Division of Information Technology, Engineering and the Environment
ISST - Transport Systems
ISST - Transport Systems
NetNoise is designed as a planning and evaluation tool specifically concentrating on the spatial distribution of traffic noise from road networks.
The program uses the CRTN (1988) procedure to calculate noise levels across a grid of receptors superimposed on the study area. This grid can be based on the network boundaries or at a user specified location. The CRTN procedure is designed to only calculate L10 one hour and 18 hour noise indices but studies by Brown (1989) have shown high correlation between L10 and Leq and this has been implemented into the program.
The program begins with a title page and then progresses to the main screen where all the calculation parameters can be set (Figure 1).
The screen layout was designed in the days when 14" monitors were predominant so if you have a larger monitor there will be space around the edge of the objects on the screen (if you use a higher resolution).
The user is guided through a calculation as controls are only enabled in the sequence necessary for that particular calculation.
The program has limited error trapping but to eliminate all possible user errors is a mammoth task so please make note of the most annoying features and let me know about them.
NetNoise was designed to be highly flexible in terms of output and input formats and at present can output files directly for use by MapInfo, ARC/INFO, Excel and Access databases. Data input for traffic volumes may be in any units and proportion of heavy vehicles can be approximated using Australian Bureau of Statistics data, commercial vehicle counts or by using the 13 NAASRA classes (now 12 AUSTROADS) classification counts.
In order to run a calculation in NetNoise, the following areas must be set up in order:
All screens will have either an accept, exit or cancel button.
The menu bar at the top of the screen operates in the same way as the tool buttons underneath it with the exception of the file option.
Selecting this option presents the user with the opportunity to exit the program, save and open *.noi files and use the file converter.
At present *.noi files just contain basic file names and paths and it is hoped that in the future all calculation parameters are saved as well. Only the main screen contains a menu bar across the top.
The file converter was written to convert files into appropriate formats after a calculation run had been completed. For example, if the user needed to view data in MapInfo but had only produced ASCII output files, instead of redoing the entire calculation the file converter could be used.
The file converter may be found under the file option in the menu bar across the top of the screen. To use the converter simply select the file formats to be converted and the files selected will be converted.
The following are used as initial settings when the user begins NetNoise:
Noise barriers may be included or omitted from the calculations. Once omitted the user has to activate the "Reset All" button to include them again. This feature is useful if no barriers exist in the network and no barrier input files have been created. If included, barriers may be toggled on and off in the propagation settings screen. Road names may also be toggled on if the user has the appropriate input file.
The screens displayed when the input button is pressed are determined by which items are checked in the input and output path areas. The user may type the file names and paths directly into the entry boxes or click the diskette to the side of the entry box thus bringing up the standard windows file selector. There is a button to make all the files on the screen become default and a clear screen and apply default button. It is after this step that files should be saved. A *.noi NetNoise file will be written containing the default names at the time the file was saved. When using NetNoise again, just open the appropriate *.noi file and the file names will be loaded as default making it unnecessary for the user to re-type the file names and paths.
The network button allows the user to view the geometry of the network and attributes relating to the network. Flows, speeds, speed limits, %heavies, gradients and road surfaces can all be displayed graphically on the network (Figure 3).
The Network Data button at the top left hand corner of the screen allows the user to view the network attributes in a tabular form. If the user is happy with the data the accept button may be selected. If the user is not happy with the data, the cancel button may be selected and the file names and paths re-entered.
Flows may be input as AADT (Annual Average Daily Traffic), 18 hour
noise, 11 and 12 hour turning count, one hour (peak) count or any user
defined units by selecting the appropriate radio
button (Figure 4).
Conversion factors will be shown on the screen depending on the noise index to be calculated. Suggested values are shown in blue text in boxes to the right of the white factor value boxes. All conversions convert the flow to either one hour or 18 hour flow units. All conversions factors are only intended as guidance and may be altered by the user.
The proportion of heavy vehicles in the traffic stream may be approximated through commercial vehicle count data, Australian Bureau of Statistics (ABS) data and NAASRA classification counts (now AUSTROADS). For the latter two cases, the user can dictate which categories or classes of vehicles can be included as heavy vehicles (Figure 5).
