Skip to main content

How Do You Find Elevations?

The person who owns two clocks never knows what time it is. The owner of one clock is certain. So it is with elevations above sea level.

Before computers I used to go to the library and get USGS maps. These usually provided the elevation data I needed. After getting computerized I bought a copy of Topo USA (Version 2.0). This generally gave good agreement with the USGS maps. Then along came the Flying Pig Marathon.

The start was at Fountain Square, in downtown Cincinnati, Ohio. The finish was in a park adjacent to the Ohio River. Both were shown on both Topo USA and USGS maps. I had fairly good agreement for the finish, at 485 feet above sea level. However, the start yielded surprising results. The USGS map showed the 550 foot contour line right on the start line near the south edge of the square. However, Topo USA showed:

Southeast corner: 686 feet
Southwest corner: 672 feet
Northeast corner: 673 feet
Northwest corner: 637 feet

In addition, a USGS benchmark shown at 543 feet was 575 feet using Topo USA.

I’d normally expect better agreement than this. I wrote to Delorme, who makes Topo USA, but did not receive a responsive reply.

Food for thought. Anybody else done any comparisons?
Original Post

Replies sorted oldest to newest

I'm lazy, and the easiest thing to do is pop in the TopoUSA CD and look it up that way. I don't lose much sleep over any possible innacuracy - I often question the usefulness of the start/finish/highest/lowest elevation figures anyway - I'm sure the average runners never sees them, although for the few rare times a record is set they are necessary. A course profile tells potential participants a lot more about the course than raw figures, unless it's to indicate that the course is so far above sea-level that getting proper levels of oxygen may be a factor.
I measured a 6 mile course a few weeks ago, one where you run past the finish line about 1/4 mile after the start. Topo USA showed the finish to be about 10 feet lower than the start, although it was obvious that the finish was higher than the start. I used the Topo USA data for the start, then estimated the finish elevation from that.
I have an old Garmin GPS 3 that shows altitude, but it sometimes changes +/- 75 feet when it's stationary. I'd rather depend on Topo USA and make adjustments when necessary.
The Marine Corps Marathon web site had a course profile on it, so I zoomed in to where I could read the elevations and printed it out to study it. It was painfully inaccurate-- it showed the start to be around 50 ft (USGS map gives 30; the profile gives a high elevation on Lee Hwy at around 180 ft (240+ ft from USGS); shows East Potomac Park somewhere between 30 and 40 feet (it's about 10); and so on. I never did find out how the profile was created-- I had the feeling that the problem may have come from a scaling error arising when they exaggerated the vertical scale for profile purposes. But after reading about Pete's experience I am wondering if the problem came from one of the software packages that give elevations (??)
City or County topo is becoming widely available in the U.S. Some Government websites offer it in a variety of formats, including AutoCAD. A proper road profile is time-consuming to generate, but with a little software and time, it is easily done.
The accuracy of ALL topographic information is assumed to be within one contour interval, meaning if the USGS says it's 1320 with a contour interval of 20', it means it's in the rage of 1310 to 1330. Since USGS started mapping, several datums have been used, and they vary from state-to-state.
I believe the way Topo USA generates the surface is an over-simplified model using only the point data on USGS, the contours are an image. I have encountered much larger errors than these in my experience.

Add Reply

Link copied to your clipboard.