The Causeway’s Safety Overhaul: Emergency Bays, Taller Railings and the Evidence


The Lake Pontchartrain Causeway’s safety improvements addressed two different problems: where a disabled vehicle could get out of traffic, and what happened when a vehicle reached the edge of the bridge. Emergency stopping bays added places to pull over. Taller railings changed the barrier along the older southbound span. Their purposes are related, but their construction histories and measures of success should be examined separately.

The most useful account follows the work from financing to construction, then asks what the available evidence can establish about its results. That approach avoids treating an announced budget as a final cost, a project goal as a proven outcome, or the completion of one contract as the completion of every improvement.

The two spans came from different construction eras.

According to the Greater New Orleans Expressway Commission, the southbound bridge was completed in 1956 and the northbound bridge in 1969. The older structure generally uses 56-foot spans, while the newer one generally uses 84-foot spans. Both form a crossing of approximately 24 miles over the lake. Causeway bridge information.

Those differences matter to an improvement project. The two roadways may look like a matched pair from a distance, but adding new sections requires connections to different existing structures. The original dimensions are part of the explanation for the different construction details on each side.

Concern about the barriers has a long documented history.

In a January 8, 1975 recommendation letter, the National Transportation Safety Board called for an evaluation of the west span’s barrier system. It wanted a system that could keep vehicles from going through or over the barrier while redirecting them with limited harm to occupants. The letter also recommended systematic review of individual crashes. NTSB recommendations H-74-40 through H-74-42.

The letter followed an investigation that also examined vessel collisions and bridge-span collapse. Those are distinct hazards from a road vehicle crossing a barrier. The recommendation is evidence that barrier performance was an identified issue decades ago; it does not, by itself, establish the intervening history of every repair, response or engineering decision.

That distinction is especially important when comparing old records with modern photographs. A recommendation’s existence does not prove that nothing happened afterward. Establishing the full implementation history would require subsequent commission records, engineering reports and the recommendation’s disposition.

The modern work had a defined financing framework.

The Legislative Fiscal Office’s October 2019 budget analysis explains that Act 481 of 2016 authorized toll-backed bonds for higher rails and safety bays. It records a revised toll schedule effective May 1, 2017 and describes how the added revenue supported the improvement program. Fiscal year 2020 GNOEC budget analysis, pages 7–8.

The report described the southbound railing project as beginning in May 2019, with an estimated cost of $40 million. It separately discussed the safety-bay work, which began in December 2018. These figures belong to a planning and budget document written while construction was underway. They should not be presented as audited final expenditures for the entire program.

Public projects can have several valid dollar figures at the same time: a statutory authorization, a bond issue, an engineering estimate, a contract amount and final spending. A sound historical account identifies which figure it is using and what work that figure covers.

Twelve bays created additional emergency stopping areas.

The fiscal report specifies six in each direction. They provide places to remove disabled vehicles from through traffic. Fiscal report’s project description.

A bay is a localized widening. It does not create a continuous shoulder along the full crossing. That distinction matters when describing the improvement to motorists: the project increased opportunities to get out of traffic, but it did not give every point on the bridge the same recovery space.

The operator’s bridge-information page locates the bays at roughly four-mile intervals, at mile markers 2, 6, 10, 14, 18 and 22. It also distinguishes them from the crossovers that connect the spans. The two features serve different operational purposes and should not be counted as though they were interchangeable. Bays and crossovers.

The construction record explains how the bays were added.

PCI Gulf South’s project profile lists December 17, 2018 as the project start, February 14, 2019 as the beginning of precast erection, April 21, 2020 as the end of that erection work, and June 12, 2020 as project completion. It identifies twelve 1,008-foot pullover lanes and a $57.5 million project cost. Safety-bay project profile.

That $57.5 million figure is for the profiled safety-bay project. It should not be labeled the combined final cost of every Causeway safety improvement. Likewise, the June completion date should not be silently extended to a separate railing contract.

The profile describes precast components and equipment working from the water to limit interference with traffic. Because the existing northbound and southbound structures differed, the team developed different widening connections. Some northbound work required nighttime lane closures. The construction account therefore supports a description of limited disruption, rather than a blanket claim that no lane ever closed. Construction methods and challenges.

Some published dimensions need their context preserved.

The operator’s public safety page describes bays ranging from approximately 600 to 1,000 feet, while the project profile describes 1,008-foot additions. The sources do not explain that difference in measurement. It could concern different portions of a bay or different summaries, but that explanation would require confirmation. The agreed and well-supported feature is twelve added stopping areas, six on each span.

A similar source check prevents a more substantial error: one national PCI project page refers to thirty bays, while the operator, fiscal report and regional project profile identify twelve. This article follows the latter, mutually consistent records. The conflicting figure should not be copied simply because it appears on an engineering-industry website. Conflicting national project profile.

The southbound railing was raised from 25 to 46 inches.

The operator confirms the installed addition described in the fiscal report. The added steel addressed vehicles reaching the bridge edge. Operator’s description; 2019 fiscal report.

Height is an understandable public measure, but it is not a complete engineering description. Connections, materials, geometry and how a vehicle interacts with the system also matter. Louisiana’s broader retrofit research illustrates this point: a 2022 LTRC summary describes an initial design that failed a crash-test criterion and a revised design that passed. That statewide research should not be represented as proof that the Causeway’s particular installed railing is identical to the tested design. LTRC Report 658 technical summary.

The operator reports improvements, but its percentages need attribution.

The Causeway’s safety page reports a 30 percent reduction in rear-end crashes and a 50 percent reduction in rollovers, and says no vehicles went overboard in the initial results it describes. Those are the operator’s reported results. The page does not supply a complete comparison period, exposure calculation or evaluation method, and it should not be read as a promise about all subsequent years. Published operator claims.

For an independent assessment, the next useful records would be annual crash counts by type and direction, traffic volumes for the same periods, installation and opening dates, and consistent definitions. A comparison should also explain changes in travel patterns, enforcement and reporting.

Consider a hypothetical example: a bridge has fewer rear-end crashes after an improvement, but it also carries substantially less traffic. The improvement may have helped, yet the raw decrease alone cannot isolate its contribution. Conversely, changes in crash severity may matter even when the total count is relatively stable. Those are separate questions that a single percentage cannot answer.

The available construction records establish what was added and when the bay project was completed. The operator’s accounts explain the intended benefits and its reported early experience. A stronger claim about a precise, lasting reduction in risk would require the underlying evaluation.

Research note: Sources checked October 16, 2026. This article separates historical recommendations, budget estimates, construction records and operator-reported outcomes. It is a review of published material, not an independent inspection, a crash-test certification or a new statistical study.