Story of missing hiker in Joshua Tree NP

[MidwestArmchair]

I've been following this for a lot of years. Forgive me if this has already been thrown out there.

What if the Saturday night/Sunday SAR was to Ewasko as China Lake was to the DV Germans?

For a moment, let's (1) accept that the ping occurred on Sunday at 0630 (about an hour after sunrise) and (2) disregard the ping radius for now.

Let's also accept: (1) The ping on Sunday morning would indicate that Ewasko was more than likely mobile at that time. (2) Ewasko would have expected a SAR by end-of-day Friday. (3) The SAR didn't begin until Saturday evening.

So, one of two scenarios occurred:

1) With no signs of a SAR, Ewasko began to attempt self-rescue on Friday evening but didn't get into cell range until Sunday morning, when his phone got one solitary ping out at an opportune moment that did not happen before and did not happen again. IIRC, this has generally been the accepted scenario, although met with perplexity. (I have trouble getting my head around it, to be honest. It's a lot of variables.)

Or

2) Ewakso decided to wait through Saturday for a SAR, planning to attempt self-rescue on Saturday evening. This hinges on him knowing a SAR would happen and potentially rationalizing that Mary didn't notify authorities until later than he expected.

My sense is #2 happened. He had limited water and got into a bad situation sometime Thursday: injured with limited mobility, lost, no cell signal, and further out than he expected to be. He decided to hunker down until the SAR came, conserving energy, water, and supplies. Any movement would occur at night, rather than in the day, again to conserve energy, supplies, and water. Then, whatever he saw or heard on Saturday evening triggered the cell phone ping on Sunday morning. Did he see or hear the SAR efforts? Did he realize that he was much further out than he realized and that he would not be found in time? Did he move closer to the SAR efforts overnight on Saturday, then try to use his cell phone as a signal mirror after sunrise on Sunday morning? (FWIW, the "lights in the saddle of the ridge" reported in the FOIA narrative are too coincidental to be a coincidence, and I wonder about that a lot. Was he trying to signal them?)

If my expanded #2 premise holds true, then where would Ewasko have reasonably been (again, disregarding the ping radius) where (1) he could see/hear/notice the SAR on Saturday evening but (2) be out of range of them enough for him to decide to move toward them and (3) be out of range enough for the SAR to not notice him. Is there a place that makes sense and hasn't been thoroughly searched yet?

My two wild theories would be:

1) he parked at the wrong trailhead (he had written directions in his car, indicating he was not overly familiar with the park) and decided to hike to his actual destination (all I can think is Keys View or Queen Mine/Lucky Boy, which is crazy). I've done that many dumb times but not in a desert in the summer.

2) he meandered down toward (not to - just toward) Coachella Valley Preserve through Covington Flats. The ping radius actually intersects down there in the jagged canyons, IIRC. The rock cairn and other clues discovered in Covington Flats may point toward this, but it's inexplicable for someone to go into that unforgiving area.

This is probably just noise; this one has been bugging me again after a lot of months, and I saw some activity here. So I'll echo others and say how incredible every effort has been to find Bill Ewasko, especially after all this time. Truly incredible.

[MidwestArmchair]

As I thought more this weekend about my mini-theory, considered the Serin Tower splash map, and read some of the early JT reports - what if Ewasko "vanished" without any confirmed trace because he never went to Quail Mountain? What if Ewasko headed east on the CHRT toward Crown Prince, got off the trail, and ended up somewhere on the eastern side of Ryan Mountain on Sunday morning?

On his itinerary, Crown Prince + Lucky Boy Vista + Johnny Lang were bracketed. He also had the Quail Springs Picnic Area noted next to this itinerary item. Johnny Lang has been searched due to the ping radius, but Crown Prince and Lucky Boy have never been in play (IIRC). Tom Mahood searched part of Ryan Mountain in JT24, also considering that Ewasko went opposite of Quail Mountain. I would argue, however, that Ewasko didn't plan to climb to the peak of Ryan Mountain: Ryan Mountain stood in his way to get back to safety.

Here is what I think he planned: park at Juniper Flats --> hike east to Crown Prince via CHRT --> north to Queen/Lucky Boy --> east to Quail Springs Picnic Area for a rest --> Johnny Lang Canyon --> back to Juniper Flats to leave. Quail Springs Picnic Area could also have been a good end point - a built-in failsafe for him, if Johnny Lang was too much for that day (which, yeah, because it's an insane, ambitious hike to begin with). (Or he did Johnny Lang first, parking at Quail Springs Picnic Area, and then decided to hit Crown Prince via Juniper Flats, explaining his late arrival there.)

