Design Guidance for Intersection Auxiliary Lanes (2014)

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.

32 State of the Practice Introduction As part of Task 2 efforts on NCHRP Project 3-102, the research team conducted a state-of-the-practice review of cur- rent design considerations. The focus of this sub-task was to identify those practices and evaluations not adequately docu- mented in traditional literature. To accomplish that objective, the research team contacted representatives from a selection of state DOTs to inquire about current practices and potential guidance needs, based on their professional experience and the policies of their respective departments. The information presented in this chapter reflects the participants’ responses as provided to the research team. Based on the findings from the Task 1 review of pub- lished literature and of state design manuals and guidance documents, the research team developed a preliminary list of topics for further study. Researchers then enlisted the help of the project panel, whose feedback guided the refinement of that list into a shorter list of three main top- ics: the design of deceleration lanes, the design of multi- ple turn lanes, and the design of right-turn islands. These three topics were the focus of the state-of-the-practice review. Researchers conducted this review through the use of a written questionnaire, distributed to the aforementioned DOT personnel. Researchers developed the questionnaire with additional review and feedback by the project panel and one additional DOT representative who did not for- mally complete the questionnaire but offered suggestions for improvements to the document. Using the finalized ques- tionnaire, researchers contacted representatives from DOTs in 12 states; these states were selected because they satisfied one or more of the following criteria: • The DOT’s existing design guidance identified in Task 1 contained information on the design of multiple turning lanes. • The DOT’s existing design guidance identified in Task 1 contained information on right-turn island design. • The state was a participant in Highway Safety Information System (HSIS). • The state was known by researchers and/or panel mem- bers to have experience with one or more of the focus topics. The selected states were California, Colorado, Connecticut, Florida, Illinois, Maine, Maryland, Minnesota, North Caro- lina, Ohio, Texas, and Washington. Researchers used the AASHTO Subcommittee on Design and Geometric Design Technical Committee member rosters to identify the initial list of representatives to contact in each of the selected states; this was supplemented by a search of DOT websites to find specific contact information. For Cali- fornia and North Carolina, researchers contacted members of the project panel instead of AASHTO subcommittee mem- bers. Researchers contacted all representatives via e-mail and included the questionnaire document as an attachment. In the e-mail, representatives were asked to respond initially to verify their willingness to participate and to state whether they preferred to complete the questionnaire by e-mail or by telephone interview. In many cases, the contacted representatives either for- warded the questionnaire to others in their office or enlisted the help of others to complete the questionnaire as a group. Indications from respondents were that staff members from Traffic Engineering or other divisions were commonly involved with auxiliary lane installations along with those who worked on roadway geometric design. Completed questionnaires were received from 11 of the 12 states contacted; no response was received from Con- necticut. For each responding state, the completed ques- tionnaire was returned via e-mail, although two respondents also asked to be contacted by phone to discuss the questions C H A P T E R 3

33 and ensure that their understanding was consistent with researchers’ expectations. The remainder of this chapter contains the responses to the questionnaires, with summary comments to note trends and patterns in the respondents’ answers. The summary tables and associated text presented in the remainder of this chapter are restatements of the responses provided by the survey partici- pants to indicate a current practice, as intended by the design of the survey. The research team makes no statement about the appropriateness of any of the policies or practices mentioned. Questionnaire Responses Deceleration Length D-1. Using the figure below, how do you define a “deceleration lane” for a left- or right-turn lane at an at-grade inter- section? Please note all of the dimensions that apply to your definition. included the dimension (c), but only 3 of the 11 respondents indicated that (c) was their definition of a deceleration lane. Two states formally used the term “deceleration length” or “deceleration distance” instead; they defined that length as the sum of (b) and (c). Other states included both the taper and the full-width portion of the lane (i.e., b plus c) as their decelera- tion distance. Table 3-1 summarizes all answers to this question. D-2. Briefly summarize your understanding of your agency’s policy or guidance for the design of a deceleration lane. Respondents’ answers typically focused on the criteria for determining whether a lane should be installed and/or how to determine the length. For the latter category, the speed of approaching vehicles was most frequently mentioned, listed either directly or implicitly by nine respondents. This includes three specific mentions of the AASHTO Green Book, which uses design speed as the criterion for deceleration length. Other respondents mentioned entry speed, design speed, or posted speed. One respondent also specifically stated that “accepted practice is for some deceleration (10 mph) to be assumed to occur in the through lane and to consider the taper as part of the deceleration lane.” To determine both the need for a deceleration lane and its appropriate length (for deceleration and/or for stor- age), a capacity or demand analysis was mentioned by five respondents. Two also included consideration of restricted geometrics or crash issues as potential installation criteria. One respondent added discussion of appropriate widths for medians, left-turn lanes, and right-turn lanes. Table 3-2 summarizes the responses received. Some respon- dents provided more than one criterion, so sums may exceed the number of overall responses (n = 11). (a) (c)(b) (d) Answer Frequency (n=11) (c) 3 (b)+(c)+(d) 3 (b)+(c) 4 (c)+(d) 1 Table 3-1. Summary of responses to Question D-1. This question was asked to determine whether the respon- dents were considering the same definition of “deceleration lane.” While deceleration can take place in any of the dimen- sions shown in the figure, the dimension (c) is the formal design element included for deceleration of turning vehicles for purposes of this research. Dimension (a) is part of the through travel lane, (b) is the lane-addition taper, and (d) is the queue storage length. Perhaps verifying the prudence of including this question, researchers received various responses. All of the responses Criteria for Determining Installation and/or Length Frequency (n=11) AASHTO Green Book (2001 or 2004) 5 Capacity 3 Entry speed 3 Restricted geometrics 2 Crash issues 2 Design speed 2 Posted speed 1 Arrival demand 1 Storage demand 1 Table 3-2. Summary of responses to Question D-2.

34 D-3. On which of the following is your agency’s policy based? (Please note all that apply.) a. AASHTO Green Book. b. NCHRP Report 279 (Intersection Channelization Design Guide). c. Harmelink’s procedures. d. TRB Access Management Manual. e. Other (please specify). Respondents were free to list as many sources as needed, and there were multiple combinations of responses. Altogether, the Green Book was cited by seven respondents as at least one source they use, the most common of the specific source choices provided, followed closely by NCHRP Report 279 (five times). However, the “other” category was the most frequently cited (eight times), and some agencies use more than one “other” source in their policy. Table 3-3 lists the responses to this question. “Other” sources cited include • AASHTO Guide for the Development of Bicycle Facilities. • AASHTO Roadside Design Guide. • California Highway Design Manual. • Colorado State Highway Access Code Manual. • FHWA Manual on Uniform Traffic Control Devices for Streets and Highways. • Florida DOT Project Traffic Forecasting Handbook. • Florida DOT Plans Preparation Manual. • Florida DOT Intersection Design Guide. • Florida DOT report Triple Left Turn Lanes at Signalized Intersections. • Ohio Location and Design Manual. • TRB Highway Capacity Manual. D-4. If constraints prevent you from installing the preferred length of deceleration lane, how do you decide which ele- ments are reduced? Is your decision based on a quantita- tive or qualitative analysis? The typical response to this question was that the deci- sion is a qualitative one based on the conditions at the site. Respondents indicated that many factors (e.g., available right-of-way, distance between adjacent intersections, and signal timing) could affect how the available deceleration lane length is allotted. Four respondents stated that they would initially assume that deceleration occurred in the through lane and would correspondingly reduce the decel- eration length of the auxiliary lane, while one stated that the taper length would be shortened. One other respondent stated that the agency’s minimum turning lane length is 100 ft but did not specify how that might be divided among the various design elements. As the previous answers might suggest, this decision is fre- quently qualitative rather than quantitative; six respondents directly answered that this was the case. Responses to this question are summarized in Table 3-4. D-5. Is there guidance on designing deceleration lanes not cur- rently included in your state policy or in the AASHTO Green Book that would be useful? If so, what? Respondents were generally satisfied with the guidance contained in their state manuals and in the AASHTO Green Book. Two respondents offered suggestions for additional material, one of which provided suggestions on various top- ics already contained within the state’s guidance documents. Seven participants responded that current guidance is ade- quate, and one provided no answer to the question. Responses are summarized in Table 3-5. Answer Frequency (n=11) (e) 2 (a) 1 (a)+(b) 1 (a)+(d) 1 (a)+(e) 1 (b)+(e) 1 (a)+(b)+(e) 2 (b)+(c)+(d)+(e) 1 all 1 Table 3-3. Summary of responses to Question D-3. Element(s) Reduced Frequency (n=11) Depends on existing site conditions 5 Deceleration length (i.e., deceleration occurs in through lane) 4 Taper 1 Minimum turn lane length is 100 ft 1 Nature of Decision Frequency (n=11) Qualitative 6 Quantitative 1 Could be either/Varies 1 Unknown 1 No answer 2 Table 3-4. Summary of responses to Question D-4.

