Mar 19 2012

Kepler Update: Earth-like Planets are “Extremely Rare”

Recently the Kepler Science Team released the results of the first 16 months of searching for Earth-like planets around nearby Sunlike stars. A statistical analysis by Space Daily (John Rehling, March 8,2012) suggests that Earth-like planets are likely to be “extremely rare” in the Galaxy.

According to Space Daily, “the pessmistic characteristic of these results suggest that to find earth-like worlds elsewhere, we should prepare to look hard – and quite possibily very hard for decades if not centuries.”

This is important because, to the public, the two most enticing drivers of human expansion into the Cosmos during the approaching new International Space Age are: 1) the discovery and exploration of Earth-like worlds, and 2) the prospects for extraterrestrial life, especially with intelligence.

The new Kepler data is consistent with analysis by JPL scientists Joseph Catanzarite and Michael Shao, that I discussed here, based on 4 months of Kepler data, but not with other more optimistic estimates.

Catanzarite and Shao — whose paper was accepted for publication in the Astrophysical Journal in 2011 — adopted an Earth Analog region characterized by a planet radius of 0.8 to 2 (Earth radii). The lower value corresponds to a mass of about 50% of Earth’s, the lower limit for retention of an oxygen atmosphere. The upper value is adopted by the Kepler scientists and, assuming Earth-like parameters, implies a planet with twice the surface heat flow of Earth and half Earth’s lithospheric thickness. Active plate tectonics and volcanism would be expected..

Catanzarite and Shao fit the Kepler transit data to power laws for both the planet radius and the scaled planet distance; they judge that the power laws are excellent fits to the data for distances from 0.2 AU to 0.5 AU (inside the HZ limits) and planetary radii from 2 to 4 (just larger than the EA range).

Using the power laws, the Kepler data set is then extrapolated into the Earth analog region defined above to obtain their estimate — based on the first 4 months of Kepler data — of the fraction of Sun-like stars likely to have Earth-analog planets: 2% +1.6%/-1.1%.

The addition of 12 more months of new Kepler data is summarizied in this table from Space Daily:

Bins of planet radius versus bins of planet period, out to the habitable zone (Earth is 365 days) are shown here; planet frequency estimates are the product of actual observed numbers of planets times their de-bias factor. This compensates for Kepler bias towards close-in and larger planets in the data.

The Kepler data show that planets with nearly exactly one Earth mass (0.9-1.1) are most likely at periods of 4-8 days, well inside a Mercury-like orbit (88 days), and also 6 bins inside the habitable zone. These are hot planets.

We can make this data more comparable to the radius limits of Catanzarite and Shao by summing appropriate bins and extrapolating them to the right (to the habitable zone) using Rehling’s median factor (0.72) relating a bin’s frequency to that of the bin to its left.

The new value — based on 16 months of Kepler data — for the fraction of Sun-like stars with Earth-analog planets is about 3.6%. This is very close to Catanzarite and Shao’s previous result, and suggests strongly again that Earth-like planets are rare.

This latest Kepler analysis reinforces:

1) An updated, anthropic Drake Equation, suggesting that high intelligence is rare in our Galaxy.
Click: Kepler, Watson, and Gott Point to the Rare Earth Hypothesis,


2) The proposition that — in defense of high intelligence — space colonization should be a high priority for humanity.
Click: Is Earth Unique? What this “Benchmark Moment” Means for ETs and Our Future

One response so far

One Response to “Kepler Update: Earth-like Planets are “Extremely Rare””

  1. Edon 08 May 2012 at 11:58 am

    So only 1% to 3% of sunlike stars have earth analog planets? That’s 1 in 33 to 1 in 100 — uncommon, but not vanishingly rare.

    So only one in 100 sunlike stars has one of these worlds…what’s the cube root of 27? It’s 3. what’s the cube root of 125? It’s 5.

    So earth analog planets will be 3 to 5 times as far away as the closest sunlike star. If we regard typical sunlike stars as 12 light years apart (Tau Ceti’s distance…taking Alpha Centauri being 4 light years away as being unusually close), that would mean the closest earth analog would be 36 to 60 light years away.

    Hi Ed,
    Thanks for your comment! I like your optimism.

    Keep in mind that these are still preliminary statistical values and refer only to “Earth-analog” worlds and not “Earth-like” planets — quite a different animal!

    However, if we use your numbers for a moment and assume that the leading edge of our expanding sphere of commercial radio radiation is about 80 light years away, the closest Earth-analog planets should know that we exist.

    If anyone is home within 40 light years we should have received their reply by now.

    Best regards,

Trackback URI | Comments RSS

Leave a Reply