Spatial and Temporal Patterns of Habitat Use by Fishes Associated with Sargassum Mats in the Northwestern Gulf of Mexico

Distribution and abundance of fishes associated with Sargassum mats in the northwestern Gulf of Mexico were examined off northern (Galveston) and southern (Port Aransas) Texas from May–August, 2000. A total of 36 species (17 families) was identified from larval purse seine collections. Individuals from seven species composed over 97% of the catch: planehead filefish Monacanthus hispidus (Linnaeus, 1766), blue runner Caranx crysos (Mitchill, 1815), gray triggerfish Balistes capriscus (Gmelin, 1789), chain pipefish Syngnathus louisianae (Günther, 1870), sergeant major Abudefduf saxatilis (Linnaeus, 1758), sargassum fish Histrio histrio (Linnaeus, 1758), and greater amberjack Seriola dumerili (Risso, 1810). Sizes were variable throughout the sampling period; however, over 95% of the species collected were in the early life stages. Temporal patterns were observed for several taxa: M. hispidus, S. louisianae, and H. histrio were more abundant in May and June, while C. crysos and A. saxatilis were more common in July and August. Abundance of dominant taxa was higher in northern waters and both abundance and diversity increased as a function of distance from shore. Natural associations of ichthyofauna were observed across zones and region samples, and most of the variation was attributed to sample month. Results of this study suggest pelagic Sargassum serves as nursery habitat and may influence the recruitment success of several species. Identification and conservation of essential fish habitat is a prerequisite to building healthy and sustainable fisheries (Rosenberg et al., 2000). To date, considerable work has focused upon characterizing habitats of demersal stocks (Lindeman et al., 2000), and these efforts have led to valuable information regarding the physical attributes and biological significance of these habitats. Unfortunately, information on habitat use of pelagic species is limited. The pelagic zone is typically characterized by its lack of physical structure and previous studies suggest that many pelagic organisms associate with structure, particularly during early life stages (Dooley, 1972; Rountree, 1990; Kingsford, 1992). Artificial structure, such as production platforms, flotsam, buoys, and fish aggregation devices (FADs), appears to function as pelagic habitat (Seaman et al., 1989; Rountree, 1990). Moreover, recent findings suggest that floating mats of brown algae (Sargassum) may be of critical importance, and the availability of these natural habitats may influence recruitment success of several species (SAFMC, 1998). Apart from qualitative surveys of pelagic Sargassum communities, the function of these mats as essential habitat has not been adequately addressed and warrants further consideration. Floating pelagic Sargassum is a ubiquitous feature of surface waters in the Gulf of Mexico. Two species of brown algae (Sargassum natans and Sargassum fluitans) comprise nearly all the floating macro-algae in the pelagic zone (Parr, 1939). Both species originate in the Sargasso Sea within the North Atlantic Central Gyre, where physical factors such as currents, gyres, eddies, and winds influence their distribution (Dooley, 1972). Sargassum mats often accumulate in large windrows, thereby forming productive ‘weedlines’ in oligotrophic waters, and these algal mats may account for 60% of the total primary production in the upper 82 BULLETIN OF MARINE SCIENCE, VOL. 74, NO. 1, 2004 meter of water (Carpenter and Cox, 1974; Peres, 1982). In addition, epiphytic cyanobacteria contribute to overall production and nutrient recycling within the Sargassum complex (Phlips et al., 1986). A succession of bacteria, hydroids, and bryozoans initially colonize the algae (Conover and Sieburth, 1964; Ryland, 1974), and these early colonizers provide the base of a food web containing a variety of invertebrates, fishes, and sea turtles (Dooley, 1972; Bortone et al., 1977, Fedoryako, 1989). Due to the physical complexity this habitat affords, Sargassum is of particular interest to scientists and fishery managers. Studies in the western Atlantic (Dooley, 1972; Settle, 1993), Japan (Edgar and Aoki, 1993), Australia (Kingsford and Choat, 1985), central Pacific (Gooding and Magnuson, 1967), and eastern Gulf of Mexico (Bortone et al., 1977) indicate that many fishes use Sargassum. The structural complexity of Sargassum acts to enhance growth by providing ample prey resources for juveniles (Hunter and Mitchell, 1967; Gorelova and Fedoryako, 1986; Edgar and Aoki, 1993), and may also enhance early life survival by reducing predation-mediated mortality (Kingsford, 1995). As a result, survivorship may increase and therefore recruitment success may be linked to Sargassum (Kingsford and Choat, 1985). The National Marine Fisheries Service (NMFS) has recently designated Sargassum as Essential Fish Habitat (EFH) for several coastal migratory species (NOAA, 