Population Structure of Columbia River Basin Chinook Salmon and Steelhead Trout

Abstract

The population structure of chinook salmon and steelhead trout is presented as an assimilation of the life history forms that have evolved in synchrony with diverse and complex environments over their Pacific range. As poikilotherms, temperature is described as the overwhelming environmental influence that determines what life history options occur. The different populations represent ecological types referred to as spring-, summer-, fall, and winter-run segments, as well as stream- and ocean-type, or stream- and ocean-maturing life history forms. However, they are more correctly described as a continuum of forms that fall along a temporal cline related to incubation and rearing temperatures that determine spawn timing and juvenile residence patterns. Population structure of chinook salmon and steelhead in the Columbia Basin, therefore, is the reflection of the genetic composition of the founding source or sources within the respective region, shaped by the environment, principally temperature, that defines life history evolutionary strategy to maximize fitness under the conditions delineated. The key in developing an effective recovery program for chinook salmon and steelhead is to recognize that measures taken must address the genetic and biological requirements of the population unit within the environmental template identified.

Keywords:

• salmonid life history
• salmonid genetics
• temperature

Acknowledgments

This project was funded by the National Science Foundation (OSR-9350539) and the Bonneville Power Administration, project number 98-004-03 (contract 98BI08319). We thank Larry Beck and Gretchen Starkey (USACE), Chuck Peven and Sandi Wrzesinski (CCPUD), and Sayre Hodgson (UW) for help compiling data. We are grateful to Steve Cramer (S.P. Cramer & Assp); Colleen Harvey Arrison (CDFG); Suzanne Hayes (ADFG); Mark LaRiviere TPUD; Wolf Dammer, Howard Fuss, Andrew Murdock, and Jim Sneva (WDFW); William Garrigues (USFS, Natches); Jim Muck (ODFW); Brent Snider (IDFG); and Thomas Troch and Carl Burger (USFWS) for their assistance in getting temperature and population data used in this report. We are also very grateful to Tracy Cone, Pete Etherton, Grant Ladouceur, Doug Lofthouse, Russ Hilland, Dave McNeil, Brian Munro, Nick Nagtegaal and David Peacock (DFO), and Brent Lister (D. B. Lister and Associates) for the temperature records and chinook salmon spawning data in B.C. rivers. We acknowledge the scores of employees for the Army Corps of Engineers, Chelan Co. PUD, and Grant Co. PUD who spent long (approximately 750,000) hours counting fish passing dam projects (those used in this study) on the Columbia and Snake Rivers over the last 60 years, and the staff of the StreamNet library in Oregon (available on the worldwide web at URL). We also gratefully recognize the untiring dedication of Keya Collins at the University of Idaho for her work in developing tables and figures, and for her review, and editing of the manuscript. Life history scenarios were developed from data on abundance, run timing, and spawning dates from Idaho Department of Fish and Game, Oregon Department of Fish and Wildlife, Washington Department of Fish and Wildlife, StreamNet, Pacific States Marine Fish Commission, Columbia River Fisheries Management Plan Technical Advisory Committee, and Department of Fisheries and Oceans of Canada. We also recognize Myers et al. (1998), Busby et al. (1996), Chapman et al. (1994, 1995), and Mullan et al. (1992) for the wealth of data that assisted in the development of this paper. Finally, we are indebted to Don Campton (USFWS) for his assistance and contributions in assessing the genetic relationships among the populations studied and for his editorial comments on the manuscript.

Notes

¹ In this paper the Columbia River Basin is divided into the lower Columbia, mid-Columbia, and Snake River regions, each of which is further subdivided into several sub-basins. We define the lower Columbia as the region extending from the mouth of the Columbia to the confluence of the Snake River. The mid-Columbia is that region extending from the confluence of the Snake River upstream to the Grand Coulee Dam. The Snake River region encompasses the entire Snake River basin.

² The Refugia Hypothesis were grouped into (1) Southwestern B.C. rivers such as the Tahuya, Vancouver and Queen Charlotte Islands; (2) Central, which included the Skeena, Kitimat, Nass and Dean Rivers, North Coast (incl. Alaska) and Yukon, Upper Fraser, Peace/Williston Rivers, Athabasca River; (3) Inland group, which included the Columbia and Snake. The Behnke hypothesis is from Behnke (1992).

³ A population segment is not analogous with the Evolutionary Significant Unit (ESU) of Waples 1991 and should not be confused with the ESU concept. The ESU designation was adopted by the National Marine Fisheries Service to represent “distinct population segments” under the Endangered Species Act, and technically serves a different purpose. Utter et al. (1995) state the “definition of an ESU by no means implies a single panmictic unit.” The population segment referred to here is a single panmictic unit, and represents the focus of recovery recommendations.

⁴ However, see Campton, Allendorf, Behnke, and Utter, 1991; Chilcote, Leider, and Loch, 1991 for comments and reply to the paper by Chilcote, Leider, and Loch, 1986.

⁵ We use the term “putatively domesticated” because there are no measures of domestication nor norms for determining domesticated stocks (Doyle, 1983). Further, domestication is a process and not a state. Fish that remain in hatcheries are subject to selection in that environment as they are under any other environment, and a fish adapting to the hatchery environment would thus be considered “domesticated.” Domestication in hatchery fish is believed to reflect three independent processes: (1) direct artificial selection by hatchery personnel, (2) natural selection in an artificial environment, and (3) relaxation of natural selection that occurs in the wild. Therefore, the mechanism for domestication would be principally selection on a set of traits related to fitness in the homogeneous environments, but also to inbreeding (i.e., absence of gene flow among populations, gradual loss of additive genetic variance). There are several studies that show that rainbow trout raised in hatcheries for the entire life cycle for several generations do poorly when released in the wild (Flick and Webster, 1976). It is unclear whether hatchery-reared fish that spend a large fraction of their life in the wild can become domesticated in the same sense (Campton, 1995; Purdom, 1994; Wohlfarth, 1993; Rhodes and Quinn, 1999).