Estuarine, Coastal and Shelf Science 147 (2014) 32e41

Contents lists available at ScienceDirect

Estuarine, Coastal and Shelf Science

journal homepage: www.elsevier.com/locate/ecss

Drifting algae and fish: Implications of tropical invasion due to warming in western Japan

Mami Yamasaki a, Mikina Aono a, Naoto Ogawa a, Koichiro Tanaka a, Zenji Imoto b, * Yohei Nakamura a, c, a Faculty of Agriculture, Kochi University, 200 Monobe, Nankoku, Kochi 783-8502, Japan b Usa Marine Biological Institute, Kochi University, 194 Usa-cho, Tosa 781-1164, Japan c Graduate School of Kuroshio Science, Kochi University, 200 Monobe, Nankoku, Kochi 783-8502, Japan article info abstract

Article history: Evidence is accumulating that the invasion and extinction of habitat-forming species alters Received 22 December 2013 coastal community structure and ecological services, but their effects on the pelagic environment have Accepted 25 May 2014 been largely ignored. Thus, we examined the seasonal occurrence patterns of indigenous temperate and Available online 10 June 2014 invasive tropical drifting algae and associated fish species every month for 2 years (2009e2011) in western Japan (Tosa Bay), where a rapid shift from temperate to tropical Sargassum species has been Keywords: occurring in the coastal area since the late 1980s due to rising seawater temperatures. Of the 19 drifting algae Sargassum species (31.6%) in drifting algae, we found that six were tropical species, whereas a study in fish global warming the early 1980s found only one tropical species among 12 species (8.3%), thereby suggesting an increase Sargassum in the proportion of tropical Sargassum species in drifting algae during the last 30 years. Drifting Tosa Bay temperate algae were abundantly present from late winter to summer, whereas tropical algal clumps occurred primarily during summer. In the warm season, fish assemblages did not differ significantly between drifting temperate and tropical algae, suggesting the low hostealgal specificity of most fishes. We also found that yellowtail juveniles frequently aggregated with drifting temperate algae from late winter to spring when drifting tropical algae were unavailable. Local fishermen collect these juveniles for use as aquaculture seed stock; therefore, the occurrence of drifting temperate algae in early spring is important for local fisheries. These results suggest that the further extinction of temperate Sargassum spp. may have negative impacts on the pelagic ecosystem and associated regional fisheries. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction recent studies have found that the invasion and extinction of seaweed species ( and fucoids) caused by long-term and Globally, increasing evidence suggests that the geographic dis- episodic climate changes have altered the local species diversity tributions of marine organisms are undergoing rapid geographic and community structure, which has negative impacts on ecolog- changes, which are associated with an increasingly warm climate ical functions and services (Harley et al., 2012; Wernberg et al., (Harley et al., 2006; Hoegh-Guldberg and Bruno, 2010; Feary et al., 2013). 2013). The increased incidence of natural populations with range Western Japan has experienced substantial ocean warming over expansions and contractions during warm or cool periods has been the last century, which may be attributable to the recent warming demonstrated widely among habitat-forming corals and of the Kuroshio Current (Wu et al., 2012), and it is a global hotspot (Lima et al., 2007; Greenstein and Pandolfi, 2008; Yamano et al., for biological change as organisms respond to the warming of 2011; Wernberg et al., 2011). Seaweeds are ecologically important coastal waters (Yamano et al., 2011; Tanaka et al., 2012; Nakamura primary producers and ecosystem engineers, which play central et al., 2013). Tosa Bay is located in western Japan and is affected roles in cold-temperate waters (Steneck et al., 2002). However, strongly by the offshore Kuroshio Current (Fig. 1a). The geograph- ical division of the bay is temperate, but the surface seawater temperature (SST) of coastal areas has increased rapidly during the winter months over the past 30 years (Fig. 1b). Together with these * Corresponding author. Faculty of Agriculture, Kochi University, 200 Monobe, increases in the SST, the abundance of tropical macroalgae Nankoku, Kochi 783-8502, Japan. E-mail address: [email protected] (Y. Nakamura). Sargassum (Fucales) species has increased substantially, whereas http://dx.doi.org/10.1016/j.ecss.2014.05.018 0272-7714/© 2014 Elsevier Ltd. All rights reserved. M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41 33

