Is my gps elevation data correct?
This topic seems to come up over and over so I figure I’ll just post it here and forward this link instead of writing 300+ word document every time. Determining the “correct” elevation gain is going to be somewhat subjective and we are limited by the consumer-grade device and its software that plots it. First topic I want to touch on is gps elevation vs. atmospheric elevation. The choice comes down to accuracy. GPS does great job determining “X & Y” position but its not so accurate in the Z, in comparison to using atmospheric pressure baseline. One can argue you can be cycling along air mass front, weaving in and out between cold and warm air masses. There are instances like this where your atmospheric baseline will be way off, but in normal comfortable weather condition an atmospheric pressure based data will be more accurate than the gps reading. Although the error difference may be miniscule for short sampling/ride, if you consider 5000+ data points for a century ride, the error can basically make your valuable data worthless. Now keep in mind I am comparing consumer version of the gps to consumer version of barometer device. There are ones used by NASA and the military that really isn’t worth discussing here.
So assuming we have a good device that plots air pressure based elevation, its now worth discussing why there can be such drastic total-elevation number between a two similar/identical devices. Chances are both data are correct, your “data recorder” is doing what its being told to do, collecting data points. The root of the problem is lack of standardization. Consider these questions:
1. How often should the device take a data in?
2. What should the minimum delta elevation be in order for a grade to be considered a hill? Novice rider and pro rider will interpret this very differently.
3. Should data point be collected based on time, distance, elevation difference, or a more complicated equation?
4. Should the elevation data be collected using same formula for when you are hiking, cycling, or riding in your car?
5. Is your elevation parameter set properly for your fitness level and your choice of locomotion?
6. How are you using your elevation data? bragging right? for your specific fitness?
For simple example, if your gps took 100 data points but your buddy’s gps took 100000 data points for the exact same loop, on a perfect day, at the same time, the guy that took 100000 data points will have a data result that shows he completed stage 18 of 2011 Tour de France! Standardization is the key if you want to compare data to your friends. Lot of good devices offer various ways to take elevation readings and you can play around with its parameters to suit your need. A cheaper device may allow you to export your elevation data in ascii format so you can really go crazy with M.S Excel.
I found Garmin to be a top notch gps system for processing elevation data. Its menu does offer reasonable flexibility with the way it collects its elevation point, but more importantly, its Garmin Connect website offers “elevation correction” option for every workout you upload. Its Garmin’s way of somewhat standardizing everybody’s elevation data. The elevation correction isn’t perfect but I think it’s the best tool available for comparing other rider’s workout. Standardizing on data-over-“x”distance would be the simplest method I can think of, because most gps offers that as an option. Keep in mind, even with the Garmin elevation correction, the data result can still be skewed by quantity of samplings and the way the sampling was taken. Like I said, its not perfect.
One last note. Please don’t bring up iPhone’s gps cyclemeter into this conversation because its basically a toy compared to real instruments like the Garmin series.