I hope everyone is having a good break. I have a bit of free time, so thought I would mention a paper I’ve just finished. It’s about the the three super-Earths that orbit a star known as GJ 9827. The system was already known to host 3 planets with radii slightly above that of the Earth, one of which was presented as one of the most massive, and densest, known super-Earths (it’s not).
Our paper was based on stellar radial velocity measurements from the HARPS-N spectrograph. If you measure radial velocities of the host star, then you can use these to infer the masses of the orbiting planets. We also had an accurate parallax measurement from Gaia, which allowed us to determine quite a precise mass and radius for the host star. In addition, after we’d submitted the paper, another paper appeared with a whole lot of extra radial velocity measurements, so we redid our analysis with all their data included too.
The basic result is illustrated on the right, which shows the radial velocity curves for planets GJ 9827 b and GJ 9827 d. I haven’t shown the result for GJ 9827 c because it’s essentially a non-detection (we do know that there must be a third planet, but we can’t really detect how it influences the radial velocity of the star).
We find that GJ 9827 b (with a radius of 1.6 Earth radii) has a mass of 4.9 ± 0.5 Earth masses, and GJ 9827 d (with a radius of 2 Earth radii) has a mass of 4 ± 0.8 Earth masses. The mass of GJ 9827 b is quite a bit lower than the first estimate, which suggested that it was one of the most massive, and densest, super-Earths. On the other hand, we find that GJ 9827 d is probably slightly more massive, and denser, than an earlier estimate.
If you have an estimate of the mass and the radius, then you can say something about a planet’s internal composition. The figure on the left shows the GJ 9827 planets on a mass radius diagram. What it shows is that GJ 9827 b has an internal composition similar to that of the Earth and Venus. It does, however, have an orbital period of only 3.6 days, so it is far too hot to be potentially habitable. GJ 9827 d has a much lower density than GJ 9827 b and clearly retains some kind of volatile atmosphere (maybe water vapour, or hydrogen/helium).
As you can almost see in the figure above, there’s a suggestion of a radius gap within the exoplanet population; small exoplanets tend to hve radii less than about 1.5 Earth radii and are rocky, or have radii above 2 Earth radii and retain a volatile envelope/atmosphere. What makes this system particularly interesting is that it hosts a planet on either side of this radius gap. What’s more, the smaller one is more heavily irradiated than the larger one. This is consistent with stellar irradiation stripping the atmospheres from very close-in, small exoplanets.
Systems like GJ 9827 will, therefore, play an important role in understanding the origin, and evolution, of these super-Earth exoplanets. Do they form in situ, or do they form further out and then migrate inwards? Do the really small ones form rocky, or did they once retain substantial atmospheres that were then stripped by irradiation from the central star? Understanding the latter is important if we want say something about the frequency of actual Earth-like exoplanets. If the known small, rocky exoplanets originally retained substantial atmospheres, that have since been stripped, then true Earth analogues are probably rarer than if these planets are primordially rocky.
Addendum:
There was something I wanted to add to this, but that doesn’t really fit within the post. It seems that many of the truly contrarian positions with respect to climate science are often promoted by scientists who are now retired (e.g., here). It always strikes me as quite arrogant to imagine that you can see something obvious that is being missed by the next generation of scientists/researchers; do these people think that the next generation are really this incompetent, or biased?
The work I discussed above is part of a programme of research involving a team of scientists from around the world. My contribution has mostly been theoretical (I run some simulations and help to write parts of some of the papers). I led the above because we were short of people to champion some of the targets. The work was, however, well outside my immediate area of expertise and it required a lot of help from others in the science team (and a steep learning curve from myself).
Most of those who were providing direct help were people at an earlier stage of their careers than I am. What struck me was just how good they are, both in terms of their technical expertise and their thinking about how to present our results. I’ve often said that I would hate to be trying to build a career in science now, because early career researchers today seem so much better than I was at their stage. The work on this paper only re-inforced that view.
