Migration Trends of Sockeye Salmon at the Northern Edge of Their Distribution

Abstract

Climate change is affecting arctic and subarctic ecosystems, and anadromous fish such as Pacific salmon Oncorhynchus spp. are particularly susceptible due to the physiological challenge of spawning migrations. Predicting how migratory timing will change under Arctic warming scenarios requires an understanding of how environmental factors drive salmon migrations. Multiple mechanisms exist by which environmental conditions may influence migrating salmon, including altered migration cues from the ocean and natal river. We explored relationships between interannual variability and annual migration timing (2003–2014) of Sockeye Salmon O. nerka in a subarctic watershed with environmental conditions at broad, intermediate, and local spatial scales. Low numbers of Sockeye Salmon have returned to this high-latitude watershed in recent years, and run size has been a dominant influence on the migration duration and the midpoint date of the run. The duration of the migration upriver varied by as much as 25 d across years, and shorter run durations were associated with smaller run sizes. The duration of the migration was also extended with warmer sea surface temperatures in the staging area and lower values of the North Pacific Index. The midpoint date of the total run was earlier when the run size was larger, whereas the midpoint date was delayed during years in which river temperatures warmed earlier in the season. Documenting factors related to the migration of Sockeye Salmon near the northern limit of their range provides insights into the determinants of salmon migrations and suggests processes that could be important for determining future changes in arctic and subarctic ecosystems.

Received October 24, 2016; accepted March 2, 2017 Published online June 8, 2017

ACKNOWLEDGMENTS

We thank the staff at the Norton Sound Economic Development Corporation for their continued fieldwork on the Pilgrim River; the JISAO (University of Washington, Seattle) for making the PDO data available; and the National Center for Atmospheric Research (Boulder, Colorado) for maintaining accessibility of the NPI data. We are grateful to J. Koch and R. Kovach for valuable insights on an earlier version of the manuscript. In addition, we thank P. Westley and an anonymous reviewer for comments that improved the paper. This project was funded in part by the Arctic–Yukon–Kuskokwim Sustainable Salmon Initiative (www.aykssi.org). Use of product or trade names is for descriptive purposes only and does not imply endorsement by the U.S. Government.