Tuesday, January 13, 2009

Clipper Street/Portola Drive neighbors: Public comment on the Bicycle Plan EIR

Clipper Street/Portola Drive area neighbors
January 11, 2008

Bill Wycko
Environmental Review Officer
San Francisco Planning Department
1650 Mission Street, Suite 400
San Francisco, CA 94103

Dear Mr Wycko:

This letter contains our response to the release of the San Francisco Bicycle Plan update and associated EIR. This letter is prepared and sent before the closing of the comment period of January 13, 2008.

The inclusion of 60 projects as a “project” in this document is inappropriate, as each project should be should be carefully designed with community participation through a detailed process and documented separately. A document this large is not only awkward, but also does not allow for adequate discussion of bicycle safety. For example, a current controversy at Octavia Boulevard and Market Street is an example of how unsafe and messy results can occur when bicycle projects are rushed without careful design.

Several of the proposals in this report significantly disrupt local traffic and buses, greatly increase greenhouse gas emissions due to delayed and rerouted vehicles, and have not been studied in sufficient depth to justify the proposed designs; others are simple, logical projects. There are many intersections not studied (especially in the A.M. peak hour) which should be studied as these projects will significantly affect the neighborhoods where the new delay will be created. Each project should be designed and evaluated carefully.

To comply with the requirements of the EIR comments and responses, we are addressing specific technical concerns and mistakes identified in the EIR. Addressing these will likely require major changes to the EIR document, and we suspect that a recirculation will be likely. One alternative may be to remove the “projects” from this document, and present those as separate studies. This would allow for more adequate studies to be made on the proposed projects and for better designs to evolve.

General Comments on Project Level Analysis of the EIR

Reports of delay at Level of Service F at “>80” seconds for traffic inadequately describes the actual delay being induced by the project. This is also inconsistent with the transit analysis methodology in the EIR, which discusses use of intersection delays of up to 100 seconds in those calculations. The Transportation Impact Analysis Guidelines for Environmental Review requires disclosure of all volume-capacity ratios at Levels of Service E or F; these are not provided and should be to bring the document into compliance. The use of “>80” is inaccurately portrays the impacts of the lane reductions on traffic. The EIR should be recirculated to show the actual estimated intersection delay, and not merely the anticipated delays as “>80” seconds.

We also request that the comment and response specifically disclose the amount of anticipated delay to the nearest second so that the decision-makers and citizens in San Francisco have full knowledge of the actual delay that they will soon experience. The Highway Capacity Manual and accompanying available software analysis packages report actual anticipated delay significantly over 80 seconds. The City Transportation Impact Analysis Guidelines for Environmental Review posted by the Planning Department require the reporting of volume-to-capacity ratios at Level of Service E or F; these are not reported, recognizing that high delays should be further illustrated---while this EIR introduces LESS technical descriptions of the effect of congestion. The EIR further discloses on Page V.A.3-15 through V.A.3-17 (transit impacts discussion) that intersection delays of 100 seconds are discussed as central to the analysis; more detailed delay information IS AVAILABLE AND IS USED IN OTHER PARTS OF THE EIR. Further, Figure V.A.3-3 (referencing the relationship between volume/capacity ratio and taken from the 2000 Highway Capacity Manual, suggest that the analysis should be able to report delay of up to over 700 seconds (over 11 minutes), so that the vehicular traffic results ARE NOT CONSISTENT WITH THE METHODOLOGIES PRESENTED IN OTHER SECONDS when they are presented as only “>80”.

The Planning Department Guidelines require that any “project” that affects any intersection over Level of Service D must have a published report that fulfills the requirements of these guidelines (page 1) of the Transportation Impact Analysis Guidelines for Environmental Review at:

These guidelines also require that the volume-capacity ratios be reported for every intersection that operates at Levels of Service E or F. There are many intersections in this report that indicate that this objective is met. The quantitative effect of the reduced capacity on to the intersection Level of Service must be more extensively documented, as set forth in the published City Guidelines for traffic studies and EIRs.

