On Tue, 27 Feb 2024 18:57:31 +0100, Bernd Rose wrote:

> Sources can either be new data
> freely donated by individuals/companies/... who created it, themselves.
> Or it can be data, that either never carried a restrictive copyright or
> ran out of it, already:
> https://wiki.openstreetmap.org/wiki/Potential_datasources

Don't get me wrong. I'm not deprecating open source maps at all. In fact, I
absolutely love the concept of open maps. That open source concept works
well for roads, which are easy to add to the open source maps but the OSM
implementation is atrocious for topographic maps (unfortunately).

I think you don't notice how terrible the OSM topographic maps are because
you have nothing better to compare them to outside of the United States.

In the United States, we already have "open" isocline maps by the USGS.
Once you use the USGS maps for real-world off-trail hiking, you'll cry if
you have to resort to the low-resolution low-detail inaccurate OSM maps.

To help you see the delta, I zoomed into the area I previously provided:
https://i.postimg.cc/nV0Ttps6/osm-vs-usgs.jpg

And then I drew a line of the same size on both maps from two known points.
https://i.postimg.cc/bYtQgB42/test1-usgs-vs-osm.jpg

To help you, I changed HSV values to make the isoclines stand out better.
https://i.postimg.cc/4yP4wZGg/test1-usgs-vs-osm-colorized.jpg

Can you please take a look at those comparisons and let me know if you're
also seeing the huge discrepancies that I'm seeing in isocline lines?

>> Hopefully not from Google as Google's publicly available topo maps are
>> horrid in terms of lack of elevation detail.
>
> Google topography is commercial and therefore cannot be used as a source
> for OSM. (Which needs to be royalty free.)

Understood. My point was that the Google topography available in their
normal web page is terrible, although the Google Earth topographic maps are
maybe better (they look like USGS maps wrapped around a globe but it has
been a very long time since I added the USGS maps to Google Earth).

> With "topographic information" you mainly seem to refer to the natural
> part of the topography. And in this mainly contour lines and hydrographic
> soil data.

You are correct that I am referring to what I need to know when I'm hiking.
Boots on the ground.

I need to know cliffs. Gullies. Ravines. Gulches. Steep slopes. Hogsbacks.
Ridges. Of course roads & structures also are nice to have on those maps.

> The OSM project concentrates on the artificial/cultural aspects of the
> topography (streets, buildings, names, and the like). To create maps
> containing natural topographic background information from the geographic
> OSM data, there's some low-res data available. Users and applications
> that need more detailed natural topography include it from other sources.

That then is probably the reason why OSM topographic maps suck so bad.
Do you see the difference in the zoomed-in section I showed you above?

The OSM maps are not only inaccurate but they're missing most great detail.

> Reasoning behind this is concentration on geographic data, that can be
> easily reviewed (and if necessary: corrected) by the participants of the
> OSM project.
>
> Contour lines, soil and hydrography data and the like are, OTOH, especially
> suited for a combination of remote sensing and modeling or need inventory
> by specialists (often with tailored equipment and very specific laboratory
> analyses, for instance for soil and aquiver mapping).
>
> Contour lines with high resolution, for instance, can be easily calculated
> from high-res DTM data. If the latter is freely available (nowadays usually
> from airborne laser scanning with meter or submeter resolution), it is
> just a matter of computing power and/or time to create the former. Here
> in Europe free availability of this data is not a problem. I can't assess
> the availability in the US, though. Shuttle Radar Topography Mission (SRTM)
> is globally available, but just with resolution of 30 meters, which isn't
> too great.
>
> Programs like OSMAnd usually combine selected and attributed data excerpts
> from the central OSM database with contour lines from another source. If
> you get higher resolution contour lines from a better source for your area
> you can use it to overwrite the standard set. Be aware though, that high-res
> contour line vector data may require quite some computing power to render
> on-the-fly.
>
> For the US, TopOSM may be a good source for OSM based maps with enriched
> natural topography:
> https://wiki.openstreetmap.org/wiki/TopOSM
> It seems to be an ongoing project, though. So you might be well-advised to
> confer with them directly.

