This information has been reproduced from Rochat (2002). The acoustical data, meteorological data, and incident log data were merged into a spreadsheet file using an in-house program called tnmval.exe. In addition to the three data files, an input file with meta data is also required to run tnmval.exe. A 5-minute averaging period was used for the data output. In addition to sound pressure level, average wind speed and direction, and average temperature, 4 different qualifiers were attached to each data block in the output:
The wind qualifications were specified according to current ANSI specifications [ANSI 1998]. The 5-minute data blocks provided a short enough time interval to expose contamination and to adequately represent the wind conditions. All data blocks that had any contamination due to BAD incident noise, were overloaded, or had VERY WINDY conditions were discarded.
The site survey data were processed in order to obtain a three-dimensional map of a measurement site. Key features in the dGPS files were identified and extracted. Video traffic data were analyzed in 5-minute blocks that coincided with the 5-minute acoustic data blocks. Traffic counts, vehicle categories, volumes and speeds were determined for each lane. These data were then used to build the site models that would correspond to each 5-minute block of measured acoustic data.