Fig. 1. (a) Map of the study site, showing the location of Tosa Bay. (b) Trends in surface seawater temperature (SST) in the coldest month (February) at the coastal area of central Tosa Bay from 1981 to 2011. (c) Distribution of temperate Sargassum species (black) and tropical Sargassum species (gray) along the coast of Tosa Bay during the late 1970s and 1990s. The continuous line indicates rich vegetation and the dotted line indicates patchy vegetation for each Sargassum species. The distribution map was modified from that reported by Hiraoka et al. (2005). (d) The box indicates the area sampled for drifting algae; (+): monthly SST sampling stations of Kochi Prefectural Experimental Station within the present study area. The thick arrow represents the main axis of the Kuroshio Current and the anticlockwise thin arrow shows the major surface current inflow into Tosa Bay (Kuroda et al., 2008). temperate Sargassum species have decreased significantly (Tanaka Sargassum were unavailable. Sargassum that have been broken or et al., 2012). In the 1970s, temperate Sargassum species domi- detached by strong waves often drift on the surface of offshore nated the coastal area of Tosa Bay (Fig. 1c). However, in the late areas, and many fish juveniles, including important fishery species 1980s, temperate Sargassum species still dominated the bay, but (e.g., family Carangidae), are attracted to the drifting algae for tropical Sargassum species began to colonize the western region, shelter and food (Kingsford and Choat, 1985; Safran and Omori, possibly because the western waters were strongly affected by the 1990; Kingsford, 1992, 1993; Dempster and Kingsford, 2004). Kuroshio Current, and they were the warmest areas of Tosa Bay. In Large numbers of drifting Sargassum are observed in Tosa Bay; the 1990s, the tropical Sargassum species expanded throughout therefore, it is expected that the algal species shift in coastal areas Tosa Bay and some temperate species beds became sparse or dis- may affect the seasonal occurrence patterns of drifting algae and appeared (Fig. 1c). Human-induced events associated with coastal associated fish. Moreover, we may also expect that possible changes development, such as increase in artificial hard structures and in the seasonal occurrence patterns of drifting algae may affect local sedimentation, are possible causes for the decline and shift in and regional fisheries because yellowtail (Seriola quinqueradiata) habitat-forming macroalgal species (Thompson et al., 2002), but juveniles are frequently associated with drifting algae in the bay the phenomenon seen in Tosa Bay has often been observed in long- and local fishermen collect these juveniles during the spring season established harbors and pristine rocky reefs; thus, an increase in for use as aquaculture seed stock (Uehara et al., 2006). Yellowtail the SST is thought to be one of the most plausible causes for the capture occurs mostly in western Japan and the Republic of Korea, shift from temperate to tropical Sargassum spp. (Tanaka et al., where farming has developed rapidly during the 1970s and 1980s. 2012). Furthermore, the annual production of farm-raised yellowtail is The tropical Sargassum spp. found in Japanese temperate reefs now three times greater than that of the fishing catch in Japan exhibit annual characteristics in terms of their life cycles (a short (Nakada, 2008). vegetation period, primarily in the spring and summer seasons), Evidence is accumulating that invasions and extinctions of which contrast greatly with the perennial characteristics of most habitat-forming seaweed species alter the coastal community temperate species (almost year-round vegetation) (Nagai et al., structure, ecological functions, and services (Serisawa et al., 2004; 2011). Therefore, it is expected that the shorter vegetation pe- Harley et al., 2012; Terazono et al., 2012; Wernberg et al., 2013; riods caused by algal species shifts may have negative effects on Voerman et al., 2013), but their effects on the pelagic environ- fish populations that rely on coastal algal beds. In Tosa Bay, for ment have been largely ignored. This study aimed to determine the example, Terazono et al. (2012) found that juveniles of the impor- importance of tropical and temperate Sargassum spp. in deter- tant fishery species Scombrops boops only recruited to temperate mining the structure of the fish communities associated with Sargassum beds during the middle of winter when tropical drifting algae. We hypothesized that fish assemblages would not 34 M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41 differ significantly between both types of drifting algae because of primarily in the temperate zone) or tropical Sargassum species their similar highly complex thallus morphology (shelter), although (primarily belonging to the subgenus Sargassum, which are the period of availability would be shorter for tropical algae than for distributed primarily in the subtropical and tropical zone) based on temperate algae, such as those observed in coastal Sargassum beds their occurrence (Yoshida, 1998). Although Sargassum patens and (Terazono et al., 2012). We suggest that further reductions in the Sargassum piluliferum belong to the subgenus Sargassum, these available period of drifting algae due to algal species shifts may species have been found widely in Japanese temperate reefs have negative impacts on local and regional fisheries. (Yoshida, 1998); therefore, we regarded these species as temperate species. When a temperate (or tropical) algal species accounted for 2. Materials and methods >80.0% of an algal clump by weight, we regarded the clump as a temperate algal clump (and vice versa) in the algal typeefish as- 2.1. Study site sociation analysis. When a temperate (or tropical) species did not account for >80.0% of an algal clump by weight, we regarded the This study was conducted in Tosa Bay, western Japan, from clump as a mixed algal clump, which was not used in the algal October 2009 to November 2011. Sampling was performed within a typeefish association analysis. All of the fish specimens were 30-km zone, parallel to the shore, at 16e20 km from the shoreline identified to the species level based on Nakabo (2002) and the (depth 120e150 m) using a research vessel (5.4 ton) belonging to standard lengths (SL) were measured to the nearest 0.1 mm. Kochi University (Fig. 1d). High numbers of drift algae were found Concentrated formalin was injected into the body cavity of each often on the current rip within this sampling area, presumably specimen, which was then preserved in 10.0% formalin. because oceanographic features facilitated the aggregation of several species of algae (movements of the algae via the anti- 2.3. Diet of fishes associated with drifting algae clockwise surface current inflow into Tosa Bay) (Ohno, 1984a; Kuroda et al., 2008)(Fig. 1d). The SST in Tosa Bay varied season- To assess whether the fish were attracted to drifting algae for ally and monthly SST data around the sampling area showed that shelter or as epiphytic food resources, the gut contents of 28 species the coldest water temperature was approximately 16 Cin were investigated. Of these, 14 species where 8 individuals con- February, whereas the highest water temperature was approxi- taining food (total 770 specimens) were subjected to food habit mately 29 C in August, during the study period. The SST data were analysis because <5 individuals are considered to yield only a expressed as the average monthly monitored SST at the three sites limited representation of food items (Nakamura et al., 2003). In the (west, central, and east) within the sampling area during each laboratory, the food items in the gut contents of each fish specimen month (Fig. 1d). Based on the monthly variation in the SST, we were identified to the highest possible taxonomic separation. The regarded JanuaryeMarch as winter, AprileJune as spring, percentage volume of each food item in the diet was estimated JulyeSeptember as summer, and OctobereDecember as autumn, visually using a 1 1 mm grid slide under a binocular microscope, which agrees with the system used by the Japan Meteorological as follows. The gut contents were squashed to a uniform depth of Agency (2013). 1 mm and the area occupied by each item was measured. The area of each item was then divided by the total area covered by the gut 2.2. Capture of fish associated with drifting algae contents to calculate the percentage volume of that item in the diet. Food resource use was expressed as the mean percentage compo- Fish associated with drifting algae were captured using a dip net sition of each item by volume, which was calculated by dividing the (diameter, 70 cm; depth, 95 cm; and mesh size, 4 mm), which sum total of the individual volumetric percentages for the item by ensured that the fish captured had a high level of spatial affinity the number of specimens examined. Specimens with empty guts with the object (Dempster and Kingsford, 2004). The size of the dip were excluded from the analysis. There were size class and/or net was determined on the basis of the usual size of drifting algae in seasonal differences in the feeding habits of some species, but these Tosa Bay (<1m3 in size and <2 kg in wet weight) (Ohno, 1984a, b; changes were not substantial. Therefore, we pooled all of the gut Taino, 2006). However, fish with a greater spatial range around content data for each species. drifting algae or a faster swimming capacity may have escaped capture or been under-sampled using this method (Dempster and 2.4. Statistical analyses Kingsford, 2004). Monthly sampling was conducted from October 2009 to November 2011. During each month, we tried to collect 10 The numbers of fish associated with drifting algae were drifting algae as far as possible within a single day during the analyzed as both the numbers per clump and per 1 kg of algae daytime (09:00e16:00 h). Drifting algae were sampled haphaz- because the size of an algal clump varied seasonally, which could ardly when they were encountered within the sampling area. We have affected the seasonal fish abundance trend. The species rich- found drifting algae easily during the spring and summer, whereas ness and abundance of fishes associated with drifting algae were they were difficult to find during the autumn and winter. Therefore, compared between algal types (temperate and tropical) and sam- although the quantity of drifting algae could not be determined pling months (the months when both types of algal clump were accurately, the total number of drifting algae collected during each collected) using a mixed model two-way analysis of variance month approximately reflected the temporal abundance patterns (ANOVA). In this analysis, algal type was considered to be a fixed of drifting algae. No samples were obtained during February and factor and the sampling month was a random factor. Before the October 2010 because of bad weather (e.g., typhoons). analyses, all of the data were transformed to √x þ0.5 to improve Immediately after collection, each drifting alga and its associ- the homogeneity of variances, as required by Levene's test. All of ated fish were placed into a numbered plastic bag (65 80 cm), the statistical analyses were conducted using SPSS (14.0J). before preservation with crushed ice on the research vessel. In the The degree of similarity of fish assemblages in drifting laboratory, all of the algal specimens were identified to the species temperate and tropical algae in the sampling months was calcu- level based on Yoshida (1998) and Mattio et al. (2009), and the wet lated using the BrayeCurtis similarity coefficient based on the weight of each alga was measured to the nearest 0.01 kg. Each algal number of individuals of each species. Prior to the analysis, the specimen was categorized as a temperate Sargassum species abundance data were fourth-root transformed (N1/4) to normalize (belonging to the subgenera Bactrophycus, which are distributed the distributions and to stabilize the variances, as recommended by M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41 35

