On Friday evening I emerged from what have been a crazy couple of months, during which I’ve been in a paper crunch. When one works on a number of different projects at any one time, with a variety of collaborators, postdocs and students, the resulting papers generally seem to see the light of day at seemingly random times. Sometimes a few papers come out more closely together than at other times.
But sometimes it gets a bit ridiculous, and that’s what happened this summer, with a lot of projects suddenly getting close to completion about the same time – five papers to finish up in the space of a couple of months. I can’t complain – I love the work – but I found this particular random bunching of projects reaching completion to be a bit much, and I haven’t been getting much free time or even time to sit back and spend time focusing on any new ideas (or even time to blog!!).
But on Friday my collaborators (Rachel Bean and Eanna Flanagan, from Cornell) and I finished up two of the papers we’ve been working on. We’ve been studying a class of instabilities that can occur if dark matter and dark energy are coupled in a nontrivial way. There are a number of models in which this happens, and indeed one might think that if one tried to include both dark components of the cosmic energy budget into a simple particle physics theory, then there might quite naturally be couplings other than gravity between them.
We’ve done a great deal of work on this problem, and our long paper is sufficiently involved that we accompanied it by a letter that covers the main points and physical interpretations without all the analysis and examples that we have in the longer paper. For an idea of what we’re about, the abstract to the longer paper reads
We consider theories in which there exists a nontrivial coupling between the dark matter sector and the sector responsible for the acceleration of the universe. Such theories can possess an adiabatic regime in which the quintessence field always sits at the minimum of its effective potential, which is set by the local dark matter density. We show that if the coupling strength is much larger than gravitational, then the adiabatic regime is always subject to an instability. The instability, which can also be thought of as a type of Jeans instability, is characterized by a negative sound speed squared of an effective coupled dark matter/dark energy fluid, and results in the exponential growth of small scale modes. We discuss the role of the instability in specific coupled CDM and Mass Varying Neutrino (MaVaN) models of dark energy, and clarify for these theories the regimes in which the instability can be evaded due to non-adiabaticity or weak coupling.
It’s very satisfying to complete a project, although also a little sad, since one usually has such fun working on them, and learns a great deal. In this case, we’re still working on more projects and have lots of new ideas, so I’ll still be enjoying the collaboration, even though this particular project is over. Plus, I can now devote a little more time to a few other ideas I’ve been working on with my graduate students, who’ve been putting in the bulk of the calculational work on them over the last couple of months.
September 11th, 2007 at 12:07 am
Nice work — important and useful paper.
So … which one of the JDEM concepts would be able to constrain these models the best, in general?
(And I’ll bet you thought plain science avoids controversy
September 11th, 2007 at 1:59 am
Typically weak lensing (also known as cosmic shear) is the best at constraining dark energy when taken alone. But since other measurements, such as supernovae, are typically highly orthogonal to weak lensing at constraining parameters, the best thing to do is definitely to perform a wide variety of experiments. This was the foundational finding of the Dark Energy Task Force, that the best thing we can possibly do is make use of a variety of dark energy measurements (supernovae, weak lensing, baryon oscillations, cluster counts, and CMB). Fortunately the methods used in collecting the data for the different experiments have a very large degree of overlap, so it’s not overly difficult to use the same instrument for multiple dark energy missions (CMB and cluster counting missions can use the same instrument, such as the South Pole Telescope, which will do both; baryon oscillations and weak lensing observations can make use of the exact same data sets, and can even overlap some with supernova measurements with a bit of compromise on the scanning strategy).
September 11th, 2007 at 5:51 am
I’d pretty much agree with what Jason said. If you know in advance that you have, say, a cosmological constant, and your goal is to measure its value as precisely as possible, then the answer might be different. But if you are looking to idenyify the origin of cosmic acceleration, distinguishing between various suggestions, then what we’ve learned over the last few years (many pf us have worked on this) is that cross-correlating geometric measures with growth measures is the way to go. And doing this in multiple ways is best. Weak lensing is indeed one of the most useful things, but you definitely need it to go hand in hand with supernovae lightcurves, precision CMB measurements, and other large scale structure probes.
September 11th, 2007 at 1:53 pm
This is one of those topics, mentioned in a recent meta-thread, for which you will perhaps get fewer comments (compared to a post about religion or beer) because many of us are unqualified to comment usefully – but as I said I’d do in that thread, here is my “wow that is extremely interesting and I can’t wait to see what people have to say about it” post.
