Predation by Caspian terns Hydroprogne caspia is a factor limiting the recovery of some Endangered Species Act‐listed populations of salmonids from the Columbia River basin, especially steelhead Oncorhynchus mykiss . This prompted the development and implementation of two separate management plans, one in the Columbia River estuary and the other in the Columbia Plateau region, to reduce the impact of Caspian tern predation on smolt survival. Caspian terns nesting at managed breeding colonies within the basin were relocated to alternative nesting islands created for terns outside the basin. Both plans were successful in significantly reducing smolt losses to Caspian terns nesting at managed sites; however, new developments have led to regression in smolt survival gains associated with tern management. Adaptive management to prevent terns at managed colonies from relocating to nest elsewhere in the basin and to improve nesting opportunities for Caspian terns outside the basin are needed to maximize the survival benefits to Endangered Species Act‐listed salmonids from management. Adaptive management is also necessary to ensure the long‐term viability of the Pacific Flyway population of Caspian terns, whose breeding population has declined by more than 50% since the two management plans were implemented.

We reviewed studies of piscivorous colonial waterbird predation on juvenile salmonids to synthesize current knowledge of factors affecting fish susceptibility to avian predators. Specifically, we examined peer-reviewed publications and reports from academic, governmental, and nongovernmental agencies to identify commonalities and differences in susceptibility of salmonids to avian predation, with a focus on mark-recovery studies in the Columbia River basin. Factors hypothesized to influence salmonid susceptibility to avian predation were grouped into four general categories: (1) salmonid species and populations, (2) environmental factors, (3) prey density, predator density, and migration timing, and (4) prey characteristics. Our review focused on predation by Caspian terns Hydroprogne caspia, double-crested cormorants Nannopterum auritum, and gull species Larus spp. as these are the most well-studied avian predators of salmonids. Results indicated that predator-prey interactions varied across salmonid species and populations and species of avian predator. Inferences across studies supported multiple hypotheses regarding predator-prey dynamics, including environmental factors that influence prey exposure to predators (e.g., river flows, turbidity, alternative prey), variation in predator and prey abundances, predator characteristics (e.g., foraging behavior, colony location), and prey characteristics (e.g., fish length, condition). Mark-recovery studies of avian predation on fish populations have greatly improved our understanding of the factors affecting fish susceptibility to avian predation, the relative contributions of abiotic and biotic factors to predation susceptibility, and the extent to which avian predation affects fish survival and the viability of prey populations. Future studies that jointly model predation and survival and the factors affecting those processes will further broaden our understanding of predator-prey dynamics and directly evaluate the effects of predation on prey population dynamics.

Abstract How predators respond to changes in prey abundance (i.e., functional responses) is foundational to consumer–resource interactions, predator–prey dynamics, and the stability of predator–prey systems. Predation by piscivorous waterbirds on out‐migrating juvenile steelhead trout (Oncorhynchus mykiss) is considered a factor affecting the recovery of multiple Endangered Species Act‐listed steelhead populations in the Columbia River basin. Waterbird functional responses, however, may vary by predator species and location, with important implications to predator management strategies. We used a 13‐year dataset on waterbird abundance across seven breeding colonies (three Caspian tern [Hydroprogne caspia], two double‐crested cormorant [Nannopterum auritum], and two California and ring‐billed gull [Larus californicus and Larus delawarensis] colonies) and steelhead tag‐recovery data (>645,000 tagged and >32,000 recovered steelhead) to quantify weekly predation probabilities and functional responses across waterbird species, colonies, and years. Weekly predation probabilities were highly variable, ranging from 0.01 to 0.30 at tern colonies, 0.01 to 0.20 at cormorant colonies, and 0.03 to 0.13 at gull colonies. Per capita predation probabilities were an order of magnitude higher at inland tern and cormorant colonies relative to estuary colonies of the same species. Terns displayed Type II functional responses across colonies and years, where predation probabilities peaked at low steelhead abundances and declined as steelhead abundance increased (i.e., predator swamping). Cormorants nesting at the large estuary colony (several thousand birds) displayed a Type III functional response, but cormorants nesting at the smaller inland colony (several hundred birds) displayed a Type II response. Consumption probabilities of steelhead by gulls remained consistent across a large range of steelhead availability, suggesting a Type I or a Type III functional response, but a lack of colony abundance data prevented quantifying functional responses. The level of tern predation combined with Type II functional responses indicate possible population‐level impacts that could destabilize small or declining prey populations. Conversely, the apparent Type III functional responses of gulls and estuary nesting cormorants are indicative of prey switching behaviors targeted at periods of high steelhead abundance. Our results illustrate the complexity of predator–prey interactions and the importance of quantifying predator‐ and location‐specific functional responses when predicting the efficacy of management strategies to enhance prey populations.

