ABSTRACT
Free-living (FL) and particle-attached (PA) bacterial communities play critical roles in nutrient cycles, metabolite production, and as a food source in aquatic systems, and while their community composition, diversity, and functions have been well studied, we know little about their community interactions, co-occurrence patterns, and niche occupancy. In the present study, 13 sites in Taihu Lake were selected to study the differences of co-occurrence patterns and niches occupied between the FL and PA bacterial communities using correlation-based network analysis. The results show that both FL and PA bacterial community networks were non-random and significant differences of the network indexes (average path length, clustering coefficient, modularity) were found between the two groups. Furthermore, the PA bacterial community network consisted of more correlations between fewer OTUs, as well as higher average degree, making it more complex. The results of observed (O) to random (R) ratios of intra- or inter-phyla connections indicate more relationships such as cross-feeding, syntrophic, mutualistic, or competitive relationships in the PA bacterial community network. We also found that four OTUs (OTU00074, OTU00755, OTU00079, and OTU00454), which all had important influences on the nutrients cyclings, played different roles in the two networks as connectors or module hubs. Analysis of the relationships between the module eigengenes and environmental variables demonstrated that bacterial groups of the two networks favored quite different environmental conditions. These findings further confirmed the different ecological functions and niches occupied by the FL and PA bacterial communities in the aquatic ecosystem.
Subject(s)
Bacteria , Bacterial Physiological Phenomena , Lakes/microbiology , Microbial Interactions , Microbiota/physiology , Bacteria/classification , Bacteria/metabolism , Biodiversity , Cluster Analysis , DNA, Bacterial/analysis , Ecosystem , Environment , Phylogeny , Species SpecificityABSTRACT
Oklahoma contains the largest number of manmade lakes and reservoirs in the United States. Despite the importance of these open surface water bodies to public water supply, agriculture, thermoelectric power, tourism and recreation, it is unclear how these water bodies have responded to climate change and anthropogenic water exploitation in past decades. In this study, we used all available Landsat 5 and 7 images (16,000 scenes) from 1984 through 2015 and a water index- and pixel-based approach to analyze the spatial-temporal variability of open surface water bodies and its relationship with climate and water exploitation. Specifically, the areas and numbers of four water body extents (the maximum, year-long, seasonal, and average extents) were analyzed to capture variations in water body area and number. Statistically significant downward trends were found in the maximum, year-long, and annual average water body areas from 1984 through 2015. Furthermore, these decreases were mainly attributed to the continued shrinking of large water bodies (>1km2). There were also significant decreases in maximum and year-long water body numbers, which suggested that some of the water bodies were vanishing year by year. However, remarkable inter-annual variations of water body area and number were also found. Both water body area and number were positively related to precipitation, and negatively related to temperature. Surface water withdrawals mainly influenced the year-long water bodies. The smaller water bodies have a higher risk of drying under a drier climate, which suggests that small water bodies are more vulnerable under climate-warming senarios.