Habitat selection and spatial segregation in three pipefish species

Estuarine, Coastal and Shelf Science 75 (2007) 143e150 www.elsevier.com/locate/ecss

Habitat selection and spatial segregation in three pipefish species Stefano Malavasi*, Anita Franco, Federico Riccato, Chiara Valerio, Patrizia Torricelli, Piero Franzoi Department of Environmental Sciences, University Ca’ Foscari di Venezia, Campo della Celestia, Castello 2737/b, 30121 Venice, Italy Received 15 October 2006; accepted 15 February 2007 Available online 25 June 2007

Abstract Habitat partitioning was investigated within a guild composed of three sympatric pipefish species, namely Syngnathus typhle, Syngnathus abaster and Nerophis ophidion. Field surveys of patterns of pipefish abundance among different seagrass habitats (each dominated by a different seagrass species) were combined with laboratory studies on habitat choice and microhabitat use in the three species. Results showed that S. typhle and N. ophidion occurred with higher abundance in the Cymodocea nodosa meadow, which is characterised by longer leaves and intermediate shoot density compared to the Zostera marina and Nanozostera noltii habitats. By contrast, S. abaster showed higher abundance in the Z. marina meadow than in the other meadows. Males of N. ophidion also showed significant habitat choice in behavioural tests, preferring long over short seagrass leaves, whereas in the other species habitat choice experiments did not show any significant results. In terms of microhabitat use, the three species tended to segregate along the vertical axis, with S. abaster spending significantly more time near the bottom, while N. ophidion and S. typhle preferred to use the intermediate and the top portion of the artificial seagrass. Results are discussed in the light of the current knowledge on habitat partitioning within fish guilds, especially in relation to vegetated aquatic systems. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: pipefish; seagrasses; habitat partitioning; microhabitat

1. Introduction Habitat selection in fish species is influenced by the degree of habitat structural complexity, the level of interspecific competition and the perceived risk of predation, as shown by several field and laboratory investigations (Werner and Hall, 1977; Savino and Stein, 1989; Utne et al., 1993; Jordan et al., 1996; Munday et al., 2001; Schofield, 2003). Aquatic vegetated systems, such as seagrass beds, represent an example of habitats characterised by a high degree of both structural complexity and spatial variability, thus offering the opportunity for sympatric fish species to partition their habitat (Curtis and Vincent, 2005). Within these systems, sympatric pipefish belonging to the family Syngnathidae are an excellent example of a taxonomic and an ecological ‘‘guild’’, showing high

* Corresponding author. E-mail address: [email protected] (S. Malavasi). 0272-7714/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.ecss.2007.02.022

levels of habitat partitioning both among and within seagrass habitats (Howard and Koehn, 1985; Kendrick and Hyndes, 2003; Curtis and Vincent, 2005). Syngnathids are cryptic fish, which occupy either the seagrass canopy or reside at the sediment-water interface, with different species often co-existing on the same seagrass beds (Howard and Koehn, 1985; Teixeira and Musick, 1995; Vincent et al., 1995; Kendrick and Hyndes, 2003; Curtis and Vincent, 2005). In the Venice lagoon, three sympatric pipefish species are abundant on the seagrass beds (Riccato et al., 2003; Malavasi et al., 2004; Franco et al., 2006a): Syngnathus typhle, Syngnathus abaster and Nerophis ophidion. Although information on life history, reproductive behaviour and trophic ecology of these three species are relatively abundant in the literature (Berglund et al., 1988; Tomasini et al., 1991; Franzoi et al., 1993; Campolmi et al., 1996; Riccato et al., 2003; Franzoi et al., 2004), little is known about the distribution of these species in different seagrass habitats and the mechanisms of habitat choice and partitioning. As reviewed by Tokeshi

