WASHINGTON, D.C. – The world’s deserts may be storing some of the climate-changing carbon dioxide emitted by human activities, a new study suggests. Massive aquifers underneath deserts could hold more carbon than all the plants on land, according to the new research.
Credit: Yan Li
Humans add carbon dioxide to the atmosphere through fossil fuel combustion and deforestation. About 40 percent of this carbon stays in the atmosphere and roughly 30 percent enters the ocean, according to the University Corporation for Atmospheric Research. Scientists thought the remaining carbon was taken up by plants on land, but measurements show plants don’t absorb all of the leftover carbon. Scientists have been searching for a place on land where the additional carbon is being stored–the so-called “missing carbon sink.”
The new study suggests some of this carbon may be disappearing underneath the world’s deserts – a process exacerbated by irrigation. Scientists examining the flow of water through a Chinese desert found that carbon from the atmosphere is being absorbed by crops, released into the soil and transported underground in groundwater–a process that picked up when farming entered the region 2,000 years ago.
Underground aquifers store the dissolved carbon deep below the desert where it can’t escape back to the atmosphere, according to the new study.
The new study estimates that because of agriculture roughly 14 times more carbon than previously thought could be entering these underground desert aquifers every year. These underground pools that taken together cover an area the size of North America may account for at least a portion of the “missing carbon sink” for which scientists have been searching.
“The carbon is stored in these geological structures covered by thick layers of sand, and it may never return to the atmosphere,” said Yan Li, a desert biogeochemist with the Chinese Academy of Sciences in Urumqi, Xinjiang, and lead author of the study accepted for publication in Geophysical Research Letters, a journal of the American Geophysical Union. “It is basically a one-way trip.”
Knowing the locations of carbon sinks could improve models used to predict future climate change and enhance calculations of the Earth’s carbon budget, or the amount of fossil fuels humans can burn without causing major changes in the Earth’s temperature, according to the study’s authors.
Although there are most likely many missing carbon sinks around the world, desert aquifers could be important ones, said Michael Allen, a soil ecologist from the Center for Conservation Biology at the University of California-Riverside who was not an author on the new study.
If farmers and water managers understand the role heavily-irrigated inland deserts play in storing the world’s carbon, they may be able to alter how much carbon enters these underground reserves, he said.
“This means [managers] can take practical steps that could play a role in addressing carbon budgets,” said Allen.
Examining desert water
To find out where deserts tucked away the extra carbon, Li and his colleagues analyzed water samples from the Tarim Basin, a Venezuela-sized valley in China’s Xinjiang region. Water draining from rivers in the surrounding mountains support farms that edge the desert in the center of the basin.
The researchers measured the amount of carbon in each water sample and calculated the age of the carbon to figure out how long the water had been in the ground.
The study shows the amount of carbon dioxide dissolved in the water doubles as it filters through irrigated fields. The scientists suggest carbon dioxide in the air is taken up by the desert crops. Some of this carbon is released into the soil through the plant’s roots. At the same time, microbes also add carbon dioxide to the soil when they break down sugars in the dirt. In a dry desert, this gas would work its way out of the soil into the air. But on arid farms, the carbon dioxide emitted by the roots and microbes is picked up by irrigation water, according to the new study.
In these dry regions, where water is scarce, farmers over-irrigate their land to protect their crops from salts that are left behind when water used for farming evaporates. Over-irrigating washes these salts, along with carbon dioxide that is dissolved in the water, deeper into the earth, according to the new study.
Although this process of carbon burial occurs naturally, the scientists estimate that the amount of carbon disappearing under the Tarim Desert each year is almost 12 times higher because of agriculture. They found that the amount of carbon entering the desert aquifer in the Tarim Desert jumped around the time the Silk Road, which opened the region to farming, begin to flourish.
After the carbon-rich water flows down into the aquifer near the farms and rivers, it moves sideways toward the middle of the desert, a process that takes roughly 10,000 years.
Any carbon dissolved in the water stays underground as it makes its way through the aquifer to the center of the desert, where it remains for thousands of years, according to the new study.
