Table of Contents -- < Backward -- Forward >
Rockfish dominate the marine recreational fishery catch in northern and central California, comprising about half the catch. Of the 62 rockfish species reported from the Pacific coast of North America (Robins et al. 1991), 57 are found off California (Miller and Lea 1972; Chen 1975, 1986).1 Of the 57 California species, 29 (51%) were encountered in the 1958-61 survey (Miller and Gotshall 1965); 43 (75%) were encountered in the 1980-86 MRFSS creel survey (36 in northern California and 39 in central California) (Appendix A).
All rockfish are viviparous and fertilization is internal (Boehlert and Yoklavich 1984). After egg developement and hatching within the ovary, larvae are extruded into the water column. Rockfish fecundity averages about 1.1 million eggs per female (Haldorson and Love 1991). The larval and juvenile stages are pelagic for several months to a year (Alstrom 1961; Miller and Geibel 1973; Boehlert 1977) and are routinely found in samples from the California Current (Ahlstrom et al. 1978; Lenarz et al. 1991; Adams 1992a). After the pelagic phase, many species use nearshore rocky reefs, kelp beds, or bays as juvenile nursery grounds (Miller and Geibel 1973; Feder et al. 1974; Gotshall et al. 1980; Gaines and Roughgarden 1987). True to their name, adult rockfish are typically associated with high-relief rocky substrate; however, some semipelagic schooling species are occasionally found over sand (Miller and Gotshall 1965; Haldorson and Love 1991).
Despite their high fecundity, annual recruitment of many rockfish species fluctuates widely. Strong year-classes dominate southern California sport catches of vermilion rockfish, blue rockfish, chilipepper, and bocaccio (Ally et al. 1991; Rogers and Bence 1992; Bence and Rogers 1993). California commercial catches of bocaccio and chilipepper exhibit dominant cohorts synchronous with other species and related to environmental changes such as ENSO (Hollowed et al. 1987).
The depth distributions of rockfish range from surface intertidal and offshore zones to waters deeper than 1000 m (Miller and Gotshall 1965; Miller and Lea 1972; Hart 1973; Miller and Geibel 1973). Miller and Gotshall (1965) described three main depth distributions and habitats for marine fish based on where they were commonly caught by the northern and central California sport fishery: kelp bed (0-46 m), shallow reef (0-46 m), and deep reef (46-107 m). Blue rockfish and olive rockfish were taken in all three habitats. Yellowtail rockfish, canary rockfish, and vermilion rockfish were found in shallow reef and deep reef habitats. Black rockfish and gopher rockfish were ascribed to kelp bed and shallow reef habitats. Brown rockfish were ascribed only to shallow reef habitat. Bocaccio, chilipepper, greenspotted rockfish, widow rockfish, and greenstriped rockfish were found mostly at deep reef habitat. Bocaccio, chilipepper, yellowtail rockfish, and widow rockfish were also described as found occasionally over deep (> 67 m) sandy habitats.
Miller and Geibel (1973) described 73 m as a transitional depth for northern and central California species composition. Blue rockfish, olive rockfish, kelp rockfish, black rockfish, and brown rockfish were described as principally taken in depths less than 73 m. Yellowtail rockfish, widow rockfish, bocaccio, chilipepper, greenspotted rockfish, rosy rockfish, and starry rockfish were ascribed to depths greater than 73 m.
To focus our analysis of the fishery, we categorized our top 16 rockfish species by the depth range where the fish are usually caught. Our categories are shallow (< 73 m), deep (> 73 m), and all depths. The depth groupings were adapted from the 73-m zonation of Miller and Geibel (1973); however, we added greenstriped rockfish to the deep-water species, gopher rockfish to the shallow-water species, and removed bocaccio from the deep-water species. Greenstriped rockfish and gopher rockfish were not included in the Miller and Geibel (1973) zonation. Greenstriped rockfish are found in depths of 61-402 m while gopher rockfish are found in depths less than 55 m (Miller and Lea 1972; Eschmeyer et al. 1983). We categorized bocaccio as an all-depths species because they were frequently caught from piers in 1980-86. We categorized canary rockfish, copper rockfish, and vermilion rockfish as all-depths species based on published habitat descriptions and depth ranges (Miller and Gotshall 1965; Miller and Lea 1972; Eschmeyer et al. 1983).
Species composition data from the 1958-61 and 1981-86 surveys support our depth groupings (Table 4). Virtually all the catch of the seven deep-water species was taken by PRBs or CPFVs. Other sources confirm that each of those species is found in waters deeper than 73 m, with reported maximum depths ranging from 128 m (rosy rockfish) to 402 m (greenstriped rockfish) (Miller and Lea 1972; Eschmeyer et al. 1983). The five shallow-water rockfishes were taken more frequently from the nonboat modes (3.5% and 5.8% in 1981-86 and 1958-61 respectively) than were the deep-water rockfishes. Two shallow-water species, brown rockfish and gopher rockfish, are reported only in waters of less than 55 m (Miller and Lea 1972). Blue rockfish and black rockfish range from the surface to 91 m, while olive rockfish range from the surface to 122 m (Miller and Lea 1972).
The percent by number of total catch in the 1980-86 MRFSS creel survey provides a quantitative assay of yearly and spatial variation in availability of each of our top 16 rockfish species. The catch data are from over 230,000 angler bags from 12 districts in northern Oregon and California. We classified the 16 rockfish species into three categories of latitudinal distribution: 1) northern, 2) southern, and 3) coastwide. Northern species were those found mainly north of Point Conception (black rockfish, canary rockfish, widow rockfish, and yellowtail rockfish). Southern species were those found mainly south of the Oregon-California border (rosy rockfish and brown rockfish) or those found mainly south of Cape Mendocino (olive rockfish, greenspotted rockfish, starry rockfish, gopher rockfish, and chilipepper). The coastwide species were those that were found throughout our study area (blue rockfish, vermilion rockfish, bocaccio, copper rockfish, and greenstriped rockfish).
Black rockfish were most available in Oregon and northern California, with no significant catch south of Point Conception (Figure 18). Black rockfish availability was remarkably high off central Oregon where they comprised up to 41.4% of the catch. The highest consistent availability occurred in Del Norte/Humboldt where contribution to the catch ranged from 14.8% to 31.2%. South of Point Conception the only occurrence of black rockfish was off Los Angeles County in 1981, which approximates the southernmost range of the species (Eschmeyer et al. 1983).
