Assessing predation risks for small fish in a large river ecosystem between contrasting habitats and turbidity conditions

Michael J. Dodrill; Michael D. Yard; William E. Pine

doi:10.1674/0003-0031-175.2.206

Home/Archive/USGS

USGS · 70170937

Assessing predation risks for small fish in a large river ecosystem between contrasting habitats and turbidity conditions

Michael J. Dodrill·Michael D. Yard·William E. Pine

View original source ↗Export metadata ↓

Abstract

This study examined predation risk for juvenile native fish between two riverine shoreline habitats, backwater and debris fan, across three discrete turbidity levels (low, intermediate, high) to understand environmental risks associated with habitat use in a section of the Colorado River in Grand Canyon, AZ. Inferences are particularly important to juvenile native fish, including the federally endangered humpback chub Gila cypha . This species uses a variety of habitats including backwaters which are often considered important rearing areas. Densities of two likely predators, adult rainbow trout Oncorhynchus mykiss and adult humpback chub, were estimated between habitats using binomial mixture models to examine whether higher predator density was associated with patterns of predation risk. Tethering experiments were used to quantify relative predation risk between habitats and turbidity conditions. Under low and intermediate turbidity conditions, debris fan habitat showed higher relative predation risk compared to backwaters. In both habitats the highest predation risk was observed during intermediate turbidity conditions. Density of likely predators did not significantly differ between these habitats. This information can help managers in Grand Canyon weigh flow policy options designed to increase backwater availability or extant turbidity conditions.

Explore related subjects

American Midland Naturalist Article

Source-reported bounding extent: 35.545635932499415° to 36.99377838872517° latitude; -114.04907226562499° to -111.2860107421875° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Michael J. Dodrill, Michael D. Yard, William E. Pine. 2016. Assessing predation risks for small fish in a large river ecosystem between contrasting habitats and turbidity conditions. https://doi.org/10.1674/0003-0031-175.2.206

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2021

Breeding bird occurrence across a gradient of graminoid- to shrub-dominated fens and fire histories

Population declines of birds affiliated with grasslands and grass-shrub habitats have been identified as a critical conservation concern. The effects of woody encroachment into grassland-dominated systems, which may contribute to species declines, has been examined mainly in upland systems. We examined occurrence of bird species across a gradient of graminoid- to shrub-dominated wetlands in the Upper Peninsula of Michigan and describe the habitat characteristics associated with their probability of occurrence at segment (100 × 100 m) and landscape (200 m buffer) scales. Because fire can suppress woody encroachment, we also examined the role of fire history, extending from <1 to >50 y on birds and habitat characteristics. We focused on seven species closely affiliated with grassland and grass-shrub habitats, but also found in wetlands: Sedge Wren ( Asio flammeus ), LeConte's Sparrow ( Ammodramus leconteii ), Savannah Sparrow ( Passerculus sandwichensis ), Bobolink ( Dolichonyx oryzivorus ), Common Yellowthroat ( Geothlypis trichas ), Swamp Sparrow ( Melospiza georgiana ), and Alder Flycatcher ( Empidonax alnorum ). We used nonparametric, multiplicative regression models to relate habitat variables and fire history to probability of species occurrence. Unlike studies in upland grasslands, none of the top models included measures of litter or graminoid height, and the highest probability of occurrence often was where graminoid cover was at intermediate rather than highest levels. Measures of woody cover or structure were important to all seven species modeled. We observed few clear differences in habitat metrics across fire history categories beyond the first year of burning. Years since last burned was in the best model for four of the seven species, despite high variability of habitat metrics within fire history categories; those species were more likely to occur in segments burned 15–19 y earlier. Infrequent light- to moderate-severity fires in fens had little effect on focal species occurrence beyond the first year after burning and little long-term effect to suppress woody cover.

Michigan↗

USGS2021

Spatial and temporal variation in length-weight relationships of age-0 Scaphirhynchus sturgeon in the lower Missouri River

Length-weight relationships can be useful tools for assessing fish condition. We developed these equations (W = a L b ) for wild-caught age-0 (4.1–12.0 cm) Scaphirhynchus sturgeon from eight reaches spanning over 750 river km of the lower Missouri River from 2014 to 2017. We used nonlinear modeling to estimate the constant ( a ) and exponent ( b ) of the LW equation for each reach to assess potential spatial differences. We also assessed long-term temporal effects by estimating these parameters by year at Lexington reach, which is located in the middle of our sampling area and was the only reach sampled all 4 y. Constant and exponent estimates from linearized regressions varied by reach and were inversely related during the spatial analyses. Similarly, parameter estimates were also inversely related and varied among years during the temporal analysis at Lexington. To account for the relationship between constant and exponent values, we used predicted weights at 2 cm increments (4.1–12.0 cm) for the spatial analysis (by reach) and for the temporal analysis (by year). During the 2014 and 2015 spatial analyses, weights varied by size but were usually higher in Lexington and Glasgow, which were the furthest upstream reaches sampled during those years. During 2016 and 2017, Lexington was the furthest downstream reach sampled but did not consistently yield relatively high predicted weights. Temporal analysis at Lexington yielded higher predicted weights for 2014–2015 compared to 2016–2017 for higher size categories (10- and 12-cm). In general our results suggest differences in body condition among reaches and years in the lower Missouri River. Further research is needed to identify the specific mechanisms driving spatial and temporal L-W relationship differences observed and to determine if differences in predicted body conditions affect long-term survival and recruitment of age-0 Scaphirhynchus sturgeon. Currently, factors influencing age-0 Scaphirhynchus sturgeon condition and growth are unknown and this work serves to highlight knowledge gaps regarding factors influencing Scaphirhynchus sturgeon recruitment.

Iowa, Kansas, Missouri, Nebraska↗

USGS2021

Limited co-existence of native unionids and invasive dreissenid mussels more than 30 Y post dreissenid invasion in a large river system

There are serious concerns for native freshwater mussel survival (Bivalvia: Unionidae) in the Laurentian Great Lakes region after populations were seemingly pushed to the brink of extirpation following the introduction of dreissenid mussels ( Dreissena polymorpha and D. rostriformis bugensis ) in the mid-1980s. The Detroit River was the first major river system in North America to be invaded by dreissenids, and unionids were considered extirpated from the river by 1998. Since then several unionid refuges (areas with relatively low dreissenid impact and surviving unionids) have been found in coastal areas of lakes St. Clair and Erie, but no documentation exists in the Detroit River. To assess dreissenid presence and potential unionid persistence, a mixture of stratified random, historical, and potential refuge sites were surveyed during summer 2019 in the Detroit River. Unionid and dreissenid habitat use was further investigated with analysis of variance and classification tree analyses. Of the 56 sites surveyed, only five sites had live unionids totaling 220 animals of 11 species. More than 2000 unionid shells of 31 species were collected from 39 sites, confirming the large and diverse unionid populations that existed prior to the dreissenid invasion. Ninety-eight percent of live unionids found showed evidence of past or present dreissenid attachment. Estimated dreissenid densities were highly variable with river location and ranged from 0 to 5673 live individuals per m 2 , with the largest densities concentrated in the upstream half of the Detroit River. Despite their previously assumed extirpation from the Detroit River, live unionids were found during this comprehensive survey. Although only 40% of the historical species within the unionid assemblage remains, our results suggest, in the right conditions, some coexistence is possible among some species of unionids and dreissenids in this large river system.

Michigan↗