Estimating carbon storage
Based on the various rates that carbon entered the desert throughout history, the study’s authors estimate 20 billion metric tons (22 billion U.S. tons) of carbon is stored underneath the Tarim Basin desert, dissolved in an aquifer that contains roughly 10 times the amount of water held in the North American Great Lakes.
The study’s authors approximate the world’s desert aquifers contain roughly 1 trillion metric tons (1 trillion U.S. tons) of carbon–about a quarter more than the amount stored in living plants on land.
Li said more information about water movement patterns and carbon measurements from other desert basins are needed to improve the estimate of carbon stored underneath deserts around the globe.
Allen said the new study is “an early foray” into this research area. “It is as much a call for further research as a definitive final answer,” he said.
###
The American Geophysical Union is dedicated to advancing the Earth and space sciences for the benefit of humanity through its scholarly publications, conferences, and outreach programs. AGU is a not-for-profit, professional, scientific organization representing more than 60,000 members in 139 countries. Join the conversation on Facebook, Twitter, YouTube, and our other social media channels.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Richard & Ferdinand,
You have provided a great deal of information, most of it understandable but perhaps some above my pay grade. Thank you, I very much enjoy this and I am learning a great deal. I choose to ignor the silly insults, you both seem passionate about this subject.
If I can be allowed a business analogy; I have worked as a consultant on a large M&A transaction, the sale ( carve out) of Allison Transmission from General Motors as project manager. Allison made more profit in that year than the rest of GM. It was said that Allison was responsible for all of GMs profit that year because without it there would be no profit for GM. This is similar to the mass balance argument.
But the above statement would be incorrect. There were other divisions in GM that were profitable, and others with even larger losses. Allison Transmission was just a part of a larger business, not the only answer to the source of GMs profits. Also, Allison Transmission shared numerous resources with GM. As a standalone entity the profits of Allison would be very different.
What I have learned here of the complexities of CO2 emissions and up takes suggest a more complex answer to the human contribution of CO2 than mass balance.
I also re-read Jaworoski’s paper. Did Calender cherry pick the CO2 measurements from the 19th and 20th Century? Was there a valid reason why so few of the available readings were used by Calender? Also, Jaworoski’s bio indicates he did have a great deal of experience with ice cores as well as experience with climate and CO2 measurements.
Gregory,
My response is below this one, except about Jaworowski.
As far as I can tell, he did a lot of work on ice cores specifically into the distributions of metals (for his work in the fall out of the radionucleides from the Tsjernobyl disaster) in the ice cores. Metal (ions) can more easily distribute through the ice matrix itself, which is impossible for CO2. Thus not comparable.
I didn’t find any work from him that he ever performed any investigation in CO2 measurements (in ice cores or else), only wrote a lot of comments on the (possible) errors in CO2 readings from ice cores, published in 1992. In 1996, Etheridge e.a. responded by publishing the results of 3 high accumulation ice cores at Law Dome, which resolved many of the objections from Jaworowski. Despite that, he repeated all his objections in 2004 and 2007, even the ones which were refuted by Etheridge e.a.
His objections contain completely wrong statements like the possibility of low CO2 levels in the core, because of migration from low (inside) to high (outside) levels and his use of the wrong column in the table of Neftel to “prove” that there was an “arbitrary shift” in data of the ice core to match the timing of the Mauna Loa data. He used the column of the age of the ice in the core instead of the age of the gas bubbles. As the pores during densification from snow to ice stay open for a long time, the average gas age in the bubbles is much younger than of the surrounding ice…
I have never met the late Dr. Jaworowski, he seemed to be a very nice person, but his opinions about ice core CO2 were and are completely out of reality.
About Callendar: he was not satisfied with the huge variability in CO2 levels taken by different researchers at different places over a year. What he did was using several a priori rules to wade through the bulk of the measurements. Some rules had merit: no CO2 measurements taken for agricultural purposes, within 10% of the bulk of the measurements, others were more controversial.