I think he changed his planned hike fairly early and jutted north toward Queen/Lucky Boy, coming to grief somewhere in the backcountry between there and the CHRT. He may or may not have seen the SAR focusing on the Quail Mountain area, made it to the southeastern side of Ryan Mountain by Sunday morning through washes and drainages, inadvertently picked up a sliver of coverage on the eastern side (there's one official spot), and didn't make it. The only search tracks I see near Ryan Mountain are helicopter fly-overs around day five of the search, plus JT24.

Besides the ping radius, there is just about as much evidence for this scenario as there is for the Quail Mountain scenario. And it could explain how a hiker vanished into thin air without one single trace for nine years, although he was out there and alive for four days (including Sunday). This theory, of course, throws out the ping radius, as it's about 20 miles from the Serin tower to the eastern side of Ryan Mountain. I don't know if there would be another cell phone tower in play that would have picked up his phone. However improbable...

Edited because I don't know my directions. *sigh*

[Ric Capucho]

Hi MidwestArmchair,

Thought I’d respond with a few thoughts about your theory, which isn’t totally out there. So don’t read this as a debunking, more of points that you wish to consider if you choose to modify your scenario.

1. Your idea that Bill consumed the missing time by doing a quick dash up and around Lost Horse has been discussed on and off for years. I say yea and others say nay. A recent surprise that weakened my resolve is that Bill likely took only two 12 ounce bottles of water with him, and if he’d just done Lost Horse then he’d either have already consumed them and needed to replenish them, or he’d have been thirsty as hell by the time he returned to his car, and there’d have been a strong compulsion to empty at least two bottles and replenish his rucksack with a bunch more.

On balance I still suspect Bill did the Lost Horse loop, but no longer would bet my ass on it.

2. The possibility that Bill marched off eastwards is credible and indeed the possibility was taken seriously by the original S&R team.

3. The cell coverage that may or may not be on the easterly flank of Ryan is unlikely to be from the Serin tower. The Verizon splashmap includes coverage from all of their towers in the area.

So now we get to the crux of it: how much do we trust the ping? Was it truly received from Serin? Or did the technician mix up his cell tower data? After countless hours reading ping info on the web, it’s unlikely that the ping tower or total ping duration was wrong, but the inferred range can be impacted by reflections from the terrain, reflections from an inversion layer in the atmosphere, and another impact that deserves its own point (see second next).

4. The ping duration means the time for the signal to go there and back, so as the speed of light is fixed, then for a given ping duration it’s impossible for Bill (at the time of the ping) to be further from Serin than a given distance. A reflection that leads to a slightly longer ping unintuitively means that Bill was closer to the tower than the ping duration might infer. So the maximum distance from Serin can only be equal or less than the ping duration infers. Not more... or could it?

5. Bear with me. Back in 2010 ping ranging was quite primitive (and going on the Nguyen/Orbeso ping ranging it ain’t much better) and the calculation procedures only recently put in place as a result of some new government legislation; the name of which escapes me. One huge variable that has to be considered is the internal latency of the phone’s electronics. So “ping” goes the tower, then the phone hears it and electronically decides how to respond, and then gets around to respond; and that internal thinking time is called latency.

So ping there plus latency plus ping back equals the total ping duration. So ranging better exclude the latency, right?

And latency isn’t fixed as phones are often doing other processing in parallel. So latency is a range of possible delays, so the ranging calculations simply use an average. But the impact of a low latency versus a high latency can have a huge impact on the ranging. So the ranging gives a plus/minus 10% error band. In this case 10.6 miles plus/minus 10%. So Tom’s website has long shown three ping arcs, one at 10.6 miles, and two more at 9.6 miles and 11.6 miles.

Let’s consider what that really means: if your cell phone is free of any other duties, and hot on the ping, then the latency is low and it gets a return ping on its way promptly. So most of the overall ping duration is down to the physical distance between tower and phone and therefore the true range is towards the upper end, in this case 11.6 miles. And if the phone was busy doing other stuff such as Candy Crush then the latency can be quite high, and that takes a bite out of the overall ping duration leaving less time for the physical distance... and the true range decreases to 9.6 miles.

Note, the layency doesn’t magically move Bill forwards and back by miles, but it does move the calculated range forward and back, and potentially misleads searchers where a mile in rough terrain may as well be a dozen.

But there’s more.

Serin is a 3G CDMA 2000 compliant tower which means it’s also backwards compatible for 2G cellphones. And here’s the rub, 2G phones have a different latency range from 3G phones. And don’t assume they were slower, because they weren’t; no fancy apps, no fancy bells and whistles to divert processing power away from its core purpose as a cellphone. So what sort of cellphone did Bill carry? 2G or 3G?

Dunno, but if it was a 2G then we could increase the average ping circle out by another 10% or a mile (average 11.6 miles plus/minus 10%).

Still a long way short of your twenty miles to Ryan, but hell your theory’s as good as anyone’s.

