• Construction Noise
• Measurement
• Noise Barriers
• Noise Compatible Planning
• Regulation and Guidance
• Resources
• Tire Pavement Noise
• Traffic Noise Model
  • TNM 25th Anniversary
  • Consistency Test Suite
  • Old Versions
    • TNM Version 3.1
    • TNM Version 3.0
    • TNM 2.1
    • TNM 2.0
    • TNM 1.1
    • TNM 1.0b
    • TNM 1.0a
    • TNM 1.0
  • TNM 2.5
  • TNM Version 3.2
• Training

Validation Report

Contents

• List of Figures
• List of Tables
• Abbreviations
• Executive Summary
• Section 1 Introduction
  • 1.1 Background
  • 1.2 Programmatic Differences between TNM 2.5 and TNM 3.0
  • 1.3 Objectives
  • 1.4 Contents
• Section 2 Measurement Sites
  • 2.1 Site Requirements
  • 2.2 Site Selection Process
  • 2.3 Description of Measurement Sites
• Section 3 TNM Modeling of Measurement Sites
• Section 4 TNM Modeling of Consistency Test Suite Sites
• Section 5 Data Culling and Matching
• Section 6 Differences between TNM 3.0 Predictions and TNM 2.5 Predictions
  • 6.1 Predictions for Measurement Sites
  • 6.2 Predictions for Consistency Test Suite Sites
• Section 7 Differences between TNM 3.0 Predictions and Measurements
  • 7.1 Comparisons for All Data
  • 7.2 Comparisons for Sites with Similar Ground Type
  • 7.3 Comparisons for Barrier/Open Sites
  • 7.4 Comparisons for Measurements at Similar Distances
  • 7.5 Variation by Site
• Section 8 Summary and Next Steps
• References
• Appendix A: Measurement Site Descriptions
• Appendix B: Measurement Instrumentation
  • B.1 Microphone System
  • B.2 Spectrum Analyzer
  • B.3 Incident Noise Log
  • B.4 Meteorological Instrumentation
  • B.5 Traffic Analysis Instrumentation
  • B.6 Site Survey Instrumentation
  • B.7 Other Instrumentation
• Appendix C: Measurement Field Procedures
  • C.1 Surveying Measurement Sites
  • C.2 Measurement System Setup
  • C.3 Measurements
  • C.4 Measurement System Dismantling
• Appendix D: Measurement Data Post-Processing
• Appendix E: Instrumentation System Reference
  • E.1 Instrumentation List
  • E.2 Configuration
  • E.3 Operation
  • E.4 System Performance Limits
  • E.5 Power Requirements
• Appendix F: Sample Data Log Sheets
• Appendix G: Comparison of Modeled Results for TNM 2.5, 2.6 and 3.0
  • G.1 TNM 2.5 and TNM 3.0
  • G.2 TNM 2.6 and 3.0
• Appendix H: Comparison of Modeled and Measured Results (Not Adjusted for Reference Microphone)
  • H.1 TNM Predictions vs. Measured Results - All Data Analyzed
  • H.2 TNM Predictions vs. Measured Results - Ground Type
  • H.3 TNM Predictions vs. Measured Results - Barriers
  • H.4 TNM Predictions vs. Measured Results – Distance
  • H.5 Variation by Site
• Appendix I: Comparison of Modeled and Measured Results (Adjusted for Reference Microphone)
  • I.1 TNM Predictions vs. Measured Results - All Data Analyzed
  • I.2 TNM Predictions vs. Measured Results - Ground Type
  • I.3 TNM Predictions vs. Measured Results - Barriers
  • I.4 TNM Predictions vs. Measured Results - Distance
  • I.5 Variation by Site