Anyway, what he managed to produce was a rather smooth curve, which decennia later was confirmed by high resolution ice cores over the same periods.
That many of the historical (wet chemical) measurements did only represent the (huge) local CO2 variability over land is clear from another long time direct discussions I had with the late Ernst Beck:
http://www.ferdinand-engelbeen.be/klimaat/beck_data.html
Gregory,
There are some parallels between a business and the carbon cycle and some differences.
The main difference is in the net yearly result, which for the carbon cycle was quite constant over at least 800,000 years and only influenced by temperature. Other influences like forest fires, volcanoes,… are not detectable within the resolution of the measurements.
It is like looking at the total GM group and seeing that the year by year gain is in exact ratio to the turnover. We can call that the “standard response” of the group.
Agreed that one can isolate any part of the group which has its own costs and gain. The difference here is that the outside player has no interconnection whatever with the rest of the group, has zero costs and only contributes to the gain. Only a few others of the group have the same one way addition (even not absolutely): volcanoes and rock weathering, but these are very small compared to the human contribution.
Moreover, there is a mechanism at work (which is also the case in the financial world) that an increase in gain above the “standard response”, is in part removed to give an increase in capital to all members of the group (- extra investments -): each of the big members (oceans and vegetation) shows an extra gain in content.
Thus while it is true that in theory one can find other ways of distribution that may be (in part) responsible for the increase, the main natural players are more sink than source (thus constantly loosing money) and the few one-way players are too small to have much influence. Only humans give huge one-way additions, increasing about every year.
Ferdinand:
You say
Only humans give huge one-way additions, increasing about every year.
No, volcanoes are not “humans” and they give “huge one-way additions” that are poorly quantified in magnitude and variability.
Please see my above answer to Gregory here: it says
Perhaps you would state the magnitude of CO2 emissions from coal seam fires and peat fires, the variability of this magnitude, and how you achieved this quantification?
Richard
Richard,
9. Production of CO2 from volcanoes (by eruption and gas leakage).
Extrapolated from several (continuous) field tests around active volcanoes (e.g. Mount Etna in Italy): less than 1% of human emissions.
And a nice explanation why volcanoes are peanuts compared to human emissions in EOS:
http://www.readbag.com/agu-pubs-pdf-2011eo240001
10. Natural forest fires, coal seam fires and peat fires.
Natural forest fires don’t contribute to the carbon cycle, just give a momentary peak. They release CO2 which was captured only a few years to a few decennia before out of the atmosphere. Thus are “carbon neutral” within the time boundary of that definition.
Peat fires are counted as “fossil” fuels if humans did use them as fuel (or started the fires as is the case in Indonesia during droughts). Thus add to the human influence.
Don’t know how much peat fires are natural, but should be counted as one-way source as the time frame between uptake and release is hundreds to thousands of years. What I found is:
http://www.wetlands.org/Whatarewetlands/Peatlands/Carbonemissionsfrompeatlands/tabid/2738/Default.aspx
Altogether global CO2 emissions amount to at least 2,000 million tonnes annualy, equivalent to 5% of the global fossil fuel emissions.
Spontaneous coal combustion (mostly during human exploration) does happen. Some figures, extrapolated from a worst case in China and extrapolated worldwide as a ratio between coal seam fires and mined coal:
http://www.ltu.se/cms_fs/1.5035!/coal%20fire%20report%20-%20final.pdf
the coal fires worldwide account for 0.31% of World CO2 emission in 2002.
Thus volcanoes, peat and coal fires together may present 7%, give it 10% to have some tolerance, of human emissions.
The main point is not how much these represent, the point is that there is a CO2 increase in the atmosphere which mimics the CO2 emissions from humans: both a factor 4 between 1959 and 2012.
There is no sign that volcanic eruptions increased (to the contrary) or that there is a 4-fold increase in worldwide peat fires (the increase in coal seam fires is too small) over that period.
Richard & Ferdinand,
Is there a layman’s explanation for why, if I want an ice core sample of today’s air, I must wait 83 years to before I can get it?
