Marine Pollution Bulletin xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul
Habitat association of larval fish assemblages in the northern Persian Gulf Mahnaz Rabbaniha a, Juan Carlos Molinero b,⇑, Lucia López-López c, Jamileh Javidpour b, Ana Ligia Primo d, Feryadoon Owfi a, Ulrich Sommer b a
Iranian Fisheries Research Organization (IFRO), Marine Ecology, Tehran-Karaj Highway, Paykan Shahr, P.O. Box 14155-775, Tehran, Iran GEOMAR Helmholtz Centre for Ocean Research Kiel, Marine Ecology/Food Webs, Duesternbrooker Weg 20, D-24105 Kiel, Germany IEO Centro Oceanográfico de Santander, Promontorio San Martín, s/n, P.O. Box 240, 39080 Santander, Spain d Centre for Functional Ecology, Department of Life Sciences, University of Coimbra, Calçada Martim de Freitas, 3000-456 Coimbra, Portugal b c
a r t i c l e
i n f o
Article history: Received 16 April 2015 Revised 9 June 2015 Accepted 11 June 2015 Available online xxxx Keywords: Fish larvae Habitat association Persian Gulf Pelagic and demersal spawners Clupeidae
a b s t r a c t We examined the habitat use of fish larvae in the northern Persian Gulf from July 2006 to June 2007. Correspondence Analysis showed significant differences between hydrological seasons in habitat use and structure of larval fish assemblages, while no differences were found regarding abundance among coralline and non-coralline habitats. The observed configuration resulted in part from seasonal reproductive patterns of dominant fish influencing the ratio pelagic:demersal spawned larvae. The ratio increased along with temperature and chlorophyll-a concentration, which likely fostered the reproduction of pelagic spawner fish. The close covariation with temperature throughout hydrographic seasons suggests a leading role of temperature in the seasonal structure of larvae assemblages. Our results provide new insights on fish larval ecology in a traditionally sub-sampled and highly exposed zone to anthropogenic pollution, the northern Persian Gulf, and highlight the potential role of Khark and Kharko Islands in conservation and fishery management in the area. Ó 2015 Published by Elsevier Ltd.
1. Introduction Understanding how fish larvae respond to space–time environmental conditions is essential to achieve adequate policies for the management of harvested populations. Analyses of observational data have shown that fish larvae dynamics is shaped by combined effects of environmental gradients, biological interactions and population connectivity. The former includes seasonal and inter annual variability of hydrographic structures that shape ichthyoplankton patterns (Perrier et al., 2012). Biological interactions relates to prey and predator densities, spawning patterns, i.e. time and location (Rakocinski et al., 1996), and food supply (Funes-Rodríguez et al., 2009), whereas population connectivity sculpt the rate of exchange among geographically distant subpopulations determining the relevant spatial scale at which populations are connected (Cowen et al., 2000). In this work we address the first issue. Coral reefs are heterogeneous environments that encompass a variety of shelter areas for fish larvae. These ecosystems offer a
⇑ Corresponding author. E-mail address: [email protected] (J.C. Molinero).
large number of potential resource axes, such as feeding, refuge and breeding, which favor the life history success and shape behavioural patterns (Paris and Cowen, 2004; Gratwicke and Speight, 2005). These systems further support high diversity and are referred as valuable nursery areas for fish and invertebrates. In the northern Persian Gulf, little is known on ecological aspects of fish larvae in the Khark and Kharko Islands, despite their potential importance in recruitment and ultimately in stock management. Yet, fisheries research in this zone has been mainly focused on taxonomy and species richness (Rabbaniha, 1998, 2002; Dehghan et al., 2000), while drivers of larval assemblages and their seasonal changes in habitat use remain elusive. These issues are fundamental to gain understanding on fish ecology, whereas their quantification bears vast implications for a sustainable management of harvested fish in an area that is heavily exposed to anthropogenic stress, i.e. oil-related pollution. Indeed, the marine habitat in the northern Persian Gulf is particularly interesting as marine populations are heavily exposed to both hydroclimate forcing i.e. high levels of temperature and salinity stress, and habitat deteriorating due to the rapid development of the region, as well as the exposure to extensive oil extraction, transportation and refinement (Sheppard et al., 2010). Understanding how fish larvae respond
http://dx.doi.org/10.1016/j.marpolbul.2015.06.028 0025-326X/Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
2
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
to such stressors is therefore crucial for their adequate management. Here we examine larval fish assemblages, we assess their habitat associations and environmental factors favoring dominant larvae groups, and aim to provide a synoptic picture of space–time patterns in the Khark and Kharko Islands, northern Persian Gulf.
2. Materials and methods 2.1. Study area and data collection The Khark and Kharko Islands are located in the northern Persian Gulf, Bushehr Province, Iran (29° 120 to 29° 200 N, and 50° 160 to 50° 210 E) (Fig. 1). Khark is a rocky island located about 57 km off the Bushehr Port, while Kharko is a smaler, sandy island surrounded by reefs patches about 60 km north-west off Bushehr. Sampling stations (maximum depth 32 m) covered both coralline and non-coralline sites, as well as the coastal area. Nine stations were located around the islands and three stations in coastal waters (Fig. 1). The sampled area included a variety of habitats that harbour a large diversity of marine life, including fish, molluscs, lobsters and sea birds (Zanjani, 2012). Sampling was carried out monthly, from July 2006 to June 2007, always at the same tidal phase (low tide) in order to account for the potential effect of tidal currents on distributional patterns of larval fish assemblages. At each sampling station, temperature and salinity were recorded in the upper 32 m depth layer, while chlorophyll-a measurements were done using a chlorophyll-a sensor in the upper 30 m depth layer. Larvae samples were taken during daylight using a 61 cm diameter bongo net of 500 lm mesh, with a digital flow meter adapted to the net mouth to measure the water volume filtered. At each station, the bongo net was towed 10 min at a constant speed of 1 knot. Oblique tows were done from near the bottom to the surface. Samples were fixed in 5% formaldehyde seawater immediately after towing and preserved in 10% ethanol. In laboratory, the developmental stage of larvae captured was registered. The scarcity of species-specific taxonomic keys for the region limited a high taxonomic resolution, therefore identification was done at the family level based on Leis and Rennis (1983) and Leis and Trnsky (1989). Adult habitat and spawning sites for the collected fish families were obtained from FishBase (Froese and Pauly, 2015). Fish data were sorted according to habitat types: coralline (5 stations), non-coralline (6 stations) and mouth of creek (1 station) (Fig. 1).
2.2. Data analysis To assess the spatial correlation of environmental conditions (i.e. temperature, salinity and chlorophyll-a) in the explored habitats we used the Mantel statistics (Mantel, 1967). The statistical significance of Mantel correlations was assessed by Monte Carlo permutation (1000 times). Hydrological patterns ascribed to temperature, salinity and chlorophyll-a were assessed by means of box-and-whisker plots to characterize the extent of monthly variations. Fish larval abundances were standardized to number of larvae per 10 m2 and diversity was calculated throughout the year by means of Shannon diversity and evenness index. Due to the non-normal distribution of data and the lack of variance homogeneity even after data transformation, we used the non-parametric Wilcoxon and Kruskal–Wallis tests to assess differences in the fish larvae abundance between hydrological seasons and habitats. Correspondence Analysis (CA) was used to evaluate the relative importance of habitats and seasonal environmental changes in larval assemblages. To compute CA we included only families with more than 2% of relative abundance (number of families used in the analysis = 10). CA was performed both seasonally and on a yearly basis. The ratio of pelagic:demersal fish larvae was calculated as the log (n1/n2), where n1 is the abundance of pelagic spawned larvae and n2 is the abundance of demersal spawned larvae. The index equals zero when the two groups show the same abundance, whereas positive values denote a higher abundance of pelagic spawned larvae and negative values indicate a dominance of demersal spawned larvae. This ratio broadly reflects the overall availability of nutrients in the marine system, and has been used as proxy of hypoxia events resulting from excess primary production and eutrophication, which negatively affect benthic/demersal fish (De Leiva Moreno et al., 2000). Environmental windows (i.e. temperature, salinity) linked to the density of dominant families were assessed by means of a non-parametric test of habitat association, as in Molinero et al. (2009). The association between dominant fish families and environment windows was displayed as traffic light diagrams to qualitatively show the extent of overlapping between the preferred environmental ranges among families. Traffic light diagrams display a colour range from dark (quantile .90) to light grey (quantile .10). Statistical analyses were performed in R (R Development Core Team, 2008) and Matlab.
