6/05/2017  |   14:00 - 14:15   |  302C

DO HYDROLOGIC METRICS ACCURATELY CHARACTERISE SPATIAL AND TEMPORAL VARIATION IN FLOW REGIMES FOR FLOW-ECOLOGY STUDIES? A fundamental step in flow-ecology research is to calculate a range of hydrologic metrics to summarise flow regimes for subsequent use in flow-ecology regression models. It is unclear how well such metrics characterise the spatial and temporal variation in flow regimes and whether this varies with the predictability of flow, making it crucial to quantify the strengths and potential weaknesses of this approach. Using discharge data from unregulated rivers across Australia, we compared how accurately hydrologic metrics characterise spatial and temporal variation in flow regimes. Distance matrices summarising spatial variation in flow metrics generally showed high correlation with distance matrices summarising spatial variation in raw daily flow records, however, this declined with decreasing predictability of flow. Correlations between temporal variation in flow at a single gauge and the metrics used to characterise it tended to be lower than the spatial analysis but also declined with predictability of flow. Hydrologic metrics will continue to be an important tool in flow-ecology research, however, there are settings where alternative approaches, such as those based on spectral analysis or functional decompositions of the hydrograph, may be more suitable.

Ben Stewart-Koster (Primary Presenter/Author), Australian Rivers Institute, Griffith University, b.stewart-koster@griffith.edu.au;


Mark Kennard ( Co-Presenter/Co-Author), Australian Rivers Institute, Griffith University, m.kennard@griffith.edu.au;


Carmel Pollino ( Co-Presenter/Co-Author), CSIRO Land and Water, Carmel.Pollino@csiro.au;


Presentation:
This presentation has not yet been uploaded.

6/05/2017  |   14:15 - 14:30   |  302C

EFFECTS OF PREDICTABLE AND UNPREDICTABLE FLOW VARIABILITY ON DESERT STREAM FISH ASSEMBLAGES In desert streams, flow regimes control fish assemblages via seasonal (predictable) and aseasonal (unpredictable) floods and droughts. Because native biotas possess a wide range of life-history, behavioral, and morphological adaptations to natural flow regimes, seasonal flows should favor native over non-native species. However, aseasonal, unpredictable flow events may represent a source of disturbance for the entire community. We tested these predictions studying fish communities in eight Arizonan streams across a hydrologic variability gradient at four seasonal time points, including two wet (Summer monsoon, winter) and two dry seasons (fall, spring). Three-pass depletion sampling, and spectral methods on long-term flow data allowed (i) determining the short-term responses of native and non-native fish populations to seasonal flow extremes, and (ii) quantifying how the long-term properties of a flow regime (i.e., the relative importance of predictable vs. unpredictable sources of flow variability) relate to fish community diversity. We found that highly seasonal streams had fewer fish species. Additionally, although native richness did not correlate with either metric of flow variability, non-native richness decreased with both increasing seasonal and interannual variation.

Ethan Baruch (Primary Presenter/Author), Arizona State University, ebaruch@asu.edu;


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


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


Presentation:
This presentation has not yet been uploaded.

6/05/2017  |   14:45 - 15:00   |  302C

GENERALIZABILITY OF STREAM FLOW EFFECTS: HOW STRONG IS SMALL-SCALE VARIATION? How generalizable are relationships between stream flow variation and fish population dynamics? This a tricky question because flow effects can vary by species, season, local habitat and geography. It is also challenging because flow effects are quite difficult to estimate as we rarely observe the direct effects of flow on fish. We present results of modeling studies for two types of data on brook trout: widespread count data from many sites vs. long-term individual-based data from one intensively-studied site. With both data types we identified seasonal effects of flow variation on body growth and survival. Results were largely convergent, identifying negative effects of high winter flow on age-0 survival and negative effects of low summer flow. While these results appear robust, the PIT tag study also revealed that seasonal flow effects varied at small spatial scales across the stream network. In general, survival was much more sensitive to flow variation in the tributaries than in the mainstem. Our results suggest that the search for generality in flow responses needs to consider life stage, season and stream order.

Ben Letcher (Primary Presenter/Author), USGS Eastern Ecological Science Center; Silvio O. Conte Research Laboratory, bletcher@usgs.gov;


Evan Childress ( Co-Presenter/Co-Author), U.S. Fish and Wildlife Service, evan.childress@gmail.com;


Matthew O'Donnell ( Co-Presenter/Co-Author), USGS, modonnell@usgs.gov;


Keith Nislow ( Co-Presenter/Co-Author), Northern Research Station, U.S.D.A. Forest Service, University of Massachusetts Amherst, keith.nislow@usda.gov;


Andrew Whiteley ( Co-Presenter/Co-Author), College Of Forestry and Conservation, University of Montana, andrew.whiteley@mso.umt.edu;


Yoichiro Kanno ( Co-Presenter/Co-Author), Colorado State University; Department of Fish, Wildlife, and Conservation Biology, Yoichiro.Kanno@colostate.edu;


Presentation:
This presentation has not yet been uploaded.

