4 - Recommendation on the use and delay-inducing traffic traffic calming devices for 17th St Bikelane Project
CITY OF BOULDER
TRANSPORTATION ADVISORY BOARD AGENDA ITEM
MEETING DATE: November 25, 2002
Agenda Item Preparation Date: November 15, 2002
SUBJECT:
Public hearing and consideration of a recommendation to City Council on the consideration of
the use of delay-inducing traffic-calming devices as part of the 17th Street bike lane project
design process.
REQUESTING DEPARTMENTS:
Public Works Department
Tracy Winfree, Director of Public Works for Transportation
Mike Sweeney, Transportation Planning and Operations Coordinator
Bill Cowern, Transportation Operations Engineer
Teresa Spears, Neighborhood Traffic Mitigation Program Liaison
Fire Department
Larry Donner, Fire Chief
Steve Stolz, Deputy Fire Chief
Police Department
Mark Beckner, Police Chief
Jim Hughs, Deputy Police Chief
Tom Wickman, Commander of Police Traffic Unit
BOARD ACTION REQUESTED:
Staff requests a recommendation from the Transportation Advisory Board (TAB) on whether
delay-inducing traffic-calming devices may be considered in the development of the 17`h street
bikelane project.
FISCAL IMPACT:
To be determined throw h public in gut and the design process.
PURPOSE:
The purpose of this memorandum is to receive a recommendation from the Transportation
Advisory Board (TAB) concerning whether city staff may consider the use of delay-inducing
traffic-calming devices in the development of the proposed 17`h street project, which includes the
Neighborhood Traffic Mitigation Program (NTMP) neighborhoods of Goss/Grove and Hillside.
Recommendations from staff and the TAB will be forwarded to City Council for action.
BACKGROUND
The portion of 17th Street, from University Avenue to Walnut Avenue, was included in the
2001/2002 street overlay program. Due to roadway impacts and probable traffic diversion from
the Broadway project, the overlay of this portion of 17`h Street was postponed until 2003.
Agenda Item#
When considering the overlay of 17'h Street and the resulting restriping of the street, staff
identified that the Transportation Master Plan - Bicycle System Plan proposes bicycle lanes be
added to 17`h Street, from Athens Street to Canyon Boulevard. To do so within the existing street
width would require eliminating parking on one side of the street.
In addition, the Goss/Grove neighborhood (17`h Street from Canyon Boulevard to Arapahoe
Avenue) and the Hillside neighborhood (17`h street from Athens Street to where 17'h Street
curves into University Avenue) are current participants in the NTMP.
Sound judgment and fiscal responsibility dictates that an integrated planning approach be taken
in considering the proposed bike lanes and the traffic mitigation request at the same time. The
neighborhoods are already in the NTMP, and would be addressed as they rose in program
priority.
In 2000, a new set of guidelines was adopted for the NTMP. These guidelines provided a flow
chart detailing the process by which a neighborhood would enter the NTMP as well as the order
in which different mitigation methods and process steps would be taken. This flow chart is
provided as Attachment A. The new guidelines placed more emphasis on public involvement
and provided check-in-points with both TAB and City Council on decisions pertaining to the use
of delay-inducing traffic-calming devices on Critical Emergency Response Routes (CERR) and
non-CERR streets outside of the six-minute emergency response time zone.
This portion of 17`h Street is a CERR. NTMP guidelines require that we obtain the TAB's
recommendation and city council direction, regarding the potential consideration of the use of
delay-inducing traffic-calming devices.
PUBLIC PROCESS
On Oct. 29, 2002, staff met with residents in the project area, including representatives from both
the Hillside and the Goss/Grove neighborhood. To notify the public of the meeting, staff used a
mailing list of approximately 700 property owners and residents along 17'h Street corridor. The
meeting was also advertised on the Public Works Web site and announced in the Daily Camera.
High traffic volumes, speeding traffic and the need for TMP-recommended bicycle facilities were
discussed at this meeting. Residents in attendance unanimously agreed it was very important to
be able to consider delay-inducing traffic-calming devices in the development of both of the
neighborhoods' proposed traffic mitigation plans. In addition to the people who attended the
meeting, staff received a-mails indicating support for the use of delay-inducing traffic mitigation
devices in both the Goss/Grove and Hillside neighborhoods on 17`h Street.
ANALYSIS
Staff's analyses have shown that delay-inducing physical mitigation is more effective at reducing
speeding traffic then non-delay inducing devices. The city of Boulder has a policy of trying to
provide emergency response within six-minutes, utilizing roadways that are designated Critical
Emergency Response Routes (CERR). The benefit of speed reduction on neighborhood streets
must be weighed against the potential impacts to emergency response time. Attachment B
Agenda Item#
details the results of speed reduction and emergency response delay impacts for different traffic
mitigation.
STAFF RECOMMENDATION
City staff from the Transportation Division, Fire Department and Police Department has
considered the findings and the input from the community, and makes the following staff
recommendation:
• Delay-inducing, traffic-calming devices should not be considered as part of the
neighborhood traffic mitigation plan on 17m Street in either the Hillside or the Goss/Grove
neighborhood.
NEXT STEPS
The TAB recommendation on this issue will be forwarded to City Council. Staff will incorporate
the results of City Council's action on this issue into the 17th Street Bike Lane project. Staff will
continue to work with the Goss/Grove and Hillside neighborhoods by beginning the conceptual
design process for the roadway, including proposed neighborhood mitigation and bicycle
facilities. Staff will then follow the remaining steps of the NTMP policy, with the proposed plan,
including the neighborhood polling process and a review for recommendation by the
Transportation Advisory Board and review for approval by the City Council.
ATTACHMENTS:
A - NTMP Flow Chart
B -Experimental Device Results - "How Did They Work"
Agenda Item#
You have a neighborhood
traffic concern.
Neighborhood Traffic Mitigation Program Process
The NTMP sends you a '
"Neighbor to Neighbor
Education Kit."
Attachment A
Education/Petition/Dat, Collection Phase
Circulate petition for participation in the NTMP, due in April each year.
- Concurrent application of educational tools
(yard signs, speed monitoring trailers, 85th percentile speed o 5 mph over speed limit
• Continue education efforts for another
neighborhood speed watch, neighborhood speed pledge). 3 months.
- Speed data collected T `
- 3 months) Remonitor traffic speeds.
(timeframe `
Decision Point
Decision Point "Revisit Problem No Problem"
"Problem No Problem" 85th percentile speed > 5 mph over speed limit
85th percentile speed > 5 mph over speed limit NO ` Yes - initiate education and enforcement phase.
Yes -transition to EducationlEnforcementphaw. No - continue educational efforts.
No - continue educational efforts.
YES YES NO
Education/Enforcement Phase
Continued application of educational tools. 85th percentile speed 16 mph over speed limit
-Application of enforcement tools Continue education efforts.
(photo radar and traditional officer speed-enforcement).
- Additional speed data collected. -
(timeframe - 6 months)
T
Decision Point
"Eligibility for engineering treatments" 85th percentile speed 5 mph over speed limit
85th percentile speed > 5 mph over speed limit NO Continue education and enforcement efforts.
Yes - continue education and enforcement and Remonitor traffic speeds as part of next annual
include project in engineering ranking phase. process.
No - continue educational and enforcement efforts.
YES j
Engineering Treatment Ranking Phase - All other projects continue education and
,
Neighborhood Needs Assessment Priority Checklist used to rank eligible projects. enforcement efforts.
The two top priority projects - begin development of engineering treatment proposal _ rojects reranked annually.
All other projects - continue educational and enforcement efforts.
All Other Projects.
11wo Top Proorqty Proffiects (or More as staff and resources al
s l
Non-CERR Streets within 6-m'hute Response Time Zone CERR Streets and Non-CERR Streets outide 6-minute Response Time Zone
Process Summary ( Process Summary
- CEAP typically will not be required. Project streets evaluated on a case-by-case basis.
Neighborhood public involvement process leading to project proposal. TAB provides recommendation to City Council on the use of delay-inducing devices.
- Both delay-inducing and nondelay-inducing devices available. - CEAP may be required.
Neighborhood ballot (residents and property owners) on proposal is final decision. Neighborhood public involvement process leading to project proposal.
(timeframe - 6 months) Neighborhood ballot (residents and property owners) determine whether CEAP proceeds to City Council II
(timeframe - 6 months) J
i
Final Decision Point Decision Point
"Project Implementation" NO "Are Delay-inducing Devices Available?"
-Neighborhood ballot (residents and property owners)' TAB recommendation to City Council.
>=60% support -install improvements. N0 -initiate non-delay design process.
Yes -initiate full design process.
<60% support - don't.
?I
YES ,
i NO YES _•-_-._m.,.___._ w -
Non-delay Inducin esign Process Delay-inducing D sign Process
Process Summary Process Summary
Neighborhood public involvement process Neighborhood public involvement process leading to
- Project not it lemented. Project Impl mented. leading to project proposal. project proposal.
Neighborhood can reapply to the (timetrame - 3 months) I { - Neighborhood ballot (residents and -Both delay-inducing & nondelay-inducing devices
NTMP in 3 Years. f homeowners) on proposal is decision-making available.
continue education and process. T - CEAP required for delay-inducing devices.
enforcement efforts.
J I - -Neighborhood ballot (residents and property owners)
on proposal determines whether proposal and
associated CEAP proceeds.
Neighborhood Ballot Area -
` Properties on or adjacent to the primary street proposed for an engineering treatment
within 400 feet of either side of the proposed device and within 1 block on the side street Decision Point
for intersection treatments (ex. traffic circles). For a cul-de-sac, the neighborhood ballot "Continue Project Consideration?" >
area expands to include all properties from the treatment to the end of the cul-de-sac. NO Neighborhood ballot (residents and property owners)
Neighborhood Ballot Voting Structure - >=60% support - continue project consideration.'
One vote per dwelling unit and one vote per property owner. <60% support - don't.
- Project not implemented.
Neighborhood can reapply to the YES i
NTMP in 3 Years.
continue education and
enforcement efforts. I
I
Final Decision Point
"Final Project Consideration"
NO
TAB/City Council Consideration of Project CEAP
With nondelay designs, step is eliminated.
r~
Project Reassessment. YE5
(After 3 years.) Project Evaluation. Project Imp' ntedT
>=60% support to remove - (After 1 year.) (timeframe - )nth,;)
Device is removed.
Attachment B
APPENDIX D
Experimental Device Results - "How Did They Work"
SECTION PAGE(S)
SECTION CI: SPEED HUMPS (INCLUDING STAGGERED SPEED HUMPS) Cl -C3
SECTION C2: RAISED CROSSINGS/FLAT-TOP SPEED HUMPS C4 - C6
SECTION C3: TRAFFIC CIRCLES (INCLUDING VOLCANO TRAFFIC CIRCLES)C7 - C15
SECTION C4: TRADITIONAL ENFORCEMENT STATISTICS C16 - C17
SECTION C5: 9TH STREET - SIGNAGE AND STRIPING STUDY C18-C19
SECTION C6: STOP SIGNS C20 - C27
SECTION C7: PHOTO RADAR SPEED ENFORCEMENT C28 - C31
SECTION C8: PHOTO RED LIGHT ENFORCEMENT C32 - C35
SECTION C9: SPEED SENSITIVE TRAFFIC SIGNALS C36 - C38
Data is current as of 1999
Originally produced as Attachment C of the 4/8/97 City Council Study Session Memorandum
Revised for 3/30/99 TAB/City Council Joint Study Session
SECTION Cl: SPEED HUMPS
DESCRIPTION
Speed humps are one of the more traditional neighborhood traffic mitigation devices. They require
vehicles to slow down to avoid discomfort when driving over a physical hump in the roadway.
