The first exoplanet discovered around a normal star in 1995 was anything but normal in comparison to our own solar system. 55 Pegasi b is a gas giant world orbiting every 4.23 days - placing it some eight times closer to its stellar parent than the planet Mercury is around the Sun. At least half as massive as Jupiter, this new planet wasn't just one of the first exoplanets, it was the first of a whole new class of objects - the hot Jupiters. How such a giant, gas laden world could exist orbiting so close to a star was a huge mystery. We still don't fully understand all the details, but it seems that planets like this must form much further away from their stars and then orbital evolution transports them inwards - through mechanisms like disk-migration, and the gravitational perturbations of other planets in a system.
These are extraordinary planets to us, but in a cosmic context they are far from unusual. Perhaps 3% of stars harbor giant objects like these on close stellar orbits; an enormous total in this galaxy of 200 billion suns. For our still fledgling efforts to find exoplanets they have been a boon, far easier (relatively speaking) to detect through their gravitational pull on their parent stars, and far more likely to be seen transiting across those stellar disks than smaller or more distantly orbiting objects. Hot Jupiters have also provided us with the first glimpses of extraterrestrial atmospheres, from the bright infrared glow of star-heated hot spots and storms, to some of the basic constituents - sodium atoms, hydrogen, carbon, and oxygen. They have also stunned us with the unexpected; some with vastly inflated atmospheres, some in retrograde orbits.
But one of their most counter-intuitive properties has generally escaped popular notice. Hot Jupiters tend to be extremely dark. This was brought into the spotlight (ahem) recently with a paper by Kipping and Spiegel in the Monthly Notices of the Royal Astronomical Society. The planet TrES2-b was discovered in 2006 as it transited its Sun-like star. TrES2-b is 1.2 times the mass of Jupiter and orbits at a mere 0.04 astronomical units, racing around with a year-length of 2.5 Earth-days. This system is also within the field of view of NASA's Kepler mission, allowing Kipping and Spiegel to dig into the extremely high-fidelity Kepler data to sniff for the variations in visible light as TrES2-b whizzes through its orbit.
Using this data it's possible to measure the difference in brightness between the day and night sides of the planet - from our point of view the night side is when it passes between us and the star, and the day side is glimpsed just before and after the planet zips around the backside of the star. Remarkably the Kepler data indicates a variation in the total light reaching us from this star-planet system between the day-night passage of only about seven parts per million. This translates into the discovery that the fully illuminated day side of TrES2-b is barely reflecting any fraction of visible starlight towards us at all. Not only that, but by modeling the possible composition and structure of the planet's atmosphere Kipping and Spiegel find that at least half of the light seen from the day side is actually coming from the planet's own warmth. This close to the star the outer atmosphere of TrES2-b is heated to over 1000 Celsius, more than enough for it to glow like an ember in the fireplace.
The conclusion is that seen head-on this planet reflects less than 1% of the visible light falling on it, perhaps only 0.04% in their best models. It is more efficient at capturing photons than black paint. The reason is that at these high temperatures the outer atmosphere of a giant planet may have no reflective condensates - no clouds - and the warm swirling gas is exceptionally good at absorbing incoming photons of light. By contrast, frigid Jupiter in our system is rich in reflective ammonia cloud structures and bounces back about 50% of light when seen head-on. Not all hot Jupiters may be quite as dark as TrES2-b, but we have yet to fully examine their properties and this could well be a common characteristic. This kind of measurement provides fantastic insight to the atmospheric state of worlds unlike anything in our own solar system - great warm gaseous orbs, soaking up photons.
The notion of a murky black planet is very intriguing, and got this result into the news, but what would TrES2-b look like if we could see it with our own eyes? This is where a bit of a reality check and perspective is useful. The starlight hitting this planet is roughly 700 times brighter than the light we receive on Earth from the Sun. So if we were sitting in orbit around TrES2-b even 1% of that light reflected towards us would still look like an awful lot to our puny eyes. The fact is that this close to the star then the sheer flood of photons will make all reflecting objects look very bright to us, even if they are darker than charcoal. What if we were instead sitting out at a respectable 1 astronomical unit (the equivalent to Earth's orbit)? We can use the planet Venus as a yardstick. Venus reflects an incredible 75% of visible light, but is about 20 times further from the Sun than TrES2-b is from its star, and is 15 times smaller in radius than TrES2-b. So how bright would TrES2-b appear to us compared to Venus in full phase?
I've brushed a few things under the carpet, but very roughly the reflected flux of photons we'd see from TrES2-b would still be about 3,000 times more than that from Venus. Since the human eye has an approximately logarithmic response to brightness this means that despite the very low reflectivity of TrES2-b it would still appear to shine as much as 3 times stronger than Venus in our sky - assuming of course you could ever separate it from the glare of the Sun given its tiny orbit. So the funny thing is that although this planet is so awfully good at absorbing light, because of its location it would not actually appear to be pitch black if we could look at it up close and personal.
Hot Jupiters can indeed be very dark, but what you end up seeing is all relative.