The entrance to a cave significant enough to get a number and a survey will eventually be marked by a numbered tag attached to a spit. This will then become the primary survey station - ie. the point where an underground survey will start, and the point to which a surface survey should go. It's worth thinking about where you would put such a tag right from the start. Unlike the first rigging bolt (often used as the first point of a survey in the past) it should be sited with a particular view to its visibility and accessibility without having to put on SRT kit. If such a point has a clear view of the majority of the sky, then this is the point to use for a GPS fix too.
If however, you are dealing with a cave at the foot of a cliff, or otherwise with a restricted view of the sky, then choose instead a good landmark with a clear view, and within one (or maybe two) survey shots of the entrance. If you have found a group of caves close together, it might be better to GPS a central point rather than get quick (but less accurate) fixes on each entrance.
Once you have chosen your point, mark it in some way (could be a spit hole or a cairn, for example - we aren't supposed to use paint any more) and place the GPS on the point. If you build a cairn, make it wide rather than high - tall cairns are knocked down by the depth of snow each winter. Give it a couple of minutes to get a fairly good fix (the first figure reported may be quite a way out, but after a couple of minutes things should settle). Then mark the point as a waypoint. If you're feeling really keen, you can set it up for averaging, which gives a more accurate fix – some GPS receivers support this automatically, and with others you can just leave it recording a track log, then record another waypoint at the same place just before you leave so it's clear to someone examining the track log when you actually left. (Averaging was once crucial to getting any kind of remotely close fix, but is somewhat less important these days now that Selective Availability has been turned off.)
While the GPS is recording your location, you can do something useful (like rigging the cave, doing a surface survey from the GPS point to the marker spit, looking for other caves, or even having lunch!) Remember to stop the waypoint averaging before moving the unit or changing the display page. Take a photo of your GPS point showing at least one of your cave entrances too.
It doesn't especially matter what display options are selected when you are getting the GPS fix, but it is important to use standard ones when writing down the reported position in the survey book. The usual systems are either Lat/Long with the WGS84 datum, which all GPSses support out of the box; or the Austrian BMN (Bundesmeldnetz) system, which should be relative to the Austrian MGI datum (Hermannskogel), which is what we use for our surveys. It doesn't matter too much what combination of parameters you actually use as long as you record what they were, since we have conversion programs that can convert coordinates between the systems; but remember to record which grid and which datum your GPS was set for when you copy the fix into the survey file. (Besides the risk of introducing severe errors, it is an extraordinarily tedious task to have to repeatedly try all the plausible combinations of grid and datum that a given set of numbers might be in, convert them all to dataset coordinates and see which ones give answers in vaguely the right place, which is what someone will have to do if you don't write the details down.)
If you want to set up your GPS to use the same coordinates the survey data set uses, which makes life a lot easier, then these are the runes to use as a "User Grid" if the unit doesn't support BMN coordinates immediately (which Garmin ones don't, for example):
|Ellipsoid:||Austrian (Bessel 1841)|
a = 63377397.155m (ΔA = 739.845)
1/f = 299.1528128 (Δf x 10,000 = 0.10037428)
|Datum:||Austria MGI (Hermannskogel)|
|Projection:||Transverse Mercator (BMN zone M31)|
|Grid parameters:||Central meridian 13°20'E|
False easting 450km
No additional scaling
Grid boundaries at 11°50' and 14°50'
|EFEC coordinate conversion
equation with respect to WGS84:
|Offsets Δx = -575m, Δy = -93m, Δz = -466m|
Rotations ωx = 5.1"; ωy = 1.6", ωz = 5.2"
Note: The table above used to incorrectly give the y rotation as 5.1". But there's a slightly more accurate version in the Coordinate Systems section.
(Technical note: the BMN grid is actually the same as Universal Transverse Mercator zone 31, but setting your GPS for UTM will give rather different coordinates, as the plateau is actually in zone 33. My understanding of the situation is that the Austrians have found it more convenient to extend one grid to cover the whole country, thus deviating from UTM for the areas of the country which are just over the grid boundary. This is an important gotcha to watch out for, since while all the other coordinate systems produce answers in recognisably different formats, UTM 33 coordinates look like dataset coordinates but are offset by a couple of kilometres. Having your GPS set to the wrong datum produces even more subtle errors - the difference between BMN grid + WGS84 datum and BMN grid + Austrian datum is an offset of around 500m to the south and 50m in altitude.)
A good way of testing that your GPS is correctly set up is to set it WGS84 Lat/Long and enter a waypoint for a point whose coordinates are known – such as the 204a tag, at 47°41.456'N 013°49.288' – and then change the settings again to use the user grid. It will now convert this point into the new grid; if you check its coordinates, it should come out as something close to 486697E, 5283699N, which are the BMN coordinates for 204a. For use in the dataset we tend to subtract the 450km offset in the easting and ignore the first two digits of the northing, giving 36697E 83699N.
Write down the figure that the GPS gives for each waypoint at the time (just in case some failure loses the data from the GPS memory – this has happened a couple of times in recent years, much to the annoyance of everybody involved). That's all you need to do at the cave. Get the GPS data downloaded to a computer next time you are in Base Camp (or Top Camp if someone has a laptop :-), and as a backup (expo computers break down surprisingly often it seems) write it down by hand on one of the A5 cave info sheets with all the other details of your cave and put that in the surveys ringbinder file.
If you want to read about the nitty gritty of converting GPS coordinates to the ones used by the Kataster system, you can do no better than read the short introduction to coordinate systems, which briefly says "it's horribly complicated and we use computer programs to do it properly". (A rather outdated first attempt at this can also be found in Wookey's Compass Points Article from 1996, which briefly says "it's horribly complicated and we don't really know how to do it properly".) Overall things have significantly improved since the early days, particularly as without the fog of the SA variation it's now easy to find out whether your GPS is set up right by just GPSsing a known point and comparing the results. However, the main point of having a GPS fix on an entrance is so we can find it again and be sure it is the same one!