The impacts should be recirculated to the public with the actual intersection delays reported for the wider public. These delays must be reported at least 100 seconds to be consistent with the transit impacts, and should be reported to be at least at delays greater than two signal cycle lengths of the approaching intersections (which suggest that delays of up to 180 seconds should be reported if the intersection has a 90 second cycle). Otherwise, the analysis reported in this Draft EIR are inadequate, inconsistent with the City’s own Transportation Impact Analysis Guidelines for Environmental Review, and do not accurately disclose the true environmental impact of the Bicycle Plan.

Queue lengths are a required consideration in the design of any street project. This EIR does not report these lengths, and is thus an inadequate Project Level report for discussion and decision-making purposes. Disclosure of traffic queue lengths of approaches with lane reductions should be reported, especially where the reductions are significant and lead to Level of Service F operations. Adjacent property owners have the right to know whether or not the bicycle plan will result in queued traffic being introduced past the front of my property. The public cannot determine any additional queue lengths that would result from the reduction of lanes. The public cannot determine whether or not the additional queues will disrupt adjacent intersections. Idling vehicles results in significant carbon monoxide emissions, which have been shown to have detrimental health effects. The introduction of additional feet of carbon monoxide represents an additional hazard, not only to adjacent properties, but to pedestrians, bicyclists and other users that must wait in the additional idled traffic. The project level analysis should report queue lengths that result from lane reductions.

The transit delay threshold of 6 minutes is too high, arbitrary and inadequately reports the impacts of additional traffic on Muni routes. Further, this is inconsistent with the analysis methodology in the Transportation Impact Analysis Guidelines for Environmental Review published by the City Planning Department, which requires the reporting of effects on the overall system capacity, and defined Transit Levels of Service. The EIR should be modified and recirculated to report the additional delay impacts on system capacity and Transit Levels of Service, and should use Transit Level of Service based-threshold (which would be substantially less than 6 minutes). There is a direct relationship between transit speed and capacity. If a bus route is forecast to experience additional delays and the number of buses assigned to a route is fixed, then the additional travel time will effectively reduce the capacity of the bus system. For example, a 60-minute round trip route with a 10-minute headway would normally have 6 buses assigned to that route during that peak hour. If delay was only an additional 5 minutes for that hour (50 seconds per bus), this would represent the need to add “a half of bus” to the route or to reduce the headways of the current buses. This represents 19 percent DECREASE in the carrying capacity of that Muni route. The Planning Department Guidelines require that any “project” that affects any intersection over LOS D must have a published report that fulfills the requirements of these guidelines (page 1 of the Transportation Impact Analysis Guidelines for Environmental Review at:

There are many intersections in this report that indicate that this objective is met. The effect of the reduced capacity on the Transit Level of Service must be documented, as set forth in the published City Guidelines for traffic studies and EIRs.

The Greenhouse Gas Emissions of the additional delay and increased VMT that result from the significant lane reductions across the City is not discussed, and could represent a significant increase in the Greenhouse Gas Emissions created by mobile sources within San Francisco. This EIR fails to address Greenhouse Gas Emissions. The Mayor’s Office and the Board of Supervisors have indicated that this is an important priority for the City, yet there is no analysis within the EIR of how the additional idling and more circuitous routing of vehicles will increase these emissions within San Francisco. The negative impacts of additional traffic congestion to Greenhouse Gas Emissions should be disclosed.

All affected property owners should be notified of projects directly in front of their homes, which appears to be a Sunshine Ordinance Violation and Planning Department procedures. We did not receive notice of how my street would change. Our neighbors would have not known had we not actually studied the plan in detail. Planning Department EIRs require notification of all affected persons within a certain distance. This qualifies as a project, and is thus subject to these requirements.

Cluster 6 Project 6-2 Option 1 Analysis Comments

We believe that Project 6-2 Option 1 is an ill-conceived, badly designed, and congestion-inducing change to a major constraint point within the City’s transportation system, and is inadequately studied within the EIR. Strategies to provide a Class 1 or Class 2 bicycle lane are available without removing a traffic lane. Specific comments on this project and the accompanying EIR analysis are provided on the following pages.

