Comments on the Environmental Assessment and Programmatic Section 4(f) Evaluation for the Longfellow Bridge Project

Comments on the Environmental Assessment and Programmatic Section 4(f) Evaluation for the Longfellow Bridge Project

March 20, 2012

Pamela Stephenson
Division Administrator, FHWA
FHWA Massachusetts Division
55 Broadway, 10th Fl.
Cambridge, Massachusetts 02142

Thomas Broderick
Acting Chief Engineer
Mass DOT Highway Division
Ten Park Plaza
Boston, MA 02116

RE: Comments on the Environmental Assessment and Programmatic Section 4(f) Evaluation for the Rehabilitation and Restoration of the Longfellow Bridge, January 2012

Dear Ms. Stephenson and Mr. Broderick:

WalkBoston is pleased to have this opportunity to discuss the Environmental Assessment for the Longfellow Bridge. We have been involved in the planning for the reconstruction of the bridge from its initiation and throughout the lengthy public process, as a member of the Bridge Task Force, and at the recent public hearing.

To reiterate our May 2009 comments, we believe that MassDOT and FHWA should take the opportunity of this very important and expensive (approximately $300m) bridge reconstruction project to make significant improvements to pedestrian and bicycle service on the Bridge and its approaches. As an overarching comment, we believe that the time is right to think about our future urban transportation network with a greater focus on transit, pedestrian and bicycle access, and a reduced focus on private auto use.

We applaud the fact that since 2009 MassDOT has made very significant positive changes to the proposed reconstruction of the bridge, very specifically incorporating major improvements to pedestrian and bicycle access on the outbound side of the bridge, major improvements in access to the riverfront on both the Boston and Cambridge sides of the bridge, and more modest pedestrian improvements on the inbound side of the bridge.

We also applaud MassDOT’s commitment to engaging with the advocacy community and the public during the planning and environmental review of the project.

WalkBoston’s most basic comments and request for further change in the design (by moving the crash barrier on the inbound side of the bridge to provide a wider space that can accommodate both pedestrians and one-way bicyclists) are contained in the handout provided by a group of advocacy organizations at the public hearing on March 1, 2012 and attached to this letter.

Our additional, more detailed, comments on the Environmental Assessment (EA) are provided below.

General Comments
The EA includes a welcome recognition of the historic role of the bridge; the parkland context of the Esplanade and river; a clear summary of the two key elements of Section 4(f) requirements to demonstrate no feasible and prudent alternative to damage to parkland and historical resources and the requirement to undertake “all feasible planning to minimize harm…” to the parkland and historic resources, and that the project will have a net benefit on these parkland resources (pp 23, 77). The EA recognizes that restoration of key historic park features is part of the purpose and need for the project and in order to do so provides clear commitments to move the eastbound Storrow off-ramp to Mugar Way, to remove a portion of Mass Eye & Ear parking from parkland, and to replace the pedestrian and bicycle bridge to the Esplanade. The EA also makes a commitment to reducing the number of outbound lanes on the bridge and adding significant walking and bicycling space. Finally, the EA also recognizes the value of the Task Force in improving the project.

One technical issue that we request to be remedied in future MassDOT environmental documents and that has confused the analysis of pedestrian impacts is the insufficient provision of dimensions on some of the drawings. The design process has resulted in confusion about the location of the crash barriers in many drawings. We note that the crash barriers are clearly shown in the midspan options (e.g. Fig. 4-7) and not shown in the pinch point options (e.g., Fig. 11-2). This brings difficulties to the analysis as we are not certain if the crash barriers at pinch points are located within the sidewalk dimension.

We have itemized our comments below for ease of reference and response.

Outbound Comments
A. Outbound sidewalks. Sidewalks in the outbound direction are proposed in the EA to be a generous 13-foot width at midspan, narrowing to a minimum 8½-foot sidewalk at Pinch Point 1. This is achieved by limiting vehicles to one lane outbound, and including an 8-foot space for a bicycle lane. The sidewalk becomes narrower at the pinch points but it unclear whether the 8½’ sidewalk width and other estimates of sidewalk widths represent clear and unobstructed distances and do not contain crash barriers.

