The ecological characteristics of fish communities were compared at the southern boundaries of the eastern and western Subarctic Pacific, based on oceanography, fishery information, and ecological features. Sea surface temperature (SST) was higher in the western North Pacific (NP) than in the eastern NP, and changes in SST showed regional and temporal alternating patterns. Cool and warm SST regimes were observed in the western NP during the early 1980s and the early 2000s, respectively, compared with warm and cool regimes in the eastern NP. Increasing SSTs were more conspicuous in the western than in the eastern NP. Catches from commercial fisheries were higher in the western NP than in the eastern NP. Small pelagic fish were dominant in the western NP, whereas demersal behaviour was common for fish populations in the eastern NP. Changes in species composition also showed contrasting characteristics between the two regions. In the western NP during the early 1980s, landings were dominated (35.8%) by sardine. After two decades, however, landings consisted of a more diverse species group. In the eastern NP, five species appeared in similar percentages (∼10% each) during the early 1980s, but hake alone made up 36.3% of the landings in the early 2000s.
Kim, S., Kang, S., Zhang, C-I., Seo, H., Kang, M., and Kim, J. J. 2012. Comparison of fisheries yield and oceanographic features at the southern boundaries of the western and eastern Subarctic Pacific Ocean. – ICES Journal of Marine Science, 69: .
Ecological components in marine ecosystems are closely linked with each other, so recruitment success or failure in a dominant fish species can result in changes in the biomass or distributions of other taxa. Prey–predator relationships and abiotic conditions, such as temperature and salinity, are major factors that determine ecosystem and fisheries productivity (Wooster et al., 1983; Ottersen et al., 2010). Yields from marine fisheries usually reflect the level of recruitment success, which in turn is strongly influenced by marine primary and secondary productivity. Unstable environmental conditions, which vary spatially and temporally, are also critical to the survival of the early life stages of fish and consequently to fish biomass and fishery yields (Drinkwater et al., 2010). Therefore, time-series data for population parameters (e.g. spawning biomass or egg production), life-history parameters (e.g. mortality or growth), and ecosystem parameters (e.g. prey abundance or predator biomass) are essential to understanding ecosystem structure and function and to predicting future fishery yields under changing environmental conditions (Kendall et al., 1996; Murawski, 2011).
Climate conditions and weather events are the result of physical interactions between the atmosphere and the oceans. Temperatures at the ocean surface and in the atmosphere are interrelated. Although certain regional seas show cooling phenomena, ocean warming has generally been observed worldwide. During the 1980s and through the early 2000s, a marked increase in temperature was observed in the northwestern Pacific, and in the northeastern and northwestern Atlantic (Belkin, 2009). Such warming may cause shifts in the distributions of fish and fisheries. Specifically, temperature effects on fish distributions seem to be evident near the periphery of a species' range (Pinnegar et al., 2010). The habitat distributions of marine fish such as walleye pollock (Theragra chalcogramma) and chub mackerel (Scomber japonicus) in midlatitudes have moved towards higher latitudes in the past few decades (Mueter and Litzow, 2008; McFarlane et al., 2009), and continuous warming in this century will likely result in many fish populations moving towards higher latitudes (Cheung et al., 2009, 2011).
Distributions as well as biological characteristics such as growth, recruitment, and survival can vary depending on the species-specific capacity for physiological adaptation (Pörtner and Knust, 2007; Pörtner and Farrell, 2008). Many marine species can adapt quickly to newly established environments, but some cannot. Some local fisheries may benefit from changes in ocean productivity or species distributions, but others may suffer from frequent extreme weather events, such as super-typhoons (Kim and Low, 2011). A step-by-step framework for modelling responses by fish to future climate changes has been created: a framework was applied to forecast summer sea surface temperature (SST) in the Bering Sea by 2050, and was used to estimate the effects of climate change on the production of fish species (Hollowed et al., 2009). Also, a practical end-to-end framework was developed to assess and forecast the ecosystem impacts of fishing and to manage fisheries under changing climatic conditions (Zhang et al., 2011).
