Sampling the water in Halema‘uma‘u - Kīlauea Volcano

Video Transcript
Download Video
Right-click and save to download

Detailed Description

USGS Hawaiian Volcano Observatory scientists Frank Younger and Patricia Nadeau talk about sampling the water at the bottom of Halemaʻumaʻu, a crater at the summit of Kīlauea Volcano. The water appeared in July 2019 and has steadily risen since then. On October 26, 2019, an unoccupied aerial system (UAS) was used to collect a sample of the water to investigate its source and composition. Limited UAS flights into the Kīlauea summit area are conducted with permission and coordination with Hawai‘i Volcanoes National Park. The information is used to assess hazards at Kīlauea's summit, and is shared with the National Park Service and emergency managers. For more information about Kīlauea Volcano’s summit crater lake in Halemaʻumaʻu, see this video: or



Image Dimensions: 4032 x 3024

Date Taken:

Length: 00:11:02

Location Taken: HI, US

Video Credits

Producer: Janet L. Babb, USGS Hawaiian Volcano Observatory, Geologist,
Producer: Katherine M. Mulliken, USGS Hawaiian Volcano Observatory, Geologist,

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.


The source of the water was a major unknown for us and a point for the sampling.

We saw the rise so steadily, we deduced that it was groundwater coming in but there’s also an influence of magmatic degassing and rainwater as well.

We needed to collect a sample of the water so we can get a full picture of how much gas is actually being released from the magma below.

Sulphur dioxide, or SO2, is a gas at volcanoes that can indicate how active a volcano is or how active it might become soon.

So the more Sulphur dioxide that comes out, we might be looking at more volcanic activity in the near future.

Unfortunately, water dissolves Sulphur dioxide so now we’re not getting accurate measurements of how much gas is coming from the deep magma because it’s dissolving into our water lake.

Leading up to the sampling there was a lot of preparatory work that we had to do in cooperation with Hawai‘i Volcanoes National Park and our pilot colleagues from mainland USGS offices.

We had to make sure we had all of the sample bottles ready, the actual sampler, the UAS itself, and we had to make sure that everything could be accomplished safely.

Our sampling team had decades of experience and was comprised of Hawaiian Volcano Observatory and other USGS scientists, Hawai‘i Volcanoes National Park, the Department of Interior Office of Aviation Services pilots, and others.

Well, we wanted to get an early jump on the day because generally first light is about the calmest weather conditions at the crater.

So we got out there at about 6 o’clock in the morning.

We all met at the Jaggar parking lot, which is a familiar spot for all of us and we caravanned down to the crater’s rim where we set up operations for drone sampling.

The sampling itself took place at about 9 a.m., so we had a little bit more wind to contend with than at the first light but we were still able to, under favorable wind conditions, collect the sample.

It was really exciting that morning.

There was a lot of reconnaissance that had to be done first.

Our pilots who would be flying the mission had never actually seen the crater with the lake in it.

So, they had to make sure that they were comfortable with our launch site and with the conditions in the crater.

The initial flight was just with a camera and that was so the pilots could get a sense of what things looked like down there and how the winds affected the flying.

It was the second flight of the day, after the pilots had established that things were safe and they were comfortable with flying, that’s when we were going to get the water.

So, we attached the water sampler, the pilots made sure everything was ok with the drone end of things and we, the HVO scientists, made sure the sampler was ready to go and then it took off.

So, it was a nerve-racking moment getting ready for flight.

We had put months of planning and logistics and writing permits into this and it was finally time to go.

So, the drone-pilot lifted off smoothly and rose up vertically in the air and I stabilized the sampler so it wouldn’t swing and he took off and it was all in his hands after that.

The sampling line and sampler itself consisted of 30 feet of polypropylene cord and a high density polyethylene sampler attached to the bottom.

It was necessary to attach the flagging so that the pilot, who was operating through a first-person viewer on a laptop screen, could tell range down to the water level.

So, the flagging was attached at 5-foot increments, at 5, 10, and 15 feet up from the bottom so he was able to, in this case, navigate in the vertical direction down into the water according to those sample flag tapes.

The sampling apparatus consisted of a sleeve of high density polyethylene that, when it is lowered into the water, is squeezed shut and excludes all air and then when it is pulled vertically, it traps water inside of it at a given depth and is returned to the surface.

The sample was collected at a depth of about 8 or 10 feet below the surface of the pond.

We sampled from that depth because we wanted to exclude any potentially diluted water at the very surface.

We wanted to try to characterize the entire lake with one point-sample, which is a tricky thing to do.

So, the reason the flagging tape looks like it was melted or otherwise compromised, is because it just got wet and it clung to the vertical cord.

It was great to see the drone actually get the water and then come back up out of the lake and hover there with the water sample but it was still nerve-wracking because it still had to get back to us.

