EGG AND LARVAL FISH PRODUCTION FROM MARINE ECOLOGICAL RESERVES.
Vetter, Supervisory Physiologist: Leader, Genetics and Physiology
H. Geoffrey Moser, Supervisory Fishery Biologist: Leader, Ecology and Systematics of Fishes
William Watson, Fishery Biologist (research): Ecology and Systematics of Fishes
Marine reserves are receiving increased attention as a tool for the preservation of biodiversity, the protection of threatened species, and. the management of exploited resources in the near shore waters of California. Different types of marine reserves, which provide different levels of protection, have been established by both state and federal agencies. Recently, four reserves were established as part of the California Marine Resources Protection Act. Two of these reserves, Vandenberg Marine Ecological Reserve and Big Sycamore Canyon Ecological Reserve are the subject of this proposal (Figure 1).
Marine reserves can function as "living museums" where pristine habitats can be viewed and studied, as source areas that produce "spill over" of harvestable resources into adjacent fishable habitat outside the reserve, and as a protected source of mature fish whose larval production "reseeds" areas outside the reserve (Rowley, 1994). The effectiveness of a marine reserve in meeting any of these three goals depends on the type of habitat preserved, the size and location of the reserve, and the life history of the species considered. The third function of marine reserves, larval production, could be the most influential for the long-term health of the entire coastal zone for fisheries that are recruitment limited, but it is the most difficult to document. This is because newly spawned eggs and larvae, which can most accurately be attributed to a specific location such as a reserve, are the patchiest and the most difficult to identify. In this proposal we will use a combination of new and established techniques to sample and identify the ichthyoplankton assemblages associated with two Southern California Reserves, Vandenberg and Big Sycamore Canyon.
2. PROJECT OBJECTIVES, TIMETABLE, and ESTIMATED COST:
The overall objective of the proposal is to determine the effectiveness of the Vandenberg and Big Sycamore Canyon Marine Ecological Reserves as source areas for the production of fish eggs and larvae that will "reseed" areas outside of the Reserves. To do this we will complete four objectives:
The research will continue for three years and support the dissertation research of one graduate student. The first half of year one (6/1/97-5/31/98) will be devoted to laboratory work to obtain diagnostic molecular identification protocols for the fish families of interest. We will also begin diver transect surveys of habitat and fish fauna from a smaller boat, the RV Ballena. The second half of year one will be devoted to the first full D.S. Jordan cruise in winter of 1997-98. The first half of the second year will be the summer 1998 cruise. The second half of 1998 will be devoted to sorting and analysis of the summer cruise and the completion of the winter 1998-99 cruise. The first half of the third year will be for the summer 1999 cruise and the second half for completion of sample analysis and the writing and publication of research results. The costs for the three years are: year one $416,025 with $75,035 from MERRP, year two $413,301 with $69,943 from MERRP and year three $134,983 with $70,069 from MERRP. Ship time is listed in years one and two and not year three but will overlap into the third year.
3. BACKGROUND AND AVAILABLE DATA:
Techniques for documenting egg and larval production from specific areas, such as reserves, are in their infancy, but developing these skills is necessary for scientifically documenting the value of marine-reserves as larval source areas. The primary dilemma is that the newly spawned eggs and larvae, which are the easiest to trace to a point of origin, are the hardest to sample because they are the most patchy, and the most difficult to identify. We will also need a full understanding of the types of habitat contained in the reserves, the current patterns that drive dispersal and recruitment, and a basic knowledge of the resident fish fauna. Some of this information is already available, and the remainder will be developed in the course of the proposed research.
The two reserves are different in their habitats, currents, and sea conditions. The Vandenberg Reserve is characterized by rocky headlands that extend a short distance into the water. The predominant bottom type is sand with fauna and flora characteristic of high currents and sand scour (D. VanTresca, pers. comm.). There are patches of hard bottom and also patches of surf grass. In general, diving conditions are often poor because the site is directly exposed to northwest winds (D. VanTresca, op. cit.). Directly to the south of the Reserve is an area of high rocky relief with extensive kelp cover.
The Big Sycamore Canyon Reserve is a low relief sand habitat with lower wave energy and lower wind stress than the Vandenberg Reserve. There are extensive areas of kelp to the south of the Big Sycamore Canyon Reserve.
For this proposal scientists working at the SWFSC constructed preliminary bathymetric maps of the two reserves by plotting data available through the National Oceanic Data Center (NODC). The map of the Vandenberg Reserve is shown in Figure 2. These maps can be improved as more sounding data are made available. Habitat information and Pt. Arguello ichthyofaunal distributions can be mapped onto these base maps via a GIS (Global Information System) format.
