A recent paper (Olson, 2013) finds a number of long-term studies were wrong about no-till practices building soil organic matter and thus sequestering carbon. The problem, says Kenneth Olson, soil scientist at the University of Illinois, is how the studies in question measured the gains or losses in soil organic carbon (SOC; organic carbon is about 50% of soil organic matter by weight). According to Olson, these long-term studies made soil carbon measurements during or at the end of the experiments which compared the results of no-till (NT1 in figure 1) to moldboard plowing (MP). They then concluded that carbon was sequestered in the soil under no-till but not in tillage systems. The figure below represents what Olson says these studies measured.
The problem pointed out by Olson is that this scenario compares everything to the carbon levels in the moldboard plow system (MP) which is assumed to be at a steady state. As Olson states, “without…pre-treatment SOC data for the baseline treatment (MP), the SOC sequestration magnitude and rate…cannot be verified.” Olson’s point is that the snapshot measurement of SOC does not tell the whole story.
He then argues convincingly that a baseline measurement of SOC is needed in all cases to determine both the sequestration rate and magnitude of both the no-till and moldboard plow systems. When this baseline measurement (A on the bottom axis) is included, as shown in the second figure below, the conclusions can be quite different.
Starting before the treatments have been applied (point A at the bottom of the graph) the SOC levels are the same (the plot averages are not significantly different). The treatments are then applied for 10 to 20 years and then the SOC levels are measured again, point B on the graph. Olson argues that because SOC is rarely steady, even over long time periods, SOC levels will have changed in both treatments. He points out that SOC levels in the moldboard plow treatment (MP) will often be lower at B than at the A, showing that carbon is still being lost in this system. Olson found in his own research, and suspects the same in other studies, that the SOC levels in the no-till also decreased, but at a slower rate than the MP soil (see the NT1 line). Therefore, carbon sequestration, as Olson defines it, “the process of transferring CO2 from the atmosphere into the soil of a land unit through unit plants [plants growing on that land unit], plant residues and other organic solids, which are stored or retained in the unit as part of the soil organic matter,” did not occur. The no-till system is losing organic carbon, but at a slower rate than the moldboard plow system. Only if the carbon levels in the NT system increased between A and B (NT2 on the chart) could it be said that carbon was sequestered.
Olson’s conclusions, if they stand up under further scrutiny (it is a peer-reviewed paper) bring up several important points. First, they highlight the fact that agricultural systems, even those that disturb the soil the least, are still degrading compared to native conditions, at least in the eastern half of the U.S. To this, I say, of course they are. The prairie did not allow for much export of food, so why is it the target for agriculture? Let’s move beyond comparing agriculture to untouched prairie and aim for something that works for us. If no-till protects the soil from erosion, slows the loss of organic matter and still produces food, then it is the best option we have.
Second, at least where soil organic matter levels were high before agriculture was introduced, the ability of agriculture to sequester carbon to mitigate greenhouse gas emissions seems to be limited. This may affect the ability of agriculture to be a player in any future carbon sequestration market.
Finally, the situation in the arid West is different. Here, where native soils are very low in organic matter, adding irrigation and high yielding crops has the potential to increase soil organic matter. However, high value vegetable production (potatoes, onions, carrots) which at present require tillage, and the complex, dynamic rotations make it unlikely that continuous no-till, the focus of many of these Midwest long term studies, will be widely adopted in these irrigated regions.
The scientific community should review Olson’s revised definition of carbon sequestration, and if they help us get a better view of reality, adopt them and adjust our course accordingly.
Reference
Kenneth R. Olson. Soil organic carbon sequestration, storage, retention and loss in U.S. croplands: Issues paper for protocol development. Geoderma, 2013; 195-196: 201 DOI: 10.1016/j.geoderma.2012.12.004
April 2014 update: an expanded and more detailed paper (press release) on this topic has been published:
Olson, K. R., Al-Kaisi, M. M., Lal, R., & Lowery, B. (2014). Experimental Consideration, Treatments, and Methods in Determining Soil Organic Carbon Sequestration Rates. Soil Science Society of America Journal, 78(2), 348. doi:10.2136/sssaj2013.09.0412
Comments
Another key factor is how deeply SOC or SOM is measured in a given study. In studies focusing just on surface soils, no-till often looks very good, but in studies looking at the soil profile down to the typical rooting depth, the story is more mixed. High-yield systems that produce a lot of residue/trash can withstand a degree of relatively deep tillage and erosion, yet still build SOC/SOM in the root zone. Fields where farmers are working organic matter into the top 8″ instead of top 2″ end up having a lot more potential for carbon sequestration and building soil quality.
