In marine ecosystems, survival can be heavily influenced by size‐selective mortality during juvenile life stages. Understanding how and when size‐selective mortality operates on a population can reveal underlying growth dynamics and size‐selective ecological processes affecting the population and thus can be used to guide conservation efforts. For subyearling Chinook Salmon Oncorhynchus tshawytscha in Puget Sound, previous research reported a strong positive relationship between marine survival and body mass during midsummer in epipelagic habitats within Puget Sound, suggesting that early marine growth drives survival. However, a fine‐scale analysis of size‐selective mortality is needed to identify specific critical growth periods and habitats. The objectives of this study were to (1) describe occupancy patterns across estuarine delta, nearshore marine, and offshore epipelagic habitats in Puget Sound; (2) describe changes in FL and weight observed across habitats and time; (3) evaluate evidence for size‐selective mortality; and (4) illustrate how marine survival of the stocks studied may be affected by variation in July weight. In 2014 and 2015, we sampled FLs, weights, and scales from seven hatchery‐origin and two natural‐origin stocks of subyearling Chinook Salmon captured every 2 weeks during out‐migration and rearing in estuary, nearshore, and offshore habitats within Puget Sound. Natural‐origin stocks had more protracted habitat occupancy patterns than hatchery‐origin stocks and were smaller than hatchery‐origin stocks in both years. Regardless of origin, subyearlings were longer and heavier and grew faster in offshore habitats compared to estuary and nearshore habitats. For all stocks, we found little evidence of size‐selective mortality among habitats in Puget Sound. These patterns were consistent in both years. Finally, the weights of subyearlings sampled during July in the offshore habitat predicted Puget Sound‐wide marine survival rates of 0.4% for 2014 and 2.0% for 2015, with stock‐specific predictions ranging from 0.18% to 11.70%.

. 2018. Trophic compression of lake food webs under hydrologic disturbance. Ecosphere 9(6):e02304. 10. 1002/ecs2.2304 Abstract. The need to protect biostructure is increasingly recognized, yet empirical studies of how human exploits affect ecological networks are rare. Studying the effects of variation in human disturbance intensity from decades past can help us understand and anticipate ecosystem change under alleviated or amplified disturbance over decades to come. Here, we use stable isotopes and an innovative analytical approach to compare the food webs of two akin lake ecosystems subject to disparate water use regimes, a pervasive, yet unappreciated stressor. We show that intensive water use (persistent, early season, rapid lake-level drawdown) can compress trophic diversity by 46%, necessitating reorganization of biostructural elements configuring lake food webs. Compression occurred over the d 13 C axis indicating erosion of basal trophic diversity, but food chain length remained intact over the period and intensity of disturbance examined. This study demonstrates the potential for water use, like other disturbances (warming, eutrophication, and invasive species), to mute opportunity for benthic-pelagic coupling and benefits to lake food webs and the inherent capacity of lake ecosystems to adapt to stress. Trophically compressed lakes may be less able to adapt to intensified water use.

The feasibility of reintroducing anadromous salmonids into reservoirs above high‐head dams is affected by the suitability of the reservoir habitat for rearing and the interactions of the resident fish with introduced fish. We evaluated the predation risk to anadromous salmonids considered for reintroduction in Merwin Reservoir on the North Fork Lewis River in Washington State for two reservoir use‐scenarios: year‐round rearing and smolt migration. We characterized the role of the primary predators, Northern Pikeminnow Ptychocheilus oregonensis and tiger muskellunge (Northern Pike Esox lucius × Muskellunge E. masquinongy), by using stable isotopes and stomach content analysis, quantified seasonal, per capita predation using bioenergetics modeling, and evaluated the size and age structures of the populations. We then combined these inputs to estimate predation rates of size‐structured population units. Northern Pikeminnow of FL ≥ 300 mm were highly cannibalistic and exhibited modest, seasonal, per capita predation on salmonids, but they were disproportionately much less abundant than smaller, less piscivorous, conspecifics. The annual predation on kokanee Oncorhynchus nerka (in biomass) by a size‐structured unit of 1,000 Northern Pikeminnow having a FL ≥ 300 mm was analogous to 16,000–40,000 age‐0 spring Chinook Salmon O. tshawytscha rearing year‐round, or 400–1,000 age‐1 smolts migrating April–June. The per capita consumption of salmonids by Northern Pikeminnow having a FL ≥ 200 mm was relatively low, due in large part to spatial segregation during the summer and the skewed size distribution of the predator population. Tiger muskellunge fed heavily on Northern Pikeminnow, other nonsalmonids, and minimally on salmonids. In addition to cannibalism within the Northern Pikeminnow population, predation by tiger muskellunge likely contributed to the low recruitment of larger (more piscivorous) Northern Pikeminnow, thereby decreasing the risk of predation to salmonids. This study highlights the importance of evaluating trophic interactions within reservoirs slated for reintroduction with anadromous salmonids, as they can be functional migration corridors and may offer profitable juvenile‐rearing habitats despite hosting abundant predator populations.Received September 3, 2015; accepted December 7, 2015 Published online April 19, 2016