The Pacific Ocean consists of a large basin with islands and regional seas that are next to continents. There is a relatively broad continental shelf and some semi-closed regional seas in the western North Pacific (NP), and a narrow continental shelf, several sounds, and an archipelago off the continent in the eastern NP. Two gyre systems are found in the open NP: one flows clockwise in a tropical area, and the other anticlockwise at high latitude (Figure 1). The warm Kuroshio and the cool nutrient-rich Oyashio currents merge at midlatitudes in the western NP and flow eastwards along 50–52°N, forming a meandering Polar Front and eddies. Near the coast of North America in the eastern NP, this current diverges towards the north (Alaska Current) and the south (California Current). Subarctic ecosystems are formed north of the Oyashio Current in the western NP and north of the Alaska Current in the eastern NP.
Comparisons of regional ecosystems in the world's oceans can contribute to the understanding of interactions between natural and anthropogenic forcing, evolutionary processes, and prey–predator relationships in marine ecosystems (GLOBEC, 1997). For example, the eastern (King et al., 2011) and western (Kim et al., 2007; Sakurai, 2007) NP ecosystems have been characterized using fisheries and oceanography information. McFarlane et al. (2009) compared fisheries and community structure between two ecosystems off Korea and Canada to outline the fishery management strategies of the two countries. Fisheries managers need to consider the effects of climate changes on the fishing industry and fish populations to conserve healthy ecosystems and to maintain sustainable yields. This study therefore sought to compare the fishery characteristics in southern boundary areas in Subarctic waters of the eastern and western NP, to identify changes in species composition in the two ecosystems during the past two decades, and to contribute to fisheries management under changing environments.
Material and methods
For this study, we selected two coastal areas in what we consider to be the southern boundary of the Subarctic Pacific Ocean: 26–46°N, 118–155°E in the western NP, and 42–55°N, 125–157°W in the eastern NP (Figure 1). To examine spatio-temporal patterns in environmental variability, temperatures at 5 m were obtained as SST during winter (January–March) from the Simple Ocean Data Assimilation (SODA; http://www.atmos.umd.edu/~ocean/) database, version 2.2.4, for the period 1981–2008. The SODA is a reanalysis dataset with monthly means reported with a 0.5 × 0.5° latitudinal and longitudinal resolution (Carton et al., 2000).
To determine fishery status and biological changes in fish communities in the eastern and western NP ecosystems in the early 1980s and early 2000s, commercial fishery statistics from four countries (South Korea, Japan, Canada, and the United States) were acquired. For the western NP ecosystem, information on commercial fisheries from South Korea and Japan were obtained from the FAO (http://www.fao.org/fishery/statistics/software/fishstat/en) database from 1981 to 2008. Maps of the distribution of chub mackerel in the early 1980s and 2000s were compiled using information on daily commercial catches, and data on the fishing grounds used by the Korean purse-seine fishery were collected from the Busan Cooperative Fish Market. Commercial catch data from British Columbia and Oregon and Washington states in the eastern NP were obtained from Canadian Fisheries Statistics (http://www.dfo-mpo.gc.ca/stats/commercial/sea-maritimes-eng.htm), USA Commercial Landed Catch, and the Pacific Fisheries Information Network (PacFIN; http://pacfin.psmfc.org/pacfin_pub/woc.php) from 1981 to 2008. Catches were categorized as pelagic fish, demersal fish, pelagic invertebrates, and demersal invertebrates. When we counted the species number in each habitat group, sharks, skates, and rays of the vertebrates were not included in the count of species number because of the ambiguity of habitat, and a collection of multispecies such as flatfish was counted as a single species. Note too that each invertebrate group such as cephalopods, crustaceans, scallops, and echinoderms was regarded as a single species. Catch statistics (i.e. annual mean catch and the proportion of the total catch) from both ecosystems were averaged for the visual comparison of species composition during the periods 1981–1985 and 2001–2005. Information on common names, fish habitats, and lifespans for the western NP fish was cited from Korean literature (NFRDI, 2005), and similar data for eastern species were obtained from FishBase (http://www.fishbase.org/search.php). Sardinops melanostictus in the western NP and Sardinops sagax in the eastern NP are referred to as sardine and Pacific sardine, respectively.
