applying the nchrp 10-76 methodology for …docs.trb.org/prp/15-4208.pdfapplying the nchrp 10-76...

Lodico, D, Donavan, P, and Rochat, J.

APPLYING THE NCHRP 10-76 METHODOLOGY FOR EVALUATING PAVEMENT AND BARRIERS FOR NOISE MITIGATION TO STATE PROJECT-BASED EXAMPLES Dana M. Lodico (Corresponding Author) Illingworth & Rodkin, Inc 1615 California Street, Suite 614 Denver, CO 80202 303-720-2355 [email protected] Paul R. Donavan Illingworth & Rodkin, Inc 1 Willowbrook Court, Suite 120 Petaluma, CA 94954 707-794-0400 [email protected] Judith L. Rochat, Ph.D. ATS Consulting 215 North Marengo Street, Suite 100 Pasadena, CA 91101 626-710-4400 [email protected]

Submitted to TRB on August 1, 2014, Revised on November 11, 2014 Word Count: 5,321 Figures: 2 Tables: 7

Lodico, D, and Donavan, P.

ABSTRACT In the National Cooperative Highway Research Program (NCHRP) 10-76 Project, methodologies were developed and demonstrated for evaluating feasibility, reasonableness, effectiveness, acoustic longevity, and economic features of pavement strategies and barriers proposed for highway noise abatement. The essential elements of the methodologies are the use of an expanded pavement Life Cycle Cost Analysis (LCCA) to examine economic features, a modified version of the Federal Highway Administration Traffic Noise Model (TNM) to integrate the noise reducing performance of pavements and barriers, and use of on-board sound intensity data as an input to the prediction model. In this paper, the developed methodology is applied to State project-based examples. Barriers, quieter pavement, and combinations of both are evaluated for feasibility and reasonableness using the policies of three state highway agencies.

Lodico, D, Donavan, P, and Rochat, J.

INTRODUCTION Over the past decade, consideration of quieter pavement for highway noise mitigation has been advanced by highway agencies and the public. As an abatement measure, barriers are costly to build, but require minimal maintenance and provide a fixed amount of noise reduction over a long period of time. In contrast, quieter pavements are often initially less expensive than barriers, but their noise reduction performance typically degrades with time. This requires shorter rehabilitation cycles to maintain their performance and adds to costs over the life of the project. For these reasons, quieter pavements are currently not approved by the Federal Highway Administration (FHWA) for noise abatement. In order to develop methodologies to account for the acoustic performance and life cycle costs of both types of mitigation measures used separately or in combination, the NCHRP 10-76 Project, Methodologies for Evaluating Pavement Strategies and Barriers for Noise Mitigation, was completed1. A previous paper2 describes the application of the Federal Highway Administration (FHWA) pavement Life Cycle Cost Analysis (LCCA) to barriers and quieter pavements. This current paper illustrates how the NCHRP 10-76 methodologies might be used in State-based highway projects, using the noise policies of California, North Carolina, and Arizona, using on-board sound intensity (OBSI) acoustic longevity data from California, North Carolina, and Arizona, and using pavement costs from the Washington State Department of Transportation (WSDOT).

SUMMARY OF METHODOLOGY The methodology developed in NCHRP 10-761 for evaluating pavement strategies and barriers for noise mitigation in highway projects includes a framework with which to compare costs of the two types of mitigation on an equivalent basis and a method of predicting the acoustic performance of the various pavement alternatives. With some additional considerations, this methodology can be used to analyze pavement and barrier strategies for noise abatement within the existing State policies. The primary barrier cost is the initial cost of construction. For pavements, a “quieter pavement” may have a lower initial cost, but to maintain its noise reducing performance, it may need to be rehabilitated on a shorter cycle than a pavement that is not intended to provide noise reduction. The method selected for capturing this temporal disparity is Life-Cycle Cost Analysis (LCCA) recommended by FHWA for evaluating pavement design alternatives. This analysis considers the costs of different pavement alternatives in initial construction and the costs of each rehabilitation and maintenance necessary over the life cycle of a highway project, which typically ranges from 28 to 50 years. The rehabilitation cycle is defined as the time period in which the pavement will deteriorate to the agency’s minimum acceptable condition. For quieter pavement options, the rehabilitation cycle would need to account for acoustic longevity. Depending on the pavement, the cycle may need to be shortened to assure that an acceptable level of noise reduction performance is maintained throughout the life of the pavement. This would assure that the FHWA criterion of maintaining the noise abatement “in perpetuity” is met3. Sound wall cost would include the initial cost of the barrier (added to the appropriate pavement cost), as well as maintenance costs such as repairs and graffiti removal. For the quieter pavement, cost would include the initial cost of the pavement relative to a reference pavement, the rehabilitation costs, and any ongoing monitoring of acoustic performance of the pavement to ensure that an acceptable level of noise reduction performance is maintained throughout the life


