Effects of global climate change on marine and estuarine fishes and fisheries Julie M. Roessig1, Christa M. Woodley1, Joseph J. Cech, Jr.1,* & Lara J. Hansen2 1 Department of Wildlife, Fish, and Conservation Biology and Center for Aquatic Biology and Aquaculture, University of California, Davis, One Shields Ave., Davis, CA 95616, USA (Phone: +1-530-732-3103 Fax: +1-530-752-4154 E-mail: jjcech@ucdavis.edu) 2 Climate Change Program, World Wildlife Fund, 1250 24th Street NW, Washington, DC 20037, USA Accepted 2 November 2004 Contents Abstract page 251 Introduction 252 The situation Model predictions Potential effects of global climate change on marine and estuarine environments Effects on fishes 257 Temperate regions 258 Physiological and behavioral effects on temperate fishes Polar regions 260 Physiological effects on polar fishes Ecological effects on temperate fishes Ecological effects on polar fishes Tropical regions 263 Effects on coral reef communities Socio-economic effects 264 Subsistence harvesters Commercial harvesters Recreational harvesters Marine protected areas Effects of rising sea levels Disease Conclusions 269 Acknowledgements 270 References 270 Key words: climate, estuarine fish, fisheries, global climate change, marine fish, temperature Abstract Global climate change is impacting and will continue to impact marine and estuarine fish and fisheries. Data trends show global climate change effects ranging from increased oxygen consumption rates in fishes, to changes in foraging and migrational patterns in polar seas, to fish community changes in bleached tropical coral reefs. Projections of future conditions portend further impacts on the distribution and abundance of fishes associated with relatively small temperature changes. Changing fish distributions and abundances will undoubtedly affect communities of humans who harvest these stocks. Coastal-based harvesters (subsistence, commercial, recreational) may be impacted (negatively or positively) by changes in fish stocks due to climate change. Furthermore, marine protected area boundaries, low-lying island countries dependent on coastal economies, and disease incidence (in aquatic organisms and humans) are also affected by a relatively small increase in temperature and sea level. Our interpretations of evidence include many uncertainties about the future of affected fish species and their harvesters. Therefore, there is Reviews in Fish Biology and Fisheries (2004) 14: 251���275 �� Springer 2005

a need to research the physiology and ecology of marine and estuarine fishes, particularly in the tropics where comparatively little research has been conducted. As a broader and deeper information base accu- mulates, researchers will be able to make more accurate predictions and forge relevant solutions. Introduction One fascinating feature of global climate change is how it relates so many facets of science that are so often segregated. To fully understand how this phenomenon affects fish, we must consider atmo- spheric science, chemistry, oceanography, physiol- ogy, and ecology. Taken a step further in relating these to people and communities, we must also consider geography, economics, and sociology. With the context so broad, one review paper cannot fully encapsulate the spectrum of implications. We focus on how global changes (particularly temper- ature-related ones) impact marine and estuarine fish and fisheries, and the people who depend on them. The amazing aspect of global climate change is the magnitude of the impact of a relatively small temperature change. An increase of a few degrees in atmospheric temperature will not only raise the temperature of the oceans, but also cause major hydrologic changes affecting the physical and chemical properties of water. These will lead to fish, invertebrate, and plant species changes in marine and estuarine communities (McGinn, 2002). Fishes have evolved physiologically to live within a specific range of environmental variation, and existence outside of that range can be stressful or fatal (Barton et al., 2002). These ranges can coincide for fishes that evolved in similar habitats (Attrill, 2002). We approach these patterns of existence by looking at three different regions of the world���s oceans: temperate, polar, and tropical. Within each region, we examine physiological characteristics common to its fishes and relate them to regional habitat characteristics. After examining predicted changes that fish and their populations will encounter, we attempt to bridge the gap between the science information and models and fish-dependent societies. Three types of harvesters exploit fish stocks: subsistence (artisanal), commercial, and recreational. These all may be impacted (negatively and/or positively) by changes in fish stocks due to climate change. Other issues affected by these global changes include boundaries of marine protected areas, low-lying island countries dependent on coastal economies, and disease (in aquatic organisms and humans). All stem from a relatively small rise in temper- ature. By examining the physiological and ecological effects on fishes in these regions, we are made aware of how much is not known about fishes and their ecosystems. There is a great need for research on the physiology and ecology of fishes, particularly in the tropics. Without an understanding of how these organisms and systems function and interact, we cannot predict how they will react to perturbation, including global climate change-related distur- bances. These gaps lead to uncertainties about fu- ture fish stocks and