Andrew has shown up an interesting flaw in both Therion and Survex. A single fix of a point, unless otherwise qualified, will always be presumed to be more accurate than multiple fixes than multiple fixes even though (I suppose, I don't have the maths) the latter should be better. Unless one is using a decent dGPS system to fix the points, it is also potentially the case that the relationship between two fixed points at entrances will be better described by the intervening underground survey data than by the GPS fixes made using consumer hand-held devices.
I will soon have to be dealing with a similar problem with a multi-entrance cave as the various lines of data join up - like Andrew's this is a dataset obtained on expo and I'll not get any more data until next year so 'soon' is a bit wrong. I think we will solve this by getting as good a fix as possible for one entrance and then using our survey data to delimit the other entrances. I'll also, however collect GPS records of the same spots and highlight the differences. The answer may be to insist that fixed point data must include an estimate of precision and that non-entry for that field would default to something worse than an exact fix. I have, in the past done this the other way around, though. I have a dataset for one relatively linear multi-entrance cave, but the data is very old and was not collected using an inclinometer, passage gradients are generally of the order of 2-3 degrees. So I used the fixed points for the entrances to skew the profile of the cave. The resulting model was a reasonable facsimile of reality. Graham _______________________________________________ Therion mailing list Therion@speleo.sk https://mailman.speleo.sk/listinfo/therion