3. Results and discussion 3.1. Environmental conditions
Fig. 1. Map of the Persian Gulf. The study area is located in the northeastern region. The inset shows the Khark and Kharko Islands and sampling stations.
No statistical differences were found in the environmental conditions relative to the habitats investigated, as revealed by the Mantel test (1000 iterations, p < 0.05), which measured the dissimilarity between the spatial locations, i.e. coralline, non-coralline, mouth of the creek. Instead, significant environmental changes were detected on the seasonal scale (Kruskal Wallis test, p < 0.05) driven by contrasting temperature variations, i.e. 40 °C in summer and 18 °C in winter (Fig. 2a). Salinity showed generally high values (42 ± 1.3; mean and standard deviation respectively), although a sharp decrease arose during the rainy season, in late summer (Fig. 2b). Chlorophyll-a ranged between 0.37 and 0.99 mg m3, with higher concentrations found in summer–autumn 2006 and spring 2007, although a large variability was observed (Fig. 2c). Likewise, conspicuous variability was observed in the monthly larval abundance, global average 7.9 ± 3.55 ind.10 m2 (Fig. 2d).
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
Fig. 2. Hydrographic conditions and larvae abundances during the period July 2006 to June 2007. Box-and-Whisker plots show monthly variations of (a) temperature, (b) salinity, (c) chlorophyll-a and (d) fish larvae at the Khark and Kharko area, northern Persian Gulf. Median (bold horizontal lines), maximum and minimum are indicated by whiskers. The length of the box accounts for the difference between the 25th and the 75th percentiles.
3.2. Habitat and seasonal changes in larval fish assemblages The development stage of collected larvae was mainly pre-flexion (90%), whereas only few were in stage flexion and post-flexion (10%). These results can be therefore used as indicators of the seasonal dynamics of the early stages of fish larvae. Twenty families accounted for the 92% of collected larvae. Clupeidae was the most abundant (29%), followed by Sigillinidae (14%) and Blenniidae (11%) (Table 1). Larvae abundance was on average 18.7 ± 0.90 ind.10 m2, and showed an evenness value of 0.56 ± 0.4. In coralline locations, we found 38 families and the average abundance was 18.7 ± 6.3 ind.10 m2, where Clupeidae accounted for 30% of the total abundance. The non-coralline habitats included 31 families with an average abundance of 14.9 ± 2.1 ind.10 m2. In these habitats, dominant families were Blenniidae and Sillaginidae. The mouth of the creek showed a lower taxonomic richness (16 families) with a mean abundance of 7.4 ± 1.3 ind.10 m2 and the families Sillaginidae and Gobiidae dominated this area (Table 1). The highest concentration of fish
3
larvae was found in the coralline zone, where larvae appeared associated with demersal habitats (Table 1). Typical coral reef fish accounted for 15% of the total abundance. The composition of larval fish assemblages identified by means of CA showed significant differences between hydrological seasons (Wilcoxon test, p < 0.05). However, we did not find unambiguous evidence on abundance changes among habitats (Kruskal Wallis test, p = 0.07). CA showed varied habitat association of larval fish assemblages (Fig. 3a). Regardless the season, the families Clupeidae and Atherinidae appeared linked with coralline sites, while Gobiidae and Sillaginidae were associated to the mouth of the creek stations. In turn, non-coralline habitats were occupied by Mullidae, Muglidae, Terapontidae, Carangidae and Blenniidae. Most of the families are associated to demersal and pelagic habitats, whereas only few were strictly coralline (Table 1). This picture changed between thermal regimes. During the high temperatures (July to October 2006 and April to June 2007; average temperature >30 °C), four families (Clupeidae, Blenniidae, Sillaginidae and Atherinidae) accounted for 67% of the larvae collected, with Clupeidae being the dominant family (39% of total abundance), which reflects the reproductive pattern of the species in the region occurring during spring and early summer (Van Zalling and Owfi, 1993). Associated with non-coralline sites we found the families Clupeidae, Atherinidae and Tripterygiidae, while Carangidae and Blenniidae were closely related to coralline sites. In turn, Gobiidae and Sillaginidae appeared linked with the mouth of the creek area. This period also showed a relatively large proportion of pelagic spawned larvae that reached up to 57% of the total abundance. The mean Shannon index was 0.83 ± 0.59, while the evenness was 0.62 ± 0.37 and the mean abundance was 24 ± 1 ind.10 m2. Overall, during the warmer period, the results of CA showed a similar structure as for the whole period (Fig. 3a). The decrease in temperature during autumn yielded a new hydrological regime that governed the period November 2006 to March 2007, when the average larvae concentration was 10 ± 0.4 ind.10 m2. The four most abundant families (Sillaginidae, Blenniidae, Mullidae and Pomacentridae) encompassed 49% of the total larval caught. Along with the decreasing temperature, the density of pelagic fish larvae dropped to 15% of the total abundance, which contrasted with a noticeable increase of demersal spawned larvae. The mean Shannon index and evenness were 0.55 ± 0.6 and 0.45 ± 0.44 respectively. In turn, larval assemblages showed a different configuration where Sillaginidae and Blenniidae were associated with coralline habitats, while Clupeidae shifted their spatial distribution to the mouth of the creek area, and non-coralline habitats were occupied by the families Sparidae, Engraulididae and Cepolidae (Fig. 3a). In reef areas, demersal spawned larvae are commonly associated with shallow environments, while pelagic spawned larvae showed a wider spatial distribution (López-Sanz et al., 2009). In our study site, we found demersal spawned larvae in dwelling either coralline habitats (i.e. Apogonidae, Belonidae and Clinidae) or dispersed through the three habitats considered (i.e. Blenniidae, Gobiidae and Trypterygiidae). In contrast, larvae from pelagic spawners were generally absent from the creek mouth area, with the exception of Carangidae and Clupeidae. Similar disparate patterns related with the spatial patterns of pelagic and demersal spawned larvae were registered in temperate waters of New Zealand (Hickford and Schiel, 2003), on the Canadian west coast (Marliave, 1986) and Off Nova Scotia (Suthers and Frank, 1991). The observed contrasting environmental configurations in part result from seasonal reproductive patterns of dominant families in the region. In line with this, we also observed conspicuous changes in structure of larval fish assemblages, as indexed by the ratio of pelagic:demersal spawned larvae. Larval assemblages displayed a dominance of pelagic larvae during warmer conditions
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
4
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
Table 1 Check list of families and mean abundance (ind.10 m2) of fish larvae in the northern Persian Gulf.
a b
Family
Coralline
Mouth of creek
Non-coralline
Relative abundance
Spawning
Habitat
Apogonidae Atherinidae Belonidae Blenniidae Bothidae Bregmaceratidae Callionymidae Carangidae Cepolidae Chaetodontidae Clinidae Clupeidae Clupeiform Creediidae Cynoglossidae Engraulididae Exocoetidae Gerreidae Gobiidae Haemulidae Hemiramphidae Labridae Leiognathidae Lethrinidae Lutjanidae Mugilidae Mugiloididae Mullidae Nemipteridae Platycephalidae Pomacentridae Sciaenidae Scomberidae Scorpaenidae Siganidae Sillaginidae Soleidae Sparidae Sphyraenidae Syngnathidae Synodontidae Terapontidae Tetraodontidae Triacanthidae Triglidae Tripterygiidae Other
0.01 ± 0.15 1.33 ± 11.79 0.03 ± 0.56 2.15 ± 6.98 0.28 ± 0.59 0.02 ± 0.42 0.12 ± 0.37 0.50 ± 1.00 0.15 ± 0.42 0.02 ± 0.44 0.01 ± 0.10 5.48 ± 41.33 0.25 ± 2.78 0.01 0.03 ± 0.28 0.27 ± 1.68 0.06 ± 0.50 0.21 ± 1.52 0.70 ± 0.94 0 0.19 ± 1.09 0.01 ± 0.21 0.06 ± 0.06 0.05 ± 0.10 0.03 ± 0.20 0.41 ± 0.48 0.02 ± 0.36 0.62 ± 0.15 0.08 0.07 ± 0.93 0.50 ± 4.26 0.01 ± 0.19 0.37 ± 2.32 0.06 ± 0.72 0.03 2.29 ± 8.61 0.08 ± 0.51 0.28 ± 0.46 0.22 ± 0.65 0.03 0.01 0.49 ± 0.50 0.03 ± 0.47 0.10 0.02 ± 0.16 0.87 ± 7.82 0.15 ± 0.88
0 0 0 0.15 ± 0.30 0.03 ± 0.10 0 0 0.10 ± 0.22 0 0 0 0.31 ± 0.53 0.02 ± 0.05 0 0.02 ± 0.05 0 0 0.10 ± 0.31 2.32 ± 6.16 0.04 ± 0.11 0 0 0 0 0.04 ± 0.11 0 0 0 0 0 0.49 ± 1.48 0 0 0 0 2.42 ± 6.72 0.26 ± 0.33 0.11 ± 0.34 0 0 0 0.15 ± 0.44 0 0.14 ± 0.33 0 0.69 ± 2.08 0
0 0.28 ± 1.08 0 2.58 ± 11.65 0.46 ± 1.91 0.00 0.17 ± 0.66 0.73 ± 2.14 0.22 ± 0.86 0 0 1.67 ± 3.89 0.07 ± 0.33 0.02 ± 0.19 0 0.30 ± 1.26 0.05 ± 0.41 0.07 ± 0.42 0.82 ± 3.10 0.00 0.10 ± 0.56 0 0.11 ± 0.55 0.10 ± 0.52 0.02 ± 0.13 0.76 ± 2.81 0 1.21 ± 5.13 0.16 ± 0.77 0.02 ± 0.17 0.41 ± 3.00 0 0.36 ± 1.41 0 0.05 ± 0.44 1.98 ± 7.86 0.00 0.46 ± 1.90 0.32 ± 0.76 0.06 ± 0.46 0.03 ± 0.23 0.82 ± 3.41 0 0.17 ± 1.39 0.02 ± 0.16 0.24 ± 1.49 0.08 ± 0.47
0.06 4.72 0.18 11.46 1.62 0.13 0.67 2.83 0.86 0.14 0.03 28.63 1.41 0.07 0.11 1.53 0.35 0.83 4.27 0.02 0.98 0.07 0.31 0.35 0.15 2.36 0.12 3.54 0.46 0.39 2.86 0.06 2.13 0.33 0.15 13.00 0.45 1.61 1.24 0.16 0.08 2.79 0.17 0.55 0.11 4.85 0.82
Demersal Demersal Demersal Demersal Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Demersal Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Demersal Pelagic –a Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic –b Pelagic Pelagic Pelagic Pelagic Pelagic Demersal –
Coral reefs Pelagic Pelagic Demersal Demersal Pelagic Demersal Pelagic Demersal Coral reefs Demersal Pelagic Pelagic Demersal Demersal Pelagic Pelagic Demersal Demersal Coral reefs Pelagic Coral reefs Demersal Demersal Demersal Demersal Demersal Demersal Demersal Demersal Coral reefs Demersal Pelagic Coral reefs Demersal Demersal Demersal Demersal Pelagic Demersal Demersal Demersal Demersal Demersal Demersal Demersal –
Male carry eggs. Eggs attached to seaweed and substrata.
and lower chlorophyll-a concentration. However, under lower temperatures and higher chlorophyll-a concentration, i.e. in late winter, demersal spawned larvae were more abundant (Fig. 3b). Previous studies have shown that the pelagic:demersal ratio broadly reflects the overall availability of nutrients in the marine system, that is the case for the Mediterranean Sea where the ratio appears correlated with the mean chlorophyll-a value, suggesting a better availability of nutrients (De Leiva Moreno et al., 2000). Generally, pelagic spawners are favored by nutrient enrichment, as it stimulates the plankton production (Caddy, 1993), while demersal ones reflect the dynamics of benthic communities, which generally responds negatively to the conditions of excessive enrichment, i.e. eutrophication (De Leiva Moreno et al., 2000). In agreement with this, in our study the ratio increased along with water temperature and chlorophyll-a concentration; this likely fostered the reproduction of pelagic spawner fish, such as Clupeidae, the numerically dominant fish family. Indeed, previous studies have reported a recurrent major peak of reproduction for this family during the same period (Van Zalling and Owfi, 1993). In addition, the Khark Island is largely exposed to oil pollution, as
suggested by the high concentrations of petroleum and polycyclic aromatic hydrocarbons in sediments (Khoei et al., 2013; Mirvakili et al., 2013). Polycyclic aromatic hydrocarbons are degraded by photo- and biodegradation processes and can permeate the food web reaching higher trophic levels. This pollution has likely a larger impact on suprabenthic communities and fauna associated to them, such as demersal fish, affecting their reproductive output. This hypothesis however should be confirmed with the appropriate data. 3.3. Environmental windows depicting the habitat use of dominant families The environmental windows relative to the abundance of the three dominant families were displayed by traffic light diagrams. The results highlighted similar thermal and salinity ranges, although the thermal window of Clupeidae was narrower than that showed by Sillaginidae and Blenniidae. The preferred depth range varied between families, with Clupeidae showing a shallower distribution (higher percentage found within the 9–12 m depth
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
5
Fig. 3. (a) Fish larvae abundance along the first two ordination axes of the correspondence analysis based on the habitat recurrence (coralline, non-coralline and mouth of creek). Results show for the whole period (top panel), as well as for warm (middle panel) and cold (low panel) seasons. The percentage of the total inertia contained in the two first dimensions was, respectively, 60.3% and 39.7% for the whole period; 67.8% and 32.2% for the warm season, and 67.8% and 32.18% for the cold season. Habitats are denoted by h1 (non-coralline), h2 (coralline) and h3 (mouth of the creek). (b) Response surface depicting the link between the ratio pelagic:demersal fish spawned larvae with temperature and chlorophyll-a gradients. (c) Habitat association of fish larvae and environmental factors. Monte Carlo simulation results show the preferred windows of the three dominant families during the survey period 2006–2007. Colours denote the strength of the association from black (intense association, p < 0.05) to grey light (no association). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
range), followed by Sillaginidae (10–18 m depth range) and Blenniidae (17–26 m depth range) (Fig. 3c). Temperature changes showed a leading role in the space–time pattern of larval fish assemblages, although some peaks of larval abundance matched chlorophyll-a peaks, which might allow species maximizing larval survival (Cushing, 1990). Similar observations were reported in the western Mediterranean Sea where the spawning period of neritic fish species was coupled with the maximum primary production (López-Sanz et al., 2009). Likewise, enhanced plankton production and maximum larval abundance have been observed in the subtropical western Pacific coasts (Franco-Gordo et al., 2008) and in East China waters (Chen et al., 2014). Further factors are however acknowledged to influence space–time patterns of fish larvae; for instance, upwelling processes (Hernandez-Miranda et al., 2003), spawning patterns of adult fishes (Sampey et al., 2004), and mesoscale currents (Pineda, 2000). In the northern Persian Gulf the main mesoscale hydrographic feature is a cyclonic overturning circulation, the Iranian Coastal Jet (ICJ) that enters through the Strait of Hormuz and establishes along the full length of the Gulf. The strength of the jet current develops in late spring-early summer and is followed by a relaxation in autumn and winter, when it becomes unstable and a clockwise circulation pattern appears in surface
layers in the northern area (Kampf and Sadrinasab, 2006). These variations in the seasonal circulation matched the timing of changes in the observed abundance pattern of Clupeidae, likely due to their larger exposure to dispersion as they mainly occupied shallow depths. Under warm conditions the relative abundance of the pelagic group reached 57% of the total assemblage, while in the cold season coastal and demersal spawned larvae were dominant (>50%). These observations stress the role of mesoscale hydrographic processes on the seasonal structure of larval fish assemblages. 3.4. Khark and Kharko Islands The marine environment of the northern Persian Gulf has long been exposed to oil pollution resulting from extensive extraction, transportation and refinement. In addition, the pace of change in resource damage warns on potential ecosystem-wide changes that may overwhelm the resilience of the system (Sheppard et al., 2010). Hence, there is a pressing need for research efforts pointing out the role marine protected areas, such as the Khark and Kharko Islands, might play for resources management, and the integration of such efforts into a large scale, long term perspective to achieve adequate policies. Here we have shown that the Khark and
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
6
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
Kharko Islands offer varied habitats that promote high densities of larval fish assemblages. This might result from the habitat structure and complexity that minimise competition by increasing habitat partitioning (Raberg and Kautsky, 2007), which in turn explains the high species richness found, as well as the high proportion of larvae in coralline habitats (40% of the total abundance). By promoting varied habitats and high species richness this zone may act as spawning centres in the area, as suggested by the large proportion of pre-flexion larval stages. Our results further pointed out that the dynamics of mesoscale currents influence the scale of connectivity of dominant families, and therefore management strategies in the Khark and Kharko Islands should be integrated into a broad spatial scale. These results provide a general picture of seasonal changes in the habitat association of larval fish assemblages and could be used as baseline for further fisheries and conservation studies in the northern Persian Gulf. Acknowledgements We are grateful to colleagues of Shrimp Research Center of Bushehr for their cooperative in cruise sampling and laboratory surveys. This work was funded by the German Research Foundation (DFG) through a visiting research fellowship to MR. References Caddy, J.F., 1993. Toward a comparative evaluation of human impacts on fishery ecosystems of enclosed and semi-enclosed seas. Rev. Fish. Sci. 1, 57–95. Chen, W.Y., Lee, M.A., Lan, K.W., Gong, G.C., 2014. Distributions and assemblages of larval fish in the East China Sea during the northeasterly and southwesterly monsoon seasons of 2008. Biogeosciences 11, 547–561. Cowen, R.K., Lwiza, K., Spounaugle, S., Paris, C., Olson, D.B., 2000. Connectivity of Marine populations: open or closed. Science 287, 857–859. Cushing, D., 1990. Plankton production and year-class strength in fish populations: an update of the match/mismatch hypothesis. Adv. Mar. Biol. 26, 249–293. De Leiva Moreno, J.I., Agostini, V.N., Caddy, J.F., Carocci, F., 2000. Is the pelagicdemersal ratio from fishery landings a useful proxy for nutrient availability? A preliminary data exploration for the semi-enclosed seas around Europe. ICES J. Mar. Sci. 57, 1091–1102. Dehghan, M.S., Savari, A., Kochnin, P., Maramazi, J.G., 2000. Abundance, diversity and distribution of fish larvae in the estuary and west coastal of Khouzestan. Fish. Sci. Bull. 4, 20–34. Franco-Gordo, C., Godínez-Domínguez, E., Freire, J., 2008. Interannual variability of the diversity and structure of ichthyoplankton assemblages in the central Mexican Pacific. Fish. Oceanogr. 17, 178–190. Froese, R., Pauly, D., (Eds.), 2015. FishBase. World Wide Web electronic publication. , version (04/2015). (accessed June 2015). Funes-Rodríguez, R., Elorduy-Garay, J.F., Hinojosa-Medina, A., Zárate-Villafranco, A., 2009. Interannual distribution of Pacific hake Merluccius productus larvae in the southern part of the California current. J. Fish. Biol. 75, 630–646. Gratwicke, B., Speight, M.R., 2005. The relationship between fish species richness, abundance and habitat complexity in a range of shallow tropical marine habitats. J. Fish. Biol. 66, 650–667.
Hernandez-Miranda, E., Palma, A., Ojeda, F., 2003. Larval fish assemblages in nearshore coastal waters off central Chile: temporal and spatial patterns. Estuar. Coast. Shelf. Sci. 56, 1–18. Hickford, M.J.H., Schiel, D.R., 2003. Comparative dispersal of larvae from demersal versus pelagic spawning fishes. Mar. Ecol. Prog. Ser. 252, 255–271. Kampf, J., Sadrinasab, M., 2006. The circulation of the Persian Gulf: a numerical model. Ocean Sci. 2, 27–41. Khoei, J.K., Bastami, A.A., Farmohammadi, S., 2013. Oil related PAH in fish and sediments in the Persian Gulf, Northern Part. World Appl. Sci. J. 24, 1378–1381. Leis, J.M., Rennis, D.S., 1983. The Larvae of Indo Pacific Coral Reef Fishes. New South Wales University Press, Sydney, p. 269. Leis, J.M., Trnsky, T., 1989. The Larvae of Indo Pacific Shore Fishes. New south Wales University Press, Sydney, p. 371. López-Sanz, Á., Vert, N., Zabala, M., Sabatés, A., 2009. Small-scale distribution of fish larvae around the Medes Islands marine protected area (NW Mediterranean). J. Plankton Res. 31, 763–775. Mantel, N., 1967. The detection of disease clustering and a generalized regression approach. Cancer Res. 27, 209–220. Marliave, J.B., 1986. Lack of planktonic dispersal of rocky intertidal fish larvae. Trans. Am. Fish Soc. 115, 149–154. Mirvakili, H.S., Zaker, N.H., Imani, F., 2013. Evaluation of oil pollution and origin in surface coastal sediments of Kharg Island in the Persian Gulf. J. Coastal Res. 65, 93–98. Molinero, J.C., Vukanic, V., Lucˇicˇ, D., Licandro, P., Souissi, S., Ibanez, F., Nival, P., Calbet, A., Christou, E.D., Daly-Yahia, N., Fernandez de Puelles, M.L., Mazzocchi, M.G., Siokou-Frangou, I., Vukanic, D., 2009. Mediterranean marine copepods. Basin scale trends of the calanoid Centropages typicus. Hydrobiologia 617, 41– 53. Paris, C.B., Cowen, R.K., 2004. Direct evidence of a biophysical retention mechanism for coral reef fish larvae. Limnol. Oceanogr. 49, 1964–1979. Perrier, C., Molinero, J.C., Gerdeaux, D., Anneville, O., 2012. Effects of temperature and food supply on the growth of whitefish Coregonus lavaretus larvae in an oligotrophic peri-alpine lake. J. Fish Biol. 81, 1501–1513. Pineda, J., 2000. Linking larval settlement to larval transport: assumptions, potentials, and pitfalls. Oceanogr. East. Pacific I, 84–105. R Development Core Team, 2008. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3900051-07-0, URL http://www.R-project.org. Rabbaniha, M., 1998. Study on Diversity and Abundance of Ichthyoplankton in the Nayband Bay (Bushehr Province). Iranian Fish. Res. Org. Press, p. 101. Rabbaniha, M., 2002. Study on Diversity and Abundance of Ichthyoplankton in the Creek-Estuary Bushehr to Genaveh. MSc. thesis. Iranian Fish. Res. Org. Press, p. 99. Raberg, S., Kautsky, L., 2007. Consumers affect prey biomass and diversity through resource partitioning. Ecology 88, 2468–2473. Rakocinski, C.F., Lyczkowski-Shultz, J., Richardson, S.L., 1996. Ichthyoplankton assemblage structure in Mississippi sound as revealed by canonical correspondence analysis. Estuar. Coast. Shelf Sci. 43, 237–257. Sampey, A., Meekan, M.G., Carleton, J.H., Mckinnon, A.D., McCormick, M.I., 2004. Temporal patterns in distribution of tropical fish larvae on the North-west Shelf of Australia. Mar. Freshwater Res. 55, 473–487. Sheppard, Ch., Al-Husiani, M., Al-Jamali, F., Al-Yamani, F., Baldwin, R., et al., 2010. The Gulf: a young sea in decline. Mar. Pollut. Bull. 60, 13–38. Suthers, I.M., Frank, K.T., 1991. Comparative persistence of marine fish larvae from pelagic versus demersal eggs off southwestern Nova Scotia, Canada. Mar. Biol. 108, 175–184. Van Zalling, N.P., Owfi, F., 1993. Resources of Small Pelagic in Southern Iranian Waters Bushehr. Iranian Fish. Res. Org. Press, Iran, p. 20. Zanjani, M.M., 2012. Ecotourism, expansion, alongside with dominant function of Khark and Kharko Islands. Int. J. Soc. Hum. Sci. 6, 478–481.
