California. Dept. of Fish and Game.

California fish and game (Volume 54, no. 4) online

. (page 2 of 10)
Online LibraryCalifornia. Dept. of Fish and GameCalifornia fish and game (Volume 54, no. 4) → online text (page 2 of 10)
Font size
QR-code for this ebook


the San Pablo Bay fishery is the most concentrated and best covered
by publicity, while the ocean fisliery is the most widespread. However,



238



CALIFORXIA FISH AND GAME



TABLE 4
Nonresponse During First Year After Tagging





Nonresponse


95 percent confidence

limit for

nonresponse


Number of
reward tags


Number of
nonreward tags


1958 . . -


0.470
0.3.-)8
0.311
0.389
0.385


0.313.5-0.6263
0.1030-0.5.534
0.1.506 0.4708
0.1808-0.5900


135
229
381
239


277


1959


919


1960

1961- . -


1498
9.50


Mean









• Fish in poor condition when tagged were excluded.



TABLE 5
Nonresponse for Fish Recovered at Various Localities





Recovery localities


Year tagged


Delta


San Pablo Ray


San Francisco
Bay


Pacific Ocean


1958


0.47
0.20
0.36
0.30
0.33


0.35
0.37
0.03
-0.25
0.13


0.26
0.40
0.35
0.63
0.39


0.75


1959


0.75


1960


. 55


1961.


0.25


Weigli ted mean. ..


0.58







annual variations from eacli locality were so lar<2:e in relation to the
differences among means that precise estimates are not possible. Hence
nonresponse was assumed to be the same for all localities. If mean
nonresponses for each area had been used, estimated exploitation rates
would have been 4.5 to 7.8% lower.

The annual nonresponse estimates are subject to considerable chance
variation (Table 4). Hence, the weighted mean nonresponse for the
4 years (0.385) was used for all years.

Representativeness of Sample

The sample of bass tagged in the western Delta during the spring
does not include immature females, since they do not migrate to the
Delta. In addition, males and larger bass may be overrepresented, the
former because males remain on the spawning grounds longer than
females; the latter because the fishing gear and methods used are
selective for larger fish (Chadwick. 1967).

Other than this, the representativeness of the sample may be judged
only by com])aring results among different segments of the sample and
between the main sample and samples taken from other segments of
the population.

Percentage returns from bass tagsed in the western Delta at different
times during the spring and at different locations there do not differ
significantly (Tables 6 and 1). These groups also have similar migra-



STRIPED BASS MORTALITY RATES



239



tion patterns (Chadwick, 1967), so that any failure of the sample
to be random in these respects would not bias results.

TABLE 6

Relationship Between Time of Tagging and Percentage of Tags
Returned During First Year *





Year^tagged


Relative
time of tagging


1958


1959


1960


1961


Early . ..


22.3

(4/11-4/30)

23.0
(4/28-5/8)

22.4
(5/8 -6/4)


13.2

(4/6 -5/7)

16.2
(5/7 -5/20)

15.8
(5/21-6/11)


14.2

(4/6 -5/10)

16.4
(5/9 -5/18)

12.3
(5/18-6/2)


12.6


M idseason

Late


(4/8 -4/25)
11.0

(4/25-5/23)



* Tagging dates are indicated in parentheses. Includes only flsh in good condition
tagged in tlie western Delta.



TABLE 7

Relationship Between Location of Tagging in the Western Delta
and Percentage of Tags Returned During First Year *



Tagging location



Broad Slough (D)..
Chain Island (B)...

False River (E)

Sherman Island (C)
Chipps Island (A)..



Year tagged



1958



21.3

(461)

24.1

(1293)

19.4

(599)

r24.3

(1538)



1959



14.7

(129)
15.7
(236)
14.0

(1204)
16.0

(1305)
15.4
(91)



1960



15.3

(340)

14.0

(473)

13.0

(1326)
16.2

(1219)



1961



11.3

(1182)
11.9
(404)



* Pish in poor condition when tagged are excluded. Numbers of tagged fish listed in
parentheses. Letters in parentheses after tagging locations refer to locations in
Figure 1.