Boulder's speed humps are generally 12 feet in length with a maximum height or peak of 4 inches.
The installation of a speed hump and the subsequent signs and pavement markings costs approximately
$1,000 per installation.
Effectiveness at Speed Reduction
Effectiveness depends upon the size of the hump and the spacing between devices. Data collected
shows that the speed humps had a positive impact on speed reduction on the following streets.
TABLE CI-1: Speed Data Comparison Before and After Speed Hump Installation
I S (S l" ~ i
Average S &tlr'e~cenfile Seed
11 MI.
hge_ Bs~fizl er Chan"ge
PialectStreeet Before
t`
Mapleton Ave (1500 block) 23 mph 21 mph -2 mph 28 mph 25 mph -3 mph
North Street (1400 block) 27 mph 21 mph -6 mph 33 mph 25 mph -8 mph
Floral Drive (2100 block) 27 mph 22 mph -5 mph 31 mph 25 mph -6 mph
(Methods used to collect traffic data are not 100 percent accurate. There are enough daily fluctuations on roadway
speed/volume and limitations in the technology used to collect the data that the information should be used for comparative
purposes only. Eighty-fifth percentile speed is the speed at which 85 percent of the vehicles are traveling at or below that
speed.)
Public Input
City staff has received many requests to place speed humps on neighborhood streets. Concern has also
been expressed from other citizens that speed humps are dangerous to maneuver, cause unnecessary
noise, degrade the beauty of their neighborhood and force vehicles to operate well below the speed
limit..
SAFETY CONCERNS
Accident trends
Speed humps are generally used in the middle of city blocks, a good distance away from any
intersections. Most accidents in the city occur at intersections. Therefore, it is unlikely that speed
humps have much affect upon accident rates. Staff is not aware of any accidents which have been the
result of any City of Boulder speed humps. Likewise, the accident rates for neighborhood streets
where speed humps would be located are so low that it is not possible to evaluate any increase or
decrease in accident rates with any degree of accuracy.
Bicyclists and Pedestrians
Bicyclists and pedestrians are not really impacted by speed humps. The one potential negative impact
on bicycles is on roadways with bike lanes or paved shoulders. For drainage purposes, speed humps
are sometimes constructed to slope down to the bottom of the curb. The slope generally occurs
through the bike lane or shoulder which requires bicyclists to travel on an uneven surface.
Emergency Response
Delay occurs when an ambulance or fire engine passes over a speed hump. Staff has studied the effects
of different speed hump sizes on emergency response. The results show that speed humps cause four
to six seconds of delay per device.
ENVIRONMENTAL CONCERNS
The effects of slowing, stopping, idling and accelerating increase automobile exhaust, noise and fuel
consumption.
Vehicle Emissions
The act of slowing and accelerating increases the amount of emissions put out by motor vehicles, Staff
has not conducted any studies on the impact of speed humps on emissions.
Noise
The slowing and accelerating associated with any of the physical mitigation devices will result in
localized noise impacts in the vicinity of the device.
Fuel_ConiumptionfResource Impacts
The deceleration and subsequent acceleration associated with speed humps probably increase gas
consumption.
PROGRAM RESTRICTIONS
Diversion of Traffic to Adjacent Streets
Staff has not studied most of the streets surrounding speed hump locations to see whether there was
diversion associated with the placement of speed humps. See Table C1-2 for specific information on
locations where a change in traffic volume as a result of speed hump installation were studied.
TABLE CI-2: Changes in Traffic Volume Before and After Speed Hump Installations
AVer'dgt'' Txla f1G' ° Cd4vF.h1Gle5;Ver
.Street z+ r f $eore AffeF Oh.ngd vu~I'el'cent Alan e+ xi
Mapleton Avenue (1500 block) 1,710 vpd 1,490 vpd -220 vpd 12.9%
North Street (1400 block) 1,050 vpd 760 vpd -290 vpd -27.6%
Floral Drive (2100 block) 900 vpd 670 vpd -230 vpd -25.6%
1998 Speed/Volume follow-up
Traffic speeds and volumes were collected in 1998 on these three roadways where the speed humps
remain. The results show that traffic volumes and speeds continue to be low on both North Street and
Floral Drive. Both speeds and volumes increased on Mapleton Avenue but have not reached
premitigation levels to date. For comparison purposes, the average speeds and 85th percentile speeds
on Mapleton Avenue and Floral Drive were 22 mph and 27 mph, respectively, on both streets. The
-Cl-
average speed and 85th percentile speed on North Street was 21 mph and 26 mph. The traffic volumes
on Mapleton Avenue, North Street and Floral Drive were approximately 1540 vpd, 520 vpd, and 570
vpd, respectively.
Staggered Speed Hump Demonstration
Staff planned a "staggered" speed hump design demonstration for fall 1998. Staggered speed humps
are designed so that rather than having both approaches to the speed hump occur at the same location,
they occur approximately 75 feet apart from each other. Thus, vehicles traveling on the roadway still
must traverse the speed hump to stay in their lane but an emergency response vehicle could circumvent
the speed humps be going into the opposing lane.
The staggered speed humps were to be placed on Moorhead Avenue and Pennsylvania Avenue.
However, a predemonstration survey of neighborhood opinions suggested that these demonstrations
did not have support in either of these neighborhoods. As a result, the city's Transportation Advisory
Board (TAB) conducted a public hearing on the proposed demonstrations and recommended to city
staff that these demonstrations be canceled. Staff notified City Council of the decision to not go
forward with the demonstrations through a October 14, 1998 Weekly Information Packet
memorandum.
-C2-
SECTION C2: RAISED CROSSINGS/FLAT-TOP SPEED HUMPS
DESCRIPTION
Raised crossings are elongated versions of speed humps (essentially a large, flat-top speed hump).
They require vehicles to slow down to avoid discomfort when driving over a physical hump in the
roadway. The raised section is much longer and allows some of the longer vehicles (such as fire
engines) to get both wheels up on the hump before the front wheel leaves the hump. Theoretically, this
minimizes the potential for undercarriage damage when these types of vehicles travel over the humps.
Boulder's raised crossings are generally 42 feet in length with a maximum height or peak of 6 inches.
The installation of a raised crossing and the subsequent signs and pavement markings cost
approximately $3,600 per installation if there are no drainage issues (which raise the cost
considerably).
EFFECTIVENESS AS A MITIGATION TOOL
Sneed Reduction
Staff has studied the effect of raised crossings as a speed mitigation device and found them to be
effective at reducing speeds. The level of effectiveness depends upon the size of the device and upon
the spacing between devices.
TABLE C2-1: Speed Data Comparison Before and After Raised Crossing Installation
1 Et i T AVCr1g*"' r rSip1,UCTGeItIjS3eeti k2
Pxolect Streei Before,SI1er Changd, fBeforeAfter'y Change
Moorhead Avenue (3100 block) 29 mph 27 mph -2 mph 34 mph 31 mph -3 mph
Moorhead Avenue (4300 block) 28 mph 27 mph -1 mph 34 mph 31 mph -3 mph
Edgewood Drive (2200 block) 29 mph 24 mph -5 mph 36 mph 28 mph -8 mph
(Methods used to collect traffic data are not 100 percent accurate, There are enough daily fluctuations on roadway
speed/volume and limitations in the technology used to collect the data that the information should be used for comparative
purposes only.)
1998 Speed data follow-up
No follow-up to these locations was conducted. The raised crossing mitigation devices on Moorhead
Avenue and Edgewood Drive have been removed.
Public lnput
City staff has received many requests to place raised crossings on neighborhood streets. Concern has
been expressed from other citizens that raised crossings are dangerous to maneuver, cause unnecessary
noise, degrade the beauty of the neighborhood and force vehicles to operate well below the speed limit.
SAFETY CONCERNS
Accident trends
See comments for speed humps, Section C 1.
-C3-
Bicyclists and Pedestrian
s
See comments for speed humps, Section C 1.
Emergency Response
As with speed humps, staff test results show that raised crossings cause between four and six seconds
of delay per device. The purpose of developing the raised crossing was to decrease the negative
impacts on fire engines. Experience and comment from the Fire Department suggest that this is not a
significant benefit and probably does not justify the additional cost.
ENVIRONMENTAL CONCERNS
See comments for speed humps, Section C1.
PROGRAM RESTRICTIONS
Diversion of traffic to adjacent streets
In 1996, staff studied the most logical adjacent streets at both raised crossing locations to see whether
there was diversion associated with the placement of speed humps. On Martin Drive, adjacent to
Moorhead, no mitigation was placed to offset potential diversion. On Floral Drive, adjacent to
Edgewood, the city constructed three speed humps to handle existing speeding issues and to offset
potential diversion. The potential for traffic diversion from the Edgewood raised crossings was also
studied on Glenwood Avenue which is parallel to but several blocks north of Edgewood.
TABLE C2-2: Traffic Volume Before and After Raised Crossing }Installation - Martin Acres
ZIH
"ag
£ i t 3^ f is 4' LI31ktt
s
' .NI t i
a t t t T
113 g
c
Stz~t , $eore Affe Cihauge ;'eicenf;Cange
-4v 0,
. , .
Moorhead Avenue (3100 block) 4,590 vpd 4,460 vpd 130 vpd -2.8%
Moorhead Avenue (4300 block) 2,810 vpd 2,620 vpd 190 vpd -6.8%
Martin Drive (3 100 block) 930 vpd 990 vpd 60 vpd +6.5%
Martin Drive (4300 block) 670 vpd 630 vpd 40 vpd -6.0%
These results suggest that the raised crossings did not divert a significant amount of traffic to Martin
Drive and may have diverted some traffic onto U.S. 36.
TABLE C2-3: Traffic Volume Before and After Raised Crossing Installation - Edgewood/Floral
sue.
t 4 { ' ( ~Y'~?etage, DtFj?f:~ 1~IA1r1G'~ { t,{
§ st c r
StreCt { *d a r a F„Bef£YIe < A"IICL t st u IIteYeeTSt ClraflgPro-
Edgewood Drive 11,140 vpd 9,690 vpd 1,450 vpd -13.0%
Floral Drive 900 vpd 670 vpd 230 vpd -25.6%
Glenwood Drive 490 vpd 480 vpd 10 vpd -2.0%
-C4-
The reduction on Floral is likely to be due to the speed humps there, rather than the raised crossings on
Edgewood. These results suggest that the raised crossings did not divert a significant amount of traffic
to Floral Drive or Glenwood Drive and may have diverted some traffic onto Iris Avenue.
1998 volume follow-up study
No follow-up to these locations was conducted. The raised crossing mitigation devices on Moorhead
Avenue and Edgewood Drive have been removed.
-C5-
Section C3: TRAFFIC CIRCLES
DESCRIPTION
Traffic circles rely upon a change in traffic control and a physical feature which vehicles must
maneuver around to slow their speed.
Since January 1995, all traffic circle locations in Boulder have all-way yield traffic control (prior to
that time, STOP signs on the side street controlled traffic). This all-way yield condition requires
vehicles to yield to traffic which is already in the circle but does not require them to stop if there is no
conflicting traffic. The traffic circles are typically 20 to 25 feet in diameter and slow vehicles to
between 15 mph and 20 mph when they are passing through them. The installation of a traffic circle
with the appropriate traffic control costs approximately $10,000 to $40,000 per intersection.
The Transportation Division has constructed six permanent small diameter traffic circles, temporary or
permanent, in order to reduce vehicle speeds. The locations studied are:
• Pearl Street & Fourth Street (1983)
• Pearl Street & Third Street (1983)
• Ninth Street & Evergreen Avenue (1994)
• Arapahoe Avenue & Fourth Street (1995)
• Arapahoe Avenue & Fifth Street (1995)
• Sixth Street & Maxwell Avenue (1995)
Two other larger diameter traffic circles were built as part of new developments. These circles were
installed during subdivision construction more as traffic control devices than as speed mitigation
devices. As such, no "before" traffic data exists. The circles in the Four Mile Subdivision and the
Noble Park Subdivision are not part of this study.
