View Full Version : AT true mileage?
Darwin again
05-25-2007, 09:58
Here's something for the map/mileage/measurement experts. I know this is a little technical, but knowing the big brains (really:rolleyes:) that lurk here, there must be an answer just waiting to spew across our screens. So here's my thing:
We all know how vague the mileages are along the AT. The distance markers, the guidebooks, the maps, none of them really match. So, thinking about this, I measured the mileage along a section of AT in the Grayson Highlands two ways:
1) I used the linear scale along the bottom of the elevation profile; and
2) I measured the length of the line that traced the actual elevation profile, the one that goes up and down with the countours.
Result: Two different mileages. The first was seven miles, the second turned out to be 8.25 miles. Measuring the mileage on the map itself might yield yet another measurement. The section contained about 1,500 feet of climbing, then 1,000 feet of descent, then 1,000 feet of climbing.)
Given that it's shorter distance to walk in a straight, level tunnel through a mountain that it is to walk up and over that same mountain, it makes perfect sense that the more mountains you walk over, the more the actual mileage deviates from a one-dimensional measure taken from a map looking straight down on it. (Another example is the diameter of a ball versus the distance from one side to the other over the curved surface (half the circumference) -- the "over" distance is longer. Example: The distance from one side to the other, over the top of a 10-inch diameter ball is about 16 inches. You can see how walking over mountains could add mileage...)
I guess my question is, how do the major guides we use, (Wingfoot, ALDHA,) compensate for that up-and-down mileage gain? Or do they? It seems the only accurate way to measure the AT mileage would be with a wheel. I know I've heard that Warren Doyle and his crew walked a wheel up the trail once.
Given all the variables, there has to be a standard measurement. What is it and how close is it to the reality of the trail? Or have I made a fundamental thinking mistake here? Are the maps and guides made to comform to wheel surveys? If so, that would mean the mileages on the elevation profiles and maps are only useful as suggestions, more useful as agreed-upon conventions than actual figures. Has anyone ever done a three-dimensional AT mileage measurement (I mean, that we know of)? Even a GPS survey wouldn't take the extra up-down mileage into account.
The proper way to measure the length of a trail is with a wheel. I believe Warren Doyle and his gang have "wheeled" the trail several times. So yes, they do measure the actual distance on the ground, taking into account the contours.
The trail has in fact grown in length over the years. In 1990 the "official" length was 2135 (or was it 2151) miles. When the trail gets re-routed off a road, the "replacement" section is often longer than the roadwalk it replaced. A classic example of that is Pond Mountain, near Laurel Falls.
Kerosene
05-25-2007, 10:26
I'm pretty confident that the official AT club mileages are minimally based on wheel measurements. There was an inveterate backpacker (he's deceased now) who spent several years laying out GPS waypoints including elevation for a large chunk of the AT, but I doubt that that data has been incorporated into some of the maps and guidebooks yet.
Elevation profiles are notoriously poor. In most cases the trail club takes elevation measurements infrequently along the trail, and then using an altimeter which are hard to calibrate. Consequently, you get profiles that are simply straight lines between two measured points, or, I suspect, the profiler tries to graphically approximate a small rise that wasn't actually measured. I trust the guidebook mileages much more than I do the elevation profiles, and generally I discount any different mileages shown by trail signs say unless some hiker has scrawled a revised mileage!
Shoot, what year was it that they added 17 miles to include Albert Mt.?
Slosteppin
05-25-2007, 19:44
I think that a wheel is a very accurate way to measure any trail, because it actually measures the distance a person walks.
OTOH, because of their construction, wheels do have a built in error - any system of measurement does. A dealer who sells to surveyors could probably give the expected error of the wheels he might sell.
I believe that a GPS set on one second updates would get a more precise and accurate (2 different things) measurement. But this would be true ONLY as long as the unit maintained lock on at least 4 sattelites. Of course, GPS also has its own types of error sources.
Very few trails would have been measured in this way. Many could not because of tree cover.
Slosteppin
Here's something for the map/mileage/measurement experts. I know this is a little technical, but knowing the big brains (really:rolleyes:) that lurk here, there must be an answer just waiting to spew across our screens. So here's my thing:
We all know how vague the mileages are along the AT. The distance markers, the guidebooks, the maps, none of them really match. So, thinking about this, I measured the mileage along a section of AT in the Grayson Highlands two ways:
1) I used the linear scale along the bottom of the elevation profile; and
2) I measured the length of the line that traced the actual elevation profile, the one that goes up and down with the countours.Therein lies the problem. That elevation profile is not the actual elevation profile. It is highly exaggerated. Each map varies, but mostly they are about six or seven times higher than acual. A real profile would be almost a flat line with just a few tiny humps. The mao will tell you how off scale the profile is.