Of the 889 bass tagged at Fremont Weir on the upper Sacramento
River, 28.3% were returned the first year. Only 23.1% of the com-
parable group of 3,891 bass from the western Delta were returned
the first year. These percentages differ significantly (x^ = 10.99,
P < 0.001). However, the April through February returns (luring the
first year (20.4 and 21.2%, respectively) do not differ significantly.
Migrations of the two groups showed an analogous difference, because
most Fremont Weir fish return to the upper Sacramento River during
the spring (Chadwick, 1967). This indicates that the difference in
percentage returns is caused by most Fremont Weir fish being sub-
jected to the intensive fishery in the Sacramento River.

Hence, mortality estimates would be biased by inadequate repre-
sentation of upper Sacramento River fish in the sample. While many



240



CALIFORNIA FISH AND GAME



bass tafrgod in tho Avostcrn Delta migrated up the Sacramento River,
this gi'ou]) is probably inndcfpiately represented, because many bass
migrate up the river before tagging began each spring. Tliis i)robably
causes returns from the Delta sample to underestimate fishing mor-
tality slightly.

Of the 88() bass in good condition tai^-ged at Prisoners Point in the
eastern San Joaquin Delta in 11*.")!). 14.(1' r Avcre returned. Eeturns from
a comparable group of 2.965 bass tagged in the western Delta ^vere
15.1% — a nonsignificant difference. Thus, the rate of return from the
"western Delta sami)le is representative of the rate among bass mi-
grating to the eastern Delta.

In summary, each of the factors affecting the validity of mortality
estimates — immiHliate tagging mortality, biases in nonresponse, de-
creased vulnerability, and underrepresentation of the Sacramento River
run — causes small ei-rors. Assuming that nonresponse is underestimated
due to downward biases exceeding the possible upward bias associated
with not stratifying estimates by area, all four errors cause fishing
mortality to be underestimated.

The first two cause natural mortality to be overestimated, while the
second two cause survival to be overestimated.



MORTALITY ESTIMATES
Spring Tagging

The number of usable nonreward disk dangler tags ranged from
3,891 to 1,609, and returns have been received regularlv from all groups
through May 1967 (Table 8).

TABLE 8
Summary of Spring Tagging in the Western Delta





Number
tagged*


Number of annual returnst


Year tagged


1958-9


1959-60


1960-1


1961-2


1962-3


1963-4


1964-5


1965-6


1966-7


1958


3,S91
2.905
3.358
1,609

11.823


890


207
450


103
200
503


52
111
201

1S8

612


27

89

158

131

405


11

01

145

117

337


12
20
59
53

144


12
35
21

75


8


1959


17


1960


33


1901


25


Totals


83



• Include.s only nonreward tags which were in good condition after tagging.
t Numbers are not corrected for nonresponse.

The estimated rate of exploitation and expectation of deaths from
natural causes were 50% higher in 1958 than in any of the next 6
years (Table 9). ]')uring that 6-year ]ieriod, the rate of exploitation
fluctuated from approximately 0.28 to 0.19, while survival rates varied
from 0.51 to 0.66. No estimates are included for 1965-66 and 1966-67
because few returns were received, and tagging during 1965 and 1966
(Inland Fisheries Branch, unpubl. data") will provide more accurate
estimates of mortality rates in these years. TIk^ 1966-67 returns are of
particular interest because they exceed the 1965-66 returns. Returns



STRIPED BASS IMORTALITY KATES



241



from the fish tagged in 1965 show a similar trend (Lee W. Miller, pers.
comm.), presumably indicating a substantial increase in the rate of
exploitation in 1966-67.



TABLE 9
Mortality Rate Estimates *





Survival
rate


Rate of
exploitation


Expectation of

death from
natural causes


Instantaneous

total
mortality rate


Instantaneous

fishing
mortality rate


Instantaneous

natural
mortality rate


1958


0.319
0.5.34
0.601


0.372
0.247
0.243
0.190


0.309
0.219
0.156


1.14

0.63
0.51


0.62
0.33
0.31


52


1959.-


0.30


1960


0.20






1961


0.662
0.592
0.511
0.557


0.148
0.208
0.208
0.208


0.41
0.52
0.67
0.67


0.23
0.25
0.39
0.36


0.18






1962


0.200
0.281
0.2.35


27


1963


0.28


1964


0.31







* Includes only striped bass tagged in the western Delta during the springs of 1958
through 1961. Estimates below line are approximations based on the assumption
that ratio of 1962 to 1961 returns is a valid estimate of 1961 survival, and that
the annual expectation of death from natural causes from 1962 through 1964 was
equal to the 1958 through 1961 mean.

Estimated variances of survival rates (equation 9) were 0.000275,
0.000516, and 0.000787, respectively, in 1958, 1959, and 1960. Thus,
survival w^as significantlv lower (95% confidence level) in 1958 than in
1959 or 1960, but it did' not differ significantly in 1959 and 1960. The
respective proportions of each year's tags which were returned during
the first year after tagging and their 95% confidence intervals (esti-
mated using equation 5) were 0.229 ± 0.013, 0.152 ± 0.013, 0.150 ±
0.012, and 0.117 ± 0.016. Thus, 1958 returns were significantly greater
than, and the 1961 returns were significantly lower than, the 1959
and 1960 returns.

The 95% confidence limits (from equation 6) for the 1959 through
1961 rate of exploitation were 0.372 ± 0.058, 0.247 ± 0.042, 0.243 ±
0.041, and 0.190 ± 0.038. Thus, the 1958 rate was significantly higher
than the rate in any subsequent year. While the 1961 estimate w^as not
significantly lower than the 1959 and 1960 estimates, the significant
differences in the proportion of tags actually returned in these years
indicate that the differences in rates of exploitation were probably real.

Rate of exploitation increased linearly with fish length in 1958 and
1960. but no similar relationship existed in 1959 and 1961 (Table 10).
The relationship was much more pronounced in 1958 than in 1960
and was due largely to the unusual catch of large bass in the Pacific
Ocean that summer (Chadwick, 1967). If bass caught in the ocean
are eliminated from the tag returns, the respective percentage returns
for the 15-17, 18-19, 20-21, 22-23, 24-26, 27-29, and 30-plus-inch
length groups become 16.9, 19.0, 21.1, 20.2, 21.5, 21.9, and 26.9.



-77541



242



CALIFORNIA FISH AND GAME



TABLE 10
Relationship Between Length at Tagging and First Year Returns *





Percentage returns from each year's tags


Fork length in inches


1958


19.59


1960


19G1


15-17


17.2

19.9

22.2

22.5

25.6

28.8

36.5
<0.001
<0.001


15.3
10.3
16.6
15.8
16.5
13.7
16.9
0.08
0.12


11.3
12.5
14.8
13.9
15.2
19.1
14.5
0.19
0.02


7 6


18-19 .


13.5


20-21


14 8


22-23-. . .


10 9


24-26


10.0


27-29.

30+. .


12.1
14 5


Overall t} valuef

X^ due to linear regression-


0.34
>0.90







* Results include only nonreward disk dangler tags from fish in good condition when

tagged.
t Data analyzed with chi-square test with overall chi-square value subdivided to

identify portion due to linear trend (Maxwell, 1961, p. 63-09). Values listed are

probabilities of getting a larger x" value by chance.

In 1960, 13.6% of the 2,472 tags on male bass were retnrned during
the first year, while 17.3% of the 886 tags on females were returned.
This difference is significant (x" =7.2, P < 0.01). The c()m])arable per-
centages in 1961 (11.0% from 1,036 males and 12.6 from 573 females)
did not differ significantly {f — 0.89, P = 0.37).

In 1961, tags from 192 of the males and 74 of the females tagged in
1960 w(n*e returned. Assuming that survival was not constant in 1960
and 1961, survival of males and females in 1960 would have been 0.7(J6
and 0.665, respectively.

Survival in 1961 can be estimated only by assuming equal sur\i\id
in 1960 and 1961. Estimates for males and females would be 0.614 and
0.847, respectively. The latter indicates an expectation of death from
natural causes of 0.027, which is unreasonably low. This probably re-
flects chance variability associated with the relatively small sample size.

While this evidence is limited, it suggests that mature males have a
lower rate of exploitation and higher survival than females.

Fall Tagging
De\\a

Of tlie 628 bass tagged in the western Delta during the fall of 1958,
128 were returned dui'ing the first year. Assuming a nonresponse of
0.385, the rate of exploitation was 0.33. This is intermediate between
the rates of exploitation estimated for bass lau'gcMl in the springs of
1958 and 1959.