Table 1 3. Results List of drifting Sargassum species collected during 2009e2011 and their relative proportion by weight. 3.1. Seasonal patterns of drifting algae Subgenus/species name Proportion Range Recorded Vegetation Code b c (%) type in Ohno in Tosa Bay no. In total, 19 species were recorded during the study period (1984a,b) (Table 1). The dominant four species by weight were Sargassum Bactrophycus horneri, Sargassum glaucescens, Sargassum yamamotoi, and Sargassum horneri 24.2 Temperate Yes No 1 Sargassum ilicifolium, which comprised 62.0% of the total weight Sargassum yamamotoi 13.2 Temperate No Yes 2 Sargassum hemiphyllum 4.0 Temperate Yes Yes 3 (Table 1). The occurrence of drifting algae was highly seasonal, Sargassum siliquastrum 3.5 Temperate Yes Yes 4 where large numbers of clumps were captured between April Sargassum ringgoldianum 2.9 Temperate Yes No 5 (spring) and September (late summer), with peak abundances in Sargassum nipponicum 1.5 Temperate Yes Yes 6 MayeJuly (middle springesummer) but few or none were captured Sargassum micracanthum 0.4 Temperate Yes Yes 7 Sargassum muticum 0.3 Temperate No Yes 8 between October and March (autumn and winter), although many Sargassum fusiforme 0.2 Temperate Yes Yes 9 clumps were collected in the autumn of 2009 (Fig. 2). Thirteen Sargassum macrocarpum 0.1 Temperate Yes Yes 10 species were temperate species, whereas the other six species were Sargassum thunbergii 0.1 Temperate No Yes 11 tropical species (Table 1). Temperate algal clumps were collected in Sargassum all 16 months sampled, whereas tropical algal clumps were Sargassum glaucescens 13.4 Tropical No No 12 e Sargassum ilicifoliuma 11.3 Tropical Yes Yes 13 captured primarily during the summer (June August) (Fig. 2). The Sargassum carpophyllum 9.5 Tropical No Yes 14 temperate species also showed species-specific seasonal occur- Sargassum piluliferum 4.7 Temperate Yes Yes 15 rence patterns: S. horneri and Sargassum nipponicum were abun- Sargassum assimile 3.8 Tropical No Yes 16 dant in spring, Sargassum piluliferum and S. yamamotoi were Sargassum patens 2.6 Temperate Yes Yes 17 Sargassum 1.8 Tropical No Yes 18 abundant in early summer, and Sargassum ringgoldianum and alternato-pinnatum Sargassum micracanthum were abundant after the middle of the Sargassum yendoi 1.2 Tropical No Yes 19 summer. The weight of a clump of drifting algae ranged from 15 to Others 1.3 3803 g with an average of 993.7 ± 895.5 g during the study period a This species was identified as S. duplicatum by Yoshida (1998). (mean ± SD, n ¼ 112). The mean weight of an algal clump did not b Vegetation in Tosa Bay during 2006e2010 (Taino, 2011). differ significantly among spring, summer, and winter (overall c Code numbers are used in Fig. 2. mean 1111 ± 889 g, n ¼ 93, Tukey HSD test, p > 0.05), but the algal clumps were significantly smaller in the autumn than those in the spring and summer (421 ± 479 g, n ¼ 19, Tukey HSD test, p < 0.05), Clarke (1993). The similarity matrix obtained was subjected to a although this occurred only in 2009. two-way mixed model permutational multivariate analysis of variance (PERMANOVA) (Anderson, 2001) with algal type as a fix 3.2. Seasonal patterns of fishes associated with drifting algae factor and sampling month as a random factor. The similarity of fish assemblages in drifting temperate and tropical algae each month During the study period, 32 fish species (4509 individuals) were were represented graphically using nonmetric multidimensional recorded from 22 families (Appendix). The dominant species were scaling (MDS). The ordinations were performed using data aver- Petroscirtes breviceps (30.0% of individuals) and Abudefduf vaigiensis aged over the replicates within each algal type for each month to (19.6%). The next most dominant species were cir- simplify the presentation and to make the habitat groupings rhifer (13.6%), Seriola quinqueradiata (8.5%), Oplegnathus fasciatus clearer. PERMANOVA and MDS were performed using the PRIMER (6.2%), and Rudarius ercodes (5.0%), which together accounted for (v6) computer package. approximately 35.0% of the total individuals. Most of the fish

Fig. 2. Relative weight of each Sargassum species within all of the algal clumps collected in each month. The total numbers of algal clumps collected in each month are shown above each column. The code number of each species is given in Table 1. White, gray or black portion represents temperate, tropical or unidentified (i.e. others in Table 1) algal species, respectively. 36 M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41