144

S. Malavasi et al. / Estuarine, Coastal and Shelf Science 75 (2007) 143e150

(1999), habitat represents one of the three main ecological axes, along which resource partitioning among closely related species should be analysed. According with this author, habitat axis should be analysed at different scales, from the largest scale of landscape to the smallest scale of microhabitat. In this paper, habitat partitioning among these three pipefish species was investigated in the Venice lagoon at two different scales: (1) by comparing abundance of pipefish species in three different seagrass meadows within the Northern basin of the Venice lagoon; (2) by testing for habitat preference and microhabitat use in laboratory experiments. To achieve this goal, we combined a field survey conducted during Summer and Autumn 2005 in the Northern basin of the lagoon, with laboratory experiments carried out in an aquarium under controlled conditions. Abundance of the three species were obtained and compared across seagrass habitats dominated by different seagrass species, Zostera marina, Nanozostera noltii Honermann and Cymodocea nodosa, after controlling for the differences among these habitats in terms of structural complexity (leaf height, leaf density and shoot density). Habitat preference was investigated by testing for fish choice between artificial seagrasses characterised by different structural complexity (such as leaf length and density) and by comparing the microhabitat utilisation in terms of the relative position held along the leaf by each species. 2. Materials and methods 2.1. Study species The three pipefish species investigated represent a large proportion of the total fish abundance in the shallow water seagrass habitats of the Venice lagoon (Franco et al., 2006a). They constitute both an ecological guild (sensu Root, 1967), as they feed on small crustaceans and share the seagrass habitats (Riccato et al., 2003; Franzoi et al., 2004), and a taxonomic guild, according with Simberloff and Dayan (1991). Syngnathus typhle (broad-nose pipefish) is a medium-size, green-brown, pipefish, with a maximum length of about 35 cm (Muus and Nielsen, 1999). S. typhle reaches sexual maturity at about 11 cm of total length, brooding lasts from 4 to 6 weeks depending on temperature (Vincent et al., 1995), and it is a sex role reversed species with males limiting female reproductive rate (Berglund et al., 1988). Syngnathus abaster (black-striped pipefish) is the smallest Mediterranean pipefish, reaching a maximum of 17 cm of total length. The sexual maturity is reached after 3e4 months with a total length of about 6 cm (Tomasini et al., 1991) and the incubation period lasts about fifteen days at a temperature of 20e21  C. The body of both species is typically slender and elongated, with males carrying eggs under their tails in a well-defined brood pouch. Nerophis ophidion (straight-nose pipefish), is a small and slender green-brown pipefish with a vermiform shape characterised by a strong sexual dimorphism with females bigger than males. It is a very cryptic species, with a weakly prehensile tail, holding the fish aligned with the seagrass. The species

is sex-role-reversed: the potential reproductive rate of females exceeds that of males, with females actively competing for access to males (Berglund et al., 1989; Rosenquist, 1990). Males bring eggs glued onto the ventral surface of their trunk, and brooding takes about 4 to 6 weeks depending on water temperature (Berglund et al., 1986). 2.2. Study area and seagrass habitats The Venice lagoon is located in the Northern Adriatic Sea and is the largest wetland coastal area of the Mediterranean basin. The lagoon is about 50 km long and 10 km wide with a longitudinal axis in the north-south direction. The total surface area is about 550 km2 with an average depth of about 0.5 m. The lagoon communicates with the sea through three inlets, Lido, Malamocco and Chioggia and an underlying network of navigation channels connects these inlets with the inner shallow areas. Tides are mainly semidiurnal and the mean tidal range is about 55 cm with a spring tide of 110 cm (Cappucci et al., 2004). The field surveys were conducted in three different stations in the northern part of the Venice lagoon (A, B and C, Fig. 1). In each station, some environmental variables were measured to obtain a more detailed description of the seagrass habitats investigated. In particular, water temperature and salinity were measured before sampling using a digital thermometer (  0.1  C) and a temperature refractometer (  1), respectively. Furthermore, water depth and seagrass coverage data were collected. Seagrass coverage was assessed on the basis of five classes, as follows: 0 ¼ 0% percentage coverage; 1 ¼ seagrass percentage from 0% to 5%; 2 ¼ from 5% to 50%; 3 ¼ from 50% to 75%; 4 ¼ from 75% to 100%. Site A (45 250 1500 N, 12 20‘4400 E). This site is characterised by a single uniform patch of Zostera marina, the meadow is about 300 m long and 40 m narrow. Water depth during sampling ranged between 1 and 1.2 m. Water temperature ranged from 25.2 by summer to 20.1  C by autumn, and salinity varied from 35 to 36. Seagrass coverage was ranked as 4 both in summer and autumn. Site B (45 260 1800 N, 12 230 1800 E). In this station Cymodocea nodosa forms an homogeneous bed on gentle sloped coarse sandy bottoms. The meadow is about 500 m long and 35 m narrow, water depth during sampling ranged between 0.80 and 1.2 m. Water temperature ranged from 24.9 by summer to 20.2  C by autumn, whereas salinity was between 35 and 36. Seagrass coverage was 4 both in summer and autumn. Site C (45 260 3300 N, 12 230 4600 E). This site is characterised by a little patch (40  40 m) of Nanozostera noltii surrounded by a mixed meadow of Zostera marina, Cymodocea nodosa and N. noltii. Water depth during sampling ranged between 0.5 and 0.9 m. Temperature varied from 24.8 in summer to 20.4  C in autumn, and salinity ranged from 34 to 36. Seagrass coverage was about 4 both in summer and autumn. As shown in Fig. 1, all the sampling stations are located within a radius of approximately 5 km. This means that the three meadows are geographically close, sharing larger scale influences such as tidal factors, water currents and climatic conditions (Jackson et al., 2002).