for people depending on them. The situation Many naturally occurring compounds from the Earth���s E crust and waters are continuously added to the atmosphere. Until recently, chemical influx and efflux have been driven by non-anthropogenic processes (Figure 1) over geological time spans, allowing organisms to evolve with their environ- ments. However, since the Industrial Revolution (19th Century), C many compounds that naturally existed in small quantities have been mass pro- duced and added to our atmosphere through anthropogenic activities (Sarmiento and Gruber, 2002). This comparatively rapid introduction of compounds has caused profound environmental alterations. Greenhouse gases, such as carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), chlorofluorocarbons (CFCs), and volatile organic compounds (VOCs), absorb incoming so- lar energy and outgoing (reflected) radiant energy. Changes in the concentrations of these and other atmospheric gases, therefore, alter the global radiation budget (Gribbin 1988 Figure 1). The retention of additional radiant energy raises atmospheric temperatures and impacts climates. These climatic changes affect the entire earth sys- tem, including ecosystems, community and popu- lation structures, and organismal ranges (Bernal, 1993 Daniels et al., 1993 Parmesan, 1996 Booth and Visser, 2001 McCarthy, 2001 Walther et al., 2002). Recent evidence outlines the magnitude of such changes in terrestrial systems (Parmesan and Yohe, 2003 Root et al., 2003). 252

Global atmospheric temperatures and CO2 concentrations have risen throughout the last 50 years (Figure 2 Trenberth, 1997 Quay, 2002). Simultaneously, the world���s oceans have experi- enced a net warming (Levitus et al., 2000 Sheppard, 2001 Fukasawa et al., 2004). Regional increases in temperature have been documented in the southwest Pacific Ocean and North N Atlantic Ocean (Bindoff and Church, 1992 Parrilla et al., 1994). For the last 20���30 years the western Mediterranean Sea temperatures have been rising (Bethoux et al., 1990), which is re- flected in the presence and abundance of ecto- thermic marine life (Francour et al., 1994). For example, off the coast of France, two thermo- philic algal species, several thermophilic echino- derm species, and some thermophilic fishes have increased in abundance, while other thermophilic species are being observed for the first time (Francour et al., 1994). Figure 1. Earth���s radiation budget. Some energy is absorbed and re-radiated downwards by the atmosphere (from Gribbin, 1988 New Scientist with permission). Figure 2. Estimated changes in annual global mean temperatures and carbon dioxide (smooth line) from a 135-year period. Earlier values for carbon dioxide are from ice cores (dashed line) and for 1957���1995 from direct measurements made at Mauna Loa, Hawaii. The scale for carbon dioxide is in parts per million by volume (ppmv) relative to a mean of 333.7 ppmv (from Trenberth, 1997 with permission). 253

Model predictions Several models have been produced recently sug- gesting outcomes to various global climate change situations. Models incorporate and combine knowledge about individual processes in a quan- titative way, yet they typically have intrinsic limi- tations because they are simplifying a complex system, and using often incomplete and inaccurate knowledge (Trenberth, 1997 Rahmstorf, 2002). Ultimately, modeling is a compromise among inclusion of processes, level of complexity, and desired resolution (Rahmstorf, 2002). Assessment of their accuracy is possible by comparing model outcomes to climate reconstructions. Climate reconstructions are based on proxies such as pollen, ocean-sediment cores, lake-level recon- structions, glacial moraines, terrestrial, and ice- borehole data (for coarse, long time scales), and tree rings, corals, ice cores, lake sediments, and historical records (for shorter time scales Mann, 2002). Each proxy has advantages and disadvan- tages, leaving no single proxy adequate for all climate reconstruction purposes (Mann, 2002). Estimates of future surface air temperature in- creases range from 1 to 7 ��C, depending on the hypothesized atmospheric CO2 contents (Daniels et al., 1993 Kwon and Schnoor, 1994 Manabe et al., 1994 Woodwell et al., 1998). Air tempera- tures are expected to increase ocean warming, most significantly in the upper 500���800 m (Bernal, 1993). However, even slight warming of deeper oceanic layers will have a huge impact on the Earth���s E energy budget due to the mass of water they contain (Bernal, 1993 Levitus et al., 2000 Stevenson et al., 2002). Ocean circulations are predicted to shift, possibly interacting with land masses, creating a north-south thermal asymmetry (Bernal, 1993). For example, the northern boundary of the Gulf Stream has shifted slightly northward in recent decades (Taylor and Stephens, 1998). Potential effects of global climate change on marine and estuarine environments It is widely accepted that due to greenhouse gases a profound change in climate will occur (Palmer and Raisanen, 2002 Schnur, 2002). How will this affect the physical environments of oceanic and estuarine ecosystems? Change in climate means there is a change in precipitation and evaporation rates, constituents of the hydrologic cycle, which affect surface runoff, and groundwater and ocean levels (Klige, 1990 Zestser and Loaiciga, 1993 Loaiciga et al., 1996). A rise in global temperature, gener- ally, would increase regional evaporation in the lower latitudes and increase regional precipitation in the higher latitudes (Klige, 1990 Zestser and