Mystic Pillar
Long have I gazed upon and admired one of the greatest images Hubble has ever taken. It's impossible not to pause and take it all in, even after all these years and all the times I've seen it. Ten years ago this was observed for Hubble's 20th anniversary. We're now approaching the telescope's 30th anniversary, which is guaranteed to be spectacular. Please do look at the original anniversary release to understand the history of this image: hubblesite.org/contents/news-releases/2010/news-2010-13.html
One of the ways I learn to appreciate HST imagery even further is by processing it myself. There's nothing quite like staring at every little detail up close and personal to feel intimately acquainted with some data. I went into this thinking there wouldn't be much different that I could do that others haven't already done... and I was wrong, which was nice.
I learned something new recently that involves subtracting light out of imagery to reveal details that are otherwise lost in a kind of bright glare. It works very well with elliptical galaxies that are very regular in shape and easy to create simple, smooth models of. The idea was still fresh on my mind when I realized that the [O III] data seemed to match up with a bright gaseous fog permeating the landscape around the Mystic Mountain. I tried applying the same technique, and to my great surprise, it worked, and really well at that. The Mountain was deeply revealed in such a way that it became more of a Pillar. Yes, it was always a pillar, but it was hard to see.
There are a few places where subtraction was obviously imperfect (at least, obvious if you flip back and forth between the two) but it doesn't detract from the overall view. I also had to mask off the PSFs (point spread functions) of the stars since subtracting them out doesn't do any good.
Anyway, hope you enjoy it. As always, it's a privilege to get to work with these amazing data.
Data from the following proposal were used to create this image:
20th Anniversary of HST Launch
Note that mosaics are already composed and ready for anyone to download and try for themselves: archive.stsci.edu/prepds/carina/wfc3/
One more note: Subtraction was not linear and applied with varying weight to each filter. It's part science, yeah, but it's also art. This means eyeballing it and adjusting it until it looks "right" to me.
Lavender screen: WFC3/UVIS F502N (I used this to put some of the subtle color variation back into the image after subtraction)
Red: WFC3/UVIS F673N-F502N
Green: Pseudo
Blue: WFC3/UVIS F657N-F502N
North is down.