Kīlauea Lower East Rift Zone 2019: Quiet but insightful

Video Transcript
Download Video
Right-click and save to download

Detailed Description

In the year since Kīlauea Volcano’s notable 2018 eruption ended, the lower East Rift Zone has been relatively quiet. But USGS Hawaiian Volcano Observatory scientists continue to gain insight into the eruption through ongoing research and monitoring. Some of the many questions asked by island residents include, Why did the fissures erupt along a linear pattern? How long will it take for the lava to solidify? Why is vegetation still dying in the area? USGS Hawaiian Volcano Observatory geologist Carolyn Parcheta explored these and other queries and shared recent observations and findings by HVO scientists in this Volcano Awareness Month talk presented in Hawai`i Volcanoes National Park on January 21, 2020. Volcano Awareness Month is spearheaded by the USGS–Hawaiian Volcano Observatory, in cooperation with Hawai‘i Volcanoes National Park, and provides informative and engaging public programs about the science and hazards of Hawaiian volcanoes. Photo caption: Aerial view of fissure 8 lava flow on Kīlauea’s lower East Rift Zone after the 2018 eruption ended. USGS photo by D. Becker, 08-31-2019.


Image Dimensions: 1920 x 1281

Date Taken:

Length: 00:48:00

Location Taken: HI, US

Video Credits

Katherine Mulliken, Geologist, Hawaiian Volcano Observatory, kmulliken@usgs.gov


Thank you all for coming tonight. It's great to be back. This is my third time doing a Volcano Awareness Month talk. I really enjoy engaging with you, so hope you enjoy this talk.

It's a little different than past years because, as the title says, this year was quiet, but insightful. We're going to have a very quick review of where we are and what Kīlauea has done in the past, and then we'll talk about some things that my colleagues and I have all been working together. There are slides from multiple other people's presentations in here that I’ve put together. While my name is on this, it’s certainly everybody at HVO who's contributed to what’s in this talk.

We are here on the Big Island of Hawaiʻi. We are up here at the National Park. Our island is made up of 5 volcanoes, for those of you who are visiting and may not know. This is Kohala, up top. Mauna Kea, with the telescopes. Hualālai. Mauna Loa, which is our largest volcano on the planet. And Kīlauea, where we are. We also have Lōʻihi down here, a submarine volcano. It's going to stay underwater for quite some time, so don't worry about seeing that one. These are all basaltic shield volcanoes, which means that they are not your typical conical cone that most people think of as a volcano. These are broad shield volcanoes and they're classic in their own right.


They typically have rift zones. In our case, they have two rift zones that come off a summit caldera. We have an East Rift Zone, that stretches all the way out here. This is what we'll be talking about today. And we have a Southwest Rift Zone that comes down here. Under Kīlauea we have at least two magma chambers, the Halema‘uma‘u shallow source, and the south caldera deeper source. Along the rift zones, this is going to come into play in our talk, there are these little stored areas of magma throughout the whole rift zone. We’ll dive into that a little later. There are also two major fault systems on our volcano here: The Koaʻe fault system, which links the upper East Rift Zone to the seismic Southwest Rift Zone, and the Hilina fault zone, which basically is the part of the edge of the south flank that moves on our volcano here.

This is a very abridged eruption history of Kīlauea. For those of you who are diehard fans of Volcano Awareness Month, it’s going to be just as odd for you as it is for me to summarize Pu‘u ‘Ō‘ō in one slide. But we're going to do it. A brief history of what Kīlauea has done, setting the scene a little more for what happened in 2018.


In the past 70 years, we started with a 1955 eruption, labeled here, these two gray spots. That was followed in 1959 by the Kīlauea Iki eruption up here at the summit in December. Then the next month, in January 1960, the Kapoho eruption happened down here on the lower East Rift Zone. LERZ stands for lower East Rift Zone and this is upper East Rift Zone here.


Through the 60s there were multiple fissures, roughly one a decade—not quite, some years had two and some years had none—but roughly one a decade, fissure eruptions on the upper East Rift Zone up towards Mauna Ulu. Then in 1969, Mauna Ulu started erupting with some lava fountaining through the rest of that year. It lasted another 4.5 years, until 1974, all effusive lava flows. There was a little fountaining here and there, but no major episodes of that. There were then some fissure eruptions in the middle rift zone, here, and the upper rift zone between 1974 and 1983.