BACKGROUND
In August 1995, the city installed trial traffic circles on Pine Street at 15th and 17th Streets, and on
Balsam Avenue at 14th and 15th Streets. Both Pine and Balsam are major streets, primary bicycle
corridors and primary emergency response routes. All four trial circles were constructed with a
simulated standard curb-edge.
EFFECTIVENESS AS A MITIGATION TOOL
Sneed Reduction
Traffic circles were placed on North Ninth Street, Arapahoe Avenue, Balsam Avenue and Pine Street
to mitigate traffic impacts. City staff has studied the effect of these traffic circles as a speed mitigation
device and found that they were effective at reducing speeds at every location. Speeds were measured
midblock between paired devices to gauge their overall impact.
-C6-
TABLE C3-1: Traffic Volume Comparison on Project Streets
i 4_ _ r ~"a'" - 3 Aerage'T)a11y ~rath+f: F" a F x_
h t A~ * { t
t ~ F~ h~ hu.i i } l ~
PSQ.(t'.y6t StCe F
er e G.hatt e t
g i~ < * . Before ' r tNl u UChange.
North Ninth Street (n/o Evergreen) 3,360 vpd 1,970 vpd -1390 vpd -41.4%
Arapahoe Ave (400 block) 2,010 vpd 1,940 vpd -70 vpd -3.5%
Balsam Ave (140011500 block) 10,910 vpd 8,280 vpd -2630 vpd -24.1%
Pine Street (1600 block) 8,660 vpd 7,280 vpd -1380 vpd -15.9%
Table C3-2 shows the change in traffic speed before and after the same conditions. The speed data
shown includes the average speed and the 85th percentile speed.
TABLE C3-2: Traffic Speed Comparison on Project Streets
P4v~rage Steed ' 1. E $klt Pewetttrl~ Speed "t'
Project Street efQie Aftez l a e $ $eftare tfler Ghar~g6
North Ninth St (n/o 28 mph 20 mph -8 mph 33 mph 23 mph -10 mph
Evergreen)
Arapahoe Ave (400 block) 28 mph 26 mph -2 mph 33 mph 28 mph -5 mph
Balsam Ave (140011500 blk) 31 mph 23 mph -8 mph 38 mph 25 mph -13 mph
Pine Street (1600 block) 30 mph 25 mph -5 mph 33 mph 31 mph -2 mph
(Methods used to collect traffic data (speed 8 volume) are not 100 percent accurate. There are enough daily fluctuations
on roadway speed/volume and limitations in the technology used to collect the data that the information should be used for
comparative purposes only.)
The data for North Ninth Street involves a comprehensive mitigation project which included four-way
stop control at Ninth and Balsam, a median treatment just north of that intersection, and the traffic
circle at Ninth and Evergreen. It is likely that all three of these devices had something to do with the
change in traffic speeds and volume on the roadway.
1998 Speed/Volume follow up
Traffic speeds and volumes were collected in 1998 on Balsam Avenue between the two temporary
traffic circles. The data suggests that while travel speeds remain low (23 mph average speed, 26 mph
85th percentile speed), traffic volume has increased. Prior to the traffic circles being placed at these
locations, the traffic volume was approximately 10,910 vpd. Following the placement of the circles,
the volume dropped to approximately 8,280 vpd. Two years later, the traffic volumes have increased
to approximately 9,570 vpd on this section of Balsam Avenue.
1998 speed and volume data were also collected on Alpine Avenue. This roadway is adjacent to
Balsam Avenue and saw the largest increase in traffic when the traffic circles were placed.
-C7-
Prior to the traffic circles being placed on Balsam Avenue, the traffic volume on Alpine Avenue was
approximately 2,610 vpd. Following the placement of the circles, the volume increased to
approximately 3,230 vpd. Two years later, the traffic volumes have increased to approximately 3,440
vpd on this section of Alpine Avenue.
Public Innut
City staff has received many comments about traffic circles. Public comment has centered on the
experimental circles on Pine and Balsam. Very few comments have been made about other permanent
traffic circles. Several people have called in support of the devices. Far more people have called to
say they do not like the devices and/or feel the devices are confusing or unsafe. In particular, there
seems to be confusion over who has the right-of-way in the intersection and there is concern about
safety for bicyclists and pedestrians. GO Boulder's Close-call Hotline has recorded many of these
safety concerns.
SAFETY CONCERNS
Accident trends
Traffic circles have been used in other cities and countries as a safety device to mitigate locations with
a high number of right angle collisions. The safety research is divided when it comes to pedestrians
and cyclists. Some studies indicate that bicyclist-vehicular accidents tend to rise at traffic circles due
to the failure of motorists to yield to circulating bicyclists. In addition, while slower traffic may make
midblock pedestrian crossing safer, traffic circles may also make it more difficult for pedestrians to
cross at the intersections themselves.
Staff has reviewed the accident history at all permanent and temporary traffic circle locations in the
city between January 1992 and December 1996. Each accident report in this time period at a traffic
circle location was reviewed to see if the accidents occurred in the intersection or were caused by
actions in the intersection, and to determine whether there are any trends. Staff also reviewed whether
accident rates have increased or decreased at traffic circle locations before and after circle construction
and whether accident rates have increased or decreased after side street STOP sign control was
replaced with all-way YIELD signs.
Permanent Traffic Circle Locations (through 1996)
The permanent locations which were in place during the study time period (1992 to 1996) do not show
a high number of accidents. The intersection of Fourth and Pearl showed a slight increase in accident
rate after the traffic control was changed from side street stop control to all-way yield control. Those
permanent locations which were installed sometime during the study time period generally showed
either a lower number of accidents per year after installation of the traffic circles or no increase in
accidents at all. However, these locations had very low accident numbers to begin with.
Temporary Traffic Circle Locations (through 1996)
The intersection of 14th/Balsam had three accidents, including two serious injury accidents during the
three and one-half years prior to the installation of the traffic circle. Two of the accidents involved
-Cg-
speeding vehicles on Balsam Avenue. No accidents occurred at this location after the installation of
the traffic circle.
The 1 5th/Balsam intersection had no accidents during the three and one-half years prior to the
installation of the traffic circle and one accident during the one and one-half years after the installation.
This accident involved a driver who claimed to have sun in his eyes and drove directly through the
traffic circle causing considerable damage to his car and to the traffic circle but no injury to himself or
anyone else.
The 15th/Pine intersection had seven accidents (four injury accidents) during the three and one-half
years prior to the installation of the traffic circle, and four accidents (no injury accidents) during the
year and one-half after installation. Most of the accidents which occurred prior to the installation of
the traffic circle involved speeding traffic on Pine Street, usually with injuries. Three of the four
accidents which occurred after the installation of the traffic circle involved ice and snowy weather as
contributing factors. All accidents were low speed collisions and did not involve any injuries.
The 17th/Pine intersection had nine accidents (five injury accidents) during the three and one-half
years prior to the installation of the traffic circle, and seven accidents (two injury accidents) during the
year and one-half after installation. Most of the accidents which occurred prior to the installation of
the
traffic circle involved speeding traffic on Pine Street, usually with injuries. All of the accidents which
occurred after the installation of the traffic circle involved ice and snowy weather as contributing
factors. All of the accidents were low speed collisions and most did not involve injuries. The two
accidents with injuries were relatively minor.
The accident information at the temporary traffic circles suggest that the probability of accidents
occurring on the two different streets is very different. Balsam Avenue has had very few accidents
prior to the installation of the traffic circles and even fewer after the circles. Pine Street had many
more accidents than Balsam Avenue prior to the circles and saw a marked increase in accidents after
the installation of the traffic circles. Other factors noted was that almost all of the accidents which
occurred at the Pine Street traffic circles had icy/snowy weather as a contributing factor and that the
severity of accidents at the two intersections dropped considerably. Both Balsam Avenue and Pine
Street are plowed as secondary snow routes.
A change in traffic control (such as the replacement of side street STOP signs with all-way yield signs)
which removes the right-of-way from main street traffic typically increases the number of accidents at
an intersection. One well-researched fact of automobile travel is that when vehicles are constantly
changing speeds (or in this case, slowing to a stop and then proceeding again), they are more likely to
get in an accident than when they are traveling at a constant speed. Staff makes traffic control changes
when they believe that severe accidents, which often result in major injuries, can be avoided or when
such changes would result in conditions to reduce speeds and reduce the chance of accidents with
injuries.
Table C3-3 details the number of accidents at each location studied over the four-year time period.
Accidents per year are shown with injury accidents in parentheses. Accidents which occurred while a
traffic circle was in place are shown with a following the number.
-C9-
TABLE C3-3: Number of Accidents at Traffic Circle Intersections
"AA z ' -
_ ~s e R'~ .Z..t s sF
7'?i ~t~ { inY~ ai"t-7 qis PF''t 3
.Loeaftort
Pearl & Third 0 0 0 0 0
Pearl & Fourth 1 * 1 * 0 1 It 2*
Arapahoe & Fourth 0 0 0 0 0
Arapahoe & Fifth 0 0 0 0 1
Sixth & Maxwell 0 1 1 0 0
Ninth & Evergreen 2 0 2 0 0
Balsam & 14th 1 0 2(2) 0 0
Balsam & 15th 0 0 0 1* 0
Pine & 15th 2(1) 2(1) 2(1) 1(1) 4*
Pine & 17th 2(2) 2(1) 4(2) 1 8(2)*
A look at the total number of accidents can be somewhat deceptive. A more appropriate measure of
the relative safety of an intersection is to look at the number of accidents which occur at an intersection
versus the amount of traffic that moves through the intersection. Traffic engineers measure this
accident rate in "accidents per million entering vehicles." The city studies such accident rates in the
"Hazard Elimination Program (HEP)" and generally considers an intersection to have a "high" accident
rate if it is higher than 2.0 accidents per million entering vehicles. The top 10 intersections on the 1996
HEP report were all higher than 1.9 accidents per million entering vehicles.
TABLE C3-4: Accident Rates
tnt rsecrion f a11y Trz fie Entering P g na Acctde~tspet, l n Acerdn mil
F (9t9CL d3{a) _@t.C`(*~',f # CEl erln 'YCh1C~~S ;
Balsam & 14th 8,100 vpd 0 0.0
Balsam & 15th 8,400 vpd 1 0.2
Pine & 15th 9,090 vpd 4 0.9
Pine & 17th 8,630 vpd 8 1.8
* - trial period of 16.5 months (502 days)
A review of the accident rate data shows that the intersection of 17th/Pine has a relatively high accident
rate for a residential intersection. A rate of 1.8 accidents per million entering vehicles would probably
put this intersection in the top 30 intersections in the city. The other three intersections have low
accident rates.
-C10-
1998 Accident information follow up
A review of accident records for 1997 and 1998 show that the total number of accidents and the
severity of accidents remained low for all traffic circle locations. There were no intersections with
more than two accidents per year and most intersections had one or less accidents per year. The
intersection of Pine Street and 17th Street showed the greatest improvement with only three accidents
occurring in the two year time period. Another significant fact observed was that since the
construction of the traffic circle on Ninth Street and Evergreen Avenue, almost five years ago, there
has not been a single accident at this location.
Compliance
For the traffic circles to operate, motorists must comply with the traffic control which requires them to
yield to traffic already in the traffic circle. Staff has made observations at several traffic circles but has
not collected any specific compliance data. Our observations suggest that compliance is still an issue
but there appears to be higher compliance in 1998 than when the traffic circles were first installed.