B. Lane widths outbound. The EA calls for one vehicular lane outbound and another parallel lane of 6’-0” reserved for cyclists and protected by a 2’-0” buffer strip. The width of the bike lane is constrained due to the unfortunate combination of a 12’-0” roadway lane width and shoulders that vary from 3’-3” to 4’-0” on the MBTA side of the vehicular lane. This configuration will result in total vehicular roadway widths of from 15’-3” to 16’0” – clearly encouraging high speeds in the traffic moving toward Cambridge. (It replicates the same mistake made on the inbound side of the bridge a number of years ago when at least one of the lanes was widened to 15’-0” thus encouraging fast traffic.) Bicyclists will be endangered by high speeds; they would be better served by a wider bike path and and a wider buffer, narrower shoulders and a vehicular lane reduced to 11’. Providing an ample and comfortable bicycle lane will also benefit pedestrians on the inbound side of the bridge because it would encourage cyclists to use this outbound facility and decrease the appeal of attempting to ride a bicycle toward Cambridge on the sidewalks of the bridge (either the upstream or downstream sidewalks).

C. Shy distances (shoulders) outbound. The report states that “design exceptions” to allow 1′ rather than 3′ shy areas in the outbound direction are not warranted, even though the use of the 1′ dimension for a shy area is proposed for the inbound direction. The extra width can and should be applied where sidewalks are narrow – particularly at the pinch points – and especially to the narrow bike lane and buffer on the outbound side.

D. Crash barriers outbound. The EA calls for crash barriers between the sidewalk and the bicycle lanes, which we support. However, the crash barrier widths are not shown at the pinch points, suggesting that the crash barriers are located directly on the sidewalk and will diminish its width. This would reduce the sidewalk width significantly, and is unacceptable.

E. Bridge intersection at Charles Circle. The intersection where the southbound ramp from Storrow Drive meets the bridge has been redesigned by Mass DOT rather dramatically – with a reduced curvature of the turn between the ramp and onto the bridge. This change allows several improvements including new parkland, removal of the existing ‘pork chop’ islands, providing a narrower street to cross and construction of a safer, more obvious path for the pedestrian and cyclist crossing. We believe that pedestrians crossing at the intersection should be protected by a signal to replace the current flashing yellow light that does little for pedestrian safety.

Inbound Comments
A. Inbound sidewalks. If a configuration similar to outbound were to be used in the inbound direction (with one lane for vehicular traffic), there could be a 9-foot sidewalk at pinch point #1, widening to 13 feet for the midspan. This is clearly a far more appropriate restoration for the most significant pedestrian element of the bridge, and is an appropriate measure to satisfy the Section 4f requirement of federal law that “all feasible planning to minimize harm” to this historic feature be taken. Unfortunately, the proposed design, which would continue to provide two slightly wider auto lanes at Pinch Point 1, results in a 5’6” sidewalk, and as shown in the preferred alternative is an inadequate “restoration,” that will not serve the many pedestrians, joggers, picture-takers, baby strollers and walking commuters adequately.

B. Crash barrier relocation. A proposal that many advocacy groups agree on is the placement of the inbound crash barriers between the vehicular travel lanes and a combined sidewalk/cycle track. This is far from an ideal solution for pedestrians at the pinch points, but offers the hope that in the future cyclists could have a safe lane of their own outside the crash barrier. Meanwhile, it is important to assure that the cycle track be clearly separated from the pedestrian area inside the crash barrier. Following construction, the sidewalk/cycle track should be evaluated for use and safety patterns. We also recommend strongly that MassDOT carry out the experiment of a single inbound lane, opening to two lanes at Charles Circle at the conclusion of the bridge reconstruction. We believe that traffic will adapt to the long construction period by shifting away from the Longfellow Bridge and that a post-construction test will reveal that a one-lane inbound configuration (with a wide bicycle lane that can also serve as an emergency lane) operates reasonably well for vehicular traffic.

C. Lane widths inbound will encourage traffic to speed. The pavement area provided for inbound vehicles in the EA include two 11-foot lanes, a 1-foot wide shoulder and a bicycle lane, which acts in part as an additional shoulder. This will result in a very wide pavement (28’-6”), which will encourage traffic to move very fast, possibly endangering cyclists. At Pinch Point 1, the travel lanes are narrowed to one 11-foot lane and one 10’-6’ lane, and a 4’-6” bicycle lane, still resulting in a wide pavement of 26’-0”.

