Sedimentary microbial communities play a critical ecological role in lotic ecosystems and are responsible for numerous biogeochemical transformations, including dissolved organic matter (DOM) uptake, degradation, and mineralization. The goals of this study were to elucidate the benthic microbes responsible for utilization of humic DOM in streams and to assess overall variability in microbial biomass and community structure over time and across multiple spatial scales in stream networks, as DOM quality and quantity will likely change with stream order. In Chapter 2, multiple spatial patterns of microbial biomass and community structure were examined in stream sediments from two watersheds; the Neversink River watershed (NY; 1st, 3rd and 5th order streams sampled) and the White Clay Creek watershed (PA; 1st through 3rd order streams sampled). Microbial biomass and community structure were estimated by phospholipid phosphate and phospholipid fatty acids (PLFA) analyses. Multivariate analysis showed that sedimentary C:N ratios, percent carbon, sediment surface area and percent water content explained 68% of the variations in total microbial biomass. Overall, the magnitude of within stream variation in microbial biomass was small compared to the variability noted among streams and between watersheds. Principal component analysis (PCA) of PLFA profiles showed that microbial community structure displayed a distinct watershed-level biogeography, as well as variation along a stream order gradient. Chapter 3 demonstrated that benthic microbial biomass was seasonally dynamic and significantly correlated to a combination of high and low flood pulse counts, variability in daily flow and DOC concentration in the White Clay Creek. Additionally, the seasonal pattern of variation observed in microbial community structure was as a result of shift between the ratios of prokaryotic to eukaryotic component of the community. This shift was significantly correlated with seasonal changes in median daily flow, high and low flood pulse counts, DOC concentrations and water temperature. Compound-specific 13C analysis of PLFA showed that both bacterial and microeukaryotic stable carbon isotope ratios were heaviest in the spring and lightest in autumn or winter. Bacterial lipids were isotopically depleted on average by 2 - 5 / relative to δ13C of total organic carbon suggesting bacterial consumption of allochthonous organic matter, and enriched relative to δ13C algae-derived carbon source. In Chapter 4, heterotrophic microbes that metabolize humic DOM in a third-order stream were identified through trace-additions of 13C-labeled tree tissue leachate (13C-DOC) into stream sediment mesocosms. Microbial community structure was assessed using PLFA biomarkers, and metabolically active members were identified through 13C-PLFA analysis (PLFA-SIP). Comparison by PCA of the microbial communities in stream sediments and stream sediments incubated in both the presence and absence of 13C-DOC showed our mesocosm-based experimental design as sufficiently robust to investigate the utilization of 13C-DOC by sediment microbial communities. After 48 hours of incubation, PLFA-SIP identified heterotrophic α, β, and γ- proteobacteria and facultative anaerobic bacteria as the organisms primarily responsible for humic DOC consumption in streams and heterotrophic microeucaryotes as their predators. The evidence presented in this study shows a complex relationship between microbial community structure, environmental heterogeneity and utilization of humic DOC, indicating that humic DOC quality and quantity along with other hydro-ecological variables should be considered among the important factors that structure benthic microbial communities in lotic ecosystems.

Inland freshwater ecosystems, though comprising a small portion of the earth's surface, are thought to be important in the global carbon (C) cycle. Carbon processing by heterotrophic microbes (bacteria) is a critical process, contributing considerably to overall ecosystem production and processing of dissolved organic carbon (DOC). This study assesses spatial variation in C processing by heterotrophic bacterioplankton in a semi-arid river network: the Rio Grande/Rio Bravo del Norte in Texas, USA. I examined how bacterial metabolism and C processing varied with spatial differences in physicochemical conditions and patterns in DOC lability in this highly impacted riverine network. Physicochemical and biological data were collected at 14 sites from March - December of 2010. I additionally analyzed phytoplankton biomass, bacterial density, bacterial community metabolic rates [bacterial respiration (BR), bacterial productivity (BP), and bacterial growth efficiency (BGE)], and C quality parameters at a subset of nine sites within this drainage. Across the drainage, hydrology and landscape position (i.e., biogeoclimatic conditions, presence of reservoirs, and groundwater contribution to flow) substantially influenced in-stream physicochemical conditions, leading to spatial patterns in bacterial density, phytoplankton biomass, and bacterial metabolism. Bacterial C metabolism was influenced by both physicochemical and C quality - quantity gradients present within the drainage. Bacterial production and BR responded to different environmental gradients, with BP being driven by C quality and inorganic nutrients. This resulted in a negative correlation between BGE and the bacterial respiration of refractory C. Results from this study indicate that natural variation and anthropogenic impacts influence the physicochemical and biotic conditions across the Rio Grande/Rio Bravo del Norte drainage and these effects have implications for C sequestration, transformation, and transport, as well as for organic matter (OM) delivery to the Gulf of Mexico.

Variation of dissolved of organic carbon concentration in stream water is a consequence of process changes in the surrounding terrestrial environment. This study will focus on 1) Identify significant environmental factors controlling the spatial and temporal variation of DOC in terrestrial ecosystems of a watershed southeast of Boston, Massachusetts; 2) Model the DOC leaching from different land cover and examine the relationship between leaching flux and in-stream DOC. Our hypothesis is variations of in stream DOC is closely related to watershed properties and environmental factors at annual, seasonal, and daily scales, especially land cover type, watershed size and hydrology. To explore the relationship of hydrology and DOC variation at ungauged sub-basin, we examined the effectiveness of using simulated stream flow from Soil Water Assessment Tool (SWAT) to study terrestrial DOC dynamics. Our results demonstrated that streamflow, drainage area, and percent of wetland and forest were particularly strong predictors in watersheds with a large proportion of developed area. The resulting linear model is able to explain about 70.2% (R2=0.702) and 65.1% (R2=0.651) of the variance of in-stream DOC concentrations at seasonal and annual scales respectively. Results also suggest that more frequent DOC sampling is necessary to establish the quantitative relationship between simulated stream flows from the SWAT and in-stream DOC concentrations at daily scale. The physically based ecosystem model developed in this study shows that DOC leaching from various land cover are highly correlated (up to 80%) with in-stream DOC by using ecological process with incorporated different hydrological pathways. It shows that leaching of DOC from soil is a significant contributor to the in-stream DOC. The production of DOC is largely controlled by the vegetation type and soil texture. Considering the hydrologic control on DOC transport with different pathways of water at finer spatial and temporal scale highlights the need to identify the quantitative relationships between water and carbon flux.