Compliance also seemed to be higher at intersections where the main street volume was lower or where
the main street and side street traffic volumes were similar. The design of the traffic circle itself also
seemed to have an effect on compliance level with the bigger traffic circles having better compliance
than the smaller circles.
Bicyclists and Pedestrians
Traffic circles are not considered to be beneficial for either bicyclists or pedestrians. The traffic circles
are designed with a narrow approach which forces vehicles to slow down as they enter. This narrow
approach also requires bicyclists to merge with motor vehicle traffic traveling in their direction.
Signage near the experimental traffic circles on Pine Street and Balsam Avenue warns vehicles to not
pass bicyclists through the traffic circle. Field observations have shown that compliance with this
request is generally good though bicyclists have called to complain that they do not feel safe in the
traffic circles. Traffic circles on roadways which are designated bicycle facilities may discourage
bicyclists from riding on these roadways.
Pedestrians crossing at traffic circles are likely receiving little benefit. Traffic circle themselves may
obscure pedestrians who are crossing at the intersection, and this may serve as a deterrent for
pedestrians. Some pedestrians have called to complain that they do not feel safe crossing near the
traffic circles. On the other hand, slower auto speeds which have been measured are likely making it
somewhat easier for pedestrians to cross the main street, particularly in the mid-block.
Emergency Response
Delay is experienced whenever an ambulance or fire engine must pass through an intersection where
they do not have the right-of-way. Traffic circles cause more delay by forcing the fire truck to
maneuver through a constrained area. Tests done by Transportation Division and Fire Department staff
show that traffic circles cause 7.5 to 10 seconds of delay per device. Also, the banking movement
required of fire trucks as they pass through the traffic circles may cause damage to the vehicle itself. In
addition to the cost of repairing the vehicle, the time spent fixing the vehicle may cause a further
impact on emergency response if other vehicles are not available to cover the service that vehicle was
providing.
-C11-
ENVIRONMENTAL CONCERNS
See comments for speed humps, Section C 1.
PROGRAM RESTRICTIONS
Diversion of Traffic to adjacent streets
Staff has looked carefully at the potential diversion associated with the placement of traffic circles.
When the permanent circles were placed at Ninth/Evergreen and the two Arapahoe traffic circles, staff
conducted before and after data collection to monitor speed and volume. Surrounding roadways of a
similar or lesser classification did not experience an increase in traffic volume or speed in the vicinity
of the Ninth/Evergreen traffic circle, despite the fact that the total volume on Ninth Street in the
vicinity of the traffic circle decreased significantly. The missing traffic was most likely diverted onto
Broadway.
Following the construction of the Arapahoe traffic circles, there was an increase in traffic on Marine
Street. The traffic volume on Marine Street was low to begin with and the increase in traffic was
small. Speeding was not a problem on this street before or after the installation of the traffic circle.
Staff also monitored the potential for traffic displacement with the construction of the temporary traffic
circles on Balsam Avenue and Pine Street.
When the traffic circles were installed on Balsam Avenue, traffic volumes increased on Elder Avenue,
Cedar Avenue and Alpine Avenue (all lower classification streets). The increase on Alpine Avenue
may be more closely associated with the placement of speed humps on North Street. Mitigation on
both Elder Avenue and Cedar Avenue (the placement of four-way stop control at three intersections)
did little to reduce speeds and had no impact on volume on the two streets. No mitigation has been
attempted on Alpine Avenue due to emergency response concerns.
When the traffic circles were installed on Pine Street, traffic volumes on Mapleton Avenue stayed
about the same and the speed went down slightly. Spruce Street actually had different speed and
volume characteristics between different blocks. In the 1700 block, the volume increased considerably
but the speeds decreased. In the 2200 block, the volumes did not increase as much but the speeds did
not decrease as much either.
TABLE C3-5: Changes in Traffic Volume on Adjacent Streets
~U3~Sat11 A etft
B yBefare' ~k$e~i ttng 1a Change ;
Elder Ave. 1,580 vpd 1,730 vpd +150 +9.5%
Cedar Ave. 560 vpd 750 vpd +190 +33.9%
Alpine Ave. 2,610 vpd 3,230 vpd +620 +23.8%
t t t :i 3. u: r Fz 4r, i. .:3 ~ 'y r2`~=.°•`k~€ lv~j~i~
PiRtree 1tig~UOIT , d 3 F BeOte er t' l~Itk*t t o G t _ ice=
Mapleton Ave. 1,420 vpd 1,490 vpd +50 +49%
Spruce St. (2200 block) 3,830 vpd 4,040 vpd +210 +5.5%
-C12-
I Spruce St. (1700 block) 3,040 vpd 3,720 vpd +680 +22.4%
(Methods used to collect traffic data are not 100 percent accurate. There are enough daily fluctuations on roadway
speed/volume and limitations in the technology used to collect the data that the information should be used for comparative
purposes only.)
Table C3-6 shows the change in traffic speed before and after the same conditions. The speed data
shown includes the average speed and the 85th percentile speed.
TABLE C3-6: Changes in Traffic Speed on Adjacent Streets
Average speed 85th Percentile Speed
P -A
e Aft, Change
.Balsam AveaMthgattoiiefire". After ` Cltange Beoi
Elder Ave. 27 mph 26 mph -1 mph 33 mph 31 mph -2 mph
Cedar Ave. 26 mph 23 mph -3 mph 33 mph 28 mph -5 mph
Alpine Ave. 31 mph 28 mph -3 mph 38 mph 34 mph -4 mph
Pit?e 8treetMiflgat}oti' , el`vre After Change liefo e ' #1ft r Change
1 25 mph 21 mph -4 mph 28 mph 25 mph -3 mph
Mapleton Ave.
Spruce St. (2200 block) 31 mph 29 mph -2 mph 38 mph 37 mph -1 mph
Spruce St. (1700 block) 27 mph 24 mph -3 mph 33 mph 28 mph -5 mph
OTHER CONSIDERATIONS
Community Concems/City Policies
Traffic circles have existed in Boulder for years, but the experimental traffic circles on Pine and
Balsam introduced the concept to a large number of drivers. They are not common in most
communities, and therefore they can be confusing to the public. In addition, when the traffic circles
were installed, our Code did not take into account the fact that it is extremely difficult to signal a
vehicle's intention to make a left turn when entering a traffic circle because the vehicle must turn to the
right to maneuver through the traffic circle first. The Transportation Division worked with the City
Attorney's office to develop an Ordinance change which would make it legal for vehicles to turn left
from a traffic circle without having to signal. This is consistent with the concept that a vehicle must
yield to any other vehicle in the traffic circle regardless of the movement they are making.
The high number of accidents at the Pine Street intersections suggest that traffic circles at intersections
with high traffic volumes, reasonable side street volumes and snow may be problematic. Some thought
should go into the impact upon the city's street maintenance resources if many traffic circles are added.
It should be noted that the 1997 and 1998 accident data showed considerably fewer accidents at the two
Pine Street locations.
Traffic circles require vehicles to slow below the speed limit to maneuver through them. While this
impact is relatively small (a reduction of five to 10 mph) compared to other devices (such as STOP
signs), it is still an issue to be considered. The goal of the NTMP is to get traffic in neighborhoods to
drive the speed limit or close to it. On every street in the NTMP there are drivers who are already
-C13-
complying with this goal and driving no faster than the speed limit. Such drivers would suffer the
same inconvenience as drivers who are currently speeding.
TIDE MODIFIED DESIGN OR "VOLCANO "TRAFFIC CIRCLE TEST•
Design Objective
The original trial traffic circles were built with 6-inch high bumper blocks to simulate a standard curb-
edge perimeter. This design forces all vehicles, including emergency response vehicles, to travel
around the circle to pass through the intersection, thus reducing vehicle speed. The required "turning"
maneuver results in an additional 7.5 to 10 second/circle delay for emergency vehicles. Staff proposed
a "volcano," or conical-shaped traffic circle, featuring a low-profile, mountable edge for further testing.
The modified edge design would need to be shallow enough to allow emergency response vehicles to
travel over the rim of the circle in a straighter path through the intersection. It was anticipated that this
would result in less delay for these vehicles at each traffic circle. At the same time, the modified edge
would have to be formidable enough to discourage other vehicles from traveling over the circle edge.
Should this happen, it would defeat the overall objective of using a traffic circle to achieve speed
mitigation.
Initial Test at the City Yards
A test run with City emergency response vehicles and a "volcano" circle was conducted at the City
Yards on October 14, 1996. The volcano circle design was based on a design used in Portland, Ore.
Staff constructed a full circle with a 4-to-1 slope and alternative edge heights of two and three inches
(on each hemisphere). The altering edge depths were provided to determine which edge height had the
least impact on response vehicles, and which, if either, would be preferable to Fire Department
personnel.
Pending Fire Department approval, the second phase of the experiment would entail on-site
construction of three volcano circles on Pine at the intersections of 19th, 22nd and 24th streets for
sustained testing.
During the initial test phase at the City Yards, Fire Department personnel were not able to traverse the
edge of the (lower) two- inch high circle perimeter at a speed sufficient to achieve less delay than now
experienced with the standard curb-edged trial circles. Fearing damage to the vehicle, drivers felt
compelled to come to an almost complete stop before mounting the rim of the circle.
Findings
Staff concluded that the trial volcano circle would have a greater negative impact on emergency
response time, if used as intended. A volcano circle design would likely equal the delay experienced
with existing designs, if drivers chose to drive around, not over, the perimeter. Since the original
objective with using a modified design was to achieve less delay at each traffic circle, staff determined
that trial volcano circles should be constructed.
-C14-
SECTION C4: TRADITIONAL ENFORCEMENT
BACKGROUND
In 1996, staff considered the neighborhoods that had applied to the NTMP and developed a list of High
Enforcement Zones (HEZ) for the Police to focus their speed enforcement efforts upon. At the time of
this study, there were approximately 38 designated HEZ areas on 30 individual streets identified on the
NTMP/ Critical Emergency Response Route (CERR) map. In the summer of 1996, traffic unit
personnel began rotating through this number of streets with peak period enforcement by four officers
three days per week and two officers two days per week. This resulted in an individual HEZ street
segment receiving full enforcement (all four available personnel) on a Tuesday and Thursday of a
week, followed by full enforcement on Wednesday of the following week. Within a 10 to I 1 week
cycle, all of the HEZs received two days of enforcement, and six HEZ streets/segments received four
days of "saturated" enforcement. Using this schedule, it would have taken over a year before a street
rotated all the way through the cycle to get four days of saturated enforcement again.
HEZ EXPERIMENT
To measure the affects of the HEZ, baseline speed and volume data was collected on two streets prior
to the two week HEZ enforcement described above. (Example: three days of enforcement the first
week and two days the second week). The two streets selected were Baseline Road and N. 26th Street.
TABLE C4-1: Autos Speeds Before and After Hez Enforcement (Using 85th Percentile Speeds)
57
' t Street ~d~?' ~ ' ~~eeks' ~ 41t1ee~`" a Speed3efort tHEI" Af1erFTFZ Afe[IE t
Street w Limsi, Eztfaremenf'nfotrxri~ent enforcement
N. 26th St (Between Iris & Kalmia) 25 34 mph 34 mph 37 mph
N. 26th St (Between Norwood & Agate) 25 37 mph 34 mph 37 mph
Baseline (Between 13th and 14th) 30 34 mph 37 mph 34 mph
Baseline (Between Grant & Eighth) 30 37 mph 34 mph 37 mph
TABLE C4-2: Percent of Autos Speeding Before and after HEZ Enforcement
A- r
J
~
,ro
r Sheet vt;' dtng oi 4 ker l3 after HEZ
of
Street LuntL : IIEZ Enforceriient nfarcetxtettn"force~nenr
N. 26th St (Iris to Kalmia) 25 74% 72% 85%
N. 26th St (Norwood to Agate) 25 78% 72% 78%
Baseline (Between 13th and 14th) 30 42% 77% 57%
Baseline (Between Grant & 30 67% 58% 62%
Eighth)
-C15-
The results for this demonstration are somewhat inconclusive. Two weeks following the enforcement,
the speeds were down in some locations, and the same or higher in others. After four weeks, all speeds
returned to their previous level or higher. This level of enforcement does not appear to be effective as
a speed mitigation device.
