Railworks America

Hello,

I was curious if anyone knows of a site that offers historical DEM, like for an area before a large dam was built and was flooded. Or maybe a program that can assist with creating something similar, say from an older topo-map. If such a thing doesn't exist yet, I'll just have to wait a while I guess. :P Thanks in advance. :)

That would be a Herculean if not impossible task, and the resulting resolution would be far less than the worst DEM currently available. Most of the current data is derived from satellites or shuttle missions, with the remainder being carried out by specially fitted aircraft, and I'm certain DEM data wasn't even considered with the advent of the computer. I suppose someone with infinite time on their hands could muddle through topo maps, but even if the DEM was right on the money, there's enough error in those maps to not make it worthwhile.

IMO, it's best to use what's available, research you given time period, and then adjust the DEM by hand.

Maybe half a decade ago, before they shut down the old website, I ran across a note stating that USGS had limited range of "backdated" DEM data sets created for areas which had been significantly altered by construction of certain features such as major dams and other construction projects. I don't know what the extent of those data were, nor where they might be found on the current website, but an e-mail them might turn up something useful. Be advised it might take them a bit of time to find it, if they actually have what you're looking for.

philmoberg wrote:Maybe half a decade ago, before they shut down the old website, I ran across a note stating that USGS had limited range of "backdated" DEM data sets created for areas which had been significantly altered by construction of certain features such as major dams and other construction projects.

I would imagine that this data would still be recent given the above info.

Hack wrote:I would imagine that this data would still be recent given the above info.

That's something I would barely hazard a guess about. IIRC these were 1/3 arc sec datasets: depending on the feature involved, I suppose it's possible that some of them coupld have been based on data from their earliest work at that resolution. Personally, I would hope they adjusted more recent data, since that tends to be higher quality at any resolution. A former colleague of mine (the same one who told me about the LIDAR project I mentioned in another thread) told me that part of the problem with producing such "backdated" data is that the Federal programs are pervasively stovepiped: if, for example, USGS wanted to issue backdated data for the Black Canyon of the Gunnison River, any terrain data below the mean low water level of the reservoir would be defined as bathymetry, and would have to be obtained from a different agency (NOAA, IIRC), while the bathymetry (and the navigational charts developed from it) tends to be measured fromt he mean high water level. Resolving the two datasets, as well as compensating the NED for the removal of the two dams would be tricky, at best, a more likely a real headache.

philmoberg wrote:That's something I would barely hazard a guess about. IIRC these were 1/3 arc sec datasets

IIRC, the NED 1/3 arc data is no older than 1998, possibly 2000 and from satellites, whereas earlier data - 1990s up to early 2000s - was obtained via shuttle missions (SRTM) and is of lower resolution. However, there may have been some classified CIA data somewhere that we're not privy to.

FWIW, when Rail Simulator was first released I had been working on a Yosemite Vally line from Merced to El Portal. Like the OP, I ran into the same issue of flooded areas after a dam had been installed in the 1930s, and had to hand-edit the DEM to match the area pre-flood. Knowing track elevations for the time (and using old topos, water tables and photos) helped in getting things to "look right". In the end, the user is none the wiser about your edits anyway.

Thanks for the information. I'll see what the USGS has to offer in regards to historical DEM's. I was curious as to what was out there. Luckily it's not the end of the world if there isn't any data available for the area I'm working on. Trinity Lake in Northern California is good sized lake so, I won't be able to adjust anything by hand, but I'll just have to work around it. Thanks again. :)