In order to compare the timing of mortality in s])ring- and fall-tagged
fish, the number of bass tagged in the spring of 1958 which were alive
at the end of November 1958 was estimated. This was done by dividing
the number of tag returns caught tlirough November 30 by the number
caught during the year to estimate the fraction of the annual exploita-
tion rate occurring between the spring and November 30. An equal
fraction of annual deaths from natural causes was assumed to have
occurred during this period.



STRIPED BASS MORTALITY RATES



243



From December 1958 through December 1959, the fraction of tags
actually returned from the 1958 spring and fall tagging (Table 11)
did not differ significantly (x" = 2-50, P = 0.12). This fact and the
fall returns being intermediate between returns from the two springs
indicate that angling and natural mortality occur at roughly propor-
tional rates within the year. This is necessary if instantaneous mor-
tality^ rates are to be estimated accurately from tag returns.

TABLE n

Seasonal Occurrence of Angling Mortality From December 1958 through December
1959 for Three Groups of Tagged Bass *



Month of recovery


Spring 1958
Delta tags


Fall 1958
Delta tags


Fall 19.58
San Pablo tags


Dec.-Feb __ . . _


0.048
0.067
0.048
0.029

0.191

2175


0.040
0.100
0.037
0.043

0.220

628


. 0.50


Mar. -May . _ _


0.128


June-Sept ._ .


. 0.56


Oct.-Dec . .


. 073


Total . .


0.307


Estimated number alive at beginning of period


673



* Expressed as the fraction of those estimated to be alive on December 1, 1958, which
were returned during given periods.

San Pablo Bay

Assuming a nonresponse of 0.385 first-year rate of exploitation esti-
mates for bass tagged in San Pablo Bay in the falls of 1958 and 1959
were 0.551 and 0.572, respectively.^ Assuming constant survival, 1958
survival and expectation of deaths from natural causes was 0.303 and
0.146, respectively. Hence, the primary difference between the Delta
and San Pablo Bay groups is in the considerably higher rate of exploi-
tation in the latter. This difference resulted largely from high returns
from San Pablo Bay tags in the spring and fall (Table 11).



DISCUSSION

Variations in Mortality Rates

Returns from the spring and fall 1958 Delta tagging indicate that
fishing and natural mortality occur at nearly proportional rates within
the year, so instantaneous and annual mortality rates can be calculated
from annual expectations of death. These estimates are included in
Table 9.

The higher rates of exploitation among San Pablo Bay fish are prob-
ably a function of migration differences. A higher fraction of fish
tagged in San Pablo Bay than those tagged in the Delta remains in
San Francisco and San Pablo Bays during the winter and spring (Chad-
wick, 1967). This presumably reflects the failure of legal-sized imma-
ture bass, which are primarily females, to participate in the spawning
migration to the Delta. The high spring returns from San Pablo Bay
fish probably reflect greater vulnerability of immature fish during the

s Actual first-year returns were 254 of 750 for 1058 tags and 38 of 108 for 1059 tags.
Second-year returns from 1958 tags totaled 77.



244 CALIFORXIA FISH AND GAME

s]);i\viiiii,u' period since iiuiture bass eal little tlioii (Stevens. 19G7 ;
Thomas, 1967). Hence, immature fisli in the Bay Area are exploited
more heavily tlian the mature fish are.

In the 4 years when both the rate ol' exploit at ion rjul exjx'ctation of
deaths from natural eauses M^ere estinuited. they -were positively cor-
related (Table 9). While th(> sample is small, it su<rfrpsts that either
fishing contributes to "natui'al"' mortality, through a factor such as
deaths from hooking, or some fishing deaths were erroneously attributed
to natural mortality. The most probable cause of the latter is luider-
estimated nonresponse. These hypotheses cannot be e\aliiafe(l with
available data but b(^th ])i-obably contribute.

One of this study's most significant findings is the magnitude of
fluctuations in the rate of exploitation. As pointed out earlier, the homo-
geneity, both in terms of geographical and seasonal migration patterns
(Chadwick, 19G7) and in terms of the proportions returned from dif-
ferent groups of tags applied in any given year, is strong evidenet' of
the validity of this conclusion. The most probable cause for these varia-
tions in exploitation rates is annual variations in catchability.'' Both
annual variations in migrations and changes in angling ef^eiency re-
lated to weather and water conditions undoubtedly changed catclia-
bility. However, sufficient facts are not available to explain the specific
changes observed in exploitation rates.

