A Storm-Surge Model Just Cost the Ferry Terminal Rebuild Eighteen Inches and $9 Million
The engineers wanted a number for the deck height. The modelers gave them a range, and the argument that followed is the actual story.
The Port Meridian ferry terminal will be rebuilt with its passenger deck at 14.5 feet above mean higher high water, eighteen inches above the height in last year’s schematic design, after a coupled surge-and-tide model produced numbers the design team could not talk its way around.
The change adds roughly $9.1 million to a $147 million project and pushes the ramp geometry into a redesign that will delay bid advertisement to March. The state ferry board approved it 5-2 on July 21, in a meeting where the technical presentation ran longer than the public comment period, which almost never happens.
The model came from the Meridian Coastal Processes Group, the same team whose bluff retreat survey landed in February, working with Harbor University’s hydraulics group under a $780,000 county and state contract signed nineteen months ago.
## What the model actually did
Previous terminal design used a design flood elevation derived from a regional stillwater surge estimate plus a fixed wave allowance and a sea level rise increment — three numbers added together, each drawn from a different document, two of which predate 2015.
The new work runs them together. The team built a nested hydrodynamic grid resolving the channel mouth at 20 meters and the terminal basin at 5, forced it with 1,200 synthetic storms drawn from a statistically resampled record, and ran each storm against six sea level scenarios and against tide phase treated as a random variable rather than a worst case.
That last choice is the one that moved the number, and it moved it in an unexpected direction. Treating tide phase probabilistically lowered the median result — most storms do not arrive on a spring high tide — but it fattened the upper tail badly. In the 2070 mid-range sea level scenario, the median annual maximum water level at the terminal came in about four inches below the old design assumption. The 1-percent-annual-chance level came in twenty-six inches above it.
“The old method was conservative about the middle and blind about the edge,” said Dr. Ines Kallenbach, the hydraulics lead. “We spent four decades designing against a bad average.”
A second finding surprised the team. Basin geometry matters more than the open-coast surge. The rebuilt breakwater and the dredged approach channel, both completed in 2016, create a resonance at roughly 22 minutes that amplifies certain surge waveforms inside the basin by up to 40 percent relative to conditions outside it. That amplification does not appear in any regional model, because no regional model resolves the basin.
The harbor is not a smaller version of the ocean, and the terminal is not a smaller version of the harbor.
## Why the board fought about it
The two no votes were not from people who disputed the science.
Board member Adaline Pruett voted no on the grounds that a 14.5-foot deck creates a ramp slope that fails accessibility standards at the lowest tides without a second transfer span, which is where a substantial share of the $9.1 million goes. “We are raising the deck to protect against a 2070 event and making the boarding worse in 2027 for every person using a wheelchair,” she said. “Somebody should say out loud that this is a trade and not a free upgrade.”
The second no came from Hector Salvatierra, who made the more uncomfortable argument.
Salvatierra, an engineer by training, said the design elevation was being set against the 1-percent-annual-chance level under a scenario the modelers themselves label mid-range, and that neither the board nor the public was told what the confidence interval around that number looks like. He asked Kallenbach directly. She gave him the answer: the 90 percent interval on the 2070 1-percent level spans about 31 inches.
“So the band is wider than the change we are voting on,” Salvatierra said. “I am not saying ignore the model. I am saying that if the honest answer is that we cannot resolve eighteen inches, then we should be building a structure that can be raised later instead of guessing high once in concrete.”
Kallenbach’s response was that she agreed with the principle and that adaptability was outside her scope. The design team’s structural lead, Corin Aday, said a jackable substructure had been priced in 2024 at $22 million and rejected as too expensive. Nobody on the board asked to see that estimate again. Salvatierra asked afterward whether anyone would, and answered his own question.
## Where the number does not travel
The team was insistent on one limitation, and repeated it in the staff report at the county’s request.
The 22-minute basin resonance is specific to the current bathymetry. Dredging schedules change it. So does the sediment that the bluff retreat survey documented moving south along the outer shore. A maintenance dredge planned for 2029 would alter the geometry enough to require the model to be rerun, and the contract contains no funding for reruns.
“Every model has an expiration date and nobody ever writes it on the box,” Kallenbach said. She has asked the county for a standing $60,000 annual line to revalidate after bathymetric surveys. It is not in the draft budget.
The group has also declined, twice, to produce parcel-level flood depth maps for the flats, despite requests from two insurers and one neighborhood association. The grid does not resolve individual streets, and Kallenbach has said in writing that publishing depths at that resolution would imply a precision the model does not have — the same position the group took when homeowners on the bluff asked for per-property retreat dates.
## What was learned, and what it cost to learn it
The county paid $780,000 for the study. The study added $9.1 million to a project and a redesign delay.
County planning director Sanjay Oduya, asked whether that made the study a bad investment, gave a number instead of an opinion. The 1996 terminal, the one being replaced, flooded twice in thirty years. Each closure cost the ferry system between $2.1 and $3.4 million in lost service and emergency repair, and the second one stranded 900 people on the outer islands for two days.
Then he gave an opinion anyway. “We are not buying certainty,” he said. “We are buying the right to be wrong in a survivable direction. That is what eighteen inches is.”
Aday has kept the jackable substructure drawings. She said she expects someone to ask for them within a decade, and she would rather they exist.
Reporting for this story was prepared for The Harbor Ledger’s science desk. Tips:newsroom@theharborledger.com