How Weather Affects the Noise You Hear from Highways (2018)

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.

58 The state of the practice in the United States is to model sound levels, evaluate noise impacts, and design noise abatement measures under neutral meteorological conditions (no wind or temperature effects). However, sound levels and the number and locations of impacted land uses could change under different meteorological conditions. The research team completed an analysis of how noise impact and noise abatement conclusions might change if highway noise studies considered meteorological effects. Analysis Sites The two analysis sites include the No-Barrier site with an assumed medium-density residential community (Site #1) and the Barrier site (Site #2), as shown in Figure 44. The medium-density community at Site #1 included four rows of single-family residences with 10 residences in each row, totaling 40 residences. The neighborhood is 180 meters deep. The near property line of the first-row residences is 40 meters from the center of the near lane of I-17, making the rear of the community 220 meters from the center of the near lane of I-17. Lots are approximately 24 meters by 41 meters for a total lot area of 0.25 acres. The TNM receivers are approximately 3 meters from each residence in the yard closest to the highway (i.e., the modeled location for the second- and fourth-row residences is in the front yard). The first-row receivers are approximately 43 meters from the center of the near lane of I-17 and the farthest receivers are approximately 180 meters from the center of the near lane of I-17. I-17 is on a slight fill through this area. The high-density residential community at Site #2 has an existing barrier. Lots at this site are approximately 17 meters by 34 meters with total lot sizes of approximately 0.14 acres. The TNM model includes receivers representing the back yards of the residences for the first full block of the community. The first-row receivers are approximately 61 meters from the center of the near lane of I-17 and the farthest receivers are approximately 214 meters from the center of the near lane of I-17. I-17 is at-grade through this area. The TNM models include terrain lines, building rows, and barriers, as appropriate. The porosity of the ground is estimated by the effective flow resistivity (EFR). Pavements, like con- crete, have high flow resistivity on the order of 20,000 cgs Rayls, while porous soils have low resistivity on the order of 100 cgs Rayls. In this case, the modeling uses 500 cgs Rayls for the soil, which is slightly less porous than for lawn (EFR of 300 cgs Rayls). The sound level differ- ences resulting from these EFRs are small, so the results would be essentially the same for lawn sites. Average pavement type was modeled per the FHWA noise regulation requirements for noise studies. C H A P T E R 4 Implications for Noise Impact and Abatement

Implications for Noise Impact and Abatement 59 The research team evaluated the measurement data at the reference microphones to identify the worst noise hour. Traffic on I-17 during the worst noise hour is approximately 7,000 vehicles per hour (vph), including 4% medium trucks and buses and 5% heavy trucks. Travel speeds are in the range of 65 mph to 82 mph. The TNM models include the traffic and speed data for the worst noise hour. Meteorological Conditions The analysis evaluated the base condition (neutral) and the four wind conditions specified in NCHRP Report 791 summarized in Table 18, and the four temperature profile classes summarized in Table 19. Tables 20 and 21 show the average sound level changes compared to neutral conditions with and without a barrier based on the measurement data discussed in Chapter 2. As discussed previously, strong downwind conditions did not exist during the measurements. Therefore, the evaluation only includes the moderate downwind condition. Source: Google Earth Site #1 (Medium-Density Residential) Site #2 (High-Density Residential) Figure 44. Analysis sites. WIND CONDITION WIND SPEED VARIABLE CLASS Strong upwind -5 m/s -2 Moderate upwind -2.5 m/s -1 Neutral 0 m/s 0 Moderate downwind +2.5 m/s +1 Strong downwind +5 m/s +2 Source: RSG for NCHRP Project 25-52. Table 18. NCHRP Report 791 wind classes.

60 How Weather Affects the Noise You Hear from Highways Table 19. NCHRP Report 791 temperature profile classes. Source: RSG for NCHRP Project 25-52. TEMPERATURE CONDITION WIND SPEED VARIABLE CLASS Strong lapse +0.3 °C/m -2 Weak lapse +0.1 °C/m -1 Neutral 0 °C/m 0 Weak inversion -0.1 °C/m +1 Strong inversion -0.5 °C/m +2 Distance (m) Sound Level Difference (dB) Wind Condition Temperature Condition Moderate Upwind Strong Upwind Moderate Downwind Strong Downwind Weak Lapse Strong Lapse Weak Inversion Strong Inversion 15 0 1 0 – 0 0 0 0 30 -1 0 1 – -1 -2 1 0 60 -2 -2 1 – -1 -2 1 1 120 -3 -3 2 – -2 -3 2 2 240 -5 -3 4 – -4 -5 4 3 480 -7 -3 6 – -5 -6 5 2 960 -8 -3 5 – -5 -6 3 0 Note: dashes indicate no data Source: RSG for NCHRP Project 25-52. Table 20. Sound level differences relative to neutral conditions based on measurement data, No Barrier. Distance (m) Sound Level Difference (dB) Wind Condition Temperature Condition Moderate Upwind Strong Upwind Moderate Downwind Strong Downwind Weak Lapse Strong Lapse Weak Inversion Strong Inversion 15 0 1 0 – 0 -1 0 0 30 0 1 0 – 0 -1 0 0 60 0 0 0 – -1 -2 0 0 120 -1 0 0 – -1 -1 0 1 150 -1 0 -1 – -1 -3 0 0 Note: dashes indicate no data. Source: RSG for NCHRP Project 25-52. Table 21. Sound level differences relative to neutral conditions based on measurement data, with Barrier.

Implications for Noise Impact and Abatement 61 Noise Impact Results The research team evaluated how noise impacts and noise abatement conclusions might change under different meteorological conditions. This required the following inputs from the user: • Impact and noise barrier feasibility and reasonableness information from an SHA noise policy. • Appropriate adjustment factors for each meteorological condition for No-Barrier and Barrier scenarios. • TNM results for a barrier design. • Distances between each TNM receiver and the study roadway. In the analysis, the TNM-predicted sound levels are adjusted for each site using the values in Tables 20 and 21. The values for modeled locations are interpolated between measurement points. For highway noise studies, the distance to modeled receivers would not normally exceed approximately 180 meters except for projects on new alignments. As shown in Table 20, moder- ate and strong upwind conditions and weak and strong lapse conditions reduce sound levels by up to 4 dB within this distance compared to neutral conditions. Conversely, moderate downwind, weak inversion, and strong inversion conditions increase sound levels up to 3 dB. The number of impacts and feasibility and reasonableness results for each meteorological condition were then calculated. Sound levels and noise reductions are not rounded; rounding could affect the results. This analysis procedure was incorporated in a spreadsheet tool used internally by the research team. Per the FHWA noise regulation, Category B residences are affected when worst-hour sound levels approach or exceed 67 dBA. Most SHAs define “approach” as one decibel below the regu- lation’s Noise Abatement Criteria (NAC), or 66 dBA. Therefore, the study used 66 dBA as the threshold for noise impact. Also, per the FHWA noise regulation, impacts also occur if the project substantially increases existing sound levels. Increases in sound levels due to changes in meteorological conditions do not constitute an impact under the FHWA noise regulation. The study evaluated existing conditions, so increases in sound levels due to a project do not apply. However, practitioners could use adjust- ments to estimate how sound levels might change at certain locations under different conditions. If noise studies did consider meteorological effects in determining impacts, then the implica- tions would be different for widening and new alignment projects. For widening projects, if the existing worst noise hour occurred under downwind conditions, then the future worst noise hour would also occur under downwind conditions. The sound level increases would not change since the same adjustments would apply to both the existing and future sound levels. However, there may be no significant existing noise source for new alignment projects. In this case, the increases from the existing worst noise hour to the future worst noise hour under downwind conditions could be high and impact distances could extend much farther than under neutral conditions. Tables 22 and 23 and Figures 45 and 46 summarize the number of impacts for Sites #1 and #2 under neutral conditions and the seven different meteorological conditions. Site #1 As indicated in Table 22, impacts occur at 20 residences under neutral conditions. These include the first- and second-row residences shown in Figure 44. Moderate and strong upwind conditions reduce sound levels at the second-row residences from 66–67 dBA to 63–64 dBA. These reduced second-row sound levels are below the NAC,

62 How Weather Affects the Noise You Hear from Highways Condition Number of Impacts Change from Neutral Neutral 20 – Moderate upwind 10 -10 Strong upwind 10 -10 Moderate downwind 20 0 Strong downwind – – Weak lapse 10 -10 Strong lapse 10 -10 Weak inversion 20 0 Strong inversion 20 0 Note: dashes indicate no data. Source: Bowlby & Associates for NCHRP Project 25-52. Table 22. Noise impact results, Site #1. Condition Number of Impacts Change from Neutral Neutral 30 – Moderate upwind 22 -8 Strong upwind 24 -6 Moderate downwind 43 13 Strong downwind – – Weak lapse 25 -5 Strong lapse 22 -8 Weak inversion 40 10 Strong inversion 40 10 Note: dashes indicate no data Source: Bowlby & Associates for NCHRP Project 25-52. Table 23. Noise impact results, Site #2. resulting in 10 fewer impacts. Weak and strong lapse conditions also reduce the number of impacts from 20 to 10. The sound level adjustment for the second-row residences is about 2 dB, which reduces the sound levels from 66 to 67 dBA to below 65 dBA. The number of impacts does not change under moderate downwind and weak and strong inversion conditions. These conditions increase sound levels at first- and second-row residences, but these residences are already impacted. The increases for the third-row residences are 3 dB for moderate downwind conditions and 2 dB for inversion conditions, but the resulting sound levels are still below 66 dBA. Impacts beyond the second row of houses at Site #1 are not predicted under any condition. Site #2 As indicated in Table 23, impacts occur at 30 residences under neutral conditions. These residences include the third- and fourth-row of residences shown in Figure 46. Moderate and

Implications for Noise Impact and Abatement 63 Source: Bowlby & Associates for NCHRP Project 25-52. Figure 45. Noise impact results, Site #1. Source: Bowlby & Associates for NCHRP Project 25-52. Figure 46. Noise impact results, Site #2.

64 How Weather Affects the Noise You Hear from Highways strong upwind conditions reduce sound levels at the impacted residences by approximately 2 dB, although the sound levels for many of the impacted residences remain above 66 dBA. The number of impacts is reduced from 30 to 22 and 24 for moderate and strong upwind conditions, respectively. The sound level adjustments for weak and strong lapse conditions are comparable to moderate and strong upwind conditions, so the number of impacts are comparable. Unlike Site #1, moderate downwind and weak and strong inversion conditions increase the number of impacts from 30 to 40 due to a 2-dB adjustment for some third- and fourth-row residences that increased sound levels above 66 dBA. Impacts beyond the fourth row of houses at Site #2 are not predicted under any condition. Noise Abatement Conclusions Conclusions regarding noise barrier feasibility and reasonableness could also change under different meteorological conditions. The research team evaluated how feasibility and reason- ableness conclusions would change for each site under each meteorological condition. The evaluation used the following criteria. • Feasibility: 5 dB noise reduction for at least one impacted receptor (the most commonly used criterion by SHAs). • Reasonableness: – Benefited receptor: ≥5 dB noise reduction (the most commonly used criterion by SHAs). – Noise Reduction Design Goal (NRDG): At least 7 dB at 50% or more of first-row benefited receptors. – Cost-effectiveness: An area per benefited residence (APBR) of 1,400 square feet (sf) (chosen to represent the middle range of values used by SHAs). Since the NRDG is a minimum criterion for a barrier design to be reasonable, the study identi- fied a design philosophy for the noise barriers. The philosophy was to design a noise barrier for each site to provide 7 dB noise reduction for all impacted first-row residences. Obtaining a uni- form noise reduction at all first-row benefited residences would be possible at both sites due to (1) the uniform nature of the communities; (2) the consistent geometric relationship between the road and the residences; and (3) the ability to extend the barrier along the road in both directions. The designed barrier for Site #1 is approximately 440 meters (1,444 feet) long and 2.5 meters (8.2 feet) tall. The designed barrier for Site #2 is approximately 580 meters (1,903 feet) long and 2.6 meters (8.5 feet) tall. These barriers are much lower than the barrier at the measurement site (20 feet [6.1 meters]) and the barrier evaluated for NCHRP Report 791 (17 feet [5.2 meters]). The adjustments could be different for a lower barrier, which would affect the abatement results although the extent is unknown. An evaluation of the existing barrier at Site #2 would shed additional light on this and could be done as part of future research. The TNM-predicted No-Barrier sound levels for each site were adjusted using the values in Table 20, and the sound levels with the designed barrier were adjusted using the values in Table 21. The adjustments for modeled locations between measurement points were interpolated linearly. Figures 47 and 48 summarize the resulting changes in barrier insertion loss (IL) for the modeled residences at each site for each of the meteorological conditions. The results indicate that barrier ILs are reduced under moderate upwind, strong upwind, and weak lapse conditions, and that barrier ILs are increased under moderate downwind, weak inversion, and strong inversion conditions. Barrier ILs under strong lapse conditions are close (within 1 dB) to neutral conditions.

Implications for Noise Impact and Abatement 65 Source: Bowlby & Associates for NCHRP Project 25-52. Figure 47. Changes in barrier IL from neutral, Site #1. Source: Bowlby & Associates for NCHRP Project 25-52. Figure 48. Changes in barrier IL from neutral, Site #2.

66 How Weather Affects the Noise You Hear from Highways Specifically, moderate upwind and strong upwind conditions reduce ILs by between 1 dB and 3 dB at both sites. Weak lapse conditions reduce barrier effectiveness by 1 dB to 2 dB at Site #1 and 1 dB to 3 dB at Site #2, while strong lapse conditions reduce ILs by 1 dB at Site #1 and 0 dB to 1 dB at Site #2. Moderate downwind conditions increase ILs by 1 dB to 4 dB at Site #1 and 1 dB and 5 dB at Site #2. Weak inversion conditions increase ILs by 1 dB to 3 dB at both sites. Finally, strong inversion conditions increase ILs by 1 dB to 2 dB at Site #1 and 1 dB and 3 dB at Site #2. Tables 24 and 25 summarize how these changes in barrier IL affect the noise barrier feasibility and reasonableness results for Sites #1 and #2 for the different meteorological conditions. Feasibility The feasibility criteria for this study is a 5-dB noise reduction for at least one impacted receptor. The noise barriers at each site provided a 5-dB noise reduction for at least one impacted recep- tor under all meteorological conditions. Therefore, the barriers are feasible under all conditions. Additionally, the barriers provided a 5-dB noise reduction for most of the impacted residences at both sites under all meteorological conditions. Therefore, the barriers would remain feasible with a higher threshold of 5 dB at 50% of impacted residences. Neutral Moderate Upwind Strong Upwind Moderate Downwind Strong Downwind Weak Lapse Strong Lapse Weak Inversion Strong Inversion Number of impacted residences 20 10 10 20 – 10 10 20 20 Feasibility Number of impacted residences receiving 5 dB reduction 10 10 10 19 – 10 10 16 10 Feasible? Yes Yes Yes Yes – Yes Yes Yes Yes NRDG Number of benefited residences 10 10 10 36 – 10 10 16 10 Number of first-row benefited residences 10 10 10 10 – 10 10 10 10 Number of first-row benefited residences receiving 7 dB reduction 10 0 0 10 – 0 0 10 10 NRDG 100% 0% 0% 100% – 0% 0% 100% 100% NRDG met? Yes No No Yes – No No Yes Yes Cost-Effectiveness Barrier area (sf) 11,765 n/a n/a 11,765 – n/a n/a 11,765 11,765 Area per benefited residence (APBR) (sf) 1,177 n/a n/a 327 – n/a n/a 735 1,177 Allowable APBR (sf) 1,400 n/a n/a 1,400 – n/a n/a 1,400 1,400 Reasonable? Yes n/a n/a Yes – n/a n/a Yes Yes Note: dashes indicate no data. Source: Bowlby & Associates for NCHRP Project 25-52. Table 24. Noise barrier evaluation results, Site #1.

Implications for Noise Impact and Abatement 67 Reasonableness For a noise barrier to be reasonable, the following conditions must be met: • The barrier must meet the NRDG in the SHA’s noise policy. • The barrier must be cost-effective in accordance with the SHA noise policy. • The benefited residents and property owners must support the noise barrier. This study evaluated the first two reasonableness criteria. Benefited Residences The number of benefited residences affects the NRDG and the cost-effectiveness of a barrier. This evaluation used a benefited value of 5 dB, meaning that residences that receive a noise reduction of 5 dB or more from the barrier are benefited. Table 24 and Figure 49 show the num- ber of benefits for Site #1 under the evaluated conditions. Under neutral conditions, the barrier benefits the 10 first-row residences. Moderate upwind, strong upwind, weak lapse, and strong lapse conditions reduce the ILs at the first row by about 1 dB. However, the ILs are still higher than 5 dB, so the number of benefited residences does not change. Moderate downwind conditions increase the ILs for some second-, third-, and fourth-row residences by 2 to 4 dB, bringing the ILs for many of these residences above 5 dB and increasing the number of benefited residences to 36. Neutral Moderate Upwind Strong Upwind Moderate Downwind Strong Downwind Weak Lapse Strong Lapse Weak Inversion Strong Inversion Number of impacted residences 30 22 24 43 – 25 22 40 40 Feasibility Number of impacted residences receiving 5 dB reduction 29 19 15 43 – 24 22 40 40 Feasible? Yes Yes Yes Yes – Yes Yes Yes Yes NRDG Number of benefited residences 48 21 15 111 – 28 28 111 105 Number of first-row benefited residences 11 11 11 11 – 11 11 11 11 Number of first-row benefited residences receiving 7 dB reduction 11 0 0 11 – 0 6 11 11 Noise reduction design goal 100% 0% 0% 100% – 0% 55% 100% 100% NRDG met? Yes No No Yes – No Yes Yes Yes Cost-Effectiveness Barrier area (sf) 16,038 n/a n/a 16,038 – n/a 16,038 16,038 16,038 Area per benefited residence (sf) 334 n/a n/a 144 – n/a 573 134 153 Allowable APBR (sf) 1,400 n/a n/a 1,400 – n/a 1,400 1,400 1,400 Reasonable? Yes n/a n/a Yes – n/a Yes Yes Yes Note: dashes indicate no data. Source: Bowlby & Associates for NCHRP Project 25-52. Table 25. Noise barrier evaluation results, Site #2.

68 How Weather Affects the Noise You Hear from Highways Source: Bowlby & Associates for NCHRP Project 25-52. Figure 49. Benefited residences, Site #1. Source: Bowlby & Associates for NCHRP Project 25-52. Figure 50. Benefited residences, Site #2. The ILs are increased slightly under weak and strong inversion conditions, and the ILs for some second-row residences under weak inversion conditions are increased from below 5 dB to above 5 dB, resulting in six additional benefits under weak inversion conditions and no change under strong inversion conditions. Table 25 and Figure 50 show the number of benefits for Site #2 under the evaluated condi- tions. Under neutral conditions, the barrier benefits 48 first-, second-, third- and fourth-row

Implications for Noise Impact and Abatement 69 residences. As with Site #1, moderate upwind, strong upwind, weak lapse, and strong lapse conditions reduce ILs at many residences, resulting in significant reductions in the number of benefited residences. Moderate and strong upwind conditions reduce the ILs for second-row residences by 1 to 3 dB, resulting in ILs below 5 dB for many residences. Similarly, moderate and strong upwind conditions reduce the ILs for third- and fourth-row residences by 2 to 3 dB, resulting in ILs below 5 dB for many of these residences. Thus, only 21 residences are benefited under mod- erate upwind conditions, and only 15 residences are benefited under strong upwind conditions compared to 48 residences for neutral conditions. Moderate downwind, weak inversion, and strong inversion conditions increase ILs at the resi- dences farther back in the community by 3 to 5 dB, resulting in many more benefits. As shown in Table 25, the number of benefited residences under moderate downwind, weak inversion, and strong inversion conditions are 111, 111, and 105, respectively—nearly all modeled residences are benefited. NRDG The NRDG for this study is at least 7 dB at 50% or more of first-row benefited residences. Tables 24 and 25 show the number of benefited first-row residences for each site. Site #1 includes 10 benefited first-row residences. Moderate upwind, strong upwind, weak lapse, and strong lapse conditions reduce the ILs for these first-row residences by approximately 1 dB from above 7 dB to below 7 dB. Therefore, these first-row residences are still benefited but none receive 7 dB, so the NRDG is 0% and the barriers are not reasonable. Moderate downwind, strong inversion, and weak inversion conditions increase the ILs for these first-row residences by about 1 dB, which does not change the NRDG. Site #2 includes 11 benefited first-row residences. Moderate upwind, strong upwind, and weak lapse conditions reduce the ILs for these first-row residences to below 5 dB, so there are no first- row benefited residences and the NRDG is 0%. Strong lapse conditions reduce the number of benefited first-row residences from 11 to 6, but the NRDG is 55% and above the reasonableness threshold. Moderate downwind, strong inversion, and weak inversion conditions increase the ILs for these first-row residences by 2 to 5 dB, which does not change the NRDG. Cost-Effectiveness The change in the number of benefited residences also changes the cost-effectiveness of the barrier. As stated previously, this study used a cost-effectiveness criterion of 1,400 sf per ben- efited residence. A barrier at Site #1 did not meet the NRDG for moderate upwind, strong upwind, weak lapse, and strong lapse conditions. Therefore, the designed barrier is not reasonable under those conditions and a cost-effectiveness analysis is not completed. As shown in Figure 51, for Site #1, the APBR is 1,177 sf (barrier area of 11,765 sf divided by 10 benefited residences) for neutral conditions. The number of benefited residences did not change under strong inversion condi- tions. Therefore, the APBR did not change, and the barrier remains reasonable under strong inversion. Moderate downwind conditions increased the number of benefits from 10 to 36 and thus reduced the APBR to 327 sf. Similarly, weak inversion conditions increased the number of benefits from 10 to 16 and thus reduced the APBR to 735 sf. The designed barrier is reasonable under these conditions. A barrier at Site #2 did not meet the NRDG for moderate upwind, strong upwind, and weak lapse conditions. Therefore, the designed barrier is not reasonable under those conditions

