On the evening of May 17, 2025, the Mexican Navy sail training ship ARM Cuauhtémoc backed away from a public dock at Manhattan’s Pier 17 and, roughly nine minutes later, reversed itself into the underside of the Brooklyn Bridge. All three masts snapped off in sequence as they struck the span’s superstructure. Two Mexican Navy cadets who had been working aloft in the rigging, América Yamilet Sánchez and Adal Jair Marcos, fell to the deck and died. Nineteen other people aboard the 277-person vessel were injured, according to the National Transportation Safety Board’s preliminary report on the accident, designated DCA25MM039.
The proximate cause was arithmetic as much as it was seamanship. The Cuauhtémoc’s air draft — the vertical distance from the waterline to the highest fixed point on the ship — measures 158 feet. The Brooklyn Bridge’s vertical clearance at mean high water is 127 feet. The 31-foot gap between those two numbers is the entire story of what went wrong, and understanding why a sail training vessel with that air draft ended up under a bridge it could not clear says as much about maneuvering physics and organizational communication as it does about measurement.
What “air draft” actually measures, and why it is non-negotiable
Every vessel with a fixed superstructure — a mast, a funnel, a radar arch — has an air draft that varies with the tide and the ship’s loading condition, and that number has to be checked against every bridge, power line, and other overhead obstruction on a planned route before departure. It is one of the most basic pieces of passage planning in commercial and naval navigation, taught early in ship-handling curricula precisely because the penalty for getting it wrong is not survivable to structures that were never designed to flex under a lateral blow from a moving hull.
The Cuauhtémoc’s masts are square-rigged steel and wood, built in 1982 to carry press of sail training cadets aloft on yards well above the height of a typical merchant ship’s bridge structure. That height is functional — it is what makes a tall ship a tall ship, and what teaches naval cadets the seamanship of climbing rigging and handling canvas at altitude. It also means the vessel has essentially no margin for error passing under the East River’s fixed crossings, several of which sit lower than the bridges further out in New York Harbor. The Brooklyn Bridge, opened in 1883, was engineered for the shipping of its era; its 127-foot clearance has not changed, and it was never going to accommodate a ship whose masts stand 31 feet taller.
How a ship backs itself into a fixed object
The clearance mismatch only became fatal because of how the ship was maneuvering. The Cuauhtémoc left its berth stern-first, a common method for getting a long vessel with no room to turn out of a tight Manhattan pier and pointed downriver, according to the NTSB’s account of the departure sequence. A sea pilot and a docking pilot were both on the open conning deck with the ship’s captain, and the tugboat Charles D. McAllister was assisting by pushing against the starboard bow to help control the turn.
As a general matter of seamanship, backing a single-screw, single-rudder vessel is harder to control than moving ahead: propeller-induced “walk” — the sideways force a rotating screw imparts to the stern, caused by asymmetric water pressure on the blades — behaves differently in reverse than ahead, and a rudder loses much of its steering authority when there is no forward wash flowing past it. The NTSB’s preliminary report does not attribute the Cuauhtémoc’s swing to this mechanism specifically; propulsion and control systems remain among the elements still under investigation. The NTSB’s preliminary findings describe the tug repositioning from the bow toward the stern in the middle of the maneuver, and the ship’s own sternway accelerating from about 3.3 knots to 5.1 knots in the final seconds before impact — speeds that gave the crew only a narrow window to arrest a swing that was carrying the stern, and the masts above it, toward the bridge. By the time the docking pilot called for additional tug assistance, there was not enough time or thrust available to stop the ship’s momentum before its rigging reached the span.
During the pre-departure master/pilot exchange, the captain reported propulsion and steering were functioning normally with no known deficiencies, according to the pilots’ account to investigators — a significant detail on its own: this points, on current evidence, toward a maneuvering and control problem playing out over a period of a few minutes in a tidal, current-affected stretch of river with limited room to correct, rather than a mechanical breakdown of an engine or rudder.
The physics of three masts breaking in sequence
Video of the collision shows the masts failing one after another rather than simultaneously — mizzen, then main, then foremast — a sequence that follows directly from the ship’s own geometry and orientation as it drifted aft under the bridge deck. Each mast met the fixed steel of the bridge’s underside at a point well above its deck-level support.
General ship-structure engineering suggests this would concentrate bending stress at the highest weak point in a mast’s structure — typically a fitting or a taper in the wood or steel spar — rather than at its base, though the NTSB has not yet published a structural analysis of this specific failure. A mast under sail is generally engineered to carry loads that pull along its length and bend it within a narrow range predicted by naval architects; a lateral impact against an immovable structure at several knots of vessel speed would apply a shock load well outside that design envelope, and rigging elements meant to flex or give way — such as backstays — would likely be unable to dissipate energy fast enough to prevent fracture.
That mechanism is consistent with what happened: each mast broke near its upper section, sending broken spars, yards, and rigging down onto the deck and into the water below, with cadets still on the rigging when it happened, though the NTSB’s own account of the structural failure mechanics is still to come.
A communication chain under scrutiny
Separately from the geometry and the mechanics, the NTSB’s preliminary report highlights a command structure that investigators are examining closely: orders from the docking pilot were acknowledged, translated into Spanish, relayed to a crew member positioned outside the ship’s enclosed navigation bridge, and then passed again to personnel inside before being executed.
Each link in that relay is a potential source of delay or distortion, and it sits within the broader operating-procedures review the NTSB has named as part of the ongoing investigation, alongside the vessel’s propulsion and steering systems and crew training and qualifications. Toxicology tests for the pilots and the tugboat’s captain came back negative, and inspectors from the New York City Department of Transportation found the Brooklyn Bridge itself sustained no significant structural damage — cosmetic and coating damage to a maintenance platform and paint, but nothing threatening the span’s integrity, according to that assessment as reported in the NTSB’s preliminary report and by Marine Log and gCaptain.
What happens next
The NTSB’s investigation remains open, and a final report with formal probable-cause findings and safety recommendations has not yet been issued. Parties participating in the inquiry include the U.S. Coast Guard, McAllister Towing, the Sandy Hook Pilots Association, and the pilotage organizations serving New York and New Jersey harbors. The Cuauhtémoc itself was held in New York for damage assessment and repairs before making the return crossing to Mexico, arriving home to a formal reception in Veracruz in late November 2025.
For naval architects and harbor pilots, the accident is likely to reinforce a long-standing but easily underestimated hazard: air draft is a fixed physical fact that maneuvering error can turn into a structural collision in a matter of seconds, in a way that ordinary collision-avoidance thinking — focused on other vessels — does not always anticipate for fixed overhead obstructions. The 31-foot gap between the Cuauhtémoc’s masts and the Brooklyn Bridge’s clearance was never in dispute; it was known before the ship ever left the dock. What the investigation is still working out is how a routine, heavily piloted departure maneuver, in a harbor navigated by tall ships and pilots for generations, produced the sternway and swing that closed that gap in the space of a few minutes once the ship began moving astern — with fatal consequences for two young sailors.