Water appears in Halemaʻumaʻu - Kīlauea Volcano

Video Transcript
Download Video
Right-click and save to download

Detailed Description

USGS Hawaiian Volcano Observatory scientists Matt Patrick and Jim Kauahikaua talk about the water that appeared at the bottom of Halemaʻumaʻu, a crater at the summit of Kīlauea Volcano, in July 2019 and continues to rise today. They address why it appeared, how it’s monitored, and its potential hazards.


Image Dimensions: 1745 x 1309

Date Taken:

Length: 00:16:26

Location Taken: Kīlauea Volcano, HI, US

Video Credits

Janet L. Babb, USGS Hawaiian Volcano Observatory, Geologist, jbabb@usgs.gov Katherine M. Mulliken, USGS Hawaiian Volcano Observatory, Geologist, kmulliken@usgs.gov. Unmanned Aircraft Systems (UAS) video by the U.S. Geological Survey and Office of Aviation Services, Department of the Interior, with support from the Hawaiian Volcano Observatory and Hawai‘i Volcanoes National Park. All other imagery is USGS Hawaiian Volcano Observatory.


In 2018, the largest lower East Rift Zone eruption and summit collapses in 200 years resulted in drastic changes on Kīlauea Volcano.


At the summit, more than 60 collapse events caused the floor of Halema‘uma‘u, a crater within the caldera, to drop more than 500 meters (1600 feet).


In July 2019, yet another change occurred at the summit—water appeared at the bottom of Halema‘uma‘u.


On September 27, 2019, USGS Hawaiian Volcano Observatory scientists talked about the water—why it appeared, how it’s monitored, and its potential hazards.




The pond was first observed on July 25, 2019, and at that time it was difficult to see.

Luckily, the aerial survey that was going on had very high-resolution images that were used to confirm that this small puddle was actually there.

The pond was very small, it was about 10 m wide, or about 33 ft across, and very shallow.

Over the last two centuries, in the history of the observation of Kīlauea crater and Halema‘uma‘u, there had not been any recorded water.

Frequent visitors to the caldera, especially in the winter, know that rainfall will pond on the crater floor in small patches.

But they’re very small and they usually evaporate during the following day if it’s not continuing to rain.

Rainfall retention is a little bit higher outside the crater because of all the ash deposits there make it more difficult for rain to percolate down into the rock.

But still, the existence of a persistent crater lake is not known over the last two centuries.

Questioning Hawaiian kupuna about whether there are mentions of water lakes in the summit have not turned up anything in mele or other recordings.

The pond has been present for two months now, the water level has slowly risen.

And today, the water pond is about 100 m long, a bit over 300 ft in the east-west dimension and about 50 m in the north-south dimension, or about 150 ft or so.

That’s about the size of a football field, so it’s a pretty big size and pretty big change over a relatively short period of time.

We’ve been tracking this water level rise very closely on a near-daily basis.

What we see is that the rise rate is relatively steady, about 6 inches per day.

The total depth of the pond is approaching 10 m or so, so a little over 30 ft deep.

The pond has this kind of greenish yellow color and it’s not uniform over the surface.

There are segments of the pond that are kind of bluer or more clear and others that are more yellow or green and opaque.

The greenish yellowish color is presumably because of sulfur.

High-resolution videos show the circulation and mixing.

But what you can see in these circulation videos is what looks like clearer, fresher water that’s coming in from the south margin, kind of up-welling and mixing into the pond. 

One other common feature on the pond surface is this steaming that we see.

That’s another testament to the fact that the pond is scalding hot.

The water table under Kīlauea Volcano has been there for decades, if not centuries.

Over the last several decades, it’s been roughly in the same place of several hundred meters below the crater floor.

It was originally discovered in 1973, when a National Science Foundation research drill hole was drilled about a mile south of where the lake is now.

It drilled down to sea level, but found water just about 500 yards below the crater floor.

The water in the well has been monitored since then off and on, and its level has varied a bit, but not a whole lot, probably less than 10 or 20 yards since 1973.  

Many people have looked to see whether the water level would be affected by summit activity.  

And there were some differences that could be ascribed to those volcanic events, but nothing very definite.

And certainly, during the collapse in 2018, many of us thought that there would be a large change in the water level, but there wasn’t.

We were all kind of stumped when there was this huge collapse that took the bottom of Halema‘uma‘u crater below where the water level was thought to have been, and there was no water there.

So, over the past year, we’ve kind of accepted that maybe there was some error in the water table estimates early on.

But in July, there was water coming back into the crater and we hypothesized that it either was the returning water level of groundwater or accumulated rain water from just the nearby surficial rainfall.

But its steady growth and its color suggest that it’s part of the groundwater system that’s recovering from the 2018 collapse.

Before everything happened at the summit, there was a fairly level water table.

The first graphic shows a profile across the crater floor in cartoon style, not to scale.

But it shows Halema‘uma‘u crater within the floor of Kīlauea caldera and the blue area below is the water table that’s about 500 yds below the Kīlauea caldera floor.

This is the way it’s been very stably, except when eruptions happen on the crater floor.

In 2008, when the magma rose up to the crater floor and initially created the lava lake, as it was rising through the groundwater it apparently developed a steam sleeve around it to insulate the groundwater from the magma, and therefore, essentially prevent more explosive interactions. 

In 2018, when the crater floor collapsed, the groundwater either dropped with the crater floor or somehow vacated that space, and it became quite a bit deeper.

After the collapse, what you show as a V-here in cartoon style, the V, by the way, is about 2000 feet deep.

The water table went down with the collapsing crater floor and it was well below the bottom of the new Halema‘uma‘u pit.

With time, that rose, and in July of this year, it became visible at the bottom of the pit and it’s continued to rise ever since.

