Bias and consequences in attempts to estimate historical salmon abundance

Author:

Cramer Steven P.1,Caldwell Lucius2

Affiliation:

1. Cramer Fish Sciences, 300 SE Arrow Creek Lane, Gresham, OR 97080, USA.

2. Cramer Fish Sciences, 7525 NE Ambassador Pl., Suite C, Portland, OR 97220, USA.

Abstract

Status of salmon populations thought to be at risk is frequently evaluated by comparing estimated abundances in recent decades with approximations of peak abundance in the era when fisheries began. However, spawner escapements of Pacific salmon were generally not surveyed until 1950 or later, so most estimates of salmon abundance before 1950 have been derived from harvest data alone. We investigated historical data on Oregon Coast coho (Oncorhynchus kisutch) to test the reliability of such estimates. In the “Recovery Plan for Oregon Coast Coho” (prepared by National Marine Fisheries Service (NMFS) in 2016), coho abundance was estimated back to 1892 by assuming that in-river landings before 1950 represented a constant 40% of the population. The first mark–recapture studies enabling estimation of harvest rates were completed in the early 1950s, shortly after regulatory action in 1947 sharply curtailed harvesting of coho. We used annual data on fishing licenses sold and length of fishing seasons before 1950 as indices of harvesting effort to reconstruct harvest rates before 1947 and determined in-river landings were near 80%, twice the level postulated by the Recovery Plan. Thus, spawner escapements before 1947 were roughly 20% rather than the 60% of adult recruitment assumed in the Recovery Plan. Actual spawner counts during the first decade of surveys (1945–1955) indicated that coho abundance was much lower and the population trend was opposite of that estimated in the Recovery Plan. However, when spawner abundance was estimated by assuming in-river harvest rates were 80% before 1947, predicted trends in spawner abundance and ocean harvest rates aligned with those from the earliest decade of studies, enabling direct estimates of spawner abundance and ocean harvest rates. Further, the revised simulations indicate that Oregon Coast coho abundance during 1892–1956 probably varied within a range similar to recent decades. We find that use of landings data to speculate about historical salmon abundance is unreliable and can lead to recovery goals that are unattainable, errant interpretation of population limiting factors, and high investment in well-intentioned but misaligned recovery actions. Correct diagnosis of population trends and their causes is best attained with contemporary data for which both precision and accuracy can be validated.

Publisher

Canadian Science Publishing

Subject

Aquatic Science,Ecology, Evolution, Behavior and Systematics

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3