Earth's 'Underground Ocean': No remnant of the Flood

I suppose every good science story deserves a creative headline. Reporting the latest research on the boundary between the upper and lower mantle, however, were catchy titles like this:

• New evidence for oceans of water deep in the Earth
• Splash! Three times as much water as ALL of Earth's oceans found TRAPPED underground
• Evidence Supports Existence of Oceans of Water in Earth
• Earth's 'underground oceans' could have three times more water than the surface

Anyone that understands young-Earth creationism and how it processes scientific reports should be able to anticipate this optimistic, yet naïve response: "Well that explains where all the water from Noah's flood went. We creationists have been saying all along that the highest peaks were once covered by water!" I've seen it pop up on several occasions, despite that neither AiG nor ICR have yet made this connection [correction: Dr. Liz Mitchell of AiG has; see comments to this post]. In any case, it may have something to do with a statement reported here (last paragraph) by Dr. Stephen Jacobsen, a co-author on the paper:

"We should be grateful for this deep reservoir... If it wasn't there, it would be on the surface of the Earth, and mountain tops would be the only land poking out."

The results from Schmandt et al. (2014) are by no means trivial, and personally I am fascinated by continuous geological discoveries so far removed from direct observation (perhaps it is the geological equivalent of deep space monitoring and theoretical physics). But the creative headlines are a bit misleading as to the nature of these deep-mantle 'reservoirs' of water. As Real Clear Science reports (along with the actual text of those articles cited above), melt zones near the transition between the lower and upper mantle (~600 km below the surface) are being produced by dehydration of a mineral called ringwoodite, which is up to ~2.5% water by weight.

Sample of diamond with a tiny inclusion of ringwoodite—the first direct evidence of a deep mantle reservoir of hydrous olivine. Image from Pearson et al. (2014).

The water is present only as single oxygen and hydrogen atoms (–OH) bonded to the most common mineral in the Earth's mantle: olivine, a ferromagnesian silicate named after its characteristic color. If mantle rocks containing ringwoodite sink below the transition zone, the ringwoodite breaks down into its constituent parts: olivine and water. Despite that the concentration of water is never high enough to make droplets of liquid water, these free molecules diffuse into the lower mantle rocks, lowering their melting point. A similar reaction occurs when we spread salt over ice. Salts break down into cations and anions (like calcium and chloride), which lowers the melting point of the ice and allows it to convert from solid to liquid without raising the temperature.

Partially melted mantle rocks (not entirely liquid, because only a small percentage actually melts) are more buoyant than those surrounding, which forces them to rise toward the surface of the Earth. Within the transition zone, however, any free water molecules would simply react with olivine and convert back to ringwoodite. This metamorphism completes the cyclic process that, according to Schmandt et al. (2014), maintains a transition zone containing ringwoodite over a partially melted boundary between the upper and lower mantle. This boundary is detectable through seismic data, according to the authors.

Is there any connection to the Flood?

It has long been known that hydrous minerals like amphibole, along with marine sediments, cause dehydration melting in the upper mantle, which is a major cause of volcanism associated with subduction zones (e.g. Japan, New Zealand, the Pacific Northwest). This process occurs primarily at much shallower depths than the mantle transition zone, however, which means that relatively little surface water is subducted to depths >525 km, where these 'underground oceans' currently exist. In addition, these reservoirs and mass transfers are all part of the global water cycle. The more water that is subducted below the surface, the more volcanism returns it to the surface. Therefore, the deep reservoirs of water hypothesized by Schmandt et al. (2014) could not possibly be remnants of a surficial flood from any point in Earth history.

Besides, the subduction of lithospheric plates occurs at rates so slow, the 'Flood waters' could not have completed even 1% of their journey since ~4,500 years ago.

How should we understand Dr. Jacobsen's statement?

In saying that "we should be grateful for this deep reservoir", Dr. Jacobsen does not imply a one-way, high-capacity conveyor between water on the Earth's surface and water in the deep mantle, which could have sequestered massive oceans. He means rather that if it were not for the constant conversion of water and olivine to ringwoodite within the transition zone, this water would have been added slowly back into surficial reservoirs through volcanism. Instead, it is locked up in minerals as solid as your own jewelry, so that Earth's water content is split between liquid reservoirs in the surface and mineral reservoirs deep underground.

So, the case against Flood geology remains: there is not sufficient water on the surface of the Earth to have covered all its continents with a worldwide flood. Where did the water come from? And where did it go?