We investigated the cumulative effects of predation by piscivorous colonial waterbirds on the survival of multiple salmonid (Oncorhynchus spp.) populations listed under the U.S. Endangered Species Act (ESA) and determined what proportion of all sources of fish mortality (1 -survival) were due to birds in the Columbia River basin, USA. Anadromous juvenile salmonids (smolts) were exposed to predation by Caspian terns (Hydroprogne caspia), double-crested cormorants (Nannopterum auritum), California gulls (Larus californicus), and ring-billed gulls (L. delawarensis), birds known to consume both live and dead fish. Avian consumption and survival probabilities (proportion of available fish consumed or alive) were estimated for steelhead trout (O. mykiss), yearling Chinook salmon (O. tshawytscha), sub-yearling Chinook salmon, and sockeye salmon (O. nerka) during out-migration from the lower Snake River to the Pacific Ocean during an 11-year study period (2008-2018). Results indicated that probabilities of avian consumption varied greatly across salmonid populations, bird species, colony location, river reach, and year. Cumulative consumption probabilities (consumption by birds from all colonies combined) were consistently the highest for steelhead, with annual estimates ranging from 0.22 (95% credible interval = 0.20-0.26) to 0.51 (0.43-0.60) of available smolts. The cumulative effects of avian consumption were significantly lower for yearling and sub-yearling Chinook salmon, with consumption probabilities ranging annually from 0.04 (0.02-0.07) to 0.10 (0.07-0.15) and from 0.06 (0.3-0.09) to 0.15 (0.10-0.23), respectively. Avian consumption probabilities for sockeye salmon smolts was generally higher than for Chinook salmon smolts, but lower than for steelhead smolts, ranging annually from 0.08 (0.03-0.22) to 0.25 (0.14-0.44). Although annual consumption probabilities for birds from certain colonies were more than 0.20 of available smolts, probabilities from other colonies were less than 0.01 of available smolts, indicating that not all colonies of birds posed a substantial risk to smolt mortality. Consumption probabilities were lowest for small colonies and for colonies located a considerable distance from the Snake and Columbia rivers. Total mortality attributed to avian consumption was relatively small for Chinook salmon (less than 10%) but was the single greatest source of mortality for steelhead (greater than 50%) in all years evaluated. Results suggest that the potential benefits to salmonid populations of managing birds to reduce smolt mortality would vary widely depending on the salmonid population, the species of bird, and the size and location of the breeding colony.

The degree to which predation is an additive vs. compensatory source of mortality is fundamental to understanding the effects of predation on prey populations and evaluating the efficacy of predator management actions. In the Columbia River basin, USA, predation by Caspian Terns (Hydroprogne caspia) on U.S. Endangered Species Act (ESA)-listed juvenile salmonids (smolts;Oncorhynchusspp.) has led to predator management actions to reduce predation; however, the assumption that reduced predation translates into greater salmonid survival, either within the life stage where predation occurs or across their lifetime, has remained untested. To address this critical uncertainty, we analyzed a long-term (2008-2018) mark-recapture-recovery data set of ESA-listed steelhead trout (O.mykiss) that were tagged (n = 78,409) and subsequently exposed to predation during smolt out-migration through multiple river reaches (spatial scales), jointly estimating weekly probabilities of steelhead survival, mortality due to bird predation, and mortality due to other causes. This concurrent estimation across time-stratified cohorts allowed for the direct measurement of the strength, magnitude, and direction of relationships between survival and Caspian Tern predation. Estimates of Tern predation on steelhead were substantial in most years, with cumulative annual estimates ranging from 0.075 (95% creditable interval = 0.058-0.099) to 0.375 (0.290-0.461). Increases in Tern predation probabilities were associated with statistically significant decreases in steelhead survival probabilities in all evaluated years and salmonid life stages (smolt out-migration and smolt-to-adult returns). Results provide novel evidence that predation by Caspian Terns may have been a super-additive source of mortality during the smolt life stage and a partially additive source of mortality to the adult life stage. Annual estimates of the difference between observed survival and baseline survival (i.e., in the absence of Tern predation) ranged from 0.052 (0.017-0.103) to 0.314 (0.172-0.459) during the steelhead smolt life stage and from 0.011 (0.001-0.029) to 0.049 (0.025-0.078) to the adult life stage. The estimated levels of compensation have important implications for predator management actions aimed at increasing the survival of endangered salmonids, and the modeling approach developed herein provides a framework to directly quantify the impacts of source-specific mortality factors on prey populations.

