Dissolved organic matter dynamics are critical to carbon budgets, land-water transfers and interactions, and the activity and ecology of bacteria. The export of dissolved organic matter (DOM) from land to water is an important component of global and regional carbon budgets, and terrestrial DOM in turn influences metabolism of aquatic ecosystems. To better understand spatial and temporal patterns in DOM dynamics and to investigate the underlying mechanisms controlling DOM processing and export, I conducted field surveys and manipulative experiments measuring DOM chemistry, microbial community composition, microbial activity, and DOM bioavailability along a terrestrial- aquatic toposequence in an arctic tundra catchment. Multivariate analyses revealed distinct DOM composition and microbial community composition in upland tussock, riparian birch-willow, and lowland wet sedge tundra soils using phospholipid fatty acid analysis and denaturing gradient gel electrophoresis analysis of 16S rRNA gene sequences. The latter method also was used to differentiate bacterial assemblages in surface waters of Toolik Lake, the main inlet stream to the lake, and a small primary stream. Experiments in which DOM from soil pore and surface waters along the terrestrial-aquatic gradient were added to a mixture of bacteria from all sites showed that bioavailability, measured as DOC-specific bacterial production (BP), differed with DOM source. Compared to experiments in which a mixture of DOM was added to each microbial community, DOM chemistry played a stronger role than bacterial community composition in controlling BP, at short time scales (hours). In longer-term (weeks) mesocosm experiments, addition of soil water DOM greatly enhanced lake BP, and path analyses showed that these effects of DOM on BP were indirect through shifts in lake bacterial community composition. Such shifts in bacterial community composition also influenced the bioavailability of DOM photo-products; initial effects of photo-oxidized DOM on BP were negative, likely due to the production of harmful byproducts, but increased contact time (days to weeks) between photoproducts and bacterial communities resulted in shifts in community composition and enhanced DOM bioavailability relative to dark controls.

Overall, the results suggest spatial heterogeneity is an important aspect of DOM dynamics across the landscape, and that bioavailability of upslope DOM or photo-oxidized DOM depends on contact time between DOM and bacterial communities. It appears that lake bacterial communities contain a full complement of functional groups with respect to C processing and variation in C supply selected for inactive groups over relatively short time scales (days to weeks). At shorter time scales, however, upslope-DOM and recently produced photo products had negative impacts on bacterial activity. Thus contact time between microbial communities and new DOM sources will determine the extent of processing as DOM moves through catchments. At low water flow, shifts in microbial community composition in response to upslope DOM inputs result in increased microbial processing and reduced export from land to water. Conversely, during high flow, short contact times prevent community level shifts and limit DOM processing, thereby increasing the proportion of DOM exported to down-slope habitats.