Fig. 3. Seasonal changes in: (a) the mean fish species richness (þSD) per algal clump, (b) mean abundance of fishes per algal clump, and (c) mean abundance of fishes per 1 kg of algae in each month in Tosa Bay. The total number of species collected in each month is shown above each mean species richness bar. The dotted lines indicate the average monthly surface seawater temperature at the three sites within the study area (Fig. 1d). Descriptive patterns of dominant fish in Fig. 3b; Abudefduf vaigiensis ( ), Petroscirtes breviceps ( ), Histrio histrio ( ), Seriola quinqueradiata ( ), Oplegnathus fasciatus ( ), Oplegnathus punctatus (-), Stephanolepis cirrhifer ( ), Rudarius ercodes ( ), Paramonacanthus japonicus ( ), Scombridae spp. ( ), Girella punctata ( ), and others (,). captured around the drifting algae were small juveniles (Appendix). The seasonal species richness pattern and the abundance per one algal clump in general paralleled that of the seawater tem- perature, with higher numbers during summer (JulyeSeptember) and lower numbers during late winter and spring (MarcheMay) (Fig. 3a, b) (KruskaleWallis test, p < 0.01 for each number among the 16 months of fish sampling). Although the trend was less clear than the above, a similar seasonal pattern was obtained for the fish abundance per 1 kg of algae (KruskaleWallis test, p < 0.01) (Fig. 3c). There were clear seasonal changes in the occurrence of the domi- nant fishes: Seriola quinqueradiata and Scombridae spp. were sometimes captured in large aggregations around drifting algae from late winter to spring (MarcheMay) (Fig. 3b), whereas Petro- scirtes breviceps and Abudefduf vaigiensis were common during the summer and autumn. S. cirrhifer juveniles were not abundant during 2010, but they dominated in the summer of 2011. Other minor species occurred primarily during the summer (others in Fig. 3b, Appendix). The relationship between the weight of a drifting algal clump and the number of fish varied among seasons (Fig. 4a, b). During the spring, when Seriola quinqueradiata and Scombridae spp. were dominant, the relationship was unclear and not significant (n ¼ 49, r ¼ 0.19, p ¼ 0.19 for species richness; r ¼ 0.13, p ¼ 0.39 for abun- dance). During the summer, when resident fish (e.g., Pomacen- Fig. 4. Relationships between the wet weights of drifting algal clumps and (a) the tridae, Blenniidae, and Monacanthidae) were dominant, the species richness and (b) abundance of associated fishes in Tosa Bay. The spring and relationship was positive and significant for both the species summer data were pooled for each sampling year (2010 and 2011). M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41 37 richness (n ¼ 42, r ¼ 0.42, p < 0.01) and abundance (n ¼ 42, r ¼ 0.52, assemblages between seasons (spring group: AprileMay 2011; and p < 0.01). During the autumn, when the algal clump size was summereautumn group: JuneeAugust and November smaller than that in other seasons, the relationship was only sig- 2009e2011), rather than algal types, with a similarity level of 35.0% nificant for abundance (n ¼ 19, r ¼ 0.60, p < 0.01) and not for for the abundance data per clump (Fig. 6a) and per 1 kg algae species richness (n ¼ 19, r ¼ 0.18, p ¼ 0.47). (Fig. 6b). The two-way PERMANOVA also detected a significant Both tropical algal clumps and temperate algal clumps were difference among sampling months but not between the algal types collected during 6 of the 16 months when fish were sampled for the abundance data per clump (PERMANOVA: Pseudo-F ¼ 0.71, (November 2009, June and August 2010, and April, May, and June p ¼ 0.59 for algal type, Pseudo-F ¼ 8.03, p < 0.01 for sampling 2011), and these months were subjected to the algal typeefish months) and for the abundance data per 1 kg algae (PERMANOVA: association analysis. The mean weight of algal clumps did not differ Pseudo-F ¼ 0.63, p ¼ 0.60 for algal type, Pseudo-F ¼ 7.83, p < 0.01 significantly between algal types during the 6 months (two-way for sampling months). Moreover, there were no differences be- ANOVA: F1,37 ¼ 0.32, p ¼ 0.59), and the number of associated fish tween the dominant fish species in the algal types, where eight of per one clump and per 1 kg algae also did not differ significantly the nine dominant species were the same in temperate and tropical between the algal types (two-way ANOVA: F1,37 ¼ 0.50, p ¼ 0.50 for algae (Table 2). species richness; F1,37 ¼ 1.11, p ¼ 0.33 for abundance per clump; ¼ ¼ e F1,37 0.14, p 0.71 for abundance per 1 kg algae) (Fig. 5a c). In 3.3. Diets of fish species associated with drifting algae terms of temporal differences, the weight of algal clumps did not differ significantly between the 6 sampling months (two-way Nine of the 14 fish species fed primarily on planktonic , ¼ ¼ fi ANOVA: F5,37 3.51, p 0.10), but the number of associated sh such as calanoid (Fig. 7). Epiphytic animals, such as har- exhibited a clear temporal difference (two-way ANOVA: pacticoid copepods and bivalves, were consumed by these nine ¼ ¼ ¼ < F5,37 4.68, p 0.05 for species richness; F5,37 6.11, p 0.05 for species, but the percentage compositions of these food items were ¼ < abundance per clump; F5,37 5.77, p 0.05 for abundance per 1 kg low (<30.0%). Petroscirtes breviceps, Stephanolepis cirrhifer, and algae) (Fig. 5aec). Multidimensional scaling of the similarities of Paramonacanthus japonicus fed on both planktonic and epiphytic the fish assemblages associated with the temperate and tropical animals. Histrio histrio fed on fishes associated with drifting algae, algal clumps during each month indicated a clear grouping of such as H. histrio, P. breviceps, Seriola quinqueradiata, and S. cirrhifer. Scombridae spp. feed on fish eggs (probably eggs of Belonoidei) and these eggs were also present in the gut contents of the other six species (Fig. 7).

Fig. 6. Multidimensional scaling of the similarity of fish assemblages [the abundance Fig. 5. (a) Mean fish species richness (þSD) per algal clump, (b) mean abundance of data (a) per clump and (b) per 1 kg algae] associated with drifting temperate and fishes per algal clump, and (c) mean abundance of fishes per 1 kg of algae associated tropical algal clumps in each month in each year. The open symbols indicate fish as- with drifting temperate algal clumps (▫) and tropical algal clumps (-) in each month semblages associated with temperate algal clumps, whereas the solid symbols indicate in Tosa Bay. The number of temperate (left) and tropical (right) algal clump analyzed is those in tropical algal clumps. Jun indicates the fish assemblage from June 2010 and given below the month. Jun2 indicate that from June 2011. 38 M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41

