re-establishing rail on the bridge

I. Executive Summary

As part of the Bay Bridge Rail Feasibility Study, this working paper reports on the available structural capacity and needed structural improvements for restoring rail on the San Francisco - Oakland Bay Bridge. This study is restricted to the seismically retrofitted West Spans, Yerba Buena Island and the proposed East Spans replacement currently being designed.

Over the 60+ year life of the structure, significant changes have been made with the 1959 removal of rail, the proposed replacement of the East Spans, and recently awarded seismic retrofit contracts for the West Spans. The potential next phase in the Bay Bridge's history, the addition of rail, would overshadow previous engineering and construction work except the original construction completed in 1936.

Four rail options were defined in Working Paper 2A.2 for this study. The vehicle types and weights ranged from BART, the lightest (least structural impacts based on vehicle weight) at 1,170 pounds per linear foot (plf) of track, to the new 'Acela' train sets developed for the Northeast Corridor 'High Speed' service weighing 1,800 plf.

Rail would not be travelling at high speeds over the bridge. Due to vertical and horizontal geometric restrictions, the operating speeds are estimated for this study to be roughly 50 miles per hour.

A. West Spans

The following ar the main findings form assessment of the West Spans:

Standard lane widths for highways are 12 feet. The West Spans are currently slightly substandard at 11'-7" and there is no shoulder. Rail widths, including emergency walkways are typically 16 to 17 feet for single track and 32 to 34 feet for double tracks. Placing rail within the existing deck widths of the West Spans would eliminate 3 lanes of traffic if rail is on one level and 4 lanes of traffic if one track of rail is on eack level. Such reductions in the number of lanes were deemed unacceptable.
Adding deck area to the West Spans to accommodate rail would be accomplished by widening the existing decks outboard of the suspension cables or hanging a new deck below the existing truss. Both positions face geometric challenges. The widening would impact air rights recently sold for a building to be constructed within 10 feet of the San Francisco Anchorage. Hanging the deck below would reduce the shipping clearance and would need Coast Guard approval.
Placing vehicular traffic on the widening was not considered. The physical barrier of the stiffening truss and suspension cables segrregates vehicle traffic in a given direction which is not the best operation scenario.
Rail would be placed on the added deck area.
The added weight from either the side-by side or below deck schemes would require improvements to the foundations, strengthening of the existing truss and towers, and possible the replacement of the current concrete roadway decks with a much lighter system.
The main cable system would need to be strengthened by adding two new main suspension cables would need to be strengthened by adding two new main suspension cables above or beside the existing cables or by adding new tower legs at the existing pier locations and suspending cables from these. The added tower legs approach would be very similar to adding a new bridge.
The added weights and structural modifications to the structure would alter the seismic behavior. Seismic retrofit beyond the construction contract currently underway would be necessary and would include foundation strengthening as well as pier, truss, and tower strengthening.

B. East Spans

Review of the design criteria and the bridge deck cross-sections has yielded the following information on the East Spans:

The current design criteria for the East Spans is to provide the structural capacity to support 5 lanes of traffic or 4 lanes of traffic and one rail track at 1,400 plf on each deck.
The bridge width is sufficient for two decks with 5 lanes of traffic and one rail track on each.
Rail is located on the interiors of the two parallel decks. This is an alignment that does not provide the greatest flexiibility in crossing Yerba Buena Island and connecting to the West Span alignments. Rail should be moved to an exterior alignment with revisions to the cable geometry to accommodate the rail envelope.
Introducing rail on the East Spans and not losing lane capacity could be accomplished by redirecting the current design team such that the current design can be altered prior to construction.

C. Yerba Buena Island

Reconstruction of the Yerba Buena Tunnel would disrupt traffic to such an extent that it was not considered in this study.
The headroom on the lower level of the tunnel is insufficient for all systems except BART.
The load carrying capacity of the upper level is insufficient for all rail systems.
Construction of new tunnel bores through the island for rail is the preferred alternative for crossing the island.

D. Costs

Very preliminary costs place the structural work alone in the $3.3 billion range. Again, this is just the costs to improve the structure without actually providing the track, track fasteners, power systems and signaling. This price covers the length from Harrison Street and out Rincon Hill near the San Francisco Anchorage, up onto the west spans, through Yerba Buena Island, across the East Spans to the Toll Plazas in the Oakland.

E. Future Study

This study concludes with approximate alignments of rail routes and projected ridership for the various rail systems. One very direct way of deciding to pursue rail on the Bay Bridge would be to compare the estimated construction costs with other means of bringing rail across the Bay, particularly tunneling or another bridge crossing.

Findings in this report should be viewed as extremely preliminary. In-depth analysis of the seismically retrofitted structures must follow this feasibility study. The analysis should use three-dimensional non-linear models to capture the full behavior and strength of the western spans under service loads and seismic excitation. Costs should be re-evaluated once details of the widening, strengthening and seismic retrofit have been developed.