Abstract

Page, H. M., Culver, C. S., Dugan, J. E., and Mardian, B. 2008. Oceanographic gradients and patterns in invertebrate assemblages on offshore oil platforms. – ICES Journal of Marine Science, 65: 851–861.

We explored variability in the composition and cover of subtidal macroinvertebrate assemblages, and the recruitment and growth rates of selected invertebrate species, on seven offshore oil and gas platforms arrayed across a gradient in oceanographic conditions in the Santa Barbara Channel, CA, USA. The major macroinvertebrate taxa (sea anemones, mussels, barnacles, tubiculous amphipods, hydroids, and sponges) were common to all platforms. However, discriminant function analysis (DFA) revealed that the assemblages of two platforms (Gilda and Gail) clearly differed from the other platforms, a pattern attributable, in part, to the presence of conspicuous exotic species (the anemone, Diadumene sp., and encrusting bryozoan, Watersipora subtorquata) on these platforms. If these exotic species were excluded from the analysis, platforms in proximity to each other generally tended to have invertebrate assemblages more similar to each other than to platforms located farther away. Spatial variation in barnacle recruitment onto ceramic plates and mussel growth rate reflected prevailing oceanographic gradients. The existence of along-channel patterns in the composition of platform invertebrate assemblages, and in invertebrate recruitment and growth associated with oceanographic gradients, suggests that assemblages attached to platforms or other artificial structures may be useful barometers of short and perhaps longer term change in ocean climate.

Introduction

The identification of factors that contribute to variation in the structure of marine assemblages has been an important focus of ecological research for decades. Much of the research has explored the role of local physical (e.g. disturbance) and biological (e.g. competition and predation) factors in influencing the structure and dynamics of benthic assemblages in rocky intertidal habitat (reviews by Menge and Branch, 2001; Sousa, 2001). More recently, studies have investigated the potential role of larger-scale or regional oceanographic conditions in structuring these assemblages (Roughgarden et al., 1988; Gaylord and Gaines, 2000; Menge et al., 2003). Spatial and temporal variations in regional oceanographic conditions have been suggested to influence intertidal assemblages through the effects of water temperature on organism life history and physiology (reviews by Hiscock et al., 2004; Somero, 2005), the transport of larval propagules by currents (Roughgarden et al., 1988; Gaylord and Gaines, 2000; Broitman et al., 2005), and the abundance of phytoplankton food available to suspension-feeders (Menge et al., 1997, 2003).

These mechanisms may also play a role in the response of coastal marine ecosystems to longer term climate change (Harley et al., 2006). Of particular interest in this regard is work linking aspects of the structure and dynamics of intertidal assemblages to gradients in oceanographic conditions along the shores of the eastern Pacific coast of North America (Menge et al., 1997; Broitman et al., 2005; Blanchette et al., 2006). These studies suggest that at scales of tens of kilometres, variation in benthic assemblage structure and dynamics can be explained, at least in part, by variation in local oceanographic conditions that affect the delivery of larval propagules and food availability.

The Santa Barbara Channel (SBC) is an ideal location to investigate relationships among biotic assemblages and oceanographic gradients. The channel, ∼100-km long and 50-km wide with a west to east orientation, is bordered on the north by the California mainland and on the south by the Northern Channel Islands (Figure 1). Circulation in the SBC is variable, but beginning in late spring, warmer waters from the Southern California Bight are typically advected into the channel through its eastern entrance. These waters move west where they meet the cooler waters of the California Current, which enters the SBC through its western entrance at Point Conception (Hendershott and Winant, 1996; Harms and Winant, 1998). As a result of this circulation pattern, a gradient of water temperature extending the length of the channel starts developing in late spring and becomes most pronounced in early summer (Otero and Siegel, 2004, Figure 2).

Figure 1.

The Santa Barbara Channel and locations of the seven offshore oil and gas platforms sampled in this study.

Figure 1.

The Santa Barbara Channel and locations of the seven offshore oil and gas platforms sampled in this study.

Figure 2.

Monthly mean sea surface temperature (SST) derived from satellite images, October 1997 to June 2001, showing the intrusion of warm water from the Southern California Bight into the SBC starting in spring. Figure modified from Otero and Siegel (2004). Black dots show approximate locations of sampled offshore oil platforms.

Figure 2.

Monthly mean sea surface temperature (SST) derived from satellite images, October 1997 to June 2001, showing the intrusion of warm water from the Southern California Bight into the SBC starting in spring. Figure modified from Otero and Siegel (2004). Black dots show approximate locations of sampled offshore oil platforms.

The mixing of warmer waters from the south with cooler waters from the north in the SBC creates a biogeographical transitional zone between the Oregonian faunal province, located to the north of Point Conception, and the Californian faunal province to the south (Valentine, 1966; Horn and Allen, 1978; Murray and Littler, 1981). Several studies have explored variation in assemblages and populations along this gradient on the California Channel Islands (Seapy and Littler, 1980; Murray and Littler, 1981; Engle, 1994; Wenner et al., 1994; Broitman et al., 2005; Blanchette et al., 2006). Studies examining assemblage patterns along the SBC in the offshore and on the mainland coast have focused primarily on fish (Horn and Allen, 1978; Love et al., 2003). Notably, Love et al. (2003) observed that assemblages of reef fish associated with offshore oil platforms and natural rocky reef outcrops in the eastern SBC are dominated by warm-temperate species, whereas fish assemblages in the western SBC are dominated by cool-temperate species.

In all, 16 offshore oil and gas platforms are arrayed along the length of the SBC, extending from near the southeast entrance to the northwest opening south of Point Conception (Figure 1). The platform structures are covered intertidally and subtidally by an assemblage of sessile and semi-mobile invertebrates typically found on inshore natural reefs and pier pilings in southern California (Wolfson et al., 1979; Page et al., 1999; Bram et al., 2005), as well as other species that are relatively rare in the inshore environment (e.g. Metridium senile).