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
Contents lists available at ScienceDirect
Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul
Habitat association of larval fish assemblages in the northern Persian Gulf Mahnaz Rabbaniha a, Juan Carlos Molinero b,⇑, Lucia López-López c, Jamileh Javidpour b, Ana Ligia Primo d, Feryadoon Owfi a, Ulrich Sommer b a
Iranian Fisheries Research Organization (IFRO), Marine Ecology, Tehran-Karaj Highway, Paykan Shahr, P.O. Box 14155-775, Tehran, Iran GEOMAR Helmholtz Centre for Ocean Research Kiel, Marine Ecology/Food Webs, Duesternbrooker Weg 20, D-24105 Kiel, Germany IEO Centro Oceanográfico de Santander, Promontorio San Martín, s/n, P.O. Box 240, 39080 Santander, Spain d Centre for Functional Ecology, Department of Life Sciences, University of Coimbra, Calçada Martim de Freitas, 3000-456 Coimbra, Portugal b c
a r t i c l e
i n f o
Article history: Received 16 April 2015 Revised 9 June 2015 Accepted 11 June 2015 Available online xxxx Keywords: Fish larvae Habitat association Persian Gulf Pelagic and demersal spawners Clupeidae
a b s t r a c t We examined the habitat use of fish larvae in the northern Persian Gulf from July 2006 to June 2007. Correspondence Analysis showed significant differences between hydrological seasons in habitat use and structure of larval fish assemblages, while no differences were found regarding abundance among coralline and non-coralline habitats. The observed configuration resulted in part from seasonal reproductive patterns of dominant fish influencing the ratio pelagic:demersal spawned larvae. The ratio increased along with temperature and chlorophyll-a concentration, which likely fostered the reproduction of pelagic spawner fish. The close covariation with temperature throughout hydrographic seasons suggests a leading role of temperature in the seasonal structure of larvae assemblages. Our results provide new insights on fish larval ecology in a traditionally sub-sampled and highly exposed zone to anthropogenic pollution, the northern Persian Gulf, and highlight the potential role of Khark and Kharko Islands in conservation and fishery management in the area. Ó 2015 Published by Elsevier Ltd.
1. Introduction Understanding how fish larvae respond to space–time environmental conditions is essential to achieve adequate policies for the management of harvested populations. Analyses of observational data have shown that fish larvae dynamics is shaped by combined effects of environmental gradients, biological interactions and population connectivity. The former includes seasonal and inter annual variability of hydrographic structures that shape ichthyoplankton patterns (Perrier et al., 2012). Biological interactions relates to prey and predator densities, spawning patterns, i.e. time and location (Rakocinski et al., 1996), and food supply (Funes-Rodríguez et al., 2009), whereas population connectivity sculpt the rate of exchange among geographically distant subpopulations determining the relevant spatial scale at which populations are connected (Cowen et al., 2000). In this work we address the first issue. Coral reefs are heterogeneous environments that encompass a variety of shelter areas for fish larvae. These ecosystems offer a
⇑ Corresponding author. E-mail address: [email protected] (J.C. Molinero).
large number of potential resource axes, such as feeding, refuge and breeding, which favor the life history success and shape behavioural patterns (Paris and Cowen, 2004; Gratwicke and Speight, 2005). These systems further support high diversity and are referred as valuable nursery areas for fish and invertebrates. In the northern Persian Gulf, little is known on ecological aspects of fish larvae in the Khark and Kharko Islands, despite their potential importance in recruitment and ultimately in stock management. Yet, fisheries research in this zone has been mainly focused on taxonomy and species richness (Rabbaniha, 1998, 2002; Dehghan et al., 2000), while drivers of larval assemblages and their seasonal changes in habitat use remain elusive. These issues are fundamental to gain understanding on fish ecology, whereas their quantification bears vast implications for a sustainable management of harvested fish in an area that is heavily exposed to anthropogenic stress, i.e. oil-related pollution. Indeed, the marine habitat in the northern Persian Gulf is particularly interesting as marine populations are heavily exposed to both hydroclimate forcing i.e. high levels of temperature and salinity stress, and habitat deteriorating due to the rapid development of the region, as well as the exposure to extensive oil extraction, transportation and refinement (Sheppard et al., 2010). Understanding how fish larvae respond
http://dx.doi.org/10.1016/j.marpolbul.2015.06.028 0025-326X/Ó 2015 Published by Elsevier Ltd.
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
2
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
to such stressors is therefore crucial for their adequate management. Here we examine larval fish assemblages, we assess their habitat associations and environmental factors favoring dominant larvae groups, and aim to provide a synoptic picture of space–time patterns in the Khark and Kharko Islands, northern Persian Gulf.
2. Materials and methods 2.1. Study area and data collection The Khark and Kharko Islands are located in the northern Persian Gulf, Bushehr Province, Iran (29° 120 to 29° 200 N, and 50° 160 to 50° 210 E) (Fig. 1). Khark is a rocky island located about 57 km off the Bushehr Port, while Kharko is a smaler, sandy island surrounded by reefs patches about 60 km north-west off Bushehr. Sampling stations (maximum depth 32 m) covered both coralline and non-coralline sites, as well as the coastal area. Nine stations were located around the islands and three stations in coastal waters (Fig. 1). The sampled area included a variety of habitats that harbour a large diversity of marine life, including fish, molluscs, lobsters and sea birds (Zanjani, 2012). Sampling was carried out monthly, from July 2006 to June 2007, always at the same tidal phase (low tide) in order to account for the potential effect of tidal currents on distributional patterns of larval fish assemblages. At each sampling station, temperature and salinity were recorded in the upper 32 m depth layer, while chlorophyll-a measurements were done using a chlorophyll-a sensor in the upper 30 m depth layer. Larvae samples were taken during daylight using a 61 cm diameter bongo net of 500 lm mesh, with a digital flow meter adapted to the net mouth to measure the water volume filtered. At each station, the bongo net was towed 10 min at a constant speed of 1 knot. Oblique tows were done from near the bottom to the surface. Samples were fixed in 5% formaldehyde seawater immediately after towing and preserved in 10% ethanol. In laboratory, the developmental stage of larvae captured was registered. The scarcity of species-specific taxonomic keys for the region limited a high taxonomic resolution, therefore identification was done at the family level based on Leis and Rennis (1983) and Leis and Trnsky (1989). Adult habitat and spawning sites for the collected fish families were obtained from FishBase (Froese and Pauly, 2015). Fish data were sorted according to habitat types: coralline (5 stations), non-coralline (6 stations) and mouth of creek (1 station) (Fig. 1).
2.2. Data analysis To assess the spatial correlation of environmental conditions (i.e. temperature, salinity and chlorophyll-a) in the explored habitats we used the Mantel statistics (Mantel, 1967). The statistical significance of Mantel correlations was assessed by Monte Carlo permutation (1000 times). Hydrological patterns ascribed to temperature, salinity and chlorophyll-a were assessed by means of box-and-whisker plots to characterize the extent of monthly variations. Fish larval abundances were standardized to number of larvae per 10 m2 and diversity was calculated throughout the year by means of Shannon diversity and evenness index. Due to the non-normal distribution of data and the lack of variance homogeneity even after data transformation, we used the non-parametric Wilcoxon and Kruskal–Wallis tests to assess differences in the fish larvae abundance between hydrological seasons and habitats. Correspondence Analysis (CA) was used to evaluate the relative importance of habitats and seasonal environmental changes in larval assemblages. To compute CA we included only families with more than 2% of relative abundance (number of families used in the analysis = 10). CA was performed both seasonally and on a yearly basis. The ratio of pelagic:demersal fish larvae was calculated as the log (n1/n2), where n1 is the abundance of pelagic spawned larvae and n2 is the abundance of demersal spawned larvae. The index equals zero when the two groups show the same abundance, whereas positive values denote a higher abundance of pelagic spawned larvae and negative values indicate a dominance of demersal spawned larvae. This ratio broadly reflects the overall availability of nutrients in the marine system, and has been used as proxy of hypoxia events resulting from excess primary production and eutrophication, which negatively affect benthic/demersal fish (De Leiva Moreno et al., 2000). Environmental windows (i.e. temperature, salinity) linked to the density of dominant families were assessed by means of a non-parametric test of habitat association, as in Molinero et al. (2009). The association between dominant fish families and environment windows was displayed as traffic light diagrams to qualitatively show the extent of overlapping between the preferred environmental ranges among families. Traffic light diagrams display a colour range from dark (quantile .90) to light grey (quantile .10). Statistical analyses were performed in R (R Development Core Team, 2008) and Matlab.
3. Results and discussion 3.1. Environmental conditions
Fig. 1. Map of the Persian Gulf. The study area is located in the northeastern region. The inset shows the Khark and Kharko Islands and sampling stations.
No statistical differences were found in the environmental conditions relative to the habitats investigated, as revealed by the Mantel test (1000 iterations, p < 0.05), which measured the dissimilarity between the spatial locations, i.e. coralline, non-coralline, mouth of the creek. Instead, significant environmental changes were detected on the seasonal scale (Kruskal Wallis test, p < 0.05) driven by contrasting temperature variations, i.e. 40 °C in summer and 18 °C in winter (Fig. 2a). Salinity showed generally high values (42 ± 1.3; mean and standard deviation respectively), although a sharp decrease arose during the rainy season, in late summer (Fig. 2b). Chlorophyll-a ranged between 0.37 and 0.99 mg m3, with higher concentrations found in summer–autumn 2006 and spring 2007, although a large variability was observed (Fig. 2c). Likewise, conspicuous variability was observed in the monthly larval abundance, global average 7.9 ± 3.55 ind.10 m2 (Fig. 2d).
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
Fig. 2. Hydrographic conditions and larvae abundances during the period July 2006 to June 2007. Box-and-Whisker plots show monthly variations of (a) temperature, (b) salinity, (c) chlorophyll-a and (d) fish larvae at the Khark and Kharko area, northern Persian Gulf. Median (bold horizontal lines), maximum and minimum are indicated by whiskers. The length of the box accounts for the difference between the 25th and the 75th percentiles.
3.2. Habitat and seasonal changes in larval fish assemblages The development stage of collected larvae was mainly pre-flexion (90%), whereas only few were in stage flexion and post-flexion (10%). These results can be therefore used as indicators of the seasonal dynamics of the early stages of fish larvae. Twenty families accounted for the 92% of collected larvae. Clupeidae was the most abundant (29%), followed by Sigillinidae (14%) and Blenniidae (11%) (Table 1). Larvae abundance was on average 18.7 ± 0.90 ind.10 m2, and showed an evenness value of 0.56 ± 0.4. In coralline locations, we found 38 families and the average abundance was 18.7 ± 6.3 ind.10 m2, where Clupeidae accounted for 30% of the total abundance. The non-coralline habitats included 31 families with an average abundance of 14.9 ± 2.1 ind.10 m2. In these habitats, dominant families were Blenniidae and Sillaginidae. The mouth of the creek showed a lower taxonomic richness (16 families) with a mean abundance of 7.4 ± 1.3 ind.10 m2 and the families Sillaginidae and Gobiidae dominated this area (Table 1). The highest concentration of fish
3
larvae was found in the coralline zone, where larvae appeared associated with demersal habitats (Table 1). Typical coral reef fish accounted for 15% of the total abundance. The composition of larval fish assemblages identified by means of CA showed significant differences between hydrological seasons (Wilcoxon test, p < 0.05). However, we did not find unambiguous evidence on abundance changes among habitats (Kruskal Wallis test, p = 0.07). CA showed varied habitat association of larval fish assemblages (Fig. 3a). Regardless the season, the families Clupeidae and Atherinidae appeared linked with coralline sites, while Gobiidae and Sillaginidae were associated to the mouth of the creek stations. In turn, non-coralline habitats were occupied by Mullidae, Muglidae, Terapontidae, Carangidae and Blenniidae. Most of the families are associated to demersal and pelagic habitats, whereas only few were strictly coralline (Table 1). This picture changed between thermal regimes. During the high temperatures (July to October 2006 and April to June 2007; average temperature >30 °C), four families (Clupeidae, Blenniidae, Sillaginidae and Atherinidae) accounted for 67% of the larvae collected, with Clupeidae being the dominant family (39% of total abundance), which reflects the reproductive pattern of the species in the region occurring during spring and early summer (Van Zalling and Owfi, 1993). Associated with non-coralline sites we found the families Clupeidae, Atherinidae and Tripterygiidae, while Carangidae and Blenniidae were closely related to coralline sites. In turn, Gobiidae and Sillaginidae appeared linked with the mouth of the creek area. This period also showed a relatively large proportion of pelagic spawned larvae that reached up to 57% of the total abundance. The mean Shannon index was 0.83 ± 0.59, while the evenness was 0.62 ± 0.37 and the mean abundance was 24 ± 1 ind.10 m2. Overall, during the warmer period, the results of CA showed a similar structure as for the whole period (Fig. 3a). The decrease in temperature during autumn yielded a new hydrological regime that governed the period November 2006 to March 2007, when the average larvae concentration was 10 ± 0.4 ind.10 m2. The four most abundant families (Sillaginidae, Blenniidae, Mullidae and Pomacentridae) encompassed 49% of the total larval caught. Along with the decreasing temperature, the density of pelagic fish larvae dropped to 15% of the total abundance, which contrasted with a noticeable increase of demersal spawned larvae. The mean Shannon index and evenness were 0.55 ± 0.6 and 0.45 ± 0.44 respectively. In turn, larval assemblages showed a different configuration where Sillaginidae and Blenniidae were associated with coralline habitats, while Clupeidae shifted their spatial distribution to the mouth of the creek area, and non-coralline habitats were occupied by the families Sparidae, Engraulididae and Cepolidae (Fig. 3a). In reef areas, demersal spawned larvae are commonly associated with shallow environments, while pelagic spawned larvae showed a wider spatial distribution (López-Sanz et al., 2009). In our study site, we found demersal spawned larvae in dwelling either coralline habitats (i.e. Apogonidae, Belonidae and Clinidae) or dispersed through the three habitats considered (i.e. Blenniidae, Gobiidae and Trypterygiidae). In contrast, larvae from pelagic spawners were generally absent from the creek mouth area, with the exception of Carangidae and Clupeidae. Similar disparate patterns related with the spatial patterns of pelagic and demersal spawned larvae were registered in temperate waters of New Zealand (Hickford and Schiel, 2003), on the Canadian west coast (Marliave, 1986) and Off Nova Scotia (Suthers and Frank, 1991). The observed contrasting environmental configurations in part result from seasonal reproductive patterns of dominant families in the region. In line with this, we also observed conspicuous changes in structure of larval fish assemblages, as indexed by the ratio of pelagic:demersal spawned larvae. Larval assemblages displayed a dominance of pelagic larvae during warmer conditions
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
4
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
Table 1 Check list of families and mean abundance (ind.10 m2) of fish larvae in the northern Persian Gulf.
a b
Family
Coralline
Mouth of creek
Non-coralline
Relative abundance
Spawning
Habitat
Apogonidae Atherinidae Belonidae Blenniidae Bothidae Bregmaceratidae Callionymidae Carangidae Cepolidae Chaetodontidae Clinidae Clupeidae Clupeiform Creediidae Cynoglossidae Engraulididae Exocoetidae Gerreidae Gobiidae Haemulidae Hemiramphidae Labridae Leiognathidae Lethrinidae Lutjanidae Mugilidae Mugiloididae Mullidae Nemipteridae Platycephalidae Pomacentridae Sciaenidae Scomberidae Scorpaenidae Siganidae Sillaginidae Soleidae Sparidae Sphyraenidae Syngnathidae Synodontidae Terapontidae Tetraodontidae Triacanthidae Triglidae Tripterygiidae Other
0.01 ± 0.15 1.33 ± 11.79 0.03 ± 0.56 2.15 ± 6.98 0.28 ± 0.59 0.02 ± 0.42 0.12 ± 0.37 0.50 ± 1.00 0.15 ± 0.42 0.02 ± 0.44 0.01 ± 0.10 5.48 ± 41.33 0.25 ± 2.78 0.01 0.03 ± 0.28 0.27 ± 1.68 0.06 ± 0.50 0.21 ± 1.52 0.70 ± 0.94 0 0.19 ± 1.09 0.01 ± 0.21 0.06 ± 0.06 0.05 ± 0.10 0.03 ± 0.20 0.41 ± 0.48 0.02 ± 0.36 0.62 ± 0.15 0.08 0.07 ± 0.93 0.50 ± 4.26 0.01 ± 0.19 0.37 ± 2.32 0.06 ± 0.72 0.03 2.29 ± 8.61 0.08 ± 0.51 0.28 ± 0.46 0.22 ± 0.65 0.03 0.01 0.49 ± 0.50 0.03 ± 0.47 0.10 0.02 ± 0.16 0.87 ± 7.82 0.15 ± 0.88
0 0 0 0.15 ± 0.30 0.03 ± 0.10 0 0 0.10 ± 0.22 0 0 0 0.31 ± 0.53 0.02 ± 0.05 0 0.02 ± 0.05 0 0 0.10 ± 0.31 2.32 ± 6.16 0.04 ± 0.11 0 0 0 0 0.04 ± 0.11 0 0 0 0 0 0.49 ± 1.48 0 0 0 0 2.42 ± 6.72 0.26 ± 0.33 0.11 ± 0.34 0 0 0 0.15 ± 0.44 0 0.14 ± 0.33 0 0.69 ± 2.08 0
0 0.28 ± 1.08 0 2.58 ± 11.65 0.46 ± 1.91 0.00 0.17 ± 0.66 0.73 ± 2.14 0.22 ± 0.86 0 0 1.67 ± 3.89 0.07 ± 0.33 0.02 ± 0.19 0 0.30 ± 1.26 0.05 ± 0.41 0.07 ± 0.42 0.82 ± 3.10 0.00 0.10 ± 0.56 0 0.11 ± 0.55 0.10 ± 0.52 0.02 ± 0.13 0.76 ± 2.81 0 1.21 ± 5.13 0.16 ± 0.77 0.02 ± 0.17 0.41 ± 3.00 0 0.36 ± 1.41 0 0.05 ± 0.44 1.98 ± 7.86 0.00 0.46 ± 1.90 0.32 ± 0.76 0.06 ± 0.46 0.03 ± 0.23 0.82 ± 3.41 0 0.17 ± 1.39 0.02 ± 0.16 0.24 ± 1.49 0.08 ± 0.47
0.06 4.72 0.18 11.46 1.62 0.13 0.67 2.83 0.86 0.14 0.03 28.63 1.41 0.07 0.11 1.53 0.35 0.83 4.27 0.02 0.98 0.07 0.31 0.35 0.15 2.36 0.12 3.54 0.46 0.39 2.86 0.06 2.13 0.33 0.15 13.00 0.45 1.61 1.24 0.16 0.08 2.79 0.17 0.55 0.11 4.85 0.82
Demersal Demersal Demersal Demersal Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Demersal Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Demersal Pelagic –a Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic Pelagic –b Pelagic Pelagic Pelagic Pelagic Pelagic Demersal –
Coral reefs Pelagic Pelagic Demersal Demersal Pelagic Demersal Pelagic Demersal Coral reefs Demersal Pelagic Pelagic Demersal Demersal Pelagic Pelagic Demersal Demersal Coral reefs Pelagic Coral reefs Demersal Demersal Demersal Demersal Demersal Demersal Demersal Demersal Coral reefs Demersal Pelagic Coral reefs Demersal Demersal Demersal Demersal Pelagic Demersal Demersal Demersal Demersal Demersal Demersal Demersal –
Male carry eggs. Eggs attached to seaweed and substrata.
and lower chlorophyll-a concentration. However, under lower temperatures and higher chlorophyll-a concentration, i.e. in late winter, demersal spawned larvae were more abundant (Fig. 3b). Previous studies have shown that the pelagic:demersal ratio broadly reflects the overall availability of nutrients in the marine system, that is the case for the Mediterranean Sea where the ratio appears correlated with the mean chlorophyll-a value, suggesting a better availability of nutrients (De Leiva Moreno et al., 2000). Generally, pelagic spawners are favored by nutrient enrichment, as it stimulates the plankton production (Caddy, 1993), while demersal ones reflect the dynamics of benthic communities, which generally responds negatively to the conditions of excessive enrichment, i.e. eutrophication (De Leiva Moreno et al., 2000). In agreement with this, in our study the ratio increased along with water temperature and chlorophyll-a concentration; this likely fostered the reproduction of pelagic spawner fish, such as Clupeidae, the numerically dominant fish family. Indeed, previous studies have reported a recurrent major peak of reproduction for this family during the same period (Van Zalling and Owfi, 1993). In addition, the Khark Island is largely exposed to oil pollution, as
suggested by the high concentrations of petroleum and polycyclic aromatic hydrocarbons in sediments (Khoei et al., 2013; Mirvakili et al., 2013). Polycyclic aromatic hydrocarbons are degraded by photo- and biodegradation processes and can permeate the food web reaching higher trophic levels. This pollution has likely a larger impact on suprabenthic communities and fauna associated to them, such as demersal fish, affecting their reproductive output. This hypothesis however should be confirmed with the appropriate data. 3.3. Environmental windows depicting the habitat use of dominant families The environmental windows relative to the abundance of the three dominant families were displayed by traffic light diagrams. The results highlighted similar thermal and salinity ranges, although the thermal window of Clupeidae was narrower than that showed by Sillaginidae and Blenniidae. The preferred depth range varied between families, with Clupeidae showing a shallower distribution (higher percentage found within the 9–12 m depth
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
5
Fig. 3. (a) Fish larvae abundance along the first two ordination axes of the correspondence analysis based on the habitat recurrence (coralline, non-coralline and mouth of creek). Results show for the whole period (top panel), as well as for warm (middle panel) and cold (low panel) seasons. The percentage of the total inertia contained in the two first dimensions was, respectively, 60.3% and 39.7% for the whole period; 67.8% and 32.2% for the warm season, and 67.8% and 32.18% for the cold season. Habitats are denoted by h1 (non-coralline), h2 (coralline) and h3 (mouth of the creek). (b) Response surface depicting the link between the ratio pelagic:demersal fish spawned larvae with temperature and chlorophyll-a gradients. (c) Habitat association of fish larvae and environmental factors. Monte Carlo simulation results show the preferred windows of the three dominant families during the survey period 2006–2007. Colours denote the strength of the association from black (intense association, p < 0.05) to grey light (no association). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
range), followed by Sillaginidae (10–18 m depth range) and Blenniidae (17–26 m depth range) (Fig. 3c). Temperature changes showed a leading role in the space–time pattern of larval fish assemblages, although some peaks of larval abundance matched chlorophyll-a peaks, which might allow species maximizing larval survival (Cushing, 1990). Similar observations were reported in the western Mediterranean Sea where the spawning period of neritic fish species was coupled with the maximum primary production (López-Sanz et al., 2009). Likewise, enhanced plankton production and maximum larval abundance have been observed in the subtropical western Pacific coasts (Franco-Gordo et al., 2008) and in East China waters (Chen et al., 2014). Further factors are however acknowledged to influence space–time patterns of fish larvae; for instance, upwelling processes (Hernandez-Miranda et al., 2003), spawning patterns of adult fishes (Sampey et al., 2004), and mesoscale currents (Pineda, 2000). In the northern Persian Gulf the main mesoscale hydrographic feature is a cyclonic overturning circulation, the Iranian Coastal Jet (ICJ) that enters through the Strait of Hormuz and establishes along the full length of the Gulf. The strength of the jet current develops in late spring-early summer and is followed by a relaxation in autumn and winter, when it becomes unstable and a clockwise circulation pattern appears in surface
