Analyses/Maps

Fish

Assemblage Analyses

Introduction

No species exists in isolation from other species or their environment. Monitoring species individually may cause managers to miss important interactions (Chavez et al., 2003; Worm and Myers, 2003; Baraff and Loughlin, 2000; Estes et al., 1998). In addition, individual species' abundances may be considered sustainable in the context of a fishery, but still be low enough to influence ecosystem dynamics and health (NMFS 2001). There has been a growing recognition that effort needs to be extended toward understanding the entire ecosystem. The National Marine Sanctuary Program is tasked with ensuring the continued health of the ecosystems contained in the sanctuaries. However, important species-species interactions and species-habitat interactions are still not well understood. Abiotic (e.g., habitat preferences toward depth or sediment) and biotic (e.g., presence or absence of prey, predators) factors can impact the importance of an area to fish. Elucidating habitat characteristics that are most important to animals, and understanding the co-occurrence of species, is a first step in determining areas that should be managed as "essential" habitats. This study aids in clarifying the interactions among species and between broad scale habitat characteristics and species on the scale of the commercial and recreational fisheries. Even though these data sets were originally deployed to collect information necessary for setting fishing limits, these data sets can provide preliminary information on multi-species interactions. Recreational hook and line drifts covering approximately one kilometer, demersal trawls on the continental shelf and slope covering one kilometer, and fifteen minute trawls in midwater, were analyzed to determine species assemblages, site groupings, and the interaction between species and locations. In addition, analyses were completed to determine larger scale environmental variables that were significantly different among identified groups. Due to limitations of the data sets, and the lack of results on individual species' distributions, habitat suitability models, for selected species were completed to complement this analysis.

Fish Assemblage Analyses Summary Maps

Integration of Assemblage Analyses	Description	PDF Map
Bathymetric Complexity	Description	PDF Map
Species Diversity of Demersal Fish	Description	PDF Map PDF Map
Integration of Community Metrics	Description	PDF Map
National Marine Fisheries Service (NMFS) Trawl in the Midwater Environment	Description	PDF Map PDF Map
Recreational Hook and Line Data from the California Department of Fish and Game (CDF&G)	Description	PDF Map
Species Richness of Demersal Rockfish	Description	PDF Map
Species Richness of Demersal Fish	Description	PDF Map PDF Map
National Marine Fisheries Service (NMFS) Demersal Trawls on the Continental Shelf	Description	PDF Map
National Marine Fisheries Service (NMFS) Demersal Trawls on the Continental Slope	Description	PDF Map
Effects of Temperature and Season		PDF

Download Assemblage Analyses Shapefiles

Download Assemblage Analyses ArcView Grids

The primary objective of the assemblage analysis is to define spatial biogeographic patterns of fishes and macro-invertebrates within the study area from Point Arena to Point Sal in California. The study is based on a synthesis of four primary databases of fish and invertebrates that were spatially comprehensive throughout the study area including: 1) the California Department of Fish and Game fishery dependent recreational fishing trips targeting rockfish (CDF&G recreational); 2) the National Marine Fisheries Service fishery independent benthic trawls on the continental shelf (NMFS shelf trawls); 3) NMFS fishery independent benthic trawls on the slope (NMFS slope trawls); and 4) the NMFS fishery independent trawls in midwater (NMFS midwater trawls). Detailed information on these surveys is given in each respective section. Databases that were limited in content and spatial coverage were not utilized in this analysis, but were used to help interpret the results.

Five specific objectives of the assemblage analysis, all of which aim to increase our understanding of the biogeography of fishes and macro-invertebrates in relationship to their environment, as well as, identify important areas or habitats within the study area, were as follows:
Id

1. Identifying spatial patterns and hot spots in community metrics (diversity and richness);
2. Determining which species tended to be caught together (species assemblages);
3. Analyzing fishing locations to determine which locations contained similar catches (site groups);
4. Resolving where the species assemblages were being caught by combining results from objectives 2 and 3 and then utilizing GIS to map the results; and
5. Identifying significant relationships between site groups identified in objective 3 and broad scale habitat characteristics (bathymetry, bathymetric complexity, and large-scale habitat classification).

Community metrics can be used to determine which areas are important to multiple species. Experts in California have acknowledged a management need to increase our understanding of fish species interactions (objective 2) (Starr, 1998) and the interactions between fish assemblages and habitat (objective 5) (Starr, 1998; Yoklavich et al., 2000, 2002). Studies exist that identify either species assemblages or site groups (see literature review section), but so far none have integrated multiple data sets, provided the interaction between species assemblages and site groups, nor presented spatially explicit results. The results of these analyses aid in defining the region's biogeography based on the spatial pattern of fishes and macroinvertebrates.

