoyed people (%A) and the percentage Lowly Annoyed people (%LA) [22]. The latter quantity is seldom used, but a choice should be made between %A and %HA on the basis of an EU working group position paper on that topic. It is important to realise that the EU is only looking for the average exposure-response relations for large groups, in stable, or slowly changing situations. That is also the case for sleep disturbance. On the basis of existing literature, a study project has been launched by the European Commission in order to choose a descriptor for sleep-disturbance, to be used for dose-effect relations in which the dose quantity will be L night. The %HA as a function of the value of L den (or related quantities) has been determined in many field surveys on road, rail and aircraft noise. Most of these surveys were carried out in the USA and Northern Europe. The average results are published by Miedema and Vos [22 ] and will also be published by the European Commission. There are no comparable results for industrial noise. Furthermore it should be doubted whether the results on the traffic noises are also applicable in polar and sub-tropical climates as occurring in Northern and Southern EU Member States. Different living patterns and different building structures may give rise to different exposure-response relations. Another situation with different dose-effect relations occurs when dwellings are especially insulated against road, rail or air traffic noise. A special research item is the determination of exposure-response relations for children at school and at home. There are strong indications that noise reduces their learning possibilities. The above gives rise to the research subjects listed in table 4. Table 4.- Research subjects related to the definition of effects and to dose-effects relations 8 Annoyance L den for traffic Road, rail and air traffic noise noises 9 Selection of one effect Shall correlate with L night. (response) descriptor for sleep disturbance. 10 Further dose -effect relations for annoyance. Northern and Southern Member Sates; houses that are especially insulated against road, rail or 11 Dose-effect relations for sleep disturbance 12 Dose-effect relations for children air traffic noise; industrial noise. The dose quantity is L night ; all EU climates; road, rail and air traffic noises; specially insulated houses. Learning capability. A project concerning subject 8 has been completed and publication by the European Commission can be expected. European projects relating to subjects 9 and 12 [32] are also underway.

10 In application of the DAMEN, Annex III of the directive will be revised to introduce such doseeffect relations (through a so-called comitology procedure associating the European Commission and the Member States). Emission Measurements Methods For Type Testing and Marking The European Directives listed in table 1 require good testing methods for the determination of a quantity which characterises the noise emission in an appropriate way i.e. in such a way that a reduction of the value of that quantity directly results in a reduction of environmental noise in practice. In this area there are many weaknesses, which shall be reduced: 1. The uncertainties of several test procedures are too large (i.e. larger than 1dB). An important example is the type testing method for motor vehicles. It is well known by experts that the freedom in the present method for the altitude of the test track, the temperature during the tests, the test track surface, the surroundings of the test track, the tyres and other factors result in a far too large uncertainty and in possibilities for manufacturers to arrange a favourable result. 2. For some vehicles/machines, the operation conditions during the test are not appropriate, because they do not sufficiently well represent operating conditions in practice. Again, the type testing method for motor vehicles is the most important example [4,23]. 3. For rail and road vehicles it is difficult to separate the direct emission from the vehicle from the emission from or via the track. Present procedures must be improved [23, 29]. Table 5 defines two research subjects covering the above. Table 5: Research related to the quality of type testing methods. 13 Quality check of (all) type testing Relevancy for practice; uncertainty methods 14 Improvement of type testing methods Greater relevancy; smaller uncertainty Some research related to subject 14 has already been done or is underway. That is for example the case for motor vehicles, aircraft and trains. A start has been made (in ISO and ICAO) to implement some results in standards. Much more should be done, however. Noise Control At The Source The DAMEN and the other European Directives stimulate the development of quieter transport systems, machines and equipment. Table 6 gives a survey. Table 6.- Research related to the reduction of noise emission 15 Quieter motor vehicles and roads Engine, tyres, exhaust, porous road surface, other silent road surfaces 16 Quieter aircraft Airframe noise, engine noise 17 Quieter rail vehicles Rolling noise from vehicles and track, aerodynamic noise, engine noise 18 Quieter recreational craft Exhaust, engine 19 Quieter lawn mowers and other outdoor equipment Engine, (exhaust), process Although much research is already carried out on quieter road traffic (subject 15), it is rather scattered and not very effective. That is much better for railway noise (subject 17) and even more for aircraft noise (subject 16). Presently, a large European project (called SILENCER) is carried out on aircraft noise and more research is being planned.

