Lava lakes mean multi-mile deep convection cells.

[halfeye]

Lava lakes are real things, but how mad are they?

There has to be a convection cell, otherwise they'd cool over and dry up, but that deep and that narrow just doesn't seem reasonable.

https://en.wikipedia.org/wiki/Lava_lake

[shawnhcorey]

Don't forget solid rock is denser than lava. When the lava cools, the rock it forms will sink where it will remelt. The surface of a lava lake would remain liquid until all the heat is out of the lake.

[factotum]

Well, yes, that's exactly what he's talking about--convection. Cooler stuff sinks, hot stuff rises to the top. He's just not sure how a proper convection cell forms in something as deep and narrow as a typical lava lake. I'm no expert on these things, but I think what might happen is that multiple convection cells would form, stacked vertically in the lake--hopefully someone who's got a more solid understanding of the physics will be along shortly to correct me.

[halfeye]

I'm not sure that cold rock is much denser than lavas, ice floating is an exception, but I'm not certain rock in lava isn't another.

The main cause of rising in lava tubes seems to be gas, either dissolved in the lava and affecting the density, or as bubbles.

I don't think the width of the lakes is a problem, they seem to be quite wide in the craters, it's the tubes up from the magma chamber that seem to be narrow.

I have kind of thought about a series of convection cells one on top of the other, but that doesn't seem that plausible to me, there's no way for gas to transfer from one cell to the next, and I don't understand a mechanism that would separate the lower from the upper and keep them separated if they were purely thermal.

[Knaight]

Well, yes, that's exactly what he's talking about--convection. Cooler stuff sinks, hot stuff rises to the top. He's just not sure how a proper convection cell forms in something as deep and narrow as a typical lava lake. I'm no expert on these things, but I think what might happen is that multiple convection cells would form, stacked vertically in the lake--hopefully someone who's got a more solid understanding of the physics will be along shortly to correct me.

Phase change and density driven movement don't need a convection cell though. As the lava cools the resulting rocks are denser, and so they sink - incredibly slowly, yes, because lava viscosity is ridiculously high at the low end and just gets more and more ridiculous, but with sinking nonetheless. Meanwhile the lava has a height determined by pressure underground, air pressure, and the height of the lava column, so that sinking mass gets replaced from below, no convection current needed.

On top of that, if the lake is gradually cooling and will gradually solidify the rocks will still form at the top first, then sink down (also the edges at the bottom, which are less visible. That can translate to it taking a long while to cool enough to remove the lake as seen from above, even as it gets shallower and shallower.

[Peelee]

I'm not sure that cold rock is much denser than lavas, ice floating is an exception, but I'm not certain rock in lava isn't another.

Water is really weird, though.

[Lord Torath]

I'm not sure that cold rock is much denser than lavas, ice floating is an exception, but I'm not certain rock in lava isn't another.

Lava is not another exception. As lava cools and solidifies, the rock it forms will be more dense than the lava it formed from*. Water molecules are polarized (positive charge at one side, negative at the other), and that polarization affects the way the liquid transitions to a solid, forming a sort of crystal lattice that decreases the density. Rock molecules don't have that type of polarization, and form different types of crystals as they solidify.

* Pumice is an exception to this, but it forms by having gas (either dissolved in high-pressure magma or from contact with water) "froth" the molten rock and cool it before it can settle back down.

[Knaight]

Lava is not another exception. As lava cools and solidifies, the rock it forms will be more dense than the lava it formed from*. Water molecules are polarized (positive charge at one side, negative at the other), and that polarization affects the way the liquid transitions to a solid, forming a sort of crystal lattice that decreases the density. Rock molecules don't have that type of polarization, and form different types of crystals as they solidify.

It's not just that water molecules are polar - lots of molecules are polar, and expansion when freezing is extremely rare even there. There's a specific combination of hydrogen bonding, the specific structure of two bonds and two free orbitals in tetrahedral form, and the size of the atoms that make up the molecule are all involved in getting this result. Even other molecules that meet most of the criteria don't display the same behavior. Ammonia, for instance, is denser as a solid than a liquid - and that's with nitrogen and oxygen having fairly similar electronegativities (nitrogen actually forms a stronger, more polar bond with hydrogen), fairly similar sizes, and with the molecules in question a shared tetrahedral geometry. One more hydrogen there though, and that peculiarity of water is gone. I suspect it also doesn't have water's bizarre phase diagram with the forward sloping solid-liquid line, but that's weirdly hard to find, mostly because it's drowned out by phase equilibrium diagrams for mixtures containing ammonia.