Andy McGuine:
Since you posted the article I assume you also did the evaluation of the paper. You did a very nice job of analyzing my SOC sequestration study and identifying the issues. Also liked you graphs which illustrate the impacts. This is the best evaluation piece I have found on the web to date. That includes 22 blogs and 6 news websites which mostly published the original article.
It would be interesting to assess dry range land SOC levels prior to agricultural use and then after irrigation and various tillage and management systems are applied for 10 or 20 years to determine if SOC sequestration does occur.
Kenneth Olson
Kenneth, I am glad that you found my review of your paper acceptable.
In response to your last comment, there have been some studies in our region comparing soil organic matter levels in agricultural fields to adjacent native soils (generally 0.4-0.7% soil organic matter). In almost every case, the fields have higher levels of organic matter, but the increases are small, ranging from 0.1-0.4% organic matter. This is after 50+ years of irrigated agriculture.
I think there is potential for further increases if tillage can be decreased, but that is easier said that done in our cropping systems.
Andy McGuire:
Thanks for providing some indication of the native SOC vs. irrigated soil after 50 years.
Most studies focus on a shorter period of time and with natural variability it is hard to determine the SOC sequestration, storage, retention and loss.
Changing the methods used to measure SOC sequestration is not going to go easy. But there will be a new paradigm. I decided to start challenging the work of the primary sources (Ohio State University soil carbon group (Drs. Lal and Dick))of much of the flawed comparison method with no pre-treatment SOC measurements and sampled only once in the 47th year on one plot area and 49th year on other plot. After 47 years of NT treatment the plots have only 55% of the SOC of the native woods. I wrote my first letter to the editor of SSSAJ and they responded in a predictable manner. I will let you interpret. This reply and comment is in the new electronic copy of the SSSAJ 77 (2)693-694)and (695-696) (March, 2013) issue.
The USDA, ARS units have been publishing SOC sequestration rates of 0.5 Mt/ha/yr for a simple switch from MP to NT. On many sites that is way to high. And even if it was stored one year the rate would drop each year thanks to aeration and microbes. Have not decided whether to challenge the ARS regional USA rates. They also only used one time sampling. These studies (many authors and 500 studies) were published in 2005 Soil and Tillage research journal and in SSSAJ in 2010 by Dr. Fransluebbers for SE US. If Cap and Trade is funded the farmers will never be able to prove these high SOC sequestration rates and they will wonder how the researchers got those rates. So do I.
Ken Olson
Two studies that looked at C sequestration in irrigated fields in Central Washington (broken out from native shrub steppe) are:
Cochran et.al. (2007). Soil carbon pools and fluxes after land conversion in a semiarid shrub-steppe ecosystem. Biology and Fertility of Soils 43:479-489
Collins, et.al. (2010). Carbon sequestration under irrigated switchgrass (Panicum virgatum) production. Soil Science Society of America Journal.
These indicate that tillage has a role in influencing SOC levels, but that residue management and carbon inputs (either through plant biomass or “artificial” C additions such as manure or compost) are the key to increasing SOC. The caveat to carbon inputs is that they come with their own life-cycle GHG costs due to fossil energy use and N2O emissions. It is important to note that N2O emissions from our irrigated systems are relatively small when compared to IPCC estimates due to how tightly N fertilizers are managed (Haile-Mariam et.al. 2008; Stockle et.al. 2012).
Knowing the initial SOC level is essential to estimating what will happen with a practice change (Kemanian and Stockle 2009). The same management practice (whether no-till, organic, etc.) can result in either gains or losses depending on the initial condition.
The reality is that the amount of SOC gained by shifting to more progressive farming systems (whether than be no-till, organic, planned grazing, etc.) are all likely far more modest than currently believed. This is especially the case with the fragile soils of Central Washington that are so low in SOM to begin with – with such high rates of annual C turnover under irrigation – that regardless of management – if we’re producing annual crops we will have definitive limits to the potential to increase SOC.
There is also some indication that another possible strategy (for some locations) may be a mix of no-till and tillage (Puraskayastha et.al. 2008), with the occasional tillage operation “burying” surface carbon deeper in the soil profile.
Nailing down the “how to measure the effect of a given practice for a given location” question is critical to understanding what the real benefit is — for the bean counting of carbon.
Whether farmers receive compensation in carbon markets in the long-run (which I doubt) likely has a lot to do with how decisions about “what and how to measure” are made. Counting soil carbon is not the same as counting electrons saved by changing light bulbs and the carbon markets that are emerging have been extremely reticent to touch soil carbon for this reason.
The benefits to the environment and agronomic systems, though, including reducing soil erosion itself (not the same thing as increasing SOC) are very real and of concern to long-term agricultural sustainability.