ABSTRACTPredation on smolts by Caspian terns (Hydroprogne caspia) has been identified as a factor limiting the restoration of some populations of anadromous salmonids (Oncorhynchus spp.) from the Columbia River basin that are listed under the U.S. Endangered Species Act. Implementation of a management plan to reduce numbers of Caspian terns nesting at the 2 largest colonies in the Columbia Plateau region of Oregon and Washington, USA, began in 2014 and is ongoing. We investigated the response of Caspian terns during 2014–2016 to reductions in nesting habitat at these 2 colonies. Management prevented terns from nesting at both colonies, and the estimated numbers of nesting pairs in the region declined significantly from a mean of 877 pairs during premanagement to 769 and 675 pairs in 2015 and 2016, respectively. The management objective of reducing numbers of nesting terns in the Columbia Plateau region to ≤200 breeding pairs was not achieved during the first 2 years of full implementation of the plan. Regional nesting success did not decline significantly following the initiation of management, and remained at levels considered sufficient to sustain the regional subpopulation. Despite the species’ capacity for long‐distance breeding dispersal, the majority of displaced terns exhibited stronger than expected philopatry to the Columbia Plateau region. Analysis of resightings of banded terns indicated that most (>80%) terns that nested in the Columbia Plateau region premanagement returned to the region in 2015 and 2016, but the proportion that returned as breeders decreased while the proportion that returned as nonbreeding floaters increased compared with premanagement. The proportion of banded terns that were unobserved also increased during management years, suggesting that more terns became floaters in the Pacific Flyway and went unobserved because they were prospecting and foraging in locations or regions where there was little or no monitoring. The unexpectedly high regional philopatry exhibited by terns during management was likely a reflection of the low availability of suitable alternative nesting habitat outside the region. Most terns that remained in the region displayed considerable flexibility in nest site selection by nesting either at a previously smaller, intermittently successful breeding colony or at a small new colony where nesting activity had not previously been recorded. As long‐lived seabirds, Caspian terns may integrate information regarding nesting success over several years before choosing to change nesting locations, longer than the 2 years of this study, especially if alternate locations are distant or intermittently available, or a history of nesting at multiple locations exists within the region. © 2020 The Wildlife Society.

To investigate the cumulative effects of colonial waterbird predation on fish mortality and to determine what proportion of all sources of fish mortality (1 − survival) was due to bird predation, we conducted a mark–recapture– recovery study with upper Columbia River steelhead Oncorhynchus mykiss that were PIT-tagged and released (N= 78,409) at Rock Island Dam on the Columbia River, USA. We used a state–space Bayesian model that incorporated live detections and dead recoveries of tagged fish to jointly estimate predation and survival probabilities during smolt out-migration to the Pacific Ocean over an 11-year study period. Estimated cumulative (all colonies combined) avian predation probabilities ranged from 0.31 (95% credible interval [CRI] = 0.27–0.38) to 0.53 (95% CRI= 0.42– 0.64) annually, indicating that avian predation was a substantial source of mortality. Of the predator species evaluated, predation by Caspian terns Hydroprogne caspia was often the highest, with predation probabilities ranging from 0.11 (95% CRI = 0.09–0.14) to 0.38 (95% CRI= 0.29–0.47). Probabilities of predation by double-crested cormorants Phalacrocorax auritus and mixed colonies of California gulls Larus californicus and ring-billed gulls L. delawarensis were generally lower than the probabilities for terns but were also substantial, with upwards of 0.04 (95% CRI= 0.03–0.07; cormorants) and 0.31 (95% CRI= 0.25–0.39; gulls) of steelhead consumed. Comparisons of total smolt mortality with mortality due to avian predation indicated that avian predation accounted for 42% (95% CRI= 30–56%) to 70% (95% CRI= 53–87%) of total mortality, suggesting that more steelhead were consumed by avian predators than died from all other mortality sources combined. Results indicate that avian predation, although not the original cause of steelhead declines in the basin, is now a factor limiting the survival of upper Columbia River steelhead. Using the analytical framework developed in this study, future studies can consider the cumulative impact of multiple mortality sources across large spatial and temporal scales to more fully understand the extent to which they limit fish survival. *Corresponding author: allen@realtimeresearch.com Received April 18, 2019; accepted July 25, 2019 Transactions of the American Fisheries Society 148:896–913, 2019 © 2019 American Fisheries Society ISSN: 0002-8487 print / 1548-8659 online DOI: 10.1002/tafs.10197

