It was recently announced that the Rosetta space mission to the comet 67P/Churyumov-Gerasimenko found that water contained by the comet did not resemble water on Earth. This has added to evidence that comets probably did not deliver the oceans to our planet billions of years ago when it first formed. In that case, where did all the water come from and how can we be so sure about where it did not come from? Interestingly, evidence is beginning to emerge that suggests the water may have been with us all along, deep below the surface.
Comets and asteroids have long been thought of as the key to the origin of Earth's oceans. When Earth first formed out of the disc of dust and gas surrounding the young Sun, the inner part of the disc would have been heated to extremely high temperatures (~1000 K, or about 700°C); hence, any water present would have vapourised. The only materials that could exist in a solid form (and therefore act as building blocks for the inner planets: Mercury, Venus, Earth and Mars) were metals, such as iron. In the outer part of the disc, temperatures were low enough for water ice to exist. Hence, for water to have reached the inner planets, it must have come from the outer solar system. This is the origin of the theory for ocean delivery via comet or asteroid strikes. However, there is a problem with this story.
The case for water delivery via asteroids is more compelling. Studies of meteorites originating from the asteroid belt show that these tend to contain water that is much closer (but not a direct match) to Earth's water composition. Asteroids contain much less water than comets, but a sustained asteroid bombardment of the early Earth could have provided the water we see around us today. Given the variation in deuterium abundances in different asteroids, the lack of a direct match could be explained by the mixing of water from various asteroids as they collided with Earth. However, while asteroids may well have played a part in the origin of Earth's water, there is another possibility emerging from Earth's inner depths.
Earlier this year, evidence for the existence of three ocean's worth of water was found in the mantle, about 700 km below the surface. This evidence came from the seismic waves generated by earthquakes across the globe. These waves travel through the planet at different speeds, depending on the type of material they pass through. Therefore, by measuring the speeds of seismic waves at different depths, scientists were able to explore the composition of Earth's deep interior. They could also infer the presence of water, because the waves take longer to pass through wet rock than dry rock. Just such a slowing down was observed at a depth of 700 km. The wave speed corresponded to the presence of a hydrous (water-containing) form of the mineral ringwoodite. It is not yet known if just an isolated pocket of this water-containing rock exists, or if this hydrated belt encircles the globe. If it does, this means that there is a lot more water on Earth than what we see covering the surface, which lends evidence to a third ocean origin theory. Rather than being delivered from elsewhere in the young solar system, some, or all, of Earth's water might have been there from the beginning after all, potentially forming via reactions between hydrogen and oxygen in the depths of the young Earth, then being bound up in hydrous minerals. Over time, as the Earth formed and cooled, high temperatures and pressures at depth could have "squeezed" the water out of the rock, allowing it to reach the surface via plate tectonics, steadily forming oceans on the surface. More recently, another group has suggested that the mineral garnet could act as a water carrier in the mantle, keeping some water trapped within the mantle for geologically long periods of time. However, they estimate that the mantle contains only another Pacific Ocean's worth of water, rather than three times Earth's surface water volume.
For now, the source of Earth's water remains a mystery. We have some ideas, but nothing concrete yet. From an astrobiological perspective this is a fascinating mystery, because water is such an essential component for life on Earth and is key to several origin-of-life-theories. Knowing exactly how and when liquid water first appeared on Earth would tell us more about the conditions that gave rise to life, and the conditions that could have given rise to life elsewhere in the solar system and beyond. Additionally, this information could tell us more about how the habitability of our planet might change in the future. In about a billion years from now, a brighter Sun is expected to kickstart a process of rapid ocean evaporation. How would having three more oceans locked away in the mantle change the rate at which water is lost? Without any water the planet would be uninhabitable. So, if there is more water down there, and if it is able to find its way to the surface, would that push back the date for the ultimate extinction of all life on Earth?