5/22/2016  |   13:30 - 13:45   |  307

USING STATE-SPACE MODELS TO INFER PROCESSES FROM PATTERNS The quantitative tools available to ecologists continue to expand and evolve. In particular, the development of multivariate state-space models (SSM) has allowed for examinations of both the intrinsic and external drivers of population viability, food web dynamics, and community stability. These models consist of two parts: 1) a process model that captures the underlying true, but unknown “states of nature”, and 2) an observation model that accounts for sampling errors. We will provide an overview of how SSM are implemented, including a discussion of necessary data and limitations to inference. We will discuss a variety of freely available software options that exist to fit SSM under both likelihood and Bayesian frameworks. We will draw on several recent examples to highlight the variety of research questions to which SSM have been applied in freshwater and marine systems. These include zooplankton in temperate lakes, salmon in boreal rivers, and fish and invertebrates in kelp forests.

Mark Scheuerell (Primary Presenter/Author), University of Washington , scheuerl@uw.edu;


Eric Ward ( Co-Presenter/Co-Author), NOAA Northwest Fisheries Science Center, eric.ward@noaa.gov;


Steve Katz ( Co-Presenter/Co-Author), Washington State University, steve.katz@wsu.edu;


Elizabeth Holmes ( Co-Presenter/Co-Author), NOAA Northwest Fisheries Science Center, eli.holmes@noaa.gov;


Presentation:
This presentation has not yet been uploaded.

5/22/2016  |   13:45 - 14:00   |  307

AN ARTIFICIAL FLOW REGIME DAMPENS TOP-DOWN EFFECTS IN A RIVERINE FOOD WEB The periodic and rapid fluctuation in streamflow due to the production of electricity by hydropower plants, or hydropeaking, represents an important alteration to the natural flow regime. Artificial flow variability is known to affect riverine geomorphology, vegetation growth, and habitat availability for aquatic faunas, but ecological responses at the food-web level remain largely unknown. We studied Chattahoochee River’s food web below Buford Dam (Georgia, USA) over the period 2003-2011. Using wavelet analyses on 15-minute flow data, we quantified the intensity of daily flow rhythms at the foot of the dam and ~25 km downstream and showed that the effect of dams on flow regime dampens downstream. Using multivariate autoregressive (MAR) models on seasonal abundance data of biota, we inferred interaction strengths among aquatic invertebrates, and between Brown trout (the most abundant predatory fish) and invertebrate prey. Food web complexity increased downstream, as did the top-down control, i.e., the negative effects of Brown trout on invertebrate prey. Our results advance the notion that novel environmental variation may alter not only species in isolation but also food-web interactions.

Albert Ruhi (Primary Presenter/Author), National Socio-Environmental Synthesis Center. , aruhi@sesync.org;


Xiaoli Dong ( Co-Presenter/Co-Author), University of California - Davis, xldong@ucdavis.edu;


Courtney Holt ( Co-Presenter/Co-Author), University of Georgia, crholt@uga.edu;


Darold Batzer ( Co-Presenter/Co-Author), Department of Entomology, University of Georgia, dbatzer@uga.edu;


John Sabo ( Co-Presenter/Co-Author), Arizona State University, John.L.Sabo@asu.edu;


Presentation:
This presentation has not yet been uploaded.

5/22/2016  |   14:00 - 14:15   |  307

ABIOTIC INFLUENCES ON FISH COMMUNITY DYNAMICS IN A U.S. SOUTHEAST RIVER The interplay between biotic and abiotic drivers of population and community dynamics is a long-standing focus of research in ecology. In highly-variable, flashy streams, flow variation (disturbance) often overrides biotic interactions (competition, predation); in contrast, according to ecological theory the latter should be relatively more important in rivers with predictable hydrographs. Here we addressed this topic using time-series analyses on long-term data of fish and flow in the Etowah River, Georgia (USA), a seasonal but predictable system. We used Fourier analyses on 55 years (1960-2014) of daily flow data to identify high and low-flow events. Multivariate Autoregressive State-Space (MARSS) models allowed asking to what extent fish abundances (1998-2014) were explained by biotic interactions and flow variation. All feeding functional groups showed significant density-dependence effects, none were negatively affected by flow variation, and two of them were actually favored by low flows. Our results support the hypothesis that biotic interactions are an important driver of fish community dynamics.