Then, in 1983, we started 3 years of 48 different lava fountaining episodes that became the Pu‘u ‘Ō‘ō eruption. There was another 32 years, beyond that, of lava effusion, and that made for a 35-year-long historic eruption. That was joined at the summit by a 10-year-long lava lake that started in 2008. As you can guess, 2018 stopped both of them.


So, those eruptions have ended. 2018 was historic in its own right, and we're going to get into that tonight.

The first day leading up to the eruption, the first day of precursors, was April 30, 2018. Pu‘u ‘Ō‘ō,  around 2:15 p.m., had this massive swelling, or inflation, and then it collapsed on itself. This is the cone and it made this typical pink ash plume for many days that week and several weeks after. That inflation that happened for about 15 minutes, cracked the west side of the cone. It was pretty rainy that day, so we couldn't see anything in our [web]cameras. The thermal camera caught a little bit. We ended up taking a helicopter out the next day to see what had happened. This was a huge signal that has never been seen to that magnitude here before.


Here's a video from the helicopter. This used to be filled with lava—you couldn’t see down here. When we took this helicopter flight, we couldn't see the bottom of this hole. It was at least 500 or 600 meters [1640 or 1968 feet] deep. There's a little dust that sluffs off and it just keeps going. You don't even see it rebound when it hits the bottom. It was very deep. It has since filled in with talus. But that was our first clue, when this collapsed, that something was happening. Then seeing how deep and how big this scar was inside the crater was really eye-opening.

I’m going to talk about this plot first. This is the rest of the afternoon of April 30, [2018] after about 2:30 p.m., there were earthquakes that started migrating down rift. Typically, a lot of our activity for the past 35 years, had been in this middle rift zone, at Pu‘u ‘Ō‘ō, or somewhere in the upper East Rift Zone connector. On this day, whatever happened at Pu‘u ‘Ō‘ō, whatever failed underneath it, allowed magma to move forward, down into the lower rift, which is down here.


On May 1, we see that earthquake front migrating to the east. Then, on May 2, it stops just short of our station WAPM, right under the subdivision of Leilani estates. To give you a sense of what this image is, it's derived from what we call InSAR. It’s a satellite that passes over and measures, it can measure very fine movements of the ground. Typically, you see it as a colorful rainbow image, but this has been processed such that you see subsidence in blue and inflation in red. These are in millimeters, so these are not huge distances even though they look very dark blue. But it gives you a sense that during this 3–4 day period, the entire middle rift zone and part of the upper rift zone was losing material in the subsurface to the lower rift zone, which was
becoming engorged.

The other thing that started happening on May 2 was that surface streets in Leilani Estates began to crack. At first it was 1 crack. Then it was 2 cracks and then 5 cracks, then 10—and we realized this is going to happen fast. Again, the ground is swelling up as material is moving under Leilani Estates.

The next day, on May 3, we have some more earthquakes in the same area. Fissure 1 opens that evening around 4:30 p.m. It crossed Mohala Street.


This is what it looked like. It's already starting to make a lava pad, but that pad is not going very far, only a couple of feet. It's pretty thick, maybe 2-3 feet tall. This is what you would call a classic fissure. It's a line, or a crack, that erupts magma. There are multiple sources of gas pushing the magma out of this crack. It's pretty sluggish. It's not what we typically think of from the pāhoehoe from Pu‘u ‘Ō‘ō. This stuff isn't going very far, but it's also in a neighborhood, so it's a good thing it's not going very far.


The next morning, literally at 1:00 a.m. the next morning, fissure 2 opened. The next few fissures, all the way to fissure 6, opened sequentially downrift, right to about where the earthquakes had stopped progressing forward.


In the middle of all that, on May 4, we had a magnitude-6.9 earthquake that was felt throughout the State. While it shook a lot of us, especially those in Puna, it didn't seem to affect those 6 fissures that were opening up, or those 5 fissures that were opening up that day. They just carried on with their sluggish spattering as if nothing was wrong.


During this earthquake the entire south flank of Kīlauea moved—this is another one of those InSAR derived diagrams—150 millimeters, or 15 centimeters [6 inches]. Everything from this line and down moved at least that much, if not more.