Variations of this magnitude in exploitation rates would cause con-
siderable variations in total catch and catch per unit of effort, inde-
pendent of changes caused by variations in po])ulation size. Thus, wide
variations in total catch and catch j)er unit of effort winch parallel
changes in exploitation rates would support the validity of the exploita-
tion rates.

Only crude estimates of trends in total catch and effort (Seeley,
Tharratt, and Johnson, 1963; Albrecht, 1964) and more precise catch-
per-unit-of-effort measurements for a few scii'ments of the fisherv
(Chadwick, 1962; L. W. IMiller and E. J. McKechnie. MS; R. J. ]\rc-
Kechnie, MS) are available for the period covered by this tagging
study. The latter provide evidence of wide fluctuations in catchability
in segments of the fishery, but neither source ])rovides evidence of
overall trends paralleling iiulicated fiuctuations in exploitation rates.
The significance of this is difficult to interpret because of imperfections
ill all measm-ements. This emphasizes the need for better measurements
of catch and effort.

ACKNOWLEDGMENTS

T thank the many Department of Fish and Game employees who con-
tributed to this study. These incdude Arnold B. Albrecht (deceased),
AVilliam ITeubach, Don LaFaunee, John B. Eobinson, and Robert J.
Totli. who hel]ied tag bass; Vincent Catania, who hung and fished the
gill nets; and Charlotte Ilotchkiss, who prei)ared correspondence with
anglers returning tags.

I also thank (i. J. Paulik and Patrick' Tomlinson foi- critically
reviewing the original manuscript and suggesting improvements in



8 Catchability is defined a."? tine fraction of the whole fi.'^h stock which is cauerht by a
unit of fishins effort. It reflects the efficiency of fishinp technifiues (vulnerability)
and the distribution of the population in respect to fishing effort (availability)
(Ricker, 1958).



STRIPED BASS MORTALITY RATES 245

analytical procedures. Formulas 6, 7, and 8 were derived in essentially
their present form by Dr. Paulik.

The program could not have succeeded without the support of the
Boards of Supervisors of Contra Costa and Alameda Counties. These
boards appropriated county funds to reimburse anglers returning re-
ward tags.

I also wish to thank the thousands of cooperative California anglers
who returned tags.

SUMMARY

Mortality rates in tlie striped bass population inhabiting the Sacra-
mento-San Joaquin river system were calculated from returns of 11,823
disk dangler tags. Fish were tagged in the Delta each spring from 1958
through 1961. Returns from these tags were corrected for nonresponse
based on tlie return of $5 reward tags.

Analysis of those returns and returns from bass tagged at other
places in the river system indicate that tag returns gave reasonably
valid estimates of mortality rates, although rates of exploitation are
probably underestimated.

The rate of exploitation ranged from 0.372 in 1958 to 0.190 in 1961.
Survival ranged from 0.319 in 1958 to 0.662 in 1962. These variations
probably reflect differences in catchability due to variations in migra-
tions and angling efficiency. Bass tagged in San Pablo Bay had a con-
siderably higher rate of exploitation, probably reflecting the different
migration pattern of immature legal-sized bass.

REFERENCES

Albreclit, Arnold B. 1964. California striped bass estimates for 1961. Calif. Fish and
Game, 50 (3) : 215-216.

Calhoun, A. J. 1952. Annual migrations of California striped bass. Calif. Fish and
Game, 38 (3) : 391-403.

. 1953. Aquarium tests of tags on striped bass. Calif. Fish and Game, 39 (2)

: 209-218.

Chadwick, Harold K. 1960. Progress report on the striped bass tagging study. Calif.

Dept. Fish and Game, Inland Fish. Admin. Rept., (60-1) : 35 p.
. 1902. Catch records from the striped bass sportfishery in California. Calif.

Fish and Game, 48 (3) : 153-177.

1963. An evaluation of five tag types used in a striped bass mortality rate



and migration .study. Calif. Fish and Game, 49 (2) : 04-83.

-. 1967. Recent migrations of the Sacramento-San Joaquin River striped bass



population. Trans. Amer. Fish. Soc, 96 (3) : 327-342.