Current flow in both of the reserves is predominantly longshore. In the surf zone flow is driven by surface waves and outside by coastal currents. There is also a strong seasonal pattern to the direction of current flow (C. Winant, pers. comm.). At the Vandenberg Reserve the flow is predominantly south during spring and summer upwelling with a reversal and northward flow from September to February. At the Big Sycamore Canyon Reserve the flow is generally from the Santa- Barbara Channel moving southeast through the Reserve, and after late summer the flow reverses and goes out the Santa Barbara Channel (Hendershott & Winant, 1996).
Over the past 40 plus years scientists at the Southwest Fisheries Science Center (SWFSC) have been engaged in systematically trying to identify all of the fish eggs and larvae that occur in the California Current. Through the California Cooperative Fisheries Investigations (CalCOFI), scientists have established a regular ichthyoplankton sampling grid with collections going back more than 40 years (e.g. Moser et al., 1993). In addition to the regular CalCOFI sampling grid, additional intensive studies have occasionally been done, including one at Point Arguello near the Vandenberg Reserve (Ahlstrom, 1965). Furthermore, William Watson has more than ten years of experience in quantitative sampling and identification of ichthyoplankton from shallow coastal waters off southern California, primarily from the San Onofre vicinity.
There are several CaICOFI sites near the two reserves which can be used to establish general long term patterns of abundance in areas near the reserves (Figure 1). Although there are clear trends of decreasing abundance with distance from shore for many coastal species, the eggs and larvae collected in the CalCOFI survey grid are already too dispersed to attribute to a particular site of origin such as a reserve (e.g. Moser et al., 1993, 1994).
Presently, Dr. Moser's group at the SWFSC can identify nearly 600 species of larval fish (Moser (ed.), 1996). However, there are eggs and early larval stages of many fishes that can only be identified to genus or family. In some cases these are rare species that when studied in greater detail will yield a diagnostic visual feature. In most cases, however, the unknown material simply does not have any diagnostic characteristics. For these problem cases we have turned to the tools of molecular biology to develop new methods of identification.
Dr. Vetter's group has used molecular methods to investigate a group of seven rockfishes of great commercial and recreational value that previously could not be identified as larvae, but can be identified as juveniles all bearing a characteristic black spot on the first dorsal fin. The seven rockfish species include the widow (S. entomelas), yellowtail (S. flavidus), black (S. melanops), vermilion (S. miniatus), blue (S. mystinus), canary (S. pinniger) and olive (S. serranoides). We have sequenced the first 800 base pairs of the cytochrome b gene for each of the seven species using polymerase chain reaction (PCR). Analysis of these sequence data have shown enough species-specifc variation to enable us to show that the seven species can be identified by differences in their cytochrome b gene sequence. This method is too time consuming for routine identification, but once sequence differences are known, easier methods can be constructed based on the known differences.
Restriction enzymes cut DNA at specific sites depending on the base sequence. Two species that differ in sequence at a restriction site will be cut or not cut, resulting in either one or two DNA fragments of differing length. If sufficiently different sites can be found between closely related species then a unique "bar code" of different sized restriction fragments can be used to identify all of the species within that group. Figure 3 shows the restriction fragment length polymorphism (RFLP) pattern for the seven species of rockfish in our initial study. Similar techniques can be devised to identify the eggs of species within other genera or family groups such as the wrasses, croakers, gobies, blennies or flatfishes that contain species characteristic of different near-shore habitat types.
Another technique that can be used to identify eggs and larvae is to simply allow them to develop to a point where they can be identified. In many cases this means allowing the eggs to hatch. On the standard CaICOFI cruises this is not done because of the large number of sites and the large number of other measurements being conducted at each site. However, it is relatively easy and routine to separate samples and incubate them until hatching or until they have developed sufficiently to be identified (Hunter, 1984). This was how many of the original ichthyoplankton descriptions were obtained (Moser (ed.), 1996).
Accurate sampling of patchy eggs and larvae has been a fundamental problem in ichthyoplankton surveys, but recent technical developments have allowed vast improvements in sampling precision. One is the multiple opening and closing net system (MOCNESS) which allows discrete samples to be taken in rapid succession. Typically, the nets are fired in a vertical succession of depths to obtain a vertical distribution but they can be fired sequentially at the same depth to continuously sample newly spawned eggs or larvae drifting away from a point of origin. Dr. Vetter has successfully deployed the MOCNESS from a crane off the side of a ship to obtain near surface samples from undisturbed water unaffected by the ship's wake and prop wash. A similar technique could be used to monitor egg vertical distribution in the vicinity of the reserves.