of the pavement. When comparing several noise abatement alternatives, a common life-cycle would be selected for all alternatives. To go along with the LCCA analysis, prediction of the acoustic performance of the various pavement alternatives is needed to complete the abatement analysis. Following the research conducted by the U.S. DOT Volpe Center under FHWA TNM Pavement Effects Implementation Study4, it was decided to utilize a research version of the FHWA Traffic Noise Model (TNM) which scales the ground level noise source strength (GLSS) with measured On-Board Sound Intensity (OBSI) levels. The initial scaling based on the older Goodyear Aquatred 3 test tire was updated to the newer Standard Reference Test Tire (SRTT) prescribed in the AASHTO TP-76 provisional test procedure5. Under this approach, state highway agencies (SHA) could input tire/pavement source levels into TNM based on their own pavement designs and acoustic longevity studies for use in traffic noise predictions. Noise levels with different barrier designs and combinations of pavement alternatives can be predicted and different scenarios to effect noise reduction can be evaluated. THREE STATE PROJECT BASED EXAMPLES This section applies the NCHRP 10-76 methodology to projects from three state highway agencies, California, North Carolina, and Arizona. In each case, OBSI data measured previously by the research team for pavements found in that state were used. The TNM models were obtained either from the state agency or their contractors. The pavements considered for the projects in this section are meant to be representative of state practices and may or may not have been candidates for consideration in the projects themselves. The examples discussed are meant to be illustrative of the application of the NCHRP 10-76 process and do not represent the outcomes of the actual highway noise studies, which were not conducted using the NCHRP 10-76 methodology. Example 1: I-580 Lane Addition, Segment 3 (California) This project proposed the addition of one High Occupancy Vehicle (HOV) lane in each travel direction of Interstate 580 over a 13.1 miles stretch between Dublin and Livermore, California. The total number of travel lanes would be increased from eight to ten lanes. Although the project was broken into several smaller projects for the purposes of the NEPA process, this analysis assumed that both eastbound and westbound HOV lanes were to be added as a single project. Out of the 13.1-mile project, three smaller segments were considered in the NCHRP 10-76 Project1, each with several potential barrier locations. The segment described here, Segment 3, extends from the Vasco Road overpass for about ½ mile to the east and is shown in the aerial photograph of Figure 1. Policy Summary The results of the abatement analysis for this example were compared to Caltrans policy6. Caltrans uses a threshold level for noise impact of 66 dBA as approaching the Federal Noise Abatement Criteria (NAC)a of 67 dBA. A benefitted receptor is defined as obtaining a noise

a The Noise Abatement Criteria, or NAC, is defined by FHWA to represent the upper limit of acceptable highway traffic noise for different types of land uses and human activities.


reduction of at least 5 dB and the design goal of 7 dB for at least one first row receptor must also be met. Caltrans uses a reasonableness allowance of $55,000 per benefitted receptor. Proposed Noise Barriers In Segment 3, three barriers are proposed as indicated in Figure 1: two barriers on the westbound side, W10 and W11, and one barrier on the eastbound side, E11. Barrier W10 is proposed to shield four (4) impacted single-family residences. Barrier W11 is proposed to shield the adjacent park. In addition, residences located north of the park, which are shielded by existing developer walls and, therefore, have noise levels slightly below the NAC threshold, are considered benefitted under some of the alternatives. Barrier E11 is proposed to shield a mobile home park.

Figure 1: Ariel photograph of California I-580 Segment 3, indicating proposed barrier locations and the number of impacted receptors in parentheses. Pavement Alternatives All eight lanes of the existing pavement are aged Portland cement concrete (PCC) with longitudinal texture (LT-PCC). The additional lanes incorporate a portion of the existing shoulder and a newly constructed pavement to provide the added lanes and shoulders. The construction options considered for the added lanes and shoulders are:

1. PCC pavement (similar to the existing longitudinally tined surface). 2. Hot mix asphalt (HMA) pavement and all lanes receiving a quieter friction course

overlay.


The following pavement alternatives were considered for the LCCA, with the present value (NPV) results indicated in the parentheses as Agency cost per mile. The present value costs are the barrier LCCA costs based on $51.61/sq ft (scaled to height and length as needed) plus the pavement present values described in NCHRP Report 7381.

1. Added PCC Lanes Only (NPV: $3,691,000) o Construct additional lanes and shoulders with PCC, similar to the surface texture

of the existing pavement. o The existing pavement is in good condition and does not require rehabilitation at

that time. o Diamond-grind all lanes (for noise and other considerations) 10 years after the

addition of the HOV lanes and every 20 years thereafter.

2. Added PCC Lanes, All Lanes Ground (NPV: $5,060,000) o Construct additional lanes and shoulders with PCC and diamond grind all lanes to

reduce the tire/pavement noise levels. o Diamond-grind all lanes on a 20-year cycle thereafter.

3. Added PCC Lanes, open-graded rubberized asphalt concrete (RAC(O)) Overlay

(NPV: $4,668,000) o Construct additional lanes and shoulders with PCC and overlay all lanes and

shoulders with a 1-in. RAC(O) overlay. o Mill the RAC(O) overlay and replace it every 9 years for noise performance.

4. Added HMA Lanes, RAC(O) Overlay (NPV: $5,353,000)

o Construct additional lanes and shoulders with HMA and overlay all lanes and shoulders with a 1-in. RAC(O) overlay.

o Mill the RAC(O) overlay and replace it every 9 years for noise performance.