We developed a state-space mark–recapture–recovery model that incorporates multiple recovery types and state uncertainty to estimate survival of an anadromous fish species. We apply the model to a dataset of outmigrating juvenile steelhead trout (Oncorhynchus mykiss (Walbaum, 1792)) tagged with passive integrated transponders, recaptured during outmigration, and recovered on bird colonies in the Columbia River basin (2008–2014). Recoveries on bird colonies are often ignored in survival studies because the river reach of mortality is often unknown, which we model as a form of state uncertainty. Median outmigration survival from release to the lower river (river kilometre 729 to 75) ranged from 0.27 to 0.35, depending on year. Recovery probabilities were frequently ≥0.20 in the first river reach following tagging, indicating that one out of five fish that died in that reach was recovered on a bird colony. Integrating dead recovery data provided increased parameter precision, estimation of where birds consumed fish, and survival estimates across larger spatial scales. More generally, these modeling approaches provide a flexible framework to integrate multiple sources of tag recovery data into mark–recapture studies.

We evaluated the impact of predation on juvenile steelhead Oncorhynchus mykiss and yearling and subyearling Chinook Salmon O. tshawytscha by piscivorous waterbirds from 11 different breeding colonies in the Columbia River basin during 2012 and 2014. Fish were tagged with both acoustic tags and PIT tags and were tracked via a network of hydrophone arrays to estimate total smolt mortality (1 -survival) at various spatial and temporal scales during outmigration. Recoveries of PIT tags on bird colonies, coupled with the last known detections of live fish passing hydrophone arrays, were used to estimate the impact of avian predation relative to total smolt mortality. Results indicated that avian predation was a substantial source of steelheadmortality, with predation probability (proportion of available fish consumed by birds) ranging from 0.06 to 0.28 for fish traveling through the lower Snake River and the lower and middle Columbia River. Predation probability estimates ranged from 0.03 to 0.09 for available tagged yearling Chinook Salmon and from 0.01 to 0.05 for subyearlings. Smolt predation by gulls Larus spp. was concentrated near hydroelectric dams, while predation by Caspian terns Hydroprogne caspia was concentrated within reservoirs. No concentrated areas of predation were identified for double-crested cormorants Phalacrocorax auritus or American white pelicans Pelecanus erythrorhynchos. Comparisons of total smolt mortality relative to mortality from colonial waterbirds indicated that avian predation was one of the greatest sources of mortality for steelhead and yearling Chinook Salmon during out-migration. In contrast, avian predation on subyearling Chinook Salmon was generally low and constituted a minor component of total mortality. Our results demonstrate that acoustic and PIT tag technologies can be combined to quantify where and when smolt mortality occurs and the fraction of mortality that is due to colonial waterbird predation relative to non-avian mortality sources.

We evaluated predation on Lost River Suckers Deltistes luxatus and Shortnose Suckers Chasmistes brevirostris by American white pelicans Pelecanus erythrorhynchos and double-crested cormorants Phalacrocorax auritus nesting at mixed-species colonies in the Upper Klamath Basin of Oregon and California during 2009-2014. Predation was evaluated by recovering (detecting) PIT tags from tagged fish on bird colonies and calculating minimum predation rates, as the percentage of available suckers consumed, adjusted for PIT tag detection probabilities but not deposition probabilities (i.e., probability an egested tag was deposited on-or off-colony). Results indicate that impacts of avian predation varied by sucker species, age-class (adult, juvenile), bird colony location, and year, demonstrating dynamic predator-prey interactions. Tagged suckers ranging in size from 72 to 730 mm were susceptible to cormorant or pelican predation; all but the largest Lost River Suckers were susceptible to bird predation. Minimum predation rate estimates ranged annually from <0.1% to 4.6% of the available PIT-tagged Lost River Suckers and from <0.1% to 4.2% of the available Shortnose Suckers, and predation rates were consistently higher on suckers in Clear Lake Reservoir, California, than on suckers in Upper Klamath Lake, Oregon. There was evidence that bird predation on juvenile suckers (species unknown) in Upper Klamath Lake was higher than on adult suckers in Upper Klamath Lake, where minimum predation rates ranged annually from 5.7% to 8.4% of available juveniles. Results suggest that avian predation is a factor limiting the recovery of populations of Lost River and Shortnose suckers, particularly juvenile suckers in Upper Klamath Lake and adult suckers in Clear Lake Reservoir. Additional research is needed to measure predator-specific PIT tag deposition probabilities (which, based on other published studies, could increase predation rates presented herein by a factor of roughly 2.0) and to better understand biotic and abiotic factors that regulate sucker susceptibility to bird predation.

Trait-selective mortality is of considerable management and conservation interest, especially when trends are similar across multiple species of conservation concern. In the Columbia River basin, thousands of juvenile Pacific salmonids Oncorhynchus spp. are collected each year and are tagged at juvenile bypass system (JBS) facilities located at hydroelectric dams, thus allowing the tracking of population-level performance metrics (e.g., juvenile survival and juvenile-to-adult survival). Several studies have suggested that juvenile salmonid survival is both size dependent and condition dependent, but little is known about trait-selective collection at JBS facilities. Trait-selective collection (e.g., length-based or condition-based selectivity) is particularly important, as inferences to population-level performance metrics may be biased if both the survival and collection processes are influenced by similar characteristics. We used a capture-mark-recapture study to investigate length-and condition-selective survival and detection probabilities for two salmonid species in the Columbia River basin. In 2014, juvenile steelhead O. mykiss (n = 11,201) and yearling Chinook Salmon O. tshawytscha (n = 7,943) were PIT-tagged, measured (FL), examined for external condition characteristics (descaling, body injuries, fin damage, or disease symptoms), and released into the Lower Granite Dam JBS facility on the Snake River to continue seaward migration. Results indicated similar trends in both length-and condition-selective juvenile survival and detection probabilities. For both species, survival probability was higher for longer, nondegraded individuals (those without descaling, body injuries, or fin damage). Trends in detection probability were also consistent across species: shorter, degraded individuals were more likely to be detected at downstream JBS facilities than longer, healthier individuals. These results suggest that similar characteristics (FL and external condition) affect survival and detection processes for PIT-tagged steelhead and yearling Chinook Salmon and that JBS facilities may selectively collect smaller, degraded individuals with lower probabilities of survival. The consistency in trait-selective survival and detection results has important management implications for several species of conservation concern.

Double-crested cormorants (Phalacrocorax auritus) have been identified as the source of significant mortality to juvenile salmonids (Oncorhynchus spp.) in the Columbia River Basin. Management plans for reducing the size of a large colony on East Sand Island (OR, USA) in the Columbia River estuary are currently being developed. We evaluated habitat enhancement and social attraction as nondestructive techniques for managing cormorant nesting colonies during 2004–2007. We tested these techniques on unoccupied plots adjacent to the East Sand Island cormorant colony. Cormorants quickly colonized these plots and successfully raised young. Cormorants also were attracted to nest and raised young on similar plots at 2 islands approximately 25 km from East Sand Island; 1 island had a history of successful cormorant nesting whereas the other was a site where cormorants had previously nested unsuccessfully. On a third island with no history of cormorant nesting or nesting attempts, these techniques were unsuccessful at attracting cormorants to nest. Our results suggest that some important factors influencing attraction of nesting cormorants using these techniques include history of cormorant nesting, disturbance, and presence of breeding cormorants nearby. These techniques may be effective in redistributing nesting cormorants away from areas where fish stocks of conservation concern are susceptible to predation, especially if sites with a recent history of cormorant nesting are available within their foraging or dispersal range. Published 2015. Wiley Periodicals, Inc. This article is a US Government work and, as such, is in the public domain in the United States of America.