Gregory,
Depends of how much snow falls down each year. At Law Dome they have 1.2 m ice equivalent or over 2 m of fresh snow per year. That makes that at bubble closing depth, that is at about 72 m depth, the ice is already 40 years old, but the average gas composition is only 7 years older than in the atmosphere, due to permanent exchanges with the atmosphere, until the pores are too small.
Then it takes a few years more to close all remaining open pores. That makes that the first fully closed ice is about 45 years old (at 80 m depth) and in that case the average air age is 10 years older than the atmosphere and represents the atmosphere with a mix of 5-15 years older air than in the atmosphere.
Thus in the case of Law Dome, if you drill a hole today and take a sample of the first layer of fully closed ice, the average gas age is from only 10 years ago.
In between you can measure the CO2 levels of the air in the still open bubbles/pores. That shows a smooth change from near the surface to bubble closing depth:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/law_dome_firn.jpg
Thus, depending of the snow accumulation rate, you can find todays air after 10 years in the ice core of Law Dome, 83 years at the Siple Dome ice core and a few thousands of years at Vostok, where only a few mm per year of ice equivalent snow falls down…
Some more background information can be found at:
http://courses.washington.edu/proxies/GHG.pdf
Gregory Lawn
You ask
Yes, there is.
The ‘adjustment’ is made to obtain an apparent agreement of the ice core data to the Mauna Loa data.
Such ‘adjustments’ of data to make fit with what is ‘wanted’ are common throughout ‘climate science’.
Of course. in physical reality there is no way that the air in the ice would all move up a distance through the ice equivalent to 83 years (or 10 years) of ice formation whatever the time taken for ice closure. In this thread I explained what actually happens in the firn here. (And the 10 year figure is not relevant because the ancient CO2 measurements are not obtained from Law Dome.)
To obtain a difference of 83 years between the ages of the ice and its entrained air would require the ice to take 166 years to seal with the average air composition of that 166 years being at the center height of the ice which solidified over that 166 years.
The Mauna Loa data has been obtained since 1958 so the oldest of this data is only 57 years old. Therefore, it is not possible to obtain an average of 166 years (or of 83 years) of data from Mauna Loa because no such data exists. Hence, it is not possible to obtain a direct comparison of ice core data to Mauna Loa data.
Richard
Richard,
You have the average gas age in the ice completely wrong: 83 years is the difference in age of gas bubbles compared to the surrounding ice (93 years down from the surface), the gas age is much younger than 83 years in the Siple Dome (even I was on the wrong leg in previous message, it is only 10 years for that core).
From the original work of Neftel (1992!):
http://www.biokurs.de/treibhaus/180CO2/neftel82-85.pdf
Here the table of ice age, gas age at the same depth and the CO2 levels measured at Mauna Loa for the same year as the average gas age:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/siple_02.jpg
As one can see, the average gas age is only from about 22 years of atmospheric air, because the air could freely move in and out the ice until the pores were too small (the air had 93 years to diffuse in and out over a distance of 68 meters, not really difficult!). The difference between ice age (counted by the number of layers) and the average gas age of the enclosed air is 83 years. That has nothing to do with the resolution of the air in the gas phase or the difference in timing compared to the atmosphere.
The Siple Dome ice core was drilled in 1983/1984, the average gas age in the first fully closed ice was 10 years younger than in the atmosphere. That makes that there is a 15 year overlap between ice core CO2 data and direct measurements at Mauna Loa.
the average gas age in the first fully closed ice was 10 years younger than in the atmosphere.
Of course must be:
the average gas age in the first fully closed ice was 10 years older than in the atmosphere in the same year…
Thus if one takes a new ice core sample today at Siple Dome (or Law Dome), one will find the same CO2 levels of 10 years ago at Mauna Loa.
Ferdinand Engelbeen:
You know you are stating a falsehood when you say
NO. You know – because you have previously admitted – that the the air in the cores is ‘smoothed’ by mixing in the firn prior to the ice sealing.
