Design Guidelines for Mitigating Collisions with Trees and Utility Poles (2022)

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.

24 This chapter presents computational examples to illustrate the benefit-cost analysis method using the results of the recommended crash prediction model. The following five examples are presented: • Benefit-Cost Example 1—Removal of an Isolated Roadside Tree • Benefit-Cost Example 2—Removal of a Continuous Group of Roadside Trees • Benefit-Cost Example 3—Relocation of an Isolated Utility Pole • Benefit-Cost Example 4—Relocation of an Extended Series of Utility Poles • Benefit-Cost Example 5—Removal of an Extended Series of Utility Poles and Replacement with Underground Utilities The examples presented here are intended to illustrate how proposed improvements can be assessed with the crash prediction model and the benefit-cost method. For any actual project that an agency wishes to assess, it is recommended that site-specific roadway characteristics, roadside characteristics, and traffic volumes be used and that implementation costs for the improvement be based on agency experience with tree- and utility-pole-related crash mitiga- tion projects. The computations for all of the examples were performed with the spreadsheet tool described in Appendix B. 6.1 Benefit-Cost Example 1—Removal of an Isolated Roadside Tree Benefit-Cost Example 1 addresses removal of a single isolated tree on the roadside of a rural 2-lane undivided highway. Table 9 summarizes the input data for the example. All of the input data are constant values except that computations were made for roadway AADT values of 1,000, 5,000, and 10,000 vpd and the distance of the tree to be removed from the outside edge of the traveled way was 6 ft and from 10 to 40 ft in increments of 5 ft. The cost of removing a single tree is estimated as $2,000. For the initial case analyzed for Benefit-Cost Example 1, with a roadway AADT of 1,000 vpd and the tree in the before condition located 6 ft from the edge of the roadway traveled way, the predicted annual crashes estimated with the crash prediction model are as shown in Table 10. The table shows both the crash predictions obtained from the model and intermediate results as derived from Equations (1) through (15). Because the single roadside tree is being removed, there will be no tree-related crashes after the improvement and, therefore, ΔNROR is equal to NROR and ΔNROR-j for each crash severity level is equal to NROR1j as defined in Equations (19) and (20). C H A P T E R   6 Benefit-Cost Analysis Examples

Benefit-Cost Analysis Examples 25   Variable Name Before Condition After Condition NROR 0.0013 0.0000 NROR-left 0.0000 0.0000 NROR-right 0.0013 0.0000 NROR-K 0.0001 0.0000 NROR-A 0.0002 0.0000 NROR-B 0.0005 0.0000 NROR-C 0.0005 0.0000 NROR-PK 0.0001 0.0000 NROR-PI-A 0.0003 0.0000 NROR-PI-B 0.0006 0.0000 NROR-PI-C 0.0008 0.0000 LikelihoodROR 1.0000 1.0000 SeverityROR-left 0.0000 0.0000 SeverityROR-right 17.8450 0.0000 RSSROR-left 0.0000 0.0000 RSSROR-right 2.8674 0.0000 Table 10. Computational results with crash prediction model before and after removal of an isolated tree 6 ft from the traveled way of a rural 2-lane highway with an AADT of 1,000 vpd. • Paved shoulder width 6 ft • Curvature Tangent alignment • Advance visibility of curve Substantial • Grade 2% • Shoulder rumble strips Present • Presence of delineation Substantial • Road surface condition Good • Skid resistance High Input Data Element Specified Value • Roadway type Rural 2-lane undivided highway • Traffic volume (AADT) 1,000, 5,000, and 10,000 vpd • Design speed 70 mph • Before condition—number of roadside trees on the right side of the road 1 • Before condition—number of roadside trees on the left side of the road 0 • Before condition—distance from outside edge of traveled way to roadside tree(s) 6, 10, 15, 20, 25, 30, 35, and 40 ft • After condition—number of roadside trees on the right side of the road 0 • After condition—number of roadside trees on the left side of the road 0 • Lane width 12 ft Table 9. Summary of input data for Example 1—removal of an isolated roadside tree.