A new sampling apparatus developed by Checkley et al. (submitted) with sampling procedures refined by scientists at SWFSC and SIO is the egg pump (Figure 4). This device allows true continuous sampling of the water mass passing under the ship. Recent studies using the egg pump have been highly successful in defining the fine-scale horizontal distribution of sardine eggs around the Channel Islands (Dotson et al., in prep.). Both the MOCNESS and the egg pump would be effective samplers of the egg production from point sources. The egg pump is less valuable for sampling larval fishes, which may be damaged by the strong water flows through the collecting mesh.
4. PROPOSED WORK:
a. Objective 1. Physical Habitat Description of the two Reserves: Marine reserves provide more protection for sedentary species than for highly migratory ones. Sedentary fish species are more likely to be associated with hard- bottom reefs, tide pools, grass beds, or kelp forest than with open sand habitat. The effectiveness of Vandenberg and Big Sycamore Canyon Ecological Reserves in providing protection for adult spawning biomass of any particular species will depend in part on the amount of the required habitat that is present in the Reserve. Therefore, one of the major objectives of our research will be to map the different habitat types and overlay the information onto the basic bathymetric map of the Reserves. This will be accomplished primarily by diver transects of the areas. Data will be plotted on the basis of GPS (Global Positioning System) coordinates entered into a GIS data management system currently being installed at the SWFSC. Data will be collected during our twice-yearly occupation of the two Reserves aboard the RV David Starr Jordan. Data will also be collected on periodic diving trips aboard the Channel Islands National Marine Sanctuary boat, the RV Ballena.
b. Objective 2. Current Patterns in the Reserves: General current flow will be determined by the ADCP (Acoustic Doppler Current Profiling) system aboard the D.S. Jordan. The ADCP is not accurate in very shallow water, but should work well to depths as shallow as 40 m, perhaps less. We will transit along the outside of the Reserves each day of sampling to determine the current flows prior to sampling, and we will continue to collect data during our sampling transects, although the quality of the data in shallower water is yet to be determined.
New developments in current drifters (C. Winant, pers. comm.) should provide an ideal way to track specific water parcels as they move through the Reserves. Development of GPS drifters that report continuously to a base station on land or aboard a stationary vessel provide much finer resolution than drifters that report a few times per day via ARGOS satellite. Because they report continuously they are easily recovered, by GPS equipped launches.
c. Objective 3. Adult and Juvenile Fish Fauna of the Reserves: Quantitative censusing of the fish fauna is beyond the scope of this proposal, but a fish faunal list is essential. Destructive methods of sampling (e.g.. hook and line, surveys or trawling) are not acceptable in the Reserves. This information will be obtained by diver transects and by live trapping with subsequent release. Pelagic juveniles and large larvae will be sample with a 10-m2 MOCNESS and epibenthic juveniles and larvae with an Auriga net. These collections will provide information on fish species recruiting to the Reserves. Juvenile and late-larval stages collected in the MOCNESS and Auriga net will be aged by John Butler of the SWFSC.
d. Objective 4. Fish Egg and Larval Production from the Reserves: This is the primary objective of the proposed research. To effectively describe the ichthyoplankton community of the reserves we need to know seasonal, areal, vertical, and diel aspects of egg and larval production. The data gathered from this portion of the research should provide an estimate of present spawning activity at the two Reserves in comparison to adjacent areas. Since both Reserves were previously fished, it is anticipated that our results will also serve as a baseline from which we can judge the development and recovery of the fish fauna within the Reserves. We will make two cruises to the two Reserves each year for two years. We will occupy the two sites for a period of two weeks in winter and two weeks in summer. We will supplement this primary data set with periodic sampling trips aboard the Channel Islands National Marine Sanctuary boat, the RV Ballena.
i. Sampling design: The basic sampling pattern will be to sample three regions at each of the two sites. These will be north of each Reserve, south of the Reserve and in the Reserve. At each of the three regions of the two Reserve sites we will sample along depth contours. Before each sampling period we will transit along the 40-m contour with the ADCP so that we will know the pattern of water flow through the Reserves. Sampling will be done around slack high tide whenever possible. Sampling of the Reserves and adjacent area will be done during the day and at night. During each ten-day cruise we will sample each of the two Reserves at least four days each. The long time series of CaICOFI (Moser et al., 1993 and 1994) and San Onofre power plant studies (Walker et al., 1987) are very good at showing the spawning times of most fish species in the area. Based on information from the CaICOFI stations nearest the two Reserves we will have a good idea of what species should be present and spawning in the general area of the Reserves prior to our sampling.