So, we just had to wait and hope that it made it all the way, that it wasn’t too heavy or anything like that.

It was really gratifying when we had actually had the water returned and in our hands on the ground.

In some ways the most tricky part of the sampling was handling the sample once it came back to us at the crater rim.

The drone hovered and was able to locate the sampler right above our collection point on the blue tarp and we stabilized it, the drone pilot released the sampling cord, and then we processed the sample down it into sterile containers right on site.

Splitting the sample on site turned out to be a little trickier than we anticipated because the apparatus involves a puncture with a straw much like a juice pouch and that was a little tricky to handle.

We spilled a little bit of water.

Things got a little messy, which was perfectly fine because we were prepared in advance with safety goggles, thick rubber gloves, and safety smocks to make sure that if anything did spill, that we wouldn’t be injured or burned or otherwise affected by the potentially very acidic water.

We collected about 750 milliliters of lake water, which is about a bottle of wine’s worth.

Once we containerized the water, we had several other flights that day in order to characterize volcanic gas emissions in the crater as well as collect more imagery and after we wrapped in the field, we brought it all back to the laboratory where we further split and filtered some of the aliquots for shipment and analyses at the California Volcano Observatory in Menlo Park.

Because this the first water lake in Halema‘uma‘u in written history, water sampling has not been part of the routine of the Hawaiian Volcano Observatory.

So, we didn’t have the equipment necessary to do the advanced analyses on the water chemistry so that’s why we sent the water to our colleagues at the California Volcano Observatory, who more routinely do water sampling at other volcanoes in the United States, like Yellowstone and Mammoth Mountain.

But we were able to measure the pH of the water, that is the acidity, right there in the field before we even got back to our cars.

pH is a measure of a substance’s acidity and the scale is from 0 to 14.

Acids are from 0 to 7, higher than 7 is a base, like soap.

Generally, volcanic lakes around the world have water with a pH that is either acidic or neutral like regular drinking water.

It turned out that the water from our sample is actually a pH of 4.2 which is right in the middle, which was a little bit unexpected.

But once we found out more of the chemistry and what was dissolved in the water, it made more sense.

Our pH ended up in the middle because the acidic water is reacting with the rocks and that keeps it in more of a middle ground of the acidity.

There were several major “aha” moments, which made it really really fun.

The first of which was the pH, that we analyzed on site, was in a range that was neither as acidic as we thought it might be nor neutral as we thought it might be based on volcano crater lakes worldwide.

So it was kind of in a middle range, we were really sort of stunned by that.

One of the main reasons we wanted to get a sample of this water was to determine where this water was even coming from.

It could have just been rainwater filling in this big pit that we had now at the summit of Kīlauea.

It could have been groundwater falling in and it could have even been water vapor that was condensing after it degassed from the magma.

We had a very quick turnaround at the laboratory in California where we got near-immediate analyses of anions, cations and then, later, water isotopes and we’re currently working on publishing a data product to publicly disseminate the information.

Initial results on some of the chemistry of the water indicate that the water originated as rainwater but didn’t fall directly into the lake, it’s rainwater that made its way into the ground surface and then flowed as ground water into the lake.

The chemistry results from our colleagues at California Volcano Observatory indicate that there is definitely Sulphur dioxide being dissolved in the lake water.

We do need to get a second sample in order to better quantify how much of that Sulphur dioxide ends up in the water.

The lake is dissolving Sulphur gasses being released from the magma below, but it’s also reacting with some of the rocks in the crater.

The chemical elements from the gasses and the rocks then combine to form yellow and green and orange substances, which give the lake it’s strange color.

Aside from just getting the water sampler, some of the footage that the drone was able to collect was really pretty amazing.

Most of our imagery at this point, because we can’t get very close to the lake, is from about a kilometer or almost a mile away.

So, having the drone be able to get so close to the lake surface, and even the crater walls, to get a sense of scale and how big that crater really is, and how big the rocks and rubble are was really pretty amazing. 

The next steps involve forming a strategy for future sampling.

Most volcanic crater lakes around the world are routinely sampled in order to analyze for hazard mitigation.

So, we’re working out a strategy to create a follow-up sample schedule and routine and work with the National Park and our other cooperators in order to make that happen.

Our first water sampling was a great success but it’s not the end of the story.

We need to keep sampling the water in the Kīlauea water lake because it may change through time and those changes could indicate different things about the degassing, what the magma is doing below, and what sort of eruptive activity we might have in the future.

Until we get another water sample at Kīlauea, the Hawaiian Volcano Observatory will continue to monitor all sorts of other data streams at Kīlauea.

We will look at seismicity, ground deformation, other volcanic gasses, and things like lake level so we’re keeping an eye on what Kīlauea is doing.