One in Five Sun-like stars have Earth-like Planets

Dr Andrew Howard explains how their observations suggest that up to one in five stars like our Sun may have Earth-like planets orbiting them. We find out why this doesn't...
31 October 2010

Interview with 

Dr Andrew Howard, University of California, Berkely


"The Blue Marble" is a famous photograph of the Earth taken on December 7, 1972, by the crew of the Apollo 17 spacecraft en route to the Moon at a distance of about 29,000 kilometres (18,000 mi). It shows Africa, Antarctica, and the Arabian Peninsula.


Chris -   Also this week, we have heard good news that we may not be alone or at least, not so lonely in the universe as we maybe first thought because researchers in America have announced that as many as 23% of stars like our Sun could have Earth-sized planets orbiting around them.  To explain how they came to this figure and what it means for the search for other worlds a bit like our own, we're joined now by Dr. Andrew Howard who is at the University of California at Berkeley.  Hello, Andrew.

Andrew -   Hello.

Chris -   Andrew, tell us first of all, how did you actually do this study?

The W. M. Keck Observatory, atop Mauna Kea, HawaiAndrew -   So we used the Keck telescope in Hawaii and essentially what we did was conduct a census of the planets orbiting the stars nearest to our own solar system.  We looked at each one of the 166 stars night after night for almost 5 years.  We didn't look at them every single night but we were able to detect the most massive planets - those planets that are even bigger than Jupiter- as well as less-massive planets - those planets that were down to a mass of about 3 times the mass of the Earth.  So we covered this enormous mass range from 3 Earth masses, all the way up to a thousand Earth masses.  We can't quite detect the Earths, which we'd really like to, but we can extrapolate their numbers based on a really clear trend that we see from having relatively few planets.  At the high mass end, we see only about 1 or 2 Jupiters per 100 stars in our survey.  Down at the low mass end, we see a lot more planets in the 3 to 10 Earth mass range.  We see something like 12 planets per 100 stars.  And so, if you just extrapolate this trend that we see over a very large mass rates down to one Earth mass planets, we estimate that about 23% of stars like the sun have a relatively close-in planet like the Earth.

Chris -   That's a big number, isn't it?

Andrew -   Twenty three is a really big number.  This number has sometimes been called Eta of Earth for the Greek character Eta.  This is a term that's appeared in Frank Drake's equation that's used to estimate the prospects for SETI.  And Eta of Earth had remained kind of a mystery.  People always wanted to know, are Earth-like planets common or are they rare and some people estimated that it was 100%.  Some people estimated that it was one in a million.  We say 23% or about 1 in 4 and to be fair, we had a little bit of an extrapolation here, and I wouldn't be surprised if the true number is 1 in 2 or maybe in 1 in 8, but it's not 1 in 100, and that's a really big improvement in our knowledge.

Chris -   If we could visit some of these distant worlds, what would they be like?  What are the prospects that they're in the same position, relative to their star, that our Earth is and therefore, have the kinds of conditions that this planet does - in other words, it's not too hot, not too cold, we've got liquid water?

Andrew -   The technique we used is the Doppler method, sometimes called the Wobble method, and it turns out that we're most sensitive to planets that are really close to their host stars.  So we were only able to detect planets out to a quarter of the Earth-sun distance.  So at that really close orbital distance which is, if you put the planets inside the solar system, they'd be inside the orbit of Mercury.  These are really hot planets.  I really seriously doubt that there's life on these specific planets, but what they tell us is that nature makes small planets commonly and deposits them at close orbital distances, and it seems reasonable to suppose that if there are planets at close orbital distances, there are probably planets a little farther out where the radiation from the Sun isn't so intense and these planets might be in the so-called Goldilocks zone or the habitable zone of planets where the temperature is not too hot, not to cold.  It's just right for liquid water and perhaps, for life to evolve.

Chris -   How does this finding sit with our understanding of how systems like our own do put themselves together and evolve in the first place?

earthAndrew -   It's really interesting that these so-called theories of planet formation, there's this dominant paradigm which is called the core accretion model.  This is a model that's been long-standing, and it predicts that most planets in all solar systems are born in the cool outer reaches far from the host star.  And the main reason is because, to efficiently make the core of a planet, you need ice and to have ice, you have to be in a cooler part of the solar system, that it's far from the host star.  So this model predicts that most planets are born out there.  Some of them grow up to be quite large, the size of Saturns and Jupiters, and some do not.  Some are a whole range of sizes - from smaller than the Earth to Earth-size to a little bit bigger, but the theory also predicts that it's basically only the largest planets, the gas giants like Saturn and Jupiter that migrate close to their host star.  So this theory made a prediction for our observations and they predicted that we would only detect large close-in planets, and we shouldn't detect very many small close-in planets.  And they actually had a name for this.  The theorists called this The Planet Desert.  They predicted there'd be almost no close-in small planets.

Chris -   But instead you've detected an oasis there, haven't you?  You've got a whole lot of them.  One in four of them have got small planets.

Andrew -   It's a tropical rainforest.  It's the opposite of a desert.  So, I think the whole theory of planet formation by core accretion, I don't think it needs to be thrown out, but I think that in particular, this migration part of it is not right and it needs to be revised.


Add a comment