An imagined habitable planet (Credit ESO/M. Kornmesser)

In 1964 Stephen Dole published a hundred and seventy-four page document for a US Air Force project at the RAND corporation in Santa Monica, California. With not a little hubris it was titled "Habitable Planets for Man", an extraordinarily detailed and prescient scientific study of the nature of worlds that might support life in the universe, their likelihood of existence, and how to find them. Reading it is like drinking a James Bond martini, bracing and head-spinning in equal measure.

Dole, who passed away in 2000 at the age of 83, was head of the Human Engineering group at RAND, dealing with the physiological and physical needs of humans in space. His paper, written at the height of the new space age, is an amazing record of the scientific enthusiasm and optimism of the time; space was humanity's new destiny, a natural piece of our ongoing evolution. And this document played a key role in elevating a discussion that is only now reaching fever pitch four decades later - which planets around other stars are actually suitable for life as we know it?

The definition of "habitability" that has become more or less standard amongst astronomers and astrobiologists is simply whether or not the surface environment of a planet can harbor liquid water, which is typically further simplified to whether or not the surface temperature is between 0 and 100 Celsius. It's awfully restrictive and it's awfully caveat-ridden (Is there any water at all? What about atmospheric pressures other than 1 bar? Do we mean everywhere on the planet?). But given the absence of anything more than the most rudimentary information about exoplanet masses and orbital configurations, it's not entirely unreasonable to at least begin with this in order to evaluate the possibilities.

So the big question of the moment is where do we stand so far in our quest for other "habitable" worlds? There are really several layers to the answer. The first is simply to do with the frequency with which we think the universe makes dense rocky planets like our own (again, a reasonable first port of call in looking for worlds that we might recognize as suitable for life). Between the incredible Kepler results and the most recent findings of other telescopic surveys (in particular the new HARPS results announced this week) the emerging answer appears to be "frequently". Rocky planets are almost ridiculously numerous - with the caveat that we have thus far only probed the heavier end of that population, worlds a few times the mass of the Earth.

Indeed, as Greg Laughlin discusses in an excellent post on reconciling Kepler results with the results of Doppler detections, planets less than 17 times the mass of the Earth (i.e. less than Neptune sized) and less than 50 day orbits seem to occur around 30-50% of all local Sun-like stars. That implies tens of billions of just this category of planet in our galaxy alone, and to be compatible with observations most of these may be of the dense rocky variety rather than the gassy 'Neptune-like' brand.

The second layer to the answer is just how many of these new worlds might satisfy our "habitability" criteria. Astute followers of the science will have noticed that the past couple years have seen an increasingly embarrassing plethora of headline grabbing "Earth like planet discovered" announcements. The most notorious of these was the announcement in 2010 of Gliese 281 g - as a 3 to 4 Earth mass planet orbiting a red-dwarf star smack dab in the orbital range that offered a fairly robust expectation of "habitability". With the system being only 20 light years away it got a lot of us excited, I'm still living down my "goosebumps" comment. It looked terrific, but this is tough astronomy, pushing the limits of technology. Later analysis and new data now seem to convincingly show that this planet was nothing more than a tricky ghost of the observations. Nonetheless, another world in this system - the putative Gliese 281 d, a roughly 10 Earth mass object - has also been held up as a potentially habitable place with equally competitive and unwarranted fanfare.

Recently though a number of other candidates have bubbled up. Some of Kepler's distant detections may be of super-Earth worlds (less than about ten times the mass of Earth) within the right orbital ranges of their parent stars to meet habitability criteria. But at distances of more than 500 light years they're hard to follow up with other instruments. More locally, a mere 36 light years away, the planet HD 85512 b is a roughly 3.6 Earth mass object orbiting a K-star about 70% the mass of the Sun and about 13% as luminous. Its orbit is quite small, whizzing around the star every 58 days, implying that it receives almost as much stellar radiation as Venus does in our solar system. Kaltenegger, Udry and Pepe have applied a rudimentary climate model to investigate whether there are any configurations of this planet that could make it "habitable". The bottom line is that it does seem plausible if various factors are in place, such as a cloud reflectivity that offsets the thermal input of the star and an assumed Earth-like atmospheric mix of water, carbon-dioxide and nitrogen.

It also seems that tweaking some of the factors we hold dear to our own Earth - such as atmospheric content, orbital shape, tidal energy input - can radically rearrange the habitability landscape for exoplanet systems in general. For example, a more primordial atmosphere with molecular hydrogen could be retained by rocky planets, significantly boosting the greenhouse effect to expand the ranges a liquid water harboring world could orbit in. Such an atmosphere might also provide an amazing chemical resource for hungry microbial life. Tidal flexure of a planet, especially on shorter orbits, could not only extend its geophysically active lifetime (helping with climate stabilization and chemical re-cycling) but might also boost surface temperatures.

And then there are the moons. We don't know how efficiently planets larger than Neptune form or capture satellites, but our own solar system hints at the possibility that such moons are going to be numerous. Big enough moons can hold on to atmospheres (Titan is an excellent, but chilly example), and be subject to many of the forcings that habitable planets are - from stellar input to tidal heating.

All of these potentially "habitable" places are those with temperate surface environments, but that could be the tip of the iceberg, quite literally. In our solar system it is a fairly good bet that the majority of liquid water environments (albeit water full of substances like ammonia) exist as subsurface oceans. Estimates of the nature of these dark reservoirs suggest as much as 10 to 16 times the volume of all Earth's oceans may exist off in the outer solar system. Extrapolate this to elsewhere and all bets are off.

So, given all of this where do we stand with habitable planets? I think it's fair to say that a few candidates are now in the bag, and that we should expect to find many more in the coming years - all the present evidence suggests that our galaxy is chock-a-block with rocky worlds of various flavors. Should we get excited each time a new "habitable" planet is claimed? Muted excitement is in order, if only because we are increasing the odds of eventually having targets where we can try to sniff out atmospheric contents and biosignatures. But we should also consider this as a preliminary round.

A while back in a Scientific American guest blog I tried to make an argument that we may find ourselves seeking the signs of life on planets not just from individual worlds but from the accumulated evidence of hundreds, if not thousands. I think that the way things appear to be going this may still be a key strategy. Which brings me right back to Stephen Dole's 1964 analysis. He estimates that roughly 5% of solar-type stars might harbor a habitable planet. Extrapolating this to the Milky Way galaxy as a whole he finds that there could be 600 million habitable planets in total. Within 34 light years of Earth he claims there could be 2 habitable planets and about 50 within 100 light years of the Earth. Incredibly it looks like he might be right.