From the 1973 drill hole we know how deep the water table is at that location, about a mile south of the current crater lake.

With geophysical information from around the crater, and the crater floor, we think that the water table may be a bit higher to the north side of the crater. 

As the water table returns, it’s seeking hydraulic equilibrium with the water table around it.

So, it probably will go up to at least the level of the water in the well to the south of the caldera, but it may go a bit higher.

We know this from examples of mine lakes where on the mainland where mines are excavated to depths greater than the water table.

And in order to continue mining, they have to pump all the water out while the mine is active.

Once the mine becomes inactive and is no longer used, the pumping stops, and they allow the groundwater to come back.

So, there are many observations of this process, of water coming back to seek equilibrium with the groundwater around it.

There have been a few models that suggest pretty much as we’ve hypothesized.

This will be a slow process, although in our rocks the rise is quite rapid compared to what it would be in a different type of rock on the mainland.

But it’s doing exactly what it should, and it should slow down as it approaches that hydraulic equilibrium level.

Today, the water table has continued to rise, we expect it to rise another 60 or 70 yards before it reaches hydraulic equilibrium with the groundwater around it.

The groundwater underneath the crater is confined by structures around it.

What’s called high level water because it’s so high above sea level, does not extend to the ocean.

For example, it cuts off about at the Koaʻe fault zone, the Hilina fault zone.

So there’s some sort of structure within those fault zones that impounds, it keeps the water high within the caldera.

The water kind of sits underneath Kīlauea caldera, and no evidence that it goes very far, and certainly not within the rift zones.

But the rift zones themselves also act as hydraulic barriers to groundwater locally.

For example, in the area between Pāhoa and Keaʻau, there’s a huge amount of groundwater going through that, but the East Rift Zone confines it north of the rift zone.

It’s a dynamic time at Kīlauea’s summit right now so we’re keeping a close eye on the water pond to look for any possible changes and we’re doing that through a number of means.

First of all, we have a webcam that we set up at the summit on the west caldera rim.

We’re also going on a near-daily basis on foot to make direct observations.

One of the most important things that we do is take measurements with the laser range finder of the water level. 

We’ll walk out a short distance to the west caldera rim, we’ll set up a tripod and set up our laser range finder and make a number of measurements of the water level.

It’s important to keep in mind that the distance there is actually…we’re pretty far away.

We‘re about 2000 feet above the water pond.

Temperature, based on the thermal camera measurements of the water surface, is about 70 degrees Celsius or about 160 degrees Fahrenheit.

That’s telling us that the water is heated by the magmatic system from depth.

One thing that we’ve we noticed through tracking is that the water temperature has been very stable.

We also rely on visual observations, taking high-resolution photographs.

We’re also monitoring on regular overflights, and that gives us a view of portions of the pond that we can’t necessarily see from the ground.

In addition to visual monitoring, field measurements, and thermal monitoring, we have our extensive network of geophysical and geodetic and geochemistry monitoring tools that are situated at the summit.

Kīlauea is one of the best monitored volcanoes on Earth and the summit network is particularly dense.

So there’s a really close, continuous view of what’s going on there.

The next step in monitoring the pond and its potential hazards is really taking a direct sample of the water to look at the chemistry.

Being able to put a constraint on how much sulfur the pond is actually absorbing would be very useful for monitoring.

At other volcanic lakes, changes in lake chemistry can sometimes be a precursor to changes on the volcano and changes deeper in the magmatic system.

So being able to take direct measurements and track the chemistry of the pond is really a fundamental part of monitoring it.

We’re working with Hawaiʻi Volcanoes National Park to look at the feasibility of how we’re going to collect that sample.

It’s not an easy thing to do, it’s a challenging location, the pond is very deep in the crater.

At Kīlauea, or any other volcano, whenever you have magma interacting with water, there’s the potential for explosive activity.

At Kīlauea summit, most likely those explosions would be relatively small, affecting just the immediate caldera floor.

We know from our observations of the coastal entries that magma or lava mixing with water of any kind, ocean water or fresh water, can result in minor explosions.

But it doesn’t always.

Sometimes the lava goes just straight into the ocean without much interaction, we’ve seen that also.

So, it’s not a guarantee that lava encountering this crater lake at the summit will create even a minor explosion, but it might, and so we should be prepared for that.

There are geological inferences that larger explosions have happened.

Modeling based on the inferences from past explosions centuries ago suggests that two things have to occur for the water lake to contribute to explosions.

That is, the existence of the lake, and the second is very fast rising magma coming to the surface.

Faster than we’ve seen probably seen in the last two centuries at the summit.

So, we should be positioned fairly well to see precursors of such activity since it’d be relatively unusual to what we’ve seen in the past.

The pond is very new.

Definitely, in the past 200 years, we haven’t seen activity like this.

So, we’re keeping a close eye on this and really looking for any kind of signs that might be potential precursors for larger scale activity in the future.

Inflation that would indicate magma rising or seismic activity that would indicate unrest in the deeper magmatic system.

It is possible, and this has happened at other volcanic lakes, that small explosions have occurred with little or no warning.

But again, those are most likely events that would just impact the immediate caldera area.

The takeaway right now is that there’s no immediate signs of imminent increased hazard at the summit.

Sulphur dioxide emission rates are low and seismicity is elevated relative to the levels before the 2018 eruption, but it’s stable.

Looking at all the indicators, it paints a relatively stable picture of the summit.

And, we’re continuing to keep a very close eye on all those indicators to see any hint of change in the future that might be a precursor to more hazardous activity.

The appearance of this lake has offered us an unprecedented opportunity to study how it forms and perhaps give us clues about why the water table was depressed in the first place.

And that may lead us to some interesting insights into the collapse itself.