S. Malavasi et al. / Estuarine, Coastal and Shelf Science 75 (2007) 143e150

145

Fig. 1. Map showing the study area with the three stations characterised by three seagrass meadows dominated by different seagrass species: (A) Zostera marina, (B) Cymodocea nodosa, (C) Nanozostera noltii. Light grey ¼ land, dark grey ¼ seagrass meadows, white ¼ water (MAV-CVN, 2005).

2.3. Structural characteristics of seagrass habitats Data related to structural characteristics of the seagrasses dominating the three stations were collected to compare structural complexity across the three habitats. As regards habitat structure, shoot density (shoots m2) and leaf density (leafs m2) were obtained, moreover leaf length was measured. Due to the high water turbidity in situ measurements were not possible, so in order to get accurate measurements on structural parameters and not to destroy a significant portion of the canopy, sampling quadrats of different size for each species were used. Thus 40  40 cm quadrats were used for Cymodocea nodosa and Zostera marina and quadrats of 20  20 cm for Nanozostera noltii as they are considered sufficient to get significant data as indicated by Buia et al. (2003). Plants were collected in three replicated sub-samples for every sampling station during each survey and then carried to the laboratory, where they were preserved at 20  C for further analysis. Shoots and leaves were then counted and leaf length was measured using a meter ( 0.1 cm). 2.4. Pipefish abundance The fish sampling was carried out on a seasonal basis during summer (July) and autumn (September) of 2005. Fishes were caught using a 2 m high and 10 m long beach seine (mesh size: 2 mm) with an extra heavy footrope to increase efficiency in seagrass (Jenkins et al., 1997), which was dragged over an area of 360 m2 (3 replicate drags, each one covering an

area of 120 m2) at each station. Among the different sampling gear the beach seine has been considered less selective for fish species than many other systems (Nagelkerken et al., 2001; Nagelkerken and Van Der Velde, 2004). Furthermore, previous studies on the fish communities of the shallows of the Venice lagoon (Franco et al., 2006a,b) indicated that the beach seine is an efficient and successful method to catch pipefish in the vegetated habitats. The three investigated pipefish species were identified and the number of individuals was counted in situ and then released. Fish density was obtained for each species in terms of number of individuals 100 m2. 2.5. Habitat preference and microhabitat use Fishes used in the laboratory experiments were captured during May 2005 with a small beach seine, transported to the laboratory and maintained in 400 L holding tanks, provided with filtered salt water. Salinity was kept between 28 and 32 and temperature varied between 18 and 22  C; the photoperiod followed natural conditions. Experiments were conducted during the day using glass tanks (capacity: 400 L) with the same characteristics of the holding tanks; the bottoms were covered with approximately 20 cm of sand, and artificial seagrasses were provided. According to previous studies, artificial seagrass is a good mimic of natural seagrasses (Schofield, 2003). Artificial seagrasses consisted of green polyester blades (5 mm width) anchored to a plastic net, which was placed into the sand. Three sets of habitat preference experiments were conducted: (1) artificial seagrass with 30 cm

146

S. Malavasi et al. / Estuarine, Coastal and Shelf Science 75 (2007) 143e150

high blades and a density of 1000 blades m2 was placed on one side of the tank to provide fish a choice between sand and seagrass; (2) choice between seagrass of different length but equal density was provided: half of the bottom received seagrass with 30 cm high blades, while the other half was provided with a seagrass with 10 cm high blades. Density was, for both seagrasses, about 1000 blades m2. (3) Choice between seagrass with equal length but different density: in this case, blades were 30 cm high but half of the bottom received a seagrass with a density of 1000 blades m2 and the other half a seagrass with a density of 3000 blades m2. These values were chosen to test for the differential abilities of the three species to discriminate between degrees of structural habitat complexity, and not to simulate punctually the variability observed in the field among the three seagrass species. Nevertheless, these values are comprised within the natural range of variation of these parameters for Mediterranean seagrasses (Buia et al., 2003). For each experimental set, 10 individuals of each species, 5 males and 5 females, were singly tested. Before each trial, the experimental fish was transferred from the holding tank to the experimental aquarium and the relative position of the fish was then recorded for 60 min. A total of 10 trials (i.e. 10 individuals) for each species were therefore run. Preliminary trials showed that a short acclimation period (about 10 min) was enough for these animals. Analysis of the videotapes allowed the percent time spent by each fish in each type of habitat to be obtained. For each experiment, new individuals were used, as they were taken from the holding tanks and then transferred to a different tank after each trial, so that no individual was used more than once in the three experiments. Microhabitat preference was assessed within those trials where seagrass with 30 cm high blades were present on both half of the bottom: this allowed the percent time spent by each fish in terms of vertical habitat occupation to be obtained, considering the vertical component of the habitat from the bottom to the apex of the leaf. The aquarium was divided into three vertical portions: (1) bottom, that is when a fish was on the bottom or within the 5 cm of the bottom, (2) intermediate portion of seagrass; that is when the fish was comprised between 5 cm from the bottom up to 10 cm of height, (3) top portion of seagrass, when a fish was between 10 and 15 cm, assuming that fish do not use the upper half of the blade. 2.6. Data analysis The field data (fish density, leaf length, shoot density and leaf density) were analysed by means of two-way ANOVA to control for possible effects both of habitat type (seagrasses) and seasonal effects (summer vs. autumn). Data were log10(X þ 1) transformed to meet the assumptions of homogeneity of variances. In many cases transformation did not produce homogeneous variances, but ANOVA was used nevertheless, as the F statistic is considered robust in relation to the assumption of heterogeneity. However, in order to compensate for the increased likelihood of type 1 error, a setting of