35 D-6. Please list up to three locations in your state with installa- tions that you would consider “best practice” sites. Respondents from five states provided a total of 10 locations or corridors for consideration. The sites are listed in Table 3-6. D-7. Do you have sites that are less than optimal? If so, what improvements do you consider needed and why? Six respondents indicated that there were sites they would like to improve; most spoke generally, though one respondent provided locations of some specific examples. The descrip- tions of the sites in each state were similar: turning lanes were not long enough and/or they were not wide enough. Lack of available right-of-way was specifically mentioned as a causal factor by two respondents. Lengthening the turning lane to improve capacity was a desired improvement listed by five respondents, while one would like to have wider lanes in some locations. Also, one respondent listed sight distance for opposing left-turn lanes as a common problem that could be improved through offsetting the turning lanes. Multiple Turn Lanes Double Left-Turn Lanes 2L-1. Does your agency’s policy include guidelines for when it is appropriate/recommended to install a second left-turn lane? Only 1 of the 11 respondents indicated that his/her agency’s policy did not have specific guidelines. Two of the remaining ten respondents provided some details in addition to their affirmative response. One stated that the recommendation to construct multiple turn lanes is based on the results of traffic studies performed during project development and later refined during preliminary design. The other stated that the decision was largely specific to the location or the demand, and that all double left-turn lanes are signalized with protected phasing; considerations included left-turn demand (especially above 300 vehicles per hour [veh/hr]), capacity, storage, and safety. This respondent also indicated that sometimes intersections have an ultimate geometry that includes a double left-turn lane but short-term fore- casts have volumes that are not yet high enough to justify more than one lane, which requires intermediate design solutions. 2L-2. If so, briefly summarize your understanding of your agency’s policy or guidance for the design of double left- turn lanes. The 11 respondents typically discussed the criteria that should be met to install a double left-turn lane, and those cri- teria were usually a turning volume that exceeded 300 veh/hr and/or the inability to provide a single left-turn lane that had sufficient length or level of service (LOS) to meet demand. Several respondents specifically mentioned that an analy- sis of operational characteristics (usually through micro- simulation) was conducted to determine both current and long-term needs. Three respondents specified that double left- turn lanes always operated under fully protected signalization. Additional Guidance Suggested Frequency (n=11) How to develop projected turning movements (and corresponding storage lengths) 1 Type of widening (e.g., symmetrical, pocket, and near-side) for left-turn treatments, left-turn warrants, left overs, and super streets (for Green Book) 1 Findings from state research projects are added to the guidance if applicable 1 None (i.e., current guidance is adequate) 7 No answer 1 Table 3-5. Summary of responses to Question D-5. State City/County Site Colorado (not specified) SH 391 (MP 0.97)/SH 8 Colorado (not specified) SH 121 (MP 6.53)/W. Quincy Ave Florida West Palm Beach State Road 7 @ Old Hammock Way Florida Tallahassee Capital Circle SE @ Tram Road Florida Tampa State Road 597 @ Stall Road Maine Winthrop Route 202 and Main Street Maine Winthrop Route 202 and Highland Avenue Minnesota (not specified) “Various turn lanes along Trunk Highway 371 between Little Falls and Baxter which incorporated 500-ft long bay lengths” North Carolina Fuquay Varina/ Wake County W. Academy St. at Coley Farm Rd North Carolina Cary Westbound SR 1010 (Ten Ten Road) at Kildaire Farm Road Table 3-6. Summary of responses to Question D-6.

