The trouble with TOPO!

Recently I've been using TOPO! software to analyze old trips and plan new ones. I've noticed problems using TOPO! to plot route profiles.

In my experience, TOPO! underestimates travel distances. The error varies from route to route, but I find that multiplying the TOPO! route distance by about 1.15 gives an acceptable approximation of the distance. Tom Chester wrote a web page about this (thanks Google!), which you can find linked below. Chester found it necessary to correct for TOPO! mileages by anywhere from 0% to +33%, depending on the route:

tchester.org/sgm/analysis/trails/chantry/topo.html

Tom Chester didn't analyze TOPO! elevation measurements, but he observed that "they clearly are so inaccurate at times as to be largely useless for producing good estimates of the elevation gain and loss for a given route."

I'm reluctant to give up on TOPO! quite so fast. I've had it produce good elevation profiles for some routes, but poor for others. The key factor, I've concluded, is whether you are using TOPO! to plot a climb or a traverse.

Below is a four-mile route plotted on Miners Ridge in the Glacier Peak Wilderness. I attempted to trace the 6000 foot contour from Image Lake to Suiattle Pass.



The route profile, which looks like a saw blade, shows elevation gain (and loss) of over 650 feet in four miles. (Ideally, both should be zero.) This route was drawn at 100% magnification. Hoping for a more accurate result, I tried drawing the route at 300% magnification (which is too tedious for regular use, in my opinion) and the elevation gain was reduced to about 400 feet - still not very good.

I think the error occurs because this route traverses a slope. The steeper the slope, the worse the error. Slight inaccuracies in either the line you draw or the hidden mapping of elevation points to the topographic image can add up quickly. For one of my trips, measured by hand and by using TOPO!, the cumulative elevation error in the TOPO! profile was +33%.

TOPO! does really well with some routes. If you plot a route up a Cascade volcano, the elevation profile will usually be very accurate, because there is little traversing. As a test, I tried plotting a route along the Hoh River in the Olympic Mountains. This 5-1/2 mile route goes from the 600 foot contour upstream to the 800 foot contour. The elevation profile is nearly perfect because the river valley is flat. A valley with steep sidewalls (or a route along a sharp ridge) would likely be much less accurate:



To work around these problems, I've tried using correction factors for TOPO! elevation measurements. If the profile can be classified as a "climb" (very little contour following), then no correction is necessary. If the profile can be classified as a "traverse" (lots of contour following), multiplying the TOPO! measurement by 0.85 yields a better result. But it's not perfect. Some "traverses" actually don't follow contours much and TOPO! measures them quite well. Others follow contours a lot and using TOPO! is problematic.

I've found that the best use of TOPO! profiles is get a "picture" of the route. This can quickly help you decide which direction to do the route - very useful for skiers. I've also become wary of guidebooks and trip reports that offer route statistics from TOPO!. I've seen some statistics that seemed way out of line, and this analysis helps explain why.

Thanks for the info Lowell -- I've been considering a TOPO! purchase recently and it's good to know the limitations...

I had an interesting conversation recently with someone at REI about some of the limitations of Topo!  I wanted something that would give an indication of slope angle.  She indicated that the Topo! sales reps were interested in user feedback.  I played around with putting Topo! measurements into an Excel spreadsheet and got unsure results.  Mountain terrain isn't very geometrically clean.

Perhaps another item for the wish list would be to be able to measure distance from point to point on a contour line.

The good thing that came out of the conversation was that she told me I could update my topo! program from the Natl Geographic Topo! site.   I believe the latest version is 3.4.3.  for the state series that I use.

Updates: maps.nationalgeographic.com/topo/upgrades.cfm

My original version of Topo! Could only measure distance in miles or kilos.  The newer version permits to measure feet and meters (probably interpolated).

I am a little disappointed that the basic data and maps are about twenty years old.  Not that the land is going to move around that much but that technology has certainly improved and we could probably get denser data points and improved landmark naming.

At the Topo! site I found a goody left behind by Carl Skoog - some trail marking on Rainier: maps.nationalgeographic.com/topo/file.cfm?fileid=390

Check out the map exchange at: maps.nationalgeographic.com/topo/search.cfm

Gave this a little more thought... Lowell mentioned that plotting a traverse is a little more troublesome than an ascent.  This pretty much follows my brief experience.  It can be more difficult to traverse because of the erosion terrain with ridges and gullies that form across a slope.

So here is my theory.  Behind all the mapping are data points, translated into visual representation.  I think the contour lines on the Topo! map are too smoothed out.  They don't quite show the ruggedness of the traverse.  However, perhaps the Topo! profile function has less smoothing and shows the gullies that the travel path cuts across as big verticle dips.

This google earth pic of the Image Lake area shows that the erosion is greater lower down on the slope.  Perhaps if Lowell plotted his line higher up on the slope, the Topo! profile would be smoother.

Thanks for the slope angle tip!  I never realized that I would have to open a second profile window to get that.  So, why don't they put it in the original profile window and why use grade? - I ain't building no road.  They should make that a preference like selecting miles or kilometers.

Anyway thanks for the tip!

So, let's see.... max avy hazzard at 78% grade....

Another thing to consider, Lowell, is that your elevation profile there seems to have a vertical exaggeration factor of about 13.  So if you reduced that a tad, it wouldn't look so saw-like.  

Good point about the DEMs, Larry.  And even if they used the highest-resolution DEMs out there, which have 10 m cells, it's inevitable that you're going to introduce a LOT of error on a traverse, whether it's drawn by hand, or else logged with recreational-grade GPS.   

I don't use my TOPO! very much, but I do like it.  I also work in the mapping industry.  Which reminds me of something I've been sort of wondering.... any other TAY-ers work in cartography or GIS stuff? 

Ethan