Canary rockfish showed a trend of higher catch to the north from Santa Barbara/Ventura northward through Oregon (Figure 18). From Los Angeles to San Diego, canary rockfish were found only infrequently and in low concentrations. The highest consistent contribution to the catch occurred off Mendocino/Sonoma (3.4% to 11.1%). The contribution to the catch off Oregon ranged from 0.03% to 12.9%. Canary rockfish range from Baja California to Alaska (Miller and Lea 1972; Hart 1973). Dark et al. (1983) estimated biomass of major rockfish species in 1980 by four International North Pacific Fisheries Commission (INPFC) areas: Vancouver, Columbia, Eureka (Cape Blanco to Cape Mendocino), and Monterey (Cape Mendocino to Monterey Bay). Percentages of total canary rockfish stock biomass were greatest in the northern INPFC areas, with the Vancouver area having 38%, Columbia 41%, Eureka 18%, and Monterey 3%.
Widow rockfish catch distribution was patchy both spatially and temporally. Patchiness was most pronounced from Del Norte/Humboldt through Oregon (0% to 2.9%) and south of Santa Barbara/Ventura (0% to 0.75%) (Figure 18). Widow rockfish range from southeastern Alaska to northern Baja California (Eschmeyer et al. 1983).
Yellowtail rockfish were taken mainly north of Los Angeles County (Figure 18). The highest and most consistent percentage of catch was in Mendocino/Sonoma (5.9% to 14%). Catch contributions off northern Oregon were occasionally high, but fluctuated greatly (0% to 11.3%). Such fluctuation may be a reflection of the variable year-class strength for the species reported by Tagart (1991). Yellowtail rockfish range from San Diego to the Gulf of Alaska (Miller and Lea 1972; Hart 1973). As with canary rockfish, estimates of percentage of total stock biomass in 1980 were greatest in the northern INPFC areas, with the Vancouver area contributing 46%, Columbia 47%, Eureka 5%, and Monterey 2% (Dark et al. 1983).
Brown rockfish had a catch distribution with a northern limit at the Oregon-California border, availability throughout California, and relatively low fluctuation in availability between years (Figure 19). The greatest availability was in the San Francisco district where catch by number ranged from 1.7% to 6.6%. Catches south of Point Conception were low, ranging from 0.04% to 1.0%. The reported range of brown rockfish extends from southeast Alaska to central Baja California (Miller and Lea 1972; Hart 1973; Eschmeyer et al 1983), which encompasses our observed range of catch.
Chilipepper were caught only south of the Del Norte/Humboldt district. Contribution to catch fluctuated widely, but was most consistent in Santa Cruz/Monterey and in Los Angeles through Orange counties (Figure 19). The highest contribution to catch (16.9%) occurred in 1986 in Mendocino/Sonoma, and followed a period of low contribution (0% to 0.02%) from 1980 through 1984. Variable year-class strength may be a factor, as discussed in the rockfish length-frequency analysis section. Chilipepper range from Baja California to Vancouver Island (Miller and Lea 1972; Hart 1973). Estimates of percentage of total stock biomass in 1980 were greatest in the southern INPFC areas, with Vancouver having 0%, Columbia 0.1%, Eureka 6%, and Monterey 93.6% (Dark et al. 1983).
Gopher rockfish had a northern boundary of take near Cape Mendocino; the only occurrence north of there was in 1984 (0.05% in southern Oregon) (Figure 19). The distribution of gopher rockfish availability was relatively patchy over time and space. A band of high contribution to the catch (4.0% to 10.7%) occurred off San Luis Obispo. South of Point Conception, contribution to the catch ranged from 0% to 1.6%. Gopher rockfish were rarely caught in northern California, where contribution to the catch ranged from 0% to 0.2%. Eschmeyer et al. (1983) described the range of gopher rockfish as extending from San Roque, Baja California to Eureka. The occurrence in southern Oregon extends beyond that range, but our data cannot document a range extension since fish were not kept for taxonomic verification.
Greenspotted rockfish had a fairly distinct northern boundary of catch near Cape Mendocino; the only occurrence north of there was in 1985 (0.04% in Del Norte/Humboldt) (Figure 19). Like gopher rockfish and chilipepper, greenspotted rockfish had a patchy temporal and latitudinal distribution; patchy temporal availability suggests variable recruitment. Catches in 1985 and 1986 in Mendocino through Monterey counties increased from previous years. Greenspotted rockfish were most available in Santa Barbara/Ventura where contribution to catch ranged from 2.1% in 1980 to 9.8% in 1984. Greenspotted rockfish range from central Baja California to Washington (Miller and Lea 1972; Hart 1973).
Olive rockfish catch distribution (Figure 20) was similar to that of greenspotted rockfish, starry rockfish, gopher rockfish, and chilipepper (Figures 19 and 20). The northern boundary of catch was near Cape Mendocino. Spatial and temporal distribution of olive rockfish was less patchy than that of greenspotted rockfish, gopher rockfish, or chilipepper. The highest availability was from Santa Cruz/Monterey through Santa Barbara/Ventura, where contribution to the catch ranged from 0.7% to 5.9%. Olive rockfish range from San Benito Island, Baja California to Redding Rock, Del Norte County (Eschmeyer et al. 1983). The one rare occurrence (0.08%) in Del Norte/Humboldt falls within that range.
Rosy rockfish were caught mainly south of the Oregon-California border (Figure 20). Fluctuation in availability between years was low. Catch distribution was centered in San Luis Obispo, where a fairly constant portion (1.6% to 2.8%) of catch was maintained. Rosy rockfish range from Turtle Bay, Baja California to Puget Sound (Miller and Lea 1972; Hart 1973).
Starry rockfish had a fairly distinct northern boundary of catch near Cape Mendocino; the only occurrence north of there was in 1982 (0.03% in Del Norte/Humboldt) (Figure 20). That boundary is well north of San Francisco, the northern range reported by Miller and Lea (1972). (Our data cannot be used to document a range extension as the fish taken were not preserved for taxonomic verification.) The highest availability was in San Luis Obispo through Santa Barbara/Ventura where contribution to catch ranged from 0.7% to 3.5%.
Vermilion rockfish were broadly distributed at low percentages of total catch (0% to 8.0%) increasing southward from central Oregon to Los Angeles County (Figure 20). The highest consistent contribution to catch occurred in Santa Barbara/Ventura (0.6% in 1982 to 4.0% in 1984). The increase in catch observed in 1985 and 1986 from Los Angles through San Diego counties may reflect the strong 1982 year-class reported by Ally et al. (1991). Vermilion rockfish range from Baja California to central Vancouver Island (Hart 1973).