At the southern boundary of the Subarctic Pacific Ocean, regional differences in SST were evident. Since the 1980s, winter SSTs have been generally higher in the western NP, ranging between 11.8 and 13.7°C with a mean of 13.0°C, whereas values in the eastern NP have been relatively lower, ranging between 6.2 and 7.8°C with a mean of 7.0°C. In the western NP, a cool phase in the early 1980s turned into a warm phase after the late 1980s, and relatively warm conditions have been sustained thereafter. In contrast, warm and cool conditions have alternated in the eastern NP since the early 1980s: warm during 1980–1987, 1991–1998, and 2003–2006, and cool during 1988–1990, 1999–2002, and 2007–2008 (Figure 2). Note that cooler temperatures appeared during relatively shorter periods. A warming tendency has been observed in both ecosystems, but the trend was more evident in the western NP. To see the changing pattern of seawater temperature between the western and the eastern NP ecosystems, we compared SSTs during the early 1980s and the early 2000s. In the early 2000s, SSTs were higher than average in the western NP, but cooler in the eastern NP than in the early 1980s. To observe the nature of these changes, we selected the two contrasting years. A comparison between the winter SST in 1984 and 2000 indicated that the western NP showed conspicuous warming, from ∼11.8 to 13.7°C, whereas the eastern NP cooled, from 7.7 to 6.7°C (Figure 2). Spatially, the temperature changes showed a contrasting pattern between the two ecosystems (Figure 3). The largest increase was found in the boundary area between the Oyashio and the Kuroshio currents off Japan, the southern East/Japan Sea, the shelf region of the East China Sea, and the central NP. In the eastern NP, although there was slight local warming off the Oregon coast, the prevalent trend was for cooling.
Pelagic fish yield was high in the western NP during the 1980s, reaching ∼7 million tonnes annually (Figure 4a). However, the yield greatly decreased through the 1990s and has stabilized at ∼3 million tonnes per year since the early 2000s. Catches of demersal fish also showed decreasing trends, from ∼2 million tonnes per year in the early 1980s to 1 million tonnes per year in the early 2000s; demersal fish generally yielded about one-third of the pelagic fish catch. Invertebrates have accounted for ∼15–25% of total fishery yields in this region. Benthic invertebrates such as clams and crabs constituted slightly larger proportions of the catch than pelagic invertebrates (mostly squids). Note that cool SSTs persisted during the high-productivity regime of the early 1980s. During the period of warmer conditions since the 1990s, the catch from both pelagic and demersal fish stocks has decreased. The catch of invertebrates was relatively stable throughout the study period, and their proportion in the total catch was greatly enhanced as a result of reductions in total catches since the early 2000s.
Fishery yields were much lower in the eastern NP than in the western NP (Figure 4b). Demersal fish catches were at a level similar to pelagic fish catches in the early and mid-1980s, but they increased rapidly from ∼150 000 t per year to 200 000–300 000 t per year after the late 1980s (Figure 4b). Pelagic fish catches, however, decreased slightly from ∼150 000 t per year during the 1980s to ∼50 000 t per year in the late 1990s, but they then increased in the early 2000s and catches have been sustained at around 100 000 t per year since then. Invertebrates fisheries included mostly demersal species (crabs and shrimps), producing ∼50 000 t annually. There has been almost no commercial squid fishing in the eastern NP. A comparison of fish catch with seawater temperature indicated that a high level or an increasing trend of especially demersal fish catch seemed to be sustained during the warm period (Figure 4b).