alpha ¼ 0.01 was used (Underwood, 1997). Tukey’s test was used for post hoc comparisons after ANOVA. The behavioural data were analysed in terms of percentage of time spent in a given habitat by a given species or gender. These data did not meet the assumptions of normality distribution and homogeneity of variances, and therefore nonparametric statistic was used. Differences in time proportion spent by a given species in either of two habitats were tested with the Wilcoxon test (Siegel and Castellan, 1992). To compare the percent time spent in a given habitat across species, the KruskalleWallis ANOVA was used (Siegel and Castellan, 1992). 3. Results 3.1. Structural and phenological characteristics of seagrass habitats Regarding the mean leaf length, no interaction was detected by the two-way ANOVA, while only the effect of station was statistically significant (Table 1, Fig. 2a). The stations B and C showed the highest and the lowest mean values respectively (Fig. 2a). As regards the structural characteristics of seagrass habitats, such as shoot density and leaf density, a significant interaction between station and season was revealed by the two-way ANOVA (see Table 1), hence differences across stations and between seasons were analysed in detail by means of one-way ANOVA. For both parameters, statistically significant differences across stations were observed only in summer (shoot density: df ¼ 2, F ¼ 52.67 P < 0.001; leaf density: df ¼ 2, F ¼ 82.77, P < 0.001; Fig. 2b and c). Station C showed the highest mean values in terms of shoot density, while the station A had the lowest mean values, with station B being intermediate (Tukey test, P < 0.01). The mean number of leaves was significantly higher in station C than in the other two stations (Tukey test, P < 0.05). Table 1 ANOVA results for structural and phenological characteristics of seagrass habitats: (a) leaf length, (b) shoot density, (c) leaf density. NS, not significant SS (a) Leaf length Season Station Interaction Error

0.82 1058.21 61.37 170.97

df 1 2 2 12

MS 0.82 529.11 30.68 14.25

F

P

0.06 37.14 2.15

NS <0.001 NS

(b) Shoot density Season 37345.3 Station 37058.9 Interaction 39069.5 Error 5854.9

1 2 2 12

37345.3 18529.4 19534.8 487.9

76.54 37.98 40.04

<0.001 <0.001 <0.001

(c) Leaf density Season 853982.4 Station 831796.6 Interaction 949806.0 Error 146922.9

1 2 2 12

853982.4 415898.3 474903.0 12243.6

69.75 33.97 38.79

<0.001 <0.001 <0.001

S. Malavasi et al. / Estuarine, Coastal and Shelf Science 75 (2007) 143e150 40

a

35

Leaf length (cm)

30 25 20 15 10 5 0

2400 2200

B

A

C

b

2000

shoot m-2

1600 1400 1200 1000 800 600 400 200 0

B

A

C

B

A

C

c

10000

leaves m-2

8000

6000

4000

2000

0

Syngnathus abaster with 48 ind. 100 m2 (mean in summer, Fig. 3b), whereas Nerophis ophidion was the rarest species with less than 27 individuals per 100 m2 (by summer, Fig. 3c). The abundance of Syngnathus abaster was significantly affected by both the season and the station (see ANOVA results, Table 2), this species being more abundant during summer and in station A (Table 2, Fig. 3). As regards Syngnathus typhle, there was a stronger effect of the station with respect to the season (Table 2), with this species being significantly more abundant in station B than in the other two stations (Tukey test, P < 0.05). A similar spatial pattern was observed for Nerophis ophidion, which was significantly more abundant in station B than in the two other stations (Table 2, Fig. 3), although for this species a significant effect of the season was also detected by ANOVA (Table 2), with the abundance being higher during summer than in autumn. 3.3. Habitat preference and microhabitat use

1800

12000

147

Fig. 2. Mean  S.E. of phenological and structural characteristics of the three investigated stations by Summer (white bars) and Autumn (black bars). B, Cymodocea nodosa; A, Zostera marina; C, Nanozostera noltii. (a) Leaf length; (b) shoot density; (c) leaf density.

3.2. Patterns of pipefish abundance in seagrass habitats All the three investigated pipefishes were found in the different seagrass habitats although with a variable total abundance. The overall most abundant species were Syngnathus typhle (about 50 ind. 100 m2 on average in summer, Fig. 3a), and

Tested pipefish preferred seagrass over sand habitat, spending more than 90% of their time in seagrasses (Wilcoxon test Z ¼ 28 P < 0.01; Table 3). As regards habitat preference based on leaf density, no habitat selection was observed by any of the three species, given the percent time spent by each species did not statistically differ between the highleaf-density and the low-leaf-density artificial habitats (Wilcoxon test, P > 0.05; Table 3). By contrast, as regards habitat preference based on leaf length, males of Nerophis ophidion showed habitat preference, by spending a statistically significant higher proportion of time in the ‘‘long’’ (long leaves) than in the ‘‘short’’ (short leaves) artificial seagrasses (Wilcoxon test, n ¼ 5, Z ¼ 2.02, P < 0.04; Table 3). In terms of microhabitat use, the three species differentially occupied the three portions of the artificial habitat, with N. ophidion and Syngnathus abaster spending around 80% of time in the low portion of the seagrass and on the bottom respectively (KruskalleWallis H ¼ 18, n ¼ 10, P < 0.01 for time spent in the intermediate portion, H ¼ 20, n ¼ 10, P < 0.01 for time spent on the bottom, Fig. 4), whereas Syngnathus typhle spent less time in the low portion of the seagrass than in the other two portions (KruskalleWallis H ¼ 6.5, n ¼ 10, P < 0.01, Fig. 4). Furthermore, our behavioural observations revealed that while S. abaster held a horizontal posture, the other two species maintained a vertical posture for most of the time, aligning their body with the artificial leaves. 4. Discussion Results of this study suggest that a certain degree of habitat partitioning occurs within the guild composed of the three investigated pipefish species. Our laboratory investigations indicate that all three species are strictly associated to seagrasses. Given the general adaptation of this fish guild to seagrass habitats, some evidence of partitioning within these habitats are shown by the present study both at the level of macrohabitat and microhabitat scales.

S. Malavasi et al. / Estuarine, Coastal and Shelf Science 75 (2007) 143e150

148

S. abaster - individuals 100 m-2

60

Table 2 ANOVA results for pipefish abundance in different seagrass habitats: (a) Syngnathus abaster, (b) Syngnathus typhle, (c) Nerophis ophidion. NS, not significant

a

50

40

30

10

B

A

C

70

b S. typhle - individuals 100 m-2

df

MS

F

P

(a) S. abaster Season Station Interaction Error

0.91 2.10 0.08 0.78

1 2 2 12

0.91 1.05 0.04 0.07

14.01 16.12 0.58

<0.01 <0.001 NS

(b) S. typhle Season Station Interaction Error

0.53 1.32 0.22 0.79

1 2 2 12

0.53 0.66 0.11 0.07

7.96 9.95 1.63

<0.05 <0.01 NS

(c) N. ophidion Season Station Interaction Error

1.10 1.98 0.47 0.84

1 2 2 12

1.10 0.99 0.24 0.07

15.72 14.15 3.36

<0.01 <0.001 NS

20

0

60 50 40 30 20 10 0

40

N. ophidion - individuals 100 m-2

SS

B

A

C

B

A

C

c

35 30 25 20 15 10 5 0

Fig. 3. Mean  S.E. of pipefish abundance in the three investigated seagrass habitats by Summer (white bars) and Autumn (black bars). B, Cymodocea nodosa; A, Zostera marina; C, Nanozostera noltii. (a) Syngnathus abaster; (b) Syngnathus typhle; (c) Nerophis ophidion.

At the macrohabitat scale, patterns of differential abundance of the three species across the three different sites were observed. While Syngnathus typhle and Nerophis ophidion tended to occur with higher abundance in the site dominated by Cymodocea seagrass, Syngnathus abaster showed higher abundance in the site dominated by eelgrass.

Differences in pipefish abundance across sites could be related not only to the different structural complexity due to the seagrass species dominating each station, rather to some other environmental factors characterising the three different sites, which could potentially affect the relative abundance of each species. However, evidence from a previous large scale study on the relationships between environmental factors and the structure of fish assemblage in the Venice lagoon suggest that the local fish assemblage associated to seagrasses is composed of the three pipefish species investigated by the present study and that this association is largely affected by salinity and the degree of seagrass coverage (Franco et al., 2006b). As these two factors do not differ among the three sites analysed in the present study, it is likely that differences in pipefish abundance are related to differences in habitat structural complexity due to the seagrass species dominating each station. Similar patterns of interspecific differences in habitat preference were found by Kendrick and Hyndes (2003) in pipefish species inhabiting Westerns Australian seagrasses, and by Curtis and Vincent (2005) in sympatric seahorses of a Portuguese lagoon. These studies suggest that mechanisms of habitat partitioning may constitute a recurrent pattern in the syngnathids guilds associated to lagoon and coastal vegetated habitats. Habitat preference exhibited by Syngnathus typhle and Nerophis ophidion at the macrohabitat scale, and the preference shown by N. ophidion male in laboratory experiments, are consistent with the results found in Stigmatopora by Steffe et al. (1989) showing that individuals tend to prefer long over short seagrass. In the Venice lagoon, as shown by the present results on phenological characteristics of seagrasses, leaves are longer in Cymodocea nodosa than in the other two species, contrary to that found in other Adriatic areas (Guidetti et al., 2002). The preference of Nerophis ophidion for long over short seagrass seems to match the morphological and behavioural