That’s fascinating, Mark.
September 11th, 2007 at 2:56 pm
Thanks tyler.
September 12th, 2007 at 2:47 pm
Whatever you do, don’t submit it to an Elsevier journal, unless you’re willing to have your speculations tested by the future evolution of the universe well before your paper is published.
September 12th, 2007 at 4:38 pm
I doubt we’ll be submitting to one of those, irrespective of the timescale involved.
September 12th, 2007 at 5:18 pm
Physics Letters B isn’t that bad…
September 12th, 2007 at 10:44 pm
Definitely not saying it is – I referee for them – but I seldom publish there.
September 13th, 2007 at 12:54 pm
JCAP or PRD?
The only problem I have with the Physical Review is that they are supposed to cover the entire field of (theoretical) physics (PRA –PRE + PRL), which includes mathematical physics. They once rejected my paper, saying that it was too specialized, suggesting that I publish it in a more specialized journal (on mathematical physics).
I ultimately did manage to get that paper published but it was repeatedly rejected from a few other journals on mathematical physics, mainly because it was written in a style more suitable for physics journals and not the more rigorous mathematical physics journals
September 13th, 2007 at 1:14 pm
Well, I think there’s bad experiences to be had all around. My first scientific paper a couple years back was published in JCAP after a huge amount of headache (which my advisor and I believed was thoroughly unwarranted). So I’m not sure it would be valid to judge an entire journal on one bad experience. The referees are, after all, individual scientists with their own opinions and biases, such that an otherwise good scientific journal may be rejected not because it lacks merit in what it attempts to do, but rather because the particular research the referee has focused on is somewhat different than the one the paper is focused on, resulting in misunderstandings that cause them to think the paper has less merit than it ought.
This is, I think, a necessary evil where scientific journals are concerned. It is much better, I think, to err on the side of rejecting papers. A good, valid scientific paper will be accepted eventually, after all. A bit of headache on the part of good scientists is, I think, a fair price to pay to ensure that bad science doesn’t attain respectability.
September 13th, 2007 at 1:35 pm
Not sure where we’ll submit, but probably PRL and PRD.
September 13th, 2007 at 5:53 pm
Jason, I agree. I guess this is the exception that proves the rule
September 15th, 2007 at 2:50 pm
Only if you think that 11-month publication delays are acceptable. This is not at all uncommon with them. These delays are totally absurd for the short papers that they publish.
By contrast, it’s fairly easy to get a full-length paper published in Physical Review D within a couple of months.
September 16th, 2007 at 12:35 pm
I’ve not experienced such long delays at all with PLB or any other Elsevier journals. I have to admit that I have only published (a few years ago) a few articles in these journals. My experience with the Physical review is also good, except for one case in which I had to transfer from PRL to PRD. The status report of that article:
September 16th, 2007 at 2:25 pm
I remember reading somewhere (must have been while reviewing something for Phys Rev!) that papers that are just 4-page-long summaries of other papers by the same authors are not supposed to be suitable for PRL.
Now we all know tons of counterexamples … so is this actually not a rule that PRL applies, is it just routinely ignored due to the overall benefit to all concerned of getting more things published in PRL, or what?
September 19th, 2007 at 9:07 am
1. Is LaTeX still as important as it use to be? I’ve heard many say, it’s OK now to send most papers in Word or PDF. Of course it depends on who wants it, I’m just looking for a scan of the patterns.
2. I’m looking for a good and cheap/free Word to LaTeX converter, that will take docs already done in Word with Equation Editor and convert to submission-grade LaTeX. If I can’t get the program, are there places that will convert. Thanks for any suggestions. Also, does Open Office latest version actually have built-in genuine LaTeX conversion?
September 19th, 2007 at 9:47 am
1. Yes. To a first approximation all theoretical physics papers are written in LaTeX.
2. Sorry, I don’t know of a converter.
September 19th, 2007 at 10:28 am
Neil, since a few years, many journals now also accept MSWord documents. Some journals require that you just send them a PDF or Postscript file.
But LaTeX is easy to learn. I wouldn’t waste any money on programs that convert documents to LaTeX. I’m not sure about the quality of the code you’ll get that way. When I export the results of symbolic Mathematica computations to TeX, I usually have to modify the LaTeX code by hand (to deal with linebreaks in equations etc.)