11 In the past decade, much progress has been made on the quieting of rail vehicles. Further research is needed however, for example on the development of cheap braking systems for freight trains that do not roughen the wheels and rails, and on the reduction of squeal noise. From all the machinery covered by the Directive on the noise emission by outdoor equipment, the lawnmower is probably causing most nuisance and improvement seems necessary. A study on this subject is underway. Costs And Benefits Of Noise Control For a well balanced noise policy it is very important that costs and benefits of certain measures are determined in a relevant and accepted way. Very much work has still to be done in this area. The approaches that are presently available neglect major aspects, consider the issue only from the point of view of one player (like the national road authority), or use unreliable data. For the European Union, the costs and benefits at macro (European) scale are of primary importance. However, costs and benefit considerations for the parties that shall develop the action plans must also be developed. A start in this area has been made in a special session and a Workshop at Inter-noise 2001 [27]. The benefits should be expressed as well as the reduction of the number of annoyed and sleep disturbed people as in Euros. For both it is essential that the EU noise mapping provides reliable data on the number of annoyed and sleep disturbed people (thus, the projects 3 to 11 are vital for the issue of benefits). In order to express the benefits in Euros, a valuation system must be developed. Table 7 provides an overview. Table 7: Research on costs and benefits 20 Valuation of changes in annoyance and sleep disturbance Effect of noise on housing prices, medical costs, lost labour days and other aspects, to be related with the number of affected people. Note: the author is sceptical about the willingness-to-pay approach, which is 21 Cost-benefit models for decision making at the EU level 22 Cost-benefit models for use at the level of agglomerations, major roads, major railways and major airports sometimes used in this context. Cost-benefit analysis of sharper emission limits, limitations for the use of noisy machinery and other EU measures (as for example the stimulation of the use of porous road surfaces). Cost benefit analysis of measures that could be considered as part of the action plans (noise barriers, traffic control, insulation of dwellings, land use, etc.). Some work related to these subjects has already been done [24,25,26,27], but major improvements are necessary. Co-operation between different disciplines (technical, sociological, economical) is essential in this area. Protection Of Quiet Areas In The Open Country According to Article 11.2 of the DAMEN the European Commission shall assess the needs for implementing a strategy for the protection of quiet areas in the open country, which provides a subject for research see table 8. Table 8.- Research related to the protection of quiet areas in the open country 23 Criteria for the quality of a quiet area Role and characterisation of the noise from in the open country different sources.

12 Some Member States have experience with the concept of quiet areas and there is some literature available. Several questions are still open, however. FINAL REMARKS The research outlined in this paper coincides to some extend with research that was already carried out in other contexts than the new EU noise policy. Examples are the research programs for further reduction of noise emission by aircraft and the study of the effect of noise on school children. Most other subjects are getting an extra stimulus, however. There are some popular research items that are missing in the above. A striking example is the combination of effects caused by noise from different sources and the related characterisation with one noise indicator. At present, this approach is not included in the EU noise policy because appropriate combination is considered to be impossible when important questions on the impact of single types of noise are not yet answered (see research subjects 8 to 12). Furthermore, in the majority of cases, appropriate decisions on noise control measures can be very well based on estimations of the number of affected people from the separate types of noise. An important role in the planning of a European research program supporting the development of the DAMEN, is played by the European thematic network CALM [28]. Regarding noise emission from civil aircraft, such planning is done by a working group of the Advisory Council for Aeronautics Research in Europe (ACARE) in co-operation with the thematic European network X-NOISE. These networks particularly consider the long-term strategic parts of the