Blue rockfish showed the widest spatial distribution of our top 16 rockfish species. Contribution to catch was high in central Oregon, Mendocino/Sonoma, Santa Cruz/Monterey and San Luis Obispo (Figure 21). The highest contribution was in San Luis Obispo in 1982 (31%); contribution there declined to 8.5% in 1985. The lowest contributions (0% to 3.0%) were from Los Angeles through San Diego counties. The large fluctuations in availability are discussed in the rockfish length-frequency analysis section and the management implications section. Blue rockfish range well beyond our study area, from Santo Tomas, Baja California to the Bering Sea (Miller and Lea 1972; Hart 1973; Eschmeyer et al 1983).
Bocaccio were caught throughout Oregon and California, but were caught most consistently and in higher percentages south of Del Norte/Humboldt (Figure 21). Contributions to the catch were highest from Santa Cruz through Orange counties but fluctuated widely (0.11% to 14.1%). Percentages of catch off San Diego were low (0.12% to 1.4%) (Figure 21). Fluctuations in contribution to catch may reflect fluctuations in cohort strength reported by Bence and Rogers (1993); cohort strength is discussed in greater detail in the length-frequency analysis section. Bocaccio range from central Baja California to the Gulf of Alaska (Miller and Lea 1972; Hart 1973; Eschmeyer et al. 1983). Estimates of percentage of total 1980 stock biomass were greatest in the southern INFPC areas, with Vancouver having 3%, Columbia 11%, Eureka 15%, and Monterey 71% (Dark et al. 1983). Such a distribution agrees with our findings.
Copper rockfish were caught throughout Oregon and California (Figure 21). Percentage of total catch was most consistent from the Oregon-California border through Los Angeles County, where copper rockfish were caught in all districts in 1980-86. High contributions to the catch were off Mendocino/Sonoma (1.3% to 15.1%) and off San Luis Obispo through Santa Barbara counties (1.2% to 4.3%). Concentrations were low (0% to 0.8%) throughout Oregon and in Orange County. Miller and Lea (1972) and Hart (1973) describe the distribution of copper rockfish as ranging from Monterey through the Gulf of Alaska. Eschmeyer et al. (1983) incorporate the range of the synonymized whitebelly rockfish (Chen 1975, 1986) and describe the range as extending from central Baja California to the Gulf of Alaska.
Greenstriped rockfish, although classified as coastwide, had a catch distribution similar to that of widow rockfish, with patchy spatial and yearly availability that was most pronounced in the north (Figure 21). Percent of the total catch ranged from 0% to 1.2% in Oregon through San Francisco, with lowest percentages occurring in 1980-83. In southern California, greenstriped rockfish were caught in all seven years; contribution to the catch ranged from 0.09% to 10.7%. The increase in contribution to the catch in 1984 in Oregon, San Francisco, and Santa Barbara/Ventura may reflect variation in recruitment. Greenstriped rockfish range from central Baja California to Alaska (Miller and Lea 1972; Hart 1973; Eschmeyer et al 1983).
In 1981-86 the average annual recreational catch of rockfish, excluding San Francisco Bay, was 3,431,100 fish weighing 2,447,700 kg, nearly a threefold increase over our adjusted 1958-61 estimate of 1,286,800 fish weighing 1,051,000 kg. Rockfish comprised 35.3% by number and 43.0% by weight of the total 1958-61 sport catch, and 54.7% by number and 54.1% by weight of the total 1981-86 sport catch (Table 5). Thus the percentage (by number) of rockfish in the total catch increased substantially between the two surveys, but average weight decreased from 0.82 kg/fish to 0.71 kg/fish.
The catch of rockfish changed differently among the various fishing modes (Table 4). The CPFV rockfish catch doubled from 1,050,000 fish to 2,087,000 fish. The proportion of total rockfish in the CPFV catch of all species did not change (85.3% in 1958-61 and 85.8% in 1981-86). The dramatic increase in PRB effort between the two surveys (Table 1) resulted in a sevenfold increase in PRB rockfish catch from 169,000 fish to 1,217,000 fish, while the proportion of rockfish in the total PRB catch of all species increased from 50.0% to 66.5%. The catch of rockfish by divers doubled from 10,000 fish to 20,000 fish, while the proportion of rockfish in the total diver catch rose from 45.3% to 50.3%. The catch of rockfish from shore rose slightly from 35,000 to 37,000 fish; the rockfish portion of total shore catch of all species also rose slightly from 3.4% to 4.4%. Pier and dock rockfish catch increased sixfold from 12,000 (16th rank) to 71,000 (seventh rank); the proportion of rockfish in the total pier catch of all species rose from 1.2% to 7.3%.
Average weight per fish during 1981-86 varied among the fishing modes (Table 5). The CPFV mode took the largest rockfish (0.76 kg/fish), followed by the PRB mode (0.68 kg/fish), and divers (0.65 kg/fish). Among the shore-based modes, beach and bank took the largest rockfish (0.65 kg/fish), followed by jetty and breakwater (0.43 kg/fish), and pier and dock (0.11 kg/fish).
Examination of our top 16 rockfishes reveals several general trends: 1) a decrease in size among the shallow-water species with no consistent decline among the deep-water species; 2) an increase in the PRB take of all-depths rockfishes relative to the CPFV take; and 3) an increase in the proportion of deep-water species to shallow-water species in the PRB and CPFV catches (Tables 4 and 5).
The decline in average weight per rockfish from 1958-61 to 1981-86 resulted primarily from declines among species of the shallow-water rockfish and all-depths group with the deep-water rockfish generally showing no consistent trend (Table 5). The average weight of shallow-water rockfish declined from 0.70 kg/fish to 0.60 kg/fish with all five species declining. That decline did not result from the increased share of rockfish taken by PRBs (19.6% to 50.8%) since the average weights of shallow-water rockfish taken by PRBs (0.62 kg/fish) and CPFVs (0.60 kg/fish) were nearly identical in 1981-86.
Average weight of the all-depths group declined from 1.29 kg/fish to 1.01 kg/fish due mainly to decreases in average weight of bocaccio (1.55 to 0.97 kg/fish), canary rockfish (0.75 to 0.71 kg/fish), and vermilion rockfish (1.78 to 1.51 kg/fish) (Table 5). The decrease in size of bocaccio can be attributed in part to recruitment of young fish, probably the strong 1984 year-class (Bence and Rogers 1993). An annual average of 44,000 juvenile bocaccio, weighing 0.05 kg/fish, were taken from pier and dock mode in 1981-86. Those fish were 24% of the bocaccio taken from all modes in 1981-86, compared to 4000 fish or 6% in 1958-61. Average weight of the deep-water rockfish did not change (0.70 to 0.71 kg/fish).