5. Added HMA Lanes, All HMA (NPV: $5,446,000) o Construct additional lanes and shoulders with HMA and overlay existing lanes

and shoulders with a 5-in. HMA overlay. o Mill 2-in. of the HMA overlay and overlay it on a 12-year cycle.

Alternative 5 (All HMA) is the most expensive. Alternative 4 (RAC(O) overlay on HMA) provides acoustic performance similar to that of Alternative 3, but at a higher cost. Based on considerations of cost and acoustical uniqueness, Alternatives 4 and 5 were not considered for further analysis. Analysis by Individual Barrier TNM was used to predict traffic noise levels for the three different pavement alternatives, using the existing LT-PCC pavement as the reference pavement. Based on TNM noise modeling, the noise levels for the existing LT-PCC pavement are about 1 dB greater than TNM Average pavement. Grinding the PCC lowers the noise level by about 3 dB. The RAC(O) is about 6 dB quieter than the new LT-PCC and about 3 dB quieter than the Ground PCC. Barrier heights ranging from 12 to 16 ft were considered, as indicated in Table 1, because 12 ft was determined to be sufficient to block the line of sight to truck exhaust stacks and 16 ft is generally the maximum allowed height in California.


A total of 4 impacted receptors were identified in the vicinity of Barrier W10. A summary of analysis results for Barrier W10 is provided in Table 1, which shows the number of benefitted receptors, the predicted noise level range for all impacted receptors, the noise reduction range provided, the total project NPV, the NPV for noise abatement, and the reasonableness allowance calculated for each alternative. In addition, whether or not each alternative would be considered acoustically feasible, cost reasonable, or design reasonable (based on the State policies) is indicated with a ‘Y’ or ‘N’ for yes or no, respectively. Effectiveness is a new term, defined in NCHRP Report 7381 and shown in the tables in this paper as the difference in noise reduction between the given alternative and the ‘most effective’ alternative (i.e., the alternative that provides the greatest noise reduction). The difference is based on the maximum predicted levels for the options being compared. Therefore, an effectiveness of 0 dB indicates the alternative that produces the lowest noise level and an effectiveness of 4 dB would indicate that the given alternative produces 4 dB more noise than the ‘most effective’ alternative. Without a barrier, the traffic noise levels are predicted to range from 68 to 77 dBA, or 2 to 11 dB above the NAC. All abatement alternatives are acoustically feasible except for grinding without a barrier. Only RAC(O) without a barrier is reasonable for cost, but this alternative does not meet the design reasonableness criteria. As a result, none of these alternatives would be considered and no abatement would be proposed.

Table 1: Feasible and reasonable requirements met under Caltrans policies with effectiveness and cost information – I-580 Segment 3, W10

Pavement Type and Barrier Height

Rec

epto

rs B

enef

ited

(out

of 4

Impa

cted

)

Pred

icte

d Le

vel

Ran

ge, d

BA

Noi

se R

educ

tion

Ran

ge, d

B

Tota

l Pro

ject

NPV

($

1000

)

NV

P fo

r Noi

se

Aba

tem

ent (

$100

0)

Rea

sona

blen

ess

Allo

wan

ce ($

1000

)

Feas

ible

Cos

t Rea

sona

ble

Des

ign

Reas

onab

le

Effe

ctiv

enes

s, dB

PCC + 0 ft 0 68-77 0 559 - - 13 PCC + 12 ft 3 66-68 2-9 1,113 554 165 Y N Y 4 PCC + Grd + 0 ft 0 65-74 3 767 207 - N N N 10 PCC + Grd + 12 ft 4 63-67 5-10 1,321 761 220 Y N Y 3 PCC + RAC(O) + 0 ft 4 62-71 6 707 148 220 Y Y N 7 PCC + RAC(O) + 12 ft 4 62-64 5-13 1,261 632 220 Y N Y 0

A total of 13 impacted receptors were identified in the vicinity of Barrier W11, although some of the alternatives result in 7 additional benefitted receptors that were not considered to be impacted. Predicted noise levels at the park behind proposed Barrier W11 range up to 78 dBA with the LT-PCC without the barrier. As shown in Table 2, the three alternatives that include barriers are acoustically feasible and meet the design goal of 7 dB, but only two of the alternatives are cost reasonable: Ground +12 ft and RAC(O) +12 ft. An increase in the barrier height for the PCC alternative up to 16 feet did not produce additional benefitted receptors and, therefore, continued to not be cost reasonable. Thus, only the two 12 ft barriers with either of the quieter pavements meet the feasible and reasonable criteria. These two alternatives are nearly


equal in effectiveness and NPV for abatement with the RAC(O) + 12 ft barrier alternative having a small advantage.

