QGIS QGis 3.6 Noosa. Creating DEM from USGS Topo Maps... How?


gO Newbie
Nov 5, 2021

First of all, I hope I am posting in the right forum. Secondly, I hope this post is acceptable here, as I am finding out quickly that even asking for help in some GIS forums is a no-go.

Basically, the key point of this post: How do I create a workable DEM in QGis? What steps are necessary?

Here's the backstory:

For the last few years, I have been working on a project for Microsoft Flight Simulator X. A video of the project is linked here:

To be as brief as possible in the description of the the project... Basically, I am recreating the Mount St. Helens areas prior to 1980. As the May 18, 1980 eruption of Mount St. Helens caused drastic elevation changes in the upper North Fork of the Toutle River and Spirit Lake region of Washington State, the existing mesh (based off of 2004 and 2009 DEM data) that Microsoft Flight Simulator and ORBX Sceneries used is creating rather harsh terrain conflicts with the pre-eruption 30m DEM I have of the peak.

In my project so far, I've realigned the shorelines of Spirit Lake to the pre-1980 contour elevation of the lake. I've also realigned Highway 504 to its pre-eruption alignment on the valley floor. I've also eliminated two large lakes that did not exist prior to 1980: Coldwater and Castle Lakes.

The terrain conflicts I am most experiencing are the elevations of Spirit Lake and the area just west of it. West of Spirit Lake, the hummock deposits down to just below Elk Rock are making the pre-eruption alignment of Highway 504 resemble a rollercoaster. Prior to 1980, Highway 504 was a gentle-sloping highway in that area with long straight stretches, and the pre-eruption alignment in the last three mile segment (just before Spirit Lake) is five hundred feet lower than the current terrain elevation. Additionally, the pre-eruption elevation of Spirit Lake is 250 feet lower than its current elevation, owing in part due to the fact that the landslide off of Mount St. Helens' north flank slid down into the lake, raising the lakebed elevation to a depth higher than that of the pre-1980 lake surface. This, in turn, significantly rose the surface elevation.

In the sim, the pre-1980 shoreline of Spirit Lake is mostly flat before it rises into post-1980 terrain, as the pre-eruption Spirit Lake was significantly smaller than it is now. Prior to 1980, the ridges that bordered the east, west, and north sides of Spirit Lake were very steep, with a steep angle of almost 50 degrees down to the lake shore in places. On the east side of the east arm, was a 200-foot-high cliff with a waterfall (Harmony Falls; now submerged)

(The terrain anomalies I mention show up in the video I linked above)

In my efforts to recreate this experience, I've asked the USGS if pre-eruption DEMs exist of the coverage area I am working on, which includes the entire region destroyed in the 1980 eruption. In my inquiries, the USGS has given the impression that the pre-eruption DEMs of the area that were created in the early and mid 1980s for studies of the blast deposits were seemingly lost or destroyed. A few USGS officials have also given me the impression that they are unwilling to help. Some have not even responded at all. And inquiries to the current contact information for authors of a study in the late 1980s who used those DEMs, back when they were with the USGS, have gone unanswered now as of a month.

The only pre-eruption DEM I have, is of the volcano itself. This DEM was authored in 1983 as part of the aforementioned USGS study on the blast deposits, but is the only remaining DEM in existence. It covers just the peak itself, to an area one mile south of Spirit Lake, and to an area just north of Swift and Yale Reservoirs south of the peak. That DEM was sourced as public-domain material from a University of Washington GIS site.

To that end, I have downloaded a series of five 15-minute USGS quad topographic maps with contour intervals of 80 feet. These topo maps were downloaded from the USGS National Map service via the TopoView utility. These 15-minute quadrangles are, as follows:

Elk Rock
Spirit Lake
Mount St. Helens

The coverage area of these topographic maps exceeds (by far) the coverage area of the 1980 eruption. As a result, I have trimmed them down to approximate the coverage area of my project. I have sourced these DEMs as the result of coming across a study by a Portland State University student, who used the exact same quadrangle maps for a DEM stufy of his own. However, I am unable to find his contact information, and emails left to the Army Corps of Engineers (a partner in that study) have gone unanswered as well.

If this can be done, then how easy/hard is it and how long could this possibly take?

More importantly, what steps do I need to do, to make this DEM a reality?