On May 6, after fissure 6 was done erupting over here, fissure 7 opened up the next morning right next to fissure 2. Then the vents started opening back uprift with the exception of 12—that  jumped the line and went back downrift. But, in general, the trend was an uprift migration.


Here's a video of fissure 11, I turned the sound off in this one because all you hear is a very loud helicopter, you can’t hear the fissure at all. Again, it's very sluggish, perhaps a little more vigorous spattering. There are multiple spots but they're all kind of red together instead of individual red spots. But it's lava that's not going very far. It's very sticky. It's very viscous.


There are two days of pause, May 10 and May 11, when we thought, “Hmm, is that it? Are we done, or is there more coming?” We assumed there's more coming because the earthquakes started migrating downrift again. They didn't go that far downrift, but they went far enough that we thought the next fissures could be downrift. Sure enough, fissure 13 opened here, 14 and the rest open sequentially downrift. Fissure 17 is offset—it’s the only one offset majorly on the whole trend.


Here's a video of what that looks like and sounds like. That was a cannon-shot sound that it was making. At the peak of its activity, it was doing that every 5-10 minutes and you could hear that for several miles away from the fissure. This was a very interesting fissure because, not only was it offset, it was erupting a slightly different composition, which we didn't know up front, but we assumed something was different because none of the other fissures were making this feature. We'll talk more about the chemistry of this later in the talk, but essentially, this is a more evolved basalt, a basaltic andesite. This more explosive area was just an andesite, there was no basaltic component to it, just andesite.

May 14 to 18. The rest of the fissures, up through fissure 23, began opening uprift again. The thought is that the earthquakes came down, something was preventing them from moving farther, the dike was able to fill that crack that they had created and erupt in here. Then the earthquakes were able to break open the rock and free up some more space, and the magma moved down here. Once that couldn't move forward anymore, it starts filling up and backing up uprift. Fissure 24, over here, opened on May 27. That was because of a series of reactivations that finally went from fissure 17 all the way back to fissure 8. Fissure 24 opened for the first time when fissure 8 reactivated for its big show in June and July.

This is a slide that one of my colleagues made showing seismicity in the lower East Rift Zone, the south flank area, and the summit. The key thing to take away from this slide is that all the seismic activity for the lower East Rift Zone happened in those first couple of days in May. From April 30 to May 18. Then everything went quiet on the rift zone. There was still some activity but compared to what we had just seen on the lower East Rift Zone, it was relatively small. And all the seismic activity went to this blue area, which is the summit. We're not going to talk about that part today, we're going to focus on the lower rift.


The deformation also did the same thing. This is based off InSAR data again, and while it shows a general subsidence across the entire East Rift Zone, it's very small, it’s two centimeters, one inch, where in previous slides we were seeing multiple inches of change. It's all kind of a wash in here. The interpretation we came to, eventually, was that if there's no geophysics, which usually means the eruption is ending, and yet we still see lava coming out and it's coming out more vigorously and faster, what does that mean? It probably means the conduit is “open.” There's no more resistance for the lava to flow and try to break rock to get to the surface. It has a path from wherever it's coming from in the rift zone or the summit, and it has a clear, easy path to fissure 8, or some of the other fissures that rejuvenated right before fissure 8.

Here's a video of fissure 20. This is our second fissure to show classic Hawaiian fountaining from a fissure source. Multiple fountains along a single crack erupting magma.


That is producing a lava flow that is moving quite quickly down from fissure 20 and around into a subdivision. It covered this area, which is probably a kilometer [0.6 miles] or little more in distance, within minutes to an hour. It was moving very fast. So much so that the helicopter pilot and I actually got out on the one of these side roads of the cul de sac. I called Civil Defense and he went to warn a few people who were still home that they needed to leave because this was coming at them very quickly.

From May 18, when our geophysics goes quiet on the lower rift, to May 27, right as fissure 8 reactivates, we have a series of uprift reactivations. Fissure 17 on May 18-19 was going strong and had an 80-meter-tall [262 feet] fountain. Fissures 16 and 18 were also active; 20 and 22 were starting to send their flows down to the sea. They did that in a day and a half. As fissure 17 started to shut down, fissure 19 and 15 started to reactivate. Then I noticed as 16 and 18 appeared to shut down, turns out they never quite shut down, but as they decreased activity, fissure 6 and 13 started to increase activity, and they produced a joint lava flow that also went to the sea.


