8084

• UCREFRP Catalog Number: 8084
• Author: Gerhardt, D. R. and W. A. Hubert
• Title: Population Dynamics of a Lightly Exploited Channel Catfish Stock in the Powder River System, Wyoming-Montana
• USFW Year: 1991
• USFW - Doc Type: North American Journal of Fisheries Management
• Copyright Material: YES

202

GERHARDT AND HUBERT

Tissues were sectioned at a thickness of 11 J,Lm,
mounted on slides, stained with Giemsa meta-
chromatic stain, and covered with Permount and
coverslips. The aging procedure followed that of
Sneed (1951) and Marzolf (1955), in which an-
nular (opaque) rings were counted to determine
the age of each fish.
All calculations of mortality rates followed
Ricker (1975). A linear regression of the natural
log offrequency against age was used to determine
total instantaneous mortality (Z), and only fully
recruited age-classes were included. To estimate
total annual mortality (A), we used the equation
A = 1 - e-Z. Tag returns from anglers were used
to estimate fishing mortality. Marking was done
throughout the 1987 and 1988 sampling season,
and recoveries were reported through 1989. No
adjustment was made for possible tag loss (Green-
land and Bryan 1974) or nonreturn of tags. The
instantaneous rates of fishing mortality (F) were
calculated for both sampling years by multiplying
the ratio of fishing deaths to all deaths by instan-
taneous mortality. The total number of deaths was
estimated by multiplying the number offish tagged
by total annual mortality offish fully recruited into
the sampling gear. Annual fishing mortality (m)
was calculated by using the equation m = 1 - e-F.
Growth and mortality statistics were used to
predict the response of channel catfish in the Pow-
der River system to variation in exploitation. To
predict changes in yield for incremental increases
of 0.05 in instantaneous fishing mortality, we used
both the Ricker and Thompson-Bell yield models
(Ricker 1975). In addition to providing estimates
of biomass, the Thompson-Bell model provided
estimates of numbers offish of each year-class and
was used to predict changes in population struc-
ture. To compute RSDs for each exploitation in-
crement, we used the frequency and average size
of individuals in each age-class generated by the
Thompson-Bell model. Instantaneous rate of
growth (G) was estimated for each age-class by the
equation

G = 10geW2 - 10g.,W\;

WI and W 2 are mean weights of successive age-
classes.

Results

Sampling was done from June 4 to August 15,
1987, and May 18 to August 15, 1988; length and
weight data were collected from 831 channel cat-
fish in 1987, and 879 in 1988. Of the channel

80
70 1987
60
>- 50 ..
U
l:
GI
:l
C"
GI
..
u..

70

60

>-
u
5i 50
:l
~40
u..

30

20

10

o

18 22 26 30 34 38 42 46 50 54 58 62 66 70 74 78
Length (em)

FIGURE 2.-Length-frequency distributions for 831
channel catfish caught in 1987 and 879 fish caught in
1988 from the Powder River system.

catfish sampled, 62% in 1987 and 76% in 1988
were longer than 30 cm. In 1987, RSD indices
were stock, 39%; quality, 15%; preferred, 20%;
memorable, 24%; and trophy, 2%. In 1988, these
were stock, 29%; quality, 18%; preferred, 19%;
memorable, 27%; and trophy, 7% (Figure 2).
Eight hundred fifty-six fish were aged. Devel-
opment of the tissue-staining technique resulted
in the loss of some spines, and others that could
not be aged with confidence were omitted in the
analysis. Estimated ages of fish ranged from 3 to
21 years. Mean lengths at a given age in 1987 and
1988 are shown in Table 1. Sex was determined
for 729 fish (416 males and 313 females) in 1988.
No difference was observed in the growth of males
and females. Channel catfish appeared to be fully
recruited into the hoop nets at age 5 (Figure 3),
and only fish 5 years of age or older were used in
subsequent computations. Using the 1987 sample,
we estimated total annual mortality to be 23% and
the instantaneous rate to be 0.26 for fish 5 years
old and older.
Of the 618 channel catfish tagged in 1987, 6 tags

j