Qi Deng (Primary Presenter/Author), Arizona State University, qdeng6@asu.edu;


Albert Ruhi ( Co-Presenter/Co-Author), National Socio-Environmental Synthesis Center. , aruhi@sesync.org;


John S. Kominoski ( Co-Presenter/Co-Author), Florida International University, jkominoski@gmail.com;


Megan Hagler ( Co-Presenter/Co-Author), University of Georgia, ms.meganhagler@gmail.com;


John Sabo ( Co-Presenter/Co-Author), Arizona State University, John.L.Sabo@asu.edu;


Presentation:
This presentation has not yet been uploaded.

5/22/2016  |   14:15 - 14:30   |  307

DROUGHT INTENSIFIES EUTROPHICATION IN THE KLAMATH RIVER: EVIDENCE FROM AN 8 YEARS OF DAILY METABOLISM Changing river flows and excess nutrients should greatly affect ecosystem processes such as river metabolism, but we have little data on this interaction. The lower Klamath River lies below a series of dams impounding high-nutrient water from the upper watershed. Tributaries dilute this water lowering nutrient concentrations downstream, and this tributary inflow varies through time and space allowing necessary variation to detect effects on gross primary production (GPP) and ecosystem respiration (ER). We estimated a daily time series of GPP and ER at 3 sites during the summer-autumn growing season. We fit state-space time series models with autoregressive error structure to metabolism to estimate the relative effects of physical drivers. At daily time scales, solar input and fraction of nutrient rich water were the strongest predictors of variation in metabolism. Interannual patterns of GPP followed the daily data showing higher production at the lower two sites as a function of the fraction of water derived from the dam. Lower tributary inputs during dry years caused higher GPP due to higher nutrient concentrations.

Robert O. Hall (Primary Presenter/Author), Flathead Lake Biological Station, University of Montana, bob.hall@flbs.umt.edu;


Laurel Genzoli ( Co-Presenter/Co-Author), University of Montana, laurel.genzoli@umontana.edu;


Presentation:
This presentation has not yet been uploaded.

5/22/2016  |   14:30 - 14:45   |  307

EXTREME NON-SEASONAL FLOODING OF STREAMS ACROSS AN ELEVATION GRADIENT – USING VULNERABILITY TRAITS TO UNDERSTAND BENTHIC COMMUNITY RESPONSES TO DISTURBANCE Record rainfall fell over the northern Colorado foothills and mountains in September 2013, creating massive flooding that attracted national media attention. This non-seasonal, extreme event created the unique opportunity to re-sample 14 recently sampled streams across a 1,200 m elevation gradient to test the hypothesis that pre/post differences in abundance and richness of benthic macroinvertebrates among the streams could be explained by variation in local disturbance intensity. At each site we calculated a reach-scale disturbance index based on cross-sectional area of inundated valley bottom at peak flow, water surface gradient, streambed resistance and inundated valley bottom roughness. We characterized all taxa in terms of species traits that reflect resilience or resistance (mobility, life stage present at flood, behavioral mode, etc.) to test if such taxa showed differential survival. We also tested if communities from low vs high elevation streams varied in vulnerability to this extreme, non-seasonal disturbance due to differences in natural flow regime caused by elevation-dependent hydroclimatology.

LeRoy Poff (Primary Presenter/Author), Colorado State University, n.poff@rams.colostate.edu;


Erin Hotchkiss ( Co-Presenter/Co-Author), Virginia Polytechnic Institute and State University (Virginia Tech), ehotchkiss@vt.edu;


Boris Kondratieff ( Co-Presenter/Co-Author), Colorado State University, Boris.Kondratieff@ColoState.edu;


Scott Morton ( Co-Presenter/Co-Author), Colorado State University, Scott.Morton@ColoState.Edu;


Rachel Harrington ( Co-Presenter/Co-Author), U.S. Environmental Protection Agency Region 8, harrington.rachel@epa.gov;


Kayce Anderson ( Co-Presenter/Co-Author), Colorado State University, kaycelu@gmail.com;


Chris Funk ( Co-Presenter/Co-Author), Colorado State University, Chris.Funk@Colostate.edu;


Alexander Flecker ( Co-Presenter/Co-Author), Cornell University, Ithaca, NY, USA, asf3@cornell.edu;


Presentation:
This presentation has not yet been uploaded.