A year after the dramatic draining and collapse of Kilauea's summit crater and lava lake, Halema'uma'u is once again beginning to fill. It happened rather suddenly: an overflight in mid-July didn't spot anything out of the ordinary. But just a week later, during a lidar survey on July 25th, a hawk-eyed helicopter pilot spotted a wee green patch at the very bottom of the crater that hadn't been there before.
Any change at an active volcano is exciting (and potentially alarming). Could it be some sort of chemical deposit? Optical illusion? Water? Most observers flying over in the following days thought water, but there were no ripples or reflections visible yet. And it wasn't like volcanologists could just stroll down to the crater floor for a look. Not only would they have to contend with several hundred meters of extremely steep, unstable loose rock, but also frequent rockfalls, and possibly a fatal accumulation of carbon dioxide if they made it that far. And sending in a drone wouldn't be feasible yet: with difficult angles, unpredictable air currents, local cultural considerations, and national park regulations, setting up a drone flight would take time.
So confirmation had to wait for August 1st, when HVO scientists conducting a helicopter overflight spotted reflections on the green patch. It was a baby pond! And it was growing!
So, why is a small patch of water so exciting? So many reasons!
We may have just witnessed the birth of a crater lake. That's always awesome, but Kilauea hasn't had one made of water in written memory, and definitely never while the Hawaiian Volcano Observatory has been there. So we'll get to watch a Hawaiian crater lake from birth to, very probably, death.
And that death, when it comes, may be quite explosive! That's partly why HVO scientists are keenly interested in whether the pond is a result of rising groundwater (most likely) or rainwater. The total water depth, including subsurface water, matters greatly when it comes to hazards from magma rising up the conduit. Shallow water, like a rainwater pond, would simply boil away without having much effect. The story could be quite different with a lake of groundwater, considering the surface depth could reach up to 70 meters (230 feet), and the saturated ground of the water table could be tens of meters deeper still.
USGS scientists at HVO predict two very different scenarios for such a sizable crater lake encountering rising magma. If the ascent is slow, the water will merely evaporate from the gradually intensifying heat, and the water lake will be replaced by lava flows - or perhaps a new lava lake (which is the scenario I'm rooting for).
Something rather more dramatic will happen if the magma rises quickly, which a volatile-rich melt propelled by expanding gas bubbles does. Lava that would, in drier conditions, provide us with magnificent fire fountains, would instead find its fragments* transferring heat to the water so rapidly that the lake would boil. Much of it could flash to steam, powering explosions that would be extremely dangerous close to the crater, and send enough ash into the air to cause a considerable bother downwind.
We know from glassy lava fragments in deposits around 400 years old that Kilauea had a crater lake and rapid quenching of lava fountain bits before. The water actually freezes the lava so quickly that the amounts of dissolved volatiles within it, like water and sulfur, are right between the amounts you find in un-erupted magma and the stuff in regular lava fountains. Neat!
So, why is this happening now, rather than right after the lava lake drained and the crater collapsed below the level of the local water table? It's not because there was a big hole like a drain left behind - as soon as the magma drained, rubble from the collapse filled the conduit. Prior to that, the incredibly hot lava lake and its magma-filled conduit had done a fine job keeping water away from the summit, and the collapse seems to have drawn the water table down. Even after the molten stuff drained, the surrounding rocks were hot! They've only just now cooled to the point where water could remain liquid there. And as the rocks cool further, the water will continue to rise, likely eventually reaching equilibrium with the surrounding water table. We will hopefully have a respectable crater lake in the end.
As the pond grows, it will have plenty of information to give us about the structure of the ground beneath Kilauea's surface. The water temperature, currently a toasty 70°C (158°F), tells us there's still a considerable amount of heat present within the summit rocks. The rate at which water fills the pond gives us valuable clues about the porosity of not only the ground immediately around the crater, but the entire region. We'll learn details about how groundwater travels through its pores, fractures, and faults that we couldn't have known otherwise. And the dissolved chemicals coloring the water will tell us about the chemistry of the rocks it's traveling through and the volcanic gasses it encounters. Currently, its blue-greens and milky greens are probably being caused by sulfides and iron ions. New colors may occur if different minerals are dissolved into it, or if bacteria begin to colonize it.
We'll know far more once technical challenges are overcome, and HVO scientists figure out how to safely get physical samples of the pond water. They also have to work within the constraints of Hawaiian culture, which sees Kilauea as a sacred place, and the National Park Service, which is responsible for preserving the area's natural beauty. They've got plenty of experience working around such challenges, so they should have solutions soon.
It won't be long before we'll have more information, and when we do, I'll be sure to update you. In the meantime, enjoy the growth of Kilauea's newest lake!
*One of the neatest facts I just learned is that the magma in a lava fountain is already fragmenting before it reaches the surface! For some reason, I always assumed it came apart only after emerging.
USGS Volcano Watch Issues:
USGS Volcanoes Facebook posts and comments: