1

Elevation data seems off (Read 621 times)

EdwardWang


    Hi there, I love runningahead.com! I have not pay much attention to elevation data before, but it seems that the data is off a lot ( gain and loss) when I tried to compare it to my GPS data and other site's (like mapmyrun.com). Here are some examples:

     

    This is a route around my house, it's a 1.7 miles loop, and it showed that a 470 feet gain and loss.

     

    http://www.runningahead.com/logs/3580278c4a65401688280e01ef29d9f0/courses/f7f4ff0316ec436cb76b867fdc9bb342

     

    both my GPS and mapmyrun indicated about 150 feet gain and loss, which seems more accurate if you look at the elevation profile.

     

    http://www.mapmyrun.com/routes/view/67286018

     

    The reason I'm starting to look at the elevation data now is that I have a San Dieguito Half and La Jolla Half Marathon on deck. Again it seems the data from here are much higher than the GPS data collected by other runners and posted on mapmyrun.com. the links are below:

     

    San Dieguito Half Marathon: 1000+ feet vs 450+

    http://www.runningahead.com/logs/3580278c4a65401688280e01ef29d9f0/courses/8c7a9d84add5433d8aa4082f434110da

     

    http://www.mapmyrun.com/routes/view/66907784

     

    La Jolla Half Marathon:  1000+ feet vs 600+

     

    http://www.runningahead.com/logs/3580278c4a65401688280e01ef29d9f0/courses/47dd51ea481c4deca7f0b0c4c2136fa8

     

    http://www.mapmyrun.com/routes/view/66909298

     

    Again I appreciate the great work you guys doing. Have a wonderful day!

     

    Edward


    A Dance with Monkeys

      Reimport your GPS data without the elevation correction checkbox checked.

       

      RunningAHEAD uses USGS reference elevation data to "correct" the data coming from a GPS unit, since the later can be very noisy. However, USGS data is often incorrect when the surveys are out of date. For the 1.7 mile loop you listed, USGS maps show big undulations that appear to have been flattened when the development in which you are running was built (compare USGS topo and satellite images to see this).

       

      MMR is probably not correcting the elevation data at time of import. Usually, the correction provides better data. Occasionally, such as in the 1.7 mile loop example, it does not.

      EdwardWang


        Hi Trent, thanks for replying. The route of that 1.7 miles loop was not imported from GPS unit, it's created manually by click Courses, New Course. btw, it seemed that I posted a wrong url for that course in my previous post, here's the right one.

         

        http://www.runningahead.com/logs/3580278c4a65401688280e01ef29d9f0/courses/f7f4ff0316ec436cb76b867fdc9bb342

         

        in fact, I just created that route again (with a new name), it still shows a elevation gain/loss of about 406 ft. while it should be about 150 ft.

         

        http://www.runningahead.com/logs/3580278c4a65401688280e01ef29d9f0/courses/aabffc2a34324077a3eb798426da9469

          Comparing the maps, I say the problem is MMR is smoothing out the elevation.  I switched over to the USGS topographic map and "walked" the route by placing the cursor on the elevation profile and moving it from the start to finish.  Every rise and dip is correlated to the topographic lines on the map.  If you look at MMR's map, it is completely smoothed out.  All the little ups and downs add up.

           

          Also, how do you know it should 150 ft and not 406?

            Comparing the maps, I say the problem is MMR is smoothing out the elevation.  I switched over to the USGS topographic map and "walked" the route by placing the cursor on the elevation profile and moving it from the start to finish.  Every rise and dip is correlated to the topographic lines on the map.  If you look at MMR's map, it is completely smoothed out.  All the little ups and downs add up.

             

            Also, how do you know it should 150 ft and not 406?

             

            Becuase GPS doesn't lie Big grin

             

            The pain that hurts the worse is the imagined pain. One of the most difficult arts of racing is learning to ignore the imagined pain and just live with the present pain (which is always bearable.) - Jeff

             

            2014 Goals:

             

            Stay healthy

            Enjoy life

             

            EdwardWang


              Comparing the maps, I say the problem is MMR is smoothing out the elevation.  I switched over to the USGS topographic map and "walked" the route by placing the cursor on the elevation profile and moving it from the start to finish.  Every rise and dip is correlated to the topographic lines on the map.  If you look at MMR's map, it is completely smoothed out.  All the little ups and downs add up.

               

              Also, how do you know it should 150 ft and not 406?

               

              Sorry for replying late, got distracted by my non-running life :-)

               

              I'm not an expert on geographic or topographic, but I think USGS topographic map is just wrong on that loop. I have been lived there for two years now and ran/walked the loop many many times. From my usual starting point (Capricorn @ Westview), I ran clockwise, it's all the downhill until the lowest point of Black Mountain @ Westview (not after that in the Westview as USGS map showed),  then Westview is almost all the uphill except a little rolling hill after Compass Point N and around Compass Point S, but not very significantly.  The uphill showed on USGS map on Capricorn toward Black Mountain is not existed at all (it's downhill so obvious that everyone can tell). So the elevation gain should be just the maximum elevation minus the minimum point and maybe add a few for the rolling hill but not much, that's why I think 150 ft is reasonable number and 400+ is way off. I'm not sure how USGS map got so wrong on that one 1). maybe it's really old? this community was built around 1996 and I notice that USGS can have map way old than that. 2). Capricorn Way (also Westview) is a elevated road which is substantially higher than the resident area the road surrounded, so a few feet deviation can have big difference on elevation.

               

              Beside my GPS watch, I tried several Android applications and sites, RunningAhead is still my favorite one and the only one have the completed log of my running. So I really think you guys did a wonderful job. Just on this elevation aspect, I think runningahead (or maybe just USGS :-) is the one has the different result with everyone else.

               

               

              http://runkeeper.com/user/edwardwang/activity/69006416

               

              RunKeeper said I climbed 875 ft over 5 loops (7.7 mi) and the elevation graphic confirmed my naked eyes (or feet :-) observation.

               

               

              http://www.trainingpeaks.com/sw/SADNWT2K2LCBKTZLLMPMLB6SFU

               

              my GPS reported a 646 ft gain on 8.17 mi of distance which is a little conservative to my like

               

               

               

              This morning, just out of curious I used HandyRun to log and post it directly to RunningAhdead, it reported a whoop 3035.8 ft gain over 5.16 mi (3 loops), that's like climb a huge mountain (don't know any in SoCal) , no wonder my Achilles tendons got toasted :-)

               

              http://www.runningahead.com/logs/3580278c4a65401688280e01ef29d9f0/workouts/09ef4f781ab34ba490dc1ae6a7440afa/map

               

              Thanks and good day!

               

              Edward