And if as you say
then you must be wrong when you say
unless, of course, you are claiming the bubbles contain vacuum and not air.
The only important truth is that the ‘adjustment’ is made to obtain an apparent agreement of the ice core data to the Mauna Loa data.
Richard
Richard,
If real field researchers show that the average gas age at bubble closing depth of 68 m (Siple Dome) or 72 m (Law Dome) are in average 10 years older than in the atmosphere, that is because they not only have calculated that from a firn densification model, but verified the model by taking field measurements, measuring CO2 levels top down in the Law Dome firn and ice from the surface to bubble closing depth:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/law_dome_firn.jpg
At 72 m depth the ice core is 10 ppmv less than in the atmosphere in 1993, that is the CO2 level in the atmosphere from 1983. That is only 10 years “older” than in the atmosphere. Not 30 years (Law Dome) or 83 years (Siple Dome), as that is the age difference with the surrounding ice at bubble closing depth. The latter has nothing to do with the average gas age, and very little with the resolution, which is less than 10 years for Law Dome and ~20 years for Siple Dome. Even if the gas exchange is over 93 years (Siple Dome), the bulk is composed of a mix of air which is from the last 20 years, as there was free movement of air in and out the firn before bubble closing.
NO. You know – because you have previously admitted – that the air in the cores is ‘smoothed’ by mixing in the firn prior to the ice sealing.
Prior to the ice sealing most of the oldest air is gone and as shown above, the average gas age is 10 years older at bubble closing depth than in the atmosphere, while the surrounding ice is 93 years old, thus a difference of 83 years between average gas age and ice age. No matter if that is practically only from the last 20 years or an asymmetric Gaussian curve, heavily weighted to the most recent years: at least half of the mix is from the last 5 years… Thus the smoothing is 10 years (Law Dome) or 20 years (Siple Dome) not 40 or 93 years.
That is what is measured, which is easy to date as the curve of the CO2 levels is exactly known since 1958, sorry that it is not what someone likes to theorize…
The only important truth is that the ‘adjustment’ is made to obtain an apparent agreement of the ice core data to the Mauna Loa data.
That kind of “truth” only is the result of an imaginary idea from someone who doesn’t have any idea how the diffusion of gases in ice cores works… The 7 ppmv / 10 years difference is what is measured…
Richard & Ferdinand,
I’ve been away completing month end financial reports, unfortunately I have am employer who finds those more important than climate science. Go figure? I hope your still paying attention.
I think the point, counter point on this subject could go on endlessly. I have learned much from both of you and thank you for all your posts. I will admit, Ferdinand, that you have moved to me closer to the center on this topic but I still remain skeptical because Richard raises considerable doubt about the lack of variability in CO2 concentrations in the past 600K to 800K years. I am still not moved by the “mass balance” argument because I find it logically flawed (not to be read as your science is flawed) and would not be comfortable using similar logic discussing the relative profitability of various divisions within a business entity. Anyway, that is my perception based on my admittedly limited knowledge on the subject of climate science.
I will continue reading here (WUWT), hopefully seeing more from both of you. I will continue learning, with an open mind, from as many reliable sources as I can find.
I am very happy to have found WUWT, what a great resource!
Gregory,
You are welcome…
Some years ago I have made a few contributions to WUWT, which triggered hundreds of comments about the same subjects. If you have the time, they are the same as my condensed web page, but the comments and answers are sometimes more informative than the article itself…
http://wattsupwiththat.com/2010/08/05/why-the-co2-increase-is-man-made-part-1/
http://wattsupwiththat.com/2010/08/20/engelbeen-on-why-he-thinks-the-co2-increase-is-man-made-part-2/
http://wattsupwiththat.com/2010/09/16/engelbeen-on-why-he-thinks-the-co2-increase-is-man-made-part-3/
http://wattsupwiththat.com/2010/09/24/engelbeen-on-why-he-thinks-the-co2-increase-is-man-made-part-4/
Happy reading…
Ferdinand
Thanks Ferdinand!