D. Lane widths inbound for emergency vehicles to pass. The stated purpose of the two 11-foot lanes is purportedly to allow for the widths of emergency vehicles, yet it seems improbable that one emergency vehicle will be overtaking another (an ambulance passing an ambulance? A bus passing a bus?) With a shoulder, if the paved vehicular lanes were 10’-6”, emergency vehicles could readily pass one another on the rare occasions they might need to do so. The same holds for buses.

E. Lane widths inbound and prior agreements with citizens. Lane widths of 10’-6” were agreed to by former Commissioner Luisa Paiewonsky and former Chief Engineer Frank Tramontozzi during the citizen participation process. We think that commitment should be honored, especially since inbound lanes are currently much narrower than that near Charles Circle, where emergency vehicles now manage, without apparent difficulty, to pass other vehicles.

F. Shy distances (shoulders) inbound. The shy distance on the MBTA side of the inbound traffic lanes is 1’-0” but takes space from the narrow sidewalks that result at the pinch points. Perhaps the shy distance could be narrower to add space to reflect the needs of pedestrians on the sidewalk, while also slowing traffic for safety as it approaches Charles Circle for safety.

G. Crash barriers inbound are clearly shown, with widths, in the EA Preferred Alternative at the midspan, but not at the pinch points. Since the sidewalks at the pinch points may be as narrow as 5’-6” there is no room for them to include the crash barriers. To do so is to reduce the clear width at the pinch point to 4’-0” – a clearly unacceptable result.

H. Lane widths over the bridge.

  •  The EA shows that one lane on the midspan of the bridge is feasible and provides traffic operations that are nearly identical to the EA’s proposed two lanes whether it leads into two lanes or three at the entrance to Charles Circle. Since Longfellow Bridge traffic has historically carried approximately 10,000 daily vehicles less than the BU Bridge (22,000 v. 33,000 ADT), we request that MassDOT document why two lanes are needed over the full length of the Longfellow when one lane over the BU Bridge in both directions has been constructed and found to be satisfactory.
  • The solution to the queuing problem is not to provide extra queue space in a second auto lane on the Longfellow Bridge. One queue lane inbound plus the auxiliary/bicycle lane is far superior, and provides head of the line accessibility for emergency vehicles, which would not be available if the lane were occupied by queue storage. The appropriate improvement to Charles Circle operations is to shorten the cycle time, so that queues will not block the intersections. This same strategy is being used in the proposed signal timing to avoid gridlock from excessive queues on the reconstructed Anderson Bridge, and MassDOT should use the same technique here. We request that the changes in timing be tested now, before reconstruction begins.

I. Lane widths inbound at Charles Circle.

  • Estimates of traffic on the inbound side of the bridge do not warrant three lanes for vehicular traffic entering Charles Circle. We would like an analysis of what Table 4-2 on p. 47 of the EA indicates: that 1 lane widening to either 2 or 3 lanes entering Charles Circle has the same volume/capacity ratio as 2 lanes entering the circle on 2 or 3 lanes. This data seems not to support the EA estimates of queue lengths backing up on the bridge (shown in the same table).
  • Inbound traffic reaching Charles Circle. The report describes, but does not deal with, the traffic and safety problems in Charles Circle that lead to gridlock and delay causing a big problem for ambulances destined for MGH from either direction of Storrow Drive. Ambulances can be stuck in the gridlock trying to go to the main MGH entrance on Cambridge Street which is backed up for two blocks into Charles Circle. Another part of the problem is the short storage distances for turning traffic and expressway access under the MBTA station.
  • Since there is already too much traffic within the Circle, it does not seem prudent to continue three lanes into the Circle from the Longfellow, especially considering that the other entrances parallel to the bridge (from either direction of Storrow Drive) have only two lanes entering into the Circle. We request an analysis of anticipated traffic volumes from the bridge compared with volumes coming from the east or west on Storrow Drive to clearly explain why three lanes are essential from the bridge.
  • One lane leading inbound from the bridge into two lanes entering Charles Circle has not been presented in the EA to show how it might perform. Such an assessment could help reviewers understand its viability and should be provided.
  • Shown by CTPS (but not included in the EA) is that a dramatic increase in traffic would result from the EA proposal of two lanes widening to three at Charles Circle, and that it would result in traffic increases on Charles, Beacon, Arlington, and Herald Streets. There has been no analysis presented of how the city streets can tolerate the loads shown by C TPS. An analysis of these traffic impacts is essential.

