CDT Map Project (free)

[postholer.com]

UTM is in base 10. Lat/Long is an arcane measuring system.

The decimal degree lat/lon graticules on my maps are base 10 as well, in .01 increments. Finding your location you would ideally split those up by '10', Leaving you with a coordinate like, 12.123, or narrowing you location down to an square area roughly 270x270 feet., which is decent accuarcy without a GPS.

While I surely understand the ease of working with base 10 numbers, it's a very poor way to communicate. The vast majoirty of people don't understand IS measurments, they undersatnd feet and miles. Just the way it is. To effectively communcate with the vast majority of hikers, I'll use miles and feet whether I like it or not. Using IS is just bravdo.

This website, esp the second paragraph, may be directed at some people

I read it. I totally agree!!!

But I'll leave it to you to convince Jerry to add lat/lon to his maps, as it is missing.

-postholer

[Mags]

The vast majoirty of people don't understand IS measurments, they undersatnd feet and miles. Just the way it is. T

-postholer

My European-born wife disagrees with you. Now, if you said the majority of Americans, that would be another story.

[postholer.com]

My European-born wife disagrees with you. Now, if you said the majority of Americans, that would be another story.

Yeah, yeah. The context of US hiking trails was implied, ie, CDT forum. Too bad SI has never gotten a foot hold here. Oh, well.

[litespeed]

So these maps vary about 22' from bottom to top and you consider this to be insignificant? If the map shown above is typical with about nine miles of trail covered, you will need about 344 maps or so to depict the entire 3100 miles of the trail. 344 X 22' yields 7568 feet of corrections, so the only way it's going to work out is if you create new scales really often, like every 30 miles or so. Even then, the only place on the map that will agree with the scale is the center of the center map. That would require at least 100 new scales for the trail. (Something to do in the tent, I guess)

Or you could use UTM. Only one scale is needed, but you probably would wear it out so you might want to carry a couple of spares.

[Another Kevin]

Hmm, I thought he was displaying the map in some sensible projection (maybe even UTM), but that the overlaid graticule was lat/lon. The scale can be more or less consistent from sheet to sheet, whatever grid is overlaid. Did I err?

[bearcreek]

I use NAD83, epsg:4269, as it's a meter more accurate than WGS84, epsg:4326
-postholer

The assertion that NAD83 is a meter more accurate than WGS84 is absolutely correct but is largely irrelevant. In order to achieve a better result than WGS84, NAD83 conversions must employ something called a NADCON transformation which utilizes large databases containing location specific information. While your mapping software may utilize a NADCON transformation, a traditional recreational GPS does not have the horsepower to do this. Instead it uses a formula called a Molodensky conversion for almost all coordinate systems, including both WGS84 and NAD83. This method can be subject to errors of up to 50 meters depending upon where you are in the US, Mexico, or Canada. Because of this, the one meter difference commonly found between the two systems is not particularly relevant while using a recreational GPS on a trail. I recently consulted on a large survey project in Florida that was having difficulties related to this very subject. It turned out that the QC expert on the job was using NAD83 with Molodensky conversions and the survey crew was using NAD83 with NADCON transformations. They were getting about 30 foot differences across the board. Needless to say, the surveyor won that argument.

This one of the reasons why some form of differential corrections to GPS surveys are vital to assure accurate results. All the data collected during the USFS mapping project of the trail was post-processed and differentially corrected using data from stationary government Continuously Operating Reference Stations (CORS) located close to the trail, or by carrier phase real time transmissions which theoretically can produce sub meter results.

[rocketsocks]

Interesting thread, learned lots, thanks. Still with all the single dangle do doe mathewsky conversions, seems the hiker is well with-in line of site of targeted area to maintain heading, lat long or ping pong, cool map guides.

Carry on men, school this fool. Meanin' me.

[postholer.com]

The assertion that NAD83 is a meter more accurate than WGS84 is absolutely correct but is largely irrelevant.