The increased take of rockfish by PRBs relative to CPFVs (Tables 4 and 5) from 1958 to 1986 resulted from rapid growth in PRB effort (Table 1). PRB catch by number of the deep-water rockfish increased from 2.6% to 10.2% while the share taken by CPFVs declined from 97.4% to 89.7%. PRB catch of the shallow-water rockfish increased from 19.6% to 50.8% while CPFV catch declined from 74.6% to 45.7%. PRB catch of the all-depths group increased from 15.0% to 28.8%, while the CPFV catch declined from 82.9% to 60.3%.
While the proportion of rockfish taken by PRBs has increased relative to CPFVs, the proportion of rockfish taken by PRBs has shown a greater increase for the deep-water rockfish than for the shallow-water rockfish. Among PRBs, catch of all seven species of the deep-water rockfish increased relative to other species taken. Together, they produced an increase from 5.5% of the catch in 1958-61 to 9.6% by 1981-86 (Table 4). Shallow-water species showed mixed results with blue rockfish, the major shallow-water species, declining from 43.0% to 30.6% and the other four species increasing. On average, shallow-water species increased from 73.3% to 77.0%. The all-depths species declined from 21.2% to 12.9%.
The CPFV catch also showed a proportional increase in deep-water rockfish and a decline in shallow-water rockfish between the 1958-61 and 1981-86 surveys. Increased percentages of chilipepper (0.3% to 11.8%), greenspotted rockfish (2.2% to 3.4%), widow rockfish (3.2% to 3.5%), and greenstriped rockfish (0.4% to 1.5%) resulted in an overall increase from 34.2% to 46.7% for the deep-water rockfish (Table 4). Several deep-water species showed minor decreases including yellowtail rockfish (22.1% to 21.4%), rosy rockfish (3.5% to 3.0%), and starry rockfish (2.6% to 2.0%). The CPFV catch of all shallow water species except brown rockfish declined, including blue rockfish (32.7% to 28.2%), black rockfish (2.9% to 2.0%), gopher rockfish (2.8% to 1.5%), and olive rockfish (6.5% to 3.2%). Brown rockfish showed an increase (1.3% to 3.2%) that was mirrored in all other modes except spear, suggesting a general increase in availability or increased targeting of the species. Bocaccio, an all-depths species, increased from 5.8% to 7.3% of the catch (Table 4), averaging 1.28 kg in 1981-86 (Table 5).
In 1981-86, rockfish comprised more than half the recreational catch by number and weight throughout northern and central California except in San Francisco Bay (Table 6). Percent by number ranged from 7.5% in San Francisco Bay to 71.9% in San Luis Obispo. Percent by weight ranged from 1.7% in San Francisco Bay to 72.6% in Mendocino/Sonoma.
Santa Cruz/Monterey had the largest share (about one-third) of the total rockfish catch (35.2% by number and 30.6% by weight) (Table 6). San Luis Obispo ranked second by number (25.3%) and was fourth by weight (17.3%). The San Francisco ocean and bay districts ranked third by number (18.4%) and second by weight (20.6%). Mendocino/Sonoma ranked fourth by number and third by weight, while Del Norte/Humboldt ranked fifth by number and weight. San Francisco (bay) produced only 2.4% by number and 0.9% by weight.
Del Norte/Humboldt was distinct among other districts in showing extreme dominance of a single shallow-water species, black rockfish, which had an IRI of 73.8 (Figure 22). Black rockfish occurred in 19.1% of angler bags and comprised 22.9% by number and 31.8% by weight of all fish sampled. Blue rockfish, another shallow-water species, ranked second among rockfishes with an IRI of 7.3 (1/10th that of black rockfish). Two all-depth species, canary rockfish and copper rockfish, ranked third and fourth with IRIs of 7.2 and 6.2 respectively. China rockfish, yelloweye rockfish, and grass rockfish ranked fifth, sixth, and seventh respectively; those species were not among the top 16 species in all districts. Yellowtail rockfish, the only deep-water species of significance in Del Norte/Humboldt, ranked eighth with an IRI of 2.7.
Mendocino/Sonoma showed a more diverse rockfish assemblage than Del Norte/Humboldt, with six species having IRIs greater than 10 (Figure 22). Blue rockfish ranked first with an IRI of 41.3, and had the highest percent by number (16.6%). Yellowtail rockfish ranked second with an IRI of 37.3, but had the highest percent by weight (13.9%) and percent frequency of occurrence in bags (12.8%). Canary rockfish ranked third with an IRI of 22.7. Almost half this value was due to a high frequency of occurrence in bags (10%). Black rockfish, copper rockfish, chilipepper, bocaccio, and brown rockfish ranked fourth through eighth with IRIs ranging from 17.7 to 6.9.
The San Francisco district showed diversity similar to Mendocino/Sonoma, with six species having IRIs greater than 10 (Figure 22). Yellowtail rockfish was the top ranking species with an IRI of 39.2; blue rockfish was second with an IRI of 28.2. Although blue rockfish had a higher frequency of occurrence (7.9%) than yellowtail rockfish (7.7%), the latter dominated with higher percentages by number and weight. Brown rockfish, canary rockfish, bocaccio, black rockfish, copper rockfish, and greenspotted rockfish ranked third through eighth with IRIs ranging from 17.9 to 6.4.
Santa Cruz/Monterey had four species with IRIs greater than 10 (Figure 22). Blue rockfish was first in number (12.7%), weight (9.3%), frequency of occurrence (7.0%), and IRI (29.0). Chilipepper ranked second with an IRI of 16.6, and was also second in number (6.4%) and weight (7.8%), but was seventh in frequency of occurrence (2.4%). Yellowtail rockfish, bocaccio, olive rockfish, canary rockfish, brown rockfish, and greenspotted rockfish ranked third through eighth with IRIs ranging from 16.5 to 5.3.
San Luis Obispo was similar to Santa Cruz/Monterey, with four species having IRIs greater than 10 (Figure 22). Blue rockfish and gopher rockfish, both shallow-water species, had the highest IRIs (39.9 and 16.4 respectively). As in Santa Cruz/Monterey, blue rockfish was first in number (17.0%), weight (13.0%), and frequency of occurrence (9.9%). Bocaccio (16.3) and yellowtail rockfish (15.3) ranked third and fourth. Yellowtail rockfish had a higher frequency of occurrence (6.4%) than both gopher rockfish (5.0%) and bocaccio (3.7%), but lower percentages by number and weight. Olive rockfish, copper rockfish, vermilion rockfish, and rosy rockfish ranked fifth through eighth with IRIs ranging from 9.7 to 7.5.
