Analysis of the forensic data from Audrey Mestre’s fatal accident has continued without pause throughout the nine days and nights that have passed since that dreadful day.?? The lead independent investigator is Kim McCoy of Ocean Sensors?? (kmccoy@oceansensors.com) a physical oceanographer with over 20 years’ experience in marine instrumentation, who has certified many prior world record attempts?? and has completed the deepest biomedical experiments to date on divers.?? Kim McCoy participated in this world record attempt as an individual, at Audrey’s request,?? to provide an independent authentication of her record depth.
McCoy, who provided the data recorder Audrey wore during the dive, was on the dive boat when the accident occurred. He??has had full access to video and still imagery, examined the sled and other gear repeatedly after the accident.?? He has interviewed everyone involved with the dive, and preparations for the dive, both at the scene and again in the days afterward. He has spoken at length with Pascal Barnabe, the bottom trimix diver, several times since Pascal returned to France.
"My duty is that of reconciling the available data by constructing the highest-probability account of what happened", McCoy explained to me today. "I want to account for as much of the data as I can, making as few untestable assumptions as I can." Kim reminded me that Einstein once recommended that theories should be kept as simple as possible — but no simpler.
The Ocean Sensors device is the analog of a commercial airliner’s flight data recorder, the so-called?? ‘black box’ upon which investigations of aviation disasters rely. The device records depth measurements on a time line, at resolutions sufficient, for example, to allow McCoy to recognize in the graph when a sled diver has tilted her head back during ascent to look up at the surface.
When I talked to McCoy on Saturday, one week after the accident, he outlined the analytical process he had been carrying out since the investigation began, back in the Dominican Republic.
"First, I draw up outlines of every plausible scenario I can think of, setting aside, for the moment, any objections which might be made.?? I cast as broad a net as I can. For example,?? I would outline a scenario in which a whale shark parked itself under the sled as it descended.?? Anything plausible is on the table. Next, I identify the component assumptions in each scenario, the things that had to have happened in order for that scenario to have unfolded.?? Then, I characterize the logical relationships between these component assumptions."
Some?? assumptions are mutually exclusive, and cannot both be true. Others have conditional relationships : if one is true, so must??be the other, or, one can only be true if the other is, too.
This process of logically ordering the assumptions provides short cuts during the next phase,??which is an iterative process of reconciling the candidate theories with the forensic data and with general knowledge of physics, fluid dynamics, and other disciplines. New questions arise, and new information is extracted from the data or is acquired by additional investigation.?? Positive confirmation, or disconfirmation,??of any component assumption ripples through the universe of scenarios, and the table is methodically cleared.
By late Saturday, October 19, Kim McCoy had reduced his solution set to 11 high-probability scenarios,??was in the process of mining additional information from the record, and had some highly specific follow-up questions for crew members.
I asked McCoy how he would characterize the distinction, if any, between the chain of causes leading to a failed dive, on the one hand, and a subsequent failed rescue.?? I framed the question by analogy to an aviation disaster, where the forensic experts might positively determine the reasons the aircraft engines failed but do not address the circumstance that every passenger was not wearing a parachute. I suggested it could be argued that this latter circumstance be taken as the cause of the passenger fatalities, rather than the engine failure.??
"I see the two as inseparable", was his reply, and pointed out that the most reasonable risk analyses did not point to any net benefit from equipping commercial jet passengers with parachutes. The physics and mechanics of commercial jetliners define the domain of effective responses to emergencies.
A world record no-limits dive is, by definition, a thing on the edge of possibility. The outcome is a function of many variables. Some of these variables have a direct impact on the outcome, others are implicated in interactions with one other, two other or several other variables. Consider this: if an outcome is a function of just 10 simple, yes-no variables, then the number of possible combinations of those 10 variables is 2 to the 10th power, or 1,024.?? Some of those 1,024 combinations yield a good outcome, the rest yield a bad one.?? In the real world, an enterprise such as the one in which??Audrey’s life was lost is rather more complex: things like weights, pressures, volumes, velocities, acceleration, drag and many others move through ranges of values.?? Kim McCoy is applying rigour, method and the experience of an entire career to the grim task of telling us what happened that day, October 12, 2002 in the waters off Bayahibe.?? " I am not in the conjecture business", he declared, " I am in the accounting business. I expect to be able to render an account within a matter of days".