6/05/2017  |   15:00 - 15:15   |  302C

EVOLVING BETTER INDICES: CAN GENETIC ALGORITHMS IMPROVE MULTIMETRIC INDEX CONSTRUCTION? For the last four decades, multimetric indices have remained a standard tool in quantitative aquatic bioassessment. During this time a set of construction guidelines has emerged to ensure that indices reliably portray the degree to which a waterbody’s biological condition differs from regional reference streams. Many steps in this standardized process rely on subjective judgment and do not ensure that the final index is better than all possible candidate indices. Here I explore how optimization methods mimicking natural selection might eliminate these shortcomings. The particular metrics included, the level to which taxa are identified and scoring routines for metrics act as genes in the DNA of any multimetric index, the fitness of which is the index’s ability to discriminate between reference and disturbed streams. In theory then, a genetic algorithm can evolve indices with higher discrimination ability. After comparing the performance of a regionalized index constructed with a genetic algorithm to one constructed using standard methods, I report the synergistic improvements evolutionary programming and ecological understanding can bring to the index construction process. Finally, I highlight future research directions for improving aquatic bioassessment methods using the evolutionary programming toolbox.

Kristofor Voss (Primary Presenter/Author), Regis University, kvoss@regis.edu;


Presentation:
This presentation has not yet been uploaded.

6/05/2017  |   15:15 - 15:30   |  302C

ASSESSING MULTIPLE STRESSORS ON STREAM DIATOM METRICS IN THE UPPER MIDWEST, USA This study assessed the influence of physical and chemical stressors on benthic diatoms in the upper Midwest, USA. The dominant stressors of interest included nutrients, herbicides, sediment, and streamflow. Habitat, chemistry and benthic diatoms were assessed at 98 sites that spanned a range of agricultural land use intensity. Boosted Regression Tree was used to examine the importance of stressors on diatom metrics. Results indicated that percent sensitive taxa, percent highly motile taxa, and percent high phosphorus taxa were the diatom metrics with the strongest model performance. For all sites combined, physical habitat and total phosphorus were the best explanatory variables; however, when sites were partitioned into fine and coarse-grain dominated streams, herbicides increased in importance in both stream types. A Classification and Regression Tree model indicated that percent sensitive taxa decreased in streams dominated by fine-grain substrate when water temperatures were over 30 C and triazine concentrations exceeded 1500 ng/L. In contrast, streams dominated by coarse-grain substrate had a higher percentage of sensitive taxa when water temperatures were below 29 C and streams were less than 0.3 meters deep.

Mark Munn (Primary Presenter/Author), U.S. Geological Survey, Tacoma, WA, mdmunn@usgs.gov;


Ian Waite ( Co-Presenter/Co-Author), U.S. Geological Survey, Portland, OR, iwaite@usgs.gov;


Christopher Konrad ( Co-Presenter/Co-Author), US Geological Survey, cpkonrad@usgs.gov;


Presentation:
This presentation has not yet been uploaded.

6/05/2017  |   15:30 - 15:45   |  

IMPACTS OF STREAMFLOW CONDITIONS ON DRIFTING INVERTEBRATES AND FISH BEHAVIOR IN A HEADWATERS STREAM Streamflow has been described as the “master variable” in stream systems because of its ability to drive processes across multiple trophic levels and levels of ecological organization. We monitored invertebrate drift abundance and diversity as a function of streamflow and related that to fish movement and growth data in the Upper Shasta River in Northern California, at impaired flowrates and un-impaired flowrates. Invertebrate drift biomass was significantly less at impaired flows compared to un-impaired flows. Non-metric multi-dimensional scaling suggested that community dissimilarity between sites increased when impaired flows dropped below 0.5 cubic meters per second. Generally, fish movement was greater at the impaired flow site, and was negatively related to the size of individual prey items and amount of prey available. This was coupled with estimated decreased growth rates at impaired flowrates. Our results suggest that fish foraging behavior may be influenced indirectly by altered streamflow through changes to amount, size and diversity of invertebrate drift. This shift, coupled with low food availability, may result in decreased energetic efficiency causing decreased growth rates.

Timothy Caldwell (Primary Presenter/Author), McBain Associates, Tim@mcbainassociates.com;


Gabriel Rossi ( Co-Presenter/Co-Author), University of California - Berkley, rossfactor@berkley.edu;


Christopher Adams ( Co-Presenter/Co-Author), Michigan Technological University, ccadams@mtu.edu;


Sudeep Chandra ( Co-Presenter/Co-Author), Global Water Center and Biology Department, University of Nevada, Reno, sudeep@unr.edu;


Presentation:
This presentation has not yet been uploaded.