If San Francisco (bay) is excluded, rockfish showed a consistent cline of decrease in mean weight from Del Norte/Humboldt (1.13 kg/fish) south through San Luis Obispo (0.48 kg/fish) in 1981-86 (Table 6). San Francisco (bay) had the smallest rockfish (0.25 kg/fish). Although the same cline of lower mean weight to the south was generally evident among each of the depth groups, trends among individual species were less consistent. The smaller sizes of some species in Del Norte/Humboldt are difficult to interpret because the number of rockfish other than black rockfish was small, and because the PRB mode, which on the average takes smaller fish than the CPFV mode (Table 5), was the only boat mode adequately sampled there in 1981-86 (Figure 9).
The shallow-water rockfish was the most important depth group in 1981-86, with a total 1,547,000 fish weighing 919,000 kg landed, or 44.4% by number and 37.3% by weight of all rockfish (Table 6). Blue rockfish was the most important species, with 823,000 fish weighing 417,000 kg. Blue rockfish also ranked first by number and weight landed among all rockfishes. Another important species was black rockfish, with 277,000 fish weighing 265,000 kg landed. Black rockfish ranked third among all rockfishes in number and weight landed.
The shallow-water rockfish catch was dominated by the San Luis Obispo and Santa Cruz/Monterey districts (Table 6); those two districts accounted for 28.9% and 28.7% respectively of the total number of shallow-water rockfish taken. Blue rockfish dominated both districts in number and weight caught, with 561,000 fish weighing 238,000 kg, over half the blue rockfish landed in all northern and central California. Although Del Norte/Humboldt ranked last in catch by number, catch by weight (206,000 kg) ranked second due to catch of large black rockfish. Most black rockfish (64%) were taken in Del Norte/Humboldt.
Blue rockfish and black rockfish, the two most frequently taken shallow- water rockfishes, showed distinct north-south clines in mean weight (blue rockfish: 0.85 to 0.41 kg/fish; black rockfish: 1.08 to 0.38 kg/fish) (Table 6). Size differences for those two species are discussed further in the length-frequency analysis section. Brown rockfish and olive rockfish showed no trends. The largest gopher rockfish were in Mendocino/Sonoma (0.62 kg/fish) and the smallest were in Santa Cruz/Monterey (0.33 kg/fish).
The seven deep-water rockfishes comprised 934,000 fish weighing 659,000 kg, or 26.8% by number and 26.8% by weight of all rockfish landed (Table 6). Yellowtail rockfish and chilipepper dominated the group. Yellowtail rockfish was the second most important species in number and weight of all rockfishes, with 434,000 fish weighing 350,000 kg. Chilipepper was fourth in number and fifth in weight with 213,000 fish weighing 167,000 kg.
Santa Cruz/Monterey had the greatest deep-water rockfish catch of all districts; 431,000 fish (46.1%) weighing 285,000 kg (43.3%) (Table 6). The dominant species by weight and number was chilipepper, with 160,000 fish weighing 118,000 kg. Yellowtail rockfish was also important in Santa Cruz/Monterey (145,000 fish weighing 104,000 kg) and also in San Francisco (119,000 fish weighing 108,000 kg). Del Norte/Humboldt had the lowest number of deep-water species (11,000 fish weighing 8000 kg); most were yellowtail rockfish.
Among the deep-water rockfishes, all species except chilipepper showed a general trend of smaller fish to the south (Table 6). Rosy rockfish were slightly smaller in Santa Cruz/Monterey than in San Luis Obispo. Size differences for chilipepper and yellowtail rockfish are discussed further in the length-frequency analysis section.
The four all-depth rockfishes comprised 443,000 fish weighing 446,000 kg (Table 6). Bocaccio was the most important species, with 180,000 fish weighing 174,000 kg. Bocaccio ranked sixth in number and fourth in weight of all rockfishes. Santa Cruz/Monterey dominated the all-depths catch with 145,000 fish (32.7%) weighing 136,000 kg (30.5%). Bocaccio was the dominant species with 90,000 fish weighing 90,000 kg taken; that catch was about half the total bocaccio landings.
All species of the all-depths group, except vermilion rockfish, were smallest in San Luis Obispo (Table 6). The small average weight of bocaccio in San Luis Obispo (0.28 kg/fish) results from the catch of juveniles from piers. The mean weight cline for the other three species cannot be explained by differences in mode since most were taken by boat modes and not from shore. Size differences in bocaccio and canary rockfish are examined in greater detail in the length-frequency analysis section.
The catch of most rockfishes, especially the shallow and all-depths species, peaked between July and September (Figure 23). Among the deep-water species, the catch of yellowtail rockfish, rosy rockfish, greenspotted rockfish, and starry rockfish also peaked between July and September. The catch of brown rockfish, copper rockfish, and chilipepper peaked in June; widow rockfish peaked in February. A minor February to March peak was apparent for a number of the other species including blue rockfish, black rockfish, olive rockfish, bocaccio, yellowtail rockfish, rosy rockfish, and greenstriped rockfish. The composite of all rockfish species showed most caught from June through October with a minor increase in February.
Of our top 16 rockfishes, six (widow rockfish, bocaccio, chilipepper, yellowtail rockfish, canary rockfish, and greenspotted rockfish) were among the top 11 commercial trawl-caught species in California during 1981-86 (Pearson and Ralston 1990). The port-of-landing areas used by Pearson and Ralston (1990) approximate our coastal county districts (Figure 1). Combined sport catch for those species was a significant portion of combined sport and trawl landings in some districts during that period (Figure 24). The portion of recreational catch for all six species increased south of Mendocino/Sonoma to Santa Cruz/Monterey, where 23% of the combined landings were recreational.
The sport catch of three of the six species exceeded trawl landings in central California districts from San Francisco to San Luis Obispo (Table 7). The proportion of canary rockfish taken by the sport fishery ranged from 53.4% to 83.9%, and for greenspotted rockfish from 45.6% to 83.3%. Yellowtail rockfish were taken mainly by the sport fishery from Mendocino/Somona (52.0%) to San Luis Obispo (74.3%). In Del Norte/Humboldt most canary rockfish, greenspotted rockfish, and yellowtail rockfish were taken by trawlers; sport landings of these species comprised 4.1%, 0.4%, and 3.2%, respectively.
Bocaccio, chilipepper, and widow rockfish were taken mostly by trawlers in a decreasing proportion to the south. Recreational catches of bocaccio and chilipepper were greatest in Santa Cruz/Monterey (14.2% and 21.5%, respectively). The largest trawl catches of bocaccio and chilipepper were in Mendocino/Sonoma, where recreational catches made up 1.9% and 2.6% of the respective landings.
