Overshooting NDL and mandatory deco stops

[taimen]

Can anyone make an educated guess how many additional dsc cases (statistically) can be expected when changing from 40/85 to 45/95.
On other words, how large is that difference in terms of relative risk.

 

[Mike1967]

Can anyone make an educated guess how many additional dsc cases (statistically) can be expected when changing from 40/85 to 45/95.
On other words, how large is that difference in terms of relative risk.

Not sure about DCS but here's an example of the difference in deco.
Scenario is 30m dive with 35min BT on 32%
40/85 has a 3min ascent to 6m with an 8min deco stop & 45/95 is the same but with only a 4min stop.

 

[taimen]

Sure it is easy to calculate and compare profiles, but the real question should be how they effect the risk.

 

[Mike1967]

Sure it it easy to calculate and compare profiles, but the real question should be how they effect the risk.

Ok dude, you're asking a question that can't be answered so I was just trying to give you something to think about.

 

[taimen]

Ok dude, you're asking a question that can't be answered so I was just trying to give you something to think about.

Honestly I expect it can be answered, even if the two of us cannot answer it. Not definitely, but based models and studies this there must be some estimates.

US Navy tables are said to have a DCS risk of approx. 2% and afaik plain Buhlmann ("100/100") is somewhere in the same ballpark. 2% risk is terribly high for recreational activity. Real life incidences are lower for those models, probably because tables are usually not dove to the limits every time.
As far is I understand, what we are doing with gradient factors (or any conservatism setting) is trying to modify that high risk model to give us acceptable risk levels, prehaps in the range of 1/1000 to 1/10000 dives.
Now the difference in DCS risk between Shearwaters rec mode conservatism levels may be double, triple, or 10-fold (from 1/1000 to 1/10000 which are typical real life incidences). Who knows. But that difference is the only reason why those settings exist.
Many recreational tables are quite aggressive compared to GF 30/70 for example. Personally I don't consider these often quite old tables a good golden standard for DSC risk.

 

[RainPilot]

As a non-deco scientist, my understanding is that it is a non-linear risk. Thus, halving the deco is more than double the risk.

The big question is, what is that risk? IIRC the old US Navy tables were based on a 3% risk of DCS, those were a GF Hi of over 100 though.

Using that, it may be fair to say that a medium conservancy puts you at 1% risk and low on 2% . Those numbers (which I pulled wholly from the air) mean very little in real terms. If you are in a DCS susceptible group, (high BMI, PFO, poor circulation, genetic predisposition or any one of a hundred other poorly understood factors) then you may be at 50% on medium and 100% on low.

Use a medium conservancy unless you suspect you have negative factors, and then assess how you feel in the hours after diving. Gradually adjust your conservancy based on bottom time vs how you feel.

Generally a person will begin to experience sub-clinical DCS symptoms before they get a "hit", as they dive more aggressive algorithms, so pay attention to your body and when in doubt, dial back.

None of this precludes you being in that group who are a hit looking for a place to happen, so take it easy out there.

 

[taimen]

Using that, it may be fair to say that a medium conservancy puts you at 1% risk and low on 2% . Those numbers (which I pulled wholly from the air) mean very little in real terms.

Wouldn't 1%-2% mean several hits on any liveaboard where people do repetitive diving for 5 days in a row, usually diving their computers close to the NDL almost every time? In real life even the most aggressive computers are at least a digit or two below that.

 

[Mike1967]

estimates

Guestimates.

US Navy tables were based on a 3% risk of DCS

Using that, it may be fair to say that a medium conservancy puts you at 1% risk and low on 2%

Probably as good a guess as any.

 

[RainPilot]

Wouldn't 1%-2% mean several hits on any liveaboard where people do repetitive diving for 5 days in a row, usually diving their computers close to the NDL almost every time? In real life even the most aggressive computers are at least a digit or two below that.

Like I said, I pulled those numbers out of the air. My point is, whatever number you might be able to get, is going to be "x % in this sample group" and the sample group will be next to meaningless for calculating your own empirical risk number. This is like saying "2 % of adults males 18-35 die of heart attacks" and thinking that means you are safe, when you eat 15 double cheeseburgers a day and weigh 500 lbs.

I actually believe that having a "real" number is counterproductive. IMO it is better to have an understanding of the ROM (rough order of magnitude) of the risk and then dial in from there. In my case, I know that, with my particular set of negative factors, the general order of magnitude for a benign NDL dive (DCS only) is sub 1%. No idea how far below, but below. I am prepared to accept that risk, and I may modify that risk depending on what i am doing (tired, dehydrated, very important dive) by amending my conservancy factors accordingly.

 

[Storker]

Guestimates

SWAGs (Scientific Wild-A** Guesses)