Accurate assessment of specific mortality factors is vital to prioritize recovery actions for threatened and endangered species. For decades, tag recovery methods have been used to estimate fish mortality due to avian predation. Predation probabilities derived from fish tag recoveries on piscivorous waterbird colonies typically reflect minimum estimates of predation due to an unknown and unaccounted-for fraction of tags that are consumed but not deposited on-colony (i.e., deposition probability). We applied an integrated tag recovery modeling approach in a Bayesian context to estimate predation probabilities that accounted for predator-specific tag detection and deposition probabilities in a multiple-predator system. Studies of PIT tag deposition were conducted across three bird species nesting at seven different colonies in the Columbia River basin, USA. Tag deposition probabilities differed significantly among predator species (Caspian terns Hydroprogne caspia: deposition probability = 0.71, 95% credible interval [CRI]=0.51-0.89; double-crested cormorants Phalacrocorax auritus: 0.51, 95% CRI = 0.34-0.70; California gulls Larus californicus: 0.15, 95% CRI = 0.11-0.21) but showed little variation across trials within a species or across years. Data from a 6-year study (2008-2013) of PIT-tagged juvenile Snake River steelhead Oncorhynchus mykiss (listed as threatened under the Endangered Species Act) indicated that colony-specific predation probabilities ranged from less than 0.01 to 0.17 and varied by predator species, colony location, and year. Integrating the predator-specific deposition probabilities increased the predation probabilities by a factor of approximately 1.4 for Caspian terns, 2.0 for double-crested cormorants, and 6.7 for California gulls compared with traditional minimum predation rate methods, which do not account for deposition probabilities. Results supported previous findings on the high predation impacts from strictly piscivorous waterbirds nesting in the Columbia River estuary (i.e., terns and cormorants), but our findings also revealed greater impacts of a generalist predator species (i.e., California gulls) than were previously documented. Approaches used in this study allow for direct comparisons among multiple fish mortality factors and considerably improve the reliability of tag recovery models for estimating predation probabilities in multiple-predator systems.

Understanding how individual characteristics are associated with survival is important to programs aimed at recovering fish populations of conservation concern. To evaluate whether individual fish characteristics observed during the juvenile life stage were associated with the probability of returning as an adult, juvenile steelhead Oncorhynchus mykiss from two distinct population segments (DPSs; Snake River and upper Columbia River) were captured, photographed to determine external condition (body injuries, descaling, signs of disease, fin damage, and ectoparasites), measured, classified by rearing type (hatchery, wild), marked with a PIT tag, and released to continue out-migration to the Pacific Ocean during 2007-2010. The PIT tags of returning adults were interrogated in fishways at hydroelectric dams on the lower Columbia River 1-3 years following release as juveniles. Juvenile-to-adult survival models were investigated independently for each DPS and indicated that similar individual fish characteristics were important predictors of survival to adulthood for both steelhead populations. The data analysis provided strong support for survival models that included explanatory variables for fish length, rearing type, and external condition, in addition to out-migration year and timing. The probability of a juvenile surviving to adulthood was positively related to length and was higher for wild fish compared with hatchery fish. Survival was lower for juveniles with body injuries, fin damage, and external signs of disease. Models that included variables for descaling and ectoparasite infestation, however, had less support than those that incorporated measures of body injuries, fin damage, and disease. Overall, results indicated that individual fish characteristics recorded during the juvenile life stage can be used to predict adult survivorship in multiple steelhead populations.