I guess my question is, how do the major guides we use, (Wingfoot, ALDHA,) compensate for that up-and-down mileage gain? Or do they? It seems the only accurate way to measure the AT mileage would be with a wheel. I know I've heard that Warren Doyle and his crew walked a wheel up the trail once.
Given all the variables, there has to be a standard measurement. What is it and how close is it to the reality of the trail? Or have I made a fundamental thinking mistake here? Are the maps and guides made to comform to wheel surveys? If so, that would mean the mileages on the elevation profiles and maps are only useful as suggestions, more useful as agreed-upon conventions than actual figures. Has anyone ever done a three-dimensional AT mileage measurement (I mean, that we know of)? Even a GPS survey wouldn't take the extra up-down mileage into account.
No expert, but... The profile is just a graph of distance traveled, as measured by a wheel or other device, and the elevation. It gives no other definitive information. Tracing this graph would always result in a mileage error, unless it were perfectly horizontal or vertical. As Frosty said the more vertical exaggeration, the more error (difference from the x-axis). The exaggeration is for easy reading. It also makes the ups and downs look steeper. Old maps of the AT did not have profiles.
If you were to trace the trail on a topo map, you would have the problem described, but on a twenty mile day the difference wouldn't be worth calculating.
Darwin again
05-25-2007, 22:35
No expert, but... The profile is just a graph of distance traveled, as measured by a wheel or other device, and the elevation. It gives no other definitive information. Tracing this graph would always result in a mileage error, unless it were perfectly horizontal or vertical. As Frosty said the more vertical exaggeration, the more error (difference from the x-axis). The exaggeration is for easy reading. It also makes the ups and downs look steeper. Old maps of the AT did not have profiles.
If you were to trace the trail on a topo map, you would have the problem described, but on a twenty mile day the difference wouldn't be worth calculating.
Thanks Groucho and Frosty!
That's pretty much the crux of the matter: the vertical exageration of the profile changes the linear measurment along the profile edge on the map. The greater the exageration, the longer the mileage will measure.
The elevation profiles tend also to be sloppy on some of the AT maps. I even found a small mountain left off an elevation profile of the trail in Vermont, probably due to a trail relo. I wish there was a good, updated, standardized, stylized set of maps for the entire AT -- alas, an impossible dream. Maps for the entire trail just like the excellent ones for the Whites in NH would be cool.
So it is a little further going over obstacles, but the difference isn't practically measurable -- or significant. I thought I'd uncovered a massive conspiracy, but just revealed my lack of cartographic prowess!:o
rusty075
05-26-2007, 01:17
The proper way to measure the length of a trail is with a wheel.
Not exactly. The proper way would be a survey crew with theodolites. Wheels are tools for estimating how far something is, but they only give you one aspect of it, the distance. Wheels can't tell you direction or elevation gain. Modern surveyor equipment could give you an elevation profile accurate to fractions of an inch. But....it would be an incredible amount of work, and way too expensive for an organization like the ATC to undertake. I wouldn't be surprised if there are surveyors maps of parts of the trail in the NP's though. The Fed's have more money to burn.
The ATC's GPS survey data that they have online does include altitude data, so you could get a pretty accurate number from it that way. I tried to do exactly that is afternoon, but all the freeware tools I downloaded to do the math crashed on the huge pathway GPS file. The thing has millions of lines of coordinates in it. There's a thread here on WB where we were just talking about the ATC GPS file a couple of days ago. If you load it into Google Earth and tilt the viewing angle way down it is actually surprising how level the trail looks from that perspective. It definitely feels a lot steeper than that when you're walking on it. ;)
Here's something for the map/mileage/measurement experts. I know this is a little technical, but knowing the big brains (really:rolleyes:) that lurk here, there must be an answer just waiting to spew across our screens. So here's my thing:
We all know how vague the mileages are along the AT. The distance markers, the guidebooks, the maps, none of them really match. So, thinking about this, I measured the mileage along a section of AT in the Grayson Highlands two ways:
1) I used the linear scale along the bottom of the elevation profile; and
2) I measured the length of the line that traced the actual elevation profile, the one that goes up and down with the countours.
Result: Two different mileages. The first was seven miles, the second turned out to be 8.25 miles. Measuring the mileage on the map itself might yield yet another measurement. The section contained about 1,500 feet of climbing, then 1,000 feet of descent, then 1,000 feet of climbing.)