Traditional Enforcement Cost
In 1998, the NTMP staff working group attempted to model how many officers would be required to
perform speed enforcement on all NTMP streets which were also Critical Emergency Response Route
(CERR) streets at a level of enforcement which would likely impact corridor speeds. Several
assumptions were made concerning how many enforcement officers would be needed to accomplish
this task and how frequently each street would need to be enforced. Based upon this model, there
would need to be an additional 70 to 117 commissioned traffic officers hired to perform this task.
Costs associated with this approach would include salaries for the new officers, equipment (weapons,
vehicles, radar, etc...) and the need to expand the current Public Safety Building to accommodate the
new officers. This would result in a one-time cost of between $2.8 million and $4.26 million. There
would also be an on-going yearly cost of between $4.24 million and $8.1 million.
In addition, there would be impacts to both the Courts and the City Attorney's office. It is anticipated
that there work load would be increased by approximately six times it's current level. There would
need to be additional clerks, judges and attorneys hired to accommodate this workload. The on-going
yearly cost of these additional employees would be between $3.5 million and $4.0 million. Another
cost consideration would be the construction of a new facility to accommodate these new employees.
There would need to be at least five more courtrooms to handle the court cases and would be
approximately 65,000 square feet in size. It is unlikely that this additional space could be
accommodated at the current facility, so a new facility would have to be constructed at a different
location. This would likely cost an additional $10 million to $15 million depending upon where it was
located. It may be possible to renovate an existing 65,000 square foot building, if a suitable location
could be found. This would reduce the cost.
In summary, it appears that this approach would cost the city between $12 million and $20 million in
up-front costs, plus an additional $7.5 million to $12 million in on-going yearly costs.
One possible funding mechanism for this effort would be a significant increase in the fines for
speeding within the City of Boulder. Based on the aforementioned officer projections, it is anticipated
that they could generate approximately 70,000 to 100,000 speeding citations per year. To fund the
annual costs of the aforementioned effort, this would result in an increase in fines of between $110 and
$120. The flaw in this approach, however, is that as enforcement officers generated this number of
citations, there would likely be a significant decrease in speeding (which is the goal of this effort).
Less speeding would mean fewer citations, and this funding source would likewise decrease.
-C16-
SECTION C5: NINTH STREET - SIGNAGE AND STRIPING STUDY
DESCRIPTION
Passive mitigation involves using devices such as signs and pavement markings to encourage traffic to
slow down. They do not physically constrain or impact vehicles on the roadway. Passive mitigation
could include speed limit signs or crosswalk markings. It can also include signs asking vehicles to
slow down, drive the speed limit, watch for children, etc. The city has used these devices sporadically
throughout the city at several resident requests. In 1996, the city studied one such corridor (Ninth
Street between Pine and Mapleton) to assess the effectiveness of such devices when used for speed
mitigation.
The installation of passive mitigation is variable depending upon what type is chosen. Signs are
typically $200 per installation. Crosswalks and other pavement markings are between $200 and $500
per installation.
EFFECTIVENESS AS A MITIGATION TOOL
Speed Reduction
Staff has studied the effect of "passive mitigation" on the Ninth Street corridor between Spruce and
Mapleton. The devices were relatively ineffective at reducing speed. These results are consistent with
feedback we have received from the public concerning signs and pavement markings and their
confidence in those devices as speed mitigation. City staff has collected speed data on Ninth Street,
between Pine and Mapleton before and after the installation of additional speed limit signs and then
again after the installation of crosswalks as well as speed limit signs. Table C5-1 outlines the results of
the before and after speed data study.
TABLE C5-1: Speed Data Comparison Before and after Passive Mitigation
i s Average Speed .e i_. h aA t e t1Ie5 efl i
P,
'kv
PFO'ect SCreeiJ Before-; -Aflei: Change :Befsrxe tex Change€,
Ninth Street (Pine to 28 mph 28 mph 0 mph 32 mph 31 mph -1 mph
Mapleton)
(Methods used to collect traffic data are not 100 percent accurate. There are enough daily fluctuations on roadway
speed/volume and limitations in the technology used to collect the data that the information should be used for comparative
purposes only.)
IMPACTS
SAFETY CONCERNS
Accident trends
Passive mitigation does not provide any type of physical obstacle to drivers. Therefore, it is unlikely
that it has much affect upon accident rates in a corridor. Staff is not aware of any accidents which have
been the result of "passive mitigation."
Bicyclists and Pedestrians
Passive mitigation has little to no impact to bicyclists and pedestrians.
-C17-
Emergency Response
Passive mitigation has little to no impact on emergency response.
ENVIRONMENTAL CONCERNS
The use of different mitigation treatments have several environmental/quality of life issues associated
with them. Specifically, the effects of slowing, stopping, idling and accelerating increase automobile
exhaust, noise and fuel consumption.
Vehicle Emissions
Since "passive mitigation" does not result in a significant slowing or stopping of traffic, it is not likely
that it has a significant impact on vehicle emissions.
Noise
Since "passive mitigation" does not result in a significant slowing or stopping of traffic, it is not likely
that it has a significant impact on noise in the corridor. Noise is directly linked to traffic speed.
Fuel Consumption/Resource Impacts
Since "passive mitigation" does not result in a significant slowing or stopping of traffic, it is not likely
that it has a significant impact on vehicular fuel consumption.
PROGRAM RESTRICTIONS
Diversion of traffic to adjacent streets
Since "passive mitigation" was so ineffective as a speed mitigation device, it is not likely that it caused
significant diversion of traffic to side streets. The corridor chosen for study did not see a significant
change in traffic volume.
OTHER CONSIDERATIONS
Public Input
City staff receives many requests to place signs and/or crosswalks requesting slower or more cautious
behavior from drivers. Staff has also received feedback in areas where these devices have been used to
suggest that they are not very effective and do not satisfy the concerns of the neighborhood concerning
speed mitigation,
Community Concerns/City Policies
Staff has concerns about the proliferation of signs and pavement markings throughout our city's
roadways. Initial installation costs are low but yearly maintenance is expensive (particularly for
pavement markings which are maintained every few years). In addition, this type of sign/marking
`pollution" can detract from. other important messages which staff is trying to convey to the driving
public. To date, there are over 6,000 signs placed throughout Boulder.
-C18-
SECTION C6: STOP SIGNS
DESCRIPTION
STOP signs are regulatory devices used by the city to assign right-of-way. The installation of two
additional STOP signs at an intersection costs approximately $200 per intersection. The addition of
two STOP bars (pavement markings) to clarify the stopping location, costs an additional $300 per
intersection.
EFFECTIVENESS AS A MITIGATION TOOL.
Sneed Reduction
Transportation engineers have debated the effectiveness of STOP signs as speed control for many
years. Considerable research into this area has been done in other jurisdictions.
In 1992, the City of Portland, Ore., performed an extensive evaluation of their STOP sign installation
warrants. As part of that evaluation, the city looked at the effectiveness of STOP signs as speed
mitigation when they were placed at locations which would not typically be considered for STOP sign
control. It found that the use of multi-way stop control MAY result in a decrease in vehicle speeds.
However, there were a sufficient number of negative tradeoffs (poor compliance, increased accident
rates, etc.) that the City of Portland did not recommend using STOP signs as speed control devices.
The City of Pueblo, Colo., recently studied the effects of STOP signs as speed control devices on
several of their intersections. The results showed that vehicle speeds were slower within 150 feet of
the STOP signs, but after that, vehicle speeds went up to the same speeds as before or higher. This
lack of effectiveness, coupled with poor compliance at the STOP signs, resulted in the City of Pueblo
concluding that STOP signs were not an effective or appropriate speed mitigation tool.
City of Boulder staff have studied the effect of STOP signs as a speed mitigation device and found that
they can be effective in certain circumstances. Typically, when the STOP signs are warranted due to
comparable main street and side street volumes, compliance is good and speeds in the area are lower.
When the STOP signs are not warranted, compliance goes down and the effectiveness of the device
decreases as well. In addition, vehicles tend to speed up between the devices, possibly to make up for
what they perceive to be "lost time" at the STOP signs.
City research, published in 1994, documented speed data in the vicinity of six multi-way stop
controlled intersections in the City of Boulder. In all cases, the speeds were well in excess of the
posted speed limit. On the average, the average speed was 3 mph over the speed limit while the 85th
percentile speed was 8 mph over the speed limit. These are speed profiles similar to or higher than
most roadways currently in the NTMP.
In 1997, city staff collected speed data at several locations where multi-way stop control was recently
installed. In 1998, Staff performed tests on two trial locations where multi-way STOP sign control was
temporarily installed. The data for these tests showed the use of STOP signs as a speed mitigation
device had varied degrees of effectiveness. Table C6-1 outlines the results of the before and after
speed data study.
-C19-
TABLE C6-1: Speed Data Comparison Before and After Stop Control Installation
.a eaxix t,, i :
.AYCTBQC speed ~95thTelCentllC speed
Pro ect Street Before t1 er k Giiarigef efore g `
1 ,f ,B % p Chan e
N. Ninth Street (n/o Evergreen) 20 mph 24 mph +4 mph 23 mph 28 mph +5 mph
13th Street (n/o High ) 26 mph 24 mph -2 mph 33 mph 28 mph -5 mph
Ninth Street (n/o University) 29 mph 28 mph -1 mph 33 mph 33 mph 0 mph
Bear Mountain (n/o Wildwood) 28 mph 28 mph 0 mph 32 mph 32 mph 0 mph
Pawnee Drive (n/o Thunderbird) 27 mph 24 mph -3 mph 31 mph 28 mph -3 mph
(The methods used to collect traffic data are not 100 percent accurate. There are enough daily fluctuations on roadway
speed/volume and limitations in the technology used to collect the data that the information should be used for comparative
purposes only.)
Staff also looked at the approach speeds of vehicles entering a multi-way stop controlled intersection to
determine the speed reduction benefit near the intersection. The results showed surprisingly little
reduction in speed within 50 feet of the STOP sign at these intersections. Table C6-2 outlines the
results of this study.
TABLE C6-2: Vehicle Speeds Within 50 Feet of Stop Signs
Aaerdge approaeh 85th prcntrlappoaeh XvlamJstdetreet
Intersectton :>p ed (CSO feet}peeOee} Vsilume distrtbuttoiV
13th & High 18 mph 20 mph 4 to 1
Wonderland & Poplar 18 mph 20 mph Unknown
Manhattan & Cimmaron 19 mph 22 mph >20 to 1
Ninth & Dellwood 18 mph 21 mph 3 to 1
* volume factored to account for high numbers of pedestrians
Public Innut
City staff has received many requests to consider using STOP signs for speed control as part of the
NTMP. In general, there has been very little concern expressed by the public concerning the general
use of STOP signs in this fashion. The Transportation Division has generally not been using multi-
way stop control unless it is warranted by high traffic volume, high accidents or sight distance
restrictions. Given the limited role that STOP signs have played in traffic mitigation, there has been
little reason to expect much comment from the public about the general use of STOP signs as traffic
mitigation.