This is useful, for example if the user does not wish to consider utes as heavy vehicles but would like to consider motorcycles (in terms of noise output). Counts made according to the AS2702-1984 vehicle noise categories can also be applied using the ABS data shell.
Propagation considers how the noise travels from the source (ie road traffic) to the receptor. This means that features between the noise source and the receptor influence the noise level at the receptor. Such features include noise barriers, rows of houses, type of ground cover and reflective surfaces.
In this screen (Figure 6), the user has the option of turning noise barriers on and off which is useful for creating before and after scenarios (this can only be done if the barriers were included initially on the main menu form).
A feature called the Radius of Influence determines the distance from which a receptor can receive noise. This is a contentious issue and NetNoise has a limit of 400m but the user may change this according to his or her own judgement (300m is recommended). To avoid some receptors having a noise level of 0 dB(A) (ie ones further than the Radius of Influence from a road) a Background Noise Level is included. This is the minimum noise level that a receptor can have and can be set for daytime, night time or a user specified value.
Ground cover is incorporated to enable the user to approximate ground absorption in certain network scenarios. This feature should be used with caution by the user as it will apply a blanket ground cover correction across the entire network being modelled. The proportion of absorbent ground cover is based on hydrological data used for runoff calculations. The user can opt to have no correction, a user specified correction or a correction based on general land use patterns within the network.
This screen (Figure 7) allows various corrections to be toggled on and off. The correction for the facade effect (as defined by CRTN), gradients (only applies to positive gradients) and Australian adjustments (based on work by Saunders et al (1983) at ARRB) can be implemented for the calculation run.
The user has the choice of implementing either the CRTN or Samuelsâ corrections for road surface or not implementing a road surface correction at all.
The CRTN correction can only be performed if the attributes input file has texture depth data for the network. The Samuelsâ corrections are based on work carried out by Dr Stephen Samuels for the RTA (NSW) and ARRB.
A final adjustment can be made by the user to apply over the entire network based on the userâs local knowledge or extenuating circumstances.
Selecting the calculate button brings up a small box with various options in it (Figure 8). The user may view a summary of all the settings made for the current calculation run of NetNoise. Selecting the calculate button brings up a gauge which indicates how the calculation is progressing. NetNoise solves spatial information and consequently as network size and grid size increase, solving time increases by a power of 3 (x3 ).
Once the calculation has taken place the write button is enabled allowing the user to write output to different file formats if these were forgotten to be set.
For example, if a calculation run had been performed with just ASCII output files and the user wished to also obtain ARC/INFO files but had forgotten to specify them, the user can set the ARC/INFO input paths (and check the appropriate output file box), select the calculate button and then select the write button thus avoiding the need to recalculate again. If, however, the user had finished the calculation and exited NetNoise, the File Converter would need to be used instead.
The output screen can show receptor based final noise levels and link based noise levels in tabular and graphical form. The user simply selects the appropriate button at the top of the screen. Cross sections through lines of receptors can also be viewed but these do not interpolate between receptor points. At the bottom of the screen a statistics button shows various statistics about the calculation run.
The noise button brings up the network viewer screen (as seen earlier under the network button) where a map of noise levels can be viewed in addition to network information by clicking the noise button (with the ear icon on it - see Figure 9). This is useful for identifying why there may be high concentrations of noise in specific areas (such as a high percentage of heavy vehicles or a steep gradient). Obviously the more receptors there are in the grid the finer will be the detail of the mapped noise distribution.
Brown A L, 1989, 'Some Simple Transformations for Road Traffic Noise Scales', Australian Road Research, Melbourne, pp309-312 v19 n4 December.
Saunders R E, Samuels S E, Leach R, Hall A, 1983, 'An Evaluation of the UK DoE Traffic Noise Prediction Method', ARRB Research Report, Melbourne.
UK DoT (Welsh Office), 1988, 'Calculation of Road Traffic Noise (CRTN)', HMSO, London.
