Effects of altered dry season length and plant inputs on soluble soil carbon

Soil moisture controls microbial activity and soil carbon cycling. Because microbial activity decreases as soils dry, decomposition of soil organic matter (SOM) is thought to decrease with increasing drought length. Yet, microbial biomass and a pool of water-extractable organic carbon (WEOC) can increase as soils dry, perhaps implying microbes may continue to break down SOM even if drought stressed. Here, we test the hypothesis that WEOC increases as soils dry because exoenzymes continue to break down litter, while their products accumulate because they cannot diffuse to microbes. To test this hypothesis, we manipulated field plots by cutting off litter inputs and by irrigating and excluding precipitation inputs to extend or shorten the length of the dry season. We expected that the longer the soils would remain dry, the more WEOC would accumulate in the presence of litter, whereas shortening the length of the dry season, or cutting off litter inputs, would reduce WEOC accumulation. Lastly, we incubated grass roots in the laboratory and measured the concentration of reducing sugars and potential hydrolytic enzyme activities, strictly to understand the mechanisms whereby exoenzymes break down litter over the dry season. As expected, extending dry season length increased WEOC concentrations by 30% above the 108 μg C/g measured in untreated plots, whereas keeping soils moist prevented WEOC from accumulating. Contrary to our hypothesis, excluding plant litter inputs actually increased WEOC concentrations by 40% above the 105 μg C/g measured in plots with plants. Reducing sugars did not accumulate in dry senesced roots in our laboratory incubation. Potential rates of reducing sugar production by hydrolytic enzymes ranged from 0.7 to 10 μmol·g^-1 ·h^-1 and far exceeded the rates of reducing sugar accumulation (~0.001 μmol·g^-1 ·h^-1 ). Our observations do not support the hypothesis that exoenzymes continue to break down litter to produce WEOC in dry soils. Instead, we develop the argument that physical processes are more likely to govern short-term WEOC dynamics via slaking of microaggregates that stabilize SOM and through WEOC redistribution when soils wet up, as well as through less understood effects of drought on the soil mineral matrix.