Saturday, April 9, 2011

Roger Revelle' work and what it shows about the carbon budget.

As we have seen from the previous blog entry, the buffering of the oceans works more efficiently the warmer the Earth is. In other words, the warmer the oceans, the quicker carbon dioxide is returned to the atmosphere.

An interesting facet of the ice age cycles is that ice age cycles are just as much about where carbon dioxide is stored as where water and ice are stored. During ice ages, a considerable fraction of the Earth's surface water is stored in ice sheets on the continents, and the oceans fall. The amount of water on the surface of the Earth doesn't change, but where the water is changes.

The same thing happens with carbon dioxide.

During an ice age, the amount of CO2 in the atmosphere falls by about 100 ppm from the 280-300 ppm in the atmosphere during interglacials to 180-200 ppm during the depths of ice ages. But where does the CO2 go?

The answer is that it goes into the sea. The oceans absorb it, and become slightly more acid. There is a cycle that slowly changes the PH of the oceans by ~0.03 The PH of the oceans is not just controlled by the acidic and alkaline compounds dissolved within it, but also by activity factors that mitigate (or increase) ionic concentrations. The chemistry for activity factors in the oceans is very complicated and I will not be going into it here, but the net effect is to slightly mitigate, or lessen, the swings in ionic concentrations in the oceans as the amount of CO2 increases or decreases. At least in the natural history of our recent ice ages. I will discuss it briefly at the end of this entry

Some of the results of this carbon cycle are counterintuitive. During ice ages, the atmosphere has less carbon dioxide, but the oceans are more acid. How can this be?

The answer is that the amount of carbon dioxide in the oceanic/atmospheric system remains broadly the same. At least the carbon does. During ice ages, more carbon remains in methane, which is trapped in methane hydrates in cold continental shelves. The total amount of carbon dioxide in the oceans and atmosphere does fall slightly, with more methane. But the amounts of carbon remain the same.

Another factor is that colder waters can hold more dissolved oxygen, and colder waters can support more life, if other trace minerals needed are present. Evidence does suggest that oceanic biological productivity was slightly greater during ice ages, and this may also have trapped some more gigatons of carbon.

But the main thing is that carbon dioxide accumulated in the oceans.

As I said, the PH of the oceans falling as CO2 concentrations fall in the atmosphere sounds counterintuitive, but it does make sense. The amount of carbon in the oceanic/atmospheric system remains broadly constant. If there is less in the air, there is more in the sea.

Carbon dioxide was not trapped on land in vegetation. Not only were millions of square miles covered under ice sheets, but today's temperate zones were much colder and drier. Forests are the main way life stores carbon on land, and there was much less forest area during the ice ages. There was increased land areas during ice ages as continental shelves were exposed (the area of ice-free land was about the same during ice ages as today) But it was mostly dry tundra or grasslands, and even in the tropics forests shrank and became patchy in restricted areas or river valleys.

*note* Tundra can trap large amounts of carbon. If it is wet tundra. We all know about the thawing tundra bogs bubbling with methane as they thaw. But dry tundra is different. It's just frozen without much biological productivity and not much buried biological matter to become peat infused with methane. The climate of the Earth was much drier during ice ages---so dry that in many places cold enough to form ice sheets, such as Siberia, they did not form. Dust deposits, or loess, show that at best much of North America, Europe, South America and the remaining temperate parts of Asia were at best semi-arid and most of these areas were true deserts.

Much of the interior of North America resembled the Gobi Desert. The Sand Hills of Nebraska were giant sand dunes. When Native Americans first became numerous ~15,000 years ago, the climate was already becoming milder and wetter, supporting more vegetation and life. And I am skipping over the raging question of when mankind arrived in the Americas, but there is not evidence of widespread populations more than 15,000 years ago.

*end note*

As the implications of Revelle's worked seeped through the geologic, oceanographic, biologic, and climatologic branches of science during the 1960s and 1970s, there was some thought that there might be a global carbon cycle, driven perhaps by a long cycle in volcanic activity over tens of thousands of years, that injected and withdrew carbon from the oceanic/atmospheric system. But it was never a widespread belief, and no evidence has been found for it.