Areal distribution: To measure total water column production of eggs we need vertical samples from the bottom to the surface throughout each of the 3 regions at the two Reserve sites. The CaIVET net sampler, used on CaICOFI cruises to measure northern anchovy and Pacific sardine egg production, does not have a wide enough net opening to collect large numbers of eggs in a vertical tow through shallow water. We will use a BONGO net (diameter 71 cm) rigged with weighted cod ends and a short cable and weight between the nets. The net will be lowered until the weight touches the bottom and then retrieved vertically to the surface. A grid will be established over the three regions at each of the Reserve sites. Four stations at three separate isobaths will be sampled a minimum of two times during the day and two times during the night. Additional samples may also be taken over sites of particular interest. Each sample will be identified by its position and time of day, within the grid. The D.S. Jordan will be more or less stationary and diver transects will be conducted from the inflatable launch during this time. The shallowest stations (<10 m) will be occupied by the Ballena. Total water column measurements are required for valid measures of egg production, but we would also like to see if there are fine scale patterns as well. It is possible that the above sampling scheme will show areas of egg concentration and species shifts over different substrates and locations, but the time required to make a series of measurements throughout the grid will erase the temporal component.
The newly developed egg pump which has been mounted on the D.S. Jordan (Figure 4) samples only at one depth (presently 3 m), but it samples continuously at a rate of 640 l/min while the ship is underway. This sampling method has already been used to detect and measure small-scale patchiness of northern anchovy and Pacific sardine eggs on a recent cruise (Dotson et al., in prep.). The egg pump will be used to sample continuously along three depth contours in each of the three regions of the two Reserves. Each of the 9 transects per Reserve (north, south and in the Reserve at three depths) will consist of about 1 nautical mile parallel to the coast. Along each transect the egg pump will produce a continuous record with the eggs collected from the trap at 10 min intervals. We will also try to use the egg pump while tracking the current drifters to measure development and hatch within a water parcel as it drifts from the reserve.
Vertical distribution: Most marine fish eggs float, but the depth of release and rate of ascent (buoyant density) varies between species, stage, and most importantly, the degree of vertical mixing (Sundby, 1983). The highest concentrations of eggs are typically found near the surface (Brewer et al., 1981), but these are not necessarily the youngest eggs which are of particular interest in this study. We will conduct a series of oblique 1-m2 MOCNESS tows to determine the vertical distribution of eggs of different species. This will be done at dawn, midday, just after sunset and at midnight.
Diel distribution: Since most fishes of interest in this study, other than the rockfishes (Sebastes), are external fertilizers there is usually a tight synchrony of mating. To successfully reproduce, most fishes have very specific times of day when new eggs are released. In some cases this is known (Lo et al., 1996; Hunter and Macewicz, 1980), and in other cases it is not. We will conduct our areal and vertical sampling programs around the clock so that we can determine the spawning times of various common species. This will be examined early on in our studies and may dictate changes in our sampling scheme so that we are more likely to encounter newly released eggs of species of interest.
ii: Sample handling: The BONGO net is a paired sampler with two identical nets side by side. One sample will immediately be washed from the net and preserved in formalin. It will be sorted and identified in the same manner as samples for the CaICOFI program. The other side of the net will be washed down and the contents placed in a 10 x 40-cm PlexiglasTM settling tube with stopcock. Salinity will be elevated to 40 ppt to speed floatation of eggs. The small surface will concentrate the buoyant eggs away from the rest of the plankton. The bottom 3/4 of the plankton sample will be drained away and preserved in ethanol so that fish larvae contained in it can be identified by genetic techniques and their otoliths read for daily age. The remaining upper 1/4 will be placed in incubation chambers. Here the eggs will be allowed to develop and hatch. Late embryos and yolk-sac larvae will be used to confirm the identification of eggs that are not easily identified as early embryos. Samples in the incubator will be examined frequently and subsamples preserved so that eventually we can make estimates of developmental rates. The temperature-dependent developmental rates will eventually be combined with our understanding of current speed and direction at the time of sampling, so that we can estimate the time and location of spawning of the eggs. Material from the incubators will also be preserved in ethanol so that the larvae can be identified by molecular techniques. 1 m2 MOCNESS samples for vertical distribution studies will be preserved in formalin with a sub-sample of eggs preserved in ethanol for identification of unknown eggs.
iii. Molecular identification of eggs and larvae: It is beyond the scope of this proposal to identify all of the presently unidentifiable eggs and larvae likely to be encountered in the Reserves by molecular techniques, but we do think we can make some significant progress toward that ultimate goal. The rockfishes (Sebastes) that we are presently studying are probably the most difficult case with over 70 species present within the CaICOFI sampling region. In our rockfish studies we can identify every species by direct sequencing but we can only apply the RFLP "bar code" approach (Figure 3; which is needed for processing large numbers of samples) to subgroups where there are only five to 10 possibilities, not seventy.