Then, as fissures 20 and 22 started to shut down, 23, 21 and 7 started producing a flow; 21 and 7 together sent a flow to the north, toward the geothermal plant, and to the south. I think had it not filled this pit crater here, it would have reached the sea as well, but quite a bit of volume went into this pit crater first. Then, as everything else shut down—6, 13, 23—fissure 8 reactivated on May 27. With that came the new fissure 24. Once that started flowing and effusing a lava flow, fissures 7 and 21 shut down as well.


During that time of reactivation, as the lower vents shut off, and that pushes magma to come out of the next available upper vent, we saw what's called a “graben” form. If you think of a keystone in an archway, a graben is essentially that kind of a shape and it drops down as magma ascends. This one is fairly small. It's 100 meters [328 feet], or a football field, across. But it dropped about 2 meters, or a little over 6 feet, over the course of 2–3 days.


The reason these grabens form is when magma is trying to rise, it has to push the walls apart. So, just like a keystone helps keep a doorway intact, the ground surface will drop and behave in much the same way as it's being pushed apart so the lava, the magma, can reach the surface. This is LIDAR based, because it's a little too fine scale for the InSAR derived image. But you can get a hint of it, there's a few blue spots in the same trend. Like I said, this is 100 meters [328 feet] across, one football field across, and several kilometers [roughly 2 miles] long. It stretched essentially from Pohoiki all the way Highway 130, through the whole subdivision.

Here's what that looks like. Over the course of a little more than 24 hours, and prior to this, it had been just a hairline crack, turned into a 1–2 foot crack and 6–12 inches of downdrop, to a 6–8 foot wide crack and 3–4 feet of downdrop. It ultimately reached about 6.5 feet of downdrop. This happened very fast. I had a resident stop me in the middle of one of my shifts and say, “is the ground is sinking?” Yes, the ground is sinking. That was our cue that a lot more magma was about to come out of whichever fissure it chose as the easiest path.


That was fissure 8, and it was active for two months. It started its first rejuvenated flow on May 27. It lasted until August 5. In this picture, which was taken a month into that time frame, on June 26, you can see how long and how wide it is. This red line is a half mile wide, or 800 meters across. The flow came up Luana Street and made a bend to the northeast here in this picture. It went 8 miles, 13 kilometers, until it reached the sea. It did that in 5 days.


The average eruption rate for those 2 months was approximately 100 cubic meters [ 22,000 gallons] per second. To put that into terms that we might recognize, for those of you who have come to these talks a lot and think about Pu‘u ‘Ō‘ō a lot, fissure 8 erupted 5 years-worth of Pu‘u ‘Ō‘ō material in 2 months. That works out to about a cubic kilometer [0.24 cubic miles], uncorrected, effused material. We haven’t corrected for the density in that.


But fissure 8 wasn't done. Even though the channel drained—the effusion rate dropped dramatically on August 4, the channel drained on August 5—there's still a little bit of lava inside the cone from August 6 to 15. There was still a few ooze-outs and ocean entry toes, as well, during this period. But there was nothing in the channel by August 6.

Then lava went away for a while. So again, we were asking ourselves, “Is this it? Is there more coming? Is this just a break? How will we know? ” Lava came back one last time September 1-5. It was very weakly active. It stayed within the fissure 8 cone.


This is a video showing just how weak it’s spattering. A nice breakout here, with the crust being pushed off. But it didn't make it into the spillway and it didn't reoccupy the channel. Then it drained away as quietly as it came back. That was finally the end of the eruption.


That brings us to 2019, which is what this talk is usually about, the past year of what's been going on. We can summarize it in just 1–2 slides because, thankfully, it's been quiet. The unfortunate thing is that some residents are still feeling the thermal effects of the intrusion. As the dike is trying to cool, the heat is moving into the surrounding ground.


This is one resident’s property—there’s a breadfruit tree right here. That was the first crack on this resident’s property, and it was also the first steaming area. We thought this was going to erupt at one point because it had a blue tinge to it, which means there's sulfur dioxide mixed into that, but it never erupted. However, a year later—this is the same breadfruit tree—all her vegetation has essentially been boiled and dissolved away by the groundwater that is so hot from this intrusion. It used to be a nice lush green field and by time we took this picture, parts of it had died. Now, the plants literally fell on the ground as their roots boiled and then they dissolve themselves. It leaves this little white crust behind on the rock.


