The Sinking of Cargo Ship El Faro

In September 2015, El Faro, a 791-foot cargo ship, set out on a run from Jacksonville, FL to San Juan, Puerto Rico. The 1200-mile trip would take five days each way and cargo included several hundred shipping containers and cars. For the captain and his crew, this was a routine trip. Two days later, El Faro was 15,000 feet below the surface, resting at the bottom of the sea with all 33 crew lost.

In the case of El Faro, there were detailed records available from the ship’s voice recorder, as well as some of the ship-to-shore communications. Using these data, the NTSB identified many causes of the sinking, but we will zero in on where we feel the decision-making was flawed and adjustments not made, even in the face of mounting evidence which led to a catastrophic result.

The obvious contributing cause to the El Faro loss was the weather. As the ship departed Jacksonville, Tropical Storm Joaquin was moving north as the ship was heading south. For an experienced captain and crew, tropical storms and hurricanes were to be respected, but with proper advance notice, could be maneuvered around or through. (When a hurricane is expected to hit one of its bases, the U.S. Navy sends all its ships out to sea to ride it out, rather than letting them get banged around in the docks.)

Before serving two terms as president of the U.S. in the 1950s, Dwight Eisenhower was a five-star general in the army and was the commander of all U.S. forces in Europe. He was responsible for the Allied Forces attacking from the sea and invading the beaches of Normandy, France on what is known as “D-Day.” D-Day was the battle in World War II that turned the war in favor of the U.S. and the Allies. 

Not much has changed since D-Day: accurate weather information is mission-critical to any sea-going operation. Weather information was available to the El Faro from several sources. The captain preferred the Bon Voyage System (BVS) weather reports, as they were sent to his email. So, El Faro was heading south at about 25 miles per hour and Joaquin, which was now Hurricane Joaquin, was heading north at 15 miles per hour. Think of your reduced reaction time in a head-on auto collision. Every second counts.

The problem was that BVS reports, although they contained useful graphics that the captain preferred, were typically six hours behind reports from other sources, the least of which being the Weather Channel. Using the speeds above, in six hours, the ship and the storm would be 240 miles closer to each other. In 12 hours, well, you get the picture on the damage that can be done relying on stale weather data, not considering other sources of weather information (and there were several available), or more importantly slowing down or changing course. 

Changing course on a five-day voyage is no big deal when you have enough warning, which El Faro did. Joaquin was already identified as a building storm. Marine experts agree that it would not have been difficult to change course and avoid the storm. Trouble was, such a change would have delayed the ship’s arrival, increased the costs of fuel and crew, and been something of a black mark against the captain.

Nevertheless, the faulty reliance on the BVS data continued to guide the decision making of an otherwise experienced and successful captain. Some of the crew were accessing Internet sites, tracking the storm, and relaying data to the captain that were more timely and unfortunately, more accurate than BVS. Things were much worse than the captain had realized because his frame of reference was distorted by his reliance on the stale BVS data and he did not factor in credible information from other sources that contradicted BVS. 

All of this led to a situation of escalating problems for El Faro. Delayed weather forecasts and the captain’s refusal to observe, collect data, process the situation, and change course all but sealed the fate of the ship and crew. In another set of decisions that did not properly incorporate the downside, course changes suggested by the crew were rejected by the captain.

Moreover, the captain’s initial call to his company’s contact on shore, the Designated Person Ashore (DPA), did not reflect the level of distress facing El Faro. It appeared that the Captain continued to be focused on the upside of an on-time delivery and was in steadfast denial of the situation.

The captain’s initial call to the DPA did not occur until 7:00 a.m., about 45 minutes before the ship went down. The DPA did not pick up, so the captain left a voice mail:

Captain Lawrence? Captain Davidson. Thursday morning, 0700. We have a navigational incident. I’ll keep it short. A scuttle popped open on two-deck and we were having some free communication of water go down the three-hold. Have a pretty good list. I want to just touch—contact you verbally here. Everybody’s safe, but I want to talk to you.

Layman’s translation:  A window popped open, and water was flowing freely into the ship. So much so that the ship was “listing” or leaning over to one side.

No big deal, right? A captain doing the right thing and checking in with the home base to let them know he had a few things going on. But and this is a big but, the ship had lost power 45 minutes earlier and was in the middle of 130+ miles per hour winds. Yes, this was a stressful situation, but this was an experienced captain and crew supported by a team on land. Somebody should have been out ahead of this with some structured thinking of what to do in each of the different scenarios. For example, with water freely entering the ship, the ship listing or beginning to roll over, no power, 40-foot seas, and hurricane-force winds, it was time to declare an all-out emergency and forget about the upside of making a cargo delivery on time.

Before the DPA could return the call, the captain called the after-hours service at 7:01 a.m. The tone was a bit different and perhaps in that one-minute interval since he left the voice mail, the enormity of the situation hit the captain:

“Oh, man! The clock is ticking! Can I please speak to a Q.I.?” 

(a QI is a Qualified Individual, the same as a DPA)

The first radio broadcast of a distress message was sent about the same time. Less than 20 minutes later, the captain called for a general alarm, the crew gathered on the starboard side of El Faro, and abandoned ship. The last recording was at 7:39 a.m.

What went on here and could this disaster have been prevented with less focus on hitting the objective? I think the answer is a resounding “yes.” Conflicting weather reports should have been reconciled, rather than fighting the collection of more data and sticking with BVS, the captain should have embraced additional weather data and listened to his crew’s updates from the Weather Channel. 

So, it is a mystery to many who have studied the El Faro sinking, why the captain waited until the last minute to acknowledge he had a serious problem. By this point in the story, you have realized that the entire cascade of errors was driven by a focus on the upside and not taking action to mitigate issues, despite plenty of data indicating a change was needed. This is a sad example of a situation where the decision maker did not even realize the consequences until it cost him and 32 others their lives.