36 Table 3-7 summarizes the responses to this question; respon- dents often had more than one component to their answers, and all responses are listed, so the summation is greater than the number of respondents. 2L-3. Does your agency’s policy discuss the design of receiving lanes on the departure of the double left-turn lane? If so, briefly summarize your understanding of your agency’s policy. Most of the survey respondents (8 of 11) indicated that their policy contained some specific guidance for receiving lanes. Typical values were 30 ft minimum and 36 ft desired width for the turning path or turning curve, which would also be the total width of the receiving lanes. Three respon- dents mentioned that when designing for trucks (e.g., WB-40 or larger), their turning curve was designed for an SU design vehicle on the inside lane and a larger design vehicle on the outside lane, rather than simultaneous turning of two large vehicles. Two respondents discussed the transition from the full width of the turning curve down to the normal lane width; one stated that the turning width was tapered down to the normal width at a ratio of 15:1 to 20:1, while the other stated that the full-width receiving lane should extend for a minimum of 1,500 ft. Table 3-8 summarizes the responses to this question. 2L-4. What is the estimated capacity of a typical double left- turn lane? a. Twice the capacity of a single left-turn lane b. 1.8 times the capacity of a single left-turn lane c. 1.5 times the capacity of a single left-turn lane d. Other (please specify) Different values have been proposed for this variable, so the research team wanted to determine what was typical based on the practices and experiences of the participating agencies. The most common answers were (b) and (d), with three responses each. Within (d), the respondents said that they determined the value site by site with a traffic or opera- tional study and/or the HCM methodology. Two respondents answered (c), and one respondent answered (a). Responses are summarized in Table 3-9. 2L-5. Compared to a typical single left-turn lane, what adjustments do you make to queue storage length, deceleration length, and/or taper length for a double left-turn lane? Respondents provided various answers to this question. Answers categorized by the three design elements are sum- marized in Table 3-10. 2L-6. Is there guidance on designing double left-turn lanes not currently included in your state policy or in the AASHTO Green Book that would be useful? If so, what? Respondents were typically satisfied with the guidance contained in their state manuals and in the AASHTO Green Book. Five respondents offered suggestions for additional material, one of which provided suggestions on various topics already contained within the state’s guidance documents. Five participants responded that current guidance is adequate, Installation Criteria/Design Feature Frequency (n=11) Consider when turning volume exceeds 300 veh/hr 6 Not possible or practical to provide single lane of sufficient length or LOS will not be met 3 Conduct evaluation of operational characteristics 3 Fully protected (green arrow) signal phasing 3 Design elements similar to single left, but with additional clearance distance to accommodate second lane 1 Consider for turning volume of 200-600 veh/hr 1 Major signalized intersections where high peak-hour left-turn volumes are expected 1 Table 3-7. Summary of responses to Question 2L-2. Typical Turning Curve Width Frequency (n=11) 36 ft (desirable); 30 ft (minimum) 4 Two receiving lanes of 15 ft each 1 35 ft 1 26-45 ft, depending on radius and percentage of trucks 1 “Full width” receiving lanes 1 No width specified 3 Table 3-8. Summary of responses to Question 2L-3.

37 Storage Length Frequency (n=11) Site-by-site/traffic analysis 3 According to HCM 2 0.6 times single lane 1 Enough to avoid through vehicles blocking entrance 1 No guidance available 1 No answer 3 Deceleration Length Frequency (n=11) Sum of storage and taper 2 Based on 15:1 taper (4:1 minimum) 1 Site-by-site/traffic analysis 1 No guidance available 1 No answer 6 Taper Length Frequency (n=11) Twice single lane 2 Same taper rate as single lane 1 Site-by-site/traffic analysis 1 15:1 taper (4:1 minimum) 1 Extended by 50 ft 1 No guidance available 1 No answer 4 Table 3-10. Summary of responses to Question 2L-5. Answer Frequency (n=11) (b) 3 (d) with traffic study/HCM 3 (c) 2 (a) 1 No answer 2 Table 3-9. Summary of responses to Question 2L-4. and one provided no answer to the question. Responses are summarized in Table 3-11. 2L-7. Please list up to three locations in your state with dou- ble left-turn installations that you would consider “best practice” sites. Respondents from six states provided a total of 14 locations or corridors for consideration. The sites are listed in Table 3-12. 2L-8. Do you have double left-turn sites that are less than opti- mal? If so, what improvements do you consider needed and why? Eight respondents provided suggestions for improvements; most spoke generally, although one respondent provided locations of some specific examples. The descriptions of the improvements in each state were similar: longer taper lengths, longer storage lengths, wider lanes, and better guidance to prevent offtracking. Triple Left-Turn Lanes 3L-1. Does your agency’s policy include guidelines for when it is appropriate/recommended to install a third left-turn lane? Only two respondents indicated that their state’s policy included guidance for triple left-turn lanes, though two others Additional Guidance Suggested Frequency (n=11) Add a design chart that shows the left-turn lengths for two left-turn lanes, along with taper lengths, as a function of design speeds (for Green Book) 1 Guidance on designing double left-turn lanes will be further developed in future publications of the state manual and will be presented in the agency’s intersection design class 1 For access control, use a barrier curbed median adjacent to the left-turn lanes and the approaching median taper 1 Directional Left Over and Super Street typical designs included in state Roadway Design Manual (for Green Book) 1 Different sets of bay taper lengths for urban and non-urban areas, and procedure to select the proper signal phasing sequence for intersections with left-turn lane overflow and/or blockage problems (based on state research project) 1 None (i.e., current guidance is adequate) 5 No answer 1 Table 3-11. Summary of responses to Question 2L-6.