research. Other pieces of the research are initiated by different Directorat-General of the European Commission ( Environment, Transport and Energy, Enterprise and the Information Society, etc.) by Member States and by Industry. Independent of the EU developments, a group of German research institutes, universities and consultants recently started a large research program on road, rail and air traffic noise [33]. It is called Leiser Verkehr, it is sponsored by the German Ministry of Research and it includes many of the subjects that are listed in this paper. Links with the EU research have been established. BIBLIOGRAPHICAL REFERENCES [1] Future European Directive relating to the Assessment and Management of Environmental Noise, based on a proposal of the European Commission presented in COM(2000)468final, Brussels, In February 2002 complete agreement between the Council of Ministers and the European Parliament was reached. Formal approval (in third reading ) and publication in the Official Journal is expected to occur in the course of [2] T. ten Wolde, The EU policy on environmental noise, Proceedings Inter-noise 2000, Nice, [3] B.M. Ross and T. ten Wolde, Global and continental policy as a support for local noise control, Noise Control Eng. J, 49(4). 2001, [4] U. Sandberg, Tyre/road noise Myths and realities, Proceedings Inter-noise 2001, The Hague, [5] International Organization for Standardization. ISO : Acoustics Description and measurement of environmental noise. Part 2: Acquisition of data pertinent to land use. Geneva: ISO, [6] Measurement of noise around Schiphol Airport. Accuracy of possible measurement systems (in Dutch), TNO report , prepared by TNO, NLR and TU Delft, June 2001, Delft. [7] S. Jonkhart, The existing methodologies for the calculation of aircraft noise load in the EEC member states, Internoise 93, Leuven, 143. [8] M. van den Berg and E. Gerretsen, Comparison of Road Traffic Noise Calculation Models, Journal of Building Acoustics, Volume 3. Number ,

13 [9] M. van den Berg and E. Gerretsen, Comparison of noise calculation models, Proceedings Inter-noise 96, Liverpool, [10] Environmental noise descriptors in Europe comparison of definitions and prediction methods, TNO report TPD-HAG-RPT , July 1996, Delft. [11] J.J.A. van Leeuwen, Noise prediction models to determine the effect of barriers placed alongside railway lines, J. Sound and Vibr., (1996) 193(1), [12] U. Isermann, On the accuracy of the computation of aircraft noise (In Dutch), Geluid, December [13] K. Attenborough et.al. Benchmark cases for outdoor sound propagation models, JASA 97, 1985, [14] J. Kragh, News and needs in outdoor noise prediction, Proceedings Inter-noise 2001, The Hague, [15] F. de Roo, Trends in sound propagation modelling, Proceedings Inter-noise 2001, The Hague, [16] E.M. Salomons, Reduction of the performance of a noise screen due to screen-induced wind speed gradients, Numerical computation and wind tunnel experiments, JASA, vol.105, 1999, [17] K.B. Rasmussen and M. Calindo Arranz, The insertion loss of screens under the influence of wind, JASA vol.104, 1996, [18] E.M. Salomons, Computational Atmospheric Acoustics, 2001, Kluwer Academic Publishers, Dordrecht, Hingham MA. [19] INM Integrated Noise Model Version, latest versions, U.S. Department of Transportation, Federal Aviation Administration. [20] F. Besnard et.al. Validation and evolution of the road traffic noise prediction model NMPB- 96-Part 1: Comparison between calculation and measurement results, Proceedings Internoise 2001, The Hague, [21] Position paper on EU noise indicators, European Commission, Brussels, ISBN X, [22] H.M. Miedema and H. Vos, Exposure-response relationships for transportation noise, JASA vol.104(6), 1998, [23] Noise emission of road vehicles. Effect of regulations. Final report, International Institute of Noise Control Engineering (I-INCE), [24] M. Vainio, Comparison of hedonic price and contingent valuation methods in urban traffic noise context, Proceedings Inter-noise 2001, The Hague, p [25] F. Elbers, Cost-benefit calculation to prevent more than 65 db(a) for railway noise in The Netherlands, Proceedings Inter-noise 2001, The Hague, [26] A. Lenders et.al., Cost-benefit analysis of rail-noise mitigation programmes at European level: Methodological innovations from EURANO and STAIRRS, Proceedings Inter-noise 2001, The Hague, [27] Report on the EC workshop on costs and benefits, see [28] Thematic Network CALM, see [29] M.G. Dittrich and M.H.A. Janssens, Measurement procedures for determining railway noise emission as input to calculation schemes, Proceedings Inter-noise 2001, The Hague, [30] B. Plovsing, Nord2000. Comprehensive model for predicting outdoor sound propagation. Experience from validating the model., Proceedings Inter-noise 2001, The Hague, [31] [32] [33]