Thanks all for weighing in!
Let me clarify a few things, I was trying not to write a novel so I wasn't totally clear. First, we're still only doing very basic runaways. I'm not sure what you count as scent work, but so far it's very basic. The way my team usually does it seems to be that we get them really solid on runaways, then add the alert, then start doing more actual scent work. Is that different from how others do it? I should add that we are by far the newest members of the team (in like a year) and the team also doesn't have a regular official trainer (though several of our members have well over a decade of experience with SAR dogs, and we bring in clinicians regularly), so I haven't really had the benefit of seeing a method or another dog trained up from the beginning.
Also, the thing about him ignoring the subject was initially a problem, not really anymore. It took us a bit to figure out how to motivate him. We currently use food rewards as he's one of those dogs who is so food motivated that he practically has an eating disorder (were initially using toys as he is also fairly toy-driven, but not enough for it to be consistent if he was distracted), and now that we've hit on the right reward he stays engaged with the subject until they disengage.
The GPS collar thing was kind of a joke, I do a lot of canyoneering and I'm definitely familiar with losing signals.
I'm not seriously worried about losing the dog himself in a long-term sense, as he has very good recall and is almost too bonded to me so if I couldn't find him, I am confident he'd come looking for me eventually. However, obviously that makes me nervous in the event that I do manage to get this dog certified and we're in a search situation, it would be terrible for him to find someone and then me not realize it because my slow butt can't keep up with him. My team also does mostly work in the mountains and canyons in some pretty rugged terrain that he can cover a lot faster than I do. On the other hand, his range when we hike off-leash isn't very far at all, but that might change when he's working.
We do have a couple of stay-and-bark dogs on our team who do fine. After more thought, I think that's the only stay alert I'd be comfortable with. The passive alert was just thrown out there in a team discussion on the subject, but it just seems too risky. I'm now leaning again to recall/refind, but I think I read in another thread that some people don't like that--is it a situational thing (like urban/disaster searching vs. wilderness search) or is there a training reason for it?
Sorry for all the newbie questions, we've only been at this a few months and there is quite a learning curve! I read and learn as much as I can but it seems like there's always more contradictory (or at least seems that way to me) information popping up.