26 Design Guidelines for Mitigating Collisions with Trees and Utility Poles For the initial case, the B/C for removing the single tree can be computed with Equation (17) as follows: [ ] ( )= × + × + × + × × = B C 0.00001 11,295,400 0.00002 655,000 0.00005 198,500 0.00005 125,600 10.594 2,000 7.8 The net benefits for removing the single tree can be computed with Equation (18) as follows: 0.00001 11,295,400 0.00002 655,000 0.00005 198,500 0.00005 125,600 10.594 2,000 $13,555 [ ] ( )= × + × + × + × × − = NB Table 11 shows the computed crash reductions and benefit-cost ratios for each value of AADT and distance from the traveled way to the roadside tree considered. The annual crash reduction estimates in Table 11, and subsequent tables of this type, assume that the trees (or utility poles) removed at any specified distance from the traveled way are the only trees (or utility poles) present on the roadside within 40 ft of the traveled way. If trees (or utility poles) at multiple distances from the traveled way are present, the trees (or utility poles) at each distance from the traveled way should be analyzed separately. The results presented in Table 11 can be easily modified to account for changes in the assumed conditions listed in Table 9. Each value of annual crash reduction and benefit-cost ratio in Table 11 can be multiplied by the ratio of adjustment factor shown in Appendix A for any specific condition to the adjustment factor shown in Appendix A for the condition of interest shown in Table 11. For example, the results in Table 11 apply to a roadway on tangent alignment. The adjustment factor for moderately curved alignment shown in Table A-6 is 1.81, while the adjust- ment factor for tangent alignment is 1.00. The ratio of these two values (1.81/1.00 = 1.81) repre- sents an adjustment factor for the difference between moderately curved and tangent alignment. Any value of average crash reduction or benefit-cost ratio in Table 11, which applies to tangent alignment, can be multiplied by 1.81 to obtain an equivalent value for moderately curved align- ment. Expressed another way, the adjustment factor ratio of 1.81 indicates that moderately curved alignment will likely result in average crash reduction and benefit-cost ratio values 81% higher than the values shown for tangent alignment in Table 11. Distance from Traveled Way to Tree (ft) Annual Crash Reduction (crashes/yr)a Benefit-Cost Ratio (B/C) Traffic Volume (AADT) (vpd) 1,000 5,000 10,000 1,000 5,000 10,000 6 0.0013 0.0061 0.0103 7.8 36.3 61.0 10 0.0011 0.0053 0.0088 6.7 31.1 52.7 15 0.0009 0.0041 0.0070 5.3 24.5 41.2 20 0.0007 0.0030 0.0051 3.9 18.3 30.7 25 0.0005 0.0022 0.0037 2.8 13.2 22.2 30 0.0004 0.0017 0.0029 2.2 10.2 17.1 35 0.0003 0.0012 0.0020 1.5 7.2 12.0 40 0.0001 0.0007 0.0013 0.9 4.1 7.0 aFor all fatal and injury crash severity levels combined. Table 11. Crash reductions and benefit-cost ratios for specific values of AADT and distance from traveled way to roadside trees in benefit-cost Example 1.

Benefit-Cost Analysis Examples 27   The following percentage changes in the values shown in Table 11 represent the effects of key differences in roadway design, traffic control, and traffic speed factors on average crash reduction and benefit-cost ratio: Horizontal Curvature (Base Condition: Tangent) • Moderate curvature 81% increase • Sharp curvature 251% increase • Very sharp curvature 502% increase Lane Width (Base Condition: 12 ft) • 11-ft lanes no change • 10-ft lanes 20% increase • 9-ft lanes 50% increase Paved Shoulder Width (Base Condition: 6 ft) • 0-ft shoulders 21% increase • 2-ft shoulders 15% increase • 4-ft shoulders no change Design Speed (Base Condition: 70 mph) • 30 mph 94% decrease • 40 mph 84% decrease • 50 mph 67% decrease • 55 mph 55% decrease • 60 mph 40% increase • 65 mph 21% decrease Presence of Delineation (Base Condition: Substantial) • Limited delineation 20% increase Road Surface Condition (Base Condition: Good) • Rough condition 40% increase 6.2 Benefit-Cost Example 2—Removal of a Continuous Group of Roadside Trees Benefit-Cost Example 2 addresses removal of a continuous group of trees, 0.5 mi in length spaced at intervals of 5 ft, on the roadside of a rural 2-lane undivided highway. Table 12 sum- marizes the input data for the example. All of the input data are constant values except that computations were made for roadway AADT values of 1,000, 5,000, and 10,000 vpd and the distance of the trees to be removed from the outside edge of the traveled way was 6 ft and from 10 to 40 ft in increments of 5 ft. The cost of removing each individual tree is estimated as $100 with economies of scale since so many trees are being removed. A total of 528 trees need to be removed for a total cost of $52,800. For the initial case analyzed for Benefit-Cost Example 2, with a roadway AADT of 1,000 vpd and the tree group in the before condition located 6 ft from the edge of the traveled way, the predicted annual crashes estimated with the crash prediction model are as shown in Table 13. The table shows both the crash predictions obtained from the model and intermediate results as derived from Equations (1) through (15).

28 Design Guidelines for Mitigating Collisions with Trees and Utility Poles Variable Name Before Condition After Condition NROR 0.0336 0.0000 NROR-left 0.0000 0.0000 NROR-right 0.0336 0.0000 NROR-K 0.0026 0.0000 NROR-A 0.0060 0.0000 NROR-B 0.0126 0.0000 NROR-C 0.0124 0.0000 NROR-PK 0.0028 0.0000 NROR-PI-A 0.0064 0.0000 NROR-PI-B 0.0156 0.0000 NROR-PI-C 0.0200 0.0000 LikelihoodROR 1.0000 1.0000 SeverityROR-left 0.0000 0.0000 SeverityROR-right 17.8450 0.0000 RSSROR-left 0.0000 0.0000 RSSROR-right 2.8674 0.0000 Table 13. Computational results with crash prediction model before and after removal of a tree group 6 ft from the traveled way of a rural 2-lane highway with an AADT of 1,000 vpd. Input Data Element Specified Value • Roadway type Rural 2-lane undivided • Traffic volume (AADT) 1,000, 5,000, and 10,000 vpd • Design speed 70 mph • Before condition—number of individual trees on the right side of the road 0 • Before condition—length of tree group on the right side of the road 0.5 mi • Before condition—number of individual trees on the left side of the road 0 • Before condition—length of tree group on the left side of the road 0 • Distance from outside edge of traveled way to roadside tree(s) 6, 10, 15, 20, 25, 30, 35, and 40 ft • After condition—number of individual trees on the right side of the road 0 • After condition—length of tree group on the right side of the road 0 • After condition—number of individual trees on the left side of the road 0 • After condition—length of tree group on the left side of the road 0 • Lane width 12 ft • Paved shoulder width 6 ft • Curvature Tangent alignment • Advance visibility of curve Substantial • Grade 2% • Shoulder rumble strips Present • Presence of delineation Substantial • Road surface condition Good • Pavement skid resistance Good Table 12. Summary of input data for benefit-cost Example 2—removal of a continuous group of roadside trees.

Benefit-Cost Analysis Examples 29   Table 14 shows the computed crash reductions and benefit-cost ratios for each value of AADT and distance from the traveled way to the roadside trees considered. The results show that, under the specified conditions, removal of trees generally becomes cost-effective between AADTs of 1,000 and 5,000 vpd for tree groups within 30 to 35 ft of the traveled way. The annual crash reduction estimates in Table 14 assume that the trees removed at any speci- fied distance from the traveled way are the only trees present on the roadside within 40 ft of the traveled way. In the case of a tree group with trees beginning at a specified distance from the traveled way and continuing back beyond 40 ft from the traveled way (e.g., a woods or forest), the results in Table 14 could also be used to quantify the crash reduction that would result from removing roadside trees for specific increments of distance from the traveled way. For example, removing trees between 10 and 20 ft from the traveled way for a road with AADT of 5,000 vpd would be expected to reduce 0.0567 crashes per year (computed from the tabulated values as 0.1345 – 0.0778 = 0.0567 crashes per year). Incremental computations of this type are not gen- erally applicable to individual trees, as opposed to tree groups, because any trees located 20 ft from the traveled way are unlikely to be located directly behind the trees located 10 ft from the traveled way. The benefit-cost ratios in Table 14 would need to be recomputed, using an imple- mentation cost applicable to the specific portion of the roadside trees being removed. In most cases where some trees are removed and other trees are not, the appropriate analysis approach is to use the modified RAP model to analyze the before and after conditions separately and then com- pute the difference in crash frequency from the results obtained. 6.3 Benefit-Cost Example 3—Relocation of a Single Utility Pole Benefit-Cost Example 3 addresses relocation of a single utility pole on the roadside of a rural 2-lane undivided highway. Table 15 summarizes the input data for the example. All of the input data are constant values except that computations were made for roadway AADT values of 1,000, 5,000, and 10,000 vpd and the distance of the tree to be removed from the outside edge of the traveled way was either 6 ft or a value from 10 to 40 ft in increments of 5 ft (as shown in Table 9). The cost of relocating the single utility pole is estimated as $14,000 based on the assumption that several additional utility poles (not necessarily in exposed positions on the roadside) also need to be relocated to maintain continuity of power lines. The relocated utility pole is assumed to be beyond 40 ft from the outside edge of the traveled way. Distance from Traveled Way to Tree (ft) Annual Crash Reduction (crashes/yr)a Benefit-Cost Ratio (B/C) Traffic Volume (AADT) (vpd) 1,000 5,000 10,000 1,000 5,000 10,000 6 0.0336 0.1571 0.2638 0.8 3.5 5.9 10 0.0288 0.1345 0.2258 0.6 3.0 5.1 15 0.0227 0.1061 0.1782 0.5 2.4 4.0 20 0.0167 0.0778 0.1307 0.4 1.7 2.9 25 0.0122 0.0571 0.0959 0.3 1.3 2.2 30 0.0094 0.0440 0.0739 0.2 1.0 1.7 35 0.0066 0.0310 0.0520 0.1 0.7 1.2 40 0.0038 0.0179 0.0301 0.1 0.4 0.7 aFor all fatal and injury crash severity levels combined. Table 14. Crash reductions and benefit-cost ratios for specific values of AADT and distance from traveled way to roadside trees in Example 2.

30 Design Guidelines for Mitigating Collisions with Trees and Utility Poles For the initial case analyzed for Benefit-Cost Example 3, with a roadway AADT of 1,000 vpd and the individual utility pole in the before condition located 6 ft from the edge of the traveled way, the predicted annual crashes estimated with the crash prediction model are as shown in Table 16. The table shows both the crash predictions obtained from the model and intermediate results as derived from Equations (1) through (15). Table 17 shows the computed crash reductions and benefit-cost ratios for each value of AADT and distance from the traveled way to roadside tree considered. Table 17, in comparison to Input Data Element Specified Value • Roadway type Rural 2-lane undivided • Traffic volume (AADT) 1,000, 5,000, and 10,000 vpd • Mean operating speed of traffic 60 mph • Before condition—number of utility poles on the right side of the road 1 • Before condition—number of utility poles on the left side of the road 0 • Distance from outside edge of traveled way to roadside tree(s) 6, 10, 15, 20, 25, 30, 35, and 40 ft • After condition—number of utility poles on the right side of the road 0 • After condition—number of utility poles on the left side of the road 0 • Lane width 12 ft • Paved shoulder width 6 ft • Curvature Tangent alignment • Advance visibility of curve Substantial • Grade 2% • Shoulder rumble strips Present • Presence of delineation Substantial • Road surface condition Good • Pavement skid resistance High Table 15. Summary of input data for Example 3—relocation of a single utility pole. Variable Name Before Condition After Condition NROR 0.0013 0.0000 NROR-left 0.0000 0.0000 NROR-right 0.0013 0.0000 NROR-K 0.00003 0.0000 NROR-A 0.0001 0.0000 NROR-B 0.0005 0.0000 NROR-C 0.0006 0.0000 NROR-PK 0.00003 0.0000 NROR-PI-A 0.0001 0.0000 NROR-PI-B 0.0006 0.0000 NROR-PI-C 0.0009 0.0000 LikelihoodROR 1.0000 1.0000 SeverityROR-left 0.0000 0.0000 SeverityROR-right 17.8450 0.0000 RSSROR-left 0.0000 0.0000 RSSROR-right 2.8674 0.0000 Table 16. Computational results with crash prediction model before and after removal of an individual utility pole 6 ft from the traveled way of a rural 2-lane highway with an AADT of 1,000 vpd.

Benefit-Cost Analysis Examples 31   Table 11 for Benefit-Cost Example 1, shows that utility pole relocation projects are generally not as cost-effective as tree removal projects because of the higher cost of relocating a utility pole. The table shows that utility relocation projects typically become cost-effective on rural 2-lane undivided highways with AADTs between 1,000 and 5,000 vpd for utility poles located within 15 to 25 ft of the traveled way. 6.4 Benefit-Cost Example 4—Relocation of an Extended Series of Utility Poles Benefit-Cost Example 4 addresses relocation of an extended series of utility poles on the road- side of a rural 2-lane undivided highway. Table 18 summarizes the input data for the example. All of the input data are constant values except that computations were made for roadway AADT Distance from Traveled Way to Tree (ft) Annual Crash Reduction (crashes/yr)a Benefit-Cost Ratio (B/C) Traffic Volume (AADT) (vpd) 1,000 5,000 10,000 1,000 5,000 10,000 6 0.0013 0.0060 0.0101 0.4 1.9 3.2 10 0.0011 0.0051 0.0086 0.4 1.6 2.7 15 0.0009 0.0041 0.0068 0.3 1.3 2.2 20 0.0006 0.0030 0.0050 0.2 0.9 1.6 25 0.0005 0.0022 0.0037 0.1 0.7 1.2 30 0.0004 0.0017 0.0028 0.1 0.5 0.9 35 0.0003 0.0012 0.0020 0.1 0.4 0.6 40 0.0001 0.0007 0.0012 < 0.1 0.2 0.4 aFor all fatal and injury crash severity levels combined. Table 17. Crash reductions and benefit-cost ratios for specific values of AADT and distance from traveled way to utility pole in Example 3. Input Data Element Specified Value • Roadway type Rural 2-lane undivided • Traffic volume (AADT) 1,000, 5,000, and 10,000 vpd • Mean operating speed of traffic 60 mph • Before condition—number of utility poles on the right side of the road 0.5 mile of utility poles spaced at 300 ft intervals (i.e., approximately 9 utility poles) • Before condition—number of utility poles on the left side of the road 0 • Distance from outside edge of traveled way to roadside tree(s) 6, 10, 15, 20, 25, 30, 35, and 40 ft • After condition—number of utility poles on the right side of the road 0 • After condition—number of utility poles on the left side of the road 0 • Lane width 12 ft • Paved shoulder width 6 ft • Curvature Tangent alignment • Advance visibility of curve Substantial • Grade 2% • Shoulder rumble strips Present • Presence of delineation Substantial • Road surface condition Good • Pavement skid resistance High Table 18. Summary of input data for Example 4—relocation of an extended series of utility poles.

32 Design Guidelines for Mitigating Collisions with Trees and Utility Poles values of 1,000, 5,000, and 10,000 vpd and the distance of the tree to be removed from the outside edge of the traveled way was 6 ft and from 10 to 40 ft in increments of 5 ft. There should be economies of scale in relocating multiple utility poles in the same vicinity. The cost of relocating utility poles along 0.5 mi of road is estimated as $30,000 based on the assumption that nine utility poles along the road, plus several additional utility poles (not neces- sarily in exposed positions on the roadside) also need to be relocated to maintain continuity of power lines. The relocated utility poles are assumed to be more than 40 ft from the outside edge of the traveled way. For the initial case analyzed for Benefit-Cost Example 4, with a roadway AADT of 1,000 vpd and the extended series of utility poles in the before condition located 6 ft from the edge of the traveled way, the predicted annual crashes estimated with the crash prediction model are as shown in Table 19. The table shows both the crash predictions obtained from the model and intermediate results as derived from Equations (1) through (15). Table 20 shows the computed crash reductions and benefit-cost ratios for each value of AADT and distance from the traveled way to roadside tree considered. The results show that relocation Variable Name Before Condition After Condition NROR 0.0116 0.0000 NROR-left 0.0000 0.0000 NROR-right 0.0116 0.0000 NROR-K 0.0002 0.0000 NROR-A 0.0009 0.0000 NROR-B 0.0047 0.0000 NROR-C 0.0057 0.0000 NROR-PK 0.0002 0.0000 NROR-PI-A 0.0009 0.0000 NROR-PI-B 0.0055 0.0000 NROR-PI-C 0.0077 0.0000 LikelihoodROR 1.0000 1.0000 SeverityROR-left 0.0000 0.0000 SeverityROR-right 17.8450 0.0000 RSSROR-left 0.0000 0.0000 RSSROR-right 2.8674 0.0000 Table 19. Computational results with crash prediction model before and after removal of an extended series of utility poles 6 ft from the traveled way of a rural 2-lane highway with an AADT of 1,000 vpd. Distance from Traveled Way to Tree (ft) Annual Crash Reduction (crashes/yr)a Benefit-Cost Ratio (B/C) Traffic Volume (AADT) (vpd) 1,000 5,000 10,000 1,000 5,000 10,000 6 0.0116 0.0541 0.0909 1.7 8.0 13.5 10 0.0099 0.0463 0.0778 1.5 6.9 11.5 15 0.0078 0.0366 0.0614 1.2 5.4 9.1 20 0.0057 0.0268 0.0450 0.9 4.0 6.7 25 0.0042 0.0197 0.0330 0.6 2.9 4.9 30 0.0032 0.0152 0.0255 0.5 2.2 3.8 35 0.0023 0.0107 0.0179 0.3 1.6 2.7 40 0.0013 0.0062 0.0104 0.2 0.9 1.5 aFor all fatal and injury crash severity levels combined. Table 20. Crash reductions and benefit-cost ratios for specific values of AADT and distance from traveled way to utility poles in Example 4.

