Plant litter tends to decompose faster in arid and semi-arid grassland ecosystems than predicted by litter decomposition models. We hypothesized that photodegradation of plant litter by ultraviolet (UV) radiation may explain discrepancies between modeled and observed patterns in decomposition in these systems. In a field manipulation experiment of UV radiation at three contrasting grassland sites, we observed significant increases in litter mass loss with UV exposure over a 2 year period. However the mechanism for this UV effect on mass loss rates is unknown. We used a series of controlled experiments to examine three possible mechanisms by which mass could be lost via photodegradation: 1) photochemical mineralization to CO₂, 2) changes in litter solubility leading to increased leaching losses, and 3) changes in litter bioavailability leading to increased microbial decomposition. In one experiment, we examined gaseous and leaching losses in a factorial experiment of litter species, sterilization, and UV radiation. To examine gaseous losses, we placed litter in air-tight microcosms fitted with either UV-transparent or UV-blocking lids under UV lamps and sampled the headspace periodically over 10 weeks. To evaluate leaching losses, a subsample of litter was leached at the end of the experiment, and the leachate was evaluated for changes in dissolved organic carbon content. In a separate experiment, we evaluated whether pre-exposing litter to radiation facilitates microbial decomposition. Sterile and non-sterile litter of two species was either irradiated with UV lamps or kept in the dark, and subsequently buried and decomposed under conditions favorable to microbial decomposition. Microbial respiration was measured by sampling headspace for CO₂.

Results/Conclusions  Based on these experiments, the only mechanism observed to increase carbon losses with UV exposure was direct mineralization of organic matter to CO₂. Rates of CO₂ production were independent of litter species or sterilization treatment, suggesting that this mechanism was abiotic. We further examined the influence of wavelength and litter density as controls on photochemically-induced CO₂ production. Surprisingly, we found that short-wave photosynthetically active radiation (PAR) was just important as well as UV radiation in producing CO₂, and rates of CO₂ production per unit mass of litter depended on litter density. Therefore, our results suggest that direct mineralization to CO₂ is the primary pathway by which litter mass is lost via photodegradation in arid ecosystems. Furthermore, PAR and litter density need to be taken into account when evaluating the role of photodegradation in litter decomposition.