The species composition of catches in each ecosystem was compared between the early 1980s and the early 2000s. In the western NP, a single species constituted a large portion of the catch in the 1980s, but a larger number of species shared a similar proportion of the catch in the 2000s (Figure 5a). For example, from 1981 to 1985, sardine catches accounted for ∼35.8% of the total catches, followed by walleye pollock (10.2%) and chub mackerel (8.6%). Some 20 years later, however, the sardine stock had disappeared and walleye pollock catch was ranked as the fifth largest catch (4.8%). On the other hand, the catch proportion of chub mackerel to the total showed a slight increase from 8.6% in the early 1980s to 9.3% in the early 2000s. Minor taxa in the early 1980s, such as anchovy (3.4%, Engraulis japonicus) and common squid (2.2%, Todarodes pacificus), became more prevalent, representing 11.3 and 8.1%, respectively, of the total catches in the early 2000s. There was also a change in species composition of the catch in the eastern NP between the early 1980s and the early 2000s. In the early 1980s, rockfish (Sebastes spp.) were the dominant group, accounting for 13.7% of the total catch, whereas Pacific herring (Clupea pallasi), hake (Merluccius productus), pink salmon (Oncorhynchus gorbuscha), and sockeye salmon (Oncorhynchus nerka) represented 8.7–10.2% of the total catch (Figure 5b). Two decades later, in the early 2000s, catch statistics showed remarkable increases in hake catches (36.3%) and decreases in rockfish (5.4%). Water temperatures were relatively warm in the early 1980s compared with the early 2000s, and ∼13–14 species of the most dominant 20 species were demersal in the region.
Although total catches were much higher in the western NP than in the eastern NP, the total number of important commercial species caught was not very different between the two areas: 132 species in the western NP and 96 in the eastern NP. In the western NP, the number of pelagic fish was the largest (40%, 53 species), and demersal fish were the second largest group (34%, 45 species), whereas these groups accounted for 23 and 41% (22 and 39 species), respectively, in the eastern NP. Species with much greater longevity and a larger number of demersal fish species were characteristic of the exploited fish community in the eastern NP, whereas shorter-lived species and a larger number of pelagic fish species were more numerous in the western NP. For example, pelagic fish species usually had lifespans of 1–8 years in both regions, but demersal fish in the western and eastern NP ecosystems lived for 8–15 and 12–114 years, respectively. The numbers of invertebrate species (and their ratios) were 34 (26%) and 35 (36%) in the western and eastern NP, respectively, some 10–20% of total catches.
Owing to the general warming trend in SST in the NP (Figure 2), poleward displacements of fishing areas have been observed in both ecosystems. Off the coasts of Oregon, Washington, and British Columbia during the early 1980s, no catch records were available for Pacific sardine (S. sagax) or albacore tuna (Thunnus alalunga), which are warm-water species. In contrast, these species ranked second and eighth, representing 9.1 and 3.1% of the total commercial catch, respectively, in the eastern NP during the early 2000s (Figure 5b). Furthermore, some cold-water species such as Pacific cod (Gadus macrocephalus) and sablefish (Anoplopoma fimbria) almost disappeared. In the western NP, chub mackerel was one of the representative warm-water species in Korean and Japanese waters. In accord with the SST increase in the western NP (Figure 2), the fishing locations of the Korean purse-seine fishery for chub mackerel moved north (Figure 6). There was more fishing activity farther north (i.e. in the Yellow Sea and along the southern part of the Korean Peninsula) in the early 2000s than in the early 1980s.
Synchrony between the rise and fall of sardine and anchovy populations has often been observed in large marine ecosystems separated by thousands of kilometres (Alheit and Bakun, 2010). In the eastern Pacific Ocean, the alternation between sardine and anchovy species is a well-known phenomenon: a cool “anchovy regime” was replaced by a warm “sardine regime” in the mid-1970s (Chavez et al., 2003). However, SSTs in the western NP changed in a pattern that was opposite to that in the eastern NP. Therefore, in the western NP, sardine species were abundant when cool SSTs prevailed in the early 1980s. Chavez et al. (2003) questioned this phenomenon because water temperature showed a reversed environmental condition to that of the “sardine regime” off California. Takasuka et al. (2008) demonstrated differences in species-specific ranges in spawning temperature between the western and eastern Pacific: the sardine population off Japan preferred lower temperatures for spawning, whereas those off California preferred warmer temperatures.
