trickier it's all got. Part of the problem is that the Earth must have formed in a region of the proto-planetary disk surrounding our baby star well within the so-called 'snow line'. Picture a thick glop of gas and dust shaped a bit like a squashed donut in orbit around a proto-sun. It's a big structure, extending out to perhaps 100 astronomical units where it thins down to almost nothing. The increasingly hot proto-sun in the middle heats up material around it, but the great donut is a pretty effective sunshade. The gas is also pressing ever inwards and as it compresses it warms up. The end result is that in the central plane, a horizontal slice through the chunky disk, it is warm towards the proto-sun and cooler and cooler the further out you go.
Somewhere, perhaps at around 4 astronomical units out, the temperature drops below 170 Kelvin (-103 Celsius). For water this represents a transition. It's the temperature at which water molecules in a vacuum will begin to stick together at an exponential rate. It's water's sublimation point. In the proto-planetary disk then water pours out of the gas phase, freezing into solids and making the snow line. Little wonder that the giant planets and moons have such an enormous water content. For little Earth, forming around 1 astronomical unit there's nary a drop in sight.
Clearly Earth had its thirst quenched somehow. The precise nature of when and how it acquired water-bearing material has long been a puzzle. An intriguing new result in Nature Geoscience by Greenwood et al, along with a nice review by F. Robert seems to provide an enormous and juicy clue. Remarkably this comes not from some fancy new mission or astronomical observation but from applying new geochemical analysis to rocks brought back by the Apollo astronauts. Ion micro probe measurements of the mineral apatite reveal not only that the Moon has a lot more water than we thought ('lot' being a relative term), that it played a role in the Moon's early geophysical life, and that most of it seemingly has a different origin to terrestrial water.
Deuterium, the heavier isotope of hydrogen, is twice as abundant in lunar water than water on Earth. The ratio of deuterium to hydrogen has long been a fingerprint of where compounds, including water, come from in the solar system. More deuterium indicates a colder chemical origin. Cometary water has a lot more deuterium than water in meteorites. Earth's oceans look a lot more like water from volatile rich meteorites and asteroids than they do the chilly water from comets. But the Moon...Well, the Moon turns out to have water that is more consistent with a cometary origin. Pounding the Moon with comets shortly after it formed could do the trick. However, this leaves a problem. Up to this point a best bet for most of Earth's water had been the deposition of material by carbonaceous chondrite type rocks, sometime following the formation of the Moon. Either the Earth dodged the thousands or millions of comets that painted the Moon or the Moon dodged the water bearing rocks that buried the Earth.
What a predicament. How can two bodies so intimately linked, one formed from the detritus of a great collision with the other, avoid having the same cocktail flavor? One possible solution, articulated by Robert, is that both Moon and Earth got moistened by comets first. Then Earth got hit by perhaps one great object that sailed past the Moon and splatted the Earth with water containing far less deuterium - diluting the terrestrial mix.
While the discussion is not over it does raise a point of acute astrobiological interest. If this scenario is correct then that final event must have provided a very significant fraction of the Earth's water. Therefore we owe much of the past 4 billion years of gloriously damp climate to that singular moment. What if it had never happened? How habitable would Earth have been? Despite our models of forming planets that seem to indicate gaining water is quite common, there is still tremendous chance involved. Oases may indeed be hit or miss.