Project 6-2 Option 1 should be removed from the San Francisco Bicycle Plan because it was developed AFTER the Notice of Preparation was issued and has not been presented in any neighborhood meetings or workshops, or scoping of appropriate intersections that should be studied. Project 6-2 Alternative 1 represents a significant modification to the Bicycle Plan made after the Notice of Preparation was issued on June 5, 2007. The change was not published until January 15, 2008. The first introduction of this project appears to be reported here:

As affected property owners, we have been given no notice about this proposed change which directly affects the roadway in front of my home. This project has not been properly developed, and has not been screened in widely-publicized public meetings in our neighborhood. Further, the impacts from Option 1 have been woefully unreported and have mistakes, and the significant impact of Option 1 should be more extensively studied, as presented below.

Project 6-2 Option 1 represents a major change to San Francisco’s transportation system and it not a minor modification to the Bicycle Plan. The reduction of the traffic movement from northbound Clipper Street to westbound Portola Drive is the sole traffic location that traffic directly can use between 18th Street (in the Castro Neighborhood) and O’Shaughnessy Boulevard (in the Glen Park neighborhood). Avoiding this intersection will require drivers to drive at least two miles of additional travel to use alternative routes, increasing local vehicle miles of travel and greenhouse gas emissions. This is THE single “bridge” across the Twin Peaks area between the east central and west central areas of the City. This intersection frequently has back-ups and queued traffic at both the AM and PM peak hours. A reduction of capacity by 50 percent at this intersection should be considered a major reduction in the overall capacity of the street system. It is similar to what would happen if 2.5 lanes of the Bay Bridge were removed for a 500 segment of roadway between Treasure Island and the remainder of San Francisco. The effects are profound for upstream traffic! Clearly, Project 6-2 should be considered in relation to the overall impact on the Citywide Circulation System. Further, drivers seeking to avoid the newly-created bottleneck will have to travel up to 3 miles out of direction (through either the Castro or Glen Park neighborhoods), increasing the impact of this project on greenhouse gas emissions contributed by the City of San Francisco.

Project 6-2 Option 1 is a discontinuous piece of the Bicycle Plan and is unsafe for bicyclists. Project 6-2 is an isolated set of bicycle lanes that are quite short and do not extend to a distance even as far as vehicles will be queued at this intersection. Bicycles will need to weave through queued traffic to reach them if Option 1 is implemented! As shown in diagrams in the Appendix of the EIR, they do not connect to proposed bicycle lanes on Clipper Street and they are running in only the westbound/northbound direction. The purpose and need for these lanes is clearly illogical because they do not connect to any other lanes and rather than encourage bicyclists sharing the roadway with vehicles, it will instead encourage bicyclists to weave between queued vehicles. Many of these vehicles will be queued through two signal cycles, encouraging more impatient behavior by the drivers in the vehicles.

Project 6-2 Option 1 does not analyze a newly-affected intersection currently operating at significant delays---Clipper Street/Diamond Heights Boulevard. The EIR is incomplete without studies at this intersection. This intersection, which currently has significant queuing, will likely experience much greater queuing and delay as traffic from Portola Drive/Clipper Street/Burnett Avenue intersection backs up into it at the PM peak hour. This will significantly increase idling delay for both vehicles and buses that travel through this intersection. It was not initially reasonable to request studies on this intersection, as the Notice of Preparation did not include the segment of Project 6-2 Option 1 between Diamond Heights Boulevard and Portola Drive, so that this intersection has not been identified as critical. The anticipated queues are not reported, so a reader is unable to determine the magnitude of the impact at this intersection. The EIR should be recirculated with this significantly-impacted intersection included.

The adoption of the Option 1 recommendation will likely lead to back-ups into and through this intersection and into adjacent neighborhoods. One probable outcome may be the requirement that this intersection also have a new traffic signal installed at this intersection. The cost of installing a traffic signal here, as well as the cost of operating the signal, and the cost of developing a coordinated signal system with signals at these two closely-spaced signals, must be disclosed as a probable outcome. The costs of installing a signal here will be significant, and can easily be avoided by lower-cost design mitigations, or by removing Option 1 from the bicycle plan. (Potential low-cost mitigations are presented below.)