I agree, it is irrelevant. Like a difference of 22 feet between lines of longitude on map top/bottoms. Decimal degree lat/lon is effectively an equal distance grid on the map scales we're talking about.

All the data collected during the USFS mapping project of the trail was post-processed and differentially corrected using data from stationary government Continuously Operating Reference Stations (CORS) located close to the trail, or by carrier phase real time transmissions which theoretically can produce sub meter results.

I am ill-equipped to have an in depth GPS conversation as I am pretty much clueless. However, a quick peek shows that only 38 CORS are within 100kms of the CDT, which makes for a large, spreadout net. Not to mention the stations blocked by terrain. Being that I'm clueless, it's hard to say how much a 3,000 mile trail benefited from GNSS corrections on a net like that.

Since we're changing the subject....

What I do know after working with the trace extensively is, the trace is *very* good. It is by no means in the realm of sub-meter accuracy. Not even close. Sure, in open areas it is, but my cheapo GPS will do that. Further, other than survey boundaries, calculating mileages or elevation gain/loss totals, sub meter accuracy for the purpose of hiking maps is *useless*. But sure, the closer the better.

If you were to paint that trail trace you have on your printed map onto the ground, it would be over 90 feet wide. Further, in a perfect world if you could pull-off a perfect sub-meter trace, that hiker with the consumer GPS would never be able to duplicate it. They would only get close to this imaginary perfect trace.

While perfection is a wonderful objective, for the purpose of hiking maps claiming sub-meter this or that is just chest-pounding.

-postholer

[bearcreek]

I agree, it is irrelevant. Like a difference of 22 feet between lines of longitude on map top/bottoms. Decimal degree lat/lon is effectively an equal distance grid on the map scales we're talking about.

I am ill-equipped to have an in depth GPS conversation as I am pretty much clueless. However, a quick peek shows that only 38 CORS are within 100kms of the CDT, which makes for a large, spreadout net. Not to mention the stations blocked by terrain. Being that I'm clueless, it's hard to say how much a 3,000 mile trail benefited from GNSS corrections on a net like that.

Since we're changing the subject....

What I do know after working with the trace extensively is, the trace is *very* good. It is by no means in the realm of sub-meter accuracy. Not even close. Sure, in open areas it is, but my cheapo GPS will do that. Further, other than survey boundaries, calculating mileages or elevation gain/loss totals, sub meter accuracy for the purpose of hiking maps is *useless*. But sure, the closer the better.

If you were to paint that trail trace you have on your printed map onto the ground, it would be over 90 feet wide. Further, in a perfect world if you could pull-off a perfect sub-meter trace, that hiker with the consumer GPS would never be able to duplicate it. They would only get close to this imaginary perfect trace.

While perfection is a wonderful objective, for the purpose of hiking maps claiming sub-meter this or that is just chest-pounding.

-postholer

You know not of what you speak.

First of all, I don't know where you came up with the 100km figure. CORS stations need to be within 200 miles of a work area to produce satisfactory results and sometimes can be as far away as 300 miles. The reason they won't work at longer distances is because they must be tracking 4-5 of the same satellites at the exact same time as the GPS. If you are too far away you will not get enough common satellites to get a solution.

There are numerous CORS stations within that distance of the trail.
There are 11 stations in NM that are close to the trail - 100 miles or less. They are:
NMDE
PO26
NMSU
SC01
AZCN
PIE1 (this one is in the building behind the PIE-O-NEER restaraunt in Pietown)
P107
NMGR
P123
CT114

There are three CORS stations within 70 miles of Crazy Cook where the trail begins.

In Colorado, all within 100 miles:
CT14
AZCN (actually a NM station in Aztec)
MC10
MC05
R301
MC08
PO31
EC01
COFC

Wyoming is a bit more difficult, but has adequate coverage. The trail is always within 100 miles of one of these:

PO32
BLW2
MAWY

In Montana, again always within 100 miles:
P706
P707
P045
P046
MTM
PLS6
MTFV
P025

If you want to actually check this out, go here. http://www.ngs.noaa.gov/CORS_Map/ It is a map of all the CORS stations in the US, and shows you how far away from them you are.