In the Reserves and surrounding nearshore habitats there should be several characteristic genera or families that contain fewer than ten species and whose species are highly associated with different near shore habitats. We will focus on the identification and quantification of eggs from these groups
One such group is the wrasses (Labridae). This group includes both the senorita (Oxyjulis californica) that is usually associated with kelp cover and is not fished, and the sheephead (Semicossyphus pulcher) that is associated with rocky bottom and kelp forests but is intensively fished by both recreational fisherman and the live-trap commercial fishery.
Another group is the basses (Serranidae) and in particular the three common Paralabrax species: the kelp bass, Paralabrax clathratus, the barred sand bass, P. nebulifer, and the spotted sand bass, P. maculatofaciatus. These species are among the few remaining abundant nearshore sportfish in Southern California. Presently their eggs and yolk-sac larvae can not be separated in field samples.
A third group is the croaker family (Sciaenidae). This group contains primarily transient species that are associated with shallow water. Some species are heavily fished while others are unexploited. Sciaenids generally inhabit sandy bottom or are generalists that occur in many habitats. The most sought after member of this group is the white seabass (Atractoscion nobilis) which is found over sand bottom and on the outer margins of kelp forests. This species is highly prized by both recreational and commercial fisherman. Presently it is depleted but recovering. The spotfin croaker, Roncador stearnsii, is common over sand bottom, while the black croaker, Cheilotrema saturnum, is more common over rocky bottom. Other members of this group which are unidentifiable as eggs include the corbina, Menticirrhus undulatus, and the queenfish, Seriphus politus. Other groups, which may be examined, are the flatfishes, blennies and gobies.
iv. Analysis and interpretation of data: The goals of this research are to provide Reserve managers and other interested parties with a description of the Reserve habitats, the fishes that inhabit them, the eggs and larvae that are being produced from Reserves, and the oceanographic conditions that determine the paths of larval export from the Reserves and recruitment to the Reserves. Although the fish fauna of the Reserves and adjacent areas may change with climatic conditions and, in the Reserves, may also benefit from recovery to a natural undisturbed state, the ichthyofaunal data provided from this research will reflect the present day characteristics of the adult spawning biomass. No other survey method will provide as complete and informative a data set on which species are present and reproductively active within the Reserves.
Our primary objective is to identify all of the ichthyoplankton and to show where they were produced. In addition, we find the possibility of developing ratios between fished and unfished species pairs (e.g. the sheephead:senorita comparison within the wrasses) within a habitat a particularly intriguing idea. The unfished species should reflect habitat quality and availability and should vary primarily with natural factors (e.g. the change in kelp habitat in warm, El Nino, conditions (Tegner et al., 1996). The fished member of the species pair should reflect anthropogenic factors, such as fishing pressure, as well as habitat quality and availability. Ideally, the Reserves will show a continued increase in the quantity of eggs from fished species that benefit from living in the Reserves in the absence of fishing.
Data analyses will include standard egg production comparisons by area, principle component analyses of recurrent groups (Moser at al., 1987) and species-habitat maps.
5. METHODS AND MATERIALS:
a. Bathymetry: L. Jacobson and collaborators at the SWFSC have been generating a detailed assessment of juvenile California halibut (Paralichthys californicus) habitat along the California coast. To do this they have used a shallow water coastal bathymetric data base compiled and maintained by the National Oceanic Data Center (NODC). They have developed a mapping program that allows visualization of bathymetric contours for regions where data is available. This mapping program is compatible with the GIS data management system (ARC/INFO) currently being installed at SWFSC. There is a substantial set of sounding data for Big Sycamore Canyon and a less extensive, but reasonably good, data set for Vandenberg. Crude maps have been generated for the available data from Vandenberg (Figure 2) and Big Sycamore Canyon. These base maps will be improved, as more sounding data becomes available. The accuracy of interpolation between points, particularly over rocky substrate, is in doubt and will be confirmed by diver transects and small boat surveys (GPS and depth sounder aboard the inflatable launch of the D.S. Jordan) during the course of this research.
b. Substrate Mapping: The bathymetric charts described above will be the basis for substrate and habitat mapping. We will swim offshore-onshore transects beneath the inflatable launch, which will be equipped with a GPS system. In this way we can precisely but inexpensively map habitat type and location. Habitat and substrate information will be overlain on the bathymetric chart via a GIS data format.
c. Current Profiling: The SIO Center for Coastal Studies has agreed to supply us with three GPS drifters and a base station that can be used aboard the D.S. Jordan. R. Lynn and other members of the Fisheries Oceanography Group at SWFSC will have primary responsibility for data collection from the ADCP and the GPS drifters during our cruises.