Table 2: Feasible and reasonable requirements met under Caltrans policies with effectiveness and cost information – I-580 Segment 3, W11


Rec

epto

rs B

enef

ited

(out

of 1

3 Im

pact

ed)

Pred

icte

d Le

vel

Ran

ge, d

BA

Noi

se R

educ

tion

Ran

ge, d

B

Tota

l Pro

ject

NPV

($

1000

)

NV

P fo

r Noi

se

Aba

tem

ent (

$100

0)

Rea

sona

blen

ess

Allo

wan

ce ($

1000

)

Feas

ible

Cos

t Rea

sona

ble

Des

ign

Reas

onab

le

Effe

ctiv

enes

s, dB

PCC + 0 ft 0 64-78 0 629 - - 9 PCC + 14 ft 3 61-70 3-8 1,356 727 165 Y N Y 3 PCC + Grd + 0 ft 0 62-75 2-3 863 233 - N N N 8 PCC + Grd + 12 ft 20 59-68 4-10 1,486 857 1,100 Y Y Y 1 PCC + RAC(O) + 0 ft 0 61-74 2-4 796 167 - N N N 7 PCC + RAC(O) + 12 ft 20 59-67 4-11 1,419 790 1,100 Y Y Y 0

A total of 16 impacted receptors were identified in the vicinity of Barrier E11. Analysis of Barrier E11, as indicated in Table 3, indicates noise levels for the existing pavement without any barrier ranging from 69 to 81 dBA, or 3 to 15 above the NAC. All three barrier alternatives meet the feasible and reasonable criteria. In this case, the PCC with the 12 ft barrier was the lowest cost, but is the least effective and benefits the lowest number of receptors. The two quieter pavement alternatives with 12 ft barriers are nearly equal in effectiveness, number of benefitted receptors, and NPV for abatement with the RAC(O) + 12 ft barrier alternative having a small cost advantage.

Table 3: Feasible and reasonable requirements met under Caltrans policies with effectiveness and cost information – I-580 Segment 3, E11


Rec

epto

rs B

enef

ited

(out

of 1

6 Im

pact

ed)

Pred

icte

d Le

vel

Ran

ge, d

BA

Noi

se R

educ

tion

Ran

ge, d

B

Tota

l Pro

ject

NPV

($

1000

)

NV

P fo

r Noi

se

Aba

tem

ent (

$100

0)

Rea

sona

blen

ess

Allo

wan

ce ($

1000

)

Feas

ible

Cos

t Rea

sona

ble

Des

ign

Reas

onab

le

Effe

ctiv

enes

s, dB

PCC + 0 ft 0 69-81 0 629 - - 12 PCC + 12 ft 10 66-71 3-10 1,252 623 550 Y Y Y 2 PCC + Grd + 0 ft 0 67-79 2 863 233 - N N N 10 PCC + Grd + 12 ft 16 64-69 5-12 1,486 857 880 Y Y Y 0 PCC + RAC(O) + 0 ft 6 65-76 4-5 796 167 330 Y Y N 7 PCC + RAC(O) + 12 ft 16 62-69 7-12 1,419 790 880 Y Y Y 0


Analysis of Segment as a Whole With Barriers W11 and E11 being located across the highway from each other, the cost of quieter pavement would be shared between the two impacted areas. As a result, a combined hybrid approach can be used for the analysis of the entire segment as one piece. Using one of the two viable options for W11 that include a quieter pavement directly opposite to the E11 barrier would result in an overall abatement NPV of $1,479,000 for the Ground PCC option and $1,413,000 for the RAC(O) option ($233,000 for the cost of grinding or $167,000 for the cost of RAC(O), plus $623,000 each for the two 12 ft barriers W11 and E11). For this case, the reasonableness allowance for the combined 36 benefitted receptors would be $1,980,000 and either hybrid solution would meet the feasible and reasonable criteria. Another hybrid option would be to apply quieter pavement over the entire segment, including the portion of the roadway adjacent to the single-family residences to the northeast of the Vasco Road Interchange, where Barrier E10 was not found to have any stand-alone feasible and reasonable alternatives. In this case, extending the total pavement length to 2,275 feet would result in a total abatement cost of $1,667,000 for the RAC(O) option ($421,000 for the cost of RAC(O) plus $623,000 for each of the two 12 ft barriers,W11 and E11). The allowance for the new total of 40 benefited receptors would be $2,200,000. This hybrid alternative is feasible, cost reasonable, meets the design criteria, and provides benefit for 4 more receptors in the area and is the most effective alternative for those shielded by W11 and E11. Example 2: I-40 Lane Addition (North Carolina) In this example case, an additional travel lane and shoulder is proposed in the median in both directions of travel of I-40 near Raleigh, North Carolina, increasing the highway from 6 to 8 lanes of travel. Policy Summary The results of abatement analysis for this example were compared to NCDOT criteria7 except in examining cost reasonableness. NCDOT uses a threshold level for noise impact of 66 dBA as approaching the Federal NAC of 67 dBA. Acoustically feasible options must achieve a noise reduction of 5 dB for at least one impacted receptor. Benefited receptors for use in reasonableness determination are receptors that receive a reduction of 5 dB regardless of impact determination. The design goal is 7 dB for at least one front row receptor. NCDOT cost reasonableness is calculated on an allowed sq ft of barrier area per benefited receptor, up to a maximum of 2,500 sq ft per each receptor. This method of reasonableness cannot be directly applied in the LCCA analysis, in which the cost of barriers are combined with and compared to pavement costs. Instead, the National average barrier cost of $35/sq ft and allowance of $37,500 per benefited receptor was used for this analysis. Proposed Noise Barriers Northbound and Southbound Barriers were proposed along I-40 to shield primarily residential receptors, as shown in Figure 2.