Review of Relevant Literature

Due to the economic importance of recreational and commercial fisheries in California, several studies have been completed that look at species co-occurrences or species interactions with their environment. NMFS publishes yearly reports on the status of demersal fish species by analyzing results from their shelf and slope trawls (Turk et al., 2001; Weinberg et al., 2002; Lauth, 2001; Shaw et al, 2000). Zimmerman et al. (2001) looked at the biomass of demersal species to determine NMFS shelf trawls that did not fish the bottom as intended. He then looked at the effect these trawls had on the estimates of biomass of selected species through time. Based on Zimmerman's results, we excluded these abnormal "water hauls" from our analyses. Williams and Ralston (2002) analyzed data from NMFS shelf trawls to determine rockfish species assemblages. The same data were used in this analysis; however, the multivariate statistical method utilized, the spatial coverage employed, and the species examined were different. The overall conclusion from Williams and Ralston was that rockfish richness was highest at a depth of 200-250 meters, where the shelf and slope meet, and that depth and latitude were the main determinants of rockfish assemblages. Jay (1996) analyzed the 1977-1992 NMFS shelf trawls to determine site groups that contained similar catches. Using 33 species of fish, he identified 23 site groups, many of which contained the same species, but with different relative abundances. Even though depth and latitude showed some influence on site groups, overall he found little association between the site groups and a suite of environmental parameters.

Gabriel and Tyler (1980) used data from the Oregon Department of Fish and Wildlife Trawl Survey and the West Coast Joint Agency Rockfish Survey to look for site groups from California to Alaska. They differentiated three large site groups: "intermediate" at less than 145 meters, "deep" between 145 and 200 meters, and "slope" greater than 200 meters deep. They found that site groups were "strongly associated with depth contours". Matthews and Richards (1991) compared gill net catches from trawlable and untrawlable areas to determine if untrawlable areas could be considered de-facto fish reserves. Even though some species overlapped, they concluded that the species assemblages were significantly different; suggesting that species assemblages determined from trawls cannot be extrapolated to non-trawlable habitats.

Only a few studies have analyzed recreational hook and line data. For a general analysis of a species' specific decline in recreational catch see Love et al. (1998), Mason (1998) or Wilson-Vandenberg et al. (1996). Mason (1995) analyzed various CDF&G recreational fishing surveys and documented trends in effort, fishing location, and species catch. She documented two principal rockfish assemblages and distinguished them by depth (less than 70 meters and greater than 70 meters). Sullivan (1995) used the CDF&G recreational fishing data (1987-1992) to determine site groups. His overall conclusion was that the rockfish management groups could be defined, and that both depth and latitude were important.

Even though the neritic zone is ecologically important, little research has addressed the midwater environment. Cailliet et al. (1979) described fish and invertebrate species co-occurrences in anchovy purse seines, and midwater trawls. An extremely thorough report by Larson et al. (1994) looked at the NMFS midwater trawl results in conjunction with local environmental conditions to determine juvenile rockfish assemblages. Their results emphasize the ephemeral nature of the pelagic environment, but they were able to document two consistent spatial trends: 1) the rockfish are larger inshore than offshore, and 2) there was a north/south gradient in species composition and abundance. Moser et al (2000) described changes in rockfish larvae abundance in CalCOFI plankton tows from 1951 to 1998 in response to adult biomass and environmental conditions. He concluded that over-fishing as well as decadal shifts in environmental conditions were affecting the stocks.

Underwater submersibles have been used to describe fish assemblages and their interaction with habitat at spatial scales relevant to the fish themselves. Yoklavich et al. (2000 and 2002) surveyed Soquel Canyon and Big Creek Ecological Reserve on the Big Sur coast, Field et al. (2002) looked at Big Creek Ecological Reserve, while Hixon et al. (1991) and Hixon and Tissot (1992) researched Haceta, Coquille, Daisy, and Stonewall Banks off the Pacific northwest. These results are very important to managers because they show fish and habitat interactions on very small scales. However, many of the results from these studies are not comparable with the current studies due to large differences in scale. Hixon et al. (1991) documented that the species composition observed from the submersibles was different than that seen in trawls. The results from these studies reveal the importance of habitat, especially rugosity, to fish species composition.

Substantial declines in the standing stock biomass of economically important rockfish species (Ralston, 1998) prompted NMFS to organize a symposium to discuss the implications of no-take areas for rockfish in September of 1997. Eleven plenary papers and six case studies are available online, and cover a range of topics. Starr (1998) provided a thorough evaluation of the potential of rockfish no-take reserves. He expressed a management need for the identification of species assemblages. Once assemblages are identified, management can address actions for adequate protection of each species assemblage. Starr also suggested protecting rectangular areas that cover 20-50 km of the coast and extend west to the edge of the continental shelf.