One of the reasons I do this, is because as I was growing up, having an interest in things Geophysical/Astrophysical, there was always a search for the “wow” factor. Not everyone’s “wow” sense is geared the same… and in some cases, the scale of stuff that people are familiar with has a lot to say about how they perceive the “wowness” of it. Grabbing that meaningful nugget of data, or of a concept that totally re-vamps your experience level is way cool. It changes your world in incremental steps… or at least how you look at it.
The difficult thing is finding usable data to ruminate on, or to have some esoteric thought wrapped up in equally esoteric language. (see “e-folding” from the last Ruminerian) It’s not that people are intentionally obtuse with the language or ideas, it’s that there is a lot of technical jargon that develops out of any technical field. (How many of you know that a “gyraline modulator” is?) This post, and the others that I have written, are geared towards the person who seeks to find out more.
This, is more.
Before I continue, a bit about SO2. SO2, Sulfur Dioxide, is a volcanic gas. It reacts with water to form H2SO4, also known as sulfuric acid. Take away the water and you get sulfate, SO4. The reaction in the atmosphere goes something like this:
SO2 1 OH 1 3H2O ═> H2SO4 (1) 1 HO2
In Ruminerian IV, I ended on a pretty interesting graphic. (well, I thought it was)
It is derived from “Stratospheric Loading of Sulfur from Explosive Volcanic Eruptions” Bluth et al (1997). This plot shows the e-folding times for SO2 to sulfate conversation, and then for sulfate removal from the stratosphere.
Where this particular model fails horribly, is in how it treats the SO2 input. It assumes one sizable lump of SO2 injected to the stratosphere. Odds are that many volcanic eruptions are not going to be just one quick blast of SO2 and the show is over. For the sake of modeling influx to the stratosphere, you can probably get away with it… but you have to always be aware that this ideal treatment is going to be incomplete. Another line of thinking is that an established vertical plume can eventually propel the gases past the tropopause if it persists long enough and has enough strength.
Revising that plot and looking at the peaks in it and the narrative that went along with it, moderate sized SO2 releases have a sulfate peak about 2.07 months after the event. In winter (for whatever hemisphere) this conversion rate can be slowed by up to 20% (Bluth et al 1997) giving a peak at about 2.27 months. (30 day months). For large eruptions the curves yield 2.78 months and 2.99 months (winter).
Okay, a lot of stuff about … something. But why?
Sulfate is an aerosol. “a suspension of fine solid particles or liquid droplets in a gas.” Smoke from a fire is an aerosol. Clouds and fog are aerosols. That brown crud drifting off of the iron pellet plant in Bahrain is an aerosol. That massive black cloud that spurted out of the stacks on a steam powered Cruiser in Mayport Florida, that then settled on the Quarterdeck of the spiffy new Gas Turbine powered Cruiser moored on the other pier… that was an aerosol. (trashed a lot of summer white uniforms as the partially burnt diesel precipitated out) Even that gunky haze that you can see over New York from 30 km at sea is an aerosol (the same for LA by the way). Fine particles suspended in a gas.
In some way form or fashion, they all act upon light that is traveling through them. Reflection, scattering, refraction, absorption. You name it. If the particles are quite small, the effects are generally in the category or Rayleigh scattering. That’s what makes those vivid sunsets or the sky blue. If they are about the size of the light’s wavelength, you get Mie scattering. That’s the effect that makes the clouds appear white.
Now I deviate. As I was growing up, I used to listen to the radio. At night I could pull in stations from hundreds of miles away… during the day time, only the closer stations would show up. I had a great uncle who was into Ham radio, and he took a partial interest in my fascination with all things electric. He gave me a copy of an ARRL handbook. I never got a ham license, but I learned everything in that book… and then some. (I wound up specializing in Electronic Warfare in the military). That late night effect that allowed me to hear stations far away, is caused by ionized layers of the atmosphere.. specifically, the ionosphere.
There are three principle layers involved, the D layer which is strongest during the day, mainly absorbs radio waves. Above that, the E layer, present during the day, acts to reflect radio waves. And above that, the F layer. It’s always present, and in the day time it tends to split into the F1 and F2 layers. This is the one that causes most of your long haul radio intercepts late at night. In CB jargon, its called “skip” because that is what the signal is doing… bouncing off of the ionosphere, back to Earth, and could bounce a second time repeating the process. (no, this is not the Van Allen radiation belts, that is something totally different) “Anomalous propagation” (the real term) can occur due to a number of causes… the sun is the main driver, but meteor showers can energize the various layers also.
This rather busy plot gives you an idea of where everything is at. Note that the vertical scale is logarithmic. Just for reference, I’ve place a few altitude events and items in there for reference… such as Felix Baumgartner’s leap altitude, and the record holder prior to that, Joseph Kittinger. Also noted are high and low altitude of the ISS, and the elevation that Mt Pinatubo erupted to during it’s strongest phase.
Now for something totally new to me. Christian Junge, Atmospheric research pioneer, released a paper in 1961 announcing he discovery of the stratospheric aerosol layer. This region is the area where the nitty gritty happens with respect to volcanoes and the climate. I have spent a few days tracking down good info on the location and the make up of the Junge layer plus some of what goes on there.
It resides at about 17 to 30 km in altitude, depending on conditions. This layer is where sulfate occurs when it forms. How dense it is depends on a number of factors… one of the strongest factors is volcanic activity. A volcano can load this layer quite quickly, and as you saw from the e-fold plot, the material can stick around for a while. One interesting thing that I found out was that the Junge layer can occur at distinct elevation nodes. During heavy volcanic activity, there can be an upper and lower node. Eventually it all settles to that lower range over a period of several months.
Yet another interesting thing about it, is that it is usually there… whether the volcanoes are running or not. There is always a background level of sulfate. This is where it gets pretty wild.
At one time, it was thought that SO2 in the atmosphere (troposphere) could drift up and cause this persistent layer. With the way SO2 plagued Los Angeles, you can bet your bottom dollar that some people were chomping at the bit to blame modern society. Many of us have sat around the Café or over at Eruptions or Jon’s Blog oogleing the OMI or TOMS SO2 vertical column data. Some of the plumes we have seen are valid volcanic events, many are not. Beijing almost always has a plume drifting out over the Pacific, one plume that was seen was slap dab in the middle of nowhere… until we found an industrial facility in the Northern reaches of Russia. (Siberian Traps fans were enthralled at possible implications) Of course Europe and The US are producers… even with the emissions standards. Couple those with the bona-fide plumes we have seen, Tolbachik, Grímsvötn (for some reason a huge plume formed over Iceland two weeks after the eruption), Puyehue-Cordón Caulle … you would think that there would be a huge effect in the Sulfate formation.
It’s not gonna happen. At least not from SO2. (Note, Grímsvötn easily punched the tropopause with it’s eruption, I’m referring to the later plume.)
SO2 is a highly reactive gas. As you can see from that plot that it only takes about two and half months for it to react out to below about 10% of what was emitted. (and that’s at the stratospheres rates, it’s probably faster in the tropopause where water vapor is quite abundant) SO2 just does not have the staying power to wind up in the stratosphere due to riding the air currents. In fact, some researchers have studied the SO2 concentration vs altitude and come up with something like this:
Don’t be fooled by that really high correlation coefficient. That’s just how well the curve fit an averaged set of multiple curves generated from the data in Meixner (1984). Think of it as a general guideline and nothing more. What is important is that SO2 trails off quite rapidly with height. It just doesn’t have the staying power.
Before I press on I would like to make mention that the Atmosphere is a highly complex dynamic system. We know a few things about it, such as large scale circulation patterns, but with as much as we do know, you can bet your bottom dollar there is just as much if not more, that is not known. Here is a tidbit that most people don’t know.
Notice the red up arrows. These are the regions where low pressure systems dominate. As air rises, the surrounding air flows inward to fill the space. Where the blue down arrows are at, high pressure systems dominate. Overall horizontal circulation of the individual lows and highs is driven by the Coriolis effect … which is due to residual angular momentum from where the air is coming from. In the Northern hemisphere, Lows rotate clockwise, highs counter clockwise (as viewed from the top). In the Southern Hemisphere, the reverse applies.
Across the world, there are regions that have what are known as “semi-permanent” features… the Icelandic Low is one, another is the Bermuda/Azores high (depending on where it happens to be at) There is no hard and fast rule about what latitude something is going to be at, this is just a generalized rendition of where the boundary regions are at.
Notice that not only is the tropopause usually low over Iceland… the general circulation pattern is lofting air to the tropopause. This also applies to the Kamchatka peninsula which is also not too far south of the Polar cell boundary. (The same for the Aleutian island volcanoes)
Now we move on to the reason for the post… (hell of a lead in eh?)
Two of the more significant volcanic eruption styles… are the massive VEI-6+ explosive eruptions… and the not so explosive VEI-6+ flood basalt events. Of the two, one would think that the huge lava flow events wouldn’t have much of an opportunity to loft stuff above the tropopause. We have already seen that SO2 doesn’t have much staying power, and tends to be scavenged out pretty quickly in the area where most of the water vapor is at… down here in our little realm of existence in the troposphere.
Yet there is a way that massive flood basalts can easily contribute to the Stratospheric Aerosol Layer (another name commonly used for the Junge layer.)
It comes in the form of a little molecule called Carbonyl Sulfide. OCS.