[halfeye]

Phase change and density driven movement don't need a convection cell though. As the lava cools the resulting rocks are denser, and so they sink - incredibly slowly, yes, because lava viscosity is ridiculously high at the low end and just gets more and more ridiculous, but with sinking nonetheless. Meanwhile the lava has a height determined by pressure underground, air pressure, and the height of the lava column, so that sinking mass gets replaced from below, no convection current needed.

That sounds a lot like a convection cell to me. All in all, a convection cell is some stuff going up and other stuff going down because of relative density. There may be some technical definition of "convection cell" that makes that not so, like there's allegedly a definintion of hardwood that makes balsa hardwood (I strongly suspect that story is silly makebelieve, but I don't absolutely know so), but stuff going up and down due to varying density due to heating sounds like a convection cell.

<edit>I just looked balsa up on Wikipedia, and what do you know, Wikipedia thinks it's a hardwood, albeit the softest one.

Being a deciduous angiosperm, balsa is classified as a hardwood despite the wood itself being very soft. It is the softest commercial hardwood.

https://en.wikipedia.org/wiki/Ochroma

[Knaight]

That sounds a lot like a convection cell to me. All in all, a convection cell is some stuff going up and other stuff going down because of relative density. There may be some technical definition of "convection cell" that makes that not so, like there's allegedly a definintion of hardwood that makes balsa hardwood (I strongly suspect that story is silly makebelieve, but I don't absolutely know so), but stuff going up and down due to varying density due to heating sounds like a convection cell.

Convection cells generally operate within the same phase, which means the stuff going up and stuff going down has to be separated - otherwise you just get rapid counter-current heat transfer and it equilibrates pretty quickly. Phase change is energy intensive enough to allow these to occupy the same space. On top of that the stuff going up isn't going up because of relative density; it's going up because of a pressure equilibrium between the fluid column and the pressurized fluid, Bernoulli equation style.

Essentially one of them needs a proper loop, one of them can be counter current flow through the same area because of different mechanisms, and that makes a big difference in terms of space needed.

[Peelee]

It's not just that water molecules are polar - lots of molecules are polar, and expansion when freezing is extremely rare even there. There's a specific combination of hydrogen bonding, the specific structure of two bonds and two free orbitals in tetrahedral form, and the size of the atoms that make up the molecule are all involved in getting this result. Even other molecules that meet most of the criteria don't display the same behavior. Ammonia, for instance, is denser as a solid than a liquid - and that's with nitrogen and oxygen having fairly similar electronegativities (nitrogen actually forms a stronger, more polar bond with hydrogen), fairly similar sizes, and with the molecules in question a shared tetrahedral geometry. One more hydrogen there though, and that peculiarity of water is gone. I suspect it also doesn't have water's bizarre phase diagram with the forward sloping solid-liquid line, but that's weirdly hard to find, mostly because it's drowned out by phase equilibrium diagrams for mixtures containing ammonia.

This sounds familiar...

Water is really weird, though.

I'm pithy.

[halfeye]

Convection cells generally operate within the same phase, which means the stuff going up and stuff going down has to be separated - otherwise you just get rapid counter-current heat transfer and it equilibrates pretty quickly. Phase change is energy intensive enough to allow these to occupy the same space. On top of that the stuff going up isn't going up because of relative density; it's going up because of a pressure equilibrium between the fluid column and the pressurized fluid, Bernoulli equation style.

Essentially one of them needs a proper loop, one of them can be counter current flow through the same area because of different mechanisms, and that makes a big difference in terms of space needed.

I'm still not groking that. The way I see it, either you have one flow up and another down, or you have solids falling through a liquid stream that's going up, and the latter seems likely to be much more slow and turbulent, and turbulence takes a lot of room. I personally suspect gases are a big part of it, I'm not sure whether as bubbles or dissolved. The pressure gradient through the height of the liquid is presumably greater than with water, because rock is denser than water.