Table 2 findings. By contrast, only one tropical Sargassum species Mean abundance (±SD) of the nine dominant fishes associated with drifting (Sargassum ilicifolium) was found in Ohno (1984, a, b) (8.3% of total temperate and tropical algae (individuals per kg) in the Tosa Baya. species richness) (Table 1), whereas we found six tropical species Temperate algae (n ¼ 26) Tropical algae (n ¼ 23) among the 19 species (31.6%), thereby suggesting that the propor- Species Mean ± SD Species Mean ± SD tion of tropical Sargassum species has increased among drifting algae during the last 30 years. Some drifting algae could have been Seriola quinqueradiata 12.1 ± 32.4 Petroscirtes breviceps 32.9 ± 47.9 Petroscirtes breviceps 11.5 ± 28.1 Abudefduf vagiensis 29.1 ± 54.3 transported from other regions, but this phenomenon probably Stephanolepis cirrhifer 3.8 ± 14.1 Seriola quinqueradiata 9.4 ± 30.0 reflects the expansion of the distribution range of tropical species Abudefduf vagiensis 3.5 ± 8.4 Stephanolepis cirrhifer 7.3 ± 15.2 and a contraction of the range of temperate species in the coastal ± ± Girella punctata 2.8 10.8 Paramonacanthus japonicus 4.8 19.3 area of Tosa Bay since the late 1980s (Tanaka et al., 2012). Rudarius ercodes 2.2 ± 7.7 Terapon theraps 4.3 ± 11.9 Oplegnathus fasciatus 1.5 ± 7.1 Girella punctata 2.9 ± 10.1 The temperate algae drifting off Tosa Bay were abundantly Paramonacanthus 0.7 ± 3.5 Histrio histrio 2.6 ± 4.0 present during the spring and summer, whereas the drifting trop- japonicus ical algae appeared primarily during the summer, and these pat- Histrio histrio 0.5 ± 1.1 Oplegnathus fasciatus 2.3 ± 5.9 terns roughly reflected the seasonal vegetation patterns of both a Data were pooled for the 6 sampling months (November 2009, June and August types of algae in coastal areas (Terazono et al., 2012). In contrast to 2010, and April, May and June 2011). the autumns of 2010 and 2011, many clumps were collected during the autumn of 2009. The mean coastal SST during JulyeOctober 4. Discussion was lower in 2009 (26.1 C) than those in 2010 (26.8 C) and 2011 (26.5 C) (data source: Kochi Prefectural Fisheries Experiment The declines and shifts in habitat-forming macroalgal species is Station); thus, the vegetation periods of both types of algae may because of climatic and human-induced events, and their effects on have been extended to late summer of 2009, thereby causing the coastal ecosystems and fisheries is a growing concern worldwide unusual occurrence of drifting algae in autumn. Clear seasonal (Steneck et al., 2002; Schaffelke et al., 2006; Harley et al., 2012; occurrence patterns were found for both types of drifting algae in Wernberg et al., 2013). However, we still have little indication of this study, but there was no repeating occurrence pattern for many the ecological and economic impacts of these changes in the pelagic of the algal species comprising the drifting algal clump (Fig. 2). environment. Thus, to the best of our knowledge, this is the first Thus, the occurrence patterns of each of these algal species could be study to report the responses of fish species to drifting indigenous attributable to a few dominant Sargassum spp. along the coast of temperate and invasive tropical algae in warming temperate Tosa Bay (small populations of most algal species) (Tanaka et al., offshore waters. 2012), the low sample size used in our study (up to 10 clumps), Sixteen out of 19 Sargassum species identified in the present and/or spatiotemporal changes of the surface current in the bay study were confirmed as present in the vegetation in Tosa Bay (Fujimoto, 1987). The peak abundances of the drifting temperate during 2006e2010 (Taino et al., 2011)(Table 1). Sargassum glau- and tropical algae occurred from late spring to early summer cescens was recently confirmed in the vegetation in Tosa Bay (S. because the thalli of many temperate and tropical algae species are Taino, personal communication) and the possible origins of the shed after their reproductive season from spring to middle summer other two species (Sargassum horneri and Sargassum ring- (Mayelate July) (Haraguchi et al., 2006; Terazono et al., 2012). In goldianum) may be Kyushu, Seso Inland Sea, and/or the Chinese warm seasons, the fish assemblages did not differ significantly coast (Ohno, 1984a, b; Komatsu et al., 2007a). Ohno (1984a) found between the temperate and tropical algal clumps, thereby sug- 12 Sargassum species in a monthly drifting algae survey in Tosa Bay gesting that the low hostealgal specificity of most fishes was during 1980e1982, where S. horneri, Sargassum hemiphyllum, possibly because both algae had similar highly complex thallus Sargassum yamamotoi, Sargassum siliquastrum, and Sargassum nip- morphology (similar shelter function). Moreover, many of the ponicum were dominant by weight, which are similar to our resident fish captured within both types of drifting algae are found

Fig. 7. Mean percentage volume of food items (%V) for each fish species. SL, standard length; and n, number of fish examined that contained food. Descriptive codes for the food items: Calanoid copepods (Cc), Appendiculata (Ap), cladocerans (Cl), Luciferidae (Lu), harpacticoid copepods (Hc), gammaridean amphipods (Gm), caprellid amphipods (Ca), bi- valves (Bi), fish eggs (Fe), and fishes (F). Items comprised <5.0% of the gut content volume for each species were included in planktonic animals (Pa) or epiphytic animals (Ea). The food items in white indicate planktonic food resources and those in gray indicate epiphytic food resources. M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41 39 in coastal temperate and tropical algal beds (e.g., Petroscirtes bre- 2011; Voerman et al., 2013). Moreover, a recent study suggests viceps, Rudarius ercodes, and Stephanolepis cirrhifer)(Terazono et al., that approximately 25% of the current seaweed flora (100e350 2012), which suggests that some fish accompanied the algae when spp.) in southern Australia will become extinct within the next 60 the thalli were detached from rocky reefs. years due to temperature increases (Wernberg et al., 2011). In The fish that aggregate with drifting algae may obtain food by Japan, approximately 20% of the area of seaweed beds (kelps and preying on epiphytic invertebrates and/or zooplankton in the sur- fucoids) was lost (18,538 ha) within 20 years since the late 1980s, rounding waters (Kingsford and Milicich, 1987; Safran and Omori, which was caused by the climate change, heavy grazing (sea urchin 1990), which is partly supported by our study. However, there and fishes), and/or anthropogenic stressors such as reduced water was no relationship between the size of the algal clumps and the quality (Fujita, 2010). The decline and disappearance of seaweed abundance of the associated fish, suggesting that many fish are beds have also been observed in many areas of Tosa Bay, where the attracted to drifting algae to obtain shelter from predators, rather size of the deforested barren areas (292.5 ha) is almost the same as than to acquire epiphytic food. Therefore, it is possible that the the vegetated areas (345 ha) in 2010 (Taino et al., 2011). Quanti- juveniles of these fish species use the drifting algae as a nursery tative data are lacking regarding the long-term decline of drifting refuge in a relatively rich food pelagic system, rather than the algae off Tosa Bay, but the expansion of deforested barren areas and surrounding open water area (Castro et al., 2002). Nevertheless, our the shorter period of availability of drifting algae due to shifts from results should be interpreted with caution because the samples temperate to tropical Sargassum species will have negative impacts were collected using a dip net (diameter, 70 cm) thereby preventing on some fish populations and regional yellowtail captures for the collection of large algal clumps (>1m3 in size). The tropical fisheries. Indeed, similar seaweed species shifts with increasing SST Sargassum spp. found on the temperate coasts are characterized by are becoming common in western Japan (Aratake et al., 2007; a lower thallus compared with temperate species (Nagai et al., Tanaka et al., 2013), and this trend is expected to continue in the 2011; Terazono et al., 2012), suggesting that the large algal future because climate projections suggest that the SST around clumps were not collected could have been temperate algal species Japan is very likely to increase in the next 100 years (Japan although there were few large clumps in the study area (Ohno, Metrological Agency, 2013). A possible solution to alleviate this 1984a, b; Taino, 2006). Further research using a purse-seine net concern is placing artificial drifting objects offshore as yellowtail (Kingsford and Choat, 1985; Massuti et al., 1999) is required to juveniles are attracted to floating objects (algae) rather than better understand the relationship between the abundance and epiphytic food resources (Fig. 7). However, climate change and the diversity of the fish community and both types of drifting algae, as subsequent biological responses are often not straightforward well as the habitat function of drifting algae for small fishes. (Duffy, 2003; Hsieh et al., 2005) and thus the yellowtail might shift The fish fauna of drifting algae was clearly seasonal and their spawning season or spawning grounds depending on the yellowtail juveniles frequently aggregated with drifting algae from changes in water temperature (Kuwahara et al., 2006; Komatsu late winter to spring (MarcheMay). During this period, the drifting et al., 2007b), but we consider that our results may help to eluci- algae primarily comprised temperate Sargassum species (Fig. 2), date the potential sensitivity of temperate pelagic ecosystems and although tropical Sargassum species were present along the coasts related fisheries to climate change. This is a first step in the (Terazono et al., 2012). This phenomenon probably occurred development of an adaptation policy that may minimize the effects because tropical Sargassum species are not detached as easily from of climate change on pelagic ecosystem services in the future. the seafloor by ocean waves as temperate species because the temperate species were taller than tropical species during this season (some temperate species had already grown and even Acknowledgments matured, whereas the tropical species had just commenced their growth) (Haraguchi et al., 2006; Terazono et al., 2012). Therefore, We are grateful to K. Yamaoka and M. Hiraoka for logistic sup- the occurrence of temperate algal clumps from late winter to early port. We are also grateful to S. Taino for assistance with the iden- spring is important for yellowtail juveniles and their capture for tification of Sargassum species. The SST data were provided by fisheries because tropical algal clumps are unavailable during this Kochi Prefectural Fisheries Experiment Station. The constructive period. comments on the manuscript by anonymous reviewers were much One of the most notable recent changes in temperate reef en- appreciated. This study was supported by a grant from the Ministry vironments throughout the world has been the dramatic decline in of Education, Culture, Sports, Science and Technology of Japan macroalgal communities (Colemen et al., 2008; Wernberg et al., (FY2009-2011). 40