70 How Weather Affects the Noise You Hear from Highways and a cost-effectiveness analysis is not completed. As shown in Figure 52, for Site #2, the APBR is 334 sf (barrier area of 16,038 sf divided by 44 benefited residences) for neutral conditions. The number of benefited residences decreased from 48 to 28 under strong lapse conditions; however, the NRDG remained above 50% and the APBR is 573 sf so the barrier is reasonable. The number of benefited residences increased significantly under moderate downwind, weak inversion, and strong inversion conditions, which significantly reduced the APBRs to 144, 134, and 153 sf, respectively. Therefore, the barrier remains reasonable under these conditions. Comparison to NCHRP Report 791 Results As discussed in Chapter 2, there are differences between the measured adjustments and the modeled adjustments generated by Harmonoise and Nord2000 for NCHRP Report 791. More consistency was found with the NCHRP Report 791 adjustments, so the research team evaluated how applying the NCHRP Report 791 adjustments might affect noise abatement con- clusions. The NCHRP Report 791 adjustments for soft site and mixed traffic (Table 14 of NCHRP Report 791) were input into the Excel tool created for this project to generate the results. The noise barrier height evaluated for NCHRP Report 791 was 17 feet (5.2 meters), which is compa- rable to the height of the barrier at the measurement location of 20 feet (6.1 meters). However, the cross-sections and traffic are different from the NCHRP Report 791 case, which used a four- lane cross-section and a speed of 60 mph. Figure 53 shows the changes in barrier IL for Site #2 using the NCHRP Report 791 adjustments. These results indicate that barrier ILs are reduced under moderate and strong upwind conditions, which is consistent with the results using the measured adjustments shown in Figure 47. However, the IL reductions are between 0 dB and 5 dB and slightly greater than the 1 to 3 dB reductions using the measured adjustments. Similarly, the NCHRP Report 791 results indicate that moderate downwind conditions increase ILs by between 2 dB and 9 dB Source: Bowlby & Associates for NCHRP Project 25-52. N/A N/A N/A N/A 1,400 sf Figure 51. APBR, Site #1.

Implications for Noise Impact and Abatement 71 N/A N/A N/A Source: Bowlby & Associates for NCHRP Project 25-52. Figure 52. APBR, Site #2. Source: Bowlby & Associates for NCHRP Project 25-52. Figure 53. Changes in Barrier IL from neutral, Site #2 using NCHRP Report 791 adjustments. compared to between 1 dB and 5 dB using the measured adjustments. The NCHRP Report 791 results indicate that ILs change less than 1 dB under lapse conditions while the results using the measured adjustments indicate reductions of 1 to 3 dB. Under weak and strong inversion conditions, the NCHRP Report 791 results indicate that ILs decrease by up to 4 dB. These results significantly differ from the results using the measured adjustments that indicate that weak and strong inversion conditions increase ILs by between 1 dB and 3 dB.