The work on the Earth's carbon budget has had two main implications for climate science and global warming, one well grounded in fact, and the other more speculative.

The first one is that yes, warming out of the ice ages does come before carbon dioxide begins to rise in significant quantities. And it does make sense and doesn't invalidate carbon dioxide as the major driver in climate change.

The reason is this. The Milanković cycles determine how much solar energy falls in the polar and temperate zones, and the tropics as well. The Milanković cycles trigger a small warming, which then increases the buffering of CO2 by the oceans. The oceans, as they warm slightly, return CO2 more quickly to the atmosphere. This increases the warming, which then increases the rate CO2 is returned to the atmosphere, and it triggers an accelerating feedback. As the climate warms and becomes wetter, tropical forests and wetlands increase, increasing wetlands and methane emissions. More warming. Methane hydrates in marginal areas become unstable and release their methane. More warming. Swamps and wetlands increase in temperate and polar zones and emit more methane--more warming. The methane is quickly oxidized to CO2 and water, but the CO2 is still a warming gas, as we know.

As ice sheets shrink, the Earth's albedo decreases and the Earth retains more heat. More warming. Less known is the fact that forests are quite dark, while deserts are reflective. Compare the Amazon rain forest to the Sahara Desert in satellite pictures. Or the Siberian taiga to the rocky wastes of northern Canada's islands.

All these effects produce enough warming to bring the Earth out of an ice age. But the key is that temperatures rise first from the Milanković cycle.

This is a point that deniers try to exploit. When they do so, you can be sure that they are either ignorant of climate processes or being deliberately dishonest. Usually they are being dishonest.

Deniers argue that because the temperature began to rise before CO2 began to rise in the atmosphere that means that CO2 is not a greenhouse gas. Or that it doesn't trigger warming. Or some such thing. No. CO2 rises as a feedback to a slight temperature increase, and then vastly increases the temperature rise far beyond what Milanković cycles can do. And the albedo effects from reductions in ice and snow cover and changes in vegetation increase temperatures still further! The Milanković cycle increases temperatures a few tenths of a degree and the feedbacks from CO2 and decreasing albedo trigger the far greater temperature rises.

Milanković cycles work because the Earth is finely balanced between different climatic states, ice ages and interglacials. Milanković cycles determine how much solar radiation reaches the polar and adjacent temperate zones. Aside from changes in the eccentricity of the orbit of the Earth, Milanković cycles do not change the total amount of solar radiation reaching the earth. When changes in the axial tilt increase solar radiation in polar zones, they decrease solar radiation in tropical zones. The total amount of solar radiation reaching the Earth remains the same. Aside from the changes in the eccentricity of the orbit of the Earth around the sun, Milanković cycles wouldn't change the temperature of the Earth at all if positive feedbacks didn't come into play when more solar radiation reaches polar and adjacent temperate zones.

The fact that the Earth's climate warms so much from subtle changes in solar radiation in polar zones shows that the positive feedbacks are very strong. And that is why what we are doing to the atmosphere is so dangerous.

This brings into play climate sensitivity. We know that Milanković cycles, aside from the minor orbital eccentricity effect, can't change the total solar radiation the Earth receives. Jule Gregory Charney (1917-1981) crafted the first definitive report on climate sensitivity in 1979. You can read it here.

Climate sensitivity compares how much a given increase in a greenhouse gas, carbon dioxide in this care, to what actually happened in the climatic record. Although Charney's report is from 1979, it is definitive. The basic physics of radiation absorption by CO2 have been well understood for decades (As I have said in previous entries, it was believed until the 1940s that CO2 in the atmosphere was saturated as far as infrared radiation absorption is concerned. In other words, that adding more CO2 would not make a difference because it already absorbed all the infrared radiation it could. I will be discussing how that was proved wrong soon in an upcoming blog entry.)

