Risks in Cave Diving

[chimie007]

the risky end a rough guess on the order of a 50% chance of surviving a dive to half gas (whether or not an accident/silt out or whatever may or may not occur).

I would argue that this a faulty assumption in general. Probably a fair assumption in no flow cave but definitively incorrect in any cave with flow (even with a little flow). I'm afraid folks will look at the plot and thing that this is scientifically correct or anywhere close to the truth.

That being said, this is an interesting look at things. At least for the sake of discussion.

[deepdiverbob]

There was an accident analysis presentation at the 2008 workshop, during the Friday evening social. That is where I got my information. I will make the suggestion to the 2009 workshop chairman.

I should have specified the Missouri workshop, sorry about that. I have already suggested it to Ed Pavey as well.

[skip]

I should have specified the Missouri workshop, sorry about that. I have already suggested it to Ed Pavey as well.

the NACD website (http://www.safecavediving.com/) has a link to the updated accident analysis. accident analysis for the new millenium by Jeffrey Bozanic. It's based on a detailed review of actual deaths, since exley's blueprint was published.

-skip

[Squirrel Girl]

In the last couple of years, I have ended most of my dives with half to 2/3 of my gas remaining. If you don't believe me, ask some of my buddies.

Quit diving with buddies like me who have crappy gas consumption rates.

[skip]

I would argue that this a faulty assumption in general. Probably a fair assumption in no flow cave but definitively incorrect in any cave with flow (even with a little flow). I'm afraid folks will look at the plot and thing that this is scientifically correct or anywhere close to the truth.

That being said, this is an interesting look at things. At least for the sake of discussion.

I don't see that the number of dives is relevant. I would think that each dive is independent of other dives, so making 100 or 500 dives does not alter the risk of any one dive. These are independent probabilities, not conditional, and not cumulative? Whether I live or died on the last cave dive has nothing to do with whether or not I will live or die on the next one....well, except if I died I wouldn't be doing another one!

Now, basing things on various rules of gas usage....at one extreme if you use more than 50% going in means you die on the way out (less than 50% left to exit), other things being equal (like same flow going in as coming out). And using less than 50% going in, you live. I don't even see that your chance of living or dying is different from these two extremes. If you use 60% going in, and have 40% coming out...or 40% going in, and 60% coming out....as long as all other things are equal you will live or die. As long as you use less or same of the total available gas going in than is needed to get out you will live.

I don't see where the number of dives previously done could possibly make a difference.

The bell curve applies to populations of individuals and independent observations, not to repeated observations from one individual. Perhaps Gary is applying the logic to how many accidents are predicted to occur in the population of cave divers given various gas consumption rules. However, whatever logic one uses based on bell curves, does not allow predictions of one individual across repeated situations.

I've looked at this and looked at this thread and just don't see Gary's logic. Maybe it's because I teach statistics and probability and we all know that those who can do and those who can't....

-skip

[Gary]

The bell curve applies to populations of individuals and independent observations,

Actually as a statistician it would be great to get another, better estimate from you. I haven't done any statistics since high school over 20 years ago (if a high school Prob/Stat class even counts for that).

To be fair though, that is where the 1/10000 number comes from and how it is being used. That is part of the reason I chose normal distribution as a basis. Extrapolating that back to apportion risk to individual dives and divers is a bit of a stretch but some method must be chosen and it seems to fit well in this instance. (OK, there are probably other better ways but I'm not a statistician ).


Perhaps one day we will have the data available under various conditions to make a much better analysis. It doesn't seem to me like it would take a tremendous population of divers diving to halves under set flow conditions in various systems to start producing some excellent data points for better analysis.

[stickydeco]

If you aren't reaching the place you want to go on 1/3rds either add more gas or improve your SAC rate. If you are conducting multiple long distance exploration dives and using safety bottles then maybe you have another option.

It used to mean that breaching 1/3rds meant no-one would dive with you. Now people are using controversial gas planning methods used by a solo diving explorer to justify cheating thirds in their everyday diving because "I'm experienced, the rules don't apply to me" and no-one seems to be concerned. What is happening in cave diving today?

Right On Baby Bob...And that's why when the freekin tanks leave my place those pups be @ a minimum 4500 PSI....whether they are 3AA, 3AL or O2. Yee-Haw

Lithium Jack

[FW]

There was an accident analysis presentation at the 2008 workshop, during the Friday evening social. That is where I got my information. I will make the suggestion to the 2009 workshop chairman.

I have just been told that there will be a discussion on Accident Analysis at the 2009 NSS-CDS Workshop by Dr Vann, and possibly Jeff Bozanic.

[deepdiverbob]

I have just been told that there will be a discussion on Accident Analysis at the 2009 NSS-CDS Workshop by Dr Vann, and possibly Jeff Bozanic.

Is this for the Florida workshop? If so, anyway it could be duplicated at the Missouri workshop?

[FW]

Is this for the Florida workshop? If so, anyway it could be duplicated at the Missouri workshop?

I asked the workshop chairman (Gene Melton) to contact Ed, and offer to do it there as well. If you bug Ed too, it would help