List of Figures

• Figure 1: Comparison of Predicted Levels between TNM Version 2.5 and Version 3.0 (Measurement Data)
• Figure 2: TNM 2.5 and 3.0 Predictions and Measured Results using Average and Specific Pavements for All Data Analyzed
• Figure 3: Comparison of Predicted Levels between TNM Version 2.5 and Version 3.0 (Measurement Data)
• Figure 4: Predictions for Measurement Sites (TNM 2.5 vs. TNM 3.0) – Grouped by Microphone Location, All Sites (Measurement Data)
• Figure 5: Predictions for Measurement Sites (TNM 2.5 vs. TNM 3.0) – Grouped by Microphone Location, Site 13CA (Default Ground is Acoustically Hard, No Barrier Present) (Measurement Data)
• Figure 6: Predictions for Measurement Sites (TNM 2.5 vs. TNM 3.0) – Grouped by Microphone Location, Site 01MA (Default Ground is Acoustically Soft, No Barrier Present) (Measurement Data)
• Figure 7: Predictions for Measurement Sites (TNM 2.5 vs. TNM 3.0) – Grouped by Microphone Location, Site 08CA (Ground Type is Acoustically Soft, Barrier Present) (Measurement Data)
• Figure 8: Predictions for Measurement Sites (TNM 2.6 vs. TNM 3.0) – Grouped by Microphone Location, All Sites (Measurement Data)
• Figure 9: Comparison of Predicted Levels between TNM Version 2.6 and Version 3.0 (Measurement Data)
• Figure 10: Comparison of Predicted Levels between TNM Version 2.5 and Version 3.0 (CTS Data)
• Figure 11: Comparison of Predicted Levels between TNM Version 2.6 and Version 3.0 (CTS Data)
• Figure 12: Comparison of Predicted Levels between TNM Version 2.6 and Version 3.0 for 3.5.3 (CTS Data)
• Figure 13: TNM 2.5 and 3.0 Predictions and Measured Results using Average and Specific Pavements for All Data Analyzed
• Figure 14. Confidence Intervals for Slope and Average Error Statistics - All Data Analyzed
• Figure 15: Comparison of TNM 2.5 and TNM 3.0 Predictions and Measured Results for Sites with Acoustically Soft Ground
• Figure 16. Confidence Intervals on Slope and Average Error Statistics - Acoustically Soft Ground
• Figure 17: Comparison of TNM 2.5 and TNM 3.0 Predictions and Measured Results for Sites with Acoustically Hard Ground
• Figure 18. Confidence Intervals on Slope and Average Error Statistics - Acoustically Hard Ground
• Figure 19: Comparison of TNM 2.5 and TNM 3.0 Predictions and Measured Results Sites without Barriers
• Figure 20. Confidence Intervals on Slope and Average Error Statistics - Sites without Barriers
• Figure 21: Comparison of TNM 2.5 and TNM 3.0 Predictions and Measured Results for Sites with Barriers
• Figure 22. Confidence Intervals on Slope and Average Error Statistics - Sites with Barriers
• Figure 23: Comparison of TNM 2.5 and TNM 3.0 Predictions and Measured Results for Measurement Locations within 125 Feet of the Center of the Nearest Lane
• Figure 24. Confidence Intervals on Slope and Average Error Statistics - Locations within 125 Feet of the Center of the Nearest Lane
• Figure 25: Comparison of TNM 2.5 and TNM 3.0 Predictions and Measured Results for Measurement Locations between 125 and 500 Feet of the Center of the Nearest Lane
• Figure 26. Confidence Intervals on Slope and Average Error Statistics - Locations between 125 and 500 Feet of the Center of the Nearest Lane
• Figure 27: Comparison of TNM 2.5 and 3.0 Predictions and Measured Results for Measurement Locations Greater than 500 Feet from the Center of the Nearest Lane
• Figure 28. Confidence Intervals on Slope and Average Error Statistics - Locations greater than 500 Feet of the Center of the Nearest Lane
• Figure A.1 - A: Site 01MA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.1 - B: Site 01MA TNM Model Description, TNM Plan and Skew Views
• Figure A.2 - A: Site 02MA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.2 - B: Site 02MA TNM Model Description, TNM Plan and Skew Views
• Figure A.3 - A: Site 03MA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.3 - B: Site 03MA TNM Model Description, TNM Plan and Skew Views
• Figure A.5 - A: Site 05CA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.5 - B: Site 05CA TNM Model Description, TNM Plan and Skew Views
• Figure A.6 - A: Site 06CA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.6 - B: Site 06CA TNM Model Description, TNM Plan and Skew Views
• Figure A.8 - A: Site 08CA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.8 - B: Site 08CA TNM Model Description, TNM Plan and Skew Views
• Figure A.9 - A: Site 09CA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.9 - B: Site 09CA TNM Model Description, TNM Plan and Skew Views
• Figure A.10 - A: Site 10CA-berm Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.10 - B: Site 10CA-berm TNM Model Description, TNM Plan and Skew Views
• Figure A.10 - C: Site 10CA-open Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.10 - D: Site 10CA-open TNM Model Description, TNM Plan and Skew Views
• Figure A.11 - A: Site 11CA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.11 - B: Site 11CA TNM Model Description, TNM Plan and Skew Views
• Figure A.12 - B: Site 12CA TNM Model Description, TNM Plan and Skew Views
• Figure A.13 - A: Site 13CA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.13 - B: Site 13CA TNM Model Description, TNM Plan and Skew Views
• Figure A.14 - A: Site 14CA Description and Photograph Source: U.S. DOT, Volpe.
• Figure A.14 - B: Site 14CA TNM Model Description, TNM Plan and Skew Views
