By: Chris Matus
Working in the GIS/Mapping field and also measuring and mapping courses for Joe along with other trails, never use a GPS to measure a course. Liam mentioned only a couple of the variables that will throw off the accuracy of the GPS. Even if you set your GPS to collect a point every second, you will typically come up short compared to the actual ground distance. A GPS is made to collect accurate locations based on a point and do very bad at figuring distance. Even a survey grade GPS that has to do differential correction, either on the fly or by post-processing the data to get an accurate location (sub-centimeter) will be bad at figuring distance.
When a GPS records a track, all it is doing is recording it's location at a certain point on the earth per the set interval. You will not see each individual point because most GPS software programs automatically connects the dots and what you end up seeing is the track line. If you were able to see each individual point, you would notice that they don't follow the line you travelled precisely. One may be 10 ft to the left of the line travelled, while the next one may be 20 ft to the right of the line. And this continues throughout the whole collection process. What the software and the GPS does is interpolate the distance. This is why you typically come up short. Though there may be times you come up long, but it's not very often.
The GPS also does not do elevation very well. There are many reasons for this and too technical for this email, but the main thing to remember is that elevation on a GPS will be around 1.5 times less accurate than the X, Y (Longitude, Latitude) of the collected point. What you can use the elevation for is just a general trend.
Now, that being said, if you are out in the open and don't have any obstacles blocking the reception from overhead or from the horizon and the satellites are evenly spaced in the sky to get good triangulation and there is no cloud cover and your not next to a lake or river and there is no weird atmospheric things going on and you travel a fairly straight line and you don't have much elevation change, then the distance recorded will be pretty close to the actual ground distance.
When I make my maps, I use a GPS only for a general location of the trail. The distances I list are from a survey wheel or a calibrated bike. And even then, I use both products very carefully making sure the wheel stays in contact with the ground at all times. By doing all this there will still be some error since most trails are typically wide enough where a person can take a slightly different path along it. To compensate for this, the trail will be measured more than once and then the average used.
If you understand the limitations of the GPS, then it's a great tool to use for a general idea of the distance you have travelled or the elevation you have traversed. Just don't take the numbers it gives you as the real deal.