NTSB Investigation: Contact of tow vessel with dock

The incident 

On November 12, 2023, about 0552 local time, the towing vessel Cindy B was pushing the loaded deck barge St. John upbound on the Columbia River at mile 53 near Clatskanie, Oregon, when the tow gradually moved to starboard out of the navigation channel and struck the Port Westward Beaver Dock. None of the three crewmembers aboard the Cindy B were injured. During the cleanup, about 2 gallons of renewable diesel fuel leaked onto the dock from a damaged pipe, with about 1 gallon going into the river; a portion of the spilled fuel was recovered. Damage to the St. John and the Beaver Dock was estimated to be about $6 million.

Analysis

While the Cindy B tow was transiting up the Columbia River pushing the loaded deck barge St. John, the tow moved gradually to starboard out of the main deepwater navigation channel and, at 0552, struck the Port Westward Beaver Dock. Deckhand 1, who was at the helm of the vessel at the time of the contact, stated that he had fallen asleep, waking only after the tow hit the dock. The vessel’s AIS track showed it began moving off course at 0544, indicating the deckhand was asleep by this time. The deckhand fell asleep during the end of his scheduled night watch, which started at 0000 and ended at 0600. Studies of shift workers have shown that workers on the night shift, similar to mariners on night watches, have a loss of alertness and increased attentional lapses when compared to dayshift workers. Shift workers are also more likely to be involved in occupational accidents, with some studies showing the risk to be almost three times that of regular day workers. Shift workers can also experience increased levels of fatigue caused by fragmented sleep cycles that negatively impact sleep quality, especially when these sleep cycles occur during times when an individual was typically awake in the preceding days.

Between November 8 and the morning of November 10, deckhand 1 had followed a normal awake/sleep cycle (awake during the day, sleeping at night) and obtained between 7 and 9 hours of uninterrupted sleep. Then, from November 10 until the casualty on November 12, deckhand 1 stood watch from 0000–0600 and 1200–1800, and he reported more-fragmented sleep patterns of just 4–5 hours each off-watch period. In addition to the general increased risk of accidents during a night watch, research of shift workers has shown that there is a greater chance of incidents during the first two nights of a night shift period.

When a person changes awake/sleep cycles in order to stand night watches or work night shifts, the person’s circadian rhythm, or biological clock, is not synchronized to their new awake/sleep cycle, a condition known as circadian misalignment. The effect is similar to jet lag and may result in excessive sleepiness during watch, at least until the body has adjusted to the change. One study found that on the first night of a night shift period, test subjects experienced cognitive impairment greater than that of a person consuming alcohol and having a blood-alcohol content (BAC) of 0.10%. On the second and third nights of the night shift period, test subjects experienced impairment greater than that of a person with a BAC of 0.05%. At the time of this report, the legal limit to operate a motor vehicle in most states is 0.08% BAC; in Utah, the legal limit is 0.05% BAC.

The risk of an accident occurring during a night watch was compounded by the fact that the contact occurred at 0552, during a period considered to be a circadian low (roughly 0200–0600), when the body is normally more fatigued and prone to diminished alertness and degraded performance. The deckhand stated that he did not feel tired before he fell asleep; however, research has shown that self-assessment of fatigue is problematic due to the noted impacts to judgment and decision-making. These impacts result in a diminished ability of the fatigued individual to detect when their performance is declining. The deckhand fell asleep due to fatigue that he did not perceive, which occurred during a night watch, at a low point in his circadian rhythm, and following a change in his awake/sleep cycle.

To mitigate the risk of an operator becoming incapacitated, including falling asleep, regulations require a pilothouse alerter system on towing vessels like the Cindy B. The towing vessel’s system was designed to activate successively louder audio and more salient visual alerts when movement was not detected in the wheelhouse for periods of 3, 6, and 10 minutes. The elapsed time from when the Cindy B tow began to turn to starboard to the time that it hit the Beaver Dock was 8 minutes. Assuming that the deckhand fell asleep before the Cindy B tow began to turn, the alerter’s lights and alarms should have activated 3 and 6 minutes after the deckhand had fallen asleep—before the tow struck the dock. However, the system did not alert during the casualty. Although postcasualty testing verified that the system operated as designed, the vessel captain determined that a VHF microphone hanging by its cord from the wheelhouse overhead could swing and trip the system’s motion detectors and reset the system timers, thus defeating the system and interrupting the activation of any indicators and alarms. Therefore, the pilothouse alerter system was rendered ineffective because a swinging VHF radio microphone in the motion sensors’ field of view defeated the system.

Probable cause

The National Transportation Safety Board determines that the probable cause of the contact of the Cindy B tow with the Port Westward Beaver Dock was the deckhand falling asleep at the helm due to fatigue that he did not perceive, which occurred during a night watch, at a low point in his circadian rhythm, and following a change in his awake/sleep cycle. Contributing to the casualty was the pilothouse alerter system not alarming to wake the incapacitated deckhand at the helm because a swinging VHF radio microphone in the motion sensors’ field of view defeated the system.

DOWNLOAD THE REPORT HERE