Mackas et al. (2007) showed “warm-and-low productivity” and “cool-and-high productivity” zooplankton community periods during the years 1979–2004 in the northeastern Pacific. They suggested that anomalous high temperatures may provide misleading environmental cues that contribute to timing mismatches between life-history events and the seasonality of ecosystem characteristics. Their observation of “cool-and-high productivity” during 1987–1991 and 1999–2002, and “warm-and-low productivity” during the periods 1992–1998 and 2003–2005, roughly matched our datasets for pelagic fish catches during the study period. Although moderate-to-strong correlations were observed between temperature anomalies and biological parameters, such as zooplankton biomass and reproduction, growth, and survival of pelagic fish and seabirds, in the northeastern Pacific (Mackas et al., 2007), this was not the case for demersal fish catches. Instead, warm temperatures seem to enhance catches of demersal fish in the eastern NP, as in the El Niño years (1983 and 1987), the early to mid-1990s, and the mid-2000s. Changes in fish biomass and the species composition of the fish community result from the recruitment success of each fish species. SST could be one of several important controlling factors that determine recruitment (de Young et al., 2010). In our analysis, we observed large changes in species composition of the catch and water temperature during the study period. Relatively fewer species constituted a larger proportion of catches in cool years and more species contributed to catches in warmer years. For example, large catches of sardine in the western NP and hake in the eastern NP were recorded in the early 1980s and the early 2000s, respectively, when water temperatures were lower than average. However, as found in the western NP, relative ratios in species composition may have been blurred by the dominance of the sardine population in the early 1980s. In the eastern NP, hake biomass increased through the 1990s when SSTs were relatively high, which indicates that large proportions of single species in catches were common even in warm years.
Climate regime shifts have frequent severe impacts on the function and structure of marine ecosystems. Interest in climate change impacts and forecasting for fisheries has been growing among natural and social scientists, stakeholders, and economists, to project societal and market changes (Hollowed et al., 2011). In the western and eastern NP, future fish production will be influenced by climate change. However, the two ecosystems have several physical dissimilarities. The atmospheric variability present in the California Current was not observed in the Oyashio and Kuroshio currents, although SST fluctuations in the western Pacific lag behind those in the east by approximately a decade (Schwing et al., 2010). As argued by Polovina et al. (2011), the changes expected in subtropical and temperate biome areas in this century will cause changes in the total fish production. Fishery yields have tended to decrease in the western NP since the last century, and the majority of the catch was of pelagic species with short lifespans that are sensitive to changes in climate and weather. Because ocean warming is expected to be dramatic in the western NP in this century (Belkin, 2009; Wu et al., 2012), rapid changes in climate may cause vigorous fluctuations in the ocean conditions that control recruitment success (Ciannelli et al., 2005; Kimura et al., 2010; Coyle et al., 2011).
Adaptive fisheries management strategies that incorporate rapid and flexible actions should be considered to deal with rapid changes in species composition and abundance. For example, in the management component of the Integrated Fisheries Risk Analysis Method for Ecosystems (IFRAME), management strategies can be selected to assess how systems will change and to address rapid climate changes or long-term changes in market forces (Zhang et al., 2011). The performance of management strategies can be evaluated by extracting ecosystem indicators from runs of ecosystem models. These indicators can be derived from the output of Ecosim models (Christensen and Walters, 2004) combined with external metrics derived from studies on ecological processes and scenarios based on fishery socio-economic analyses (Zhang et al., 2011). Fisheries management in the eastern NP, where demersal fish with greater longevity are dominant, needs to be adaptive and should consider conservation strategies for the spawning biomass of long-lived demersal fish (McFarlane et al., 2009). To conduct reasonable and effective fishery management under changing climatic conditions, we need to increase our general understanding of climate impacts and conduct interdisciplinary research to resolve knowledge gaps related to the effects on fish and marine ecosystems.
This work was funded by the Korea Meteorological Administration Research and Development Program under Grant CATER 2012-7160 and NFRDI RP-2012-FR-010. We also thank Yong-Woo Lee who kindly consulted on fishery statistics for the Pacific Northwest of the United States.