Widow rockfish had the largest trawl landings of the six species, averaging 4916 MT; sport landings were 38 MT (0.8%). The largest trawl catch was in Del Norte/Humboldt, averaging 2685 MT (0.03% sport), and the lowest was in San Luis Obispo, averaging 52 MT (13.1% sport).
Length data for blue rockfish spanned the years 1959 through 1986, and were the most extensive of any of our top 16 rockfishes. Prior to 1980, blue rockfish lengths were most consistently sampled from CPFVs in Santa Cruz/Monterey and San Luis Obispo. We therefore restricted our length-frequency comparisons among modes to 1980-86, and our comparisons among districts in 1960-63 to the CPFV mode. Uniform sampling among modes during 1980-86 allowed combining the CPFV and PRB data for purposes of comparing districts and analyzing recruitment patterns. Similarity in yearly recruitment patterns between Santa Cruz/Monterey and San Luis Obispo allowed us to combine CPFV and PRB data from those districts for a long-term (1959-86) analysis of recruitment, growth rate, and birth years of major cohorts.
The two-way ANOVA test found that in 1980-86 mean lengths of blue rockfish in northern California (344 mm) and central California (295 mm) were significantly different (p < 0.00001), that mean lengths of the CPFV catch (316 mm) and the PRB catch (299 mm) were significantly different (p < 0.0001), and that the interaction of area and mode was also significant (p < 0.0001). Comparison of size distributions showed that in the central districts, CPFVs took larger blue rockfish (mean length of 303 mm) than PRBs (mean length of 286 mm) (Figure 25). The distributions were significantly different (KS test, p < 0.0001). Comparison of the northern districts showed no substantial difference in mean length between the two modes; the distributions were not significantly different (KS test; p = 0.8204).
In central and northern California, 50% of male and female blue rockfish are sexually mature at 270 mm and 290 mm, respectively (Echeverria 1987). Thus during 1980-86 most of the PRB and CPFV catch in the northern districts, and about half the PRB and two-thirds the CPFV catch in the central districts, had reached the size at which 50% are sexually mature (Figure 25).
We examined length and recruitment patterns among districts in 1960-63 for the CPFV mode and in 1980-86 for the CPFV and PRB modes combined. Dominance by one or more strong year-classes in 1960-63, reported by Miller and Geibel (1973), was apparent in Santa Cruz/Monterey and in San Luis Obispo, but not to the north (Figure 26). The length-frequency modes of strong year-classes and their progression from year to year are most distinguishable in San Luis Obispo, but are also apparent in Santa Cruz/Monterey. In a selected comparison for a representative year (1962), the length-frequency distributions in Santa Cruz/Monterey and San Luis Obispo were significantly different (KS test; p < 0.0001), however both distributions were platykurtic and skewed to the right. The 1962 length-frequency distributions for San Luis Obispo, Monterey/Santa Cruz, and Mendocino/Sonoma-San Francisco were all significantly different from one another (p < 0.0001 for each comparison). The results suggest that in 1960-63, the catches in San Luis Obispo and Santa Cruz/Monterey were both dominated by strong year-classes, and the degree of dominance was less in San Mateo to Mendocino counties
In 1960-63 mean length was slightly greater in San Luis Obispo (309 mm) than in Santa Cruz/Monterey (301 mm) and much greater to the north (336 mm). Two-way ANOVA comparison by area and year showed the differences were significant by area (p < 0.0001), year (p < 0.0001), and interaction (p < 0.0001).
In 1980-86 smaller sizes and dominant year-classes were again evident in the central but not the northern districts (Figure 27). In the central districts a mode at 320 mm in 1980 progressed to 340 mm by 1983 and disappeared in 1984. Average length declined from 308 mm in 1983 to 275 mm in 1984. The distribution in 1983 was skewed to the left by the dominance of the older cohort(s) and became skewed to the right in 1984; the distributions were significantly different (KS test; p < 0.0001). In the northern districts the distributions, though based on small sample sizes, showed no dominance by strong
year-classes. The 1983 and 1984 distributions in the northern districts were not significantly different (KS test; p = 0.271). As in the 1960-63 period, mean lengths were greater in the northern districts (344 mm) than in the central districts (295 mm). ANOVA comparison by area and year showed the differences were significant by area (p < 0.00001) and year (p < 0.0001). Interaction between factors was also significant (p < 0.0001).
Mean lengths of blue rockfish taken from CPFVs showed no long-term decline from 1960 through 1986 (Figure 28). In the northern districts mean length increased from 336 mm in 1960-63 to 345 mm in 1980-86. The distributions in both periods were skewed to the right and platykurtic but significantly different (KS test; p < 0.0001). In the central districts mean length decreased from 304 mm in 1960-63, when the distribution was strongly bimodal and skewed to the left, to 291 mm in 1966-72, when the distribution was less platykurtic and skewed to the right. In 1980-86 mean length increased to 303 mm and the distribution regained a more platykurtic shape.
Those results appear to contradict our earlier observation that the mean weight of blue rockfish decreased slightly for all of northern and central California from 1958-61 (0.56 kg/fish) to 1981-86 (0.51 kg/fish)(Table 5). However, that decrease was for all modes combined, and the above comparison was for CPFVs.
Combined PRB and CPFV data from the central districts show three periods when length-frequency modes, each representing one or more strong year-classes, could be followed through time: 1959 to 1964, 1970 to 1972, and 1980 to 1983 (Figure 29). Two periods, 1967 to 1970 and 1984 to 1986, show no evidence of length-frequency modes progressing through time.
Miller and Geibel (1973), whose study spanned 1959 to 1971, interpreted lack of older year-classes between 1966 and 1971 to overfishing of localized populations. Post-1971 data suggest that environmental factors play a role in producing strong year-classes that are large enough to withstand existing levels of fishing pressure. Examination of the four major length-frequency modes (Figure 29) shows fluctuations in population growth rates under varying environmental conditions. The growth rate, as depicted by shift to the right of the length-frequency modes, was most pronounced for B, C, and A and lowest for D. Also of interest was the abrupt disappearance of D following 1983, while A showed a more gradual attenuation that could be followed for larger sizes and presumably older fish.
A major environmental change during the years of mode D was the 1982-83 ENSO, which had major impacts on numerous fish and invertebrate species (Dayton and Tegner 1989). As nearshore planktonic feeders, blue rockfish are more dependent on cycles of upwelling and downwelling than other fishes (Hobson and Chess 1988). The ENSO in the spring of 1983 showed reduced upwelling in central California (Roughgarden et al. 1988). Blue rockfish in the Monterey area were unusually emaciated during 1982 and 1983 (D. VenTresca, CDFG, pers. comm.). Bodkin et al. (1987) observed three incidents of fish mass mortality in central California in 1982-83 that corresponded to periods of high-amplitude, long-period swells; 72% of the observed mortalities were blue rockfish that averaged 283 mm in length. The depressed growth rate from 1981 to 1983 and termination of the mode following 1983 may reflect the collective impact of the ENSO with an increase in fishing pressure since the 1960s.
Ford - Walford plot analysis applied to growth intervals of modes A through D produced a Von Bertalanffy growth equation of
lt+1 = 0.888 lt + 48
where
l¥ = 427.
Our growth rates were slower than those obtained by Miller and Geibel (1973), who used both scale and tag-and-recapture methods for central California fish, and much slower than those by McClure (1982), who used surface otolith aging for Oregon fish (Figure 30). Surface aging of otoliths may have underestimated age at size. Recent work on rockfish age determination shows surface aging of otoliths to be less reliable than a "break-and-burn" technique (Kimura et al. 1979; Pearson et al. 1991). The larger sizes at age observed in Oregon for both males and females could also reflect increased growth rates in the north. There is no general trend of increased growth by latitude among rockfish (Haldorson and Love 1991). Some species of rockfish such as splitnose rockfish and olive rockfish have been found to have higher growth rates to the north while others such as canary rockfish have not (Love 1978; Boehlert and Kappenman 1980). Tag-and-recapture methods used by Miller and Geibel (1973) are considered reliable but do not account for long-term environmental effects such as ENSO. Haaker et al. (in prep.) found that growth rates decreased dramatically for tagged red abalone Haliotis rufescens during the 1982-84 ENSO when compared to the preceding four years off Santa Rosa Island in southern California.
Applying our Von Bertalanffy growth rate estimate to length-frequency modes A through D and assuming 1) each mode represents a single cohort and 2) environmental conditions were similar throughout the life span of each cohort, we estimated the following birth years: A-1946, B-1954, C-1962, and D-1968 or 1969. Our results clearly suggest blue rockfish recruitment in central California varies greatly; the significance is discussed in context with other species later in this paper. Our growth rate estimate is from a period that included depressed growth during the 1982-84 ENSO. In assigning birthdates to A and B, we assumed similar effects during the previous major ENSO of 1957-58. Modes C and D did not experience a similar event and may represent fish born in 1963 and 1970 respectively, based on growth rates from Miller and Geibel (1973) tag data.
In northern California during 1980-86, the mean length of trawl-caught yellowtail rockfish was 48 mm longer than CPFV-caught fish, which in turn was 29 mm longer than PRB-caught fish (Figure 31). The northern California trawl length-frequency distribution was monomodal while both the CPFV and PRB distributions were bimodal. The CPFV distribution was skewed to the left by a dominant mode of larger fish while the PRB distribution had a dominant mode of smaller fish. In central California, mean length of the trawl catch was similar to that of northern California, but interpretation is difficult due to small sample size (n = 272). Pearson and Ralston (1990) found similar average lengths of trawl-caught yellowtail rockfish in Morro Bay and Eureka catches, and the longest average lengths in the Bodega Bay catch. Mean lengths of the CPFV and PRB catches were respectively 68 mm and 69 mm smaller in central California than in northern California. The length-frequency distributions of CPFVs and PRBs in central California were less clearly bimodal than in northern California and were strongly skewed to a mode of smaller fish. Two-way ANOVA comparisons of mean length differences were significant by northern California versus central California (p < 0.0001), mode (p < 0.0001), and interaction (p < 0.0001). In spite of univariate similarities among modes within areas, all paired KS comparisons were significantly different. A cline of larger sizes and older ages to the north among yellowtail rockfish was first reported from a 1977 federal trawl survey which encompassed our study area and extended to northern Washington (Fraidenburg 1980).
In central and northern California, 50% of male and female yellowtail rockfish are sexually mature as 6- and 7-year olds at 350 mm and 360 mm length, respectively (Echeverria 1987). Thus, during 1980-86, approximately half the northern California PRB and CPFV catch exceeded the size at which 50% are sexually mature, while very little of the central area recreational catch reached that size (Figure 31).
Yearly length-frequency plots show modes of dominant year-classes in the recreational catch in both central and northern California in 1980-86 (Figure 32). Mean length of the northern California trawl catch decreased from 476 mm in 1980 to 436 mm in 1986; the length-frequency distribution had one major mode during those years. Age-at-size data were unavailable for the recreational fishery, but careful examination of modes suggests they may correspond in northern and central California. As with blue rockfish, the older mode (A) decreased abruptly in 1983 in central California. The following year catch shifted to a mode of younger fish (B) at 260 mm in 1984 that grew to 300 mm by 1986. In northern California the mode of older fish (A) did not decline after 1983 but persisted into 1986, attenuating more gradually. In the north the mode of younger fish (B) did not become important until 1985 and was dominant in 1986 at about 310 mm.
Tagart (1991), in a detailed stock analysis of yellowtail rockfish in the trawl fishery, reported that ages (based on otoliths) showed that a single dominant mode, comprising 1974, 1975, and 1976 cohorts, led landings during the 1980-86 period as the last of the strong 1963 cohort was fished down in the Eureka/S. Columbia INPFC area. He also described the 1980 and 1981 year-classes as major cohorts that were important but would not be part of the commercial size distribution during 1980-86. Lenarz and Echeverria (1986) described decreases in gonadal and visceral fat in yellowtail rockfish in 1983 (during the 1982-83 ENSO). As with blue rockfish, the effects of starvation brought on by the ENSO may have contributed to the decrease in a mode of larger fish in central California following 1983. Alternative explanations for our results include greater fishing pressure or availability of young recruits in central California. Greater effort in the central area could have eliminated the older cohort (A) earlier than in the north where lower fishing pressure allowed mode A to dominate catches for an additional two years. Likewise increased availability of young fish by a relatively larger recruitment event in central California could produce the same effect. The most plausible explanation, considering the parallel decline in blue rockfish, is a combination of stress from fishing pressure and ENSO on adult stocks.
Pearson and Ralston (1990) described a decline in average size among commercially taken yellowtail rockfish during 1978-88 that they attributed to fishing down a virgin stock. Such a decline was apparent in our northern trawl length data but was obscured in the recreational take by the succession of dominant modes (Figure 32). A regression comparison for northern California using ANCOVA produced slopes of decline for trawlers (-9.2 mm/year) and recreational boats (-4.7 mm/year) (Figure 32). The slopes were significantly different from 0 (p < 0.001) and from each other (p < 0.0001).