• Figure A.15 - A: Site 15CA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.15 - B: Site 15CA TNM Model Description, TNM Plan and Skew Views
• Figure A.16 - A: Site 16MA Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.16 - B: Site 16MA TNM Model Description, TNM Plan and Skew Views
• Figure A.17 - A: Site 17CT Description and Photograph, Source: U.S. DOT, Volpe
• Figure A.17 - B: Site 17CT TNM Model Description, TNM Plan and Skew Views
• Figure B - 1: Microphone System, Source: U.S. DOT, Volpe
• Figure C - 1: Example Instrumentation Setup for Open Area Site; Plan View
• Figure C - 2: Example Instrumentation Setup for Open Area Site; Profile
• Figure C - 3: Example Instrumentation Setup for Barrier Site; Profile
• Figure C - 4: Line of Microphones and Meteorological Systems at Site 12CA, Source: U.S. DOT, Volpe
• Figure E - 1: Instrumentation Diagram for LDL2900 System
• Figure E - 2: Instrumentation Diagram for LDL 820 and Sony TCD-D100 System
• Figure E - 3: B&K Deltatron Microphone System
• Figure F - 1: Blank Site Checklist
• Figure F - 2: LDL 2900 Spectrum Analyzer System Log
• Figure F - 3: LDL 820 Sound Level Meter and Sony TCD-D100 DAT Recorder System Log
• Figure F - 4: General Site Log
• Figure G - 1: TNM 2.5 Predictions vs. TNM 3.0 Predictions using Average Pavement – All Data
• Figure G - 2: TNM 2.5 Predictions vs. TNM 3.0 Predictions using Specific Pavement – All Data
• Figure G - 3: TNM 2.6 Predictions vs. TNM 3.0 Predictions using Average Pavement – All Data
• Figure G - 4: TNM 2.6 Predictions vs. TNM 3.0 Predictions using Specific Pavement – All Data
• Figure H - 1: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – All Data
• Figure H - 2: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Soft Ground
• Figure H - 3: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Hard Ground
• Figure H - 4: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Sites without Barriers
• Figure H - 5: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Sites with Barriers
• Figure H - 6: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations within 125 Feet of the Center of the Nearest Lane.
• Figure H - 7: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations between 125 and 500 Feet of the Center of the Nearest Lane
• Figure H - 8: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations Greater than 500 Feet of the Center of the Nearest Lane
• Figure H - 9: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 01MA
• Figure H - 10: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 02MA
• Figure H - 11: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 03MA
• Figure H - 12: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 05CA
• Figure H - 13: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 06CA
• Figure H - 14: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 08CA
• Figure H - 15: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 09CA
• Figure H - 16: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 10CA
• Figure H - 17: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 11CA
• Figure H - 18: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 12CA
• Figure H - 19: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 13CA
• Figure H - 20: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 14CA
• Figure H - 21: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 16MA
• Figure H - 22: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 17CT
• Figure I - 1: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – All Data, adj.
• Figure I - 2: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Soft Ground, adj.
• Figure I - 3: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Hard Ground, adj.
• Figure I - 4: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Site without Barriers, adj.
• Figure I - 5: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Site with Barriers, adj.
• Figure I - 6: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations within 125 Feet of the Center of the Nearest Lane, adj.
• Figure I - 7: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations between 125 and 500 Feet of the Center of the Nearest Lane, adj.
• Figure I - 8: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations Greater than 500 Feet of the Center of the Nearest Lane, adj.
• Figure I - 9: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 01MA, adj.
• Figure I - 10: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 02MA, adj.
• Figure I - 11: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 03MA, adj.
• Figure I - 12: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 05CA, adj.
• Figure I - 13: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 06CA, adj.
• Figure I - 14: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 08CA, adj.
• Figure I - 15: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 09CA, adj.
• Figure I - 16: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 10CA, adj.
• Figure I - 17: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 11CA, adj.
• Figure I - 18: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 12CA, adj.
• Figure I - 19: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 13CA, adj.
• Figure I - 20: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 14CA, adj.
• Figure I - 21: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 16MA, adj.
• Figure I - 22: TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 17CT, adj.

