(Updated*) Mega-Earth!, or Quirks, Quarks, and how to score zero on a sci-journalism check list but ace the test.

Artistic license at work, Kepler-10c

Those of us in the science journalism business, especially ones such as yours truly paid to provide criticism and bloviation on the trade’s output, recognize that we don’t really have many hard and fast rules beyond the universal ones of good citizenship: to be honest and fair, to not steal , to pursue and examine events that tend to expand knowledge and edge toward truth, and to not put an audience to sleep. Among the fuzzier indications of a well-done science yarn is that it should reflect more than one opinion, stem from a paper that passed peer review and better yet landed in a good journal, is based on more than one observation, and spells the scientists’ names correctly along with their institutions and the reason the news is in circulation at this particular moment.

Over the weekend I happen to have been sitting at lunch next to a very accomplished physicist who worries how well the public is getting news of scientific development with such nuances as clues to whether it is reliable. We talked about ways qualities of news stories are measured. I reeled off a few and told him that if there were a way to score news stories that reduces to a numerical indicator of trustworthiness, science writers would 1) Scoff at the notion one could do this meaningfully and 2) Compulsively read each collection of scores with relish and some dread. I was particularly pleased with myself for mentioning that guideposts to a good job include multiple sources to get outside opinions on the news event, reference to where and how the news entered the public conversation, sensible indication of the news’s importance, and indication whether peer review was involved. I just thought of another tip to diligence: if a lead author is a student or junior scientist, whip-smart and highly-published or not, the reporter might best also talk with and perhaps also cite a senior co-author.

It didn’t take long to get reminded of the real world. Here’s a story that does a bangup job. Yet it gets no check mark next to several standard gauges of excellence:

  •    Canadian Broadcasting Corp. (radio) Quirks & Quarks – host Bob McDonald, producer Jim Handman: Mega Earth ; The link goes to a site with a podcast listen-to button. The segment takes a little less than ten minutes. It is, as usual for this show, a worthwhile listen.

The show ran this past Saturday. Five days earlier, at the American Astronomical Society meeting in Boston, a team of 35 European and American astronomers reported that an ancient planet more than 500 light years away is 17 times as massive as Earth. But it is only about 2.3 times greater in diameter. It is about half-again as dense as Earth. Thus it would appear to be mostly rock, its core squashed far tighter than Earth can manage with its gravity field. By inference, it has little or no atmosphere. This is a puzzle for the amusement of experts. Planet formation theory has convincing explanations for how small rocky planets evolve, and for how much-larger planets with huge thick atmospheres evolve. But such a massive, rocky world with a thin atmosphere is a stumper.

Kepler space telescope’s data revealed its existence and size a few years back with recordings of the amount of light from its star that it blocks during each 45-day orbit. The team that reported the new results a week ago, using a highly sensitive spectrometer called HARPS-N in the Canary Islands, then figured out its mass. They calculated it from the Doppler shift of its host star, Kepler-10, as planets tug it this way and that (the planet in question is called Kepler-10c). The team includes big shots in astronomy and planetary sciences including Swiss astronomers Michel Mayor and Didier Queloz, whom astronomy writers will recognize as the first, nearly 20 years ago, to deduce alien planets from Doppler measurements of their stars’ wobbles.

OK then. The CBC show did not, of course, spell out names or anything for radio listeners. It had but one source, Xavier Dumusque*, first author and a post-doc at the Harvard-Smithsonian Center for Astrophysics. The program did not mention that the work is by a team of dozens, the AAS meeting in Boston, or whether the paper has been accepted in a journal (I see no sign in the preprint of it even being in a journal’s mailbox). So, you know what? So what is what. All those rules and guideposts in science writing have escape clauses.

First, in this case, when science news breaks the weekly Quirks & Quarks (like NPR’s and Ira Flatows’ Science Friday and The Science Show with Robyn Williams in Australia) is often late out of the gate. Its audience is a science-savvy stratum of the public, many of whom will have read something about most of its topics. Also, radio just doesn’t work well if it tries to recite all the formal particulars of science news, such as why it’s news now and everybody’s affiliations. Plus, the CBC website itself – as seen at the link above to the podcast – provides further links to a few other media stories on the topic, to the CfA press release, and to a pdf of the paper itself on astronomy’s and physics’s digital watering hole, the arXiv server (the release and paper are in Grist below). The radio program is a question-and-answer exercise with one prime source, much like what a print outlet may do to add depth to its standard news reports. The CBC interview works well and has many strengths, including a guest, Dumusque, who explains the science in good, enthusiastic style. The piece is responsible, honest, fair, accurate, and will put few in its audience to sleep.

       * Dr.Dumusque by the way was in the news at least once before, about a year and a half ago, with an episode of science performed near the edge of instrumental ability to observe things. He was lead author on a paper in Nature reporting possible detection of a planet in the nearby Centauri system. The conclusions about Alpha Centauri-B’s planet apparently remain in play. Astronomy and science writers generally should get a bang from a science writer’s blog on the expert conversation about the statistical analysis that the paper depended upon:

  •    ZastrowPhysics (writer Mark Zastrow): The FAP trap: FAP, by the way, means false alarm probability. This report is fun and fair.

Some other stories on Kepler-10c early last week:


  • Washington Post – Joel Achenbach: Kepler space telescope spies a ‘Mega-Earth’ ; Good job. Achenbach has one reason this planet is surprising that most other stories do not: Not only does it appear to be rocky, very massive, and nearly airless, but its star (and the planet) appear to be about 11 billion years old. Still hanging around in our own galaxy yet dating to less than 3 billion years after the Big Bang. Achenbach reports that some theorists have felt that is too ancient (presumably due to the rate at which stars explode and seed the universe with heavy elements) to expect rocky planets. He also interviewed a well known scientists not affiliated with the study, and id’d Dimitar Sasselov as a co-author of the study. Many outlets just quoted him, presumably off the press release, but didn’t say he is on the team.

Grist for the Mill: Harvard-Smithsonian Center for Astrophysics press release ; arXiv paper,

Leave a Reply

Your email address will not be published. Required fields are marked *