ASUO and GREBES travel-award recipient Lorien Reynolds
After at least 30 yrs of observation, empirical investigation, and dynamic modeling, soil and soil organic matter dynamics remain largely a mystery, as is their potential response and feedback to global change. I was privileged to be invited to speak at a session on “Soil Change and Soil Organic Matter (SOM) Dynamics in the Anthropocene” at the 2013 Fall Meeting of the American Geophysical Union. My talk was entitled “The DIRT on Q10: In situ depletion of labile-inputs does not increase temperature sensitivity in a laboratory incubation”, combining a 20 yr long litter experiment (Detritus Input and Removal Treatments, DIRT) with a long-term (525 days) laboratory incubation to test the temperature sensitivity (Q10) of decomposition at differing content and putative quality of carbon.
The concept of temperature sensitivity has become a particular sticking point in the SOM literature. My findings, that Q10 did not increase with the depletion of putatively labile substrates, stands in direct contradiction to many other recent publications. Several talks during the session I was invited to also concerned Q10, though methods varied and their conclusions were consistent with the previously published studies, and thus contradictory to my own. I presented my data with great hope of stirring discussion about how and why my results are so different and emphasized my growing doubt that 1) SOM quality is the primary control of decomposition rate and temperature response and 2) experiments of this type can in fact determine whether or not this is true. Unfortunately my presentation passed almost without comment. I can only surmise that either my conclusions were so far off base as to render them irrelevant or the audience was too taken aback to respond in kind at the time.
The two presentations prior to my own had both directly concerned Q10. The first was a short (3 weeks if I recall correctly) laboratory incubation of soils that had been warmed in situ for several years, and thus could be assumed to have ‘burned off’ some labile carbon as inputs were not increased, and thus have relatively higher non-labile, and presumably more temperature sensitive, SOM. The presenter concluded that Q10 had in fact increased due to warming and putative decrease in SOM quality. The second presentation brought up a very interesting conceptual shift that in light of my findings and that of the previous presenter, I found particularly intriguing. The presenter emphasized that Q10, while measured experimentally and used to parameterize models of SOM decomposition is itself a response variable effected by microbial populations, enzyme energetics, substrate availability, etc., none of which are typically considered when it is measured. I should break here to emphasize something that may not be obvious to any but another ecosystem ecologist, namely that ecology, being interested in how integrated systems respond to change, frequently treat those systems as a kind of ‘black box’ which is then perturbed (i.e., experimentally altered) and the effect observed. Thus our results are often used to infer mechanism rather than measure it directly. Such is the case of my soil incubation as I did not, and could not, measure the microbes, enzymes, or carbon quality directly during the course of the experiment, much less the interactions thereof. Thus my results, like those of any incubation of this kind, cannot be used as direct evidence of a change in SOM decomposition rates or controls. A need to shift toward a more mechanistic examinations of the controls on SOM decomposition, such as by actually measuring the controls of Q10 rather than attempting to infer them from ‘black box’ ecology is apparent, at least to me.
In addition to the question of temperature sensitivity there were several interesting presentations on the controls of SOM decomposition and how they are incorporated into models. Without belaboring the details, I believe there was a clear trend toward a deeper, more mechanistic view of these dynamics, as well as an excellent discourse on the necessary checks and evaluations of global scale claims for local scale dynamics.
In this same spirit, though not in the same session, I sat in on a presentation by Schlesinger, who was among the first to examine SOM decomposition and its dynamics, evaluating the claim that North American deserts may be substantial carbon sinks. The recently published data had been obtained in a variety of ways, including eddy-covariance and direct soil carbon metrics. Schlesinger systematically deconstructed the argument, pointing out that the projected carbon sequestration would have been high even for a highly productive system, and impossible for a low-productivity system such as a desert. Non-biological means of carbon sequestration via the translocation of carbon through the soil profile as CaCO3 also failed to demonstrate a large enough pool to make up for the discrepancy. Again I was struck by the necessity to closely examine mechanism in the face of improbable results from what I am calling ‘black box’ ecology.
In summary, I was both frustrated and encouraged by my experiences at AGU this past December. Though my own efforts failed to engage my colleagues, I found that mine was not the only voice calling for a shift in perspective and a retreat from the over-emphasis of purely inferential results.