Recommendation for developing an accurate trail system map? (Urban and TIGHT.)- Mtbr.com
Results 1 to 12 of 12
  1. #1
    Trail Cubist
    Reputation: SWriverstone's Avatar
    Join Date
    Sep 2009
    Posts
    1,255

    Recommendation for developing an accurate trail system map? (Urban and TIGHT.)

    There is a need in our town to develop an accurate map of a trail system in a local park that is very tight and spaghetti-like. The "spaghetti" factor---combined with dozens of shortcut trails, cutoffs, connectors, and oxbows where the trail doubles back on itself with just 3-5 feet between either side makes using standard GPS tracks REALLY difficult.

    For starters, it's pretty much impossible to get a single clean GPS track because there's no way to ride every trail just once. As most here know, Garmin GPS tracks just aren't accurate enough for a trail system like this---duplicate tracks and even waypoints are all over the place.

    Has anyone else ever tried mapping a system like this? How did you do it?

    I'm thinking that getting the $$$ (or $,$$$) to rent (or pay someone to do it with) a commercial-grade GPS that's accurate to less than 1 meter is the way to go (like one of the big Trimble GPS's with the separate antenna/base station).

    But I'm not a GIS person or a surveyor—maybe there are other ways to do it?

    We are fortunate to have extremely high-resolution aerial photography of the park (at 4" resolution), and that shows the gravel walking paths and parts of the main biking trails pretty clearly. (So we could create a solid basemap of the park with that.) But there are still large areas in the high-res aerials where the trails are obscured beneath tree canopy.

    Any ideas?
    Thanks,
    Scott
    29er wheels are dangerous. They may cause you to go faster which can result in serious bodily injury. —Jim311

  2. #2
    Dirt Monkey
    Reputation:
    Join Date
    Apr 2012
    Posts
    278
    I have made a map for a similar densely packed trail system. Consumer grade GPS equipment is pretty useless for the reasons you mentioned. I just went out with a sketchbook and guesstimated where the trails were located using a topo map and aerial imagery. Unless you need accuracy better than a foot for some kind of surveying purposes, the method I mentioned should work well enough if you have a few unique terrain features or landmarks to anchor the sketches on.

    If high res. aerials are available you might try to track down LIDAR elevation data for the same area and see if you can find the trails. LIDAR has the ability to map terrain through the canopy and might be able to pick up the slight variations in ground cover (or lack of) above the trails or even the different surface texture of the trails vs forest floor.

    Also, from my experience, a map showing that many densely packed trails may not be very useful to a user without good on site signage which is also shown on the map.

  3. #3
    Trail Cubist
    Reputation: SWriverstone's Avatar
    Join Date
    Sep 2009
    Posts
    1,255
    Thanks aero---that's kinda the direction I've been heading in. It's just a lot of work! LOL And you're right about a "spaghetti map" not being very good on its own...but we do plan to put up trail markers/signage, eventually. The map is needed just so we have a baseline document we can use to indicate primary trails, secondary trails, shortcuts and connectors to be closed off, establish loops of varying difficulty, etc.

    Without a map it's pretty hard (especially with a "spaghetti system") to compare notes on trails or try to explain to someone which part of the trail has that mudhole that needs armoring.

    LIDAR is a good idea. I've seen some of that and it is pretty amazing how much of the surface it picks up. I'll dig around and see what I can find!

    Scott
    29er wheels are dangerous. They may cause you to go faster which can result in serious bodily injury. —Jim311

  4. #4
    WillWorkForTrail
    Reputation: Cotharyus's Avatar
    Join Date
    Jun 2012
    Posts
    4,842
    The trail system I'm working on has had it's perks. Because it's a city project, and they came to me and asked me to do it, they have opened up all sorts of resources to me. Last Friday the city streets department GPS/GIS expert was "loaned" to me. With Timble unit in hand, we went through what's been built, notes were made, etc. - but the GPS wasn't the end of the story. All of that data was put into Arcmap (ArcGIS) and fine tuned. Turns out in areas where trails are close together, sometimes even the good GPS hardware can have overlaps. But a good operator can massage the data and come up with good stuff. I've now got maps in hand to quality check.

    Short version: Use whatever GPS you want. Then put the data into something that lets you fine tune it. That's the only way you get a good map with GPS. But I can't help you much beyond that, because up til now, all my maps of all my trails have been rubbish because I'm simply no good at making them.

