The decomposition of plant litter is one of the most important ecosystem processes in the biosphere and is particularly sensitive to climate warming. Aquatic ecosystems are well suited to studying warming effects on decomposition because the otherwise confounding influence of moisture is constant. By using a latitudinal temperature gradient in an unprecedented global experiment in streams, we found that climate warming will likely hasten microbial litter decomposition and produce an equivalent decline in detritivore-mediated decomposition rates. As a result, overall decomposition rates should remain unchanged. Nevertheless, the process would be profoundly altered, because the shift in importance from detritivores to microbes in warm climates would likely increase CO(2) production and decrease the generation and sequestration of recalcitrant organic particles. In view of recent estimates showing that inland waters are a significant component of the global carbon cycle, this implies consequences for global biogeochemistry and a possible positive climate feedback.

The effect of resource subsidies on recipient food webs has received much recent attention. The purpose of this study was to measure the effects of significant seasonal seaweed deposition events, caused by hurricanes and other storms, on species inhabiting subtropical islands. The seaweed represents a pulsed resource subsidy that is consumed by amphipods and flies, which are eaten by lizards and predatory arthropods, which in turn consume terrestrial herbivores. Additionally, seaweed decomposes directly into the soil under plants. We added seaweed to six shoreline plots and removed seaweed from six other plots for three months; all plots were repeatedly monitored for 12 months after the initial manipulation. Lizard density (Anolis sagrei) responded rapidly, and the overall average was 63% higher in subsidized than in removal plots. Stable-isotope analysis revealed a shift in lizard diet composition toward more marine-based prey in subsidized plots. Leaf damage was 70% higher in subsidized than in removal plots after eight months, but subsequent damage was about the same in the two treatments. Foliage growth rate was 70% higher in subsidized plots after 12 months. Results of a complementary study on the relationship between natural variation in marine subsidies and island food web components were consistent with the experimental results. We suggest two causal pathways for the effects of marine subsidies on terrestrial plants: (1) the "fertilization effect" in which seaweed adds nutrients to plants, increasing their growth rate, and (2) the "predator diet shift effect" in which lizards shift from eating local prey (including terrestrial herbivores) to eating mostly marine detritivores.

The flux of emerging aquatic insects from streams can provide a significant energy subsidy to riparian web‐building spiders. However, despite the high temporality of aquatic insect emergence, the effects of such aquatic insect dynamics on spider distribution are poorly understood. To examine the relationship, the aquatic insect flux from a headwater stream in a northern Japanese deciduous forest was experimentally manipulated by using a greenhouse‐type covering, during May to July. Under natural conditions, the aquatic and terrestrial insect abundances dramatically decreased and increased from May through July, respectively. The experimental reduction of aquatic insect flux depressed the density of horizontal orb weavers (Tetragnathidae) in both May and June, but not in July when aquatic insects were scarce, indicating a temporal limitation on spider distribution by aquatic insect flux. In contrast, the densities of both vertical orb weavers (Araneidae) and sheet weavers (Linyphiidae) were unaffected by the manipulation throughout the study period. These various responses, differing among months or spider guilds, may be attributed to the degree of specialization for aquatic prey in the spiders and their mobility in response to aquatic insect flux. The experimental results provided direct evidence that the temporal dynamics of aquatic insect flux, as well as spider characteristics, were primary factors determining the distributional patterns of riparian web‐building spiders.