Project 6-2 Option 1 should be considered in light of the effects during the A.M. peak hour at both affected intersections. The EIR is incomplete without an A.M. peak hour analysis, and the A.M. peak hour congestion appears to be much worse than the PM peak hour congestion. As a neighbor, I routinely witness vehicles needing 2 or 3 cycles to clear this Portola Drive/Clipper Street/Burnett Avenue during the A.M. peak hour. It appears that this movement has more congestion in the A.M. peak hour than in the P.M. peak hour. Traffic from the signal at this location backs up at least two to three blocks, and often extends past Duncan Street on northbound Diamond Heights Boulevard, and almost reaches High Street on westbound Clipper Street---well through the Clipper Street/Diamond Heights Boulevard intersection. This has not been previously identified as needing study as the Notice of Preparation issued for the plan did not include the lane reduction in this option. There is a significant impact to traffic flows at the A.M. peak hour when reducing this lane, and this has not been studied or reported in the EIR. Studies at the AM peak hour should be presented.

Project 6-2 Option 1 appears to have a significant transit impact for Projects 6-2 Option 1 and 6-5, and mistakes in the calculation are presented in the EIR; this section must be corrected and the corrections should include a more detailed discussion of how the impact was calculated to fully understand where the error is located. The transit impacts discussed in the Bicycle Plan EIR on Page V.A.3-645 and V.A. 3-546 are in error. The report indicates that delay is 3.4 minutes “for each route” (Routes 48 and 52) then proceeds to report a cumulative delay also at 3.4 minutes. If each route is forecast to experience a 3.4 minute delay, the combined impact would be 6.8 minutes---which then becomes a significant impact.

The report inaccurately states that the Route 52 operates at a 15-minute headway, when it actually operates at a 10-minute headway during the time period used for the analysis (P.M. peak hour). (The 15-minute headway is the condition during the A.M. peak hour.)

Further, the analysis states that it is based on delays in one direction. However, the level of service for the adjacent intersection is reported as an average for all movements in the intersection. It is improper to discuss transit delay only in one direction for what is an average condition at the intersection. The delays should either be analyzed for that specific approach (in which case one direction would be fine) or the delay should be calculated as if the bus route passes through in both directions. This is a significant math error in this instance, as the author is mixing overall intersection delay with approach delay; this significantly underreports the impacts to the transit system. Correcting this math error would result in a peak hour impact of either 6.8 or 13.6 minutes for transit service, depending on how the inconsistency in the report presented in the above paragraph is explained.

As defined in the Highway Capacity Manual, if the LOS goes from E to F, queued traffic will not be able to clear the intersection, including buses. If every bus will miss an entire signal cycle, this will result in at least 60 seconds of delay per bus to allow for the Portola Drive traffic to move through the intersection. If there are 11 buses at peak hour having to wait 60 additional seconds, this is an impact of 11 minutes total at peak hour, which exceeds the 6 minutes of delay at peak hour criteria established in the methodology. Clearly, this impact in the EIR is underestimated and the analysis of the potential delays from this project are clearly too little, and this represents a significant impact well above and beyond the artificial 6 minute threshold presented in the EIR criteria.

Transit will also be impacted by additional delays discussed previously at the Clipper Street/Diamond Heights Boulevard. This intersection, which currently has significant queuing, will likely experience much greater queuing and delay as traffic from the Portola Drive/Clipper Street/Burnett Avenue intersection backs up into it at both the AM and PM peak hours; this will significantly increase idling delay for buses that travel through this intersection. This additional delay should be reported in the transit impacts and a determination of whether or not this will further deteriorate transit speed and reliability should be further disclosed.

The impacts of this project to the Transit Level of Service, required in the Planning Department Transportation Impact Analysis Guidelines for Environmental Review, are not discussed. As noted in the general comments above, additional delay has an effect on transit capacity, and this effect is not presented for this project. The Transit Level of Service calculations should be presented in order to fulfill the requirements of these guidelines.

The three routes in this area---Routes 37, 48 and 52---have packed buses at peak hours. Standees are common and sometimes riders are actually unable to board buses. Increasing bus travel times would increase overcrowding on these line, as the slower speeds would mean that bus frequencies would have to be decreased. This could also jeopardize the recent Muni restructuring proposal, which has bus routes carefully designed to be able to operate within certain headways; this plan would jeopardize the extensive work already done to set up the new routes in the restructuring. For these reasons, the Transit Level of Service Analysis, required in transportation impact studies, should be examined in this EIR.

There is no attempt to mitigate Alternative 1 for Project 6-2 when low-cost, feasible design alternatives exist. There is no reason to take one of the left-turn lanes from northbound Clipper Street to westbound Portola Drive for bicyclists. Available low cost, feasible mitigations are clearly available that would provide a Class 2 bicycle lane at this same location! Further, the project may create the need to install a signal at the Diamond Heights Boulevard/Clipper Street intersection (not evaluated in the draft EIR), which would be more costly than other mitigations available.

Possible mitigations include:

1. Conversion of the southbound receiving lane to a single lane at the Portola Drive/Clipper Street/Burnett Avenue, accomplished by shifting the very small concrete median further westward/southward, adding the additional northbound left-turn lane back into the intersection, restriping southbound/eastbound Clipper Street to be one lane, and to remove one through movement on the southbound Burnett Avenue approach. In fact, removing one southbound/eastbound lane could provide enough pavement for a bicycle lane in the other direction!

2. Widening of the northbound approach to the Portola Drive/Clipper Street/Burnett Avenue intersection to allow for bicycle lanes to be added, but without eliminating the second left turn lane. There is adequate right-of-way (the parcel diagram attached is from SFGIS files showing the property line follows this comment).

3. Creation of a Class 1 bicycle path directly between Noe Valley and the Portola Drive Corridor. A Class 1 bicycle path facility would enable bicyclists to completely avoid the need for Project 6-2. Alternative routes could be a path that uses (a) the “scenic overlook” property between High Street and Portola Drive (1 blocks north of the Clipper Street intersection), or (b) the Market Street underpass at the top of 24th Street, which would tie into Portola Drive at Corbett Avenue. This would be a more desirable and attractive Class 1 bicycle facility connecting Noe Valley to the Portola Drive corridor, improving the bicyclists connectivity to the Noe Valley business district.

Comments to Other Sections

We have restricted our comments to one general and one specific project in the bicycle plan. However, as San Francisco residents, we believe that there are serious design mistakes made in this plan. There are many instances where the turning radii of buses (both Muni and tour buses) cannot be met in the narrow lanes, so that buses may sideswipe other vehicles or bicycles on the roadway. Examples include Project 6-5 where Portola Drive curves are so sharp that Muni and tour buses will be unable to stay in their lane if they are narrowed. We already witness this problem on Portola Drive and several other streets today. The designs of these projects suggest that turning radii are not an issue when they are.

There are examples where the “projects” are not fully diagrammed in the report, but are only described as cross-sections (such as Project 5-6 on Cesar Chavez Street between US Highway 101 and Valencia Street). This does not represent an adequate project description and thus should be not considered for acceptance within the EIR.

We are disappointed that the Bicycle Plan does not “seize the moment” to provide separate Class 1 bicycle facilities, enabling a safer and more desirable experience for residents and inspiring new bicyclists. Bicycle routes in other Bay Area counties and bicycle systems in European countries such as Denmark and the Netherlands are increasingly geared to separating bicycles from traffic, rather than merely aligning bicycle lanes on streets next to vehicles placed in narrow lanes. Bicycle lanes provide dangerous situations to bicyclists, including risks from people opening doors from their parked cars, or people driving into the bicycle lane from the narrowed traffic lane.

One lost opportunity is with Portola Drive. The entirety of Portola Drive (which has frontage roads and remaining open space) could be completely redesigned from property line to property line to turn this facility into a signature parkway for San Francisco. Instead, bicyclists are only given a narrow corridor while higher-speed vehicles travel by them. This does not encourage more people to become bicyclists, but merely satisfies requests of existing bicyclists to have the lane! This plan clearly is avoiding adequate consideration of improvements which could require the City to do more than restripe lanes.


To address the myriad of impacts and issues with the projects in the Bicycle Plan should not be studied and environmentally cleared at a citywide level. The plans should be implemented in coordination with Neighborhood Circulation Plans, or detailed design discussion studies for each of the project “clusters.” The appropriate design and implementation of the projects in this EIR should be as a neighborhood or cluster document, rather than a single citywide EIR for the 30 proposed projects. Finally, the public deserves to be informed of the real costs or benefits of lane reductions for every project---to not only vehicles, but to transit and to greenhouse gas emissions.


Clipper Street/Portola Drive area neighbors

Labels: , , ,