Benefit-Cost Analysis Examples 33   of utility poles can be cost-effective, even for roads with AADTs as low as 1,000 vpd, when the utility poles are near the traveled way. Because of economies of scale in the cost of utility pole relocation, projects involving relocation of multiple utility poles appear to be more cost-effective than projects involving relocation of a single utility pole, as can be seen from comparing the results in Tables 17 and 20. 6.5 Benefit-Cost Example 5—Removal of an Extended Series of Utility Poles and Replacement with Underground Utilities Benefit-Cost Example 5 addresses removal of an extended series of utility poles on the road- side of a rural 2-lane undivided highway and their replacement with underground utilities. Table 21 summarizes the input data for the example. All of the input data are constant values except that computations were made for roadway AADT values of 1,000, 5,000, and 10,000 vpd and the distance of the tree to be removed from the outside edge of the traveled way was 6 ft and from 10 to 40 ft in increments of 5 ft. The input data for Benefit-Cost Example 5 in Table 21 are identical to the input data for Benefit-Cost Example 4 in Table 18. The only difference between the two examples is the cost of providing underground utilities, which is substantially higher than relocation of the utility poles. The cost of removing utility poles along 0.5 mi of road and replacing them with underground utilities is estimated as $1,165,000 based on the cost estimates in Section 5.7 of this report. Provi- sion of underground utilities would clear all utility poles from the 0.5-mi project site. For the initial case analyzed for Benefit-Cost Example 5, with a roadway AADT of 1,000 vpd and the extended series of utility poles in the before condition located 6 ft from the edge of the traveled way, the predicted annual crashes estimated with the crash prediction model are the same as those shown in Table 20. Input Data Element Specified Value • Roadway type Rural 2-lane undivided • Traffic volume (AADT) 1,000, 5,000, and 10,000 vpd • Mean operating speed of traffic 60 mph • Before condition—number of utility poles on the right side of the road 0.5 mile of utility poles spaced at 300 ft intervals (i.e., approximately 9 utility poles) • Before condition—number of utility poles on the left side of the road 0 • Distance from outside edge of traveled way to roadside tree(s) 6, 10, 15, 20, 25, 30, 35, and 40 ft • After condition—number of utility poles on the right side of the road 0 • After condition—number of utility poles on the left side of the road 0 • Lane width 12 ft • Paved shoulder width 6 ft • Curvature Tangent alignment • Advance visibility of curve Substantial • Grade 2% • Shoulder rumble strips Present • Presence of delineation Substantial • Road surface condition Good • Pavement skid resistance High Table 21. Summary of input data for benefit-cost Example 5—removal of an extended series of utility poles and replacement with underground utilities.

34 Design Guidelines for Mitigating Collisions with Trees and Utility Poles Table 22 shows the computed crash reductions and benefit-cost ratios for each value of AADT and distance from the traveled way to roadside utility poles considered. The results shown in Table 22 confirm the indications in the literature (see Section 5.7) that replacement of utility poles with underground utilities is often not cost-effective, but is typically justified on aesthetic grounds. Distance from Traveled Way to Tree (ft) Annual Crash Reduction (crashes/yr)a Benefit-Cost Ratio (B/C) Traffic Volume (AADT) (vpd) 1,000 5,000 10,000 1,000 5,000 10,000 6 0.0116 0.0541 0.0909 < 0.1 0.2 0.3 10 0.0099 0.0463 0.0778 < 0.1 0.2 0.3 15 0.0078 0.0366 0.0614 < 0.1 0.1 0.2 20 0.0057 0.0268 0.0450 < 0.1 0.1 0.2 25 0.0042 0.0197 0.0330 < 0.1 0.1 0.1 30 0.0032 0.0152 0.0255 < 0.1 0.1 0.1 35 0.0023 0.0107 0.0179 < 0.1 < 0.1 0.1 40 0.0013 0.0062 0.0104 < 0.1 < 0.1 < 0.1 aFor all fatal and injury crash severity levels combined. Table 22. Crash reductions and benefit-cost ratios for specific values of AADT and distance from traveled way to utility poles in Example 5.