From the Charney report we know that climate sensitivity greatly increases temperature swings from changes in atmospheric carbon dioxide alone would do. And has done. The question humanity faces is how powerful these positive feedbacks will be in a warming world.

Revelle's research, and other research by scientists later, is disquieting on several fronts. Back then there was 50 times as much dissolved CO2 in the oceans as in the atmosphere. It is now about 40 times as much, as atmospheric CO2 has increased so quickly. We have added nearly 3 trillion tons of carbon dioxide to the atmospheric/oceanic system. Despite the swings in oceanic chemistry between ice ages and interglacials, the oceans are already far more acid (or less alkaline) than in the ice ages. As temperatures rise, the efficiency of carbon dioxide return to the atmosphere increases. The oceans hold hundreds of trillions of tons of carbon dioxide. Could the large increases in temperature in store turn the oceans into net carbon dioxide emitters?

We don't know.

The warmer water becomes, the less gas it can dissolve. That runs counter to what we know in daily life, because we know about how solids behave in water. Cold water doesn't dissolve much sugar, and dissolves it much more slowly than hot water.

With gases it is different. Molecules vibrate faster and travel faster as temperatures warm. In fact the motion of molecules defines temperature in our daily lives. Solids dissolve more readily in liquids as the temperature rises because the break off from the surface of solids and are incorporated into the liquid.

For gases, the greater the temperature, the more rapidly the gas molecules travel, and the more easily they can escape the liquid. That is why the warmer water is, the less dissolved gas they can hold. That is why CO2 buffering becomes more efficient as temperatures rise. Rising temperatures will also decrease the amount of oxygen dissolved in water, with impacts on biological productivity. We don't know how oceanic life will adapt to warmer, more acidic conditions. Photosynthetic algae do remove a lot of CO2 and convert it to oxygen. Algae decompose on the surface when they die, and the carbon within them remains part of the open carbon system. But animals feeding on them can sink to the ocean floor when they die, as well as diatom shells. Could the warming and acidification of the oceans decrease biological productivity enough to significantly reduce the amount of carbon that settles to the sea floors?

We don't know.

The chemical processes of the Revelle Effect are well known, and assuming no major changes in biological activity, the impact of the Revelle effect is quantifiable. All measurements agree that CO2 buffering and return to the atmosphere increases by 6%-8% for every 1° C. That is a closer agreement than many in science.

But the possibility of some sort of biological threshold being reached--a cliff--where biological productivity decreases enough to increase the Revelle effect much more than expected is not something we can safely ignore.

Here are some links to the chemistry of oceanic carbon dioxide buffering:

From the IPCC

Some of the major chemical reactions in oceanic carbon dioxide buffering from Columbia University.

And a more detailed paper on the chemical reactions of oceanic carbon dioxide buffering by Chuixiang Yi, Peng Gong, Ming Xu and Ye Qi.

Sunday, April 3, 2011

Roger Revelle

The investigation into atmospheric CO2 and its interaction with the oceans is a long and complex story. Dozens of oceanographers and chemists contributed in this work, and the chemistry is also very complex. However, Roger Revelle is the central figure in this field of research. At least I consider him so. At any rate, he was the first to show that mankind's addition of CO2 to the atmosphere would not be absorbed by the oceans quickly, and was able to work out why.

This had been a question since Arrhenius. It was known since the Challenger Expedition of the 1870s that the oceans contain large amounts of CO2, and that oceans are alkaline world-wide. The Challenger expedition showed that the oceans contained ~50 times as much CO2 as the atmosphere. One of the main objections to anthropogenic global warming is that the oceans are alkaline---and should therefore absorb CO2 easily.

But proponents of anthropogenic global warming, such as Arrhenius, Alfred Wallace, and Callendar raised an interesting question. If the oceans really could absorb all the anthropogenic emissions of CO2 easily, why didn't the oceans absorb all the CO2 that is in the air now? In other words, since the oceans hold 50 times as much CO2 as the atmosphere, why didn't the oceans just absorb the 51st molecule, and then have the Earth freeze into a snowball?