Loaiciga, 1993 Manabe et al., 1994 Palmer and Raisanen, 2002). Shifts in the evaporation/precip- itation regime could have significant consequences to the continents, including worsening conditions for flood control and water storage (Loaiciga et al., 1996 Milly et al., 2002). In addition, excess runoff (in relation to evaporation) will contribute to groundwater levels (Zestser and Loaiciga, 1993 Manabe et al., 1994). Approximately 6% of the total water influx to the oceans and seas comes from direct groundwater discharge (Zestser and Loaiciga, 1993). An increase in the amount of groundwater entering the ocean would lead to a net gain in oceanic volume. In addition to in- creased groundwater discharge, meltwater from glaciers may contribute to increasing ocean vol- ume (Klige, 1990 Daniels et al., 1993 Sch��tt Hvidberg, 2000 Stevenson et al., 2002). Finally, as water temperatures rise, the volume of the oceans will also increase due to thermal expansion (Daniels et al., 1993 Stevenson et al., 2002). Increased oceanic volume and concomitant sea level rise have tremendous implications for coastal environments. Sea levels have risen (0.1���0.3 m over the past century) in conjunction with the rising global temperature (Wigley and Raper, 1987 Liu, 2000 IPCC, 2001), but with a time lag of 19 years (Klige, 1990). Depending on model factors, predicted increases range from 0.3 to 5.0 m, possibly inundating almost 1 million km2 of coastal land (Klige, 1990 Daniels et al., 1993 Liu, 2000). This rise is occurring at a faster rate than plants can colonize and establish wetland habitat (Daniels et al., 1993 Stevenson et al., 2002). Therefore, many tidal wetlands, estuaries, mangroves, and other shallow-water habitats may be lost if climate change continues at the predicted rates. An increasing water column depth affects the complex interactions of the hydrodynamic processes that take place in the coastal environment. Tides and tidal currents, distribution of turbulent energy, shoreline configuration, near-shore depth distribution, sedimentation 254

patterns, and estuarine���river interactions will be affected (Liu, 2000). Another major consequence of a changing cli- mate is the likely perturbation of oceanic circula- tions. Currents are driven directly by winds (upper layer of ocean), fluxes of heat and freshwater (thermohaline circulation), or by the gravitational pull of the sun and moon (tides Rahmstorf, 2002). Thermohaline circulation is the deep ocean water (200 m) that is conveyed in slow large-scale circulations, driven by water density, which is dependent on heat and salinity (Figure 3 Garrison, 1996). Although there is much debate on the predicted future of this circulation (Hansen et al., 2004), many global climate change models suggest weakening, and possibly complete break- down, of the thermohaline circulation, particularly in the Atlantic Ocean (Bernal, 1993 Manabe et al., 1994 Sarmiento et al., 1998 Plattner et al., 2001 Vellinga and Wood, 2002). Furthermore, sugges- tions that a rise in sea level may also decrease the formation of North Atlantic deep water (NADW) will directly impact massive ocean water circula- tions (Mikolajewicz et al., 1990). This is caused, in part, by increased density-driven stratification of the upper water column in the higher latitudes, which decreases vertical mixing and convective overturning (Sarmiento et al., 1998). Evidence of this has been uncovered in the Gulf of Alaska. Here, surface temperatures have been rising from warmer air temperatures, while salini- ties have been decreasing from melting ice, thereby decreasing the water���s density. Meanwhile, deep waters have had little change in density, leading to increased ocean stratification and decreased for- mation of mixed layers (Whitney and Freeland, 1999 Freeland and Whitney, 2000). This halts the convective flow that drives thermohaline circula- tion, and consequently disturbs the circulation of nutrients and heat that these deep waters contain. In the Pacific Ocean, increased stratification could increase the frequency of El Nino/Southern Oscillation (ENSO) events and more extreme cli- matic variations (Timmermann et al., 1999). ENSO events are characterized by an intrusion of warm water from the western equatorial Pacific into the eastern equatorial Pacific, where it causes a rise in sea level, higher sea surface temperatures, and a weakened thermocline, which is associated with reduced primary productivity (Miller and Fluharty, 1992). Thermohaline circulation is inti- mately linked to the carbon cycle and deep ocean ventilation, and any change in either would further disrupt the carbon cycle and biogeochemistry of the coupled system (Bernal, 1993 Manabe et al., 1994 Sarmiento et al., 1998). The oceans act as an immense carbon sink. The amount of carbon stored in the oceans is regulated by atmosphere���sea gas exchange, carbonate equi- libria, ocean circulation, and marine organisms (Plattner et al., 2001 Sarmiento and Gruber, 2002 Sabine et al., 2004). Increasing water temperature decreases the solubility of CO2, resulting in the slowed uptake of atmospheric CO2 (Kwon and Figure 3. Highly simplified cartoon of the global thermohaline circulation. Near-surface waters (red) flow towards three main deep- water formation regions (yellow ovals) ��� in the northern Atlantic, the Ross Sea, and the Weddell Sea ��� and recirculation at depth. Deep currents are shown in blue, bottom currents in purple. Green shading indicates salinity above 36&, blue shading indicates salinity below 34& (from Rahmstorf, 2002 with permission). 255