Thank you for that information, which is new to me, and which discusses the
granularity when they say "Hillshading and contour lines are derived from
the 1/3 arc-second National Elevation Dataset from USGS."
https://www.usgs.gov/search?keywords=Elevation

I generally use the 1:24K maps for backcountry off-road off-trail hiking,
where I have to look up what it means to be using 1/3 arc-second data.
<https://data.usgs.gov/datacatalog/data/USGS:3a81321b-c153-416f-98b7-cc8e5f0e17c3>

That implies about 10 meter resolution, but I'm not sure how that compares
to the USGS 1:24000 maps that I habitually load into PaperMaps on Android.
https://www.nwcg.gov/course/ffm/mapping/52-map-scale

I didn't know about that project as I had already long ago given up on the
existing OSM topographic maps, but I always wondered why almost all people
didn't seem to understand when I said the OSM topo maps are atrocious.

I think they don't know how great the USA USGS topographic maps are, so
they don't know enough to assess how terrible the OSM topographic maps are.
https://i.postimg.cc/4yP4wZGg/test1-usgs-vs-osm-colorized.jpg

Looking at the Wiki for TopOSM you referenced, the first thing I noticed is
they use the USGS map data, so that's a good sign in terms of accuracy.
https://www.usgs.gov/faqs/what-types-elevation-datasets-are-available-what-formats-do-they-come-and-where-can-i-download

I appreciate you pointed that effort out as it seems the way they differ
from the source USGS topographic data is they strive to *present* the maps
in a "more pleasing" way - which is fine by me as

Unfortunately, when I tried to download the OSM 1/3rd arc-second maps from
the wiki page referenced, they all came up with a 404 not found error.

>> While I'm sure most people follow established trails, I shun them.
>> I pick a point on another ridge a mile LOS away and head toward it.
>
> If this was the case, you'd have no use for a program explicitly created
> for the navigation along artificial lines, like OSMAnd. In the scenario
> you described first, OTOH, you wanted to follow a pre-defined track. (It
> doesn't matter that this track would be hand-drawn by yourself and would
> perhaps not follow any real lane or trail.) And you wanted to get routing
> information whenever you leave this track (too much) or whenever you have
> to take a major turn of direction. If you still want to go this way, but
> find no source for better suited contour lines for your area, you may
> need to use a two-fold approach: Use OSMAnd for navigation commands along
> your predefined track and use another program for the visualization of
> your current position on (raster) USGS topo maps. If you just rove the
> wild with a rough idea of direction and target, you can forgo navigation
> completely...

You are astute in that most of backcountry navigation is dead reckoning,
but often you're deep in a gully or on your hands and knees in a thicket,
where dead reckoning is much harder due to lack of farfield visual cues.

Also just physically staying on a brushy hogsback seems so easy when you
look at the problem set on a map, but inevitable small errors tend to
multiply such that you'll be on the wrong side of that hogsback within
minutes if you're not constantly following the previously drawn track.

The drawn track is especially important when you're aiming to cross ridges
at an optimal cleft, since once you aim downhill at the wrong spot, the
terrain quickly becomes almost impossible to traverse in a matter of feet.

About the only time you go in a straight line is when you're rappelling.

Anyway, I appreciate the advice and the new information from you that the
OSM topographic maps are based on the USGS 1/3rd arc-second USA data.

May I ask for you to give me advice on two questions that are still open?

1. Do you see the huge difference that I see in these two zooms?
https://i.postimg.cc/bYtQgB42/test1-usgs-vs-osm.jpg
https://i.postimg.cc/4yP4wZGg/test1-usgs-vs-osm-colorized.jpg

The reason I ask is I hear from everyone how useful the OSM topo maps
are & yet, when I use them, I find them lacking in almost every way.

(But I had not known about the 1/3 arc-second OSMTopo maps, which I
don't have yet - but which I will search to find a source of today.)

2. Can you help me compare the "granularity" of the OSMTopo
1/3rd arc-second data versus what I'm currently using, which is the
1 inch on the map = 24,000 inches on the ground?

Are they the same?
Or different?
Which is more granular?

Thanks for all your help and advice!