38 suggested that there might be conditions under which a site- specific analysis of conditions would produce a triple left-turn as an alternative. The remaining respondents did not have such guidance and, in fact, often suggested that such treatments were discouraged or prohibited, typically because of truck traffic. 3L-2. If so, briefly summarize your understanding of your agency’s policy or guidance for the design of triple left- turn lanes. The two affirmative responders in Question 3L-1 stated that they both referred to guidance by the Florida DOT. The two conditional responders both stated that they reviewed intersections on a site-by-site basis. 3L-3. Does your agency’s policy discuss the design of receiving lanes on the departure of the triple left-turn lane? If so, briefly summarize your understanding of your agency’s policy. One affirmative respondent provided these criteria: • Need to design so the relative turning movement distribu- tion at a downstream intersection does not compromise the ability of the receiving lanes to store the left-turning vehicles. • Three downstream lanes need to be available to receive the left-turning traffic for at least 300 ft from the inter- section, and at least two continuous downstream lanes must exist beyond that point. If possible, continuous downstream receiving lanes should be provided to avoid a lane drop. • The receiving leg should have a raised median island at least 2 ft wide to provide drivers on the inside lane with a visual point of reference to guide the vehicle through the left-turn maneuver. The other affirmative respondent stated that they specify 15-ft lanes at the throat, dotted lines for guidance through the turn, and a minimum width of 15 ft at the center of the curve. One conditional respondent stated that his/her agency’s goal in conducting microsimulation analysis was to opti- mize the amount of width on the receiving lanes to accom- modate the design vehicle in each turn lane individually but not simultaneously. The other site-specific response emphasized lanekeeping and lane continuity issues and concerns. 3L-4. Compared to a typical single left-turn lane, what adjust- ments do you make to queue storage length, deceleration length, and/or taper length for a triple left-turn lane? One affirmative respondent stated that the taper length would simply be the additional length created by the same taper rate used for the other lanes and all other design ele- ments would be determined after analysis of site-specific con- ditions (e.g., geometry, right-of-way, and traffic). The other three respondents offered similar comments. 3L-5. Is there guidance on designing triple left-turn lanes not currently included in your state policy or in the AASHTO Green Book that would be useful? If so, what? The two affirmative respondents suggested material from Florida. One recommended a 2002 FDOT report entitled “Triple Left Turn Lanes at Signalized Intersections,” while the other suggested that existing FDOT guidelines would be fur- ther developed in future editions of that state’s policies and design classes. One respondent with no current agency policy advised that inclusion of basic parameters and other control criteria would be helpful to designers, and another referred to a DOT-sponsored research project. State City/County Site Colorado (not specified) SH 391 (MP 0.97)/SH 8 Colorado (not specified) SH 391 (MP 4.23)/W. Alameda Ave Colorado (not specified) SH 95 (MP 12.09)/W. 72nd Ave Florida West Palm Beach State Road 7 @ State Road 80 Florida Tallahassee Capital Circle SE @ Blair Stone Road Florida Tampa State Road 580 @ Waters Avenue Maine Augusta Route 202 (Western Ave) and Armory Street Maine Augusta Whitten Road at Route 202 Maine Portland Route 1 at the Route 703 Connector Minnesota Roseville SB Trunk Highway 51 at County Road B North Carolina Leland/Brunswick County US-17 SuperStreet Corridor North Carolina Cary Tryon Road at Kildaire Farm Road North Carolina Chapel Hill/Orange County US 15-501 and Erwin Road Washington Olympia Black Lake Blvd for both the left-turn channelization SB for EB or SB 101 ramp Table 3-12. Summary of responses to Question 2L-7.

39 3L-6. Please list up to three locations in your state with tri- ple left-turn installations that you would consider “best practice” sites. Respondents from four states provided a total of eight locations or corridors for consideration. The sites are listed in Table 3-13. 3L-7. Do you have triple left-turn sites that are less than opti- mal? If so, what improvements do you consider needed and why? One state’s respondent indicated that there may be less- than-optimal sites in operation; needed improvements might include additional storage due to increases in traffic volumes that have exceeded expectations. One state offered specific examples of sites that could be improved. Double Right-Turn Lanes 2R-1. Does your agency’s policy include guidelines for when it is appropriate/recommended to install a second right- turn lane? Four of eleven respondents stated that they had guidance for installation of double right-turn lanes; two of them also indicated that the guidelines were similar to those for double left-turn lanes. 2R-2. If so, briefly summarize your understanding of your agency’s policy or guidance for the design of double right- turn lanes. Nine participants provided responses to this question, indicating that, even if they do not have specific guidance in their agencies’ policies, relevant guidelines in place can be applied if needed. Table 3-14 summarizes the responses to this question. 2R-3. Does your agency’s policy discuss the design of receiving lanes on the departure of the double right-turn lane? If so, briefly summarize your understanding of your agency’s policy. Five respondents stated that their policies discussed receiv- ing lanes; of those five, three stated that the guidance was similar to that for double left-turn lanes. One stated simply that free right turns were discouraged, and one said that the design was somewhat contradictory because it challenged the designer to accommodate large trucks while still trying to minimize radii for intersection and clearance widths. 2R-4. Compared to a typical single right-turn lane, what adjust- ments do you make to queue storage length, deceleration length, and/or taper length for a double right-turn lane? While double right-turn lanes were not that common among the respondents, almost all of them offered a response as to Policy Highlights Frequency (n=11) Site-by-site analysis 3 Length is the same as double left-turn lanes 1 Mirror image of double left-turn lanes 1 Similar to double left-turn lanes; consider priority of pedestrian movements, use channelizing islands, and discourage right-turn on red 1 Double right-turns are discouraged, but they can be considered when a single turning lane does not have adequate length, storage, or level of service 1 Increase turning radius to 200 ft 1 Rarely used, but would require larger radius and throat width 1 No answer/Not applicable 2 Table 3-14. Summary of responses to Question 2R-2. State City/County Site Colorado (not specified) SH 128 (MP 7.97) Colorado (not specified) E. Orchard Road @ SH 83 (MP 66.56) Colorado (not specified) S. Chambers Road @ SH 83 (MP 66.92) Florida Lake Worth State Road 7 @ State Road 802 Florida Wellington State Road 7 @ State Road 882 Florida West Palm Beach State Road 7 @ State Road 704 Minnesota Minneapolis The southbound exit ramp from I-494 to Trunk Highway 62— southbound to eastbound turning movement Washington (not specified) Southbound Interstate 5 and Eastbound State Route 512 Table 3-13. Summary of responses to Question 3L-6.

40 appropriate adjustments to make. Answers are summarized in Table 3-15. 2R-5. Is there guidance on designing double right-turn lanes not currently included in your state policy or in the AASHTO Green Book that would be useful? If so, what? Five respondents offered suggestions for additional guid- ance, as shown in Table 3-16. 2R-6. Please list up to three locations in your state with double right-turn installations that you would consider “best practice” sites. Respondents from four states provided a total of seven locations or corridors for consideration. The sites are listed in Table 3-17. 2R-7. Do you have double right-turn sites that are less than opti- mal? If so, what improvements do you consider needed and why? Five agencies responded to this question; the common thread was the need for improved storage, along with possible deceleration length, taper length, and/or pedestrian accommo- dation. One agency described two specific sites, while others did not know of exact locations. Island Design for Right-Turn Lanes I-1. Does your agency’s policy include guidelines for when it is appropriate/recommended to install a channelizing island at a right-turn lane? Ten of the eleven respondents stated that they have avail- able guidance on channelizing islands in one or more of their policy/guidance documents. The remaining respondent indicated that his/her agency has some guidance in an Access Permit Manual, although it was primarily used by developers and not agency designers. I-2. If so, briefly summarize your understanding of your agen- cy’s policy or guidance for the design of a channelizing island adjacent to a right-turn lane. State City/County Site Colorado (not specified) SH 83 (MP 65.31)/E. Arapahoe Rd Florida Wellington State Road 7 @ State Road 882 Florida West Palm Beach State Road 7 @ State Road 704 Florida Tallahassee State Road 20 @ State Road 373 Maine Auburn Veterans Bridge Connector at Route 4 Maine Topsham 295 NB off ramp at Route 196 Washington Olympia Double-right from ramp terminal for WB or NB 101 to Black Lake Blvd Table 3-17. Summary of responses to Question 3L-6. Adjustments Frequency (n=11) Similar to double left-turn lanes 4 Site-by-site analysis 2 Half the storage length 1 Per HCM 1 No guidance available 1 Not applicable/No answer 2 Table 3-15. Summary of responses to Question 2R-4. Additional Guidance Suggested Frequency (n=11) General guidance, with emphasis on pedestrian considerations 1 Design chart that shows the right-turn lengths for two right-turn lanes, along with taper lengths, as a function of design speeds; develop and illustrate warrants for double right-turn lanes 1 Describe conditions and solutions for when the capacity of a double right-turn is exceeded 1 Additional design considerations that must be addressed for successful operation when changing from single to double turning lane 1 Findings from DOT-sponsored research project “Development of Guidelines for Triple Left and Dual Right Turn Lanes” 1 None (i.e., current guidance is adequate) 5 No answer 1 Table 3-16. Summary of responses to Question 2R-5.

41 Respondents provided a wide variety of answers, reflec- tive of the open-ended nature of the question. Responses are summarized below: • Traffic islands can be beneficial in separating lanes of traffic, directing a traffic stream toward a specific direction, or providing pedestrian refuge areas. Traffic islands can also be used to discourage/prohibit certain movements. • Islands are included in intersection design for the follow- ing purposes: – Separating conflicts. – Control of angle of conflict. – Reducing excessive pavement areas. – Regulating traffic and indicating proper use of inter- section. – Arrangements to favor a predominant turning movement. – Protecting pedestrians. – Protecting and storing turning and crossing vehicles. – Locating traffic control devices. – Access control. The size of the islands should be a minimum of 50 sq ft in urban areas, and 75 sq ft in rural areas; 100 sq ft mini- mum preferred in all areas. Triangular islands should be a minimum of 12 ft (preferably 15 ft) on a side after round- ing of corners. Islands shall be delineated, depending on size. Design Manual references triangular island layouts in Chapter 9 (Intersections) (pages 634–639) of the 2004 Green Book. • Policy is based on geometrics. At larger intersections where large turning radii are required due to vehicle mix, chan- nelizing islands may be appropriate. In locations where mast arms may be required due to proximity to the coast, channelizing may be used to site mast arm poles. • Provide a smooth, free-flowing alignment both into and out of the channelization. • Turning roadways are channelized areas (separated by an island) which allow a moderate-speed, free-flowing right turns away from the intersection area. The designer should consider using turning roadways when 1. It is desirable to allow right turns at speeds of 15 mph or more; 2. The angle of turn is greater than 90°; 3. The volume of right turns is high, the turning move- ment is from a high-volume road, or it is desirable to remove right turns away from a signal; 4. It is desirable to reduce the intersection paved area. As a guide, if an island with a turning roadway will be at least 75 square ft (urban) or 100 square ft (rural), then a turning roadway should be considered; and/or 5. Pedestrian volumes are high and a pedestrian refuge is a desirable feature. • There are two conditions for using a channelized right-turn lane. One would be a free right-turning movement and the other is when we are trying to protect the right-turning movement from last-minute right-turn lane encroachment at an intersection which would negatively affect the through movement in an intersection. • They may be warranted in the following situations: 1. When it is necessary to accommodate semitrailer or large buses. 2. When intersections are skewed. 3. When it is desirable to allow right turns at speeds of 15 mph or more with an acceleration lane. 4. When it is necessary to accommodate traffic control signal installation. • Complete Street efforts under development and improved curb ramp designs are addressing/providing improved right-turn island designs. Most locations where this will be pursued will be retrofit intersections or locations with new development improvements with pedestrian, modal, and bicycle access. Traditional approach was mobility- based and aimed at free flow or yield optimal conditions; the evolving guidance we are using is more modal-sensitive and improves vehicular alignment for a lower speed but hopefully safety approach and turn in a multi-modal environment. • Channelizing islands control and direct traffic into the proper paths for the intended use and are an important part of intersection design. They may be of many shapes and sizes, depending on the conditions and dimensions of the intersection. A common form is the corner triangu- lar shape that separates right-turning traffic from through traffic. Channelizing islands are used at intersections for the following reasons: – Separation of conflicts. – Control of angle of conflict. – Reduction in excessive pavement areas. – Indication of proper use of intersection. – Favoring of a predominant turning movement. – Pedestrian protection. – Protection and storage of vehicles. – Location of traffic control devices. These islands should be placed so that the proper course of travel is immediately obvious and easy for the driver to fol- low. Care should be given to the design when the island is on or beyond a crest of a vertical curve, or where there is a sub- stantial horizontal curvature on the approach to or through the channelized area. Properly placed islands are advanta- geous where through and turning movements are heavy.