What I’m getting at is that the idea that a small number of retired scientists can see major, obvious, problems in climate science that are somehow being missed by active researchers today is just utterly bizarre. Even more so, when you consider some of the ridiculous ideas that they promote.
...and Then There's Physics 
Excellent post, and completely agree with the addendum – there are a lot of very bright people out there, and it is hard enough to stay near the state of the art in your own field, so the chances of overturning the fundamentals of another are vanishingly small. If you are going to try, then at the very least you need to take the argumentsof those telling you why you are wrong very seriously indeed!
ATTP, presume this can only be done for a star in our galaxy and possibly only for those in the spiral arm near to us.
Well done.
Could you explain to me why dark matter seems to have no effect on us or our near neighbours and our calculations but somehow affects all other galaxies we can see.
If this statement is correct of course.
As to the other, these young whipper snappers, we can show ‘em a thing or two.?
angech,
Yes, we can only do this for stars that are relatively close. This system if 30 pc (parsecs) from us. The distance to the centre of our galaxy is 8000 pc.
The reason dark matter doesn’t impact us and our neighbours is that its mass only really dominates on large scales. On the scale of the Solar System, the dark matter mass is low enough that the dominant mass is the Sun, which then largely determines the orbits of the planets.
” I’ve often said that I would hate to be trying to build a career in science now, because early career researchers today seem so much better than I was at their stage.”
yup
>There was something I wanted to add to this, but that doesn’t really fit within the post. It seems that many of the truly contrarian positions with respect to climate science are often promoted by scientists who are now retired
Isn’t the same true of retired politicians? something to do with no longer having to toe the party line?
bill,
No.
‘As to the other, these young whipper snappers, we can show ‘em a thing or two.?’
How to make a fool of ourselves with our hubris? ;o)
Great post.
A bit of serendipity since I am currently reading a book by Karel Schrijver, “One of Ten Billion Earths”. In the book there are detailed explanations of how exoplanets are discovered and how planetary systems may be formed. Excellent layman’s guide to astrophysics. It also included some sobering observations about the the odds of discovering other lifeforms in our solar system and the probable extinction of our own species due to mismanaging our own planet’s resources.
Regarding your addendum (and honestly meaning no snark and no offense), I would like to take this opportunity to repeat my sharp criticism of one of your defenses of a paper you co-authored with Bart Verheggen and John Cook. In it, you highlighted the differences in confidence levels regarding anthropogenic climate change between those with higher numbers of publications and those with lower.
At the time I attempted to point out to you that younger scientists in all probability benefited from improvements in pedagogical techniques, an advanced knowledge base and other factors. You very much seemed to take the other side of the argument at Bart’s blog. I am very happy to see you champion the cause of younger scientists today. I wish we could go back in time and re-evaluate the discussion we had at Bart’s using your addendum as a re-starting point.
Tom,
That is a rather uncharitable interpretation of what I said. I didn’t mention career stage and was referring only to groups of researchers, not to individuals. Here, for example, is a representative comment.
and here
Tom,
I’ve posted your comment and responded, but given that I’d really rather not have another food fight about consensus studies, maybe we can simply leave it at that.
That’s fine with me, ATTP. Happy holidays.
Well the addendum is only partly right in my view. In any simulation field, the young tend to be optimistic and to have limited experience which biases their view of how skillful models are. For example, in a CFD graduate level course, most assignments involving computation will use easy test cases for which the answer is pretty well defined and the instructor can judge easily if the student got that right answer and why or why not. Only a really good instructor will delve into things like transition to turbulence.
It is true that younger scientists tend to be a lot better with the new and powerful aides to computation, such as the R statistical package, visualization tools, and things like MATLAB which is quite powerful these days. They are also more adept at using the new generation of numerical analysis packages for example for numerical optimization. There can be very powerful in the right hands.
However, younger scientists sometimes lack theoretical mathematical or statistical understanding and may have trouble correctly interpreting results from those powerful tools. They also tend to have a very binary view of the universe dividing it into “right” results and “wrong” results.