This was a nagging question for oceanographers, but that scientific field was consumed by another controversy. As I wrote in a previous blog entry, the oceans have contained roughly the same salt concentration as today for billions of years. Once it was realized that the Earth was billions of years old, the main question for oceanographers was how do the oceans get rid of their salt? Even now some aspects of that question have not been solved, although we now have a broad picture of how salt can be evaporated and buried under sediments in shallow seas and estuaries. But during much of the 20th century, the salt question was the major question in oceanography.

Roger Revelle (1909-1991) was an oceanographer with the Scripps Institute of Oceanography. During the mid 1950s he was part of a team studying how fast the oceans 'turn over', the seawater at the surface sinking to the depths and deep ocean waters rising to the surface. This was suddenly an important question. The Japanese were in an uproar over nuclear testing and radioactive pollution of their Pacific fishing grounds.

There had been rising anxiety in Japan already about nuclear fallout (Japan had great nuclear anxiety in any case from their experience with the atomic bombing of Hiroshima and Nagasaki less than 10 years earlier.) In 1954 two incidents occurred.
The first, and most serious, was the irradiation of the Daigo Fukuryū Maru (q.v) and later that year, the release of the movie Gojira, which we know as Godzilla, rushed into production after the Daigo Fukuryū Maru incident.

The United States Navy rushed a study to find out how fast the oceans 'turned over' and carried radioactive fallout to the depths. Revelle and his team determined that the oceans turned over over several hundred years (that is a bit wrong---we no know that the oceans turn over in about 3,000 years). That data showed the oceans turn over fast enough to absorb and remove CO2 from anthropogenic emissions. And yes, 3,000 years is fast enough also to dissolve most CO2 in the oceans and keep atmospheric CO2 from rising much.

But Revelle went further. The question of why the oceans didn't absorb all CO2 nagged at him. And his research in the field gave him knowledge and access to a tool oceanographers didn't have. The nuclear tests in the Pacific created lots of radioactive carbon isotopes. Carbon isotopes as great as C-22 and as low as C-8 were created. Most of these had a half life of microseconds or less. But C-11 (carbon 11) has a half-life a little over 20 minutes. Revelle didn't use the C-11 created by nuclear explosions---almost all would be gone in a couple days--too quick to visit an explosion site, with lots of other longer-lived radioisotopes around. But C-11 did give him an idea---create CO2 using C-11 and see how it interacted with ocean water. It was radioactive enough to be very easily traceable, but not too fast to decay immediately. And also, in a day or two almost all the C-11 would be gone. So it wasn't a disposal hazard.

The chemistry he found was amazingly complicated. Seawater is not just salt, it is a complex soup of many thousands of chemicals dissolved within it. And it also has living organisms. So there are thousands of reactions that CO2 can make with the different chemicals in seawater.

It had been suspected that the oceans had a buffering mechanism. What Revelle found was that in many cases, CO2 combined with chemicals in the seawater and created volatile compounds that promptly evaporated back into the air. Once back in the atmosphere, the CO2 would encounter free oxygen, or be dissociated by ultraviolet light, and create CO2. When he raised CO2 concentrations slightly, to 350 ppm or 400 ppm in the atmospheric samples over the tanks of seawater, molecules containing the radioactive C-11 were returned back to the air in significant amounts, while significantly less C-11 remained in the seawater. C-11 decays too quickly for longer studies, so Revelle switched to C-14, with a half-life of 5,730 years. In 1955-56 he determined that when CO2 in the atmosphere increased, about half of what the oceans absorbed would be evaporated out via volatile organic compounds within a year.

In a paper he co-authored with Dr. Hans Seuss (1909-1993) (no, not that Dr. Seuss) Revelle wrote in a few sentences at the end that assuming that CO2 emissions stayed at 1957 levels, CO2 would rise in the atmosphere about 40% (to 440 ppm) over the next few centuries and stabilize.

This was mind-blowing. 440 ppm was a big rise! And certainly enough to warm the Earth's climate considerably! Revelle was not a climatologist or meteorologist, and did not realized the implications of what he had written. But others did.