72 How Weather Affects the Noise You Hear from Highways Frequency of Meteorological Conditions The frequency and duration of the various meteorological conditions might also be a key fac- tor to consider. Neutral conditions may rarely occur in areas where windy conditions are typical throughout the year. Upwind or downwind conditions may be prevailing at project sites in these areas. Thus, the variability and frequency of meteorological conditions is important because that information is directly related to how often land uses are affected, and the percentage of time that a noise barrier may provide the desired noise reduction. The research team obtained meteorological data for the measurement sites from the NWS’s Automated Surface Observing System (ASOS), which is widely available throughout the United States. Table 26 and Figure 54 summarize the results. As shown, downwind conditions existed approximately 35% of the time while upwind conditions existed approximately 37% of the time. There were minimal wind effects approximately 29% of the time. Note that this table only accounts for wind conditions. The temperature profile also effects the refraction of sound. The table uses a simplified stability class as a proxy. The stability class is a function of cloud cover and whether it is day or night. Stability Class Strong Downwind Moderate Downwind Zero Wind Weak Upwind Strong Upwind S1 (day) 0.5% 2.9% 2.7% 3.2% 0.4% S2 (day) 1.3% 6.9% 6.0% 7.2% 1.2% S3 (day) 1.0% 4.2% 3.6% 4.1% 0.9% S4 (night) 1.2% 6.7% 6.5% 7.3% 1.1% S5 (night) 0.8% 9.3% 10.6% 9.4% 0.7% Total 4.9% 30.1% 29.4% 31.2% 4.4% Source: RSG for NCHRP Project 25-52. Table 26. Prevailing conditions at measurement location. Source: RSG for NCHRP Project 25-52. 0% 5% 10% 15% 20% 25% 30% 35% Strong Downwind Moderate Downwind Zero Wind Weak Upwind Strong Upwind % o f  m e S5 (night) S4 (night) S3 (day) S2 (day) S1 (day) 4.9% 30.1% 29.4% 31.2% 4.4% Figure 54. Prevailing conditions at measurement location using NWS ASOS data.

Implications for Noise Impact and Abatement 73 Strong winds are not typical at the location and there is no clear prevailing condition during the day (S1, S2, and S3). Upwind, downwind, and zero-wind conditions all occur with approxi- mately the same frequency under each daytime stability class. Similarly, there is no clear prevail- ing condition during the night (S4 and S5). Upwind, downwind, and zero-wind conditions all occur with approximately the same frequency under each nighttime stability class. Sound levels and barrier effectiveness at the measurement locations fluctuate on a regular basis as conditions change. However, other locations will experience conditions unique to the area. Locations in windy areas may have a prevailing upwind or downwind condition so the sound levels and barrier performance might be more consistent, such as near a coast where a land- and sea-breeze often repeat. Summary The state of the practice in the United States is to model sound levels, evaluate noise impacts, and design noise abatement measures under neutral meteorological conditions (no wind or tem- perature effects). If highway noise studies included meteorological effects, the research results indicate that noise impacts would generally increase under downwind and inversion conditions and decrease under upwind and positive lapse conditions. The noise abatement conclusions would also change. The results indicate that the barrier would be less effective under upwind and lapse conditions and much less likely to meet the rea- sonableness criteria in the SHA noise policy. Sound levels under these conditions are also lower than for neutral conditions, so this would not be an appropriate design condition per the FHWA noise regulation. The evaluated barriers are much lower than the barrier at the measurement location and the barrier evaluated for NCHRP Report 791, which could affect the abatement results. Future research could evaluate the existing barrier at Site #2. The results also indicate that the barrier designed for the “worst noise hour” under neutral conditions is more effective under downwind and inversion conditions: ILs are higher, which increases the number of benefited residences. Therefore, the barrier is feasible and reasonable under downwind and inversion conditions. Differences exist between the adjustments measured for this research and the NCHRP Report 791 adjustments based on the Nord2000 model. Both the measured adjustments and the NCHRP Report 791 adjustments indicate that barrier ILs increase close to the barrier for downwind conditions. The most pronounced difference between the measured adjustments and the NCHRP Report 791 adjustments is for inversion conditions. The NCHRP Report 791 adjustments indicate that the barrier is less effective during inversion conditions, which is contrary to the measured adjust- ments. As discussed in Chapter 2, the likely reason that the barrier is effective under downwind conditions is that the barrier is relatively tall. At an effective height of 7.6 meters (25 feet), the sound monitors were in the acoustical shadow zone of the barrier, even out to the furthest distance of 150 meters (492 feet). Even under strong refractive conditions, the sound rays bend downward, but not enough to affect the sound levels within this distance. In addition, during the measure- ments, there may have been less turbulence than was assumed in research performed for NCHRP Report 791. Turbulence “pulls” sound down into the shadow zone and increases the sound relative to nonturbulent conditions. Highway noise barriers can also create higher wind shear on the leeward side of the barrier, creating increased downward refraction. This may not have occurred here, or the barrier may have been tall enough such that the impact was small.