[Knaight]

I'm still not groking that. The way I see it, either you have one flow up and another down, or you have solids falling through a liquid stream that's going up, and the latter seems likely to be much more slow and turbulent, and turbulence takes a lot of room. I personally suspect gases are a big part of it, I'm not sure whether as bubbles or dissolved. The pressure gradient through the height of the liquid is presumably greater than with water, because rock is denser than water.

Solids flowing through a liquid stream takes significantly less room than a stable convection current. As for the pressure gradient through the height of the liquid, it's greater than with water - for the surface-reaching column itself that's a direct scale with density. The pressure at the bottom of the lava lake is immense however.

As for turbulence, remember that turbulent flow is associated with high Reynolds numbers, which is calculated from velocity, viscosity, density, and characteristic length. Particularly relevant is that viscosity is inversely proportional to Reynolds number, and velocity proportional. Extremely low velocity movement of extremely high viscosity fluids like we're talking about (it's a lake, not an eruption) is almost certainly laminar, unless there is a truly huge characteristic length involved.

[Yoshi-TRM]

I'm currently building a career studying geodynamic convection modeling, so I can shed some light on this subject:

Lava lake convection works much like the larger process of terrestrial plate tectonics in miniature. Flow in lava lakes generally reaches Reynold's numbers of approximately 30, and material of basaltic viscosity generally needs Re>2000 for turbulent flow to begin. Therefore, lava lakes can be assumed to flow in a laminar fashion. Flow in lava lakes is primarily driven by the ascendance of warmer, thermally buoyant magma from a vent at depth (and yes, the lava does rise primarily because of thermal buoyancy: a density difference of 6 kg/m^3 is enough to drive magma convection in lava lakes).

Once this rising magma reaches the surface of the lava lake and makes contact with the air, radiative cooling and heat loss begins, crystallizing a surface "skin" of solidified basalt. While this solidified basalt will be more dense than the rising lava, it does not have enough negative buoyancy to sink. Instead, it will be pushed aside by the rising magma plume and travel laterally along the surface of the lava lake. During this time, the lava skin will continue to cool, thickening as it travels outward from the lava plume epicenter. This thickening at the surface will continue until gravitational stress on the basalt crust (due to it being more dense than the underlying basalt) exceeds the rigid crust's yield strength, at which point the crust will break and sink into the lava lake, much like the lithospheric-scale process of tectonic plate subduction.

This sinking, cooled basalt will pile up on the floor of the lava lake, and may end up (temporarily or permanently) sealing off all or part of the magma source vent. Convection in lava lakes is thought to occur on a local scale, and the main body of the magma chamber, while connected to the lake directly, is not expected to fully participate in convection, other than providing the rising plume that drives convection.

If you want to dig deeper and can handle some jargon and math, here's the paper I used as a source for this post:

Harris, A.J., 2008. Modeling lava lake heat loss, rheology, and convection. Geophysical Research Letters 35: LO7303. doi: 10.1029/2008GL033190.

I believe it's an open access paper, and should be clarifying for those who have a background in physics.

[Henry1122]

I don't know that cool shake is a lot denser than gammas, ice drifting is an exemption, however i am not sure shake in magma is not another, the primary driver of ascending in magma tubes is by all accounts gas, either broke up in the magma and influencing the thickness, or as air pockets...

[Yoshi-TRM]

Volcanic degassing is an important factor in volcanic eruptions, but only for rhyolitic and andesitic magmas of a viscosity high enough to trap the degassed bubbles in the flow. In those cases, the trapped bubbles do reduce the effective density of the magma and create a negative pressure gradient that drive the magma plug upward (These magmas are so thick that they hardly act like a fluid at all). However, the basaltic magmas that produce lava lakes are much less viscous, and allow any degassed volatiles to escape from the magma, therefore lessening the gas' contribution to the flow.

Source: "Volcanic Degassing", eds. Oppenheimer, Pyle, and Barclay. Geological Society Special Publication 213, 2003.