Given that it's shorter distance to walk in a straight, level tunnel through a mountain that it is to walk up and over that same mountain, it makes perfect sense that the more mountains you walk over, the more the actual mileage deviates from a one-dimensional measure taken from a map looking straight down on it. (Another example is the diameter of a ball versus the distance from one side to the other over the curved surface (half the circumference) -- the "over" distance is longer. Example: The distance from one side to the other, over the top of a 10-inch diameter ball is about 16 inches. You can see how walking over mountains could add mileage...)
I guess my question is, how do the major guides we use, (Wingfoot, ALDHA,) compensate for that up-and-down mileage gain? Or do they? It seems the only accurate way to measure the AT mileage would be with a wheel. I know I've heard that Warren Doyle and his crew walked a wheel up the trail once.
Given all the variables, there has to be a standard measurement. What is it and how close is it to the reality of the trail? Or have I made a fundamental thinking mistake here? Are the maps and guides made to comform to wheel surveys? If so, that would mean the mileages on the elevation profiles and maps are only useful as suggestions, more useful as agreed-upon conventions than actual figures. Has anyone ever done a three-dimensional AT mileage measurement (I mean, that we know of)? Even a GPS survey wouldn't take the extra up-down mileage into account.
The most accurate measurement is produced by a wheel in the hands of a conscientious human, someone who walks in as straight a line as possible along trail and is careful about not letting rocks and fallen logs distort the distance.
I don't know how the rest of the world does it, but all of Maine has been measureed this way, most of it several times. Each of the five overseers of the trail has a wheel or access to a wheel. Each relocation is carefully measured and the results sent to ATC for inclusion in the data book and used in the description of the trail as listed on the back of each of the seven Maine maps. I strongly suspect other maintaining clubs use the same technique.
All the major guides are based on the maintaining club data as sent to ATC for inclusion in the data book. Wingfoot occasionally reports mathematical or omission errors and adjusts his guide accordingly. All the measurements are subject to human error. Measuring wheels measure feet or meters, not miles. Someone does the simple math, but errors are possible. The wheel I use counts only to 5,000 feet or so. I have to keep a paper record of each way point, add them all up and then calculate the miles.
A GPS survey is the least accurate way to measure mileage in mountainous and hilly regions. I currently use the latest -- and excellent -- Delorme software and GPS device. It's mileage figures never agree with the figures I find from using a wheel. And these days I'm not measuring the Appalachian Trail, but my town land trust trails, where the spread between the highest and lowest points is less than 200 feet.
Nor would measuring the elevation profile give accurate distances. These are based, I believe on contours from government Topo maps that typically range from 20 feet to 100 feet. ATC and maintainance clubs then deliberately distort them to emphasize changes in elevation. Also there are a lot of ups and downs in these contour intervals that are essentially erased when the profiles are drawn.
Weary
Shoot, what year was it that they added 17 miles to include Albert Mt.?
I don't know, but we should be eternally grateful.
soulrebel
05-26-2007, 08:32
the trail is far too long to be measured...:sun
warren doyle
05-26-2007, 09:30
In 1990, three members of the 1990 Appalachian Trail Circle Expedition took turns wheeling the trail. One person wheeled at least two-thirds of the trail while the other two shared the remaining third. We did this as a volunteer service to and for the trail.
The ATC, thanks to Collins Chew, gave us two measuring wheels that were the same and we carefully calibrated them pre-expedition. The extra measuring wheel was carried in the support van and used when the original one became unrepairable.
Take the combination of the same type and make wheel used and only three people measuring, along with one person's intimate knowledge of the trail and subsequent relocations from 1973-1990, I have no doubt that this measurement was about as reliable, valid, meticulous and consistently accurate as an entire trail measurement could be. When we handed in our lengthy (about 50-60 pages) final document to the ATC in September 1990, they had a benchmark to mark additional post-1990 mileage changes from.
Unfortunately, it was also during September 1990 that two southbounders were murdered in a shelter just south of Duncannon. Three former thru-hikers (and ALDHA members) took it upon themselves, on a volunteer basis, to capture the murderer, which the ATC/NPS were unable to do, still walking south on the trail near Harpers Ferry more than a week later. Although they were praised as 'citizen-heroes' by both the PA and FL law authorities, the ATC official who labeled the three 'vigilantes' (and still hasn't apologized to them) caused the three well-meaning individuals to drop their ATC memberships. Because I spoke up for and supported these three hikers against this injustice, all our hard work on measuring the trail was almost ignored and underutilized by the ATC and its maintaining organizations.
PS: As to the question about when they put the trail over Albert Mt., I walked over Albert Mt. in 1973. The only relocations in that area (Standing Indian) that have ocurred during the last 34 years is the section just south of Deep Gap, and between Deep Gap and about halfway up Standing Indian Mt. (along with a minor relocation just south of Mooney Gap).