Those residents are in Ala‘ili and they've been feeling this for the better part of a year and a half now. This started for them during the eruption. They saw it progress through the eruption and it's been getting worse after the eruption. Recently Halekamahina residents have called us and said they started noticing some similar effects that we recognized from Ala‘ili this fall, for the last 5–6 months. Then Leilani Estates called in December and said folks near fissures 9 and 10 are noticing the temperatures on their property are rising, is everything okay? The unfortunate answer is, yes, everything is okay… but here's what's happening.


Here's the areas that are affected. This is Ala‘ili area. This is the Leilani Estates area that was recently affected. And this is the Halekamahina area.

In our geologic cartoon, we have a dike that's coming up to the surface. It doesn't break the surface in this cartoon, it is just an intrusion to help conceptually with the process.


Thermally, right above the intrusion is very hot, magmatic temperature hot. Hot steam coming out of this crack. We're talking anywhere from 200 to 500 Celsius [390 to 930 Fahrenheit], so very hot. Much higher above the boiling temperature than anyone would want to approach. But just a few yards away, 20–30 yards away, the ground feels normal, feels cool. It doesn't feel hot at all. It's just these cracks that are really hot. Fast forward, whatever time frame you want—6 months, a year—this heat will slowly start leaching into the ground rock.


This spot will cool down, but these spots will warm up. So, this has basically cooled to the boiling point, it seems to be buffered there. We haven't seen the temperatures change on these properties above or below the boiling point too much. But now that 20 yards away that used to be cool, also has steaming water and is just shy of the boiling point. That's because the heat from this dike is slowly migrating outward into the ground as this tries to solidify and crystallize.


Unfortunately, that means that farther away areas you thought wouldn't be impacted at all by the heat are now starting to warm up. So, a lot of residents in these affected areas are asking themselves, is there anything we can do? What should we do? Unfortunately, our best guidance is the 1955 eruption where there were steam vents that were still warm and spa attractions as recently as 2017, 2018. So, it can stay hot for several decades. That's just how the heat is trying to leave the dike so that it can solidify fully.

As I mentioned at the beginning, this eruption was unprecedented for Kīlauea in the past 200 years of its activity. The total flow field area, which is in pink here, is 13.7 square miles. The new lava delta area—this is the old shoreline that’s dashed here, so this area that extends out into the sea—is 875 acres.

It was the most voluminous eruption for two months of time in the past 200 years. Pu‘u ‘Ō‘ō did erupt more volume, but it was over 35 years. It's the highest sustained lava discharge rate ever measured from Kīlauea. And it's the highest SO2 [sulfur dioxide] rates measured at Kīlauea. In addition, the 6.9 earthquake was the largest we'd had since 1975.

Let's talk about the gas for slide or two, here. Our gas team started measuring SO2 output from Kīlauea in 1979. When the Pu‘u ‘Ō‘ō eruption started in 1983, we averaged about 1 mega-tonne, or 1 million tonnes, of SO2 per year. When the summit eruption and the lava lake started in 2008, that bumped up to 2-3 million tonnes of sulfur dioxide per year from Kīlauea.


However, you can see this giant bar over here on the side the plot. That is solely 2018, and that is solely from the lower East Rift Zone, when that eruption started in May, and especially as it picked up in June, the summit and Pu‘u ‘Ō‘ō areas basically dropped to zero sulfur dioxide output. It was within the noise, we weren't sure if were actually getting a reading or if it was just within the noise. So, 10 million tonnes of sulfur [dioxide] for the year came out of Leilani Estates. 2019 is down there, with 13,000 tonnes for the year. Much reduced. We have the best air quality on this island that we've had in a long time. That is the lowest rate since these measurements have started being recorded by our gas team.


That's the summary of the past year and the 2018 eruption. Short and to the point. HVO has been using this quiet year (A) to move into our temporary facility and (B) to start addressing some of these questions and gain some insights on what happened and why, and try to find some answers so we can understand future potential eruption crises better.


We touched on this already. What does it mean when the seismicity and deformation stop, even though the lava still flowing? It means the conduit is open and that there's likely more material coming. Until we see some other sign, like the gas drop or the lava just stops coming out. It's really a fluid dynamic and a magma static head that's continuing that material to erupt.