d. Diver Surveys and Live Trapping: Fish fauna in the Reserves will be assessed via the same GPS-directed offshore-onshore transects used to map substrate and habitat. Fish traps will also be deployed along the same transects. Traps will be left in place and monitored by divers. R. Vetter and five others at SWFSC are certified NOAA divers available for swimming GPS directed transects. Fish will be released from traps and traps rebaited in situ to minimize disturbance of the fish fauna within the Reserve. It is beyond the scope of this proposal, which focuses on ichthyoplankton, to quantitatively assess the fish fauna of the Reserves. Our fish surveys are designed to obtain fish species lists for the Reserves and adjacent areas.
e. Bongo Sampling: The bongo sampler is a well-known and thoroughly documented plankton sampler. The bongo, equipped with 0.505-mm mesh nets, will provide the quantitative, integrated water-column samples of fish eggs and larvae required for production estimates (one side of the paired sampler), and fish eggs for rearing and for use with molecular identification techniques (the other side of the sampler).
f. MOCNESS Sampling: The MOCNESS sampling system (Wiebe et al., 1985) is ideal for collecting discrete samples of ichthyoplankton. The 1-m2 net with 0.505-mm mesh is good for collection of eggs and larvae and the 10-m2 net with 3-mm mesh has been demonstrated to be as good a collector of large larvae and young juveniles which are not very susceptible to most traditional samplers. In this study the MOCNESS would be used to collect samples at 15-min increments at a forward speed of about 1-knot. The net would be deployed off the side of the D.S. Jordan so that it could sample in shallow waters undisturbed by the ship's wake.
g. Auriga Sampling: The Auriga net is an epibenthic sampler that has been used extensively in nearshore, soft-bottom, ichthyoplankton sampling off the southern California coast during the past two decades, but is little known elsewhere. The Auriga is a large, wheeled sampler well suited for use in the soft-bottom epibenthic stratum, where most other towed samplers cannot be used. This sampler will be equipped with a 0.505-mm mesh net and towed at ca. 1 m/s over a distance of 400 m.
h. Egg Pump Sampling: The egg pump is a relatively new sampling device that should be ideal for the purposes of this study (Checkley et al., submitted). As presently configured on the D.S. Jordan the opening is 3-m below the surface and collects water at a rate of 640 l/min. Fine-scale horizontal distribution data will be obtained using the egg pump.
i. Egg Incubation: Samples will be placed in upright glass cylinders of 15-cm by 6-cm that have a NitexTM bottom secured with an "O" ring. We have used these tubes repeatedly in the course of UV exposure experiments with good egg and larval survival. This design retains eggs and larvae but permits free exchange of seawater with the bath. These cylinders (about the size of an aluminum soda can) are placed in racks in a 10001 Living StreamTM temperature-controlled live tank. Samples will be preserved initially and at different intervals up to and through hatch. The hatched larvae in most cases will be used to identify the eggs. Eggs or larvae not identifiable by visual means will be identified by molecular means.
j. Visual Identification: Late stage embryos and yolk-sac stage larvae will be identified using traditional morphological and pigmentation characters (e.g. Moser (ed.), 1996) whenever possible. We anticipate that the majority will be identifiable by these means, but are certain that not all eggs and larvae will be.
k. Molecular Identification: Many eggs and early larvae simply do not possess reliable visual characters that allow identification. The basic rational for genetic identification is that adult specimens can be used to determine a species-specific genetic characteristic that can then be used to identify the eggs or larvae of that species. Many investigators have independently arrived at variations on this basic approach to address a range of long-standing problems in larval recruitment. Our studies to date have been with the rockfishes (Sebastes) and have focused on two approaches that rely on sequence difference in the mitochondrial cytochrome b gene. The first approach, described in the background section, uses sequence differences to develop species-specific RFLP banding patterns (Either et al., 1995; Taylor, 1997). Another approach developed in our laboratory relies on differential PCR amplification of primer pairs that differ in sequence at the first base. If two fishes are known to differ (from the adult sequence) then one will be amplified and the other will have a poor match between the primer and the DNA extracted from the larvae. This approach works well when two species differ by only a single base pair and has been applied successfully to the separation of rockfishes within the subgenus Sebastomus (Rocha-Olivares, 1996).
The main objective of this study will be to develop identification systems for the taxonomic groups considered in this proposal. In addition to the rockfishes, this would include the wrasses, seabasses, flatfishes, and croakers. This would not be an exhaustive phylogenetic consideration of any of these groups. Rather we would include only those species known to spawn in the area of the Reserves.
Adult specimens would be sequenced using our ABI automated sequencer. Five samples of each species will be sequenced to be reasonably sure that we have species-specific invariant sites. There are many potential genes and approaches that can be used, but based on our experiences with the scorpaenid fishes we are reasonably confident that the cytochrome b gene contains the right amount of variation for separating species without a lot of within-species differences to confuse the analysis. Since each major taxon contains only a few possible choices we are hopeful that the RFLP approach can be used as the basis of identification.