In 1998, staff experimented with multi-way stop control at two locations where main street and side
street volumes had a ratio of less than 5:1, and where potential traffic diversion would be unlikely to
occur or would likely occur on a higher classification street. Considerable public comment was
received as a result of both of these installations. There were many citizens who expressed support for
-C20-
the placement of the STOP signs. Many other citizens called to state that they did not supportlfavor
the use of STOP signs at these locations. Most of the comments acknowledged that the STOP signs
probably did little to reduce corridor speeds. However, there was also acknowledgment that the STOP
signs did slow down speeds in the direct vicinity of the STOP sign. There was support for the STOP
signs as a method to help facilitate pedestrian crossings and as a tool for discouraging traffic from
traveling through these neighborhood streets. Several comments discussed how the STOP signs made
it easier to traverse the intersection, especially for young children and elderly pedestrians.
Concern was expressed about the potential safety issues associated with the poor compliance at the
STOP signs. There were also some concern about noise and pollution impacts as a result of the STOP
signs. There was considerable concern expressed about the inappropriateness of this type of traffic
control at the locations selected. Several comments reflected the concern that the driver did not feel
that it was appropriate to have to stop at this location. At least one comment reflected the fact that the
driver was not speeding when they encountered the stop sign and was upset at being forced to stop
when it was clear there was no traffic on the side street.
There was a protest of the STOP signs on Bear Mountain Drive that lasted throughout the entire
demonstration. Drivers of vehicles passing through the STOP signs, at various hours of the day and
night, would honk their horns in protest. This prompted at least one citizen living near the intersection
to contact staff on several occasions and request that the signs be removed immediately.
SAFETY CONCERNS
Compliance
Compliance is another issue which has received considerable discussion and study by transportation
engineers.
In 1992, the City of Portland, Ore., studied the effects of STOP signs on several of their streets. The
results showed that compliance was very poor at locations which were not warranted by national
standards. As a result, Portland concluded that STOP signs should not be installed for the sole or
primary purpose of reducing speed due to the potential negative impact on traffic safety. A recent study
by the City of Pueblo, Colo., resulted in the same conclusions.
City of Boulder Transportation Engineering staff evaluate requests for STOP signs based on national
standards and professional judgement. Staff will post STOP signs at locations which meet the national
standards in the Manual on Uniform Traffic Control Devices (MUTCD) and/or seem safe and
reasonable.
Since the creation of the NTMP, staff has been asked to experiment with multi-way stop control at
several locations which did not meet MUTCD standards, primarily for the purpose of traffic mitigation.
Transportation Engineering staff generally considers a STOP sign to be warranted when some
clarification to right-of-way is necessary (a low volume street intersecting a high volume street) or
when the physical characteristics of the intersection require that all vehicles stop to better provide for
safe operation. Generally, compliance at a STOP sign is better when the amount of traffic on the main
street and side streets are relatively high and when the volumes on the main street and side streets are
similar. This is due to the fact that the probability of a main street vehicle encountering a side street
vehicle at the intersection is higher. When these situations do not exist, lack of compliance can occur.
-C21-
In 1994, the City of Boulder published an article (January 1994 - Institute of Transportation Engineers
(ITE) Journal) detailing findings on several aspects of multi-way STOP sign affects, at intersections in
the City of Boulder. Between 0 percent and 16 percent of all traffic entering the 13 intersections
studied ran the STOP signs on the main street. The percentage was higher for T-intersections than
four-way intersections and also for side streets with very low volumes. The study also showed that an
additional 60 percent to 89 percent of all traffic on the main street rolled through the intersection,
traveling at 5 MPH or less.
In 1997, Transportation staff performed similar compliance studies on several intersections with multi-
way stop control. In 1998, staff performed similar compliance studies on the two intersections for
which experimental multi-way stop control had been placed. Findings showed a higher rate of
compliance than the 1994 study. However, there were still a high number of vehicles who ran the
STOP signs (similar numbers to the 1994 study). The first three locations in Table C6-3 were made
multi-way stop controlled intersections within the last year. The next four locations have been multi-
way stop controlled for several years. The last two locations were the two intersections where staff
experimented with multi-way stop control in 1998.
TABLE C6-3: Stop Sign Compliance
' , d r ; „ ,5 ' e Matril de 9",
U`j ° , Al~f'ealc PereeRt P1~ a` ~~erCCRtt~~ `~txeet,yolnme
lIltetSeP.loR i ti s gtIIS r?..TQfdl.. -tif',$k(?RS- ~,dfQfal? + orShYbU#Ifi7R'
Ninth St & Forest Ave 6 vehicles 5% 13 vehicles 9% 2.5 to 1
Ninth St & Dellwood Ave 1 vehicle <1% 6 vehicles 2% 3 to 1
13th St & High St 4 vehicles 1% 7 vehicles 2% 4 to I*
Wonderland & Poplar 8 vehicles 8% 27 vehicles 16% 5 to I
Manhattan & Cimmaron 8 vehicles 5% 12 vehicles 9% >20 to 1
Ninth St & Maxwell Ave. 4 vehicles <1% 0 vehicles 0% 4 to 1
Sixth St & Arapahoe Ave. 2 vehicles 1% 6 vehicles 4% Even
Bear Mountain & Wildwood 16 vehicles 14% N/A N/A 4 to 1
Pawnee & Thunderbird 26 vehicles 10% N/A N/A 2 to 1
* volume factored to account for high numbers of pedestrians
Most of the intersections chosen for this study did not have traffic volumes which would warrant multi-
way stop control using national standards. The intersection of 13th Street and High Street received
multi-way stop control to facilitate crossing of the children at Casey Middle School and Sacred Heart
School. Both of the North Ninth Street intersections and the intersection of Manhattan and Cimmaron
had multi-way stop~control installed as traffic mitigation. The intersections of Wonderland and Poplar
and Ninth and Maxwell both have sight distance issues at the intersection which suggest the need for
multi-way stop control. City records do not discuss the purpose for placing multi-way stop control at
-C22-
the intersection of Sixth and Arapahoe. Bear Mountain and Wildwood, and Pawnee and Thunderbird
were temporary installations and have since been removed.
Accident trends
In 1992, the City of Portland, Ore., studied the effects of STOP signs on several of their intersections.
They found that when multi-way stop control was installed at locations which did not meet MUTCD
national warrants, accidents typically increased. As a result, the City of Portland concluded that STOP
signs should not be used for the sole and primary purpose of speed control due to the potential negative
impact on traffic safety.
In some situations, the Transportation Division does recommend the use of STOP signs to mitigate
safety problems. Generally, these are problems of sight distance or right-of-way control which are
resulting in accidents. In 1996, the city installed multi-way stop control at the intersection of Ninth
Street and University Avenue, to mitigate a right-angle collision accident problem. Two years later,
the number of right-angle collisions had decreased but the number of rear-end collisions increased.
The severity of accidents at the intersections dropped considerably.
The use of multi-way stop control typically increases the total number of accidents at an intersection
while decreasing the severity of those accidents. Staff uses STOP signs when they believe that severe
accidents, which often result in major injuries, can be avoided or when such a changes would result in
conditions to reduce speeds and reduce the chance of accidents with injuries.
Compliance issues outlined in the section above bring into question the safety of using STOP signs,
when compliance is likely to be low. The probability of a serious accident occurring increases when
vehicles are ignoring the traffic control which defines right-of-way in the intersection. Vehicles which
are relying on that traffic control to make good decisions about when to proceed are put at risk when
other vehicles ignore the same traffic control.
Staff has reviewed the accident history at several intersections which have multi-way stop control. The
findings are inconclusive. Some locations have relatively high number of accidents with multi-way
stop control while others have little or no accidents. These findings are consistent with national
findings. The only common trend seemed to be that the higher the main street volume, the higher the
number of accidents. The City of Boulder studies accident rates at intersections through the "Hazard
Elimination Program." An accident rate which is higher than 2.0 accidents per million entering
vehicles is considered to be unusually high and merits evaluation of possible accident reducing
mitigation measures.
The results of Table C64 show that the intersection of Ninth and Maxwell has a relatively high rate for
a residential intersection. A rate of 1.8 accidents per million entering vehicles would probably put this
intersection in the top 30 locations in the City. The intersections of Balsam/19th and Walnut/33rd also
have high accident rates which would put them in the top 50 locations in the city. The intersections of
Wonderland/Poplar and Sixth/Arapahoe have moderate accident rates; the intersection of
Manhattan/Cimmaron has been without incident in the time period studied.
-C23-
TABLE C64: Accident Rates at Multi-way Stop Intersections
-ant street
Acazdent§rei'yar Acid)`'tior
Ihfersech6n} yQar huerage} entertngvuoles otusle dtstnbutton=
Ninth & Maxwell 5 1.8 4 to 1
Sixth & Arapahoe 2 1.3 Even
33rd & Walnut 4 1.5 3 to I
Balsam & 19th 5 1.5 1.5 to 1
Wonderland & Poplar 1 1.3 4 to 1
Manhattan & Cimmart 0 0.0 >20 to 1
Bicyclists and Pedestrians
Bicyclists and pedestrians using intersections with multi-way stop control should be at no more risk
than any other user, provided the traffic control is receiving good compliance. However, staff's
observations show that many bicyclists do not comply with the STOP signs themselves and therefore
put themselves at risk from side street traffic. Pedestrians are at risk when vehicles do not comply with
STOP signs, because pedestrians are less likely to be seen by an oncoming car.
Emergency Response
Delay is often experienced whenever an ambulance or fire engine must pass through an intersection
which they do not have the inherent right-of-way. In 1998, city staff studied the effects of STOP signs
on emergency response. Studies were done at two low volume intersections where trial STOP signs
had been installed. These studies were done in the off-peak time periods. The results of these studies
showed that the delay to emergency response in this situation was negligible (less than one second on
average). The only time-significant delay occurred was when there were two or more other vehicles at
the intersection, and this condition did not occur often.
In 1999, staff studied the effects of STOP signs at a higher volume intersection. Studies were done at
the intersection of 19th Street and Balsam Avenue during the p.m. peak hour. These studies showed
that there was a considerable amount of emergency response delay resulting from the traffic control.
Delay at the intersection varied between 5 and 12 seconds. The fire engine used in the experiment
could not approach the intersection as fast as they would in an uncongested condition, and they had to
slow to an average of 12 mph as they traversed the intersection itself.
The impact that STOP signs have on emergency response is a varied effect. It seems to depend largely
upon the traffic volume, the time of day and the geometrics of the intersection (whether there are
parking aisles, neckdowns, etc...). However, it appears that on any roadway where there is higher
traffic volumes (1,000 to 2,000 vehicles per day or higher), there is likely to be emergency response
delay during some portions of the day, as a result of STOP signs.
The Fire Department has indicated that multi-way stop control may be acceptable from an emergency
response perspective on some lower volume roadways but should not be used on higher volume
-C24-
roadways as it will cause a delay in their emergency response time. The Fire Department's Master
Plan states that any increase of delay upon their emergency response routes is "unacceptable."
ENVIRONMENTAL CONCERNS
The use of different mitigation treatments has several environmental/quality of life issues associated
with them. Specifically, the effects of slowing, stopping, idling and accelerating increase automobile
exhaust, noise and fuel consumption.
VYthicle Emissions
The act of stopping one's vehicle and accelerating back to speed has an impact on the amount of
emissions put out by the vehicle. A study of 10 multi-way stop controlled intersections in Michigan
showed that the average additional emissions per intersection each year was approximately 128,750
pounds of carbon monoxide, 7,920 pounds of hydrocarbons and 8,300 pounds of oxides of nitrogen.
Using carbon monoxide emissions as an example, this represents an increase of approximately 0.2
percent per intersection per year over the annual emissions of Boulder Valley. Use of stop control as
speed mitigation on all NTMP streets would likely result in as many as 50 additional multi-way stop
control intersections. This would result in an increase of almost 7.5 percent of the annual emissions of
Boulder Valley or an additional 8.8 tons per average winter weekday.