How does magmatic chemical variation affect eruptions? This first point is really about the structure of the volcano, because there's more that controls an eruption than structure. There's also the chemistry. In this work, I've taken a few slides from our colleague Cheryl Gansecki at UH-Hilo. She did a fantastic job during the eruption with her colleagues and with HVO, essentially doing petrology real-time. She kind of pioneered that with a proof-of-concept in this eruption.


Let's look at how different chemistries affect what the eruption does. This is a repeat from her talk last Thursday night. If any of you were there, you'll recognize a few of these slides. Again, we can look back to the 1955 eruption that was in the same area. Our 2018 eruption was in this gap right here. They noted in 1955 that the lavas seemed cool, and that they might have been stored for a while. They had more crystals, they're sticky and viscous, like our first 15 fissures were in 2018. The lava produced later on in the 1955 eruption was more hot and fluid and covered a lot more area, as was the case in 2018, as well.

We're back to these little pods of magma that are stored in the rift zone. Why are they stored there? Any eruption that happens never erupts 100 percent of its material. There’s always something that's left behind. Depending on how warm that something can stay, it could be remobilized later.


We call these magma storage regions. There’s one under Makaopuhi Crater that drained during this event. Then there are probably numerous others that we don't even know about. There are half dozen or so that we do know about. It's likely residue from earlier eruptions, and for the 2018 eruption, the chemistry seems to match an evolved 1955 [magma]. So, the 2018 material came in, but to get out, first it had to push out what was left of the 1955 magma.


When the magma is in the rift zone, without an intrusion to help push it along, it’s not going to erupt again on its own, which gives it time to cool and crystallize and solidify. But, as we know, since [2018] erupted evolved 1955 [magma], that process can easily take 7 decades. That's how long this has been sitting there, and it came out again.  


Here's one of Cheryl’s chemistry plots. Don't be intimidated. It's actually easy. She was looking at calcium oxide as a proxy for temperature. So, she's measuring this and and she can determine the temperature of the lava that it erupts at. Fissures 1-15, here in the yellow, all erupted pretty cool for Kīlauea—around 1110 Celsius [2010 Fahrenheit]. Normal composition for Pu‘u ‘Ō‘ō just a month prior to this event is up here. So, this a cooler temperature.


Fissure 16—we're going to skip 17 for a second—and 18, 19, 20, were a little warmer. Remember that was as we started to turn the page from sticky magma to more fluid and less viscous magma and hotter magma.


Then, when our geophysics die away in the lower rift, we start seeing lava temperatures that are much hotter. They’re not quite Pu‘u ‘Ō‘ō temperatures, they’re just shy of it. That remains true for the rest of the eruption when fissure 8 turns on and for the whole time that was active. Why is that a little cooler than Pu‘u ‘Ō‘ō if we think Pu‘u ‘Ō‘ō material is actually coming out at this point? Well, it has to travel twice as far, so it's going to lose a fraction of its heat in that transit underground.  


Now, fissure 17. Let's talk about that. So that erupted really cold (for lava), especially from Kīlauea. Fissure 17 did that. This is the initial fissure 17 samples. Later in its life, before it totally waned away, it did warm up a little bit. This is, again, because of that chemistry issue. These are all evolved 1955 basalt that’s getting pushed out. This is basalt that we're used to seeing from Pu‘u ‘Ō‘ō and maybe even a little bit from the summit. But this is basaltic andesite and andesite, it’s naturally a slightly colder magma. 


Since it’s sitting in the rift zone, and we don't know from what eruption they came from. There are some options, but I'm not sure that Cheryl and her group have pinned it down yet. Magma is not only cooling, it has time to evolve. As crystals form, they leave behind silica dioxide, and that's the difference between basalt and andesite—how much silica dioxide is in the liquid rock. The more silica, the more viscous it is. The more viscous it is, the more explosive it is. So, new eruptions can push out the older magma, but it does come with the consequence of potentially increased hazards.


That's our insight number 2: new magma types can lead to new eruption styles, like a big explosive boomer that you saw and the cannon shot that it produced. It also comes with the hazards that that produces, where lava bombs are flying much farther than they otherwise would have. This is a slightly denser rock; it punctured through a roof, I believe, and a wall, and unfortunately, broke someone’s leg. It has some extra hazards that you wouldn't think of for basaltic eruptions.


