RESEARCH ARTICLE Balancing competing life-stage requirements in salmon habitat rehabilitation: between a rock and a hard place Joseph Merz 1,2 , Lucius Caldwell 1 , Michael Beakes 1 , Chris Hammersmark 3 , Kirsten Sellheim 1,4 Gravel augmentation is often applied to rivers and streams to rehabilitate salmonid spawning and incubation habitat. However, the effect of gravel size on salmon spawning utilization and embryo survival during incubation is not well understood. We conducted an experiment on a regulated and previously mined Northern California salmonid-bearing stream in which different sized gravel (small, medium, and large) patches were placed into the stream’s degraded spawning reach. We documented Oncorhynchus tshawytscha (Chinook salmon) spawning activity within the three gravel sizes for two seasons. In addition, we deployed Chinook salmon embryos into each gravel size patch and allowed them to incubate until estimated emergence time. Although all experimental gravel sizes were predicted to be within the spawning population’s mobilization capabilities, model results indicated the probability of salmon building redds decreased as substrate size increased. Conversely, embryo survival increased as gravel size increased. A possible mechanism of disparate Chinook salmon embryo survival is provided by an observed decrease in embryo survival correlating with greater presence of embryo predators (leeches), which are associated with smaller gravel. Our results indicate a parent-offspring confict in optimal spawning gravel size for Chinook salmon, and suggest that an intermediate gravel size would maximize overall reproductive success across both spawning and incubation life stages. Key words: Chinook salmon, embryo survival, gravel augmentation, parent-offspring confict, salmon embryo incubation, spawning habitat rehabilitation Implications for Practice • The response of different salmon life stages to restoration actions is neither uniform nor unidirectional, and differs among life stages with potentially conficting needs. • To optimize salmonid habitat restoration effectiveness, resource managers must consider such competing require- ments of multiple life stages. • The specifc gravel size ranges predicted by this study can be used during the design of gravel augmentation projects to jointly optimize female Chinook salmon spawning activity and embryo survival. Introduction River restoration is a multibillion U.S. dollar industry encompassing infrastructure removal, channel and foodplain modifcation, fow manipulation, and nutrient and sediment management (Bernhardt et al. 2005; Roni et al. 2005). These actions aim to mitigate anthropogenic effects including habitat simplifcation and disruption of fuvial and sediment dynamics associated with river regulation and mining (Nelson et al. 1987). Coarse sediment (e.g. gravel) is frequently added to impaired streams to improve salmonid (Salmonidae) habitats (Merz & Setka 2004) by increasing spawning opportunities, food production, and habitat connectivity (Gore et al. 1998; Merz & Ochikubo Chan 2005). Coarse sediment augmentation can beneft multiple salmonid life stages (Merz et al. 2004; Zeug et al. 2013; Sellheim et al. 2016). However, despite several studies evaluating effects on spawning and incubation (Sear & DeVries 2008; Johnson et al. 2012; Roni et al. 2016), the infuence of specifc sediment size on spawner site selection and subsequent embryo survival has not been well examined. Previous research demonstrates that Oncorhynchus tshawytscha (Chinook salmon) females build redds (nests) within a range of suitable water depths and velocities (Geist & Dauble 1998). Studies by Riebe et al. (2014), Zeug et al. (2013) and others have also shown that sediment size can limit Chinook salmon ability to successfully build redds. A female spawner’s physical ability to excavate a redd is limited by the upper grain size of the channel bed surface (Kondolf & Wolman 1993) and the female’s body length (Riebe et al. 2014). Although smaller sediments are easier for females to mobilize, they can Author contributions: JM, KS conceived, designed, and implemented the feld experiment; LC, KS, MB, JM analyzed the data; CH performed the 2D modeling; JM, LC, KS wrote and edited the manuscript. 1 Cramer Fish Sciences, 3300 Industrial Boulevard #100, West Sacramento, CA 95691, U.S.A. 2 Department of Ecology and Evolutionary Biology, University of California Santa Cruz, 100 Shaffer Road, Santa Cruz, CA, 95060, U.S.A. 3 CBEC Eco Engineering Inc, 2544 Industrial Boulevard, West Sacramento, CA, 95691, U.S.A. 4 Address correspondence to K. Sellheim, email kirstens@fshsciences.net © 2018 Society for Ecological Restoration doi: 10.1111/rec.12900 Supporting information at: http://onlinelibrary.wiley.com/doi/10.1111/rec.12900/suppinfo Restoration Ecology 1