List of Tables

• Table 1: 16 Measurement Sites (see Appendix A for site details).
• Table 2: Effective Flow Resistivity used for values of Noise Reduction Coefficient (NRC)
• Table 3: Constants for A-weighted sound-level emissions and 1/3rd-octave-band spectra
• Table 4: Average Differences between TNM 2.5 and TNM 3.0 Predictions for Measurement Sites – Grouped by Microphone Location and Site (Microphone Location - Near: <100 ft. horizontal, Low: <15 ft. vertical)
• Table 5: Average Differences between TNM 2.6 and TNM 3.0 Predictions for Measurement Sites – Grouped by Microphone Location an Site
• Table 6: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – All Data
• Table 7: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Soft Ground
• Table 8: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Hard Ground
• Table 9: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Sites without Barriers
• Table 10: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Sites with Barriers
• Table 11: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results for Measurement Locations within 125 Feet of the Center of the Nearest Lane
• Table 12: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results for Measurement Locations between 125 and 500 Feet of the Center of the Nearest Lane
• Table 13: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results for Measurement Locations greater than 500 Feet of the Center of the Nearest Lane
• Table 14: Meta Data and Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results
• Table H - 1: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – All Data
• Table H - 2: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Soft Ground
• Table H - 3: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Hard Ground
• Table H - 4: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Sites without Barriers
• Table H - 5: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Sites with Barriers
• Table H - 6: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations within 125 Feet of the Center of the Nearest Lane
• Table H - 7: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations between 125 and 500 Feet of the Center of the Nearest Lane
• Table H - 8: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations Greater than 500 Feet of the Center of the Nearest Lane
• Table H - 9: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 01MA
• Table H - 10: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 02MA
• Table H - 11: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 03MA
• Table H - 12: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 05CA
• Table H - 13: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 06CA
• Table H - 14: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 08CA
• Table H - 15: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 09CA
• Table H - 16: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 10CA
• Table H - 17: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 11CA
• Table H - 18: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 12CA
• Table H - 19: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 13CA
• Table H - 20: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 14CA
• Table H - 21: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 16MA
• Table H - 22: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 17CT
• Table I - 1: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – All Data, adj.
• Table I - 2: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Soft Ground, adj.
• Table I - 3: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Acoustically Hard Ground, adj.
• Table I - 4: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Site without Barriers, adj.
• Table I - 5: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Site with Barriers, adj.
• Table I - 6: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations within 125 Feet of the Center of the Nearest Lane, adj.
• Table I - 7: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations between 125 and 500 Feet of the Center of the Nearest Lane, adj.
• Table I - 8: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – Measurement Locations Greater than 500 Feet of the Center of the Nearest Lane, adj.
• Table I - 9: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 01MA, adj.
• Table I - 10: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 02MA, adj.
• Table I - 11: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 03MA, adj.
• Table I - 12: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 05CA, adj.
• Table I - 13: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 06CA, adj.
• Table I - 14: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 08CA, adj.
• Table I - 15: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 09CA, adj.
• Table I - 16: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 10CA, adj.
• Table I - 17: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 11CA, adj.
• Table I - 18: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 12CA, adj.
• Table I - 19: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 13CA, adj.
• Table I - 20: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 14CA, adj.
• Table I - 21: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 16MA, adj.
• Table I - 22: Summary Statistics for TNM 2.5 and 3.0 Predictions vs. Measured Results using Average and Specific Pavements – 17CT, adj.