Next question. Why did the fissure 8 fountain pulse? Matt and I, and many of our colleagues, noticed that several of the fountains, but especially the fissure 8 fountain, was sitting there doing its thing when it would go up and down like every 5–10 minutes. The question was why? What is that all about? The next few slides are from Matt’s talk from last year, and some of his new work that we co-published together this year.


This is a video, sped up 5 times. This is courtesy of our UAS team that was here during the eruption. They are USGS, not HVO-specific, but they were a huge help in monitoring the eruption. You can see that, at times, fissure 8, which is here, is sending a lot of material through the spillway into the channel, which starts about here. We're getting some convection, it looks like turbulent flow. Then, after 5–10 minutes—remember this is sped up so it's going to be a few seconds here—it slows down. We start seeing more crust and we lose our turbulent eddies; they're almost all gone except for that one. It stays like this, nice and calm and flowing a little more slowly. Then it'll speed up again.


Matt finally wrapped his head around it and was able to figure out that this is very similar to a process that we've seen before called “gas pistoning.” The only difference is that gas pistoning, when we’ve seen it at Mauna Ulu, Pu‘u ‘Ō‘ō, and Halema‘uma‘u, it’s confined, it’s in a pit. So, you see the lava rise up. Then when the gas can finally break the surface, the gas all comes out and the lava drops back down. It's a cyclic rise and fall of the lava surface, depending on the gas build-up and the gas release.


But if you're in a fountain that has a spillway where a lava can leave, and it's not confined, it manifests itself a little differently. Matt was able, after watching hours and hours of video and tediously analyzing it, he was able to figure out that as the fountain height goes up, more SO2 comes out the fountain, so our gas flux goes up. And the seismic tremor goes up. Keep in mind this is all smaller than the early May seismic tremor, but it is still there.


As a result of that, the lava level in the channel goes down, and the bulk effusion rate goes down, and the velocity of the lava in the channel goes down. But when the fountains start to die, or start to wane, and come down in height, the SO2 stays trapped in those fountains a little longer. The seismic tremor goes down. The lava level in the channel goes up, because all that material, instead of going up and releasing the gas, is now coming out and over into the spillway. So, the bulk effusion rate goes up, and the velocity of the channel goes up. This happens on the order of minutes, roughly every 10 minutes or so you can see this change happens. Sometimes it's more dramatic than others, but this seems to happen on a cyclic pattern just like gas pistoning.


Here’s a video of that. This is the low-flow time where the fountains are high, just off the screen.  Higher, they got about 50 meters [164 feet] around this time. Then you'll see it starting to speed up again. The lava is reaching higher because the fountains are dropping. It's coming down the spillway faster, so it banks up and builds up these ramparts in the back here. We can use gas pistoning to help explain this behavior in the lava channel that, for a while during the eruption, was kind of enigmatic for us.

Last question. Why did the fissure 8 lava delta not collapse? I've been asked that many times. I'm not sure we entirely know the answer, but I'll tell you where my thoughts are on this. This is something that my colleague, Mike Zoeller, and I are working on right now.


To refresh your memory about deltas, we are going to jump back to [Pu‘u ‘Ō‘ō episode] 61g in 2017. When a pāhoehoe lava delta comes, it has to dribble over a cliff, then find the sea floor and start building itself up. It does this by putting layer on top of layer on top of layer. When it interacts with the water, it makes this fine sediment base, as well. So, it may have some lava flow in there and it may be some particles, and a mix thereof. Whereas up top, it's all lava flows. As you can imagine, this gets heavy, and is not very stable.


At some point, you develop a zone of weakness. That manifests itself on the surface, up here, as cracks in the delta. Just because we see a crack, we don't know how deep it extends. All we know is that this land is unstable, and because it's cracking, it might fail sooner than later.


When it does, it can either fail partially or in whole, and take some of the previous sea cliff with it. Either way, when hot water from the ocean entry interacts with freshly exposed lava, you can get big explosions, and they have hurt and killed people in in Hawaiʻi in the past.

What’s different about 2018? Why didn't that happen? These graphics—I love them, they're wonderful—but I will note that they're two times height exaggerated. So, I've redrawn it to be one-to-one. And I've drawn this one from Kapoho to also be one-to-one, so that we're comparing apples to apples here.