Ric

[Myth]

Speaking of Nguyen/Orbeso, the thing I'm turning over in my mind these past few months is this: from memory the ping in their case was at one point thought to have come from the general area above Quail Springs, wasn't it? It generated an SAR search out there - I remember discussion on here about how that lent some coverage help to Bill's case.

But, as far as we know, they were never anywhere close to that area. If they were it would be very odd. It is a long walk between Quail Springs and where they were found, and you'd have to cross the main road through the Park.

That's leading me to think that something might be up with that Serin tower and its pings - it isn't very trustworthy for whatever reason, and it likes the area above Quail Springs.

[OtherHand]

As far as I know, the info was never released as to which system's phones resulted in the bogus pings for Nguyen/Orbeso. The fact that the pings were given as lat/long values suggests it was a provider other than Verizon, or that Verizon has now upgraded their location recording setup. Remember it's been a long time since Bill's pings. Oh, and of course we don't know for sure if it was the Serin tower the Nguyen/Orbeso phone connected to. There are several other towers out there that are candidates.

Setting aside lat/long, in terms of distance, the two recorded Nguyen/Orbeso pings were 8.2 and 11.2 miles from the Serin tower (assuming that is the correct tower). They were found 10.25 miles from the tower. Average the two pings and you get 9.7 miles, not too far off. But I'd be real interested to learn why the lat/longs were so hosed up.

[Myth]

As far as I know, the info was never released as to which system's phones resulted in the bogus pings for Nguyen/Orbeso. The fact that the pings were given as lat/long values suggests it was a provider other than Verizon, or that Verizon has now upgraded their location recording setup. Remember it's been a long time since Bill's pings. Oh, and of course we don't know for sure if it was the Serin tower the Nguyen/Orbeso phone connected to. There are several other towers out there that are candidates.

Setting aside lat/long, in terms of distance, the two recorded Nguyen/Orbeso pings were 8.2 and 11.2 miles from the Serin tower (assuming that is the correct tower). They were found 10.25 miles from the tower. Average the two pings and you get 9.7 miles, not too far off.

That is a good point, OtherHand. We don't know if it was Verizon, we don't know if it was the Serin tower. It just struck me that that area was indicated again.

But I'd be real interested to learn why the lat/longs were so hosed up.

Me too! With the time I don't have I am planning to do some thinking on that and see if something like using an incorrect projection, an offset error, something ( ???? ) could account for it.

I'm a programmer by trade. I've dealt with many bugs ( some not even caused by me! ) where the results were baffling and the explanation, once you had all the data and could really look it over, simple and sound. I'm just itching for an environment where I could actually have all the data and reason it out. Of course, that environment will never be in place for us.

[Perry]

Orbeso's phone was with AT&T. I believe it was the tower that can be seen from within the park because myself and Sh44 (both with AT&T service) had reception only when we could see that tower. Here's a map analyzing the error when guessing which tower it was:
https://caltopo.com/m/5R79

The southernmost tower in the map is what is visible in the park and strongly correlated with our AT&T reception. That would put their final location beyond both radii and all of their hiking significantly beyond the Ping 1 radius. The radius of Ping 2 matches very closely with the car which would explain why, according to early news reports, the cell data suggested the parking lot.

In a past post in another thread I believe, Otherhand suggested the direction could be based on comparing signal strengths of 120-degree directional antennas at a tower. Obviously not a very accurate method.

[Ric Capucho]

I’ll see if I can root it all out, but the Orbeso ping(s) resulted in *two* unlikely areas being mass searched.

The first was a rectangular area in Upper Covington just west of the dirt road. The second (I remember a day later) was indeed up above Quail Springs.

It makes no sense that Serin would have been the ping tower as the Upper Cov location is very dissimilar from QS in terms of ranging or triangulation or anything else. I never got round to seeing which tower would be equidistant from either location. Might get around to it now.

Ric

[Myth]

Thank you for putting together that map, Perry! Nice. That makes a lot of sense.

[Ric Capucho]

Looking at Perry’s excellent map, I can’t help seeing a slightly different interpretation.

My assumption last year was that the pings were somehow triangulated between two different towers, but then we’d be talking about four Orbeso pings in total, as pings aren’t somehow shared between towers. Duh.

Looking at the Serin arcs, the errors start to make more sense (to me anyway), as their true resting place was 10.31 miles from Serin and Ping 1 was calculated at 8.18 miles from Serin and Ping 2 was calculated at 11.3 miles. I’m ignoring the radial errors which were totally misleading (I have my theories, but wanna think them out a little more). That would mean that the first ping calculated out at 20% less than the true location and the second ping calculated out at 10% further out than truth. All from Serin.

Keep in mind that Orbeso definitely had a smartphone, so plenty of internal latency to muddy the waters.

I’m thinking Serin, again. And if so, wondering what it may mean for Bill.

Ric