German, E. R., and D. A. LaFaunce. 1955. A comment on the use of red tags on
fish. Calif. Fish and Game, 41 (1) : 119-120.

Hallock, Richard J., D. H. Fry, Jr., and Don A. LaFaunce. 1957. The use of wire
fyke traps to estimate the runs of adult salmon and steelhead in the Sacramento
River. Calif. Fish and Game, 43 (4) : 271-298.

Hansen, Morris H., William N. Hurwitz, and William G. Madow. 1953. Sample
survey methods and theory. Vol. I. Methods and applications. John Wiley and
Sons, New York, 638 p.

Hawk, Philip B.. L. Oser, and William H. Summerson. 1949. Practical physiological
chemistry. 12th ed. The Blakiston Co., Philadelphia, 1,323 p.

Jolly, G. M. 1905. Explicit estimates from capture — recapture data with both death
and immigration — stochastic model. Biometrica, 52 (1 and 2) : 225-247.

Lawler, G. H., and G. F. M. Smith. 1903. Use of colored tags in fish population
estimates. Jour. Fish Res. Bd. Canada, 20 (6) : 1431-1434.



246 CALIFORNIA FISH AXD GA^IE

Lindley, D. V. 19G5. Introduction to probability and statistics from a Bayesian
viewpoint. Part I. Probability. Cambridge Univ. Press, 259 p.

:Max\voll, A. E. 1901. Analyzing (lualitative data. .Tobn Wiley and Sons, N.Y., 1G3 p.

Kicker, W. E. 1958. Handbook of computations for biological statistics of fish popu-
lations. Bull. Fish. Res. Bd. Canada, (119) : 300 p.

Seeley. Charles M., Robert C. Tharralt, and Richard L. .Johnson. 1963. California
inland angling surveys for 1959 and 19(50. Calif. Fish and Game, 49 (3) : 183-190.

Stevens, Donald E. 1967. Food habits of striped bass, Roccus saxaiilis, in the Sacra-
mento-San .Toaquin Delta, p. G8-96. In Jerry L. Turner and D. W. Kelley (com-
pilers), ecological studies of the Sacramento-San .Toaciuiii Delta. I'art II. Calif.
Dept. Fish and Game, Fish Bull. (13G).

Thoiuas. .Tohn L. 19G7. The diet of juvenile .-ind adult striped bass, Roccus saxa-
tilis. in the Sacramento-San Joaquin River system. Calif. Fish and Game, 53
(1) : 49-G2.



Calif. Fish and Game, 54(4) : 247-256. 1968.



SOME ENDOPARASITES OF FISHES FROM THE

SACRAMENTO-SAN JOAQUIN DELTA,

CALIFORNIA^

STEPHEN R. EDWARDS^ and FUAD M. NAHHAS

Department of Biological Sciences

University of the Pacific

Stockton, California

A survey of 236 fishes representing 26 species in fresh waters of the
Sacramento-San Joaquin Delta revealed the presence of a number of
helminths. The present paper discusses or describes the following
species. Trematoda: Posthodiplostomum minimum centratchi (MacCallum,
1921; Dubois, 1936) Hoffman, 1958 from Archoplites interruptus (Girard),
Chaenobryttus gu/osus Cuvier, Lepomis cyanellus Rafinesque, and L.
macrochirus Rafinesque; Neascus sp. from Hysterocarpus traski i^traskii;
Neascus sp. from Lepomis cyanellus; Pisciamphistoma stunkardi (Holl, 1929)
Yamaguti, 1934 from Chaenobryttus gulosus and Lepomis gibbosus Lin-
naeus; Alloglossidium corti (Lament, 1921) Mueller, 1930 from Ictalurus
catus (Linnaeus) and Lepocread/um californianum sp. n. from Roccus saxa-
tilis (Walbaum). Cestoda: Cora//obofhri'um giganteum Essex, 1927 from
Ictalurus catus; C. fimbriatum Essex, 1927 from Ictalurus catus and I. punc-
tatus Rafinesque, Proteocep/ta/us sp. from Ictalurus catus; Bothriocephalus


2 4 5 6 7 8 9 10

Online LibraryCalifornia. Dept. of Fish and GameCalifornia fish and game (Volume 54, no. 4) → online text (page 2 of 10)