You'll notice that at the zero marker, which is where the coastline is, a pāhoehoe delta has a steep drop. It has to build itself up until it reaches a good equilibrium spot. It’s still over-steepened here. This arrow implies that the current could erode some of this more fine-grained sediment and help undermine the delta.


If we look at Kapoho Bay and the first part of the delta that formed there, it's pretty flat. So, I'm not worried about lava that filled in the bay collapsing, per se. It’s not until it reaches the edge of the shelf and drops down the deeper levels in the ocean that it starts to approach this slope. But even so, on this slope. This lava flow, even though it's an ʻaʻā flow, is going to behave a little differently than pāhoehoe, which is over-steepened onto this slope. I think, inherently, pāhoehoe might be more inclined to collapse, as a delta, than ʻaʻā flows.


We still have more work to do to support this, but we haven't seen any cracks form in the Kapoho ʻaʻā delta, nor did we during the eruption. Maybe one crack. We typically didn't see the features that we were looking for to say a collapse is coming, in who knows, days, weeks, months, but the cracks are there.


The other interesting thing that we noticed—I have some pictures on the next two slides for you—is that, right here, there's a little bit of breakouts at the ocean entry that does one of these pāhoehoe things and it overlaps on top of the ʻaʻā. In general, the channel will just flow straight into the ocean. That told me that it's entering some protected … it's protected at the top somehow. It’s not interacting with the water. It's not exploding. It did on a few occasions, but fairly rare for the whole eruption. It went out—this is a real bathymetry profile of pre-eruption and from boats who came out to do surveying—it went out 2.5 km [1.5 mi] away from … the edge of the old coastline.


So, the thought is, where's all this material going? Many of us at HVO asked that … where's all this material going? Why isn't it exploding when we know these little tiny things can cause explosions from pāhoehoe deltas. What I realized is that this is inflating and lifting up out of the sea. It didn't do that all the time, but … the next slide, I'll show you an example where we did see it lift up. We saw that twice actually, and it was fairly interesting.


But the point is that this is a cohesive flow as it extends underwater, whereas the pāhoehoe is onlapping layers and layers. So, perhaps this is actually a more stable delta anyways, even though it already has more shallow slope to begin with.


This is shortly after the lava entered Kapoho Bay. We're going to focus on this feature here, it's got a peak in the middle and two dips.  We're going to try to track that through June. Here we go. I put an arrow there to help. It's going to get more and more subtle as time goes on, but it's here. I want you to focus on this next lobe that came out and formed a new part of the delta. We're going to look at what happens to the water between these two. So, in this picture down below, it's here.


Then on June 11, a few days later, it's here. On June 14, it's isolated from the ocean, and it's slightly higher than the ocean.


June 15, not much change. June 22, it's evaporating away, and it is 1–2 feet above sea level at this point. That's something that I don't think we've ever seen in pāhoehoe delta, where it lifts up from the sea, as opposed to piling on top and going into the sea and diving down.


The last example I have of this is from July. We were doing a routine flight—at this point we weren't allowed to fly over the open ocean because, it was such a long stretch where we have no emergency landing that we decided it's not worth it, but this was right on the edge. This was the coastline the day before, and then suddenly this “island” appeared. But really, this just rose up, it's connected to the delta underneath. There's only a couple of feet of water here. And it’s erupting lava. It’s still a fissure 8 lava that has just come down through the delta underneath and back up. Again, we have another example of something rising up, as opposed to piling on top and diving down.

The possible insights from last year that we're going to keep working on for this year… that the submarine slope was shallower in Kapoho for that delta, and perhaps ʻaʻā deltas not only grow by onlapping, they also grow by inflating their core. So, maybe this is more stable. There's more work that needs to be done, but that’s a potential interesting insight into how lava deltas work. With that… I forgot the coquis were in this this video ...

I would love to thank you for joining me and HVO tonight. We really enjoy serving the community and look forward to doing that in the future. Regardless of where our new home is, it'll be on this island.

A big thanks to Civil Defense and the Puna residents who are still welcoming us into your communities and letting us take measurements. Also thank you to Janet, she does a great job organizing all this stuff for us and it’s a lot of fun for us to give these talks, but she’s like the unsung hero of all this.

Thank you, everybody, and I’ll take your questions.