Appendix

Mean abundance (±SD) (individuals per kg) and body size of fishes associated with drifting algae in Tosa Bay during the study period. "" indicates not collected.

Family Species Total abundance Month a Standard length (mm) Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. MinMax Mean ± SD

Antennariidae Histrio histrio 114 e 1.5 ± 4.5 1.7 ± 1.8 1.3 ± 1.5 1.2 ± 2.5 4.8 ± 4.9 0.05 ± 0.1 e 0.4 ± 1.0 1.5 7.0e95.0 33.2 ± 22.8 32 (2014) 147 Science Shelf and Coastal Estuarine, / al. et Yamasaki M. Fistulariidae Fistularia commersonii 1 eeeeee0.4 ± 1.4 eee97 97 Syngnathidae Syngnathus schlegeli 2 eee0.04 ± 0.1 0.02 ± 0.1 eee ee96.6e118.1 107.4 ± 15.2 Syngnathoides biaculeatus 2 eeeeee0.4 ± 1.4 e 0.1 ± 0.3 e 93.0e121.0 107 ± 19.8 Exocoetidae Exocoetidae sp. 1 eeee0.06 ± 0.3 eee ee15.1 15.1 Apogonidae Apogon semilineatus 1 eeeeeee0.9 ± 2.9 ee13 13 ± Coryphaenidae Coryphaena hippurus 2 ee0.1 ± 0.4 e 0.03 ± 0.1 eee ee32.0e82.4 57.2 35.6 Carangidae Seriola quinqueradiata 385 6.4 ± 9.1 33.5 ± 53.0 14.2 ± 21.0 0.2 ± 0.4 e 0.4 ± 1.5 ee ee10.0e86.0 28.0 ± 12.2 Seriola dumerili 9 e 0.7 ± 2.3 0.1 ± 0.4 0.04 ± 0.2 0.3 ± 1.3 1.0 ± 3.2 ee ee17.0e36.0 28.7 ± 5.8 Lobotidae Lobotes surinamensis 1 eeeee0.4 ± 1.5 ee ee55 55 Pomacentridae Abudefduf vagiensis 882 ee0.7 ± 1.4 0.7 ± 1.3 21.6 ± 19.8 61.1 ± 65.9 4.5 ± 7.6 102.7 ± 114.3 9.9 ± 13.2 7.4 6.0e48.0 12.4 ± 3.4 Pomacentrus coelestis 1 eeeeeee0.9 ± 2.9 ee12 12 e e ± Teraponidae Terapon theraps 99 eee0.24 ± 0.7 0.4 ± 1.0 9.0 ± 16.2 2.2 ± 6.9 10.9 ± 14.9 0.9 ± 2.6 10.0 57.0 18.6 7.2 Labracoglossidae Labracoglossa argentiventris 4 eeee0.1 ± 0.3 1.1 ± 2.6 ee ee9.0e15.5 13.6 ± 3.1 Kuhliidae Kuhlia marginata 1 eeeeeee0.9 ± 2.9 ee16 16 Oplegnathidae Oplegnathus fasciatus 281 e 1.3 ± 4.6 0.02 ± 0.07 5.7 ± 10.2 7.7 ± 9.7 0.2 ± 0.6 e 0.2 ± 0.5 ee7.7e41.6 15.8 ± 4.3 Oplegnathus punctatus 46 0.3 ± 0.4 3.1 ± 10.1 0.7 ± 1.0 1.0 ± 1.2 eeee ee8.0e78.7 23.1 ± 13.2 Kyphosidae Kyphosus vaigiensis 49 e 0.03 ± 0.09 ee1.1 ± 2.1 4.6 ± 6.9 e 9.0 ± 21.7 ee9.0e56.0 17.5 ± 8.0 ± eeeee eee ± Girellidae Girella punctata 172 e 15.3 ± 20.8 0.091.4 7.0 27.0 17.3 2.9 Nomeidae Psenes cyanophrys 1 eeee0.07 ± 0.3 eee ee31.5 31.5 Blenniidae Petroscirtes breviceps 1349 eee0.2 ± 0.4 21.9 ± 23.9 76.0 ± 44.5 11.3 ± 8.5 135.6 ± 195.3 27.6 ± 36.3 42.6 9.0e51.0 21.7 ± 5.1 Sphyraenidae Sphyraena flavicauda 2 eeeeeee1.8 ± 5.7 ee2023 21.5 ± 2.1 Scombridae Scombridae spp. 73 10.4 ± 14.7 eeeeeee ee88.0e114.0 97.9 ± 5.1 fl eeeeee ± eee

