Discussion: USGS Topo & Hill Shading

I've been playing around a bit with the USGS National Map Viewer, and I'm trying to get a sense of where the data is coming from.

The contour intervals shown on the digital National Map, say 20 feet, match up with the previous paper versions. The hill shapes are generally the same, too. But in the areas where I'm looking (central Washington state), the old paper versions show more contour detail (smaller hills and depressions) than the newer digital contours do. Where did the detail go?

A related question has to do with the data used to create the hill shading. The National Map seems to agree with what's shown in Google's Terrain View. But in an area where I'm looking, the hill shading has more detail than the contours of the National Map, but it doesn't quite match up with the older paper versions. Where is this detail coming from?

I haven't played around with the GIS downloads yet, but if I were, which version of the terrain would I get? Something closer to the original paper, or the National Map contours, or the hill shading?

I don't know what they're thinking with the new 7.5 minute series. These new maps suck. From the few that I've seen, the new ones are an embarrassment. I hope the USGS gets some funding to hire some cartographers (apologies to real cartographers in the USGS).

I realize the new maps should be digital rather than paper, but I just don't have words for how disappointed I am with the new maps. Seriously, the new Apple Maps app for iPhone, that is automatically generated at an arbitrary scale, has better, more artistically pleasing maps than these new digital pdfs. (Obviously, the Apple Maps aren't topos, but come on.)

The link I'm talking about is this: http://store.usgs.gov/

then click on "map locator and downloader". (Oh, and the download interface kinda sucks too, although if you figure it out you can download old versions of maps back over 100 years in some cases.)

I've been looking at the 7.5min series as well as the 15 min and 100K series. I've been using the 100K series to plan some long bike rides. Just from what I can see on the geo-pdf, the actual paper maps are beautiful. Unfortunately, the ones I want were last updated in 1986.

The shaded relief you see in The National Map is raster data derived from NED.

NED 1/3" data is almost exclusively derived from sampling the old paper quads (extract contour lines, then interpolate). The process is not reversible: when you derive contour lines from this data, you don't get the old contours back, small hills and depressions are missing and pointy reentrants are less pointy. Why? don't know. Backstreet Boy took one look at the result and went with old paper maps for Joshua Tree two years ago.

Use the EarthExplorer site and look for 1/9 arc-sec data.

Thanks for the responses so far!

In EarthExplorer, where do I look for the 1/9 arc-sec data? It's not obvious where that is.

I guess my next question is, if I want a USGS-derived basemap, what is my best source?

On the (USGS national map) viewer page, in the advanced tab, leftmost button -- download by bbox, choose elevation, and 1/9 if it exists otherwise 1/3. Then probably some combination of GDAL tools will get you a DEM that OCAD can read. I've done this some time ago, will have to experiment to get all the details.

The NED is already a DEM. I don't have newer OCADs, so I derive contours from the DEM in TNTmips. I am cheap (because my participants are cheap), so can only use the free TNTmips. So, I have to manually tile the DEM in the National Map Viewer. This is still possible for 1/9" data, but will go away mid-April, and the 1/3" data is no longer user-tileable, all you can get is one-degree-square chunks, which don't fit into the free nor the cheaper paid TNTmips.

So, your best bet, lacking the newer OCADs (10+), is to fork over $400 for Global Mapper. You won't have any problems converting anything between pretty much any possible format, nor with the file size. The full-versioned TNTmips can also do all of that, but is much more expensive.

I guess my next question is, if I want a USGS-derived basemap, what is my best source?

As far as I know...

- data from lidar.cr.usgs.gov if offered for the area in question,

- otherwise, 1/9 arc second NED data if it exists for the area in question,

- otherwise, older USGS quads are, as has been noted above, often better than 1/3 arc second NED data. Newer ones created, IIUC, by automated photogrammetry, not so much.

I thought they take information from the 10m-grid DEM, which is publicly downloadable.

also lidar.cr.usgs.gov is strictly out of date for WA

Of one area I'm thinking about, there's no LIDAR (the closest the Puget Sound Lidar Consortium comes is within about 5 miles).

Kevin helped me out with the USGS stuff, and now I have an OCAD file with basemap contours, which I think will work out ok. The contours aren't quite as detailed as the ones on the paper maps, but they are pretty close and it'll work out well. Plus, the contours here aren't too complicated anyway; the orienteering appeal is in the rock detail.