ABBREVIATIONS

Executive Summary

The Federal Highway Administration (FHWA), conducted a validation study to quantify the agreement of FHWA's Traffic Noise Model Version 3.0 (TNM 3.0) with measured highway noise data and also with predictions from TNM 2.5.

The FHWA Traffic Noise Model (TNM), first released in 1998, has been a valuable tool for highway traffic noise analysis and barrier design. Given the utility and impact of TNM, it is necessary to validate the sound level predictions to ensure accuracy. In 2002, the FHWA completed a comprehensive Validation of TNM 2.5. This validation included collection and analysis of traffic noise data at seventeen highway sites around the country (the validation dataset). These sites had characteristics of those most commonly modeled by TNM users: open areas next to the highway with acoustically soft ground (e.g., lawn); open areas with acoustically hard ground (e.g., pavement or water); and areas next to the highway with an open area behind a single noise barrier.

The most recent version of TNM, version 3.0, includes improved acoustical computations as well as minor changes and updates. These improvements include:

• Bug Fixes
  • One affecting ground impedance averaging (Bug 1)
  • One affecting the selection of highest path points (Bug 2)
• The removal of interpolation/extrapolation at lowest and highest one-third octave bands
• The replacement of horizontal divergence with combined horizontal and vertical divergence
• A change in the manner in which elemental triangles are determined
• Updated computations for Day-night equivalent Level (Ldn) and Day-evening-night equivalent level (Lden)
• Added percentile level computations, L10 and L50
• Changes to vehicle speed computations to make them more robust
• The correction of a few coefficients in the reference emission level (REMEL) database
• The standardization of effective flow resistivity (EFR) and noise reduction coefficient (NRC) values

Because of these changes, it is advisable once again to validate the model against the measured validation dataset in order to ascertain the model's accuracy in predicting practical site conditions. This report contains the results of this validation study; leveraging the original validation dataset, to compare and contrast measured data against TNM 3.0 predictions, providing continuity and consistency of analyses. Following the methodology of the TNM 2.5 validation study, TNM 3.0 predictions were generated for all 5-minute time blocks using input files from the TNM 2.5 studies, exported to TNM 3.0 compatible format. Predictions were first generated for all 5-minute time blocks using Average pavement. As an additional analysis set, predictions were also generated with the TNM specific pavement (Dense Graded Asphalt Concrete (DGAC), Open Graded Asphalt Concrete (OGAC), or Portland Cement Concrete (PCC)) that best matched the site. Although other factors are confounded with pavement type, comparing data modeled with specific pavement type against measured data can provide additional insight by accounting for one known deviation between sites as modeled and actual measurements.

Changes in predictions between TNM 2.5 and TNM 3.0 were first examined, as it is important to understand how the model improvements affect results under a variety of conditions, as well as providing historical context. A direct comparison of levels predicted by TNM 2.5 and TNM 3.0 for the measured validation dataset is shown in Figure 1. This figure shows that TNM 3.0 predicts values about 1 dB lower than TNM 2.5. In the figure, the colored circles represent individual 5-minute model computations (color coding is given in the legend); the blue dashed line shows a first-order linear regression between the two datasets; the blue dotted lines indicate the 95-percent prediction interval for any new computations; and the solid black line indicates where all results would fall if both models gave the same predictions for all analyses. Although the statistical parameters (noted in the top left corner of the graphic) indicate good average agreement, individual predictions are somewhat varied between models, since 95-percent of the variation falls within a +/-2.78 dB range.