Here, we explore variability in subtidal invertebrate assemblages on these platforms across oceanographic gradients on scales of tens of kilometres, in the absence of the habitat heterogeneity that characterizes natural rocky reefs. We utilized the vertical structure of platforms for sampling and for conducting experiments at depths that are less influenced by spatial variation in aerial exposure and wave action. We hypothesized that invertebrate assemblages and the demographic attributes of sessile animals (recruitment and growth) on offshore platforms would vary along the SBC in association with prevailing gradients in oceanographic conditions. To explore this hypothesis, we compared the assemblages of macroinvertebrates attached subtidally to the structure of seven platforms arrayed along the length of the SBC, and examined patterns of barnacle recruitment and mussel growth among these platforms and relationships with selected environmental factors.

Material and methods

Study sites

We conducted our study at seven of the oil and gas platforms located in the SBC. The study platforms were arrayed geographically offshore from Oxnard, CA, in the southeast, northwestwards ∼65 km towards Point Conception (Figure 1), encompassing a range of water depths (29–225 m) and distances from shore (2.9–14.4 km; Table 1). The platforms differed in size (Table 1), but their general configuration is similar, with a subtidal portion consisting of steel, vertical, oblique, and horizontal cross members, together with conductor pipes through which the wells are drilled. The platforms were chosen based on accessibility (some have restricted access) and broad spatial coverage (65 km) of the SBC. Subtidal reef habitat at the water depths of our study platforms is rare along this portion of the Channel.

Table 1.

Characteristics of study platforms.

Variable Gina Gail Gilda Grace Hogan Houchin Holly 
Location along channel (km) 12 15 19 33 36 65 
Year of installation 1980 1987 1981 1979 1967 1968 1966 
Distance from shore (km) 5.0 13.2 11.9 14.4 5.1 7.0 2.9 
Water depth (m) 29 225 64 97 46 49 64 
Platform size (m2 on bottom) 560 5600 2340 3120 1444 1444 1728 
Variable Gina Gail Gilda Grace Hogan Houchin Holly 
Location along channel (km) 12 15 19 33 36 65 
Year of installation 1980 1987 1981 1979 1967 1968 1966 
Distance from shore (km) 5.0 13.2 11.9 14.4 5.1 7.0 2.9 
Water depth (m) 29 225 64 97 46 49 64 
Platform size (m2 on bottom) 560 5600 2340 3120 1444 1444 1728 

Water temperature

To examine spatial variation in water temperature, we attached one electronic (HOBO™) temperature-logger to each of the seven platforms at a depth of 15 m. Water temperature was recorded hourly, and the loggers were retrieved and downloaded ∼3-monthly, spanning summer (June–August 2001), autumn (September–November 2001), and spring (March–May 2002).

Spatial variation in invertebrate assemblages

To explore spatial variation in patterns of invertebrate distribution and abundance among oil platforms along the SBC, we used a Nikonos V 35-mm camera fitted with a water-corrected rectilinear 15-mm lens to sample the invertebrate assemblage photographically, following methods modified from Leichter and Witman (1997), Coyer et al. (1999), and Witman and Smith (2003). The camera and two strobes were mounted on a PVC frame designed to photograph 0.25 m2 quadrats. Quadrats measured 41 × 62 cm internal diameter (0.25 m2), to accommodate the dimensions of the platform legs and conductor pipes. The distribution and abundance of invertebrate taxa were measured by photographing a single 0.25 m2 quadrat located on the inside and outside of the four corner legs and on four randomly selected conductor pipes at depths of 12, 18, and 24 m, for a total of 48 photoquadrats per platform. Disturbance of the invertebrate assemblage by wave action or platform-cleaning occurs mainly in the upper 6–8 m (Page et al., 1999). Photographs were taken between August and November 2001.

In the laboratory, we identified and estimated the percentage cover of major invertebrate taxa within each photoquadrat, using point-contact methods. The percentage cover of all taxa was estimated by projecting the photographic slide images onto 100 randomly located points and recording contacts to the lowest possible taxonomic level (Table 2). Because invertebrate assemblages on the platforms may be several centimetres to decimetres thick, only organisms occupying the surface layer were counted. These organisms were often attached to secondary substratum (e.g. mussels and encrusting bivalves). Mussels and encrusting bivalves were therefore undersampled in the photoplots because they were often covered by other species. We also recorded data on different categories of non-living substrata (e.g. bare steel), if present.

Table 2.

Taxa identified and quantified in terms of percentage cover in photoplots, and taxa identified, but quantified under higher taxa.

Higher taxon Quantified (percentage cover) Identified (not quantified) 
Chlorophyta Green (filamentous)  
Rhodophyta Red (filamentous)  
 Red (bladey)  
 Red (branching)  
Porifera Sponges Halichondria panicea, Spheciospongia confoederata, Haliclona sp. 
Cnidaria   
 Hydrozoa Hydrozoans Plumularia, Aglaophenia 
 Anthozoa Anemones  
 Anthopleura sp.  
 Corynactis californica  
 Diadumene sp.  
 Metridium senile  
 Urtincina sp.  
 Unknown  
Bryozoa Unknown encrusting  
 Watersipora subtorquata  
 Erect Crisia complex, Bugula neritina 
Mollusca Bivalves  
 Mytilus sp. Mytilus galloprovincialis, M. californianus 
 Crassadoma gigantea  
 Small non-mytilid, unknown  
Annelida Sabellidae  
Crustacea   
 Cirripedia Barnacle Megabalanus californicus, Balanus sp. 
 Amphipoda Unknown tubiculous Erichthonius spp. 
 Decapoda Cancer antennarius  
 Loxorhynchus sp.  
Echinodermata Pisaster sp. Pisaster ochraceus, P. giganteus 
 Cucumaria sp.  
 Ophiuroid  
 Strongylocentrotus purpuratus  
Ascidiacea Styela montereyensis  
 Unknown colonial  
Higher taxon Quantified (percentage cover) Identified (not quantified) 
Chlorophyta Green (filamentous)  
Rhodophyta Red (filamentous)  
 Red (bladey)  
 Red (branching)  
Porifera Sponges Halichondria panicea, Spheciospongia confoederata, Haliclona sp. 
Cnidaria   
 Hydrozoa Hydrozoans Plumularia, Aglaophenia 
 Anthozoa Anemones  
 Anthopleura sp.  
 Corynactis californica  
 Diadumene sp.  
 Metridium senile  
 Urtincina sp.  
 Unknown  
Bryozoa Unknown encrusting  
 Watersipora subtorquata  
 Erect Crisia complex, Bugula neritina 
Mollusca Bivalves  
 Mytilus sp. Mytilus galloprovincialis, M. californianus 
 Crassadoma gigantea  
 Small non-mytilid, unknown  
Annelida Sabellidae  
Crustacea   
 Cirripedia Barnacle Megabalanus californicus, Balanus sp. 
 Amphipoda Unknown tubiculous Erichthonius spp. 
 Decapoda Cancer antennarius  
 Loxorhynchus sp.  
Echinodermata Pisaster sp. Pisaster ochraceus, P. giganteus 
 Cucumaria sp.  
 Ophiuroid  
 Strongylocentrotus purpuratus  
Ascidiacea Styela montereyensis  
 Unknown colonial  

Barnacle recruitment

To investigate spatial variation in barnacle recruitment, we used unglazed ceramic terracotta tiles (15 × 15 cm). Two tiles were attached to a PVC frame, and four frames were suspended vertically between adjoining conductor pipes at a depth of 15 m, starting in June 2001. Each tile possessed a smooth and a grooved side. As barnacle recruitment is typically higher on rugose surfaces (Crisp, 1984), we used the grooved side (grooves 1-mm wide) as the sampling surface. Tiles (n = 8 per platform) were retrieved after 3 months and transported to the laboratory, where all barnacles were identified and counted. Data from the two tiles of each frame were averaged to form a single replicate from each frame. Recruitment was assessed in summer (June–August 2001), autumn (September–November 2001), and spring (March–May 2002).

Mussel growth rate

To explore spatial variation in the growth rate of the mussel, Mytilus galloprovincialis, among platforms, we enclosed ten mussels of ∼30 mm shell length in a vexar mesh cage, and attached one cage to each of the PVC frames described above (n = 4 replicate cages per platform). Mussel shell-length was measured initially and following 3 months of deployment, after which the cages were retrieved and replaced by cages of new mussels of ∼30 mm shell length. Growth rates were calculated from the difference in mean shell length of the ten mussels between the beginning and end of each experiment standardized to 30 d of deployment. Mussel growth rate was assessed in summer, autumn, and spring, as above.

Statistical analysis

The data on percentage cover of taxa were arcsine-transformed [p′ = arcsin(√p)] before statistical analysis (Zar, 2003). We tested for significant differences in invertebrate assemblage composition and cover across platforms using multivariate analysis of variance (MANOVA). We also examined differences in assemblage patterns among platforms using canonical discriminant function analysis (DFA), and explored relationships between these patterns and physical variables using multiple regression analysis. Mobile taxa such as crabs and starfish were excluded from statistical analysis, as were macroalgae (∼5% cover). Individual invertebrate species with low overall cover or that were difficult to identify from photographs were grouped into higher taxa (e.g. tubiculous amphipods and sponges) for statistical analysis. Statistical analyses were conducted using SPSS 11.5 (SPSS, Inc.).

Results

Water temperature

Mean water temperature decreased ∼3°C from the southeast, platform Gail, to the northwest, platform Holly, in summer 2001 (Figure 3). However, water temperatures at one of the southern platforms, Gina, were quite variable during this period, and lower overall than at the three other southern platforms (Gail, Gilda, and Grace). An along-channel gradient in temperature was also evident, though less pronounced in autumn 2001 and spring 2002.

Figure 3.

Water temperature at the study platforms during summer and autumn 2001, and spring 2002. Platforms are listed along the x-axis from the most southeast (Gina) to the most northwest (Holly) in the SBC.

Figure 3.

Water temperature at the study platforms during summer and autumn 2001, and spring 2002. Platforms are listed along the x-axis from the most southeast (Gina) to the most northwest (Holly) in the SBC.

Distribution and abundance of selected taxa

Across all platforms, the most widely distributed and abundant higher taxa, together accounting for >80% of the total cover in our photoquadrats, were anemones (e.g. Corynactis californicus, M. senile), tubiculous amphipods, hydroids (Plumularia sp., Aglaophenia sp.), and sponges (e.g. Halichondria panicea and Spheciospongia confoederata; Figure 4). Other widespread taxa included mussels (Mytilus californianus and M. galloprovincialis), barnacles (Megabalanus californicus and Balanus spp.), and tunicates. Exotic species that were rare or absent in local natural habitats were conspicuous and abundant on two of the platforms; the encrusting bryozoan Watersipora subtorquata was observed only on platform Gilda, and the anemone Diadumene sp. was recorded only on platform Gail (see Page et al., 2006). Filamentous red algae were the most widely distributed algal taxon, but in general, the cover of algae was low (∼5%).

Figure 4.

Percentage cover of the anemones, Corynactis californicus and Metridium senile, sponges, tubiculous amphipods, encrusting bryozoans, and hydroids among study platforms. Data were derived for each leg or conductor pipe from photoplots taken inside and outside, and across depths of 12, 18, and 24 m. Data for encrusting bryozoans include the exotic species Watersipora subtorquata recorded only from platform Gilda. Mean values ± 1 s.e.

Figure 4.

Percentage cover of the anemones, Corynactis californicus and Metridium senile, sponges, tubiculous amphipods, encrusting bryozoans, and hydroids among study platforms. Data were derived for each leg or conductor pipe from photoplots taken inside and outside, and across depths of 12, 18, and 24 m. Data for encrusting bryozoans include the exotic species Watersipora subtorquata recorded only from platform Gilda. Mean values ± 1 s.e.

The structure of invertebrate assemblages varied significantly among platforms (p < 0.001, F = 14.761, d.f. = 102, MANOVA). Anemones had higher cover overall (up to 50–60%) than most other invertebrates, but the dominant taxon varied with location (Figure 5). Corynactis californicus was the dominant anemone on platforms at the southeast end of the channel (e.g. Gina, 50 ± 18%, mean ± 1 s.e.); cover of this anemone tended to be lower on platforms to the northwest (e.g. 5 ± 2% at Holly). At Gail, where mean cover of C. californicus was 30 ± 14%, another anemone, the exotic species Diadumene sp., was also abundant (26 ± 7%). In contrast, mean cover of M. senile was generally highest at the northwest platforms (Holly, 61 ± 6%) and lower to the southeast (Gina, 3 ± 3%; Figure 5). An exception to this pattern was evident at Hogan, where Metridium cover was just 3 ± 1%.

Figure 5.

Results of canonical DFA of invertebrate assemblages on the seven study platforms: (a) all species, (b) exotic species excluded. Each data point represents one leg or conductor pipe.

Figure 5.

Results of canonical DFA of invertebrate assemblages on the seven study platforms: (a) all species, (b) exotic species excluded. Each data point represents one leg or conductor pipe.

The cover of tubiculous amphipods, hydroids, and mussels on platforms also generally increased with increasing distance along the channel from the southeast to the northwest (Figure 5). For example, the cover of tubiculous amphipods was 18–20% on Hogan and Houchin, but <5% on Gail and Gilda. The cover of sponges was variable, with greatest cover at Gail (25 ± 10%) and the two most northerly platforms (Houchin and Holly, 14–19%). The bryozoan Watersipora subtorquata occurred only on Gilda, with a mean cover of 14 ± 8%.

Assemblage patterns

DFA revealed that the invertebrate assemblages of platforms Gail and Gilda were clearly different from the other platforms (Figure 5a; taxa used in the analysis indicated in Table 1). Canonical discriminant functions (CDF) 1 and 2 explained 73% of the variation in the data. We used the correlation values in the structure matrix produced by the DFA to identify the most important taxa contributing to the discrimination among platforms. Cover of the exotic anemone Diadumene sp. was highly correlated (r = 0.850, p < 0.001, Pearson's correlation) with CDF1, and an important source of separation of Gail from the other platforms along the CDF1 axis. Cover of the exotic bryozoan Watersipora subtorquata was correlated with CDF2 (r = 0.453, p < 0.001) and an important source of separation of Gilda from the other platforms along the CDF2 axis (Figure 5a).

To explore the effect that inclusion of the two abundant exotic species Diadumene sp. and Watersipora subtorquata had on the patterns observed with DFA, we repeated the DFA, but excluded these species from the analysis (Figure 5b). Removal of the latter from the analysis reduced variability in the Gilda data, and assemblage patterns at this platform tended to become more similar to those at Gina and Grace. In contrast, the structure of the invertebrate assemblage at Gail remained distinct from the other platforms (Figure 5b). Cover of sponges (r = 0.618, p < 0.001, Pearson's correlation) and M. senile (r = 0.444, p < 0.001) were positively correlated with CDF1, and important contributors to the discrimination among platforms along the CFD1 axis. Cover of sponges (r = −0.450, p < 0.001) and Balanus (r = −0.444, p < 0.001) was negatively correlated with CDF2, and important contributors to the separation of Gail from the other platforms along the CDF2 axis.

Assemblage patterns and environmental variables

To explore relationships between assemblage patterns and environmental variables, we used the mean value (centroid) of CDF1 for each platform (computed from data in Figure 5b), and the independent variables of location along the channel, water depth, proximity to shore, and platform size (Table 1) in stepwise multiple regression analysis. Before this analysis, we tested for co-linearity among the independent variables. There was a significant correlation between platform size and both water depth (p < 0.001, r = 0.974, Pearson's correlation) and proximity to shore (p = 0.049, r = 0.758). However, depth and proximity to shore were not significantly correlated (p > 0.1). Therefore, we excluded platform size from the analysis, but included water depth and distance from shore. There was no relationship (p > 0.1) between variation in CDF1 and any of the independent variables if the data from Gail were included in the analysis. If platform Gail, which is located in much deeper water than the next shallowest platform, was excluded, a significant amount of variation in CDF1 was explained by location along the channel (p = 0.047; Table 3, Figure 6).

Figure 6.

Relationship between CDF1 and location of platforms along the SBC. The r2 value was calculated excluding data from platform Gail. Taxa most positively or negatively correlated with CDF1 are also shown on the y-axis.

Figure 6.

Relationship between CDF1 and location of platforms along the SBC. The r2 value was calculated excluding data from platform Gail. Taxa most positively or negatively correlated with CDF1 are also shown on the y-axis.

Table 3.

Summary of results of backwards stepwise multiple regression analysis evaluating the relationship between CDF1 (see Figures 5 and 6) and water depth, distance from shore, and location along the SBC.

Model Effect Coefficient Standard error t p 
Constant −2.454 2.054 −1.195 0.355 
 Water depth −0.083 0.083 −0.998 0.423 
 Distance from shore 0.452 0.478 0.944 0.445 
 Location along channel 0.132 0.067 1.961 0.189 
Constant −1.659 1.839 −0.902 0.434 
 Water depth −0.010 0.029 −0.331 0.762 
 Location along channel 0.075 0.030 2.516 0.086 
Constant −2.166 0.898 −2.412 0.073 
 Location along channel 0.073 0.026 2.833 0.047 
Model Effect Coefficient Standard error t p 
Constant −2.454 2.054 −1.195 0.355 
 Water depth −0.083 0.083 −0.998 0.423 
 Distance from shore 0.452 0.478 0.944 0.445 
 Location along channel 0.132 0.067 1.961 0.189 
Constant −1.659 1.839 −0.902 0.434 
 Water depth −0.010 0.029 −0.331 0.762 
 Location along channel 0.075 0.030 2.516 0.086 
Constant −2.166 0.898 −2.412 0.073 
 Location along channel 0.073 0.026 2.833 0.047 

Data from platform Gail excluded.

Barnacle recruitment

Three species of barnacle recruited to the plates deployed at the platforms: Balanus trigonus, B. regalis, and Megabalanus californicus (Figure 7). Recruitment of B. trigonus, which took place during summer and autumn, was greatest on plates at the two most southeasterly platforms (Gina and Gail), declining with distance towards the more northwesterly platforms. M. californicus recruited during autumn and spring. During autumn, there was no apparent gradient in recruitment along the SBC, whereas during spring, this species and B. regalis recruited primarily at the most southeasterly platform (Gina), with markedly less recruitment onto plates deployed at the other platforms (Figure 7).

Figure 7.

Recruitment of barnacles, Balanus trigonus, Megabalanus californicus, and B. regalis onto ceramic tiles deployed at a depth of 15 m for ∼3 months at each platform. n = 4 per platform. Mean values ± 1 s.e.

Figure 7.

Recruitment of barnacles, Balanus trigonus, Megabalanus californicus, and B. regalis onto ceramic tiles deployed at a depth of 15 m for ∼3 months at each platform. n = 4 per platform. Mean values ± 1 s.e.

Mussel growth

The growth rate of Mytilus galloprovincialis was most rapid during summer and at the southeastern platforms, declining with distance towards the more northwestern platforms (Figure 8). In general, mussels grew more slowly in autumn and spring and without the pronounced spatial pattern along the channel evident in summer. There was a significant correlation between mussel growth and location in the channel during summer (p < 0.05, r = 0.774, Pearson's correlation), but not during any of the other seasons. During summer, mussel growth was also positively correlated with mean temperature (p < 0.01, r = 0.900). However, there was no correlation (p > 0.05) between mussel growth and distance from shore or water depth of the platform during any time of the year.

Figure 8.

Growth rate of mussels, Mytilus galloprovincialis, in cages deployed at a depth of 15 m for ∼3 months at each platform. Mean initial shell length of mussels was ∼30 mm. n = 4 cages of ten mussels at each platform.

Figure 8.

Growth rate of mussels, Mytilus galloprovincialis, in cages deployed at a depth of 15 m for ∼3 months at each platform. Mean initial shell length of mussels was ∼30 mm. n = 4 cages of ten mussels at each platform.

Discussion

Although the major macroinvertebrate taxa (e.g. sea anemones, mussels, barnacles, tubiculous amphipods, hydroids, and sponges) were common to all platforms, the relative abundance of these taxa (as percentage cover) varied along the SBC, such that platforms in proximity to each other tended to have invertebrate assemblages more similar to each other than to platforms located farther away. We propose that along-channel variation in platform invertebrate assemblages reflects, in part, regional oceanographic gradients created through the advection of warmer water into the channel from the south, via the Inshore Counter Current and Southern California Eddy during late spring and summer. The general pattern of decreasing temperature at platforms from east to west along the channel (Figure 3) was consistent with the channel-wide temperature patterns reported by Otero and Siegel (2004), based on monthly satellite SST data from 1997 to 2001. The lower mean and greater variability in water temperatures at Gina than at the other southern platforms (Gail, Gilda, and Grace) in summer and autumn 2001 were probably related to its proximity to Hueneme Canyon. This submarine canyon, located ∼2.5 km to the east of Gina, extends to a depth of >400 m and could serve as a source of cooler water to the nearby shallow water shelf.

There are two scenarios by which along-channel variation in platform assemblages could be influenced by variation in oceanographic conditions. First, assemblage patterns could reflect optimal physiological temperature preferences or tolerances of component species (Barry et al., 1995). Under this scenario, water temperature could influence species distributions through physiological effects on growth, reproduction, and/or survival (Barry et al., 1995; Stillman, 2003; Somero, 2005). Unfortunately, few data are available on relationships between temperature and the distribution of platform species, and many have wide distribution ranges extending the length of the California coastline and beyond. There is some evidence that this explanation applies to the anemone Metridium senile, which was most abundant at the northeastern platforms and apparently has an affinity for cooler water, with a distribution range extending from southern California to Alaska and both sides of the North Atlantic (Morris et al., 1983). The anemone Corynactis californica, which was most abundant on the southern platforms, is considered a “southern” species by Barry et al. (1995), occurring primarily in southern and central California.

Evidence to support a difference in the temperature affinities of Corynactis californica and Metridium senile is also available from another survey of six platforms—three north of Point Conception in the southern Santa Maria Basin, and three in the eastern SBC (Continental Shelf Associates, 2005). These platforms varied in distance from shore (7.5–16 km) and water depth (74–225 m) and two of the platforms (Grace and Gail) were also sampled in our study. Because of differences in study design and data presentation, their results are not statistically comparable with ours. Nevertheless, those authors reported Metridium to be more abundant, in terms of percentage cover, than Corynactis on the legs of the platforms north of Point Conception. Conversely, Corynactis is more abundant than Metridium on the legs of platforms surveyed in the SBC, south of Point Conception.

Our second scenario would be that assemblage structure is influenced by different recruitment rates along the SBC. The prevailing westward flow that brings warmer waters from the Southern California Bight into the channel during late spring and summer could transport invertebrate larvae into the channel from the south. Of the three barnacle species that settled on our plates, Balanus trigonus has the clearest subtropical affinities (Werner, 1967). Its higher recruitment at the southern vs. the northern platforms is consistent with the prediction that west-flowing water masses transport longer-lived planktonic larvae of southern taxa into the channel. The recruitment of Megabalanus californicus and B. regalis at Gina during autumn also suggested that these larvae were transported into the SBC from the south.

The potential importance of prevailing currents and larval transport in structuring platform assemblages is also consistent with observations of a link between patterns of barnacle recruitment onto settlement plates in the rocky intertidal and oceanographic conditions along the shore of Santa Cruz Island (Broitman et al., 2005; Blanchette et al., 2006). Barnacle recruitment was strongest along the eastern shore of the island, which is influenced by intrusions of warmer water from the south, whereas recruitment was less along the western shore, which is subject to persistent influxes of cold, recently upwelled water with apparently few larvae (Broitman et al., 2005). In addition, macroalgae associated with cooler water tended to be more abundant at the western end of the island. In this regard, platform assemblages differed from those of natural habitats in the low overall abundance of macroalgae (cover rarely exceeded 5%), probably in part attributable to shading by the structure and competition for space with sessile invertebrates (Bram et al., 2005).

Larval transport via currents may be less important in influencing the distribution of those members of platform assemblages that have limited dispersal ability. Such species have short pelagic larval stages (anemones, bryozoans, and hydroids), asexual reproduction (anemones and sponges), or crawl-away juveniles (amphipods), which would be “self-seeding” and less dependent on the vagaries of current flow for recruitment (but see Sammarco et al., 2004). At least two exotic species are included in this group, the anemone, Diadumene sp., and the encrusting bryozoan, Watersipora subtorquata, each of which occurs on a different platform. These species were probably introduced via barges or boats, have limited pelagic dispersal ability, but were found at high cover and appear to be superior competitors for space on the platforms (Page et al., 2006).

The invertebrate assemblage of platform Gail did not conform to our general along-channel gradient pattern in assemblage structure, even when the exotic anemone Diadumene sp. was excluded from the analysis. This platform was unique in being much larger (∼2× seabed area) and located in a water depth substantially deeper (161 m deeper) than the next shallowest platform (Table 1). The invertebrate assemblage there was characterized by a high cover (∼20%) of sponges, which were less abundant on the other southeastern platforms. Sponges typically have limited pelagic dispersal ability, but once established can expand their cover through asexual reproduction. Local environmental conditions (e.g. current flow and food availability) may have been conducive to the establishment and growth of sponges on Gail compared with the other platforms, but data to evaluate this possibility are not available.

It was beyond the scope of this study to differentiate between the relative importance of regional and local processes in structuring platform assemblages, or to evaluate whether the importance of these processes varied among platforms. Indeed, our photographic methods only sampled the surface layer of the assemblage, precluding such an analysis. Biological interactions, such as competition and predation, must play an important role along with oceanography in shaping platform assemblages (Wolfson et al., 1979; Dugan, 2000; Bram et al., 2005). For example, barnacles that recruit to bare space can be covered by mussels or overgrown by sponges or other colonial species (Bram et al., 2005), but no information exists on whether these interactions vary among platforms along the SBC. The invertebrate assemblage may develop to a greater thickness on platforms closer to shore in shallow water than on platforms farther from shore in deeper water (MBC Applied Environmental Sciences, 1987), suggesting that these assemblages could be subject to less disturbance (or slower rates of recruitment). Proximity to shore/water depth, however, was not a satisfactory predictor of invertebrate assemblage patterns at our study platforms.

The indication that platform invertebrate assemblages are influenced by variable oceanographic conditions is congruent with observations that the composition and relative abundance of reef fish on shallow-water platforms along the SBC is related to oceanography (Love et al., 2003). Reef fish assemblages at platforms in the western SBC are dominated by a cool-temperate assemblage of rockfish (Scorpaenidae) and surfperch (Embiotocidae), whereas those at platforms in the eastern SBC are dominated by a warm-temperate assemblage consisting primarily of damselfish (Pomacentridae), wrass (Labridae), and sea chub (Kyphosidae). This trend was also evident at natural inshore rocky outcrops along the SBC, although local habitat features (outcrop relief, presence of giant kelp Macrocystis pyrifera), which influence fish distribution and abundance, modified biogeographical patterns.

The growth rate of caged mussels, Mytilus galloprovincialis, varied along the SBC during summer, growth rates being fastest at the southeastern platforms. Growth was strongly correlated with the east-to-west temperature gradient in the channel at that time of year. However, mussel growth was not correlated with water temperature during spring (cool temperatures) or autumn (warm temperatures), or for data grouped across the three seasons. Moreover, mean water temperature varied considerably (12.6–15.8°C) among seasons at one location (Holly), whereas mussel growth varied little across these periods. Therefore, gradients in water temperature per se clearly do not explain satisfactorily the spatial and temporal variation of mussel growth in the channel.

Alternatively, spatial and temporal variation in mussel growth rate may reflect patterns in food availability, or the interaction of food availability and water temperature (Seed, 1976). Temporal variation in the growth rate of M. galloprovincialis correlated with chlorophyll a and particulate organic carbon concentrations at platform Holly, after incorporation of a 3-week time-lag (Page and Hubbard, 1987). In our study, the rates at Holly varied less than at the other platforms. This platform is closest to Point Conception, an area of upwelling and regular phytoplankton blooms, where northern waters first enter the SBC; it could experience episodic pulses of food throughout the year without sustained periods of low food availability (Page and Hubbard, 1987).

The existence of along-channel patterns in the composition of platform invertebrate assemblages, and in invertebrate recruitment and growth, suggests that assemblages attached to platforms or other artificial structures may be useful as barometers of short-term and longer term change in ocean climate. Offshore platforms typically have lifespans of decades; three of our study platforms have been in place for ∼40 years. Within this time-span (42 years), surface-layer ocean temperatures off the coast of California have increased by 0.8°C uniformly in the upper 100 m (Roemmich, 1992). Also over this time-scale, there has been a decrease in the biomass of macrozooplankton in waters off southern California (Roemmich and McGowan, 1995), and declines in abundance and shifts in the composition of temperate reef fish assemblages have been observed in the Southern California Bight (Holbrook et al., 1997; Brooks et al., 2002).

We propose that in the short term, interannual changes in oceanographic conditions that affect water temperature and the movement of water masses (e.g. upwelling, El Niño–Southern Oscillation or ENSO) may influence platform assemblages through effects on invertebrate recruitment and growth. For example, we observed less invertebrate recruitment on settlement plates attached to platform Houchin associated with the colder waters of La Niña of 1999 than the following year (Bram et al., 2005). In the longer term (decades), changes in climate that alter ocean currents, upwelling regimes, and temperature (e.g. multiyear ENSO events, Pacific Decadal Oscillations or PDO, and global warming) may shift the composition of platform invertebrate assemblages through similar effects on population dynamics. Longer term monitoring of platform invertebrates could provide insight into the responses of benthic assemblages to climate change. For biogeographic transition zones, such as the SBC, such monitoring would permit an evaluation of the concept that these zones are particularly susceptible to shifts in the composition of marine species driven by ocean climate (Helmuth et al., 2006).

Acknowledgements

We thank J. Bram, M. Newnham, and A. Willis for assistance in the field and laboratory, and M. Nishimoto, D. Schroeder, and P. Raimondi for discussions. We also thank Venoco, Inc., Pacific Operators Offshore, LLC., and Nuevo Energy for access to their platforms, and the Minerals Management Service, US Department of the Interior, and the Coastal Marine Institute, University of California, Santa Barbara (UCSB) for support. The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing official policies of the US Government, either expressed or implied.

References

Barry
J. P.
Baxter
C. H.
Sagarin
R. D.
Gilman
S. E.
Climate-related long-term faunal changes in a California rocky intertidal community
Science
 , 
1995
, vol. 
267
 (pg. 
672
-
675
)
Blanchette
C. A.
Broitman
B. R.
Gaines
S. D.
Intertidal community structure and oceanographic patterns around Santa Cruz Island, CA, USA
Marine Biology
 , 
2006
, vol. 
149
 (pg. 
689
-
701
)
Bram
J. B.
Page
H. M.
Dugan
J. E.
Spatial and temporal variability in early successional patterns of an invertebrate assemblage at an offshore oil platform
Journal of Experimental Marine Biology and Ecology
 , 
2005
, vol. 
317
 (pg. 
223
-
237
)
Broitman
B. R.
Blanchette
C. A.
Gaines
S. D.
Recruitment of intertidal invertebrates and oceanographic variability at Santa Cruz Island, CA, USA
Limnology and Oceanography
 , 
2005
, vol. 
50
 (pg. 
1473
-
1479
)
Brooks
A. J.
Schmitt
R. J.
Holbrook
S. J.
Declines in regional fish populations: have species responded similarly to environmental change?
Marine and Freshwater Research
 , 
2002
, vol. 
53
 (pg. 
189
-
198
)
Continental Shelf Associates.
Survey of invertebrate and algal communities on offshore oil and gas platforms in southern California: Final report
Prepared for the U.S. Department of the Interior, Minerals Management Service, Pacific OCS Region, Camarillo, CA. OCS Study MMS 2005-070
 , 
2005
pg. 
232
  
plus appendices
Coyer
J.
Stellar
D.
Witman
J.
A Guide to Methods in Underwater Research
The Underwater Catalog
 , 
1999
2nd edn.
Ithaca, NY
Shoals Marine Laboratory, Cornell University
Crisp
D. J.
Costlow
J. D.
Tipper
R. C.
Overview of research on marine invertebrate larvae, 1940–1980
Marine Biodeterioration: An Interdisciplinary Study
 , 
1984
Annapolis, MD
Naval Institute Press
(pg. 
103
-
126
)
Dugan
D. S.
Rock crab growth on an oil platform
2000
CA
MSc thesis, California Polytechnic State University, San Luis Obispo
pg. 
108
 
Engle
J. M.
Halvorson
W. L.
Maender
G. J.
Perspectives on the structure and dynamics of nearshore marine assemblages of the California Channel Islands
The Fourth California Islands Symposium: Update on the Status of Resources
 , 
1994
Santa Barbara, CA
Santa Barbara Museum of Natural History
(pg. 
13
-
24
)
Gaylord
B.
Gaines
S. D.
Temperature or transport? Range limits in marine species mediated solely by flow
American Naturalist
 , 
2000
, vol. 
155
 (pg. 
769
-
789
)
Harley
C. D. G.
Hughes
A. R.
Hultgren
K. M.
Miner
B. G.
Sorte
C. J. B.
Thornber
C. S.
Rodriguez
L. F.
, et al.  . 
The impacts of climate change in coastal marine systems
Ecology Letters
 , 
2006
, vol. 
9
 (pg. 
228
-
241
)
Harms
S.
Winant
C. D.
Characteristic patterns of the circulation in the Santa Barbara Channel
Journal of Geophysical Research
 , 
1998
, vol. 
103
 (pg. 
3041
-
3065
)
Helmuth
B.
Mieszkowska
N.
Moore
P.
Hawkins
S. J.
Living on the edge of two changing worlds: forecasting the responses of rocky intertidal ecosystems to climate change
Annual Review of Ecology, Evolution, and Systematics
 , 
2006
, vol. 
37
 (pg. 
373
-
404
)
Hendershott
M. C.
Winant
C. D.
Surface circulation in the Santa Barbara Channel
Oceanography
 , 
1996
, vol. 
9
 (pg. 
114
-
121
)
Hiscock
K.
Southward
A.
Tittley
I.
Hawkins
S.
Effects of changing temperature on benthic marine life in Britain and Ireland
Aquatic Conservation: Marine and Freshwater Ecosystems
 , 
2004
, vol. 
14
 (pg. 
333
-
362
)
Holbrook
S. J.
Schmitt
R. J.
Stephens
J. S.
Changes in an assemblage of temperate reef fishes associated with a climate shift
Ecological Applications
 , 
1997
, vol. 
7
 (pg. 
1299
-
1310
)
Horn
M. H.
Allen
L. G.
A distributional analysis of California coastal marine fishes
Journal of Biogeography
 , 
1978
, vol. 
5
 (pg. 
23
-
42
)
Leichter
J. J.
Witman
J. D.
Water flow over subtidal rock walls: relation to distributions and growth rates of sessile suspension feeders in the Gulf of Maine. Water flow and growth rates
Journal of Experimental Marine Biology and Ecology
 , 
1997
, vol. 
209
 (pg. 
293
-
307
)
Love
M. S.
Schroeder
D. M.
Nishimoto
M. M.
The ecological role of oil and gas production platforms and natural outcrops on fishes in southern and central California: a synthesis of information
US Department of the Interior, US Geological Survey, Biological Resources Division, Seattle, Washington, 98194. OCS Study MMS 2003-032
 , 
2003
MBC Applied Environmental Sciences.
Ecology of oil/gas platforms offshore California. OCS Study MMS 86-00094
Pacific Outer Contenental Shelf Region of the Minerals Management Service, US Department of the Interior, Washington, DC
 , 
1987
Menge
B. A.
Branch
G. M.
Bertness
M. D.
Gaines
S. D.
Hay
M. E.
Rocky intertidal communities
Marine Community Ecology
 , 
2001
Sunderland, MA
Sinauer Associates, Inc
(pg. 
85
-
130
)
Menge
B. A.
Daley
B. A.
Wheeler
P. A.
Strub
P. T.
Rocky intertidal oceanography: an association between community structure and nearshore phytoplankton concentration
Limnology and Oceanography
 , 
1997
, vol. 
42
 (pg. 
57
-
66
)
Menge
B. A.
Lubchenco
J.
Bracken
M. E. S.
Chan
F.
Foley
M. M.
Freidenburg
T. L.
Gaines
S. D.
, et al.  . 
Coastal oceanography sets the pace of rocky intertidal community dynamics
Proceedings of the National Academy of Sciences of the USA
 , 
2003
, vol. 
100
 (pg. 
12229
-
12234
)
Morris
R.
Abbott
D.
Haderlie
E.
Intertidal Invertebrates of California
 , 
1983
Stanford, CA
Stanford University Press
Murray
S. N.
Littler
M. M.
Biogeographical analysis of intertidal macrophyte floras of southern California
Journal of Biogeography
 , 
1981
, vol. 
8
 (pg. 
339
-
351
)
Otero
M. P.
Siegel
D. A.
Spatial and temporal characteristics of sediment plumes and phytoplankton blooms in the Santa Barbara Channel
Deep-Sea Research II
 , 
2004
, vol. 
51
 (pg. 
1129
-
1149
)
Page
H. M.
Dugan
J. E.
Culver
C. S.
Hoesterey
J. C.
Exotic invertebrate species on offshore oil platforms
Marine Ecology Progress Series
 , 
2006
, vol. 
325
 (pg. 
101
-
107
)
Page
H. M.
Dugan
J. E.
Dugan
D. S.
Richards
J. B.
Hubbard
D. M.
Effects of an offshore oil platform on the distribution and abundance of commercially important crab species
Marine Ecology Progress Series
 , 
1999
, vol. 
185
 (pg. 
47
-
57
)
Page
H. M.
Hubbard
D. M.
Temporal and spatial patterns of growth in mussels Mytilus edulis on an offshore platform: relationships to water temperature and food availability
Journal of Experimental Marine Biology and Ecology
 , 
1987
, vol. 
111
 (pg. 
159
-
180
)
Roemmich
D.
Ocean warming and sea level rise along the southwest US coast
Science
 , 
1992
, vol. 
257
 (pg. 
373
-
375
)
Roemmich
D.
McGowan
J.
Climate warming and the decline of zooplankton in the California Current
Science
 , 
1995
, vol. 
267
 (pg. 
1324
-
1326
)
Roughgarden
J.
Gaines
S.
Possingham
H.
Recruitment dynamics in complex life-cycles
Science
 , 
1988
, vol. 
241
 (pg. 
1460
-
1466
)
Sammarco
P. W.
Atchison
A. D.
Boland
G. S.
Expansion of coral communities within the northern Gulf of Mexico via offshore oil and gas platforms
Marine Ecology Progress Series
 , 
2004
, vol. 
280
 (pg. 
129
-
143
)
Seapy
R. R.
Littler
M. M.
Power
D. M.
Biogeography of rocky intertidal invertebrates
The California Islands. Proceedings of a Multidisciplinary Symposium
 , 
1980
Santa Barbara, CA
Santa Barbara Museum of Natural History
(pg. 
307
-
323
787 pp
Seed
R.
Bayne
B. L.
Ecology
Marine Mussels: Their Ecology and Physiology
 , 
1976
New York
Cambridge University Press
(pg. 
13
-
65
)
Somero
G. N.
Linking biogeography to physiology: evolutionary and acclamatory adjustments of thermal limits
Frontiers in Zoology
 , 
2005
, vol. 
2
  
1; doi:10.1186/1742-9994-2-1
Sousa
W. P.
Bertness
M. D.
Gaines
S. D.
Hay
M. E.
Natural disturbance and the dynamics of marine benthic communities
Marine Community Ecology
 , 
2001
Sunderland, MA
Sinauer Associates, Inc
(pg. 
85
-
130
)
Stillman
J. H.
Acclimation capacity underlies susceptibility to climate change
Science
 , 
2003
, vol. 
301
 pg. 
65
 
Valentine
J. W.
Numerical analysis of marine molluscan ranges on the extratropical north-eastern Pacific shelf
Limnology and Oceanography
 , 
1966
, vol. 
11
 (pg. 
198
-
211
)
Wenner
A. M.
Dugan
J. E.
Hubbard
D. M.
Hochberg
F. G.
Sand crab population biology on the California islands and mainland
Third California Islands Symposium. Recent Advances in Research on the California Islands
 , 
1994
Santa Barbara, CA
Santa Barbara Museum of Natural History
(pg. 
335
-
348
)
Werner
W. E.
The distribution and ecology of the barnacle Balanus trigonus
Bulletin of Marine Science
 , 
1967
, vol. 
17
 (pg. 
64
-
84
)
Witman
J. D.
Smith
F.
Rapid community change at a tropical upwelling site in the Galápagos Marine Reserve
Biodiversity and Conservation
 , 
2003
, vol. 
12
 (pg. 
25
-
45
)
Wolfson
A.
Van Blaricom
G.
Davis
N.
Lewbel
G. S.
The marine life of an offshore oil platform
Marine Ecology Progress Series
 , 
1979
, vol. 
1
 (pg. 
81
-
89
)
Zar
J. H.
Biostatistical Analysis,
2003
4th edn
Delhi
Pearson Education