QUOTE (Phil Stooke @ May 28 2006, 04:10 PM)
...
(another edit... just overlaying points on the map. It's unfortunate that the route has been so close to the line of site to Beacon... you really wouldn't get very much parallax anyway. Go sideways a bit, drivers! - maybe this isn't so convincing after all)... but this will change soon. Corner crater has a very different viewpoint.
Well, after spending (wasting?) all Sunday morning working on this problem in true Mars geek style, I see Phil has scooped me! Here's what I've come up with. To summarize: Far-siders, be afraid. Be very afraid.
Near siders, rejoice!
After pointing out in my last post that there wasn't enough parallax with the L/R pancam separation to settle this, it occured to me (and Phil!) that the greatest horizontal parallax along the route is actually much greater - several tens of metres actually. So I found the two sols that maximized the parallax. As you can see from this route map they are sols 809 and 816 (the yellow lines terminate at those sols' positions; green dots are positions for all sols for which the beacon was visible):
Click to view attachmentYou can also see from that map the parallax displacement that must occur between features on the near and far rims when you flip between the sol 809 and 816 views - it's just the separation between the two yellow lines on the near rim, which is about 20 metres. At the sol 816 distance, that corresponds to about 65 pancam pixels.
So I took the sol 809 pancam of the beacon, enlarged it by 15% so the image scale matched that of the sol 816 pic (on sol 809 Oppy was 15% farther from the near rim than on 816). I then made a flicker gif of the two images. I also added white dots to show how much parallax the beacon
would have had if it had been on the far rim, based on the 65 pixel value from above. The result is clear:
Click to view attachmentYou can see that even though the dark features are vague,
a 65 pixel shift would have been completely obvious! There's a small shift of the beacon relative to the dark streak to the left, but that's entirely consistent with the fact that the dark streak extends quite a bit closer to us than the rim edge.
Conclusion: the beacon is on the near rim, and quite close to the edge. We have plenty of parallax already to prove it. This method is different from Tesheiner's, and the results are stronger I think. The beacon's not obvious on the orbital map because it's not tall/steep enough to cast shadows.
Here's an analglyph version of the flicker gif. You can see that had the beacon been on the far rim (white dots) it would've looked much farther away:
Click to view attachment