I recently visited Jerry Melillo's warming plots at the Harvard Forest, and I don't know why, but I was amazed by how much the leaf litter had decomposed over the past few months. This got me thinking about the structure of carbon of the remaining litter. The historic view has been that labile sugars go first, then cellulose, then lignin, but what about the waxes and other compounds which don't fit into these categories? To a certain extent, our understanding of litter decomposition has been limited by our ability to detect and distinguish these other compounds, but tools like NMR are changing that.
The husband-wife duo of Nishanth Tharayil and Vidya Suseela at Clemson have once again teamed up with Baoshan Xing at UMass to study the fine chemistry of litter. In one of their previous papers, the authors noted that reducing precipitation increased the relative abundance of tannins in red maple litter; in this paper they looked at how the chemistry of Japanese knotweed litter, that horrible invasive which is actually quite delectable when young, changes through time in litter subject to different warming and precipitation treatments at the Boston Area Climate Experiment.
In this experiment, the authors made "old" and "new" litter bags by harvesting Japanese knotweed which had either been decomposing upright following senescence the previous year, or just-senesced stems. They were placed at the edges of the high (~+3C) and ambient temperature plots, under drought (50% of precipitation removed year-round), ambient, or wet (an extra 50% of rain applied during the growing season) treatment, and harvested at four time points over a period of three years. By pulling peaks from various methods I don't understand (DRIFT Spectroscopy, and 13C cross-polarization magic angle spinning NMR spectroscopy), they confirmed that decomposition of recalcitrant litter is generally more temperature sensitive than more labile stuff. Again, as previously noted, decomposition is greatest when supplemental precipitation is applied in conjunction with warming. But perhaps the most interesting point was that while the effect of climate treatment on overall decomposition rate did not differ between new and old litter, specific (recalcitrant) compounds did decompose more fully in the older litter, which the authors cite as evidence that initial litter chemistry does matter.
As my PI pointed out, there are a few problems with the way in which this experiment was designed. First, the authors used litter from a plant not found in the warming experiment, collected at a site a hundred miles away. The authors state that this litter was chosen because all litter is clonal, and therefore should have been initially identical, but wouldn't litter taken from one of the trees at or near the experimental site not be adequate? Perhaps the problem would be that litter allowed to decompose in-situ for a year would start with clearly different microbial communities than the litter which had just-fallen from a tree. That said, we know that there is often microbial succession during decomposition, and therefore the litter going in old probably had a different microbial community associated with it than the "new" litter, whether or not it was harvested from immediately adjacent stems from a clonal population. In this instance, differences in decomposition with litter starting chemistry could be due to the presence of a microbe at the litter source site initiating decomposition of compounds that microbial populations are not as well-suited to break down. For instance, there could be fungi at the source litter site which are much less abundant at the experimental warming site because it is a well-trodden former agricultural field with relatively low organic matter content.
My PI also pointed out that using litter which starts from the same plant but is in different stages of decay is not a particularly biologically informative way to answer a question. She said that using various kinds of leaf litters which naturally differ in their starting chemistry, as occurs in ecosystems today and potentially exacerbated by climate-induced shifts in species composition and litter chemistry, would make the results of the experiment more useful in ecosystem carbon models. I believe the litter bags for that experiment are decomposing in-situ as I type.
As I alluded to above, my main beef with the paper was, of course, that they didn't talk about whether the microbial community differed between warming treatments. I am interested in knowing whether the differences in decomposition are due to purely physical effects, or whether changes in the microbial community also played a role. I would also like to know if succession of microbes on the different litter ages and in the different plots followed the same pattern, just being accelerated in some instances, or whether the communities were completely different.
This is a kind of chronic problem with ecologists; they generally ignore microbes (or suppose what is happening without validating the assumption). Almost every paper I read, I hope with all my heart that they have soil cared for nicely and kept in an ultra-low freezer somewhere. But anytime I inquire, the answer is no. If they want to really understand what happened in their system, it is their loss.
But please, if you are doing anything involving soil, or are doing a study in which you expect soil or litter-degrading microbes to be adhered to your object of interest, please flash freeze that soil/litter at collection and place in a -80C freezer. If you don't have access to one, contact me before you collect your samples and I will send a self-addressed cooler with dry ice, and I will fill my PI's freezer with random samples as long as I can without her noticing.