6. NEED FOR. THE PROJECT AND POTENTIAL USE OF RESULTS:
Marine Reserves are becoming an increasingly common solution to address problems of overexploitation and degradation of coastal habitat in California. Often the placement of reserves, and the size of reserves, is based on political rather than scientific considerations. While small reserves may benefit small sedentary invertebrates, the potential for marine reserves to contribute to the harvestable marine resources of California depends to a large degree on the size of the reserve and its capacity to retain and protect spawning adults. The success of increased larval production in supplementing areas that are fished will also depend on the location of the reserve and the pattern of larval drift. Thus siting of reserves is also an important consideration. Despite the public appeal of marine reserves, if they are to be more than aquatic zoos for divers we must prove that reserves actually function as larval source areas. This is very difficult, but the combination of new and traditional techniques presented here is an exciting approach to this difficult problem.
1. This proposal will provide support for one graduate student (C. Taylor) to complete the research portion of a doctoral dissertation on the molecular identification of eggs, and the fine-scale vertical, areal, and diel production of eggs from marine reserves.
2. It will provide managers of the Vandenberg and Big Sycamore Canyon Reserves with maps of bathymetry, habitat maps, current patterns, and a list of fish fauna of the Reserves.
3. It will provide a objective and scientific evaluation of the contribution of the two Reserves in terms of egg production and "reseeding" of areas outside the Reserves.
4. It will provide advice for managers considering the placement, and minimum size requirements, of future marine reserves.
7. DISSEMINATION OF RESULTS: Final dissemination of results will be through published peer-reviewed journals that specialize in fisheries and oceanography. We also propose to host a symposium dealing with reserves as part of the annual CaICOFI meeting. This could be in two or three years when either Cal. Fish and Game or SWFSC is responsible for organizing the symposium. This format would provide an opportunity for all MERRP researchers to get together and summarize their results. The papers from the symposium would be published in the CalCOFI Reports.
8. BUDGET JUSTIFICATION:
1. Graduate student support: The support of a graduate student is vital to the success of the project. The student (C. Taylor) has developed the methods originally proposed by Dr. Blaise Eitner and has shown that they are reliable and accurate for rockfishes. Data from the icthyofaunal surveys and molecular larval identification will form the basis of her Ph.D. research.
2. Salary for SWFSC employees: Some salary support is required for all DOC-NOAA contracts for administrative time as well as the time of scientists and technicians. This project is of great benefit to the statutory and research obligations of NOAA, so base funding can be devoted to cover much of the manpower requirements.
3. GPS-current drifters: These drifters permit continuous tracking over short distances and are recoverable. The drifters can be placed in a specific location (e.g. a kelp bed) to provide likely drift patterns of eggs. SWFSC does not possess this type of drifter and they are vital to the success of the project.
4. MOCNESS nets: Nets wear and break. This project should be self-supporting with regard to the expendable portion of the samplers. Repair of the net frames, flow meter, computer control, and conducting cable are the responsibility of the SWFSC.
5. Auriga net: SWFSC does not possess an Auriga net. Although an Auriga net is not absolutely essential to this project, its use would greatly enhance the value of the final product.
6.Supplies for Genetic Analyses: The SWFSC has a state-of-the-art molecular biology laboratory under the direction of R. Vetter. We have an ABI Autosequencer, two 48 well Perkin-Elmer PCR machines, and a fully equipped laboratory for storage, extraction, and analysis of DNA sequences. The automated sequencer is more cost effective, in terms of supplies, than manual sequencing but supplies are still expensive. The primary costs are for the reagents needed for PCR amplification, and for fluorescent end labeling of the sequencing reactions. The supply budget includes the supplies needed to initially sequence the adult (positively identified) specimens for development of species-specific RFLP patterns, as well as for the amplification and RFLP analyses of eggs from the field samples.
Ahlstrom, E.H. 1965. A report on evaluation of fish resources of the Pt. Arguello area Part I: Fish larvae of the Pt. Arguello area. Agreement AT(49-7)-2428 for the Div. of Biology and Medicine, U.S. Atomic Energy Comm. 118 pp.
Brewer, G.D., Lavenberg, R.J., and McGowen, G.E. 1981. Abundance and vertical distribution of fish eggs and larvae in the southern California bight: June and October 1978. Rapp. P.-v. Reun Cons. int. Explor. Mer 178: 165-167.
Checkley, D.M., Jr., Ortner, P.B., Settle, L.R., and Cummings, S.R. A continuous underway fish egg sampler. Submitted to Fisheries Oceanography.
Dotson, R., Griffith, D., & Checkley, D.M., Jr. (in prep.) Fine-scale horizontal distribution of sardine eggs sampled by an onboard pump system.