In general, in my experience, more experienced modelers and developers (particularly among those with rigorous mathematical training) tend to have a more rigorous scientific attitude. However I have found turbulence modelers to be by far the most honest and reliable in their opinions. They also tend to be older and to be life long learners. Those who run codes and generate results make their careers by getting “good” results. It is natural to develop a faith that if the code is run right, the answer will be right. There is also a natural selection by promotion practices of businesses and universities for those who hold this view.
Tom,
You too.
dpy,
My suggestion isn’t that early career people have nothing left to learn, or that more senior researchers don’t have valuable insights, or experiences. It was more the idea that early career researchers are missing things that seem obvious to a small group of, mostly, retired researchers. This seems an utterly bizarre implication.
Just to understand the text right, a “rocky planet” is one without a significant atmosphere? So the Earth would not qualify as a rocky planet?
Yes, I was wondering about that too.
Victor,
No, the Earth would be rocky in this context. In this context, we’re talking about atmospheres that substantially change the apparent radius (i.e., as determined by how much of its host star’s light that it blocks) and, hence, change the bulk density.
The idea that R and Matlab are new aides to computation seems somewhat ironic to me. ;o)
In my own field, early career researchers only really differ from mid- and late career researchers in the number of papers read rather than written, which means the are sometimes less aware of wheels that have been repeatedly invented already. Differences in optimism, mathematical rigour, binary interpretation etc. are more likely down to individual variability in human nature. However, that might be confirmation bias.
> That might be confirmation bias.
My own is that DavidY found a way to allude to himself first and foremost.
I would love to go take a closer look at these planets, but I don’t fly because of the CO2 footprint issue with flying, so I guess I will just have let these destinations slip away along quite a few others. I am sort of landlocked in a radius of 500 miles around Portland/Seattle. Luckily for me, there is a lot of beautiful planet to explore in that limited range.
Also, wrt to the hubris of the retired scientist: maybe a little loss of ego and status in retirement makes some “experts” yearn for a return to the spotlight? What better way to earn some spotlight time, than to say that mainstream, active scientists have missed something that a retired expert now sees so clearly. It seems to me there is an easy solution for the retired expert: publish in peer-reviewed forum. I don’t think there is any reason a retired scientists cannot publish, but it may be hard to gather the data to support some arcane hunch.
Happy, etc. to all here at ATTP. May we all do well and good in the coming cycle around my favorite star.
Mike
Of course R and Matlab are not new, but they have gotten very powerful over the last 10-15 years as new capabilities, scripts, etc have been developed. They are now almost comprehensive programming languages in which complex simulation and analysis codes can be written.
I’ve been fortunate to work with quite a few senior (60-90) year old mathematicians and scientists and uniformly they are fantastic both technically and personally. There is selection involved here. Those who survive a 40 year career and still are strongly motivated by science and mathematics tend to be the best of the crop (certainly smarter than I am).
My experience here ATTP is that when a group of senior scientists who have shown a long distinguished career and are still active say that something needs further work, they are usually right. Particularly with regard to model flaws experience leads to superior knowledge of the importance (and large number) of negative results that often go unreported.
[Peddling. -W]
dpy,
You are rather twisting this whole issue somewhat. My intent wasn’t to start a discussion about the relative merits of old versus young scientists (technically, I was also trying to use early career, rather than young). Of course there are many senior, in some cases, retired researchers who have important insights into a topic. Similarly, there will be early career researchers who lack the experience to necessarily appreciate all of the subtleties of a topic. However, given my experience, I would be incredibly surprised if a retired researcher who takes a rather contrarian stance on an important, and well-studied, topic has noticed something obvious that has not been noticed by those who are still active researchers. This is especially true if you consider that a great deal of what some of these people do promote is clearly nonsense.
Thanks, got is now. Even if they are hard to see it must be so nice to suddenly have a huge sample of planets to study. For me the detection of exoplanets came completely out of the blue, I had never expected that to be possible. Did you see it coming that that would one day be possible?
Freedom is just another word of nothing left to lose. Once you go with pensions you no longer have to care whether what you claim makes sense. If you do not have or no longer have an intrinsic motivation to do good science you can make any claim to fool the public.
I wonder how science was before 1900, when it was mostly done by independently wealthy people and not by professionals. They would have the freedom to ignore all quality standards, on the other hand if you do not get paid why waste your time on science if you are not intrinsically motivated. Was science much hindered by crackpots or were they rare?
Victor,
Not specifically. I only really got involved in exoplanet research a few years after the first one was discovered.
DPY MATLAB has been a mature programming environment for a long time, as have the add-ons like simulink. Personally, I tend to write my own tools, rather than use the toolboxes, some of which are good, others rather less so.
Like Prof Salby, or Essenhigh, or Harde or Dyson etc.? In my experience, they tend not to be right if they think they know what is lacking in some other field of research, rather than their own.
“If an elderly but distinguished scientist says that something is possible, he is almost certainly right; but if he says that it is impossible, he is very probably wrong.” – Arthur C. Clarke
Well, ATTP, a lot of what passes for discussion in politicized fields in my opinion relies on selecting what you want to highlight and not looking too carefully elsewhere.
A lot of what senior skeptics have said it seems to me is legitimate and based on experience. I think Dyson is focused on our ignorance of the carbon cycle. Happer questions whether increasing CO2 is a net negative. Lindzen criticized the climate models as do most of the senior skeptics. On this latter point they have a good point. There is a huge literature on nonlinear dynamical systems and fluid dynamics that goes back 50 years lending this critique a lot of credence. You probably have a list of questionable things these people have said. I’m more interested in the substantive and credible points being made.
My point is just that a lot of the less deep work about “running” codes is deeply biased and suffers from huge positive biases. Senior people are much less likely to fall for the “after I tuned the parameters, the grid, and adjusted the data, the answer was right” mode of argument. They will want to see if the agreement is robust for many cases and not just a tuning exercise. There is a growing body of literature on structural instability showing that the “attractor will fix it” ideas are wrong too.
“My own is that DavidY found a way to allude to himself first and foremost.”
Yes the whole thing was rather funny and unselfaware, in a bad way not a good way.
DPY “I think Dyson is focused on our ignorance of the carbon cycle.”
So precisely what ignorance about the carbon cycle justifies Dyson’s skeptical stance?
dpy,
Those senior scientists are essentially the people I’m talking about. You may want to reflect on you regarding them as making some good points?
Np bill, scientists don’t, never have and never will toe the party line. Teachers and lecturers maybe, teaching to the textbook, but for a working research scientist that’s a sure path to obscurity.
Are you sure that’s not just him expressing his own ignorance dpy? What with him not being a biologist, biochemist, oceanographer, isotope expert or marine geochemist? That’s a decade of undergraduate study on its own, even without getting into research. Is that really how he spent his retirement?
Interesting. So he’s never heard of Ice Ages. Maybe he should start with something simple, like Ice Age: Dawn of the Dinosaurs.
Lindzen criticized the climate models as do most of the senior skeptics.
How long did it take for the Iris Effect to be disproved by observations? Two years? Three? Probably less because they needed time to analyses the observations and get the papers published.
dpy
In my experience, the exact opposite is the case.
It’s absolutely astonishing what transparently wrong complete nonsense “senior sceptics” are prepared to push.
Here’s Judith Curry quote mining from the IPPC to defend a post published claiming 50% of the CO2 rise could be “natural”.
Amazing stuff which opened my eyes to how senior scientists are prepared to defend the indefensible.
Note the worst thing Prof. Curry wrote on that thread:
Yes, that’s right, conservation of mass is dismissed as “naive zeroth order”. And again here:
Of course DPY will selectively ignore evidence of Prof. Curry making ridiculous scientific arguments.
Rather good example of a senior figure calling for more work to be done on something outside their immediate field of expertise, whilst at the same time demonstrating a complete lack of understanding of the most basic concepts of the field they wish to criticize.
> Amazing stuff which opened my eyes to how senior scientists are prepared to defend the indefensible.
First four comments:
[Centinel2012] Interesting analysis appears to be real science! So that means no one will consider it.