    Side note: The city has 10' and 2' LIDAR data in their ArcGIS system. It proves what I've been saying the whole time about the topo map we were given for proposal/development - which is that it's wrong. LIDAR is freaking awesome.

  5. #5
    Trail Cubist
    Reputation: SWriverstone's Avatar
    Join Date
    Sep 2009
    Posts
    1,255
    I think I could develop a pretty good map myself using my Garmin GPS ** if ** I can figure out a way to ride *most* of the trails in one pass---so I end up with a clean track. Even if the track is off or overlaps (as you pointed out Cotharyus), as long as it's ONE track I can tweak it using aerials beneath it.

    I've tried doing this with multiple tracks---or tracks that included multiple passes over certain trail sections---and you end up with a freaking mess. Then it becomes incredibly difficult to know what is just GPS deviation or what is an actual nearby (but different) trail.

    I just need to take the time to go out and ride the system specifically trying to figure out how to ride them all. I might even consider putting up temporary flagging or chalk arrows or some way to help me remember which way to turn at the 80 or so intersections, LOL.

    Scott
    29er wheels are dangerous. They may cause you to go faster which can result in serious bodily injury. —Jim311

  6. #6
    mtbr member
    Reputation: Joules's Avatar
    Join Date
    Oct 2005
    Posts
    3,293
    How much mileage total are you talking about? Seems to me like using consumer-grade gps, instead of trying to get one line and tweaking it (which from what you describe wouldn't be any better than just freehand drawing lines on a map), I'd rather have many passes over the same trail to "average" - if you had 5 tracks of the same trail you could with some confidence draw a line down the middle.
    I'd just go back and forth across one trail till I had 4-5 passes, then start a new log and hit another trail.

    If you were trying to map 100 miles of trail, I don't think that would be feasible. But shorter trails might be especially if you could enlist some help.

  7. #7
    mtbr member
    Reputation: thickfog's Avatar
    Join Date
    Oct 2010
    Posts
    1,271

    Re: Recommendation for developing an accurate trail system map? (Urban and TIGHT.)

    MI
    Quote Originally Posted by SWriverstone View Post
    There is a need in our town to develop an accurate map of a trail system in a local park that is very tight and spaghetti-like. The "spaghetti" factor---combined with dozens of shortcut trails, cutoffs, connectors, and oxbows where the trail doubles back on itself with just 3-5 feet between either side makes using standard GPS tracks REALLY difficult.

    For starters, it's pretty much impossible to get a single clean GPS track because there's no way to ride every trail just once. As most here know, Garmin GPS tracks just aren't accurate enough for a trail system like this---duplicate tracks and even waypoints are all over the place.

    Has anyone else ever tried mapping a system like this? How did you do it?

    I'm thinking that getting the $$$ (or $,$$$) to rent (or pay someone to do it with) a commercial-grade GPS that's accurate to less than 1 meter is the way to go (like one of the big Trimble GPS's with the separate antenna/base station).

    But I'm not a GIS person or a surveyor—maybe there are other ways to do it?

    We are fortunate to have extremely high-resolution aerial photography of the park (at 4" resolution), and that shows the gravel walking paths and parts of the main biking trails pretty clearly. (So we could create a solid basemap of the park with that.) But there are still large areas in the high-res aerials where the trails are obscured beneath tree canopy.

    Any ideas?
    Thanks,
    Scott
    Contact c0nsumer here on this board. He does this locally for our trails in Michigan and is a very helpful and knowledgeable person.

    He makes awesome maps.

    Some of his work:

    https://nuxx.net/blog/2014/01/05/new...ea-north-unit/

    https://nuxx.net/blog/2012/08/19/add...openstreetmap/

    https://nuxx.net/blog/2013/03/06/the...a-trail-guide/

    Other issues with mapping:

    https://nuxx.net/blog/2013/08/02/goo...outing-issues/

  8. #8
    Dirt Monkey
    Reputation:
    Join Date
    Apr 2012
    Posts
    278
    Quote Originally Posted by Joules View Post
    I'd rather have many passes over the same trail to "average" - if you had 5 tracks of the same trail you could with some confidence draw a line down the middle.
    I'd just go back and forth across one trail till I had 4-5 passes, then start a new log and hit another trail.
    There may be software which can average multiple tracks. Many of the professional grade GPS units use an averaging scheme, GPS signal quality control, and other techniques in the field to improve accuracy. The best thing about averaging is that your tracks will become better over time (given you always ride the same route).


