User:Henry

From Tauwiki

2009 August 30 - my new text has disappeared! And where is image I just uploaded?

Dick

2009 May 2 : Been quite some time since I made the remarks below! Sujatha and Professor Murthy have made a lot of progress, particularly on understanding "airglow" which I guess we are using as a generic term for terrestrial as opposed to astrophysical sources.

TeX is input for equations:

The Wiki shows good progress!

A few very general comments:

We have a very new situation with SANDAGE showing, I think conclusively, that there is no significant extragalactic contribution to the high-galactic-latitude UV background.

The conservative (i.e. correct) approach is now to try hard to account for ALL of the Sandage signal as starlight forward-scattered by dust.

With g~0.6 we will potentially get contributions from stars very far from our target:

So what is most urgent is for Sujatha to continue and elaborate the modeling process (on which she has done so much already).

Single scattering I expect should be entirely adequate. This, despite the fact that the E(B-V) for SANDAGE is so enormous.

That is because we see no sign of the dense dust cloud in our image, which not only means that a significant extragalactic contribution is excluded, but also that there can be no significant back-scattered UV from the dust cloud.

(The Henyey-Greenstein predicts none, with g = 0.6, but of course Henyey-Greenstein is merely a model, and a mathematical one at that --- no physics in it!).

In the modeling, the value of the albedo can simply set at a = 0.3 and there is no need to vary it, since the predicted signal is simply directly proportional to the albedo.

Now that I think of it, it might be as well to set the albedo equal to one. It can't be more than that, and our aim is to see what the MAXIMUM signal can possibly be, for UV starlight forward-scattered by dust.

The sources for UV forward-scattered by dust are fixed and are well-established in the database that the model uses. There is nothing to vary there!

In fact the only thing that can be varied is the distribution of the dust between us and the dense cloud.

But do we even know how much dust there IS between us and the cloud? In that sense, SANDAGE is about the worst possible target for this modeling exercise, since it is so hard to know where the RELEVANT dust is located and how much there is!

It does seem to me that the modeling program should be modified to automatically produce a Standard Report. That report would look rather like the Table that Sujatha already has in the Wiki, but would include additional information.

In particular, the report should indicate How Much Dust was assumed in the model, and Where the Dust was Assumed to Be (distance, or distances). This information is essential for distinguishing one output from another (you need to know the input!)

I do believe that the report should also include a list of the contributions from each star (say, each star that contributes more than one percent to the predicted signal). I suggest that for each such star, the star be identified, its luminosity listed, and its angular distance from the target listed, and of course its absolute and percentage contribution to the signal.

I must say that for the task of testing whether the total high-galactic-latitude signal can simply be due to starlight scattering from dust, our Target One is much better than SANDAGE, because its E(B-V) is so low! All you need to do is set albedo=1 and vary the distance of the known amount of dust, and see if you can match our GALEX brightnesses!

Keep in mind that, no matter the important result we have in hand from SANDAGE, there is still a big puzzlement: the "ledge" effect: that with Voyager, for both SANDAGE and Target One, we see little (SANDAGE) or nothing (Target One) below about 1216 Å. Yet, there is a strong signal on both targets longward of that wavelength. On the face of it, that says that most of the signal CANNOT be starlight scattered from dust, because we KNOW the albedo and g values do not change significantly over 912 - 1800 Å.

As I say, I do think the best dust-scattered-starlight testing is with Target One (l = 331.1, b = -71.6) rather than SANDAGE.

For SANDAGE, however, there is additional extremely important modeling that does need doing, and that is the establishment of quantitative limits on both the extragalactic contribution, and the back-scattered starlight contribution. This modeling is beyond what the standard program does; it means putting in the detailed dust distribution over the target (from Schlegel) and blasting it from behind with an assumed uniform extragalactic background and seeing what partially-absorbed signal emerges. This will definitely require multiple scattering! And the other part is blasting it from the front with a uniform flat (like a hot star) spectrum and seeing what it takes to get a back-scattered signal. It does seem to me that at least the first of these is essential for a paper on SANDAGE since the main theme of the paper will be the discovery that there is no significant extragalactic background (to our great surprise!).

I had suggested an ApJ Letter, with focus simply on that discovery ("no extragalactic"). Reading over all of the above, I now wonder if we would not be better off with a full paper covering all of these issues. That would not fit in a Letter. And it would have to include Target One if we go that route.

Well, enough for now! This is going very well indeed! I am very excited about these results!