Balistidae Pseudobalistes avimarginatus 1 0.08 0.3 22 22 e Sufflamen fraenatus 1 eeeee0.6 ± 2.2 ee ee30 30 41 e ± Canthidermis maculata 3 ee0.2 ± 0.7 e 0.08 ± 0.2 ee0.2 ± 0.7 ee8.3 26 17.3 8.9 Monacanthidae Aluterus scriptus 2 eeee0.1 ± 0.5 eee ee58.6e61.2 59.9 ± 1.8 Rudarius ercodes 225 eee1.1 ± 3.8 3.7 ± 6.2 10.4 ± 21.6 2.8 ± 5.9 2.7 ± 6.0 0.6 ± 1.7 1.5 5.0e21.0 8.8 ± 2.9 Thammaconus modestus 3 eee 0.1 ± 0.6 eee ee12.0e17.1 15.2 ± 2.8 Stephanolepis cirrhifer 614 ee0.8 ± 1.6 15.8 ± 21.6 7.9 ± 9.5 5.7 ± 14.1 ee ee5.9e37.1 14.8 ± 4.7 Paramonacanthus japonicus 181 eee0.5 ± 1.9 0.5 ± 0.9 10.9 ± 27.7 23.6 ± 28.7 3.0 ± 6.5 ee5.0e22.0 10.9 ± 3.3 Diodontidae Diodon eydouxii 1 eeee0.04 ± 0.2 eee ee43 43 a Data were pooled for the years (2009e2011). M. Yamasaki et al. / Estuarine, Coastal and Shelf Science 147 (2014) 32e41 41

