Friday, April 24, 2015

Volcanic Eruptions and Climate

Many people have been awestruck by the amazing images and video coming out of Chile this week, where the Calbuco volcano has dramatically erupted.  Here's an example I lifted from weather.com:
(David Cortes Serey/AFP/Getty Images)
The dramatic ash plumes from multiple eruptive phases have been visible on satellite, as these blog posts from the CIMSS satellite center illustrate.
This eruption has been throwing a lot of ash into the air and undoubtedly is emitting a lot of gasses, including greenhouse gasses like carbon dioxide, and producing suspended particles like sulfate aerosols.  These emissions can noticeably alter global temperatures if the eruption is large enough and in the right place.  How does this work?  And is the Calbuco eruption enough to change our climate trajectory?

You might think that by emitting a lot of carbon dioxide, the effect of a large volcanic eruption would be to warm the planet on average since, after all, carbon dioxide is a greenhouse gas.  However, even if you sum up all of the carbon dioxide emissions from all the volcanoes on earth, it is relatively small compared to other sources of carbon dioxide (like human beings).  Using USGS estimates, global volcanic emissions of CO2 amount to an average of 0.26 gigatons of carbon per year.  In comparison, just cars and trucks alone are estimated to emit 3.0 gigatons of carbon per year.  The EPA estimates total global human emissions of carbon to be some 32 gigatons of carbon per year.  What about a single BIG volcanic eruption?  That same USGS page has an estimate of the carbon emissions from the Mount Pinatubo eruption in 1991, one of the largest eruptions in recent history.  Their estimated emissions from that single eruption amount to 0.05 gigatons---much smaller than human emissions.  As such, we really can't detect any warming due to volcanic emissions...

In fact, we tend to see strong cooling of the atmosphere in the years following a large eruption.  Here's a reconstruction of the global mean temperature from the Berkeley Earth group, using mostly direct weather observations.
Their estimates get very noisy before the early 1800s because we had very few weather observations globally at that time.  You can also see a dark line on top of their fit, which is an estimate of the global mean temperature including only two things to force it: carbon dioxide changes and volcanic eruptions.  They've identified several of the larger eruptions that had a noticeable impact on the temperature.  Note that these eruptions are associated with sharp spikes in the temperature.  For a period of a few years after each eruption, the global temperature cooled.  Why is that?

It mainly has to do with another big emission from volcanoes---sulfur dioxide, which undergoes chemical reactions in the atmosphere to produce sulfate aerosols.  Sulfate aerosols are very good reflectors of solar radiation, so when they get lofted high up into the atmosphere, they reflect away some of the incoming sunlight.  This, in turn, leads to a temporary cooling of the earth's surface below while those aerosols are up there.  There are a number of other ways in which these aerosols affect the climate, but this reflection is one of the big ones.

Which brings us to another question---if volcanoes emit a lot of sulfur dioxide, why do only some volcanoes seem to have this big effect on climate and others don't?  Not all volcanic eruptions are created equal---some just emit more sulfur than others.  But location of the eruption also plays a big role.  I've plotted a map of the locations of all the eruptions listed in the above figure and a few others for reference.

On that map, the red triangles are the locations of eruptions that are known to have produced significant global cooling.  Notice anything about their locations?  They are all located in the tropics, between 20 degrees south and 20 degrees north.  It turns out that large tropical volcanic eruptions are more likely to produce cooling on the earth's surface.

To get a prolonged period of cooling, you have to get the sulfur lofted high enough into the atmosphere, you want it to cover a relatively large area of the globe, and you want it to stay there for a long period of time.  Volcanoes are typically capable of lofting emissions into the stratosphere, which is high enough for them to be effective at reflecting incoming sunlight and staying aloft a long time (i.e., they're lofted above the dynamic weather systems and thunderstorms that could efficiently "rain out" all of the sulfur).  The stratosphere has its own circulation, called the "Brewer-Dobson Circulation".  It's actually a fairly simple arrangement:
Deep, persistent tropical convection below helps lead to rising motion in the tropical stratosphere.  This air then slowly moves north and south, away from the equator, and then sinks again in the mid to upper latitudes.  The entire process is somewhat slow, taking a couple years to fully cycle.

So, if you wanted to inject sulfur into the stratosphere in a place where you would know that it would be lifted up high, spread out across the globe and stay there for a few years, you would want it to be injected in the tropics.  In the mid to upper latitudes, the general motion in the stratosphere is downward, so it would be hard for any injected sulfur to spread out much or stay aloft for very long.  This is why large tropical eruptions usually produce the biggest impact on climate.

Occasionally an extratropical volcano can make a difference.  Note in the temperature record above there was a strong response to the Laki volcano in the late 1700s.  Now, the Laki volcano is in Iceland, well away from the tropics.  Why does it show such a large impact?  For one, those temperature records are based on where we had instruments, and in the late 1700s that was mostly in Europe which is directly downstream from the Laki volcano.  Still, the effect was significant and felt across the northern hemisphere, including in the United States.  One difference is that the Laki eruption was very long in duration---it erupted nearly continuously for eight months.  That's a long time to be pumping sulfur into the atmosphere, regardless of where it's being emitted.  So it is possible for an extratropical volcano to have an observable, long-term effect, but it doesn't happen often.

Which brings us back to the Calbuco eruption this week.  Calbuco is at 41 degrees south latitude, well outside the tropics.  It's worth noting that other nearby volcanoes that have erupted in recent years (like Villarica and Caulle Cordon) have not caused observable global temperature decreases; their emissions mainly stayed in the far Southern Hemisphere. So, it's unlikely we'll see a global temperature change due to this eruption, but it's still fascinating to watch.