[DannyT] Thank you for your work and for sharing. Would you be willing to offer a bio? This might be of value in addressing Centinel2012’s concern.
[Judy] From Fred’s LinkedIn profile: environmental engineer, retired, formerly with the US EPA ASRL, research expertise on the effects of air pollutants.
[grog] This is an interesting article that will need time to digest. Maybe the EPA background explains why our engineer freind thing running averages and running diffs are a low-pass filter of choice.
In fairness, FerdinandE, who wasn’t impressed, it also retired. His luckwarm concerns regarding mainstream science lie elsewhere.
***
FWIW, if we’re to indulge DavidY’s peddling, I duly submit that:
(1) We are on the Internet and we have time on our hands.
(2) The Internet attracts both the young and the old.
(3) Both contrarians and libertarians are over represented.
(4) The most immediate putative reference for an experienced modeler with rigorous mathematical training and more rigorous scientific attitude is DavidY himself.
I find that the main advantage of a senior scientist is knowledge of the 90% that never gets published because of a lack of time or lack of sexiness. There is a reservoir of observations that were made and not published and a reservoir of observations that were meant to be made until the funding ran out. Early career scientists are more able to make use of the new tools and ideally the older scientist can point to interesting problems.
An older scientist out of his field has no such advantage, hence the emeritus effect.
Talking about exoplanets. ATTP, what do you think about this video about rogue planets with live on them?
This YouTube channel is a good example that science is more fascinating than people just making stuff up.
Perhaps not so uncommon in astrophysics where amateurs contribute and are typically taken very seriously by other senior scientists in the field?
Examples from the past year:
1. Supernova discovery by amateur Victor Buso
2. New type of aurora discovered by an amateur group known as the Alberta Aurora Chasers
3. Asteroid detected passing earth within moon’s orbit by amateur Michael Jäger
4. Seemingly lost NASA satellite redetected by amateur Scott Tilley
Victor,
The idea that there are rogue planets is entirely reasonable. It’s pretty clear that planetary systems can be pretty dynamic and that some planets will be ejected into space. The discussion on the possibility of life seems plausible, but I don’t think we know enough to really know if anything like this is actually likely.
Good musicianship, good history, deserves another run in context!
What Mitch said
Graduate training is designed to pass lore from advisers to students. You learn much about things that didn’t work and therefore were never published [hey Prof. I have a great idea!…Well actually son, we did that back in 06 and wasted two years on it], whose papers to trust, and which to be suspicious of [Hey Prof. here’s a great new paper!… Son, don’t trust that clown.] In short the kind of local knowledge that allows one to cut through the published literature thicket. http://rabett.blogspot.com/2006/10/amateur-night.html
Of course if you are out of field you should talk to your colleagues in field before stepping in it
Izen, that video made my day. I’ll have to fix it up on speakers and play it for dad (PW); he’ll be amused. Thanks!
(And people think science is boring or incomprehensible.)
Eli,
Thanks, I’d forgotten about that post of yours. It does provide a pretty good description of how people can fall into traps if they’re not aware of the history, and background, of a topic.
Thank you, izen.
Rogue planets
Maybe “lonely planets” is a better term.
Recently I saw an announcement, maybe Astronomy Now, maybe Phys.org, of a sighting of a lonely planet. Decades ago I read the summary of a computer study of the stability of the solar system in which Mercury picked up and wandered off. By the way, the summary stated that Venus and Terra stabilize each other, comforting even if I don’t understand it.
David,
I tend to refer to them as free-floating planets. The lonely planet mentioned in that article might have been PSO J318.5-22.
Earlier this year, there was this announcement:
http://www.astronomerstelegram.org/?read=11448
Followed by this:
http://www.astronomerstelegram.org/?read=11449
Perhaps we need to buy a planetary drive from the Outsiders?
Just don’t try to build your own (spoiler alert for the Fleet of Worlds series).
Pingback: 2018: A year in review | …and Then There's Physics
Pingback: An attempt to do a Bayesian estimate of climate sensitivity | …and Then There's Physics