S. Malavasi et al. / Estuarine, Coastal and Shelf Science 75 (2007) 143e150

149

Table 3 Time percent (means  S.E.) spent in each artificial habitat by males and females of each species (Syngnathus abaster, Syngnathus typhle, Nerophis ophidion), within each experimental trial (see text for details) Species

Sex

S. abaster S. typhle N. ophidion

M F M F M F

n

5 5 5 5 5 5

Expt. 1

Expt. 2 % Sand

% High density

% Low density

% Short leaves

% Long leaves

98.7  0.7 99  0.5 95  1.3 98.3  1.3 98  0.9 95  0.7

1.3  0.9 1  0.5 5  1.3 1.7  1.3 2  0.7 5  0.5

57  23 59  18 38  16 34  16 61  23 16  16

43  23 40  18 62  16 66  16 39  23 83  16

56  22 28  19 33  17 60  18 98  0.9 40  24

44  22 72  19 66  18 40  18 2  0.9 60  24

traits of this species. This species has a weakly prehensile tail allowing the alignment of the body with the leaves, and longer leaves are probably more advantageous with respect to this cryptic behaviour, as also suggested by Howard and Koehn (1985) and Steffe et al. (1989). Intersexual differences observed in our experiments for N. ophidion are probably related to the influence of reproductive biology on the more general behavioural differences between sexes, as suggested by Roelke and Sogard (1993) for Syngnathus fuscus. These authors found out that males of this species tend to reduce their activity and to remain in protected habitats, and this seems to explain the inter-sexual differences found by us in N. ophidion. By contrast, neither Syngnathus typhle nor Syngnathus abaster showed any significant preference in behavioural tests, although they seem to have different habitat preference according with our field results. This should mean that different factors from those tested in the laboratory are acting on the habitat preferences of both species. As pointed out by Schofield (2003) and Warfe and Barmuta (2004), habitat use may be influenced by a combination of several factors, such as structural complexity, predation and competition. Simple and linear relationships between patterns of distribution and abundance in the field and habitat choice based on structural

100

80

% Time

60

40

20

0 S.ab

Expt. 3

% Grass

S.ty

N.op

Fig. 4. Mean  S.E. of time percent spent by each species (S. ab, Syngnathus abaster; S. ty, Syngnathus typhle; N. op, Nerophis ophidion) in each vertical portion of the artificial seagrass habitat (black bars: bottom; white bars: top portion; dotted bars: intermediate portion).

characteristics in the laboratory are not therefore to be expected. A clear pattern of segregation found by the present work is at the level of the vertical partitioning occurring among the three species. While Syngnathus typhle and Nerophis ophidion occupy the high and the intermediate portion of the canopy, Syngnathus abaster strongly prefers the proximity to the bottom. A similar partitioning pattern has been observed by Kendrick and Hyndes (2003), where the two congeneric Stigmatopora tended to occupy the canopy, whereas Vanacampus poecilolaemus tended to lay among the leaf detritus on the bottom. In our case, the two congeneric Syngnathus species are those which tend to diverge more in terms of microhabitat use, with their microhabitats completely separated. This seems to be consistent with the hypothesis of the ‘‘generic spread’’, outlined by Tokeshi (1999), i.e. that congeneric species tend to occur in different communities avoiding coexistence. Both the tendency to the horizontal segregation, observed in the field, and to the vertical partitioning, observed within the artificial seagrasses of the experimental tanks, agree with this hypothesis. Patterns of microhabitat segregation among sympatric syngnathid species have to be related, as also stated by Curtis and Vincent (2005), to inter-specific differences in foraging strategy. As suggested by previous studies on feeding habits of Mediterranean pipefish species (Franzoi et al., 1993; Campolmi et al., 1996; Franzoi et al., 2004), snout morphology is strictly related to foraging strategies, leading morphologically divergent species to use different microhabitats. In particular, the longer and higher snout enables Syngnathus typhle to catch both faster and larger pelagic preys, while the shorter, cone shaped snout of Syngnathus abaster and Nerophis ophidion allows them to capture small preys hidden in the vegetation, as clearly shown by comparative diet analyses carried out on these three species by Franzoi et al. (2004). A pattern of food niche segregation, which parallels morphological differences in snout length and probably microhabitat use, has been also found in two congeneric Syngnathus species co-occurring in eelgrass beds of lower York River (Teixeira and Musick, 1995). Thus, the observed vertical partitioning along the canopy seems to parallel these differences in foraging strategies. A detailed comparative analysis of foraging strategies in these species should therefore contribute to a better comprehension of the coexistence mechanisms among them.