A significant increase in emissions would be in direct conflict with the goals of the Transportation
Master Plan which attempts to lower the tons of carbon monoxide emissions from 118 tons per day to
between 35 to 60 tons per day.
N ise
The slowing and accelerating associated with any of the mitigation devices will result in localized
noise impacts in the vicinity of the device. Since STOP signs require the greatest degree of slowing
and acceleration to resume speed, the noise impacts are greatest with the use of this device (over the
others discussed in this document). Also, since STOP signs increase the amount of time that vehicles
are present at the intersection this will prolong the length that the additional noise is present.
Fuel Consumption/Resource Impacts
Fuel consumption and resource impacts are other environmental factors to be considered. A study
performed in California in 1982 found that the deceleration and acceleration for each stop that an
average passenger car makes would result in an increased consumption of 0.0173 gallons of fuel. On a
roadway carrying 10,000 vehicles per day, this would translate to 173 gallons of fuel per day or 63,145
gallons per year. Using the example above of 50 intersections over the course of the NTMP, this
would be an additional 3,157,250 gallons per year in the City of Boulder at a cost of over $4.0 million
to be absorbed by the users of these intersections.
PROGRAM RESTRICTIONS
Diversion of Traffic to adjacent streets
Staff does not have sufficient data to draw a conclusion about the diversion potential for STOP signs
when used as a speed control device. Staff has not recommended employing STOP signs as a speed
control device at any location in the city where there was not a logical higher classification street to
which traffic could be diverted. A specific example is Ninth Street with Broadway as a parallel facility
to handle diversion.
-C25-
OTHER CONSIDERATIONS
Community Concems/City Policies
The City of Boulder follows national standards for STOP sign installation, published by the Federal
Highway Administration, which has been adopted by ordinance as official policy concerning the use of
traffic control devices. This document discusses the "appropriate" uses for STOP signs and multi-way
stop control.
Specifically, it cites issues with right-of-way, high side street traffic volumes and documentable safety
issues as reasons to employ STOP signs. These national standards state that "STOP signs should not
be used for speed control." Concerning the placement of multi-way stop control, national standards
state that it should ordinarily be used only where the volume of traffic on the intersecting roadways is
approximately equal.
-C26-
SECTION C7: PHOTO RADAR ENFORCEMENT
DESCRIPTION
Photo-radar is an automated speed enforcement device using radar to determine the speed of vehicles
interconnected with cameras to photograph vehicles traveling faster than the set violation threshold. In
the City of Boulder photo-radar demonstration, the device is mounted in a van and deployed at
different locations throughout the day.
EFFECTIVENESS AS A MITIGATION TOOL
5peed Reduction
The photo-radar unit is an excellent data collection device, logging the speed of all vehicles observed;
speeding or not. An examination of data compiled by the photo-radar device during deployment was
conducted. The speeding and violation percentages for deployments throughout the demonstration
time line were analyzed. The analysis was performed on individual streets and streets grouped by
functional characteristics. Functional groups included:
Arterials;
Collectors;
r Locals;
School Zones; and,
Critical Emergency Response Routes (CERR)
Results of this analysis indicate that on a majority of the streets and street groups, speeding and
violation percentages are decreasing over time as continued deployment occurs. On a somewhat
unusual note, the percent speeding (those going above the posted speed limit) seemed to be more
impacted than the violation percentage (those going above the threshold which triggers the photo radar
equipment). However, there were enough locations that did not show a reduction in speeding or
violations to make the results of this analysis conclusive. Additionally, since the photo-radar unit is
performing the data collection, its presence may have influenced driver behavior and the data as a
result.
Staff conducted separate independent speed studies on five neighborhood streets in June 1998 before
implementation of the photo-radar demonstration. After six months of deployment, follow-up studies
were performed in December 1998 without the photo-radar unit present to evaluate whether the
residual affects of photo-radar had reduced vehicle speeds. (See Table C7-1 below) Results of these
studies do not indicate that efforts to date have reduced vehicle speeds on these facilities. On each of
the streets studied, there was no change to the average or 85th percentile speed.
-C27-
TABLE C7-1: Speed Data Comparison Before and After Photo Radar Implementation
-ifi to Average , to*. a}tlr a centtle p
s eed ...a
- a` v t v, .v i ..per ;i s
PIOeetJlreet at a f tu` sr+ v _1-.
ai. q;$e€ore it `After '!.}rati_a~rAQc ~Ai=ter Ghange:
Ninth Street (2100 block) 27 mph 28 mph +1 mph 31 mph 32 mph +1 mph
Mapleton (2200 block) 24 mph 25 mph +1 mph 30 mph 30 mph 0 mph
Balsam (1700 block) 29 mph 29 mph 0 mph 33 mph 33 mph 0 mph
Brooklawn (900 block) 26 mph 28 mph +2 mph 31 mph 33 mph +2 mph
Alpine (1400 block) 29 mph 30 mph +l mph 34 mph 34 mph 0 mph
(The methods used to collect traffic data are not 100 percent accurate. There are enough daily fluctuations on roadway
speed/volume and limitations in the technology used to collect the data that the information should be used for comparative
purposes only.)
In conclusion, after six months of citation deployment there is no statistically conclusive evidence that
the photo-radar demonstration to date has reduced speeding in the City of Boulder.
Limitations of our data collection and the short time frame studied may not provide an accurate picture
of the device's effectiveness. In addition, staff does not have a clear sense yet of what level of photo
radar enforcement would likely begin to change driver behavior. At approximately 1,000 to 1,500
citations issued per month, there is still quite a bit of the population that has not had any experience
with photo radar enforcement. As time passes, and more and more citations are issued, people who
speed frequently on neighborhood streets should be receiving their second (third, fourth, etc )
citation, and perhaps the cumulative effect of multiple violations will affect their driving behavior
when a single ticket (or none) would not. It is possible that photo radar would yield a higher success
with a change in either the number of units deployed in the city or longer term use of this tool (i.e., the
use of this tool for a year rather than six months may show different results).
Public Input/Community Support
In February 1998, the city set up a phone message line for citizens to ask questions and give their
thoughts on photo radar. Since February, thousands of messages have been left on the phone line with
both support and nonsupport for the use of photo radar. More of the input received was opposed to
photo radar rather than for support of photo radar. The number of daily calls received has varied since
the message line was set up. Higher volumes of calls occurred when the message line was first set up,
photo radar began and after articles were in the press. Citizens have also sent letters regarding their
support or nonsupport for photo radar.
To gauge public support of the use of automated enforcement techniques (photo-radar and photo red
light) statistically significantly surveys were commissioned through CPPA. A survey was performed
in June 1998 prior to beginning the demonstration to gauge initial sentiment. A follow-up survey was
performed in November 1998 to reassess public support and determine if public opinion had shifted. A
short summary is provided below.
-C28-
;u P-1:5- Jg~
F.i,
Do you favor the use of Photo-radar in Boulder? June 1998 November 1998 Change
Strongly favor 20% 26% +6%
37% 48% +11%
Somewhat favor 17% 22% +5%
Neither favor nor oppose 9% 9% 5% 5% -4% 4%
Somewhat oppose 15% 16% +1%
54% 47% -7%
Strongly oppose 39% 31% -8%
Total 100% 100% -
Mean Rating (1= strongly oppose, 5= strongly favor) 2.1 2.4 +0.3
In the initial survey, the majority of respondents indicated that they were either "somewhat" or
"strongly" opposed to the use of photo-radar. The follow-up results indicate a statistically significant
shift in public support for the use of the technology, with respondents now equally split between favor
and opposition.
When asked why they favored or opposed the use of photo radar, approximately half said they favored
it because it may decrease speeding/accidents. Those who opposed photo radar were most likely to cite
fears of an invasion of privacy or "big brother;" more than 40 percent mentioned this reason in
November while approximately 30 percent of those opposed gave this reason in June.
Interestingly, while support for photo-radar increased from June to November, perceptions about the
effectiveness of the device in reducing speeding remained unchanged. When asked how effective they
believed photo radar would be in reducing speeding in Boulder, over half thought it would be very or
somewhat effective in both June and November.
SAFETY CONCERNS
The use of photo radar is very similar to traditional enforcement. The fundamental difference is that it
is automated. As such, staff does not believe there any significant safety issues associated with this
tool.
Bicyclists and Pedestrians
Staff does not believe that this tool has any negative impact of bicycle use or pedestrian activities.
Emergency Response
Staff does not believe that this tool has any negative impact on emergency response.
ENVIRONMENTAL CONCERNS
Since the use of photo radar speed enforcement is a passive mitigation device, it does not have several
of the environmental/quality of life issues associated with many of the physical mitigation devices or
-C29-
traffic control restrictions. Specifically, the photo radar unit does not cause artificial effects of slowing,
stopping, idling and accelerating, which in turn increase automobile exhaust, noise and fuel
consumption.
Vehicle Emissions
The photo radar unit does impact this area in one fashion. On particularly hot or cold days, the vehicle
is left running throughout the deployment to protect the officer monitoring the unit from significant
temperature impacts. This practice has a relatively small vehicle emissions impact.
Noise
Staff does not believe that the use of photo radar has any noise impacts.
Fuel Consumption/Resource ImImpacts
The photo radar unit impacts this area in one fashion. On particularly hot or cold days, the vehicle is
left running throughout the deployment to protect the officer monitoring the unit from significant
temperature impacts. This practice will have a relatively small fuel consumption impact.
PROGRAM RESTRICTIONS
Diversion of Traffic to adjacent streets
Staff does not believe that the use of photo radar has resulted in any significant traffic diversion.
OTHER CONSIDERATIONS
Community Concerns/City Policies
For further detail on citizen input and community concerns please refer to the Photo Radar Survey
Results which can be found in Appendix F of the March 30, 1999 TAB/City Council Joint Study
Session memorandum.
-C30-
SECTION C8: PHOTO RED LIGHT EXPERIMENT
DESCRIPTION
Photo red light is an automated enforcement tool used to monitor and ticket vehicles who run red lights
at specific intersections. The device involves the use of a mounted camera at an intersection and a
series of loop detectors in the roadway surface to monitor speed and the time of the vehicles approach,
compared to the phasing of the traffic signal. If the loops detect that a vehicle is entering the
intersection during the red phase for that approach movement, the camera takes a picture of the vehicle
and driver, and a citation is issued to the driver.
EFFECTIVENESS AS A MITIGATION TOOL
Red Light Violation Reduction
Violation percentage data was collected at six intersections, prior to the installation of the photo red
light cameras. The cameras were then deployed at three intersections (four approaches) in the
following locations: Valmont Road/47th Street (WB), 28th Street/Arapahoe Avenue (S/B and W/B)
and Table Mesa Drive/Foothills Parkway (WB). Data collection was conducted in January 1999 at
each of the intersections to see whether compliance had increased on: a) the approaches with photo red
light; b) other approaches in the same intersection; and c) approaches on other intersections without
photo red light enforcement.
The results of these studies are shown in Table C8-1. The violation data is summarized for all
movements in the intersection. The numbers represent violations occurring during one peak hour.
TABLE C8-1:
Ytolation,perceptageomp_rison Be#ore aril After Phttto Red Lrghf lmplementahon "
k
Percment Red 1L
ghtV iolatio~r',
Intersectton= Hefore [fret ge
28th Street & Arapahoe Road 79 45 34
(photo red fight)
47th Street & Valmont Road 4 8 +4
(photo red light)
Table Mesa & Foothills/RTD 4 6 +2
(photo red light)
28th Street & Canyon Boulevard 39 51 +12
(no photo red light)
Ninth Street & Canyon 5 14 +9
Boulevard
(no photo red light)
-C31-
Baseline Road & Broadway 23 23 0
(no photo red light)
The results of this study suggest that there may be some reduction of red light violations at
intersections using photo red light. However, only one intersection with photo red light deployment
actually showed a reduction in violations. The data does not support the theory that use of photo red
light is decreasing red light violations at other intersections. None of the three intersections without
photo red light deployment showed any increase in compliance.