Carbonyl Sulfide can be considered as an intermediate between CO2 (carbon dioxide) and CS2 (carbon disulfide). It has a really long persistence in the troposphere… accounting for up to 80% of the sulfur gases present. I’ve seen residence times ranging from 4 years, to 7.1 to 11 years. Basically, it doesn’t like to react. This gives it time to wander throughout the different atmospheric flows and become well distributed. And a really interesting thing happens when it is hit with ultraviolet light of about 200 to 270 angstroms. (UV-C). The bonds begin to break and it dissociates. Once it does that it forms CO2 and S2… the S2 then reacting with the H2O and OH radicals forming H2SO4… the sulfate.
Hello aerosol haze.
Okay… we have a mechanism not involving SO2 that can make sulfate. Some of the largest sources are the oceans, fossil fuel usage, even the making of concrete. (via a catalytic reaction). In general, the background level of the aerosol is not that big of a deal unless something radically increases the amount there… like an large explosive volcano. Or, a really big flood basalt event. (Eldga, Skaftar, Krafla, Þjórsá lava or any of the huge flow fields that pop up in Iceland from time to time)
Remember, OCS is ultra stable in the troposphere, but once it gets to the stratosphere where the UV-C can get at it, hello Aerosol Haze.
Enjoy!
GEOLURKING
This article has gone through about 4 revisions before I actually wrote it. I hope you were able to read it without dozing off. If you did, it’s no big deal. I doze off reading what I think is really interesting stuff from time to time.
Note: The energy in a photon packet (or wave packet depending on how you look at it) is determined by it’s wavelength. The shorter the wavelength, the more energy per packet. 200 and 270 angstroms are the wavelengths that OCS best dissociated at when exposed to it. I don’t know why, but the ratio of the length of the two bonds is pretty close to the ratio of the differences in those two wavelengths. It’s about 1731 times the length of the bond in both cases. Why? I don’t know. I just found it interesting.
As noted there were about four iterations of this post before I actually wrote it. Here is some stuff didn’t make it in, but deserves to be mentioned. (well, since I already did the plots for it)
Stepping back from Carbonyl Sulfide… and back to Sulfur Dioxide and the usual way that volcanoes can affect the Junge layer. NASA GISS has a few models they play with. One is a compilation of the “Stratospheric Aerosol Optical Thickness” (What they have against Christian Junge is beyond me, the Junge layer is where most of this stuff is at.) One of the data products is something called the “Tau Line” and represents the average thickness at 550 nm. (that’s pretty much in the middle of “green” light at 520–570 nm.)
http://data.giss.nasa.gov/modelforce/strataer/
For those of you who are chomping at the bit over the Roaring 40′s, nothing really shows up, but they have some nice graphic of sulfate blooms and spreads for various volcanoes over the years. They also have that tau line data set.
First, let’s look at some of the more recent party poppers.
This is a plot of the Tau Line (Aerosol Optical Depth) in relation to a few volcanoes that have gone off recently. Notice that the hemisphere that received the brunt of the sulfate load depends on what volcano erupted.
Also notice that the shape of the curve pretty much follows the decay rate. The lag time between the eruption and the sulfate peak is noted. For the most part, it follows the growth and decay curves at the beginning of the post. Personally, I thought that was pretty neat.
So.. how do they compare to some known atmosphere shakers? Volcanoes such as El Chichón or Pinatubo?
El Chichón, at 17.36°N, had most of it’s effect in the Northern Hemisphere. According to Wikipedia, the Mauna Loa observatory registered a larger drop in Solar radiation transmittance than Pinatubo. However, Pinatubo (15.14°N) had a longer duration of it’s drop. It also had better coupling to both hemispheres. It also had 4.8 times the output of bulk tephra (using GVP Data).
Comparing them with those diminutive spikes over at the right hand side of the plot… those are the ones shown in the previous plot.
How is that for perspective?
Analyses and visualizations used in this [study/paper/presentation] were produced with the Giovanni online data system, developed and maintained by the NASA GES DISC. (Specifically, the tropopause elevation data)
http://disc.sci.gsfc.nasa.gov/giovanni/overview/index.html#maincontent
“Stratospheric Loading of Sulfur from Explosive Volcanic Eruptions” Bluth et al (1997)
http://www.geo.mtu.edu/~raman/papers/BluthJG.pdf
“The role of carbonyl sulphide as a source of stratospheric sulphate aerosol and its impact on climate” Brühl et al (2012)
http://www.atmos-chem-phys.net/12/1239/2012/acp-12-1239-2012.html
“The Vertical Sulfur Dioxide Distribution at the Tropopause Level” Meixner (1984)
http://link.springer.com/article/10.1007%2FBF00114130?LI=true#page-1
“A ThreeDimensional Global Model Study of Carbonyl Sulfide in the Troposphere and the Lower Stratosphere” Kjellström (1998)
http://link.springer.com/article/10.1023%2FA%3A1005976511096?LI=true#page-1