Enhancing the survival of juvenile salmonid (Oncorhynchus spp.) is a priority objective to recover populations of Columbia River salmonids listed under the U.S. Endangered Species Act (ESA). In the Columbia River estuary, a significant mortality factor for juvenile salmonids is predation by doublecrested cormorants (Phalacrocorax auritus) nesting at East Sand Island. The U.S. Army Corps of Engineers is considering management alternatives to reduce this mortality. Understanding the factors that influence cormorant predation is important to understanding the potential consequences of various management strategies. We used principal components regression (PCR) to evaluate the relationship between several annual measures of cormorant predation and a combination of colony size and environmental covariates. The environmental factors considered included large-scale climate indices (Pacific Decadal Oscillation, El Nino/Southern Oscillation Index, North Pacific Gyre Oscillation, Pacific Northwest Index), regional climate measures (sea surface temperature, upwelling strength, upwelling timing), and variables describing conditions during freshwater and estuarine outmigration (river discharge, spill at hydroelectric dams, measures of salmonid smolt survival to the estuary). These covariates potentially influenced both the susceptibility of salmonids to cormorant predation and the abundance and distribution of marine forage fish and their availability as alternative prey for cormorants nesting in the estuary. Measures of cormorant predation spanned a 15-year period (1999 – 2013) and included (1) predation probabilities for multiple steelhead (O. mykiss) and Chinook salmon (O. tshawytscha) populations derived from recoveries of salmonid passive integrated transponder (PIT) tags at the cormorant colony, (2) estimates of annual consumption of steelhead and yearling Chinook by cormorants derived using bioenergetics modelling, and (3) the observed percentage of the cormorant diet that consisted of salmonids. We also related cormorant diet composition to purse seine catches in the estuary during 2007 – 2012 to assess how predation on salmonids is related to availability of alternative, non-salmonid prey and to examine cormorant selectivity of salmonids relative to other available prey. PCR analyses indicated that environmental factors explain a substantial proportion of the annual variability seen in several measures of cormorant predation on Columbia River juvenile salmonids. Cormorant colony size was an important explanatory factor in most regressions; however, it never explained more than 17% of the variability in any annual measure of cormorant predation on salmonids. In aggregate, environmental factors explained a greater proportion of the annual variability in cormorant predation than did colony size; in particular, river discharge and the North Pacific Gyre Oscillation (NPGO) were prominent environmental explanatory factors. Based on comparisons to estuary purse seine catches, cormorants appeared to take salmonids in proportion to their relative availability in the Columbia River estuary, not their absolute abundance. Conversely, changes in absolute abundance of alternative prey, both marine and freshwater/estuarine forage fishes, did influence how much cormorants relied on salmonids as prey. While colony size is an important determinant of cormorant impacts on salmonid populations, environmental conditions that regulate the availability of alternative prey might outweigh the effects of changing colony size in any given year. Potential management efforts to reduce the size of the double-crested cormorant colony on East Sand Island to benefit ESA-listed salmonids would best be evaluated in the context of environmental conditions, particularly if evaluation occurs on an annual basis, with special attention given to river discharge and the NPGO. Multiyear data sets following any implementation of management would likely be more useful to evaluate potential benefits.

Abstract We investigated colony size, productivity, and limiting factors for five piscivorous waterbird species nesting at 18 locations on the Columbia Plateau (Washington) during 2004–2010 with emphasis on species with a history of salmonid (Oncorhynchus spp.) depredation. Numbers of nesting Caspian terns (Hydroprogne caspia) and double-crested cormorants (Phalacrocorax auritus) were stable at about 700–1,000 breeding pairs at five colonies and about 1,200–1,500 breeding pairs at four colonies, respectively. Numbers of American white pelicans (Pelecanus erythrorhynchos) increased at Badger Island, the sole breeding colony for the species on the Columbia Plateau, from about 900 individuals in 2007 to over 2,000 individuals in 2010. Overall numbers of breeding California gulls (Larus californicus) and ring-billed gulls (L. delawarensis) declined during the study, mostly because of the abandonment of a large colony in the mid-Columbia River. Three gull colonies below the confluence of the Snake and Columbia rivers increased substantially, however. Factors that may limit colony size and productivity for piscivorous waterbirds nesting on the Columbia Plateau included availability of suitable nesting habitat, interspecific competition for nest sites, predation, gull kleptoparasitism, food availability, and human disturbance. Based on observed population trends alone, there is little reason to project increased impacts to juvenile salmonid survival from tern and cormorant populations. Additional monitoring and evaluation may be warranted to assess future impacts of the growing Badger Island American white pelican colony and those gull colonies located near mainstem dams or associated with Caspian tern colonies where kleptoparasitism is common.