In 1980-86 lengths of black rockfish from CPFVs and PRBs from central California were narrowly distributed with a mean length of 311 mm, while those of northern California were more broadly distributed with a mean length of 402 mm (Figure 33). The central California distribution had nearly normal kurtosis while the northern California distribution was platykurtic. The distributions were significantly different from each other (KS test, p < 0.0001).
In northern and central California, 50% of male and female black rockfish are sexually mature at a length of 360 mm and at 410 mm as 6- and 7-year olds, respectively (Echeverria 1987). Thus nearly all the central California catch, and approximately one-third the northern California catch, were smaller than the size at which 50% are sexually mature (Figure 33).
In spite of the small the sample sizes, a strong mode of young fish was evident in central California, but not in northern California (Figure 34). The central California mode grew approximately 20 mm each year from 1984 to 1986. The availability of black rockfish increased in 1984 and 1985 due to what appears to be an isolated cohort (Figure 18).
The largest bocaccio taken were from northern California, where mean lengths of the CPFV catch and trawl catch averaged 573 mm and 537 mm, respectively (Figure 35). Fish were smaller in central California, where mean lengths of CPFV and trawler catches were 418 mm and 491 mm, respectively. Pearson and Ralston (1990) found a cline of larger bocaccio from Monterey to Eureka in trawl catches averaged from 1978 to 1988. The central California CPFV length-frequency distribution was strongly skewed to the right by a strong mode of young fish. The length-frequency distributions of the CPFV and trawl catches were significantly different in both northern and central California (KS test; p < 0.0001).
In central and northern California 50% of male and female bocaccio are sexually mature as 3- and 4-year olds at 420 mm and 480 mm, respectively (Echeverria 1987). Thus much of the 1980-86 central California CPFV catch was sexually immature. Most of the central California trawl catch and the northern California CPFV and trawl catches had reached sexual maturity (Figure 35).
Two strong modes were apparent in the annual length-frequency distributions of recreationally caught fish from central California and trawl-caught fish from northern and central California (Figure 36). Trawl fishery age data (otolith break-and-burn technique) (Bence and Rogers 1993) indicate the two modes represent the strong 1984 (A) and 1977 (B) year-classes. The strong mode of young fish for the central area is the 1984 cohort that was first observed in 1984 as juveniles (approximately 240 mm in length) taken from piers.
The much larger average sizes found in northern California are partially the result of earlier recruitment to the central California recreational and trawl fisheries of the dominant year-classes. Both the 1984 and 1977 cohorts were recruited a year earlier in the central California trawl fishery than in the northern California trawl fishery. Unlike blue rockfish and black rockfish, bocaccio did not show greater dominance by strong year-classes in central California than in northern California.
Like bocaccio, chilipepper from northern California were much larger than those from central California; the northern California trawl catch averaged 425 mm while the central California trawl and CPFV catches averaged 391 mm and 381 mm, respectively (Figure 37). All three distributions were significantly different from one another (KS test; p < 0.0001).
In central and northern California, 50% of male and female chilipepper are sexually mature as 3-year olds at 310 mm and 340 mm, respectively (Echeverria 1987). Thus most fish taken by CPFVs and trawlers were sexually mature.
Yearly length-frequency distributions showed more than one mode in both northern and central California, though the modes in the central California recreational boat catch were mixed and less discernible than for bocaccio (Figure 38). A distinct mode of young fish among central California recreational boats and not among trawls suggests that recruitment to the recreational boat fisheries occurs at a smaller size. Rogers and Bence (1992), using otolith break-and-burn technique on trawl-caught fish, identified 1975 and 1984 as two dominant year-classes. The strong mode of young fish (A) in the recreational boat catch in 1986 probably represents the 1984 year-class.
In northern California in 1980-86, size distributions and mean lengths of PRB-caught and CPFV-caught canary rockfish were similar, while trawl-caught fish were much larger and had a more platykurtic distribution (Figure 39). The northern California CPFV and PRB distributions were skewed to the right by a similar dominant mode at 340 mm; the distributions were significantly different (KS test; p = 0.0336). Mean length of the CPFV catch (367 mm) was slightly greater than that of the PRB catch (356 mm) but much less than that of the trawl catch (482 mm). In central California, mean length of the CPFV catch (361 mm) was greater than that of the PRB catch (301 mm). Both distributions were skewed slightly to the left and were significantly different (KS test; p < 0.0001).
A study by Boehlert and Kappenman (1980), encompassing our study area, showed that canary rockfish growth rates did not vary by latitude. This implies the smaller sizes we found in central California represent catches of younger fish. Fifty percent of canary rockfish in central and northern California are sexually mature at 400 mm as 7-year olds for males and 440 mm as 9-year olds for females (Echeverria 1987). Thus most of the recreationally caught canary rockfish in both northern and central California were sexually immature, with a higher proportion of immature fish in central California.
Mean length of both recreationally caught and trawl-caught canary rockfish declined from 1980 to 1986 (Figure 40). Mean length in the northern California recreational boat catch declined from 390 mm in 1980 to 352 mm in 1986. Mean length in the northern California trawl catch declined from 498 mm to 469 mm during the same period. In central California, low sample sizes in 1980-84 obscured any trends, and an apparent strong year-class mode increased mean length from 311 mm in 1984 to 326 mm in 1986. A similar occurrence in the northern California recreational boat catch in 1985-86 suggests the strong year-class was also present there. Pearson and Ralston (1990) described the declines in mean length of trawl-caught canary rockfish from 1978 to 1991 in California as indicative of fishing down of surplus stocks. An ANCOVA comparison between northern California trawlers and recreational boats showed parallel significant declines, with slopes of -4.87 mm/year and -4.89 mm/year for trawlers and recreational boats, respectively (Figure 40). Both slopes were significantly different from 0 (p < 0.00001) but not from each other (p = 0.983).
Mean lengths of central California sport-caught canary rockfish were greater in 1959-60 than in 1980-86; the decline in northern California was not as pronounced (Figure 40). The declines affecting the stock taken by recreational fisheries appear greater in central California than in northern California.
Due to small sample sizes, brown rockfish length-frequency data should be examined cautiously. Between 1959-60 and 1980-86, mean length of brown rockfish appears to have declined substantially (Figure 41). In northern California, mean length of the CPFV catch declined from 389 mm to 344 mm, while mean length of the PRB catch declined from 363 mm to 274 mm. In central California mean length of the PRB catch declined from 402 mm to 310 mm.
In central and northern California 50% of male and female brown rockfish reach sexual maturity as 5-year olds at 310 mm (Echeverria 1987). Our results indicate that in 1959-60 most brown rockfish taken had reached sexual maturity, but by 1980-86 few fish taken from shore and about half taken from PRBs had reached the size at which 50% are sexually mature (Figure 41).