干旱区生态环境知识资源中心

The mechanisms of plant litter decay in drylands are poorly understood, limiting the accuracy of nutrient-cycling models for these systems. We monitored the decay of 12 leaf litter types on the soil surface of the Sonoran Desert for 34 months and assessed what traits predicted mass loss and how exposure to different wavebands of sunlight influenced mass loss. Mass loss varied considerably among litter types, ranging from 42%-96% after 34 months in full sunlight. Traditional indices of litter quality (e.g., initial C:N or lignin:N ratios) failed to predict differences in mass loss among litter types. The strongest predictor of mass loss was the microbial respiration rate of initial litter, which explained 45%-54% of the variation in loss among litter types. Microbial respiration rates were not correlated with traditional indices of litter quality, but were positively correlated with the water-soluble fraction in litter and concentrations of dissolved organic C in this fraction. Traditional indices of litter quality failed to predict decay likely because they did a poor job of predicting microbial degradability of litter, not because microbial degradation was a minor driver of decay. In all radiation-exposure treatments, water-soluble fractions and respiration rates increased through decay and were several times higher after 34 months than initially. Hence, labile pools and microbial degradability of litter increased through decay in contrast to traditional views that labile pools decline and constrain microbes. Litter exposed to UV or UV through blue radiation wavelengths, lost on average 1.3 times or 1.5 times more mass, respectively, than litter not exposed to these wavebands. The magnitude of this photodegradation was greater in litter types that had higher initial concentrations of hemicellulose and cellulose per unit surface area. Litter exposed to full sun had higher water-soluble fractions and usually had higher respiration rates, illustrating that sunlight accelerated microbial degradation by increasing labile pools. The processes driving litter decay appeared to differ appreciably from mesic systems and involved strong couplings between abiotic and biotic drivers.