Upper Perdition Falls

Back in October of last year (2021), I did a post about Perdition Falls, which is an unofficial name for the seasonal waterfall just to the right of Multnomah Falls. In that post I briefly mentioned something about there being another smaller waterfall on the same creek, just upstream and right next to the long-closed Perdition Trail. The trail closure means there's no legal way to visit the upper falls in person. It didn't occur to me that they might be visible from below, at least. But I happened to be at Multnomah Falls last week for lunch and as soon as I got out of my car some dusty weird corner of my brain went "Wait a minute, what's that?". There's really no mistaking it once you know it's there, and as a bonus that makes it easy to tell where the old trail was (and still is, unofficially).

The mildly weird part is that I don't recall ever seeing it before. This is only mildly weird, because my powers of observation are... a bit off. So it's entirely possible I've seen it regularly for years and it just sort of never registered somehow. But it also may have been completely hidden by trees before the 2017 fire, and it seems as though you can only see it from the I-84 parking lot and not from any closer, and I can probably come up with even more excuses if I have to.

In any case, I think I've located the falls on the state LIDAR map right around here, and I used the latitude & longitude from that placemark to create the embedded Google Map here, for anyone who has, I dunno, rocket boots and can get there without using the trail. The advanced technique of clicking around the area on the LIDAR map and guessing what spots might the top and bottom of the falls gives me height numbers in the 20-40 foot range most of the time, with the wide error bar completely due to the human in the loop.

The original dataset has latitude & longitude numbers out to a whopping eleven digits, which I actually had to trim down to six to make Google Maps happy. Which got me wondering just what these decimal places mean in terms of physical distance. One degree latitude comes to about 69 nautical miles, or 111.1 km, anywhere in the world, while the length of one degree longitude varies by latitude: It's the same 111.1 km at the equator, but around 79km at the 45th parallel, and a bit shorter than that in the gorge (the number eventually goes to zero at the poles). Going with the latitude number, one decimal place is 1/10 degree, or 11.1km, while a change in the sixth decimal place is 11.1cm. At eleven decimal places (or 10^-11 degrees), a change in the last digit is a distance of 1111 nanometers, which just so happens to be about one wavelength of near-infrared light of the type typically used in LIDAR. So while 11 digits looks impressive, I'm not sure how many of those are actual significant digits.

So I stumbled off a Google tangent at that point, as I tend to do, so here's a short list of links mostly about lasers that I'm not even going to try to relate back to the subject of this post.

• Slides from a Portland State geography class explaining how airborne LIDAR works, aimed at people who might be using the data later in the term.
• A 2018 paper calculates theoretical accuracy limits for LIDAR in self-driving cars, and comes up with something around 0.1mm.
• A 2019 paper proposes that better resolution can be achieved by ditching the lasers in favor of spooky quantum magic with entangled photons.
• Which in turn leads to Wikipedia articles about things like quantum metrology (which LIGO might use someday) and quantum lithography (which might be used in chipmaking someday)
• A paper about a specialized Leica camera designed for LIDAR applications. This particular model was from way back in 2011, so it's useful information in case they start showing up at Goodwill or something.
• There seems to be a lot of overlap between LIDAR vendors and major defense contractors, so you're always just a few clicks away from stuff like this article. One of the lesser-known Geneva conventions from the 1970s bans using lasers to blind people, and another bans any weapons that cause 'undue suffering'. The article argues that there's a way around these unfortunate legal obstacles, which is to use a laser powerful enough to instantly vaporize whoever it's used on, so they don't suffer first. And no suffering means there's no problem and you can go around lasering people to your heart's delight. Of course those lasers don't actually exist yet, not in airborne form at least, so (as usual) a massive federal research program is needed in order to bring this inspiring dream to life.