The status of the double-crested cormorant (Phalacrocorax auritus) in western North America was last evaluated during 1987-2003. In the interim, concern has grown over the potential impact of predation by double-crested cormorants on juvenile salmonids (Oncorhynchus spp.), particularly in the Columbia Basin and along the Pacific coast where some salmonids are listed for protection under the United States Endangered Species Act. Recent re-evaluations of double-crested cormorant management at the local, flyway, and federal level warrant further examination of the current population size and trends in western North America. We collected colony size data for the western population (British Columbia, Washington, Oregon, Idaho, California, Nevada, Utah, Arizona, and the portions of Montana, Wyoming, Colorado and New Mexico west of the Continental Divide) by conducting aircraft-, boat-, or ground-based surveys and by cooperating with government agencies, universities, and non-profit organizations. In 2009, we estimated approximately 31,200 breeding pairs in the western population. We estimated that cormorant numbers in the Pacific Region (British Columbia, Washington, Oregon, and California) increased 72% from 1987-1992 to circa 2009. Based on the best available data for this period, the average annual growth rate () of the number of breeding birds in the Pacific Region was 1.03, versus 1.07 for the population east of the Continental Divide during recent decades. Most of the increase in the Pacific Region can be attributed to an increase in the size of the nesting colony on East Sand Island in the Columbia River estuary, which accounts for about 39% of all breeding pairs in the western population and is the largest known breeding colony for the species (12,087 breeding pairs estimated in 2009). In contrast, numbers of breeding pairs estimated in coastal British Columbia and Washington have declined by approximately 66% during this same period. Disturbance at breeding colonies by bald eagles (Haliaeetus leucocephalus) and humans are likely limiting factors on the growth of the western population at present. Because of differences in biology and management, the western population of double-crested cormorants warrants consideration as a separate management unit from the population east of the Continental Divide. Published 2014. This article is a U.S. Government work and is in the public domain in the USA.

Identification of the factors that influence susceptibility to predation can aid in developing management strategies to recover fish populations of conservation concern. Predator-prey relationships can be influenced by numerous factors, including prey condition, prey size, and environmental conditions. We investigated these factors by using juvenile steelhead Oncorhynchus mykiss from the Snake River (Pacific Northwest, USA), a distinct population segment that is listed as threatened under the U. S. Endangered Species Act. During 2007-2009, steelhead smolts (n = 25,909) were captured, examined for external condition characteristics (e. g., body injuries, descaling, external signs of disease, fin damage, and ectoparasite infestations), marked with passive integrated transponder (PIT) tags, and released to continue their out-migration. Recoveries of PIT tags on a downstream colony of Caspian terns Hydroprogne caspia (n = 913 tags) indicated that steelhead susceptibility to Caspian tern predation increased significantly with decreases in steelhead external condition, decreased water discharge, and decreased water clarity. Susceptibility to Caspian tern predation also increased with increasing steelhead fork length up to 202 mm but then decreased for longer steelhead. Recoveries of PIT tags on a downstream colony of double-crested cormorants Phalacrocorax auritus (n = 493 tags) indicated that steelhead susceptibility to double-crested cormorant predation increased significantly with declining external condition of steelhead, and that steelhead of hatchery origin were more susceptible than their wild counterparts. Results indicate that steelhead susceptibility to avian predation is dependent on fish condition and length and is influenced by river conditions and rearing environment.

San Francisco Bay is a proposed relocation site for some of the Caspian terns Hydroprogne caspia currently nesting at the world's largest colony for the species in the Columbia River estuary and consuming salmonids listed under the U. S. Endangered Species Act (ESA). However, several runs of salmonids listed under the ESA occur in San Francisco Bay and managers are concerned that increased Caspian tern predation may pose a threat to the recovery of these fish. We used a bioenergetics modeling approach, employing estimates of tern energy requirements and proportions of energy supplied by various prey types, to estimate the consumption of juvenile salmonids by Caspian terns nesting on Brooks Island in central San Francisco Bay during 2008 and 2009. Estimated salmonid consumption was similar to 205,000 smolts (95% confidence interval, 175,000-245,000 smolts) in 2008 and similar to 167,000 smolts (144,000-191,000 smolts) in 2009. The interannual difference in smolt consumption was due to the smaller size of the tern colony and lower nesting success in 2009. Estimated predation rates on ESA-listed Central Valley spring-run Chinook salmon Oncorhynchus tshawytscha (0.1%) were lower than those on unlisted fall-run Chinook salmon (1.0%). Continuation of the current downward trend in the number of Caspian terns nesting on Brooks Island and the resulting reductions in salmonid predation would not be sufficient to reverse salmonid declines in San Francisco Bay. The proposed enhancement of the Brooks Island Caspian tern colony to 3,000 individuals would at most cause declines in annual population growth rates of 0.28% for fall-run Chinook salmon and 0.02% for threatened spring-run Chinook salmon, assuming that the mortality from tern predation is 100% additive.