Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
7-1 CHAPTER 7. CONCLUSIONS This Final Report for ACRP 02-52 represents the efforts of the Research Team investigating how to improve the modeling of aircraft noise propagation over mixed or hard impedance surfaces in the AEDT noise model. The deliverables for the ACRP 02-52 project included in this report: 1. A technical report documenting the entire research effort, including the research methodology, results, and a prioritized list of additional related research needs. 2. Detailed documentation of model improvements to address impedance variability of ground surfaces to improve the noise prediction accuracy of AEDT. 3. A supplement to the AEDT User Guide to help practitioners incorporate the effect of impedance variability of ground surfaces. The first item is this report. The second is contained in Appendix F. The supplement to the AEDT User Guide can be found in Appendix G. As laid out in this report, the chosen method to calculate lateral attenuation was the use of straight ray theory. The justification for this selection lay in the numerous studies showing it a valid model for propagation in isotropic atmospheres, its adaptation in current noise models, and the fact that it has already been used in the development of AEDT. The best choice for modeling of the impedance of the ground in AEDT is to utilize the one-parameter method. The information available to AEDT users on the characteristics of ground surfaces is limited. A pre-defined selection of surface characteristics based on ground cover is the most practical option to enable users to have good choices without undue burden. The National Land Cover Database (NLCD) provides the ground cover categories over the entire continental United States. The estimated flow resistivities of the land cover classifications presented here would benefit from a measurement program to validate the flow resistivity estimates. As the information stands, using the NLCD estimates with the validation data gathered from airport noise monitoring systems resulted in average estimates of the lateral attenuation to within 1 dB of the measured values. The resolution of the ground coverâs characteristics between and around the source and receiver needs to be defined accurately. Particularly important is to not interpolate flow resistivities between grid points and the definition of boundaries between land and water. The first point is necessary because interpolation between two flow resistivities may introduce a value that does not exist, as in the case at a land-water boundary. The second point is important in ensuring that the weighting of the different flow resistivities in finding the average is performed accurately. If a grid point with an estimate of the flow resistivity of water is near the shoreline of a river, then the area associated with that value may extend on to land. Because the potential for differences in flow resistivity is greatest between water and any other surface, careful attention should be made to how accurately that boundary is represented in the ground coverâs flow resistivity representation. Evaluating the difference between the average flow resistivity calculated in a Fresnel zone ellipse and the average flow resistivity along the major axis of the ellipse that connects source and receiver resulted in a difference of approximately 10% for most frequency bands. The advantage of the solution for the ellipse boundaries in real world coordinates is that the average flow resistivity inside the ellipse based on the intersection of the areas of the grid cells is an exact solution; whereas, the accuracy of the interpolation of the average flow resistivity along a profile cut depends upon grid resolution. In order to implement variable ground impedance modeling in AEDT, one must reduce the segment lengths to be commensurate with the variation in ground surface characteristics. As detailed in the section of the report on Integrated Modeling Issues (Section 6.5), there is an uncertainty associated
7-2 with applying what is essentially a point-to-point theory to an integrated noise model. The flight track segment size should not be made smaller than the level associated with an acceptable uncertainty. The lateral attenuation applied to the new segments should be the maximum lateral attenuation from either the point of closest approach or the segment end points to the receiver. The study found little error in applying the lateral attenuation for three points on a line segment (both ends and the closest point of approach) for segments up to 2,000 feet in length. In conjunction with the above, measurements around multiple transitions of surface types with in situ measurements of the surface impedances are needed. While the application of Fresnel zones in the Transportation Noise Model do indicate that the use of straight ray theory with the one-parameter model of ground impedance accurately predicts lateral attenuation across mixed-impedance surfaces, it would be prudent to further the validation with aircraft noise. In closing, this research concluded that: ï· Using the single-parameter model of ground impedance based upon flow resistivity estimates of categories in the National Land Cover Database along with straight ray theory as presented here accurately calculated the average lateral attenuation of aircraft operations to within 1 dB of measurements. ï· The methods presented here can be incorporated into AEDTâs architecture with minimal changes. In addition to decreasing segment lengths of input tracks and importing the additional terrain properties necessary for estimating ground impedance, one can make use of the existing framework of AEDT to add this capability. ï· The AEDT user should easily be able to see the changes in AEDT noise estimates with and without the application of this method. While the presentation here relies on comparison of theory with measurements, once implemented in AEDT, the effect that applying the correction to sound propagation over non-soft ground should be evident in the change of contour levels and location. For airport operators knowledgeable about deficiencies in noise estimates when compared to measurements of noise propagating over water or mixed ground types, a comparison of noise results when applying this theory versus not applying it in AEDT may help operators explain the discrepancies.