Headwaters are critical reservoirs of microbial diversity for fluvial networks

Katharina Besemer Gabriel Singer Christopher Quince Enrico Bertuzzo William Sloan Tom J. Battin 0 School of Engineering, University of Glasgow , Glasgow G12 8QQ , UK 1 WasserCluster Lunz GmbH , Dr Carl Kupelwieser Promenade 5, 3293 Lunz am See , Austria 2 Department of Limnology and Oceanography, University of Vienna , Althanstrasse 14, 1090 Vienna , Austria 3 Laboratory of Ecohydrology, School of Architecture, Civil and Environmental Engineering, Ecole Polytechnique Fe de rale Lausanne , 1015 Lausanne , Switzerland Articles on similar topics can be found in the following collections ecology (1804 articles) environmental science (300 articles) microbiology (57 articles) Receive free email alerts when new articles cite this article - sign up in the box at the top right-hand corner of the article or click here References Subject collections Email alerting service rspb.royalsocietypublishing.org Research Cite this article: Besemer K, Singer G, Quince C, Bertuzzo E, Sloan W, Battin TJ. 2013 Headwaters are critical reservoirs of microbial diversity for fluvial networks. Proc R Soc B 280: 20131760. http://dx.doi.org/10.1098/rspb.2013.1760 Subject Areas: ecology, microbiology, environmental science Authors for correspondence: Katharina Besemer e-mail: Tom J. Battin e-mail: Electronic supplementary material is available at http://dx.doi.org/10.1098/rspb.2013.1760 or via http://rspb.royalsocietypublishing.org. Katharina Besemer1,2, Gabriel Singer1,2, Christopher Quince3, Enrico Bertuzzo4, William Sloan3 and Tom J. Battin1,2 1. Introduction & 2013 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/3.0/, which permits unrestricted use, provided the original author and source are credited. high beta diversity compared with mid-sized steams [8], an observation that is supported by experimental work with protozoan metacommunities [11]. These patterns may be attributable to large environmental variation among headwaters [12], their spatial isolation limiting dispersal [11] and their high abundance within fluvial networks [13,14]. Furthermore, stream confluences, as conspicuous nodes in the fluvial network, have been postulated to augment biodiversity of a network by way of accumulating species from multiple catchments and, thus, increasing the size of the metacommunity from which local communities assemble [5,11,12]. As posited by the network dynamics hypothesis, strong gradients of channel geomorphology across confluences may also increase habitat heterogeneity and community variation, which would have a knock-on effect on network scale biodiversity [15]. However, empirical observations supporting these conjectures are sparse. Field studies have revealed elevated fish diversity around confluences [16,17] and, similarly, laboratory work on protozoan metacommunities evoked that dispersal increases diversity in experimental confluences characterized by higher connectivity [11]. In streams and rivers, microbial life is dominated by benthic biofilms, which control key ecosystem processes [10]. The biodiversity of these biofilms results from the interplay of local environmental conditions and the dispersal dynamics of microorganisms from the source community suspended in the streamwater [18]. Microorganisms are primarily passive dispersers [19]; the directionality of the water flow generating asymmetrical dispersal, together with the dendritic network structure, are therefore likely to influence microbial diversity patterns [20]. Understanding microbial biodiversity patterns at the scale of entire fluvial networks is of paramount importance, especially since headwaters are increasingly under threat by burial, mountain-top mining and inter-basin water transfer [21,22]. In this study, we investigated patterns of microbial alpha and beta diversity in benthic biofilms throughout a fluvial network. We leaned on the concept of metacommunity (i.e. a set of local communities linked by dispersal) ecology [2] to guide our understanding of microbial diversity. Specifically, we predicted higher alpha diversity downstream than upstream of confluences because of increasing metacommunity size [23]. Furthermore, based on the converging structure of fluvial networks [7], we hypothesized that microbial alpha diversity increases from headwaters downstream, a pattern that may be amplified by significant downstream dispersal of small organisms with water flow [11]. We also predicted that microbial beta diversity decreases from headwaters downstream because of dispersal limitations [11] and pronounced habitat variation among headwaters [12]. 2. Material and methods (a) Study area and field survey We sampled benthic biofilms from 114 streams within a pre-alpine catchment (River Ybbs, Austria; 254 km2; 1893532 metres above sea level (m.a.s.l.); figure 1). Catchment geology is dominated by dolomite (82%) and karst; forests (82%) and alpine meadows (11%), characterized land use, bare rock, agricultural areas and settlements constitute minor parts of the catchment (7% in total). Streams were sampled during a one-week period in winter after prolonged baseflow. This was to ensure rather stable and homogeneous hydrological conditions throughout the fluvial network and to sample mature biofilms with reduced successional dynamics [24]. Discharge ranged from less than 1 l s21 in the smallest headwaters to 2282 l s21 in the fifth-order stream during the survey. To assess the relevance of confluences for biodiversity patterns, we primarily sampled tributary pairs upstream of their confluence and the recipient streams downstream of their confluence (figure 1). Recipient streams were sampled 1020 times the channel width or at least three riffle-pool sequences downstream of the confluence [25] to ensure complete mixing of the streamwater, while retaining the characteristics of the confluence environment [15]. Mixing of streamwater was confirmed by measuring electrical conductivity. Sampling was primarily designed to cover important confluences, while equally representing all orders and sizes of streams. A number of additional samples were taken at the inflow and outflow of lakes to complete the picture of the network. Stream channel depth, width, slope, velocity and discharge were measured in the field following standard procedures. The dimensionless Froude number was calculated as an integrative descriptor of streambed hydraulics [14]. A digital elevation model, rigorously ground-truthed, served to compute network metrics, hydrologic distances between sampling sites, the size of subcatchments and land use (see the electronic supplementary material, methods). Streamwater was analysed for NO3, NH4 and PO4 concentrations and dissolved organic matter (DOM) was characterized using fluorescence and spectrophotometric techniques (electronic supplementary material, methods). From each site, 612 stones (14 cm in diameter) (...truncated)