Paleoecological studies on variability in marine fish populations: A long-term perspective on the impacts of climatic change on marine ecosystems Bruce P. Finney a, ⁎, Jürgen Alheit b , Kay-Christian Emeis c , David B. Field d, 1 , Dimitri Gutiérrez e , Ulrich Struck f a School of Fisheries and Ocean Sciences, University of Alaska Fairbanks, Fairbanks, AK, USA b Leibniz Institute for Baltic Sea Research, Warnemünde, Germany c Institute for Biogeochemistry and Marine Chemistry, Hamburg University, Germany d Monterey Bay Aquarium Research Institute, Moss Landing, CA, USA e Direccion de Investigaciones Oceanograficas, Instituto del Mar del Per, Callao, Peru f Museum für Naturkunde, Berlin, Invalidenstraße 43,10115 Berlin, Germany abstract article info Article history: Received 11 May 2007 Received in revised form 19 June 2008 Accepted 2 December 2008 Available online 20 February 2009 Keywords: Marine fish remains Marine sediments Paleoclimate Paleoceanography Holocene Northeastern Pacific California Current System Humboldt Current System Benguela Current System The use of historical fishing records to understand relationships between climatic change and fish abundance is limited by the relatively short duration of these records, and complications due to the strong influence of human activity in addition to climatic change. Sedimentary records containing scales, bones or geochemical proxies of variability in fish populations provide unique insights on long-term ecosystem dynamics and relationships with climatic change. Available records from Holocene sediments are summarized and synthesized. The records are from several widespread locations near or along the continental margins of the South Atlantic and Pacific oceans, including Alaska, USA (Pacific salmon), Saanich and Effingham Inlets, British Columbia, Canada (pelagic fish), Santa Barbara Basin, California, USA (Northern anchovies and Pacific sardines), Gulf of California, Mexico (Pacific sardines, Northern anchovies and Pacific hake), Peru upwelling system (sardines, anchovies and hake), and Benguela Current System, South Africa (sardines, anchovies and hake). These records demonstrate that fish population sizes are not constant, and varied significantly over a range of time scales prior to the advent of large-scale commercial fishing. In addition to the decadal-scale variability commonly observed in historical records, the long-term records reveal substantial variability over centennial and millennial time scales. Shifts in abundance are often, but not always, correlated with regional and/or global climatic changes. The long-term perspective reveals different patterns of variability in fish populations, as well as fish–climate relationships, than suggested by analysis of historical records. Many records suggest prominent changes in fish abundance at ca. 1000–1200 AD, during the Little Ice Age, and during the transition at the end of the Little Ice Age in the 19th century that may be correlative, and that were likely driven by major hemispheric or global reorganizations in the earth's climate system. Additional sedimentary records of marine fish abundance and corresponding paleoenvironmental conditions are likely to further enhance our understanding of marine ecosystem dynamics. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Historical catch records of marine fishes reveal large variability over multi-decadal timescales virtually everywhere where available (Francis and Hare, 1994; Mantua et al., 1997; Schwartzlose et al., 1999; Alheit and Bakun, 2010-this issue). However, determining the factors responsible for such variability in these relatively short records is challenging. Correlations between records of fish stock abundance in different regions, and between records of fish stock abundance and climatic and oceanographic time series suggest large-scale environ- mental drivers (e.g., Mantua et al., 1997). However, many of these period capture only one major “cycle” of variability and are confounded by the potential effects of fishing and other anthropo- genic factors. Records that extend beyond the available historical period are needed to accurately define variability that occurs over multi-decadal time scales. Longer paleoecological records can better define the nature of such variability and help understand the processes underlying fluctuations in fish stocks. Further, longer records are less strongly influenced by human activities, and encompass a broader range of climatic states and variability (e.g., Mann et al., 1999; D'Arrigo et al., 2005). Estimates of past changes in pelagic fish populations have been derived in most studies from fossil evidence such as preserved scales and bones. In the case of Pacific salmon, which return to freshwaters to spawn and die after marine residence, isotopic, biological and geochemical sedimentary records from their spawning areas (lakes) Journal of Marine Systems 79 (2010) 316–326 ⁎ Corresponding author. Present address: Department of Biological Sciences, Idaho State University, Pocatello, ID, USA. Tel.: +1 208 282 4318; fax: +1208 282 4570. E-mail address: finney@isu.edu (B.P. Finney). 1 Present address: Hawaii Pacific University, College of Natural Sciences, 45-045 Kamehameha Hwy., Kaneohe, HI, USA. 0924-7963/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jmarsys.2008.12.010 Contents lists available at ScienceDirect Journal of Marine Systems journal homepage: www.elsevier.com/locate/jmarsys