Depth Gauge Error and how it relates to the Cave Survey

[jlillest]

I've spent my fair share of time discussing the accuracy of a grade 5 survey and the instruments that we use for dry cave survey. From this, I have a pretty good understanding of how my compass readings and 10ft knots play into the error of a dive survey. But what I don't have a good understanding of is what role my depth gauge plays in the error of the system and part of this is because I don't know how my dive computers actually work.

So, if I were to use my analog depth gauge, I know that it is calibrated to sea level and sea water, and I can use conversion tables for my sump dives (obviously fresh water) at altitude. That would leave the precision of the reading at +-1ft plus the error that the conversion applies to the reading (minimal).

I have a DR computer that says it compensates for altitude, but within ranges. So is there a direct way to figure the max precision of the depth readings for this or a general dive computer? I also have a pretty hardy but basic bottom timer, can I apply the conversion factors for saltwater to freshwater to get a basic idea of what the actual depths are based on my readings?

So, how do you confirm that a computer or bottom timers readings are accurate. I know how to calculate the errors in a Suunto compass and clino (used for dry survey, precision of +-1 degree), and to calibrate them when they have that functionality, but how do you tell if a specific depth gauge is correct? (hoping for something better than just comparing two gauges side-by-side to tell that one or the other is wrong. How do you know that both aren't wrong?).

This may have been discussed in a survey book or in a workshop at one of the dive org seminars, if so can somebody point me to some good resources on depth gauge error and its effects on a cave survey?

Now, a certain percentage of folks that started reading this are thinking that these are pretty silly questions since my depth gauge readings are, despite any calibration, altitude and water density errors, is far better than my 10-foot knots and the 5-degree increments on my compass. Dive surveys are a far cry from the grade 5 dry cave surveys and even further from a professional land surveyor's standards, but it can be somewhat important when you're trying to estimate where the cave passage is going and how far you are from a certain feature, like new entrances. I'm also aware of techniques like radio location and how it is used for finding entrances, establish wireless comms and the sort. I'm just hoping to get some answers as to the precision of any particular depth gauge and, if it's never really been discussed before, make this information available to the other nerdy survey types like myself.

Thanks,
-Jon

[FW]

Maybe use a fiberglass tape, and calibrate your depth gauge in a quarry?

[mpoucher]

Its the difference in depth that counts so your error is constant regardless of altitude. You just have to establish absolute depth at the beginning.

[jason]

And if you're doing repeat surveys in the same underwater passage - find a fixed point underwater that you can use as a datum. The water levels can change and, without a datum, you'll end up with a much larger vertical error in your survey.

[dogwatch]

"......Use a pressure pot with a window so that you can accurately record foot by foot to chart each devices reading with the given pressure - for which that you can control."

This is essentially what i did when i had to report the depth of sample we had collected to FL DNR. After the dive i put my analog gauge in a pressure pot. I then knew the error at various depths (not linear) at that point. i.e from 170-200 FSW, subtract 4 ft ... from 200-220, subtract 5 ft, etc

Much the way that you have a known and variable error for a timepiece in celestial navigation or a compass mounted on your boat

[apitkin]

And if you're doing repeat surveys in the same underwater passage - find a fixed point underwater that you can use as a datum. The water levels can change and, without a datum, you'll end up with a much larger vertical error in your survey.

Jason and Mike are both right. In most cases, you start the survey at the cave entrance and the guideline forms the datum. Provided you tie subsequent surveys correctly into your first one, you should be OK. The difference in depth between the start and end of each shot forms the 'clino' reading, which of course can be converted to an angle with the shot length. Software like Compass does this for you automatically.

But in practice the horizontal errors are likely to be substantially larger than the vertical ones.

Andy

[DeepSea]

Considering I have no idea what a grade 5 survey is

Take a look at the "Accuracy" section here - http://en.wikipedia.org/wiki/Cave_survey

[icestac]

very interesting thread. I had not even though about the variation in water level between days, but it makes sense once you say it

mathematically estimate exactly

Can I use is quote? I love it!

[jlillest]

Its the difference in depth that counts so your error is constant regardless of altitude. You just have to establish absolute depth at the beginning.

So assuming that we always get a depth reading at our first point at the start of our survey, which is the final point of the last survey, then we have established a system where consistent error in depth readings is unimportant.

From dogwatch's comment though, it appears that analog depth guages can have non-linear error. So my follow-up questions are:

1. How are dive computers calibrated (typically)?
2. When does the amount of error in a depth gauge become significant in a survey? I wouldn't think that 10ft knots, +-5 degrees on compass readings would even come close, but assume a fiberglass tape (+-0.1 inches) and a compass with +-2 degrees or a BCRA Grade 3 survey.
3. Is there a layman's way to calibrate a dive computer (or at least determine it's error) that doesn't require fancy machinery? (FW mentioned calibrating in a quarry, how?)

-Jon

[FW]

...Is there a layman's way to calibrate a dive computer (or at least determine it's error) that doesn't require fancy machinery? (FW mentioned calibrating in a quarry, how?)

-Jon

I have never tried it, but you should be able to float the end of a tape at the surface, and weight the other end. Then compare the tape to the depth gauge every 10'. Then make a correction table. Don't forget that most depth gauges read in FSW (or meters), so you will have to correct for that. However Ii think Mike is right, it is the difference, more than the absolute depth, that is important.