Eitner,.B.J., Kimbrell, C.A., and Vetter, R 1995. The systematics and larval identification of selected species of rockfishes (Sebastes spp.) using molecular analysis of the cytochrome b gene. Contributed presentation, S. Cal. Acad. Sci. Annual Meeting, Fullerton, CA, May 5-6, 1995 and ASIH Annual Meeting, Edmonton, Canada, June 15-19, 1995.
Hanan, D., Mullin, M., and Tillman, M. 1996. Review of some California fisheries for 1995. CalCOFI Rep. 37: 7-21.
Hendershott, M.C., and Winant, C.D. 1996. Surface circulation in the Santa Barbara Channel. Oceanography 9: 114-121.
Hunter, J.R. 1984. Synopsis of culture methods for marine fish larvae. In: Ontogeny and Systematics of Fishes. Moser, H.G., Richards, W.J., Cohen, D.M., Fahay, M.P., Kendall, A.W., Jr., and Richardson, S.L. (eds.). Spec. Pub. 1, ASIH.
Hunter, J.R., and Macewicz, B.J. 1980. Sexual maturity, batch fecundity, spawning frequency, and temporal pattern of spawning for the northern anchovy, Engraulis mordax, during the 1979 spawning season. CalCOFI Rep. 21: 139-149.
Knudson, T. and Vogel, N. 1996. Pacific Blues. In: Sacramento Bee, 12/22/9612/26/96.
Lo, N.C.H., Green Ruiz, Y.A., Cervantes, M.J., Moser, H.G., and Lynn, R.J. 1996. Pacific sardine egg production and spawning biomass. CalCOFI Rep. 37: 160174.
Moser, H.G., Charter, R.L., Smith, P.E., Ambrose, D.A., Charter, S.R., Meyer, C.A., Sandknop, E.M., and Watson, W. 1993. Distributional atlas of fish larvae and eggs in the California Current region: Taxa with 1000 or more total larvae, 1951 through 1984. CalCOFI Atlas 31.
Moser, H.G., Charter, R.L., Smith, P.E., Ambrose, D.A., Charter, S.R., Meyer, C.A., Sandknop, E.M., and Watson, W. 1994. Distributional atlas of fish larvae and eggs in the California Current region: Taxa with less than 1000 total larvae, 1951 through 1984. CalCOFI Atlas 32.
Moser, H.G. 1996. Introduction. In: The early stages of fishes in the California Current region. Moser, H.G. (ed.). CalCOFI Atlas 33. pp. 1-72.
Moser, H.G. (ed.). 1996. The early stages of fishes in the California Current region. CalCOFI Atlas 33.
Moser, H.G., Smith, P.E., and Eber, L.E. 1987. Larval fish assemblages in the California Current region, 1954-1960, a period of dynamic environmental change. CalCOFI Rep. 28: 97-127.
Rocha-Olivares, A., and Vetter, R. 1995. Molecular systematics of rockfish species (Sebastes spp.) of the subgenus Sebastomus using mitochondrial DNA sequences. Contributed presentation, ASIH Annual Meeting, Edmonton, Canada, June 15-19, 1995.
Rocha-Olivares, A. 1996. A novel method for species identification of the early-life stages of Sebastomus rockfishes (Sebastes), Scorpaenidae. Contributed poster ASIF Annual Meeting, New Orleans, LA, June 1996.
Rowley, R.J. 1994. Case studies and reviews: Marine reserves in fisheries management. Aquatic Conservation: Marine and Freshwater Ecosystems 4:233-254.
Sundby, S. 1983. A one dimensional model for the vertical distribution of pelagic fish eggs in the mixed layer. Deep Sea Res. 30: 645-661.
Taylor, C. 1997. Larval identification of rockfishes (genus Sebastes) using mitochondrial DNA techniques. WSN contributed paper, La Paz, Mexico, Jan 6-10, 1997.
Tegner, M.J., Basch, L.V., and Dayton, P.K. 1996. Near extinction of an exploited marine invertebrate. TREE 11: 278-280.
Tegner, M.J., Dayton, P.K., Edwards, P.B., and Riser, K.L. 1996. Is there evidence for long-term climatic change in southern California kelp forests? CalCOFI Rep. 37: 111-126.
Walker, H.J., Jr., Watson, W., and Barnett, A.M. 1987. Seasonal occurrence of larval fishes in the nearshore southern California bight off San Onofre, California. Estuarine, Coastal Shelf Sci. 25: 91-1109.
Wiebe, P.H., Morton; A.W., Bradley, A.M., Backus, R.H., Craddock, J.E., Barber, V., Cowles, T.J., and Flierl, G.R. 1985. New developments in the MOCNESS, an apparatus for sampling zooplankton and micronekton. Mar. Biol. 87: 313-323.