J. The existing pedestrian underpass near the Boston end of the Bridge should be maintained as it could become an option for future pedestrian connections.


K. The proposed construction period interim traffic pattern may increase difficulties for pedestrians in adjacent neighborhoods.

  • The proposal for Longfellow to operate one-way towards Boston shifts Cambridge-bound traffic through the severely-congested Leverett Circle and Craigie/Prison Point intersections. (See Figure 4-3) We do not believe this should be allowed to happen, as it would drive the Cambridge-bound traffic into the East Cambridge residential neighborhood. All such moves have potential impacts on pedestrians using sidewalks along these streets.
  • A two-way operation would be more consistent with the desire to provide an evacuation route, would avoid exacerbating the badly-congested Leverett Circle and Prison Point intersections, and avoid spillover traffic into Cambridge and Boston neighborhoods. For the essential need to provide ambulance access to MGH, one lane should be provided in each direction throughout construction. (Since both inbound and outbound sides of the bridge have roadways of about 30’ in both directions, it is not clear why there is no room for two-way auto and bicycle lanes during construction, by continuing to tolerate the existing skimpy sidewalk and bike lanes for three more years during construction, perhaps with a 20-mph speed limit.)
  • Any proposal to limit Red Line closing to weekends, with bus service from Park Street to Charles Street to Kendall, ignores the lack of a lane for the Cambridgebound buses over the Longfellow, so implicitly involves routing the buses via Leverett Circle and Craigie Bridge on their way to Kendall. Bus stops for this proposal would be critical for adequate access by pedestrians.
  • There is no mitigation proposed for the 3-year constraint of Longfellow to one lane. Yet the nearby Green Line is supposed to provide 3-minute frequency (double the current schedule) by 2014. That service could be improved earlier, and connected to Kendall with a frequent shuttle. Proper mitigation of the 3-year constraint on auto use is essential to public safety and access to MGH, and would permanently reduce the auto demand on the bridge. We would like to see an analysis of how these issues are to be addressed and how MassDOT intends to measure the impacts of vehicle diversions. Since roughly two-thirds of Longfellow Bridge users come from the Red Line catchment area, it would be beneficial to try to capture drivers displaced during the construction as riders permanently using the Red and Green lines.

L. The MassDOT decision to use a design-build approach to construction affords some interesting possibilities. One is that the very difficult rebuilding of the structural support for the roadway and the Red Line approaching Charles Street Station requires interference with both lanes of Storrow Drive and the adjacent pedestrian paths in the Esplanade. The procedure should allow the contractor to suggest ways that might allow experimentation with the Eye & Ear parking lot and with the potential closures of Storrow Drive in one direction or both during construction.

M. Reconvening the Task Force is appropriate to inform an expeditious design-build process and to review interim and final access plans, particularly regarding an improved pedestrian overpass and transit and roadway mitigation measures to improve pedestrian, bicycle, and auto flow in Charles Street Circle. This can be done in parallel with the restoration of the bridge structure, to avoid delay.

N. Traffic counts of vehicles, pedestrians and bicycles should be closely monitored throughout the construction period to determine impacts on patterns of use.

O. The EA for the Longfellow project omits a meaningful description of the construction period impacts on Storrow Drive and the Esplanade, and any restoration of the historic park that the bridge occupies. Given the significant work required to provide structural integrity to the Longfellow Bridge directly over Storrow Drive, significant disruption of Storrow Drive and the Esplanade land will occur during the construction process. When Storrow Drive was built in its current configuration, it displaced a key element of the Esplanade, specifically the use of the land under the first archway adjacent to the Charles River as an essential connection between large portions of the Esplanade to either side of the Longfellow Bridge. The resulting roadway with its reverse-curved alignment (especially in the westbound direction) is unsafe. This is due to off-ramps and on-ramps in the westbound direction connecting to the Storrow Drive high-speed lane. This can all be corrected if Storrow Drive is relocated below the single arch of the bridge as proposed in the Esplanade 2020 Vision.

Thank you for the opportunity to comment on this important project. We would be pleased to answer any questions you might have.


Wendy Landman                               Robert Sloane
Executive Director                             Senior Planner

Comments are closed.