Employing computerised predictive techniques
With increasing pressure to construct on brownfield sites, residential developers are having to consider constructing on sites with inherent noise issues. In order to maximise the development potential of these sites, the existing noise environment must be accurately characterised and the impact of existing and predicted future noise sources assessed.
In many instances, these predictions are most effectively undertaken using computerised predictive and mapping techniques. This process can then help inform the optimum layout and design of the development and any necessary noise-related mitigation measures. As a result, businesses are looking to engage an acoustic consultant with the necessary skills and resources to provide this package of services and prepare the appropriate noise assessment report as part of a comprehensive Environmental Impact Assessment.
Why use computer modelling?
Computerised prediction techniques may be used for various reasons, many of which are driven by the perceived limitations of direct monitoring. This article sets out to assess the benefits and pitfalls of these techniques for assessing noise levels for planning purposes.
Direct monitoring by its very nature produces a snapshot, limited by location and time. The use of numerous monitoring points over extended time periods in order to increase the accuracy of this ‘snapshot’ is not only likely to be costly but may also have logistical issues. Indeed, over the years noise consultants have discovered that the preferred location for setting up monitoring equipment may not be available for one reason or another, including concerns over leaving expensive equipment on site in vulnerable situations.
Examples of typical use of modelling
In order to demonstrate the benefits of computerised modelling techniques two ‘real life’ situations are presented below. The first deals with the impact of a major road on a proposed high-rise development. The second looks at the issues associated with moving an existing air conditioning plant adjacent to a housing scheme.
Situation 1: Assessing the impact of motorway noise on a proposed high-rise development.
A developer is considering purchasing a plot of land and demolishing an existing warehouse and constructing several high-rise units in an inner city location. The proposed development location is however less than 100m from a motorway flyover. Noise Levels were monitored at this location over several days at 1.5m above ground level in free-field conditions with an unobstructed view of the adjacent motorway in order to determine existing ambient conditions, a summary of the results is plotted in Figure 1.
The results of the monitoring exercise indicated that existing daytime noise levels would place the site in Planning Policy Guidance Note 24 (PPG24) ‘Planning and Noise’ Noise Exposure Category NEC B (Leq,16h 55 dB(A) – 63 dB(A)). PPG24 states that for NEC B:
“Noise should be taken into account when determining planning applications and, where appropriate, conditions imposed to ensure an adequate level of protection against noise.”
At the noise levels recorded at ground level the Environmental Protection Department of the local authority had indicated that they would be of a mind to recommend approval of the scheme, but had expressed concerns relative to noise levels at upper floors.
The site survey indicated that the structure of the flyover included a 2m high parapet which was actually providing screening of the traffic at ground level. It was also apparent that the screening provided by the structure would not be effective at higher levels. Given the inner city location and the lack of suitable elevated monitoring points, the most effective means of producing an indication of likely noise levels at the facades of upper floors was considered to be computerised modelling techniques. A computerised model was therefore set up to assess the likely issues.
Traffic flow data for the motorway provided the basis for assessing the noise emissions associated with the adjacent flyover. A detailed survey of the surrounding area produced height data of other significant buildings that may influence the reception of noise at the proposed site; the survey also revealed other noise sources that may affect the ambient noise level at the site. All the information was entered into the model and a predictive assessment was carried out for various heights at the façade of the nearest proposed receiver building.
The predicted noise levels based upon monitored traffic flows correlated well with monitored noise levels. The predicted noise level contour plan at ground level and at second floor level are shown in Images 1 and 2 respectively, with tabulated results for all levels summarised in Table 1.
|Façade Levels||Height (m)||Daytime Noise Levels (L eq,16h )||Night time Noise Levels (L eq,8h )|
Contours produced by the model indicated that noise levels from about level 2 and above would be considered as being in PPG24 NEC C, for which the PPG24 guidance states:
“Planning permission should not normally be granted. Where it is considered that permission should be given, for example because there are no alternative quieter sites available, conditions should be imposed to ensure a commensurate level of protection against noise.”
Whilst it may have been feasible to have convinced the Environmental Protection Department that the project should go ahead, the results of the modelling exercise raised issues relating to development zoning and the appropriate level of acoustic insulation provision within the proposed dwellings.
In practice, little could be done in terms of external screening and it was clearly not possible to construct an acoustic barrier of sufficient height to protect the upper floors. There was also no scope to incorporate a non-residential barrier block in front of the proposed dwellings. The model did indicate however that the building itself would provide significant acoustic protection for facades to the rear.
With this information it was clear that in order to maximise the development potential, consideration would have to be given to the configuration of the internal room layouts within each of the dwellings. The BRE/CIRIA Document ‘Sound Control for Homes’ provides some guidance for the internal planning of a barrier block under these circumstances. The document recommends that “Sensitive rooms should be protected from external noise by good room planning. Less noise-sensitive areas should be interposed between sensitive rooms and the noise source”.
Guidance on acceptable noise levels inside dwellings is given in BS8233 “Sound insulation and noise reduction for buildings, A Code of Practice”, which reflects current World Health Organisation guidelines, with established criteria for good and reasonable internal noise levels based upon occupancy patterns.
Situation 2: Assessment of the impact of a rooftop HVAC plant room on the surrounding residential area.
As part of a complete building refurbishment project the occupiers of an office block were considering installing a new air conditioning system. It is intended to move the plant from its current housing on the western boundary of the site and install the new HVAC plant within a rooftop plant compound.
The local authority has indicated that the noise levels associated with the new air conditioning plant should be 10 dB(A) below the existing background noise levels L90 as monitored outside the façade of the nearest dwelling. Given the type of work carried out in the offices, the plant was assumed not to operate overnight.
Direct monitoring was used to determine baseline noise levels outside the nearest dwelling. The monitoring indicated that background noise levels L90 were typically in the order of 45 dB(A) during the quieter periods over the working day.
Computerised predictive techniques were then used to determine the maximum plant noise levels permissible to ensure that the proposed noise ‘condition’ would not be infringed. Modelling techniques were also used to determine the benefits of moving the plant compound away from the nearest residential units.
Given the distance of the nearest receiver from the single HVAC unit, the noise was modelled as a point source centred at 0.5m above either ground level or roof top level. Existing buildings and fences in the vicinity were included in the model, as were the different types of terrain comprising both hard standing and some soft landscaping.
A notional maximum sound power rating of the HVAC plant (PWL) of 76.1 dB(A) was derived from the modelling to preserve the stated noise limit at the shared boundary. This rating did not take account of any possible attenuation from screens or barriers. At roof top level the maximum unprotected plant sound power rating was predicted to range from between 80 and 90 dB(A) depending on the precise location of the unit. The noise levels from the existing plant are shown in Image 3, together with the noise impact from the proposed new plant at three possible locations across the roof in Images 4 to 6.
|Road traffic noise||Railway traffic noise|
|Characteristics of primary noise source||Vehicle flows, velocities and % of heavy goods vehicles (HGV)||Rail vehicle type(s) and frequency of services|
|Secondary source features||Road gradients and surface types||Train velocities, type of track, e.g. continuously welded|
|Ground model features||Elevation of road and general topography and buildings||Track position relative to cuttings and embankments|
|Acoustic features||Acoustic barriers/ reflectors||Acoustic barriers/ reflectors|
|Noise-sensitive receivers||e.g. nearest dwelling facades||e.g. nearest office facade|
Hence, the computerised assessment permitted acoustic constraints to be set for the procurement of the plant. In this particular situation spectral characteristics of the proposed plant were also modelled to enable the designer to determine the most effective means of additional acoustic screening, should it be required.
Existing available computerised predictive techniques
Various proprietary computer-based programs exist with modules to assess the impact of road, railway and general site noise and other noise issues, for which an appropriate level of background knowledge and training is required.
For road and railway traffic noise assessment the UK’s Department of Transport published the industry-standard methodologies in Calculation of Road Traffic Noise (CRTN) and Calculation of Railway Noise (CRN). An idea of the input requirements to these models is given in Table 2.
The road traffic characterisation is typically derived from a mixture of publicly available sources and site-specific evaluation. For instance, flow data is often published in terms of Annual Average Daily Flows (AADF), which is averaged over the year but may not represent the specific design conditions. Similarly, the AADF sensor points may not relate directly to the site location being considered or the precise breakdown of vehicle types. Noise levels will vary significantly with the percentage of HGVs. Velocities can be monitored directly or default values used based upon road types defined by CRTN.
For a preliminary appraisal, the National Physics Laboratory (NPL) website http://www.npl.co.uk/acoustics/techguides can provide a rapid assessment of the possible noise impact of road traffic noise using CRTN methodology. The website does not allow the production of noise contours but does allow consideration of attenuation as a result of alterations of distances from the noise source, the influence of barriers and other means of attenuation including ground absorption. The results in accordance with CRTN are presented in terms of hourly or 18 hour L10 and not in terms of Leq.
For railway traffic, the operators’ timetables typically provide the basic scheduled train movements, although this data is unlikely to include details of the type of rail vehicle used, or the number of carriages. On occasions it is possible to obtain further data for vehicle types from websites, e.g. www.railfaneurope.net. Freight movements can also provide a significant source of noise. These may include regular overnight movements when the lines are not in use for scheduled passenger services. Unfortunately, there is often very little information in the public domain relating to these movements and a site visit will be necessary to confirm many of these details including typical velocities close to the site.
Modelling of plant noise
Where computerised modelling of plant noise is required specifically for construction sites the methodology usually follows that outlined in BS 5228 Part 1: “Noise and vibration control on construction and open sites Part 1. Code of practice for basic information and procedures for noise and vibration control”. In addition to the normal parameters used for noise modelling, plant noise levels are required to be provided either in terms of a Sound Power Level Lw dB(A) or equivalent sound pressure level Leq at a pre-defined distance. There is also a requirement to enter an indication of the proportion of operational time for each item of plant this is defined in BS 5228 as the ‘percentage on time’. BS5228 does provide a resource for generic noise levels associated with typical items plant that can be used as base data for input into computer models.
One of the most significant issues associated with modelling of plant noise is the definition of the noise source in particular whether the noise source be a ‘point’, ‘area’ or a linear source. Failure to define noise sources properly can lead to a significant under-assessment of its impact.
Pitfalls associated with modelling
As with all computerised modelling techniques the resulting information is only as good or as accurate as the information entered into it. Where a model depends upon relative heights between a noise source and a barrier or some other receiver to accurately compute noise levels, particular care should be taken in assessing these values, whether from digital Ordnance Survey or site topographic survey plans. These plans can also be used for marking up items of acoustic interested observed during site visits.
Computer modelling should wherever possible be validated against some appropriate form of on-site monitoring in order to confirm that the results produced by the program are of the right order of magnitude.
Available software programs for modelling noise environments and predicting future impacts provide acoustic specialists with the tools to help developers and businesses optimise the layout, design and construction of property developments and plant and equipment installations. This input is most valuable if sought at the earliest possible stage of the project planning and implementation cycle, when designs and procurement specifications can be influenced.
Published: 01st Sep 2007 in AWE International