150

S. Malavasi et al. / Estuarine, Coastal and Shelf Science 75 (2007) 143e150

References Berglund, A., Rosenquist, G., Svensson, I., 1986. Mate choice, fecundity and sexual dimorphism in two pipefish species (Syngnathidae). Behavioural Ecology and Sociobiology 19, 301e307. Berglund, A., Rosenquist, G., Svensson, I., 1988. Multiple matings and paternal brood care in the pipefish Syngnathus typhle. Oikos 51, 184e188. Berglund, A., Rosenquist, G., Svensson, I., 1989. Reproductive success of females limited by males in two pipefish species. American Naturalist 133, 506e516. Buia, M.C., Gambi, M.C., Dappiano, M., 2003. I sistemi a fanerogamemarine. In: Gambi, M.C., Dappiano, M. (Eds.), Manuale di metodologie di campionamento e studio del benthos marino mediterraneo. Biologia Marina Mediterranea 10, pp. 145e198. Campolmi, M., Franzoi, P., Mazzola, A., 1996. Observations on pipefish (Syngnathidae) biology in the Stagnone Lagoon (West Sicily). Publicacion Especiales Instituto Espanol de Oceanografia 21, 205e209. Cappucci, S., Amos, C.L., Hosoe, T., Umgiesser, G., 2004. SLIM: a numerical model to evaluate the factors controlling the evolution of intertidal mudflats in Venice Lagoon, Italy. Journal of Marine Systems 51, 257e280. Curtis, J.M.R., Vincent, A., 2005. Distribution of sympatric seahorse species along a gradient of habitat complexity in a seagrass-dominated community. Marine Ecology Progress Series 291, 81e91. Franco, A., Franzoi, P., Malavasi, S., Riccato, F., Torricelli, P., Mainardi, D., 2006a. Use of shallow water habitats by fish assemblages in a Mediterranean coastal lagoon. Estuarine, Coastal Shelf and Science 66, 67e83. Franco, A., Malavasi, S., Zucchetta, M., Franzoi, P., Torricelli, P., 2006b. Environmental influences on fish assemblage in the Venice lagoon. Chemistry and Ecology 22, 105e118. Franzoi, P., Maccagnani, R., Rossi, R., Ceccherelli, V.U., 1993. Life cycles and feeling habitus of Syngnathus taenionotus and S. abaster (Pisces, Syngnathidae) in a brackish bay of the Po River Delta (Adriatic Sea). Marine Ecology Progress Series 97, 71e81. Franzoi, P., Riccato, F., Franco, A., Torricelli, P., 2004. Dietary differences in three pipefish species (Osteichthyes, Syngnathidae) related to snout morphology. Biologia Marina Mediterranea 11, 592e594. Guidetti, P., Lorenti, M., Buia, M.C., Gazzella, L., 2002. Temporal dynamics and biomass partitioning in three Adriatic Seagrass Species: Posidonia oceanica, Cymodocea nodosa, Zostera marina. Marine Ecology PSZN 23, 51e67. Howard, R.K., Koehn, J.D., 1985. Population dynamics and feeding ecology of pipefish (Syngnathidae) associated with eelgrass beds of Western port, Victoria. Australian Journal of Marine and Freshwater Research 36, 361e370. Jackson, E.L., Rowden, A.A., Attrill, M.J., Bossy, S.F., Jones, M.B., 2002. Comparison of fish and mobile macroinvertebrates associated with seagrass and adjacent sand at St. Catherine Bay, Jersey (English Channel): Emphasis on commercial species. Bulletin of Marine Science 71, 1333e1341. Jenkins, G.P., May, H.M.A., Wheatley, M.J., Holloway, M.G., 1997. Comparison of fish assemblages associated with seagrass and adjacent unvegetated habitats of Port Phillip Bay and Corner Inlet, Victoria, Australia, with emphasis on commercial species. Estuarine, Coastal and Shelf Science 44, 569e588. Jordan, F., Bartolini, M., Nelson, C., Patterson, P.E., Soulen, H.L., 1996. Risk of predation affects habitat selection by the pinfish Lagodon rhomboids (Linnaeus). Journal of Experimental Marine Biology and Ecology 208, 45e56. Kendrick, A.J., Hyndes, G.A., 2003. Patterns of abundance and size distribution of syngnathid fishes among habitats in a seagrass-dominated marine environment. Estuarine, Coastal and Shelf Science 57, 631e640. Malavasi, S., Fiorin, R., Franco, A., Franzoi, P., Granzotto, A., Riccato, F., Mainardi, D., 2004. Fish assemblages of Venice Lagoon shallow waters: an analysis based on species, families and functional guilds. Journal of Marine Systems 51, 19e31. MAV-CVN, 2005. Ministero delle Infrastrutture e dei Trasporti, Magistrato alle Acque di Venezia, tramite il suo concessionario Consorzio venezia

