Estimating Escapement for a Low-Abundance Steelhead Population Using Dual-Frequency Identification Sonar (DIDSON)

Abstract

Estimating the escapement of small populations of steelhead Oncorhynchus mykiss (ocean-migrating rainbow trout) is challenging and requires innovative methods. Difficulties arise from rare and episodic occurrence of individuals, high turbidity levels during migration, and the need to minimize jeopardy to the fish, which have led to a lack of population estimates, especially in California. Here we show that dual-frequency identification sonar (DIDSON) can be used to produce escapement estimates for a small steelhead population in Scott Creek, Santa Cruz County, California, with virtually no impact on the fish. The DIDSON uses sound to form near-video quality images and passively monitors fish without the need to handle them or constrict passage. We deployed a DIDSON and recorded steelhead passage over three spawning seasons (2008–2010). We used a decision support tool to analyze DIDSON images and compared the resulting estimates (153, 57, and 84) with raw weir counts (50, 23, and 40), mark–recapture estimates (293 ± 9, 126 ± 12, and 109 ± 34) generated over the entire migration period, and adjusted mark–recapture estimates (201, 74, and 85), which coincided with the period of DIDSON deployment. The DIDSON and weir estimates were restricted to a smaller sampling window due to the installation of downstream migrant traps causing increased incidences of milling fish that interfered with the DIDSON results late in the migration season. The DIDSON estimates were two to three times higher than weir estimates and 23% to 55% less than the full-season mark–recapture estimates. The adjusted mark–recapture estimates followed the same trends as the full-season mark–recapture estimates and were correlated with the DIDSON estimates. We conclude that DIDSON is an effective tool to generate steelhead escapement estimates, but it is important to collect data over the entire migration season and to consider fish behavior and potential species identification issues during analysis.

Received October 6, 2011; accepted May 21, 2012

ACKNOWLEDGMENTS

Funding for this study was provided by a Joint Institute of Marine Observations grant between the National Oceanic and Atmospheric Administration and the University of California–Santa Cruz, and the Fisheries Restoration Grant Program. Field assistance was provided by Mark Jessop, Lewis Barnett, Alison Collins, Michael Feldman, Morgan Bond, Danielle Frechette, Andrew Jones, Jeff Perez, Ann-Marie Osterback, Nicolas Retford, and Andrew Sobieraj. Landowner access was provided by California Polytechnic State University's Swanton Pacific Ranch. We appreciate insightful comments from David Boughton, Dave Rundio, and Steve Lindley on earlier drafts of this manuscript. We are also grateful to the associate editor, two anonymous reviewers, and Patrick Connolly for providing constructive comments that greatly improved the manuscript. Reference to trade names does not imply endorsement by the U.S. Government.