<meta-note>
Arghman and other readers of this post: You may already know some or all of what follows--hopefully it will be useful to at least some of you.
</meta-note>
DougPaul said:
You can also simply record waypoints at your points of interest.
Some GPSes, including the 60CSx, have the capability of averaging a position before recording the waypoint to increase the accuracy. (There are short-term and longer-term (hours) errors--this will only allow one to reduce the short term-errors.) So it might be worth averaging for up to a minute or so. (Keep the GPS stationary or mount it on a pole or tripod for best results.)
I'll take whatever accuracy I can get out of the system, the same way people who use trackpoints+interpolating will take whatever accuracy they can get out of that system.
When recording waypoints, one might be able to increase the accuracy by using an external antenna (or mount the GPS) on a pole or tripod to give it a better skyview. (Or at least hold the GPS in the optimal orientation (vertical for the 60CSx) clear from and above your body and the foliage.)
As noted earlier, people using tracks should pay attention to how often track points are recorded to improve their accuracy. An external antenna mounted on one's head might also help... (Or just hold the GPS in a good position and orientation for long enough to record trackpoints before and after the picture. This probably requires a tripod or two people so it doesn't move while taking the picture.)
One advantage of recording waypoints over using the track is that the user is more likely to use the GPS carefully while recording the critical data than if he simply assumes the data is being recorded automatically.
Note 1: a consumer-grade external antenna is not necessarily better than the internal antenna--it just allows one to put the antenna in a good location while operating the GPS from a poorer location.
Note 2: Surveyors generally use an external choke-ring antenna on a tripod. It might help to use one with a consumer GPS, but it would cost more (eg $5K) and be harder to carry than the consumer GPS itself.
http://www.trimble.com/infrastructure/gnss-choke-ring-antenna.aspx?dtID=overview http://en.wikipedia.org/wiki/Choke_ring_antenna
Note 3: GPSes tend to be more accurate in the dead of night (say, ~2hrs after sunset to ~2hrs before sunrise) because the ionosphere is more stable. Dual-frequency GPS will improve this significantly and may be available in the next 5-10 years. (Yes, you will have to buy a new GPS to use it.)
GPS absolute accuracy may be rated at 10m (what is it, 99% confidence <10m?), but in practical terms it's been better than that. I have had several surveys where the plants in question have been very difficult to find by eye, and I have to wander around in the general area, doubting the accuracy of the GPS point I took a year or two earlier.... until I find the spot, usually within a couple of meters of where the GPS receiver indicates.
The accuracy is generally stated as 95% confidence < 10m. (This is about 2 standard deviations.) This is equivalent to 50% confidence < ~5m.) The modern GPSes have gotten a bit better at dealing with non-ideal skyviews than the older ones.
The 60CSx manual gives accuracy specs of <10m, 95% and 3-5m, 95% with WAAS. FWIW, I'm not convinced that WAAS is actually helpful in the woods. (I have some 12-hour data runs showing no improvement in the standard deviation.) The WAAS satellites are also low in the sky and may be difficult or impossible to receive in the woods.
One might note a common theme in the above--the accuracy of the GPS depends on how one uses it. (Just because it can give you a lat and lon in a wide variety of conditions doesn't mean they are equally accurate.) A professional user might check the predicted satellite constellations to make sure the measurements aren't made during a bad satellite constellation and might also record the raw data and determine the position by post-processing when the orbital and clock errors are known and published. Another technique is to accurately survey a reference point and use it to determine satellite signal corrections when measuring nearby locations (Differential GPS (DGPS).) The current state-of-the art is sub-centimeter accuracy.
Doug