References Massuti, E., Morales-Nin, B., Deudero, S., 1999. Fish fauna associated with floating objects sampled by experimental and commercial purse nets. Sci. Mar. 63, 219e227. Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of Mattio, L., Payri, C.E., Verlaque, M., 2009. Taxonomic revision and geographic dis- variance. Austral Ecol. 26, 32e46. tribution of the subgenus Sargassum (Fucales, Phaeophyceae) in the western Aratake, H., Shimizu, H., Watanabe, K., Yoshida, G., 2007. Long-term change in and central pacific islands based on morphological and molecular analysis. Sargassum-bed distribution along the coast of Kushima-city, southern part of J. Phycol. 45, 1213e1227. Miyazaki Prefecture, Japan. Bull. Miyazaki Prefect. Fish. Res. Inst. 11, 1e13. Nagai, S., Yoshida, G., Tarutani, K., 2011. Changes in species composition and dis- Castro, J.J., Santiago, J.A., Santana-Ortega, A.T., 2002. A general theory on fish ag- tribution of algae in the coastal waters of western Japan. In: Casalengno, S. (Ed.), gregation to floating object: an alternative to the meeting point hypothesis. Rev. Global Warming Impacts e Case Studies on the Economy, Human Health, and Fish Biol. Fish. 11, 255e277. on Urban and Natural Environments. InTech, Shanghai, pp. 209e236. Clarke, K.R., 1993. Non-parametric multivariate analysis of changes in community Nakabo, T. (Ed.), 2002. Fishes of Japan with Pictorial Keys to the Species, English ed. structure. Aust. J. Ecol. 18, 117e143. Tokai University Press, Tokyo, p. 1749. Coleman, M.A., Kelaher, B.P., Steinberg, P.D., Millar, A.J.K., 2008. Absence of a large Nakada, M., 2008. Capture-based aquaculture of yellowtail. In: Lovatelli, A., brown macroalgae on urbanized rocky reefs around Sydney, Australia, and Holthus, P.F. (Eds.), Capture-based Aquaculture. Global Overview. FAO, Rome, evidence for historical decline. J. Phycol. 44, 897e901. pp. 199e215. FAO Fisheries Technical Paper. No. 508. Dempster, T., Kingsford, M.J., 2004. Drifting objects as habitat for pelagic juvenile Nakamura, Y., Horinouchi, M., Nakai, T., Sano, M., 2003. Food habits of fishes in a fish off New South Wales, Australia. Mar. Freshw. Res. 55, 675e687. bed on a fringing coral reef at Iriomote Island, southern Japan. Ichthyol. Duffy, J.E., 2003. Biodiversity loss, trophic skew and ecosystem functioning. Ecol. Res. 50, 15e22. Lett. 6, 680e687. Nakamura, Y., Feary, D.A., Kanda, M., Yamaoka, K., 2013. Tropical fishes dominate Feary, D.A., Pratchett, M., Emslie, M., Fowler, A., Figueira, W., Luiz, O.J., Nakamura, Y., temperate reef fish communities within western Japan. PLoS ONE 8, e81107. Booth, D.J., 2013. Latitudinal shifts in coral reef fishes: why some species do and Ohno, M., 1984a. Algological observation on the floating seaweeds of off shore others don’t shift. Fish. Fish. http://dx.doi.org/10.1111/faf.12036. water of Shikoku Island in Japan. Bull. Jpn. Soc. Sci. Fish. 50, 1653e1656. Fujimoto, M., 1987. On the flow types and current stability in Tosa Bay and adjacent Ohno, M., 1984b. Observation on the floating seaweeds of near-shore waters of waters. Umi Sora 62, 127e140 (in Japanese with English abstract. southern Japan. Hydrobiologia 116/117, 408e412. Fujita, D., 2010. Current status and problems of isoyake in Japan. Bull. Fish. Res. Safran, P., Omori, M., 1990. Some ecological observations on fishes associated with Agency 32, 33e42. drifting seaweed off Tohoku coast, Japan. Mar. Biol. 105, 395e402. Greenstein, B.J., Pandolfi, J.M., 2008. Escaping the heat: range shifts of reef coral Schaffelke, B., Smith, J.E., Hewitt, C.L., 2006. Introduced macroalgaeA growing taxa in coastal Western Australia. Glob. Change Biol. 14, 1e16. concern. J. Appl. Phycol. 18, 529e541. Haraguchi, H., Yamada, C., Imoto, Z., Ohno, M., Hiraoka, M., 2006. Species Compo- Serisawa, Y., Imoto, Z., Ishikawa, T., Ohno, M., 2004. Decline of the Ecklonia cava sition of Sargassum Beds on the Coast of Ogisaki, Kochi, Southern Japan. In: population associated with increased seawater temperatures in Tosa Bay, Bulletin of Marine Sciences and Fisheries, vol. 24. Kochi University, pp. 1e9. southern Japan. Fish. Sci. 70, 189e191. Harley, C.D.G., Hughes, A.R., Hultgren, K.M., Miner, B.G., Sorte, C.J.B., Thornber, C.S., Steneck, R.S., Graham, M.H., Bourque, B.J., Corbett, D., Erlandson, J.M., Estes, J.A., Rodriguez, L.F., Tomanek, L., Williams, S.L., 2006. The impacts of climate change Tegner, M.J., 2002. forest ecosystems: biodiversity, stability, resilience and in coastal marine systems. Ecol. Lett. 9, 228e241. future. Environ. Conserv. 29, 436e459. Harley, C.D.G., Anderson, K.M., Demes, K.W., Jorve, J.P., Kordas, R.L., Coyle, T.A., 2012. Taino, S., 2006. Distribution and species composition of the floating seaweeds along Effects of climate change on global seaweed communities. J. Phycol. 48, the coastal waters of Kochi Prefecture, southern Japan. Kaiyo Mon. 38, 590e594. 1064e1078. Taino, S., Tanaka, K., Hiraoka, M., 2011. Distributional Survey of Seaweed beds in Hsieh, C., Glaser, S.M., Lucas, A.J., Sugihara, G., 2005. Distinguishing random envi- Kochi Prefecture Sea Area, pp. 158e178. Annual report of Kochi Prefectural ronmental fluctuations from ecological catastrophes for the North Pacific Fisheries Experimental Station. Ocean. Nature 435, 336e340. Tanaka, K., Taino, S., Haraguchi, H., Prendergast, G., Hiraoka, M., 2012. Warming off Hiraoka, M., Ura, Y., Hagaguchi, H., 2005. Relationship between seaweed beds and southwestern Japan linked to distributional shifts of subtidal canopy-forming seawater temperature in the Tosa Bay. Aquabiology 27, 485e493. seaweeds. Ecol. Evol. 2, 2854e2865. Hoegh-Guldberg, O., Bruno, J.F., 2010. The impact of climate change on the world’s Tanaka, T., Yoshimitsu, S., Imayoshi, Y., Ishiga, Y., Terada, R., 2013. Distribution and marine ecosystems. Science 328, 1523e1528. characteristics of seaweed/seagrass community in Kagoshima Bay, Kagoshima Japan Metrological Agency, 2013. Weather Statistics Information: Global Environ- Prefecture, Japan. Nippon. Suisan Gakkaishi 79, 20e30. ment and Climate. Available: http://www.data.kishou.go.jp/climate/index.html Terazono, Y., Nakamura, Y., Imoto, Z., Hiraoka, M., 2012. Fish response to expanding (accessed 2013 December). tropical Sargassum beds on the temperate coasts of Japan. Mar. Ecol. Prog. Ser. Kingsford, M.J., Choat, J.H., 1985. The fauna associated with drift algae captured with 464, 209e220. a plankton-mesh purse seine net. Limnol. Oceanogr. 30, 618e630. Thompson, R.C., Crowe, T.P., Hawkins, S.J., 2002. Rocky intertidal communities: past Kingsford, M.J., Milicich, M.J., 1987. Presettlement phase of Parika scaber (Pisces: environmental changes, present status and predictions for the next 25 years. Monacanthidae): a temperate reef fish. Mar. Ecol. Prog. Ser. 36, 65e79. Environ. Conserv. 29, 168e191. Kingsford, M.J., 1992. Drift algae and small fish in coastal waters of northeastern Uehara, S., Taggart, C.T., Mitani, T., Suthers, I.M., 2006. The abundance of juvenile New Zealand. Mar. Ecol. Prog. Ser. 80, 41e55. yellowtail (Seriola quinqueradiata) near the Kuroshio: the roles of drifting Kingsford, M.J., 1993. Biotic and abiotic structure in the pelagic environment: seaweed and regional hydrography. Fish. Oceanogr. 15, 351e362. importance to small fishes. Bull. Mar. Sci. 53, 393e415. Voerman, S.E., Llera, E., Rico, J.M., 2013. Climate driven changes in subtidal kelp Komatsu, T., Tatsukawa, K., Filippi, J.B., Sagawa, T., Matsunaga, D., Mikami, A., forest communities in NW Spain. Mar. Environ. Res. 90, 119e127. Ishida, K., Ajisaka, T., Tanaka, K., Aoki, M., Wang, W.D., Liu, H.F., Zhang, S.D., Wernberg, T., Russell, B.D., Thomsen, M.S., Gurgel, F.D., Bradshaw, C.J.A., Zhou, M.D., Sugimoto, T., 2007a. Distribution of drifting seaweeds in eastern Poloczanska, E.S., Connell, S.D., 2011. Seaweed communities in retreat from East China Sea. J. Mar. Syst. 67, 245e252. ocean warming. Curr. Biol. 21, 1828e1832. Komatsu, T., Mikami, A., Matsunaga, D., Sagawa, T., Tanoue, H., Boisnier, E., Wernberg, T., Smale, D.A., Tuya, F., Thomsen, M.S., Langlois, T.J., de Bettignies, T., Suzue, M., Kusaka, T., Ishida, K., Michida, Y., 2007b. Influence of global warming Bennett, S., Rousseaux, C.S., 2013. An extreme climatic event alters marine on seaweed forests and drifting seaweeds. Kaiyo Mon. 39, 336e342. ecosystem structure in a global biodiversity hotspot. Nat. Clim. Change 3, 78e82. Kuroda, H., Shimizu, M., Hirota, Y., Ambe, D., Akiyama, H., 2008. Surface current and Wu, L., Cai, W., Zang, L., Nakamura, H., Timmermann, A., Joyce, T., McPhaden, M.J., vertical thermal structure on the continental slope in Tosa Bay. Estuar. Coast. Alexander, M., Qiu, B., Visbeck, M., Chang, P., Giese, B., 2012. Enhanced warming Shelf Sci. 64, 81e91. over the global subtropical western boundary currents. Nat. Clim. Change 2, Kuwahara, H., Akeda, S., Kobayashi, S., Takeshita, A., Yamashita, Y., Kido, K., 2006. 161e166. Predicted changes on the distribution areas of marine organisms around Japan Yamano, H., Sugihara, K., Nomura, K., 2011. Rapid poleward range expansion of caused by the global warming. Global Environ. Res. 10, 189e199. tropical reef corals in response to rising sea surface temperatures. Geogr. Res. Lima, F.P., Ribeiro, P.A., Queiroz, N., Hawkins, S.J., Santos, A.M., 2007. Do distribution Lett. 38, L04601. shifts of northern and southern species of algae match the warming pattern? Yoshida, T., 1998. Marine Algae of Japan. Uchidaroukakuho, Tokyo, p. 1222. Glob. Change Biol. 13, 2592e2604. 本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。 图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具