It created a big scientific controversy. Unlike today, it played out in scientific circles and was largely unreported to the public. Objections were made, and then refuted. Perhaps the C-11 (with a half-life of 20 minutes, it is very radioactive) was killing the seawater microbes and they were releasing volatile organic compounds when they died. But research by others using C-14 quickly showed this was not the case. By 1960 Revelle's work was accepted.

One of the scientists Revelle worked with was Dr. Charles Keeling. Keeling, whom we all know, was inspired by Revelle to create his famous Mauna Loa carbon dioxide measuring observatory. It also combined with other work that I will be talking about in another blog entry, about how CO2's apparent saturation in its IR bands was not really saturated after all to show that increasing CO2 in the atmosphere would result in global warming. This has been the consensus for the past 50 years, and is the consensus now.

Revelle's guess about future CO2 concentrations in the atmosphere was a gross underestimate. He assumed that CO2 emissions would remain close to their 1957 levels. At the time, that was not a ridiculous assumption.

During the previous generation, there had been two terrible world wars. Continuous, progressive, exponential economic growth had not been the reality. In 1950, industrial production in the Soviet Union, Germany, France, Italy and Japan was lower than it was in 1913! There was no real reason to expect that the future would be different from the past. What Revelle didn't realize was that beginning in the mid 1950s the world had entered an unprecedented economic boom---with CO2 emissions more than doubling between 1957 and 1972. From 1973 to the late 1990s, emissions slowed in their rate of increase, but did not stop rising. And the growing boom in China and India has caused CO2 emissions to rise more rapidly in the past 15 years than during the pause from the mid 70s to the mid 90s.

In short CO2 emissions rose on an annual basis

1957-1973 6%
1974-1997 2%
1998-2011 3.5% (4%+ 2005-2010, despite the recent recession)

The combination of unprecedented economic growth, Revelle's CO2 chemistry work, Keeling's Curve, the discovery that IR absorption by CO2 was not saturated led by 1965 to a scientific conference in Boulder, CO about anthropogenic global warming--the first scientific conference with that as the main topic. I'll blog about that soon.

Revelle's work was also incomplete. He identified some of the main chemical pathways CO2 dissolved in the oceans returns to the atmosphere in volatile compounds. But many other oceanographers and chemists have been working since to identify other pathways--there are tens of thousands of them! Individually, most are trivial, but together they make a significant fraction of carbon dioxide atmospheric return. They vary by differences in temperature, local differences in chemicals dissolved in the oceans, and of course, different organisms present in the surface waters.

Also, some of the major chemical reactions that return CO2 to the atmosphere had already been discovered during the 1940s and 1930s. But it wasn't realized that they were a major part of the flux of carbon between the atmosphere and oceans. What Revelle largely did was discover that these reactions returned a large amount of carbon to the atmosphere quickly. There is a distinction between discovering a chemical reaction and discovering its magnitude and importance.

One alarming thing is that this return of carbon dioxide to the atmosphere is a powerful feedback. The warmer surface temperatures rise, the more quickly organic compounds evaporate and return to the atmosphere. Also, as temperatures rise, more compounds become able to evaporate.

Research shows, and is unanimous in agreement, that for each 1 C° rise in the surface ocean temperature, the return rate of CO2 to the atmosphere increases by 7%.

When I said unanimous, I was not quite right. All models show an increase from 6-8% per degree Celsius. What's really interesting is that if the increase is closer to 8% then more CO2 will be returned to the atmosphere, warming it faster, and cause more CO2 to be emitted by the oceans. This has big implications on temperature and oceanic acidification. It results in a damned either way situation. A slightly faster CO2 evaporation rate will result in faster warming, but less oceanic acidification. A slower CO2 evaporation rate will result in less warming, but more acidification. A difference between 6.2% and 7.8% is pretty close agreement, but running the calculations out to 2100 and beyond results in quite different states for the atmosphere and the oceans. As always, we need more research!

Roger Revelle

I will add further information in this blog entry later on some of the chemical pathways of CO2 return to the atmosphere.