Physical Characteristics Influencing Nearshore Habitat Use by Juvenile Chinook Salmon in the Sacramento River, California

Abstract

Urban and agricultural development has resulted in drastically modified riverine corridors that are often considered to be detrimental to the recovery of anadromous salmonid populations. Although mitigation features (e.g., large wood and shallow-water areas) are frequently incorporated in flood control infrastructure to offset the impacts of streambank stabilization, little is known regarding their effectiveness and the habitat characteristics associated with enhanced nearshore rearing conditions in large rivers. We evaluated two measures of habitat use by emigrating juvenile Chinook Salmon Oncorhynchus tshawytscha relative to different shoreline types (rock revetment [riprap], mitigated, and natural) and analyzed associations between environmental characteristics and habitat occupancy using a large number of presence/absence samples in the lower Sacramento River, California. We found both measures of habitat use to be significantly higher at natural shorelines and those including mitigation features than at shorelines consisting predominantly of rock revetment. A predictive logistic regression model suggested that the density of woody material and inundated terrestrial vegetation, depth, and substrate type significantly affected habitat occupancy. Despite a moderate predictive capability (62% of correctly classified records in a leave-one-out simulation), the model was useful in identifying habitat characteristics associated with significantly increased habitat use in this large, low-gradient river, most notably the presence of instream cover (wood or vegetation), gently sloping streambanks, fine substrate, and variable nearshore current velocity. Conversely, habitat occupancy by juvenile Chinook Salmon diminished with large, rocky substrate and increased depth, characteristics favored by introduced predatory Smallmouth Bass Micropterus dolomieu. This study illustrates the value of incorporating mitigation features and identifies characteristics that enhance habitat use by emigrating juvenile Chinook Salmon.Urban and agricultural development along large, salmon-bearing streams of the Pacific Northwest has resulted in the need for and construction of extensive flood control features, such as levees and weirs, to protect valuable agricultural lands, public infrastructure, and private property. As a consequence, nearshore riverine habitat, which is

Do rearing salmonids predictably occupy physical microhabitat?

“…Regression analysis has been used to evaluate the correlation between suitability values estimated by HSC functions and salmonid biomass or density (Wesche et al 1987;Beard and Carline 1991;Beecher et al 2002), while chi-squared tests have been used to validate the transferability of HSC functions between rivers (Thomas andBovee 1993, Mäki-Petäys et al 2002;Guay et al 2003). For probabilistic-based biological models, Akaike's information criterion, pseudo-R 2 values, and the Hosmer-Lemeshow statistic have been used to evaluate and compare the goodness-of-fit of different models, while metrics of selectivity, sensitivity, and errors of omission and commission are commonly used to test classification accuracy (Tiffan et al 2006;Hatten et al 2009;Hatten et al 2016;Tiffan et al 2016;Hellmair et al 2018).…”

“…Compared to the large number of biological models developed and evaluated for salmonids and other aquatic organisms, there have been relatively few studies that have tested the ability of microhabitat suitability models to accurately and reliably predict where these species are more or less likely to occur using independent observational data. Although probabilistic-based microhabitat suitability models have been tested against independent observations of spawning (Geist et al 2008;Hatten et al 2009;Hatten et al 2016) and rearing (Guay et al 2000;Tiffan et al 2006;Tiffan et al 2016;Hellmair et al 2018) salmonids, the metrics used in these tests require a categorical prediction of habitat suitability (i.e., presence or absence). These metrics are therefore unable to evaluate the wide variety non-probabilistic predictive models commonly used in ecohydraulic modelling worldwide.…”

“…Microhabitat suitability models developed using probabilistic-based biological models have been tested against independent observations for their ability to predict the presence and absence of spawning (Geist et al 2008;Hatten et al 2009;Hatten et al 2016) and rearing (Guay et al 2000;Tiffan et al 2006;Tiffan et al 2016;Hellmair et al 2018) salmonids. Test metrics include Cohen's kappa, percentages of microhabitat correctly classified as presence and absence, and errors of commission and omission.…”

Evaluating the influence of environmental and biological factors on migration behavior and residence duration of wild subyearling Chinook Salmon in a fjord estuary using miniature acoustic transmitters

“…Habitat use and migration phenology may also be affected by environmental factors (Hellmair et al 2018; Frechette et al 2021). In ectotherms, temperature is considered the “master” factor controlling behavior through its influence on physiological processes.…”

Shade affects magnitude and tactics of juvenile Chinook salmon antipredator behavior in the migration corridor

“…Habitat structure, including woody debris and submerged aquatic vegetation, has been presumed to both decrease predation risk by providing prey refuge and increase predation risk by providing predator habitat (Zajanc et al 2013;Henderson et al 2019). Migrating juvenile salmon also can have a variable reaction to shade-they may avoid passing through shade cast by anthropogenic structures, but also preferentially hold in shaded reaches with complex habitat (Kemp et al 2005;Zajanc et al 2013;Ono and Simenstad 2014;Hellmair et al 2018). Habitat structure and shade are more constrained to the shoreline in the migration corridor.…”