In addition, the photo red light cameras have been tracking violation information on the four
deployment approaches since they were activated in October 1998. Problems during the program start
up resulted in a lack of data for the month of October so there is only data available for November.
The violation information being tracked is summarized in Table C8-2.
TABLE C8-2:
Photo Red light Violation Data - November 1998
fnferseetto"pproaeh' „ rol txots
d rR .N { i ¢ d,
pec;eCl Perfeeted eue a"feF
Valmont & 47th Street - 291 26 8.9% 0.44 1.47
westbound approach per hour seconds
Table Mesa & Foothills/RTD 188 41 21.8% 0.26 1.83
- westbound approach per hour seconds
28th Street & Arapahoe - 242 77 31.8% 0.40 2.09
westbound approach per hour seconds
27.5% l.26 1.74
28th Street & Arapahoe - 894 T-246,
southbound approach per hour seconds
"Violations detected" are the number of vehicles observed running a red light on each approach during
the month. "Violations perfected" are the number of clear images with appropriate information that
could be used to issue citations. "Violation rate" is the rate at which detected violations become
perfected violations. "Violation frequency" is the number of red light violations detected per hour of
deployment.
The cameras were not functioning every hour of the month. The deployment time was influenced by
the need to change film in the camera and by camera malfunctions throughout the month. "Violation
time" is the average amount of time after the red light activated that violations occurred. Most traffic
signals in the City of Boulder have between one and two seconds of all red time (time where the traffic
signal is red in all directions) following each phase.
The data suggests that, on average, approximately one vehicle per two hours is running a red light at
these four intersections, and when they do so, they are typically entering the intersection between 1.5
seconds and 2.0 seconds after the light has changed.
-C32-
Public Input/Community Support
The opinion surveys conducted in June and November of 1998 regarding automated enforcement also
asked citizens about photo red light. Familiarity increased greatly since June, with the percentage of
respondents who had heard of photo red light increasing from 34 percent to 66 percent. Both the initial
and follow-up surveys indicate that the majority of respondents either "somewhat" or "strongly" favor
the use of photo red light. The follow-up survey indicates that this support has grown. A tabular
summary is provided below.
. - ~ PntrlicSvpporf'forttief3se;ufPhotv~ie~LgLtti~'~'-~.,,. ' Do you favor the use of photo Red light in Boulder? June 1998 November 1998 Change
Strongly favor 32% 39% +7%
57% 63% +6%
Somewhat favor 25% 24% -1%
Neither favor nor oppose 7% i% 5% 5% -2%
Somewhat oppose 13% 10% -3%
36°.~ 32% -4%
Strongly oppose 23% 22% -1%
Total 100% 100%
Mean Rating (1= strongly oppose, 5= strongly favor) 3.3 3.5 +0.2
The reasons most often given for supporting photo red light were increased safety and decreased
accidents as well as increased compliance with traffic laws. Over one third of respondents opposed
photo red light due to the fear of "big brother" or an invasion of privacy. Almost 60 percent of
respondents thought photo red light would be "somewhat" or "very effective" in reducing accidents
when asked in November. This represented a decline compared to June, when 70 percent of
respondents rated photo red light as effective.
SAFETY CONCERNS
Accident reduction
Several of the intersections chosen to receive the photo red light mitigation were selected due to a high
number of accidents resulting from red light violation. The number of accidents resulting from red
light violations were considered in all candidate intersections. Table C9-3 summarizes the average
accidents per month, prior to the deployment of photo red light and following the deployment of photo
red light at six of the candidate intersections. The intersections of 28th Street and Arapahoe Road,
471:4 Street and Valmont Road, and Table Mesa Drive and Foothills Parkway/RTD Driveway, each
have photo red light enforcement deployed on one or more approaches (a total of six positions). The
intersections of 28th Street and Canyon Boulevard, Sixth Street and Canyon Boulevard, and Baseline
Road and Broadway, were all candidate locations which did not receive photo red light deployment.
-C33-
TABLE C8-3:
Red I 1ght aeetdents CompansoQ before and After PhototedigttT"implementation'
"k , { 3 X99&Redltht' '`cctdentrates
a
„"F `~TTg, 'l`''t
iz ~';'..fi''~ uy#Ei F s a.;
y~sr y P'~
~L41.JG,4i"x'fa w~,n~... e!>re n, tiltE~*jl • ~i~*'.rt ft ~~a~.,.y a v
28th Street & Arapahoe Road 0.22 0.33 +0.11
(photo red light)
47th Street & Valmont Road 0.56 0.33 -0.23
(photo red light)
Table Mesa & Foothills/RTD 0.33 0.00 -0.33
(photo red light)
28th Street & Canyon Boulevard 0.67 0.67 None
(no photo red light)
Ninth Street & Canyon Boulevard 0.00 0.00 None
(no photo red light)
Baseline Road & Broadway 0.00 0.33 +033
(no photo red light)
There is a small trend towards a reduction in accidents resulting from red light violations at
intersections with photo red light deployed at the intersection. There does not appear to be any
reduction in these types of accidents at intersections without photo red light deployment. However,
with only three months of deployment, there is insufficient data to draw any statistically significant
conclusions on accident reduction benefit.
Compliance
The purpose of this device is to increase compliance when a vehicle encounters a red light. Staff feels
that there can only be beneficial safety results from deployment of photo red light enforcement.
Bicyclists and Pedestrians
Staff does not believe that this tool has any negative impact of bicycle use or pedestrian activities.
Emergency Response
Staff does not believe that this tool has any negative impact on emergency response.
ENVIRONMENTAL CONCERNS
Staff does not believe that this tool has any negative impact on air pollution, fuel consumption or noise
impacts.
PROGRAM RESTRICTIONS
Diversion of Traffic to adjacent streets
Staff does not believe that the use of this tool has resulted in any significant traffic diversion.
-C34-
SECTION C9: SPEED SENSITIVE TRAFFIC SIGNALS
DESCRIPTION
R&st in All-Red Speed Sensitive Signal Operation
The speed sensitive traffic signal demonstration involved taking an existing traffic signal (the
intersection of Pine Street and 20th Street) and introducing an influence on the timing of the traffic
signal related to the approach speeds of the vehicles on the roadway. The traffic signal would have red
light indication in all directions, unless a vehicle was approaching the intersection. If that vehicle were
driving the speed limit (monitored by a series of inductive loops in the asphalt), then the signal
indication would change to green before the vehicle entered the intersection. If the vehicle was
speeding, then the signal indication would not change to green until well after the vehicle reached the
intersection. Signs indicating that speeding would influence the signal control were posted on
approaches to the intersection.
The "rest in all-red" speed sensitive traffic signal demonstration involved modifying an existing traffic
signal at the intersection of Pine Street and 20th Street to introduce an influence on the operation of the
traffic signal related to the approach speeds of the vehicles on the roadway. This location was chosen
in part because "rest in all-red" signal operation does not favor one street over the other, making it
appropriate for intersections of streets with similar traffic volumes, such as Pine Street and 20th Street.
The traffic signal was equipped with advance detection loops in addition to the standard detection
loops at the stop bars and was programmed to display a red light in all directions when no vehicles are
being detected. When a vehicle approached the intersection, the advance loops measured the vehicles
speed. If the vehicle is traveling at or below the 25 mph speed limit, the controller is notified of the
arriving vehicle. Unless another vehicle is already being served on the cross street, the signal turns
green before the vehicle arrives at the intersection, and it proceeds without delay. If the vehicle is
traveling above the speed limit, it is ignored until it stops at the stop bar, where it is detected in the
traditional manner.
Advance loops were also installed in the bike lanes on all four approaches to allow approaching
bicyclists to trigger a green light, and "SPEED SENSITIVE SIGNAL" signs were posted to indicate
that speeding would influence the signal operation.
EFFECTIVENESS AS A MITIGATION TOOL
Speed Reduction
Speed data collection was performed on Pine Street, east and west of the intersection, prior to the
change in signal operation. A follow-up study approximately four months after the signal change was
also performed. The results of these studies are shown in Table C9-1.
-C35-
TABLE C9-1: Speed Data Comparison Before and After Photo Radar Implementation
z+ bS1Gejlf.~?,,'°, ai~s~'kEStC'rrienIei'. speed,
v & aI i~.^F.° 'F ,s'-4F 7 '~~eyayY ~.i `max'
rxF F7 c ,rte- J c fia { rrfi a
pjL
ha (BefpB e hang ?:~efOYe er t s Change
Plcr~eottree
Pine Street (1900 block) 25 mph 25 mph 0 mph 29 mph 29 mph 0 mph
Pine Street (2000 block) 27 mph 26 mph -1 mph 31 mph 31 mph 0 mph
(The methods used to collect traffic data are not 100 percent accurate. There are enough daily fluctuations on roadway
speed/volume and limitations in the technology used to collect the data that the information should be used for comparative
purposes only.)
It should be noted that both of these sections of Pine Street also received photo radar enforcement
during the speed sensitive signal experiment. The data in Table C9-1 suggests that the speed sensitive
signal experiment did not change vehicle speeds.
Public lnl2
The input received on this speed mitigation demonstration was mostly positive. Reasons cited for
support included no emergency vehicle delay, doesn't endanger cyclists and it punishes or rewards
drivers immediately. Concerns included expense and possible air quality impacts.
SAFETY CONCERNS
Compliance
Some concern was expressed that the use of a traffic control device in this fashion would result in less
compliance with the red light indication and ultimately result in a less safe operation of the traffic
signal. Red light compliance data was collected on Pine Street, during the AM and PM peak hours,
prior to the change in signal operation. A follow-up study was performed during the experiment. The
results of this study showed that the overall number of red light violations did not increase during the
experiment. Staff does not believe that the use of this tool results in any significant safety impacts. .
Bicyclists and Pedestrians
Staff does not believe that this tool has any negative impact of bicycle use or pedestrian activities.
Emergency Response
Staff does not believe that this tool has any negative impact on emergency response.
ENVIRONMENTAL CONCERNS
See comments for speed humps, Section C 1.
PROGRAM RESTRICTIONS
Diversion of Traffic to adjacent streets
Staff does not believe that the use of this tool has resulted in any significant traffic diversion.
-C36-
REST IN GREEN SPEED SENSITIVE SIGNAL OPERATION
An alternative to the "rest in all-red" type of speed sensitive signal is "rest in main street green." This
variation of speed sensitive signal operation may be more appropriate at intersections of streets with
unequal traffic volumes where there is a higher volume "main street" and a lower volume "side street."
As with typical traffic signal operation, a "rest in main street green" speed sensitive signal normally
displays a green light for the main street unless traffic is detected on the side street. However, a "rest
in main street green" signal has detection loops on the main street to measure the speed of approaching
vehicles and turns the main street signal red when a vehicle traveling over the speed limit is detected,
requiring it to stop.
The existing traffic signal at Baseline Road and 17th Street was identified by staff as an appropriate
location for a demonstration of "rest in main street green" speed sensitive signal operation. However,
limitations in the existing traffic signal controller software used in Boulder have prevented the
demonstration from being implemented. The controller software is scheduled to be replaced in 1999 as
part of the signal system upgrade, and the replacement software is currently under development. Staff
is seeking to include the capability for "rest in main street green" operation in the new software and
will proceed with the demonstration when it is completed and installed later this year.
-C37-