# Calculate Geometry function returns underestimated distances

I'm making a map of the Colorado Trail and obtained a set of waypoints taken down the entire trail. I projected the set of waypoints from WGS84 (unprojected) to NAD 83 UTM13N after adding to the data frame which is also in the latter projection). Coordinates display as northing/easting meters as expected. Everything looks fine but using Calculate Geometry to compute trail length (total and individual section lengths) returns erroneous underestimates of published lengths (the published lengths have been measured with a distance wheel). Not even close-- off by nearly 100 miles for the whole trail.

I computed surface length using a 1/3 arc sec DEM but this in no way accounts for the error. I can't figure this out but I feel it has something to do with the projection and how it relates to the initially unprojected waypoints.

Can anyone clear this up as I am out of ideas?

• How close is one waypoint from another? May 1, 2016 at 17:06
• This sounds like a How Long is the Coast of Britain problem. Them switchbacks will get you in the end. May 1, 2016 at 17:31
• Too far by at least two orders of magnitude, possibly three, to compare with a distance wheel. May 1, 2016 at 17:54
• I've been hiking in the Canadian Rockies, and a half-mile interval there could show a simple vertical climb (a la Appalachian Trail), without the 1/4 mile switchbacks to gain that vertical. May 1, 2016 at 18:04
• check out arc map's "add z information"; if the issues is related to the movement on the z axis perhaps that tool could help come up with a better measurement (pro.arcgis.com/en/pro-app/tool-reference/3d-analyst/…)
– crld
May 1, 2016 at 19:30