    For what the maps will be used for, I bet free handing the trails onto a map will be less work while conveying the needed information. As already said, once you get the GPS data you will need to clean up the paths and export them in a usable format.

    Have you thought about reducing the number of trails that will be shown on the map? On the map I previously mentioned, I left out many of the unnecessary "shortcut" type trails that were no use to people trying to follow the main loops. We then went out and started closing down the trails that weren't shown on the map.

  9. #9
    I should be out riding
    Reputation: ACree's Avatar
    Join Date
    Sep 2004
    Posts
    1,842
    Using multiple gps files and the 'make map' function in topofusion is a good starting point.

  10. #10
    mtbr member
    Reputation:
    Join Date
    Jul 2006
    Posts
    2,673
    Over the past 10 years I've mapped pretty much every trail in my area for my trail site. The one I show below was a particularly difficult one for reasons similar to what you described. There are a couple of trail systems in my area that are like this. Unfortunately there's no simple way. I use GPS tracks to start, but then I need to manually adjust them quite a bit based on my memory of the trails and multiple trips back to verify questionable areas (Now I also will take some photos and video of the trickiest areas where multiple trails meet or come close together). A good topographic map can really help tweak things if the terrain varies enough, as can Google Earth and anything else I can get my hands on.
    "Accurate" is a relative term though. I wouldn't dare claim my maps are 100% accurate (in fact I have a recently revised version of this map I' haven't gotten around to uploading yet), but they are plenty accurate for the purpose they are intended.
    At some point the spiderweb of trails gets so complicated that even if the map is accurate, it's very difficult to use when you're on the ground anyway if you don't have marked intersections (this park doesn't).
    Recommendation for developing an accurate trail system map? (Urban and TIGHT.)-map_wi_kenosha_petrifying_springs.png
    Warning: may contain sarcasm and/or crap made up in an attempt to feel important.

  11. #11
    Trail Cubist
    Reputation: SWriverstone's Avatar
    Join Date
    Sep 2009
    Posts
    1,255
    Interesting trailville---and yes, that trail system is very similar to the one we have. I agree "accurate" is a relative term, and also agree that a map would be useless without signage.

    @ACree: I have TopoFusion, and if the "make map" feature you're talking about is the same as the "Create Trail Network" feature, I've tried using it and haven't found it to be useful for trails this tight---it really goofs up the tracks and moves some of them wildly out of place. There are two sliders for controlling how it does this, and I admit I don't understand either of them and haven't played with them a lot...so I might go back and give it another shot.
    29er wheels are dangerous. They may cause you to go faster which can result in serious bodily injury. —Jim311

  12. #12
    I should be out riding
    Reputation: ACree's Avatar
    Join Date
    Sep 2004
    Posts
    1,842
    Yes, that's the feature. I have used it to combine gps files from multiple rides on various routes in a trail system and create a very basic map. I thought it was supposed to average out the track where something had been ridden multiple times. I'm no expert on the software though. Might be worth reaching out to the developer.

    Quote Originally Posted by SWriverstone View Post
    Interesting trailville---and yes, that trail system is very similar to the one we have. I agree "accurate" is a relative term, and also agree that a map would be useless without signage.

    @ACree: I have TopoFusion, and if the "make map" feature you're talking about is the same as the "Create Trail Network" feature, I've tried using it and haven't found it to be useful for trails this tight---it really goofs up the tracks and moves some of them wildly out of place. There are two sliders for controlling how it does this, and I admit I don't understand either of them and haven't played with them a lot...so I might go back and give it another shot.

Similar Threads

  1. The McCoy Flats Trail System (60 miles of trail in eastern Utah) is looking for input
    By Jasonoutside313 in forum Trail Building and Advocacy
    Replies: 5
    Last Post: 05-16-2014, 09:49 PM
  2. Recommendation for street/urban/touring
    By GrooveRite in forum Wheels and Tires
    Replies: 0
    Last Post: 06-17-2013, 07:47 AM
  3. Replies: 12
    Last Post: 01-10-2012, 10:56 AM
  4. Tire recommendation for urban bike
    By MI-29er in forum 29er Bikes
    Replies: 0
    Last Post: 12-15-2011, 04:58 PM
  5. New light system recommendation
    By Radius in forum Lights and Night Riding
    Replies: 24
    Last Post: 09-16-2011, 02:40 AM

Members who have read this thread: 0

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

THE SITE

ABOUT MTBR

VISIT US AT

© Copyright 2020 VerticalScope Inc. All rights reserved.