Lidar elevations

[GowerGulch42]

With regard to "known, established benchmarks", they were subject to surveying and mapping errors. Not to mention, vertical datums changing over the years.  Or, maybe I misunderstood what you meant by benchmarks?

There are some notable errors on the USGS topo maps where one or more contour lines are missing.  A great example is the Coxcomb Mountains range in California. There are three high points in close proximity.  And the 7.5' map shows the lowest of them as being the highest!  

Some of the surveying methods used since the late 1800s, namely Leveling Lines, are actually as accurate as modern systems. They use the same benchmark locations but their actual measurements (for the science of geodesy and the shape of the earth) are located at the same places decade after decade, which makes them very useful in my field of tectonics. This network of leveling lines forms part of the foundation of modern geodesy in the USA, and does not cover every single disc you find on a peak. Up until the last ~10-20 years topo maps were using the same elevation data that they'd been using for decades, and some of these are getting resurveyed and updated with time. Likely, the next edition of the Coxcomb 7.5' topo will have different measurements.

I brought up these benchmarks because their positions are extremely well constrained, usually within a few millimeters, by GPS measurements which can take 20 minutes to two hours or more. In lidar these can used as control points to make sure your data is spatially correct - i.e. fits to the world with the projection you want to use. I'm sure the folks doing this for peaks are more than capable of figuring out how to do it right!

Since it's been a few years, maybe I'll take a crack at Badwater again. There's a freeware called QGIS with which you can view and play with lidar data which you can download off the USGS National Map or OpenTopography, among others.