Nuova, Attivita` di monitoraggio ambientale della laguna di Venezia, MELa2dRilievi e monitoraggi della vegetazione acquatica sommersa (fanerogame marine e macroalghe) in laguna di VeneziadRapporto finale. Technical report, 123 pp. Munday, P.L., Jones, G.P., Caley, M.J., 2001. Interspecific competition and coexistence in a guild of coral-dwelling fishes. Ecology 82, 2177e2189. Muus, B.J., Nielsen, J.G., 1999. Sea Fish. Scandinavian Fishing Year Book, Hedehusene, Denmark, 340 pp. Nagelkerken, I., Van Der Velde, G., 2004. A comparison of fish communities of subtidal seagrass beds and sandy seabeds in 13 marine embayments of a Caribbean island, based on species, families, size distribution and functional groups. Journal of Sea Research 52, 127e147. Nagelkerken, I., Kleijnen, S., Klop, T., van den Brand, R.A.C.J., Cocheret de la Morinie`re, E., van der Velde, G., 2001. Dependence of Caribbean reef fishes on mangroves and seagrass beds as nursery habitats: a comparison of fish faunas between bays with and without mangroves/seagrass beds. Marine Ecology Progress Series 214, 225e235. Riccato, F., Fiorin, R., Franco, A., Franzoi, P., Libertini, A., Pranovi, F., Torricelli, P., 2003. Population structure and reproduction of three pipefish species (Telostei, Syngnathidae) in a seagrass meadow of the Venice Lagoon. Biologia Marina Mediterranea 10, 138e145. Roelke, D.L., Sogard, S.M., 1993. Gender-based differences in habitat selection and activity level in the Northern pipefish (Synagnathus fuscus). Copeia 2, 528e532. Root, R.B., 1967. The niche exploitation pattern of the blue-grey gnatcatcher. Ecological Monographs 37, 317e350. Rosenquist, G., 1990. Male mate and female-female competition for mates in the pipefish Nerophis ophidion. Animal Behaviour 39, 1110e1115. Savino, J.F., Stein, R.A., 1989. Behavior of fish predators and their prey: habitat choice between open water and dense vegetation. Environmental Biology of Fishes 24, 287e293. Schofield, P., 2003. Habitat selection of two gobies (Microgobius gulosus, Gobiosoma robustum): influence of structural complexity, competitive interactions, and presence of a predator. Journal of Experimental Marine Biology and Ecology 288, 125e137. Siegel, S., Castellan, N.J., 1992. Statistica Non Parametrica. McGraw Hill, Milano (Italy), 477 pp. Simberloff, D., Dayan, T., 1991. The guild concept and the structure of ecological communities. Annual Review Ecology and Systematics 22, 115e143. Steffe, A.S., Westoby, M., Bell, J.D., 1989. Habitat selection and diet in two species of pipefish from seagrass: sex differences. Marine Ecology Progress. Series 55, 23e30. Teixeira, R.L., Musick, J.A., 1995. Trophic ecology of two congeneric pipefishes (Syngnathidae) of the lower York River, Virginia. Environmental Biology of Fishes 43, 295e309. Tokeshi, M., 1999. Species Coexistence, Ecological and Evolutionary Perspectives. Blackwell Science, Tokyo, Japan, 544 pp. Tomasini, J.A., Quignard, J.P., Capape, C., Bouchereau, L., 1991. Facteurs du success reproductif de Syngnathus abaster Risso 1826 (Pisces, Teleostei, Syngnathidae) en milieu lagunaire mediterraneen (lagune de Mauguio, France). Acta Oecologica 12, 331e335. Underwood, A.J., 1997. Experiments in Ecology. The Logical Design and Interpretation using Analysis of Variance. University Press, Cambridge, 504 pp. Utne, A.C.W., Aksnes, D.L., Giske, J., 1993. Food predation risk and shelter: an experimental study on the distribution of adult two-spotted goby Gobiusculus flavescens (Fabricius). Journal of Experimental Marine Biology and Ecology 166, 203e216. Vincent, A.C.J., Berglund, A., Ahnesjo, I., 1995. Reproductive ecology of pipefish species on eelgrass meadow. Environmental Biology of Fishes 44, 347e361. Warfe, D.M., Barmuta, L.A., 2004. Habitat structural complexity mediates the foraging success of multiple predator species. Oecologia 141, 171e178. Werner, E.E., Hall, D.J., 1977. Competition and habitat shift in two sunfishes (Centrarchidae). Ecology 58, 869e876.