42 • Where the inner edges of pavement for right turns at inter- sections are designed to accommodate semitrailer combina- tions or where the design permits passenger vehicles to turn at 15 mph or more (i.e., 50 ft or more radius), the pavement area at the intersection may become excessively large for proper control of traffic. In these cases, channelizing islands should be used to more effectively control, direct, and/or divide traffic paths. Physically, islands should be at least 50 ft2 in urban and 75 ft2 for rural conditions (100 ft2 preferable for both) in size and may range from a painted to a curbed area. • Traffic islands perform the following functions: – Channelization islands control and direct traffic movements. – Divisional islands separate traffic movements. – Refuge islands provide refuge for pedestrians and bicyclists crossing the roadway. – Islands can provide for the placement of traffic control devices and luminaires. – Islands can provide areas within the roadway for land- scaping. • Location of traffic islands can help reduce wrong-way movements at intersections with freeway ramp terminals, especially side-by-side on/off ramps. Intersection islands also have pedestrian accessibility requirements that must be met in design. I-3. When you use a channelized right-turn lane without an acceleration lane, which of the designs in [Figure 2-7] is more common? Responding state agencies were almost evenly split on this answer, with six agencies stating that the design on the left of Figure 2-7 was their preference, and four indicating that the design on the right was more common. One agency did not answer this question. a. What are typical site characteristics that led you to select the design you chose? Characteristics that respondents identified in response to this question varied, with vehicle and pedestrian volumes, intersection size, and available sight distance being the most common. All of the provided characteristics, and the fre- quency with which they were stated, are listed in Table 3-18. Respondents could choose as many characteristics as they wished, so the sum of the frequencies is greater than the 10 agencies that actually answered Question I-3. b. What are the site characteristics that would lead you to select the other design? Responses to this question were similar to those of the previous question. Responses, and the frequency with which they were provided, are listed in Table 3-19. I-4. Is there guidance on designing channelizing islands at right-turn lanes not currently included in your state policy or in the AASHTO Green Book that would be useful? If so, what? One agency representative suggested that information on the interaction of right-turn channelization with regard to the impact on pedestrians would be useful. Another respondent said that his/her agency does not have any sug- gestions for additional guidance, but the bureau respon- sible for snow removal is opposed to channelization for reasons that have nothing to with the need for channel- ization. The remaining agencies either had no suggestions (seven respondents) or did not answer the question (two respondents). I-5. Please list up to three locations in your state with installa- tions that you would consider “best practice” sites. Respondents from three states provided a total of 10 loca- tions for consideration. The sites are listed in Table 3-20. Site Characteristic Frequency (n=10) Traffic volume 4 Pedestrian accommodation 4 Sight distance for turning drivers 3 Large intersection 3 Design speed 2 Turning radius 2 Intersection angle 2 Available right-of-way 2 Types of vehicles using the intersection 1 Receiving lanes 1 Downstream acceleration lane 1 Signal control 1 Location of traffic control devices 1 No answer 1 Table 3-18. Summary of responses to Question I-3a. Site Characteristic Frequency (n=10) Traffic volume 3 Sight distance for turning drivers 3 Pedestrian accommodation 2 Design speed 2 Intersection angle 2 Receiving lanes 2 Insufficient right-of-way 1 Types of vehicles using the intersection 1 Large intersection 1 Turning radius 1 Downstream acceleration lane 1 Upstream deceleration lane 1 Signal control 1 None/No answer 2 Table 3-19. Summary of responses to Question I-3b.

43 I-6. Do you have sites that are less than optimal? If so, what improvements do you consider needed and why? Five respondents offered ideas for improvements based on their experiences. Those responses follow: • Increase the turning radius to accommodate WB-67. • Pedestrian treatments where a raised island may not be practical. Currently reviewing shape of islands for Elder Road User program. • Improve island configuration, sight distance, intersection angle, lane tapers, and acceleration lane. • Some sites constructed under a previous standard design (using a yield-condition design similar to Figure A) in Question I-3 with a 1:15 closing taper rate on the receiv- ing roadway) have some issues with rear-end crashes. They are typically being refitted with newer standard designs upon reconstruction, but the crash frequency has typically not been high enough to warrant stand- alone projects. • Some sites with concerns about vehicle/pedestrian inter- action. These are reviewed to determine what is causing the concern and if improvements can be made, such as improving the sight distance or other features. State City/County Site Colorado (not specified) SH 391 (MP 0.97)/SH 8 Colorado (not specified) SH 391 (MP 4.23)/W. Alameda Ave Colorado (not specified) SH 119 (MP 59.04)/E. 3rd Ave Florida Tallahassee Appalachee Parkway @ Capital Circle NE Florida Tampa State Road 60 @ US 92 Florida West Palm Beach State Road 80 @ Military Trail Maine Winthrop Route 202 and Main Street Maine Winthrop Main St. at Routes 41 & 133 Maine Freeport Route 136 at Route 125 Maine Freeport Route 1 and Desert Road Table 3-20. Summary of responses to Question I-5.