The collision in the early hours of Tuesday morning resulted in a large part of the structure collapsing into the water with six construction workers presumed dead.

The data recorder has been recovered from the Singapore-flagged ship and will be analysed, US National Transportation Safety Board (NTSB) chair Jennifer Homendy said.

What do we know about the moments before the crash?

The operators of the Dali cargo ship issued a mayday call that the vessel had lost power moments before the crash, but the ship still headed toward the structure at "a very, very rapid speed," Maryland Governor Wes Moore said.

Just before the crash, Dali suffered "momentary loss of propulsion", the management company Synergy Marine Pte Ltd, was quoted as saying, according to the Maritime and Port Authority of Singapore.

Read more:
Rescue ongoing after ship strike causes bridge to collapse
Listen to the full despatch call from the incident

"As a result, it was unable to maintain the desired heading and collided with the Francis Scott Key bridge," the statement added.

The 985-foot-long (300-metre-long) vessel struck one of the 1.6-mile (2.6 km) bridge's supports, causing it to break and fall into the water within seconds.

Previous issues

An inspection of the Dali last June at a port in Chile identified a problem with the ship's "propulsion and auxiliary machinery," according to Equasis, a shipping information system. The deficiency involved gauges and thermometers, but the website's online records did not elaborate.

The most recent inspection listed for the Dali was conducted by the US Coast Guard in New York in September. The "standard examination" did not identify any deficiencies, according to the Equasis data.

What have experts said?

City University professor John Carlton told Sky News that while fuel may indeed have been the cause of the power loss, it is usually tested beforehand to ensure compatibility with the engine.

"It is difficult to speculate on the true cause of this incident until the investigators have completed their work," the director of maritime studies said.

"Certainly there may have been a problem with the fuel but usual practise is to send a sample of the fuel to a laboratory test house to determine its chemical properties before burning it in the engine so as to prevent major damage to or untoward behaviour of the engine.

"However, there may have been several other possible causes for this accident and this is where the present investigation will throw light on the root causes of the events that took place."

According to David McFarlane, director of Maritime Risk and Safety Consultants Ltd, the "most likely" cause of the crash was a failure in the ship's machinery.

He told Sky News on Tuesday: "The first thing that springs to my mind is: was there a sudden fault with the ship's engines or the steering gear?"

Mr McFarlane said a human error was less likely because of the number of people who would have been on duty.

"There should be no room for one-person errors because one of the other people should jump in and say 'hang on'," he said.

"The most likely cause of this is a failure in machinery or steering gear, but we just won't know until the authorities have been on board. And even then, they're unlikely to say what's been going on for some considerable time."