1996). Nevertheless, studies assessing the value of Sargassum as habitat are limited, particularly in the northwestern Gulf of Mexico where large floating mats are predominant surface features on the continental shelf. The aim of this study was to evaluate the role of Sargassum as nursery habitat of fishes in the northwestern Gulf of Mexico. Spatial and temporal patterns of habitat use were examined to determine the function and dynamics of the Sargassum complex. Specifically, patterns of abundance and assemblage diversity were assessed at two spatial scales. Inshore versus offshore comparisons were analyzed for small-scale spatial patterns, while northern and southern regions of the northwestern Gulf were investigated for large-scale differences. In addition, monthly changes in fish abundance and composition were examined to evaluate temporal stability and identify important periods of recruitment to the Sargassum complex. MATERIALS AND METHODS SAMPLING DESIGN.—This study was conducted in coastal and offshore waters of the northwestern Gulf of Mexico (Fig. 1). All samples were obtained within the following boundaries: north (29∞ 07.82¢ N, 94∞ 43.31¢ W), south (27∞ 37.55¢ N, 96∞ 35.84¢ W), east (28∞ 58.52¢ N, 93∞ 55.19¢ W), and west (27∞ 46.81¢ N, 96∞ 45.05¢ W). Waters 15–70 nm off Galveston and Port Aransas, Texas were designated as the offshore north and offshore south regions, respectively. Regions are separated by approximately 200 mi and oceanographic differences between the two include currents, winds, bottom topography, and nutrient concentrations (Smith, 1980a; Sahl et al., 1993). In addition, an inshore north zone (< 15 nm) off Galveston, Texas was sampled and compared to the offshore north zone (same as the offshore north region). This inshore north zone is heavily influenced by physical and biological processes occurring within the Galveston Bay estuary, such as advection of riverine and bay discharges, while the offshore north zone is governed more by continental shelf processes, such as upwelling (Temple et al., 1977; Smith, 1980b). Sargassum and associated fauna were sampled from May through August 2000, which was the duration Sargassum was present in the area. Replicate samples (3–5) were collected monthly from the offshore north region, offshore south region, and the inshore north zone. All mats were haphazardly chosen and samples were taken from 0800–1500 h using a larval purse seine (20-m long, 3.383 WELLS AND ROOKER: HABITAT USE BY FISHES ASSOCIATED WITH SARGASSUM MATS m deep, 1000 mm mesh). The purse seine was deployed as the boat encircled the chosen mat, and once completely around the mat, the net was pursed. Next, Sargassum was removed (while the net remained in the water), and fishes were funneled into the cod end, collected, and frozen on dry ice. Mat volume (length ¥ width ¥ depth) and GPS locations were recorded at each sample location. Environmental parameters measured included depth, sea surface temperature, salinity, dissolved oxygen, and water clarity. Fishes were sorted and identified to species in the laboratory and standard length measurements were taken to the nearest 0.1 mm. DATA ANALYSIS.—Relative abundance of fishes was expressed as catch per unit effort (CPUE), representing the number of fishes caught per larval purse seine sample. Seven species comprised 97% of the overall catch, and statistical analyses were limited to these taxa. Effects of location and date on CPUE estimates of each species were examined using a two-way analysis of variance (ANOVA). Two-way ANOVAs were also used to examine differences for several other dependent variables; these included environmental conditions, mat volume, sizes, and diversity estimates. Tukey’s honestly significant difference (HSD) test was used to determine a posteriori differences (a = 0.05) among means. The assumption of homogeneity of variances was examined using Levene’s test and residual examination. Normality was examined using a probability plot of residuals versus expected values. CPUE data were log (¥ + 1) transformed to minimize heteroscedasticity. Regional August comparisons were not performed due to a lack of samples in the offshore south region (Port Aransas). In addition, sizes of all seven species were compared; however, sizes of only four species were statistically analyzed between the offshore north and offshore south regions due to a low number of individuals in the south region. Patterns of diversity using all species collected were investigated by Shannon diversity (H') and evenness (J') indices (Zar, 1984). The need to use a Figure 1. Map of sampling locations along the Texas Gulf coast. IN represents the inshore north zone, ON represents the offshore north zone and region, and OS represents the offshore south region. 84 BULLETIN OF MARINE SCIENCE, VOL. 74, NO. 1, 2004 rarefaction method was deemed inappropriate due to the similar number of samples among zones, regions and months. Diversity measures were estimated using the following equations: