Habitat Groups and Island-Mainland Distribution of Kelp-bed Fishes off Santa Barbara, CA

The islands off southern California diversify the coastal environment by doubling the length of shoreline and extending coastal habitats (Horn 1974). Geologically, and perhaps biologically, these islands may be classified into northern (bordering the Santa Barbara Channel) and southern groups (Hewatt 1946, Valentine and Lipps 1967, Weaver and Doerner 1967). Yet classification of the insular shore fish fauna is more complex. On a broad scale, distributions of shore fishes are influenced by water temperature and associated currents: the cool, southerly-flowing California Current offshore, and the warmer, inshore countercurrent and eddy (Hubbs 1967, 1974, Neushul et al. 1967). Also, the specific assemblage of fishes in a given area will depend very strongly on the habitat structure there. And finally, isolation at islands may be brought about either by differential transport of species having planktonic larvae (Kanter 1980, Seapy and Littler 1980) or by chance transport of species that have no planktonic dispersal stage (cf. Haldorson 1980). Our study analyzes the effects of habitat on the composition of one element of the inshore fish fauna, the kelp-bed Fishes, and applies this analysis to compare the kelp-bed fish assemblages at Santa Cruz Island with those from the adjacent Santa Barbara mainland. Some papers in this symposium dealt with large-scale biogeography of inshore organisms on the California Islands (Seapy and Littler 1980, Silva 1978). Even though we do not address this problem directly, we realize that interpretations of habitat effects must consider geographic affinities of the fauna. On the other hand, habitat effects may confound broad-scale geographic effects (cf. Kanter 1980, Littler 1980). Our objective, therefore, is to show how assemblages of kelp-bed fishes may be classified into particular habitat groups, and how differences in structural habitat affect the composition of fish assemblages making up such groups. We did our study off Santa Barbara, at the southern end of a transitional zone between a warm-temperate biota to the southeast and a cool-temperate biota at San Miguel Island and north of Point Conception (cf. Hewatt 1946, Hubbs 1948, 1960, 1974. Neushul et al. 1967, Quasi 1968b. Ebeling et al. 1971). The mixed composition of the fauna reflects water tempera-lure and exposure to currents. The California Current carries cool water seaward past Point Conception, although a small branch of this current feeds a counterclockwise eddy in the western part of the Santa Barbara Channel (Reid 1965, Kolpack 1971). This eddy meets warmer currents from the southeast at the eastern end of the Channel, near Santa Barbara and Santa Cruz Island (Kolpack 1971). Therefore, even though oceanographically complex, our study areas are warmer and more exposed to southern currents than is San Miguel Island at the western end of the channel. Given the geographic affinities of the fish fauna in our study areas, we investigated the influence of structural habitat on the composition of fish assemblages in and about beds of giantkelp(Macrocystis). In this way, we hoped to explain any "island effect" on these assemblages. as expressed by differences in species abundance and composition between Santa Cruz Island.. and the Santa Barbara mainland. Inshore habitats vary along several environmental gradient (Limbaugh 1955, North 1963, Quast 1968a, Frey 1971). For example, density of giant kelp varies with depth and several other factors (Neushul et al. 1967, North 1963, 1971, Quast 1968a. Pearse and Lowry 1974). A depth-related gradient in dominant plants extends from surfgrass, boa kelp, or Pterygophora-Eisenia communities inshore, through giant kelp and red algae at intermediate depths, to depauperate plant communities in deeper water (Clarke and Neushul 1967, Neushul et al. 1967). Such biotic gradients follow abiotic gradients in temperature, light, wave surge, and productivity (Quast 1968a, Pequegnat 1964). A gradient in substrate type extends from flat and soft bottoms of sand and/or mud, through flat and hard bottoms, to high-relief rocky reefs; the "turf" of sessile animals and plants that covers these reefs varies with degree of water movement, silting, scouring, light penetration, and grazing (McLean 1962, Pequegnat 1964, Turner et al. 1965, 1968, Clarke and Neushul 1967, North 1971, Pearse and Lowry 1974, Neushul et al. 1976). Much like a forest, giant kelp provides a vertical gradient along which animals tend to stratify in the water column. Kelp stipes in midwater and the dense canopy near the surface provide shelter, food, and landmarks for a variety of fishes (Hobson 1965, Quast 1968b, Feder et al. 1974, Alevizon 1976). Limbaugh (1955) and Quast (1968b, 1968c) analyzed kelp-bed fish assemblages mainly in the San Diego area of southern California. Miller and Geibel (1973) and Burge and Schultz (1973) analyzed such assemblages off central California, north of Point Conception. These investigators evaluated specific responses of individual species to their natural environment and were concerned with how assemblages may respond to changes in structural habitat. Excepting a few fragmentary observations and species lists (Hewatt 1946, Clarke and Neushul 1967, Neushul et al. 1967, Quast 1968c), however, there was almost no published information from the Santa Barbara area. We supplement these seminal studies of southern and northern regions by taking a more synthetic approach to analyzing species assemblages in the less well-known transitional region off Santa Barbara. Thus, although we must interpret our results in terms of behavior of particular species, our results provide an overall view of changing fish assemblages. Hopefully, our synecological approach reveals general trends not immediately obvious from autecological studies. We compared kelp-bed fish assemblages sampled at different localities along Santa Cruz Island with assemblages sampled at different localities along the Santa Barbara mainland. We first determined the structure of assemblages by identifying subgroups of species ("habitat groups") that tend to associate with different positions on environmental gradients in and about areas of reef and kelp. We then compared the density, diversity, and composition of assemblages among localities, between mainland and island, and between seasons. With this information, we were better able to distinguish and explain any island effect on the assemblages, in light of the faunal complexity of the region.