Pavement Alternatives The existing lanes are constructed of PCC with a semi-random transverse texture. For the widening project, the following three scenarios were considered with the present value (NPV) results indicated in the parentheses as Agency cost per mile.

1. Added PCC Only (NPV: $4,186,000) o Construction of new lanes and shoulders with PCC (transverse tined). o Future rehabilitation includes diamond grinding of all lanes on a 20 year cycle.

2. Added PCC Lanes, All Lanes Ground (NPV: $5,427,000)

o Construction of new lanes and shoulders with PCC, which is then diamond ground together with the existing lanes.

o Future rehabilitation includes diamond grinding of all lanes on a 20 year cycle.

3. Added PCC Lanes, HMA Overlay (NPV: $5,942,000) o Construction of new lanes and shoulders with PCC and overlay of all lanes with

1-inch thick, 9.5 mm HMA. o Future rehabilitation includes mill and overlay all lanes with 1-inch thick, 9.5 mm

HMA every 9 years.

Figure 2: Location of Barriers on I-40 highway widening project near Raleigh, NC, indicating proposed barrier locations and the number of impacted receptors in parentheses.


Analysis by Individual Barrier For the I-40 example, the existing semi-random transverse tined PCC pavement was used as the reference pavement. Based on TNM noise modeling, the noise levels for the existing transverse tined PCC pavement are about 3 dB greater than TNM Average pavement. Grinding the PCC lowers the noise level by about 3 dB. The HMA is about 6 dB quieter than the new transverse tined PCC and about 3 dB quieter than the Ground PCC. Barrier heights ranging from 12 to 16 ft were considered, as indicated in Table 4. A summary of analysis results for the Southbound Barrier is provided in Table 4. A total of 43 impacted receptors were identified for the Southbound Barrier. As shown in Table 4, all of the Southbound Barrier alternatives, except for the grinding only alternative, produce a feasible noise reduction of 5 dB or more. Cost reasonableness was confirmed for four alternatives: PCC +16 ft, PCC Ground + 12 ft, HMA with no barrier, and HMA with a 12 ft barrier. For the case with the existing PCC and a 14 ft barrier, the added height over the 12 ft barriers for the quieter pavements did not result in a sufficient number of benefited receptors to generate the required allowance. Only the four barrier alternatives met the design goal reduction of 7 dB or more, although the HMA without a barrier provided reductions of up to 6 dB. Three alternatives met all of the criteria: Ground PCC with a 12 ft barrier, HMA with a 12 ft barrier, and the existing PCC with a 16 ft barrier. Of these alternatives, the HMA +12 ft is the most acoustically effective, resulting in noise levels that are 3 dB lower than Ground PCC +12 ft and 6 dB lower than PCC +16 ft. The HMA + 12 ft alternative also resulted in about twice as many benefited receptors as any other option. However, the NVP for the HMA +12 ft alternative is higher than the NVP for Ground PCC + 12ft and the PCC +16 ft alternatives by $151,000 and $273,000, respectively. The PCC + 16 ft is the least expensive option, but is the least effective of the alternatives that meet all criteria, with the fewest benefitted receptors.

Results of the analysis for the Northbound Barrier are shown in Table 5, which is proposed to shield a total of 52 impacted receptors. Again, all of the alternatives except for grinding alone are acoustically feasible and all alternatives with barriers except PCC + 14 ft meet the criterion

Table 4: Feasible and reasonable requirements met under NCDOT policies with effectiveness and cost information –I-40 Southbound Barrier


Rec

epto

rs B

enef

ited

(out

of 4

3 Im

pact

ed)

Pred

icte

d Le

vel

Max

& A

vg, d

BA

Noi

se R

educ

tion

Ran

ge, d

B

Tota

l Pro

ject

NPV

($

1000

)

NV

P fo

r Noi

se

Aba

tem

ent (

$100

0)

Rea

sona

blen

ess

Allo

wan

ce ($

1000

)

Feas

ible

Cos

t Rea

sona

ble

Des

ign

Reas

onab

le

Effe

ctiv

enes

s, dB

PCC + 0 ft 0 77/67 0 1,231 - - 9 PCC + 14 ft 21 74/64 3-9 2,082 851 810 Y N Y 6 PCC + 16 ft 29 74/63 4-11 2,203 972 1,110 Y Y Y 6 PCC + Grd + 0 ft 0 74/64 3 1,596 365 - N N N 6 PCC + Grd + 12 ft 38 71/62 5-10 2,325 1,094 1,448 Y Y Y 3 PCC + HMA+ 0 ft 35 71/62 4-6 1,747 516 1,335 Y Y N 3 PCC + HMA + 12 ft 70 68/60 7-12 2,476 1,245 2,648 Y Y Y 0


for cost reasonableness. The HMA without a barrier is under the allowance and is cost reasonable but it does not meet the design goal requirement. The HMA with the 12 ft barrier is most acoustically effective by 3 to 6 dB as compared to the other alternatives that met all of the criteria and benefits the greatest number of receptors, but it has the highest NPV cost. Again, the PCC + 16 ft is the least expensive option, but is the least effective, with the fewest benefitted receptors.

Table 5 Feasible and reasonable requirements met under NCDOT policies with effectiveness and cost information –I-40 Northbound Barrier


Rec

epto

rs B

enef

ited

(out

of 5

2 Im

pact

ed)

Pred

icte

d Le

vel M

ax

& A

vg, d

BA

Noi

se R

educ

tion

Ran

ge, d

B

Tota

l Pro

ject

NPV

($

1000

)

NV

P fo

r Noi

se

Aba

tem

ent (

$100

0)

Rea

sona

blen

ess

Allo

wan

ce ($

1000

)

Feas

ible

Cos

t Rea

sona

ble

Des

ign

Reas

onab

le

Effe

ctiv

enes

s, dB

PCC + 0 ft 0 76/66 0 2,962 - - 11 PCC + 14 ft 51 70/61 5-10 5,010 2,048 1,940 Y N Y 5 PCC + 16 ft 62 70/60 5-11 5,303 2,340 2,352 Y Y Y 5 PCC + Grd + 0 ft 0 73/64 2-3 3,841 878 - N N N 8 PCC + Grd + 12 ft 90 68/59 7-11 5,596 2,633 3,402 Y Y Y 3 PCC + HMA+ 0 ft 33 70/61 4-5 4,205 1,243 1,265 Y Y N 5 PCC + HMA + 12 ft 98 65/58 8-12 5,960 2,998 3,702 Y Y Y 0

Analysis of Segment as a Whole Considering the alternatives that use quieter pavements in combination with barriers, the cost of the quieter pavement would be shared between both sides of the freeway. As a result, combined alternatives can be assessed. To analyze these alternatives, the cost of the pavement for the greatest length needed to mitigate both impacted areas (3,738 ft for the Northbound Barrier) is added to the cost of the two barriers to produce the NPVs shown in Table 6. These costs account for the barriers and the pavement, with the quieter pavement included only once because it affects receptors on both sides of the highway. The total NPV cost for the alternatives using Ground PCC and HMA are close to the NPV of the PCC + 16 ft alternative. For comparison between the individual barrier analyses and the combined analysis for the HMA + 12 ft barrier, the summed NPV cost from Tables 4 and 5 is $4,243,000, as compared to $3,727,000 for the combined analysis in Table 6. Taking into account the total number of benefited receptors in the combined analysis as a sum of those reported in Tables 4 and 5, the HMA + 12 ft barrier alternative results in the lowest cost/benefited receptor at $22,185. Consideration of the cost/benefitted receptor may be useful in comparing abatement alternatives having relatively close NPV costs and noise reduction performance. Another alternative to consider would be building the Northbound Barrier and using the HMA pavement without the Southbound Barrier. The HMA pavement without a barrier will produce 35 benefited receivers (Table 4) and the Northbound Barrier will produce another 98 for a total of 133 benefited receivers for a total cost of $2,998,000 ($1,243,000 + $1,755,000). This alternative meets all criteria, has the lowest NPV, and has a cost/benefited receptor of $22,541.


Table 6: Cost and allowance results for combined pavement and barrier analysis considering both sides of the highway

Example 3: New 8-Lane Highway (Arizona) In this project, an existing uncontrolled access, 4-lane section of state highway in Arizona will be completely replaced with an 8-lane access controlled freeway as an extension of an existing freeway. Policy Summary For this example, ADOT policy8 was used. ADOT defines “approaching” the 67 dBA NAC as 3 dB below this criteria, or 64 dBA. For acoustic feasibility, at least half of the impacted receptors must receive a 5 dB reduction. To determine cost reasonableness, benefited receptors are allowed up to $49,000 in costs for noise abatement with the cost of barriers set at $35/sq ft for barriers on grade. To meet design goal, at least half of the benefited receptors in the first row must receive a 7 dB reduction. Proposed Noise Barriers This study segment extends over a length of about 13,500 ftb. The westbound side of the right-of-way contains relatively high-density residential receptors in several subdivisions and a barrier is considered for this entire project length adjacent to the residential receptors. The eastbound side of the proposed freeway does not have residential receptors, but has a recreational use area that is considered a sensitive receptor. For this recreational receptor, a barrier with a length of 2,632 ft parallel to the barrier on the opposite side is considered. Pavement Alternatives The following pavement scenarios are considered with the present value (NPV) results for the 13,500 ft long pavement alternative indicated in the parentheses as Agency cost per mile.

1. LT-PCC (NPV: $27,013,000) o Construct all new lanes and shoulders with longitudinally tined PCC. o Diamond-grind all lanes on a 20-year cycle.

2. LT-PCC with 1” asphalt rubber friction course (ARFC) Overlay (NPV: $34,471,000)

b An aerial is not included for this example due to the ongoing and controversial nature of this project.


o Construct all new lanes and shoulder with PCC and overlay all travel lanes with 1 in. thick ARFC.

o Mill and overlay all lanes with 3/4 in. ARFC every 9 years.

3. HMA with 1” ARFC Overlay (NPV: $30,627,000) o Construct all new lanes and shoulder with HMA and overlay all travel lanes with

1 in. thick ARFC. o Mill and overlay all lanes with 3/4 in. ARFC every 9 years.

Analysis of Segment as a Whole For this analysis, new longitudinally tined PCC pavement that produced predicted levels about 1½ dB higher than TNM Average pavement was considered as the acoustic baseline. For noise abatement, barriers with heights of 16 and 10 ft on both sides of the freeway were considered, as indicated in Table 7. A hybrid analysis is used, accounting for the quieter pavement affecting both sides of the highway. However, with one barrier being much shorter than the other, there was little cost savings with the shared pavement costs. The total number of impacted receptors for both sides of the highway is 249. The results, when compared to ADOT criteria are provided in Table 7.

Table 7: Feasible and reasonable requirements met under ADOT policies with effectiveness and cost information – 8 lane new highway construction


Rec

epto

rs B

enef

ited

(out

of 2

49 Im

pact

ed)

Pred

icte

d Le

vel M

ax

& A

vg, d

BA

Noi

se R

educ

tion

Ran

ge, d

B

Tota

l Pro

ject

NPV

($

1000

)

NV

P fo

r Noi

se

Aba

tem

ent (

$100

0)

Rea

sona

blen

ess

Allo

wan

ce ($

1000

)

Feas

ible

Cos

t Rea

sona

ble

Des

ign

Reas

onab

le

Effe

ctiv

enes

s, dB

PCC + 0 ft 0 80/75 0 27,013 - - 16 PCC + 16 ft 224 67/65 13/10 37,114 10,101 10,976 Y Y Y 3 PCC + 10 ft 224 72/68 10/7 33,326 6,313 10,976 Y Y Y 8 PCC + ARFC 120 75/71 5/4 34,471 7,458 5,880 N N N 11 PCC + ARFC + 16 ft 249 64/62 16/13 44,572 17,559 12,201 Y N Y 0 PCC + ARFC + 10 ft 249 69/66 13/9 40,784 13,771 12,201 Y N Y 5 HMA + ARFC 120 75/71 5/4 30,627 3,614 5,880 N Y N 11 HMA + ARFC + 16 ft 249 64/62 16/13 40,728 13,715 12,201 Y N Y 0 HMA + ARFC + 10 ft 249 69/66 13/9 36,940 9,927 12,201 Y Y Y 5

As shown in Table 7, all of the barrier alternatives meet the acoustic feasibility and design reasonableness criteria. The acoustic performance of the ARFC on either the PCC or the HMA pavement is the same, but the NPV of the HMA is lower than the PCC alternative. For the ARFC only cases, the feasibility criterion was not met, as a 5 dB reduction was only achieved for 120 out of 249, or 48%, of the receptors (50% is required under ADOT policy). Of the feasible and design reasonable alternatives, only three alternatives also meet the cost reasonableness criteria: PCC + 10 ft, +16 ft, and HMA + ARFC + 10 ft. Of these, PCC + 16 ft barrier is most acoustically effective and the PCC + 10 ft barrier is the least effective but it has the lowest NPV


costs. The PCC + 10 ft alternative provided the lowest cost per benefited receptor because the number of benefited receptors does not change with the decreased barrier height, although the effectiveness is reduced by 5 dB. The HMA + ARFC with a 10 ft barrier alternative comes in between the PCC alternatives in terms of cost and effectiveness but it produces the highest number of benefited receptors. Again, consideration of the cost/benefitted receptor may be useful in comparing abatement alternatives having relatively close NPV costs and noise reduction performance.

CONCLUSIONS From the research reported here, it was found that the FHWA pavement LCCA process, using the research version of TNM to account for pavement based on OBSI data, could be readily adapted to state highway projects for the purposes of comparing noise abatement that use barriers, pavement strategies, or combinations of both. As described in this paper, using OBSI acoustic longevity data, the analysis can be extended to the life cycle of the highway project. The case studies presented indicate the potential cost savings and impact reduction that could be achieved by considering barriers and quieter pavement together. This is further confirmed in the NCHRP 10-76 Project1. Under current policy, quieter pavement alone did not typically meet the design goal requirements, although it offered considerable cost advantages in some cases when compared to the barrier alternatives. As a result, new policies may need to be considered to allow for quieter pavement noise abatement in cases where barriers are not feasible. For the two lane addition examples, quieter pavements combined with low-height barriers were found to be the most acoustically effective alternatives and to result in the most benefitted receptors. For the North Carolina I-40 case, traditional PCC pavement with a higher barrier was less expensive as compared to the quieter pavement with low-height barrier alternatives; however, it was also less effective with fewer benefitted receptors by about half. For the California I-580 case, none of the traditional PCC options were found to be feasible and reasonable. In both examples, when barriers were evaluated on both sides of the highway, the cost of quieter pavement would be shared between the two impacted areas. As a result, cost reasonableness is improved when analyzing both sides of the freeway together because of the lower combined cost for the same number of benefitted receptors. In the California I-580 case, an area that would not have qualified for any noise abatement when assessed as a stand-alone portion of the project barrier could be benefitted by the extension of the quieter pavement from an adjacent qualifying area. This hybrid alternative, which included quieter pavement for the entire segment and construction of two of the three proposed barriers, was found to be feasible, cost reasonable, meet the design criteria, benefit 4 additional receptors in the area, and is the most effective alternative for the areas that did qualify for barriers. For the Arizona new construction case, again quieter pavement alone was unable to meet the design goal. This design goal could have been met had the uniform transverse tine texture that was previously used as a standard highway pavement by ADOT been considered as the baseline; OBSI levels for the uniform transverse tine PCC were typically 3 to 5 dB higher than for a longitudinally tined PCC pavement9. For this example, longitudinally tined PCC with a 16 ft barrier was found to be most acoustically effective, but with the highest NPV cost and PCC with the 10 ft barrier was found to be least effective, but with the lowest NPV costs. Both PCC options resulted in the same number of benefitted receptors. Only one quieter pavement alternative with a 12 ft barrier was found to meet all criteria. This alternative came in between


the PCC alternatives in terms of cost and effectiveness, but produced the highest number of benefited receptors. In all three examples, several alternatives with different NPV costs, effectiveness, and benefitted receptors would meet all criteria. For such cases, a rational approach for trading off NPV cost and effectiveness needs to be considered. Combining barriers and pavement in the analysis can also facilitate lower cost solutions and more benefitted receivers than considering either barriers or pavement alone. The cost per benefited receptor may be an appropriate criterion for comparing alternatives with different numbers of benefited receptors.

IMPLEMENTATION CHALLENGES Some of the overall challenges of implementing the NCHRP 10-76 methods to state departments of transportation are described in Cost-Benefit Analysis – Noise Barriers and Quieter Pavements10. These include: 1) development of a noise evaluation process that includes both noise barriers and quieter pavements, 2) upgrading the public release version of TNM to include the OBSI-related pavement assessment capabilities, and 3) implementation of a pilot program to evaluate the methods presented and improve the process. ACKNOWLEDGEMENTS The research reported here was funded by NCHRP under Dr. Amir Hanna. Research Team members also contributing to the work include Dr. Linda Pierce, P.E. of APTech, and Harvey Knauer, P.E. of Environmental Acoustics, Inc.


REFERENCES

1 Donavan, Paul R., “NCHRP Report 738: Evaluating Pavement Strategies and Barriers for Noise Mitigation”, Transportation Research Board, Washington, D.C., 2012.

2 Donavan, Paul R., “Methodologies for Evaluating the Lifecycle Cost and Acoustic

Performance of Barriers and Pavement for Highway Noise Reduction”, Transportation Research Board, Washington, D.C., 2012.

3 Shrouds, J., “Information: Highway Traffic Noise – Guidance on Quiet Pavement Pilot

Programs and Tire/Pavement Noise Research”, http://www.fhwa.dot.gov/environment/noise/qpppmem.htm, Office of Natural and Human Environment, Federal Highway Administration, U.S. Department of Transportation, January 19, 2005.

4 Rochat, J, Hastings, A., Read, D., and Lau, M., FHWA Traffic Noise Model (TNM)

Pavement Effects Implementation Study: Progress Report 1, Report Nos. FHWA-HEP and DOT-VNTSC-FHWA-12-01, U.S. Department of Transportation, Volpe National Transportation Systems Center, January 2012.

5 Standard Practice for Measurement of Tire/Pavement Noise Using the On-Board Sound

Intensity (OBSI) Method”, TP 76-11 (Proposed), American Association of State Highway and Transportation Officials, 444 North Capitol Street N.W., Suite 249, May 2009.Washington, D.C. 2001.

6 Caltrans Traffic Noise Analysis Protocol, California Department of Transportation, May

2011. http://www.dot.ca.gov/hq/env/noise/pub/ca_tnap_may2011.pdf 7 Traffic Noise Analysis and Abatement Manual, North Carolina Department of Transportation,

Human Environment Unit, Raleigh, NC, July 2011. 8 Noise Abatement Policy, Arizona Department of Transportation, Environmental Planning

Group, Phoenix, AZ, July 2011. 9 Donavan, P., and Scofield, L., An Evaluation of the Effects of Different Portland Cement

Concrete Pavement Texturing on Tire/Pavement Noise, Proceedings of Noise-Con 2003, Cleveland, OH, June 2003.

10 Vanchieri, Cori (Rapporteur), Cost-Benefit Analysis - Noise Barriers and Quieter Pavements:

A Workshop Sponsored by the INCE Foundation, the Noise Control Foundation, and the Transportation Research Board Committee ADC40, 2014.

applying the nchrp 10-76 methodology for …docs.trb.org/prp/15-4208.pdfapplying the nchrp 10-76...

Documents