Robert Buchkowski has written a blog post about his Journal of Ecology paper on how interactions between green and brown food webs impact plants and nutrient cycling. 

Herbivores indirectly impact plant growth by altering the speed at which nutrients are recycled through the soil and back into plant tissue. Changes in plant leaf chemical or physical quality is one of the mechanisms through which herbivores change nutrient recycling rates. The legacy of these quality changes in living leaves are carried down to the soil where leaves and nutrients are recycled. Sometimes litter quality increases with herbivory because of rapid lush growth or incomplete resorption of nutrients, while in other cases it decreases because of defensive responses used to protect the leaves. Regardless of the direction, decomposers and plant growth should respond to changes in the availability of recycled nutrients.

We designed an experiment to measure the recycling effect from reductions in litter quality, to reductions in litter decomposer feeding, through to reductions in nutrient recycling and plant growth. The only problem was that our data did not support the hypothesis.

Cage with goldenrod growing over the first year of the experiment. Leaf litter from these plants with and without grasshoppers and fertilization was used in our experiment the following year.

To understand why we didn’t see a strong recycling feedback, it is important to recount our experimental design. Our experiment occurred in two phases wherein we grew goldenrod plants (Solidago altissima) in a glasshouse with and without grasshoppers (Melanoplus femurrubrum) feeding on them. We also fertilized half of the plants with nitrogen containing an isotope that could be used to track the nitrogen through the soil. In the winter, we harvested the plant leaf litter. We found that herbivores increased leaf toughness and nitrogen content, while fertilization increased only nitrogen content.

Field mesocosms the day of harvest. We measured litter decomposition, woodlice abundance, microbial activity, and plant growth in each mesocosm.

For the second phase of the experiment, we decomposed that leaf litter in the field with and without the addition of a common detritivorous woodlouse (Trachelipus rathkii). At the start and end of the growing season we estimated or measured litter, woodlouse abundance, microbial activity, soil nitrogen, and plant biomass.

We expected to see woodlice quickly eating fertilized, no-herbivory litter and recycling nitrogen into the soil to promote plant growth. Instead, we found that woodlice did eat more fertilized litter, but these differences did not lead to consistent changes in soil nitrogen and plant growth. The changes in plant growth were wholly unrelated to litter decomposition. Rather, plant growth was explained by plant community, earthworm abundance, microbial activity, and was negatively associated with woodlice.

The model comparisons we made between tight and diffuse nitrogen recycling. Letters show the different nitrogen pools (P = plants, M=microbes, L= litter, I = woodlice, and S= soil). Stars in the diffuse cycling case show the buffering processes that help explain our field results.

We were surprised by the weak relationship between litter decomposition and plant growth, so we built a mathematical model to understand why these two processes might be uncoupled. Our model tested two alternative hypotheses proposed by previous researchers: (1) litter nitrogen recycles directly to a plant available pool and (2) litter nitrogen recycling is diffuse and buffered by other nitrogen sources. Other nitrogen sources included external inputs, soil nitrogen, and plant nitrogen storage. We found that the second diffuse model fit our experimental data better. Furthermore, we used the model to demonstrate that changes in litter nitrogen recycling do impact plant growth. However, relatively small rates of external nitrogen inputs or large stockpiles of plant or soil nitrogen can overcome these differences in litter recycling.

Our results led us to a refined perspective on how animals, especially those feeding in different parts of an ecosystem, will change nutrient recycling and plant growth. We did discover important effects of these animals on litter quality and nutrient release. Yet, we also came to appreciate the importance of considering external nutrient sources, internal nutrient stores, and the current state of the plant community.

Our results suggest that animal communities persisting for years or decades may be necessary to develop feedback effects strong enough to rise above background and anthropogenic sources of nutrients.

Robert Buchkowski, Yale School of Forestry & Environmental Studies, USA

Read the full paper online: Nitrogen recycling in coupled green and brown food webs: weak effects of herbivory and detritivory when nitrogen passes through soil