The comment about "blocked terrain" is perplexing. As long as the satellites at the CORS and at the GPS track 4-5 of the same satellites the CORS corrections will work well enough for mapping a trail. (within 5 meters) The reason they call it "Post Processing" is because the corrections are not made in the field, but rather in the office later. The data is stored on servers and correlated via software later. It could be months or years later. This is not in real time. There is no radio contact between CORS stations and the units on the ground. It does not matter what lies between them. All that matters is that they be close enough to the GPS to receive data from 4-5 of the same satellites.

There are times when the local terrain and tree cover makes the survey difficult and in those case we relied on unprocessed SBAS data, which we always collected. Instead of using CORS, we would turn on the Russian GLONASS system to double the number of satellites being tracked, and corrected it in real time via SBAS satellites 133, 135, and 138 which are geostationary over the equator. These are the same satellites they use to land jets with without human help. This method works great in the trees - better than CORS. In this case, the corrections are made in the field in real time.

As far as accuracy, you are absolutely correct that in difficult environments the best one can expect from a moving GPS is 5 meters or so. To achieve that you need a sub meter rated GPS. GPS receivers are rated in the most ideal locations possible. However, that cheapo GPS you are bragging about WILL NOT produce submeter results under any circumstances. You get better results with better gear. That is why land surveyors don't survey with Garmins. Our primary unit was a Ashtech Promark 10. Google it.

Finally, that trace you are using for your maps was deliberately degraded and thinned to the point of inaccuracy in order to make a demonstration line that would work quickly for fly over type presentations in Google Earth. I helped in that effort. In fact, I would say that your trail line is really very sketchy because of that. Individual track points are so far apart that they skip switchbacks, corners, etc. It is not really suitable for navigation at all.

Our mapping database has over 1,000,000 track points, and that is the one I use to make maps. It shows a great deal of detail compared to the Google Earth file, which was never intended for making maps.

[postholer.com]

There are numerous CORS stations within that distance of the trail.

Like I said twice, I'm clueless regarding the finer workings of GPS. However, I did say there are 38 stations within 100kms of the trail, your list shows 33. I agree with you and still you chastise? Interesting.

Finally, that trace you are using for your maps was deliberately degraded and thinned to the point of inaccuracy in order to make a line that would work quickly for fly over demonstrations in Google Earth. I helped in that effort. In fact, I would say that your trail line is really lousy because of that. Individual track points are so far apart that they skip switchbacks, corners, etc. It is not really very good for navigation at all. Our database has around 1,000,000 points, and that is the one I use to make my maps. It shows a great deal of detail compared to your Google Earth file.

Sorry Jerry, no such luck. The original, available data hasn't been downgraded.

The trace I use for my google maps was intentionally downgraded BY ME to a minimum of 30 feet between points. I reduced the point count down to 288,769 from my source of which has 1,979,541 points. Having 2 million points on a google web map would only kill it. Further, the CDTC data was corrected with data I have access to reflecting the most recent changes to the trail. This is the data I used to make my printed maps, determine mileages and make elevation profiles.

Find out for yourself. Download the CDTC data and you have an idea of what I have.

-postholer

[bearcreek]

Enough has been said. Best of luck with your project.

[bamboo bob]

Spinning-head-spinning...........................

[postholer.com]

I updated the map image in the initial post in this thread.

The scale has been increased to 1:36K from 1:39K, the major/minor contours have increased from 400/80 to 200/40, the actual contours lines have been reduced in thickness as well.

Let's see, I've also added one additonal grey shade to the urban colors bringing it to 4 shades. It's a real advantage being able to change everything about the maps. Compare them with 40 year old scanned topo base map with 100/20 contours that look like blurred smudges. Not to mention unreadable elevation labels.

-postholer