Tag Archives: lava lake

Is the Kilauea lava lake less dense than water?

According to a new study, the answer is yes. In a paper published this week in Earth and Planetary Science Letters, lead author Daniel Carbone says that, based on gravity measurements in the lava lake at Hawaii’s Kilauea volcano, they’ve worked out that the bulk density of the stuff in the lava lake (that includes lava, dissolved volatiles, and gas) is an astonishingly low 0.95 g/cm3. For those keeping track, that’s less than the density of water (1 g/cm3).

Thermal image of Kilauea’s lava lake (Carbone et al., 2013)

How is this possible? Carbone and colleagues suggest that the vapor-melt ratio (that is, how much gas there is in the lava lake compared to how much lava) could be very high to account for the low density. Since the density of gas is so much lower than that of lava, more gas means lower density. Intrigued, I decided to do the math. Just how much gas does it take to get a lava lake with a density of 0.95? Here’s a quick run-down of my calculations. If anyone spot’s an error or wants to improve upon this back-of-the-envelope calculation, please do so in the comments below!

Let’s assume:

  1. The density of a Kilauea tholeiite at 1200° C is about 2.66 g/cm3 (reference: Lange and Carmichael, 1990. Reviews in Mineralogy)
  2. To simplify things, let’s use a vapor consisting only of H2O, which makes up most of the budget of volcanic volatiles (other components would be CO2, SO2, H2S, F, and Cl)
  3. The density of water vapor is around 0.0006 g/cm3 (reference)

Now we can calculate the bulk density like so:
Bulk Density (lava and gas) = [Density of Lava]*[Weight fraction of lava in the bulk] + [Density of water vapor]*[Weight fraction of water vapor in the bulk]


ρ_bulk = ρ_lava * F – ρ_H2Ovapor * (1-F)

Plugging in the above numbers gives us a weight fraction of lava of about 35%, meaning that gas must make up 65% of the Kilauea lava lake (by weight) to account for such a low density of 0.95. That’s equivalent to about 70 mol% gas and about 70 vol% gas. That’s a LOT of gas, and it raises a LOT more questions: how does the magma hold itself together without simply degassing it’s huge volumes of gas? The paper is an intriguing one whose results have certainly caught the attention of a lot of volcanologists. What do you think?

Check out the paper published in EPSL here: Continuous gravity measurements reveal a low-density lava lake at Kīlauea Volcano, Hawai‘i

Erebus science – 3D imaging with LiDAR

Our next interview is about a long running project at Erebus. Three dimensional imaging through LiDAR (Light Detection and Ranging) terrestrial scanning technology is a useful way to look at changes in the Erebus landscape – within the crater, and around the ice caves. Drea Killingsworth is the latest student to undertake the scanning work, which began with a scan of the Erebus lava lake in 2008.

Drea did her undergraduate degree at Washington State University in Pullman, Washington, and is now doing her Masters’ degree at New Mexico Tech. For the past two field seasons at Erebus, she has been working with Jed Frechette, from the LiDAR Guys in Albuquerque, New Mexico, and with Marianne Okal and Brendan Hodge, from UNAVCO. And not only has she been doing her own fieldwork but, as expedition chef, she has also been keeping the team fed!

Drea in Warren Cave

Drea in Warren Cave

Drea’s project on Erebus has included scanning the lava lake, the main crater, and two ice cave systems: Warren, and Mammoth/Cathedral, both from the inside and on the surface. At Warren Cave, by combining her work this year with previous years’ LiDAR scans and a survey carried out by Aaron Curtis and Nial Peters, Drea can get a detailed image of how things have changed over the last four years.

A 3D image of Mammoth Cave produced from this year’s LiDAR scans (D. Killingsworth)

With several years of lava lake and crater data, it is also possible to find out how the lake surface level has fluctuated, to quantify its surface area, and follow changes in the shape of both the lake and the crater. LiDAR allows us to see parts of the crater that aren’t normally visible when walking around the crater rim.

Jed with the Optech at the crater rim (December 2010)

Volcanofiles: How does LiDAR work?

Drea: The instrument fires a laser (either visible green or near-infrared light), which is reflected off surfaces and returns to the scanner. Using the amount of time it takes to receive the reflection (or backscatter) when it returns, and the known speed of the laser, it is possible to map the point on the surface where the reflection occurs. Thousands of such points (about 50,000 per second) form a point cloud in three dimensions over a ~7-minute medium-resolution scan at each scanner position. In the ice caves, scans from over 40 scan positions throughout the cave are registered together, to form a detailed 3D model of the entire accessible portion of the cave.

Scanning in Warren Cave (December 2011; Photo: D. Killingsworth)

Screenshot while scanning in Warren Cave (December 2011; Photo: D. Killingsworth)

Volcanofiles: Can you tell us about the different LIDAR instruments you use? What are some of the features you have to consider when selecting the right instrument for a particular scan?

Drea: We used three different scanners on Erebus this field season: the Leica Scanstation C10, which has a class 3R green laser (with 532 nm wavelength), the Riegl VZ400 and the Optech ILRIS-3D, which both use near infrared (with 1535 nm wavelength).

All of the scanners have similar accuracy (to < 1 cm) and cold ratings (functioning in 0° C – 40 °C) but, because of the different wavelengths, power capabilities and designs, they have different ranges and fields of view (FOV). The Leica is very useful for cave mapping because it has a wide FOV of 360° x 270° while its smaller range of < 1 m – 300 m allows for manoeuvrability in tight squeezes.

The Leica scanner inside Warren Cave – that green light it its laser! (Photo: D. Killingsworth)

The Optech has a maximum range of 1200 m but a FOV of only 40° x 40°. In the caves we are scanning entire rooms so a large field of view allows us to use fewer scan positions to scan a given area. For the lava lake, we do a series of scans from a single FOV, so aren’t limited by this smaller area. The longer range of the near infrared scanner is needed to reach the lava lake, which is 300 m down from our scan position at the crater rim.

Volcanofiles: In addition to doing scans of the caves, you have been working on combining GPS data with LIDAR scans, and matching surface features to those underground. Has this been done before? Can you describe your instrument setup for us?

Drea: The GPS does not work beneath the ice of the caves because it cannot receive satellite messages. To accurately locate the caves on the slopes of Erebus, we had to set up known survey positions outside the caves on the surface. Four known surface positions were scanned using LiDAR and a traverse was made down through the cave entrance to four more survey positions set up inside of the cave. By combining the exterior and interior scans, the location of the interior points can be extrapolated from the GPS locations of the exterior points.

Entrance to Warren Cave, through which equipment has to be lowered in (December 2011; D. Killingsworth)

This is not as easy as it sounds! Access to the caves is by rappel and all equipment must be lowered by rope and pulley. To tie the scans together, we had to set up the scanner on a level tripod at the edge of the cave entrance and scan down into the cave to hit an interior target, also on a level tripod. The scanner and target then had to be reversed, carefully lowering the scanner into the cave without moving either of the level tripods.

The LiDAR target at the entrance of Warren Cave (Photo: D. Killingsworth)

The now-known exterior and interior survey points are semi-permanent and can be used for later scans to locate the caves in relation to a coordinate system.

Volcanofiles: You’ve spent quite a bit of time in very different Erebus environments. What are some of the challenges specific to your work on the crater, and in and above the caves?

Drea: In the caves, the biggest issue we face is the change in temperature and humidity between the outside entrance and the inside of the cave. On the surface, the air is very dry and the average temperature is -20°C before wind chill. Inside the caves, temperatures are just above freezing and humidity is extremely high, causing problems with fogging of lenses in the scanners as they are moved from cold to warm areas.

These differences can also cause problems in the opposite direction as water condenses and freezes onto the equipment (and the equipment operators!). The ice caves are not formed completely of ice – the floors of the caves are volcanic rock, covered by a layer of decomposing volcanic glass particles from lava bombs. A day of ice cave scanning involves lugging the scanner, tripods, targets and other equipment through tight squeezes, and navigating and levelling equipment by the light of a headlamp.

The lava lake presents a different set of challenges. The scanners are rated to operate down to 0° C -but the average temperature at the rim, before wind chill, can be around -30° C.

LiDAR gear being carried up to and around the crater rim. (December 2010; L to R: Laura Howald, Kayla, and Jed)

We have had to come up with some very interesting make-shift ‘cosies,’ involving everything from hand warmers to garbage bags, just get our equipment to turn on. Wind and cold makes setting up and levelling tripods difficult.

Scanning the lava lake (L to R: the Riegl, Brendan, and Marianne; Photo: D. Killingsworth)

Despite any difficulties, it is such a privilege to be working on such exciting science in such an amazingly beautiful place.

Volcanofiles: Thanks to Drea for talking about her work, sharing her photos, and for the delicious cooking! To finish, here’s another image from this year to show the kind of results she’s been coming up with.

Image of the lava lake from the 2012-13 field season. One of the interesting results of this year’s LiDAR work is finding out the size of those spatter ramparts – they’re over 5 m high! (D. Killingsworth)

Amazing View of the Erebus Lava Lake

Erebus lava lake, 2011-2012 field season

This most recent image of the Erebus lava lake was taken just days ago from the crater rim, some 340 meters line-of-sight to the lake. The lake has been shrinking in size, and at this point is about 30 meters across. Photo: Clive Oppenheimer