The variation is due to the combined effects of all changes between TNM 2.5 and 3.0. Further exploration was completed to understand the major contributors to the variation. An analysis of predictions found that the largest differences occurred at sites with multiple ground types, especially at larger source-to-receiver distances, while smaller differences occur at sites where multiple ground impedances are not part of the computations (sites with a single ground type). As a result, it was concluded that the first bug fix accounts for a significant portion of the observed differences between TNM 2.5 and 3.0 for the measured validation dataset.

In addition to reanalyzing the original measured validation dataset, sites that are included in the Consistency Test Suite (CTS) were also used to explore changes in predictions between TNM 2.5 and TNM 3.0. The CTS sites have a large number of receivers and can show how differences in TNM versions change more continuously as a function of distance or location. These sites also provide a greater diversity of acoustically significant structures and traffic flow. These models, however, do not include measurement data. For these model sites, it was found that on average TNM 3.0 predicts values about 2 dB lower than TNM 2.5. Although the statistical parameters indicate good average agreement, individual predictions are somewhat inconsistent between models, since 95-percent of the variation falls within +/- 5.33 dB range. This larger prediction interval is not surprising, as the CTS sites incorporate a greater number TNM's features, and thus a greater number of improvements between TNM 2.5 and TNM 3.0 can potentially affect results.

A research version of TNM (TNM 2.6), which included several bug fixes including the fix to the ground impedance averaging bug but not the remainder of the TNM 3.0 improvements, was utilized to identify causes of variation outside of the bug fixes. The average Root Mean Squared (RMS) difference between TNM 2.6 and 3.0 is small, 0.43 dB, compared to 1.42 dB between TNM 2.5 and 3.0, so, from a sound pressure level perspective, 70% of the variation is due to bug fixes (especially the fix to the ground impedance averaging). Further exploration of the data found that much of the remaining variation was the result of receivers at somewhat extreme locations: 1) large heights (up to 60 ft above ground), 2) large distances (up to 1600 ft from the roadway), and 3) positions near the end of the roadway oriented in a line perpendicular to the roadway. Computations for these receivers were affected by the following TNM 3.0 improvements:

• The replacement of horizontal divergence with combined horizontal and vertical divergence,
• The removal of interpolation/extrapolation at lowest and highest one-third octave bands, and
• A change in the manner in which elemental triangles are determined.

Finally, comparison of the measured validation dataset and model predictions was performed. The initial comparison, shown in Figure 2, includes all data in aggregate and was performed for four model variations: 1) TNM 2.5 with Average pavement (top left), 2) TNM 3.0 with Average pavement (bottom left) 3) TNM 2.5 with specific pavement (top right) and 4) TNM 3.0 with specific pavement (bottom right). The formatting is similar to Figure 1, except in this presentation the levels on the x-axis correspond to measured data instead of a second set of modeled data.

Figure 2 shows that TNM 2.5 (top two graphics) on average over predicts the measured data while TNM 3.0 (bottom two graphics) under predicts the measured data. This trend is consistent over the range of measured sound levels; the offset between the solid black line and dashed regression line is nearly constant. Modeling with specific pavement types does not change the overall picture greatly; the same general trends are visible in both sets of results. Modeled with average pavement, TNM 2.5 on average over predicts these data by 0.53 dB, while TNM 3.0 under predicts these data by -0.53 dB. Modeled with specific pavements, the average error is shifted upward to 0.69 for TNM 2.5 and -0.36 for TNM 3.0; a result of modeling 5 of the 14 sites with PCC pavement. Because PCC typically has much higher sound pressure levels than Average pavement, while DGAC has only slightly lower sound pressure levels than Average pavement for the same traffic, one would expect an upward shift in these results modeled with specific pavements.

To determine if TNM 3.0 is performing better or worse for a specific type of site, comparisons were also performed for several sub-sets of the data: sites with acoustically hard or soft ground, sites with or without a barrier, receivers at near, medium and far distances, and on a site-by-site basis. These results are summarized as follows:

• Sites with acoustically soft ground: TNM 3.0 and TNM 2.5 have similar magnitudes in average prediction errors. TNM 3.0 on average under predicts soft ground by 0.81 dB with Average pavement and by 0.47 dB with specific pavements. TNM 2.5 on average over predicts soft ground by 0.23 dB with Average pavement and by 0.56 dB with specific pavements.
• Sites with acoustically hard ground: TNM 3.0 has smaller magnitudes in average prediction errors. TNM 3.0 on average over predicts hard ground by 1.11 dB with Average pavement and by 0.25 dB with specific pavements. TNM 2.5 on average over predicts hard ground by 2.29 dB with Average pavement and by 1.4 dB with specific pavements.
• Sites without a barrier (Open sites): TNM 3.0 has smaller magnitudes in average prediction errors. TNM 3.0 on average over predicts open sites by 0.4 dB with Average pavement and under predicts by 0.11 dB with specific pavements. TNM 2.5 on average over predicts open sites by 1.98 dB with Average pavement and by 1.46 dB with specific pavements.
• Sites with a barrier: TNM 3.0 has larger magnitudes in average prediction errors. TNM 3.0 on average under predicts sites with barriers by 1.15 dB with Average pavement and by 0.53 dB with specific pavements. TNM 2.5 on average under predicts sites with barriers by 0.44 dB with Average pavement and over predicts by 0.15 dB with specific pavements.
• Comparisons for Measurements at Similar Distances: Distance effects are especially difficult to analyze in the aggregate because all possible confounding factors tend to show up in each distance category. For example, data for acoustically hard and soft ground, and for sites with and without barriers are present for receivers at most distances. Thus, interactions between distance and ground type, barrier presence, and pavement type may be present at all distances. Even so, it is still useful to consider how each model is performing at various distances.
• Distances less than 125 feet: TNM 3.0 has smaller magnitudes in average prediction errors. TNM 3.0 on average under predicts near distances by 0.17 dB with Average pavement and over predicts by 0.19 dB with specific pavements. TNM 2.5 on average over predicts near distances by 0.78 dB with Average pavement and by 1.15 dB with specific pavements.
• Distances between 125 and 500 feet: TNM 3.0 has larger magnitudes in average prediction errors. TNM 3.0 on average under predicts middle distances by 1.21 dB with Average pavement and over predicts by 1.2 dB with specific pavements. TNM 2.5 on average over predicts middle distances by 0.18 dB with Average pavement and by 0.12 dB with specific pavements.
• Distances greater than 500 feet: TNM 3.0 has larger magnitudes in average prediction errors. TNM 3.0 on average under predicts far distances by 0.78 dB with Average pavement and by 1.37 dB with specific pavements. TNM 2.5 on average over predicts middle distances by 0.03 dB with Average pavement and under predicts by 0.55 dB with specific pavements.
• Site-by-site basis: In general, the trends for individual sites mirror trends for the groupings noted above. TNM 2.5 tends to over predict measured data while TNM 3.0 tends to under predict measured data; modeling with specific pavements increases overall levels when sites include PCC pavements.

Overall, when predictions were compared to measured data, both TNM 2.5 and TNM 3.0 performed similarly, but not identically. In general, the correlation and variance were not significantly different; however, biases as measured by average errors showed that TNM 2.5 tends to over predict measured data by about 0.5 dB while TNM 3.0 tends to under predict measured data by about 0.5 dB. The differences between the predicted results (TNM 2.5 to TNM 3.0) were generally found to be the result of a correction to the computation of Fresnel ellipses used to account for changes in ground impedance between source and receiver. This primarily affected sites with acoustically soft ground. It should be noted that the Fresnel ellipse correction may also affect the pre-computed sub-source corrections used to project the reference sound level (REMELs) data at 50 feet back to the source; these were derived during TNM's original development. This may explain why TNM 3.0 is predicting about 1 dB lower than TNM 2.5 on average for the measured validation dataset. It is recommended that the sub-source corrections be recomputed using TNM 3.0 and the datasets in this report be compared again to document any improvements obtained.