layers in the northern area (Kampf and Sadrinasab, 2006). These variations in the seasonal circulation matched the timing of changes in the observed abundance pattern of Clupeidae, likely due to their larger exposure to dispersion as they mainly occupied shallow depths. Under warm conditions the relative abundance of the pelagic group reached 57% of the total assemblage, while in the cold season coastal and demersal spawned larvae were dominant (>50%). These observations stress the role of mesoscale hydrographic processes on the seasonal structure of larval fish assemblages. 3.4. Khark and Kharko Islands The marine environment of the northern Persian Gulf has long been exposed to oil pollution resulting from extensive extraction, transportation and refinement. In addition, the pace of change in resource damage warns on potential ecosystem-wide changes that may overwhelm the resilience of the system (Sheppard et al., 2010). Hence, there is a pressing need for research efforts pointing out the role marine protected areas, such as the Khark and Kharko Islands, might play for resources management, and the integration of such efforts into a large scale, long term perspective to achieve adequate policies. Here we have shown that the Khark and
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
6
M. Rabbaniha et al. / Marine Pollution Bulletin xxx (2015) xxx–xxx
Kharko Islands offer varied habitats that promote high densities of larval fish assemblages. This might result from the habitat structure and complexity that minimise competition by increasing habitat partitioning (Raberg and Kautsky, 2007), which in turn explains the high species richness found, as well as the high proportion of larvae in coralline habitats (40% of the total abundance). By promoting varied habitats and high species richness this zone may act as spawning centres in the area, as suggested by the large proportion of pre-flexion larval stages. Our results further pointed out that the dynamics of mesoscale currents influence the scale of connectivity of dominant families, and therefore management strategies in the Khark and Kharko Islands should be integrated into a broad spatial scale. These results provide a general picture of seasonal changes in the habitat association of larval fish assemblages and could be used as baseline for further fisheries and conservation studies in the northern Persian Gulf. Acknowledgements We are grateful to colleagues of Shrimp Research Center of Bushehr for their cooperative in cruise sampling and laboratory surveys. This work was funded by the German Research Foundation (DFG) through a visiting research fellowship to MR. References Caddy, J.F., 1993. Toward a comparative evaluation of human impacts on fishery ecosystems of enclosed and semi-enclosed seas. Rev. Fish. Sci. 1, 57–95. Chen, W.Y., Lee, M.A., Lan, K.W., Gong, G.C., 2014. Distributions and assemblages of larval fish in the East China Sea during the northeasterly and southwesterly monsoon seasons of 2008. Biogeosciences 11, 547–561. Cowen, R.K., Lwiza, K., Spounaugle, S., Paris, C., Olson, D.B., 2000. Connectivity of Marine populations: open or closed. Science 287, 857–859. Cushing, D., 1990. Plankton production and year-class strength in fish populations: an update of the match/mismatch hypothesis. Adv. Mar. Biol. 26, 249–293. De Leiva Moreno, J.I., Agostini, V.N., Caddy, J.F., Carocci, F., 2000. Is the pelagicdemersal ratio from fishery landings a useful proxy for nutrient availability? A preliminary data exploration for the semi-enclosed seas around Europe. ICES J. Mar. Sci. 57, 1091–1102. Dehghan, M.S., Savari, A., Kochnin, P., Maramazi, J.G., 2000. Abundance, diversity and distribution of fish larvae in the estuary and west coastal of Khouzestan. Fish. Sci. Bull. 4, 20–34. Franco-Gordo, C., Godínez-Domínguez, E., Freire, J., 2008. Interannual variability of the diversity and structure of ichthyoplankton assemblages in the central Mexican Pacific. Fish. Oceanogr. 17, 178–190. Froese, R., Pauly, D., (Eds.), 2015. FishBase. World Wide Web electronic publication. , version (04/2015). (accessed June 2015). Funes-Rodríguez, R., Elorduy-Garay, J.F., Hinojosa-Medina, A., Zárate-Villafranco, A., 2009. Interannual distribution of Pacific hake Merluccius productus larvae in the southern part of the California current. J. Fish. Biol. 75, 630–646. Gratwicke, B., Speight, M.R., 2005. The relationship between fish species richness, abundance and habitat complexity in a range of shallow tropical marine habitats. J. Fish. Biol. 66, 650–667.
Hernandez-Miranda, E., Palma, A., Ojeda, F., 2003. Larval fish assemblages in nearshore coastal waters off central Chile: temporal and spatial patterns. Estuar. Coast. Shelf. Sci. 56, 1–18. Hickford, M.J.H., Schiel, D.R., 2003. Comparative dispersal of larvae from demersal versus pelagic spawning fishes. Mar. Ecol. Prog. Ser. 252, 255–271. Kampf, J., Sadrinasab, M., 2006. The circulation of the Persian Gulf: a numerical model. Ocean Sci. 2, 27–41. Khoei, J.K., Bastami, A.A., Farmohammadi, S., 2013. Oil related PAH in fish and sediments in the Persian Gulf, Northern Part. World Appl. Sci. J. 24, 1378–1381. Leis, J.M., Rennis, D.S., 1983. The Larvae of Indo Pacific Coral Reef Fishes. New South Wales University Press, Sydney, p. 269. Leis, J.M., Trnsky, T., 1989. The Larvae of Indo Pacific Shore Fishes. New south Wales University Press, Sydney, p. 371. López-Sanz, Á., Vert, N., Zabala, M., Sabatés, A., 2009. Small-scale distribution of fish larvae around the Medes Islands marine protected area (NW Mediterranean). J. Plankton Res. 31, 763–775. Mantel, N., 1967. The detection of disease clustering and a generalized regression approach. Cancer Res. 27, 209–220. Marliave, J.B., 1986. Lack of planktonic dispersal of rocky intertidal fish larvae. Trans. Am. Fish Soc. 115, 149–154. Mirvakili, H.S., Zaker, N.H., Imani, F., 2013. Evaluation of oil pollution and origin in surface coastal sediments of Kharg Island in the Persian Gulf. J. Coastal Res. 65, 93–98. Molinero, J.C., Vukanic, V., Lucˇicˇ, D., Licandro, P., Souissi, S., Ibanez, F., Nival, P., Calbet, A., Christou, E.D., Daly-Yahia, N., Fernandez de Puelles, M.L., Mazzocchi, M.G., Siokou-Frangou, I., Vukanic, D., 2009. Mediterranean marine copepods. Basin scale trends of the calanoid Centropages typicus. Hydrobiologia 617, 41– 53. Paris, C.B., Cowen, R.K., 2004. Direct evidence of a biophysical retention mechanism for coral reef fish larvae. Limnol. Oceanogr. 49, 1964–1979. Perrier, C., Molinero, J.C., Gerdeaux, D., Anneville, O., 2012. Effects of temperature and food supply on the growth of whitefish Coregonus lavaretus larvae in an oligotrophic peri-alpine lake. J. Fish Biol. 81, 1501–1513. Pineda, J., 2000. Linking larval settlement to larval transport: assumptions, potentials, and pitfalls. Oceanogr. East. Pacific I, 84–105. R Development Core Team, 2008. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, ISBN 3900051-07-0, URL http://www.R-project.org. Rabbaniha, M., 1998. Study on Diversity and Abundance of Ichthyoplankton in the Nayband Bay (Bushehr Province). Iranian Fish. Res. Org. Press, p. 101. Rabbaniha, M., 2002. Study on Diversity and Abundance of Ichthyoplankton in the Creek-Estuary Bushehr to Genaveh. MSc. thesis. Iranian Fish. Res. Org. Press, p. 99. Raberg, S., Kautsky, L., 2007. Consumers affect prey biomass and diversity through resource partitioning. Ecology 88, 2468–2473. Rakocinski, C.F., Lyczkowski-Shultz, J., Richardson, S.L., 1996. Ichthyoplankton assemblage structure in Mississippi sound as revealed by canonical correspondence analysis. Estuar. Coast. Shelf Sci. 43, 237–257. Sampey, A., Meekan, M.G., Carleton, J.H., Mckinnon, A.D., McCormick, M.I., 2004. Temporal patterns in distribution of tropical fish larvae on the North-west Shelf of Australia. Mar. Freshwater Res. 55, 473–487. Sheppard, Ch., Al-Husiani, M., Al-Jamali, F., Al-Yamani, F., Baldwin, R., et al., 2010. The Gulf: a young sea in decline. Mar. Pollut. Bull. 60, 13–38. Suthers, I.M., Frank, K.T., 1991. Comparative persistence of marine fish larvae from pelagic versus demersal eggs off southwestern Nova Scotia, Canada. Mar. Biol. 108, 175–184. Van Zalling, N.P., Owfi, F., 1993. Resources of Small Pelagic in Southern Iranian Waters Bushehr. Iranian Fish. Res. Org. Press, Iran, p. 20. Zanjani, M.M., 2012. Ecotourism, expansion, alongside with dominant function of Khark and Kharko Islands. Int. J. Soc. Hum. Sci. 6, 478–481.
Please cite this article in press as: Rabbaniha, M., et al. Habitat association of larval fish assemblages in the northern Persian Gulf. Mar. Pollut. Bull. (2015), http://dx.doi.org/10.1016/j.marpolbul.2015.06.028
