Stock Assessment of Connecticut River Shad: Examination of Fishing and Predation Effects on the Recent Stock Decline

Table 3. Parameter estimates and equilibrium overfishing thresholds (F_msy, N_msy) for Connecticut River shad derived from the Gompertz external production model. The model was fitted to the LTS Robust regression model. The standard error (SE) is given for each parameter estimate (a, b), as well as the coefficient of determination (r²). Overfishing thresholds (F_msy, N_msy) are presented with 80% CI.

LTS Robust Regression Model
Parameter Mean SE P

a 3.611 1.00 < 0.0004

Overfishing Thresholds

________________________________________________________________________

Threshold Mean 80% Confidence Limit

Table 5. Population estimate and 95% confidence limits (CI) of striped bass in the Connecticut River from April through June, 1994 based on the Petersen mark-recapture method (Savoy 1995). Note that this table was taken from Savoy and Crecco (2004).

-----------------------------------------------------------------
Statistic Estimate or Number
-----------------------------------------------------------------

Number Tagged (M) 346 ^1/

Number Recaptured (R) 26 ^2/

Recreational Catch in River (C) 30,610 ^3/

Population Estimate (P) 407,300

95% Confidence Limits 269,400 to 604,100 ^4/

1/ Marked fish were reduced by 50% (from 692 tagged fish to 346 fish) to reflect a combination of factors including tag loss, tag-induced mortality and migration of tagged fish from the River to LIS.
2/ Number of recaptures increased to reflect a 60% (from 16 recaptures to 26 recaptures) reporting rate.
3/ Striped bass recreational catch in the Connecticut River was taken from the MRFSS and information in the 1994 Volunteer Angler Survey.
4/ 95% confidence limits were based on a Poisson distribution of recaptures (Appendix 2, Ricker 1975).

Table 6. Results of striped bass dietary studies in the Connecticut River during 2005 and 2006 conducted by Mr. Justin Davis as part of his Doctoral dissertation. The dietary data (234 stomach samples) are expressed as the percentage of river herring and shad biomass in the stomachs of striped bass distributed by 10 cm length groups.

Striped Bass
Size class (cm) % biomass river herring % biomass shad
------------------------------------------------------------------------------------------------------------
< 30 0 0
30 – 39 0.8 0
40 – 49 1.7 0
50 – 59 2.5 3.7
60 – 69 18.3 5.9
70 – 79 41.1 6.3
80 – 89 34.4 8.3
90 – 99 9.2 45.5
> 100 5.8 82.4

Table 7. Parameter estimates and approximate equilibrium overfishing thresholds for Connecticut River shad derived from the Steele-Henderson production model with predation effects. The Steele-Henderson production model (equation 6) with striped bass predation effects was fitted by least squares iterative re-weighting with lognormal error using data for 1981-2005. The standard error (SE) is given for each parameter such as the intrinsic rate of shad population increase (r), shad carrying capacity (K in numbers), striped bass per capita consumption (c), and shad stock size at which striped bass satiation (A) takes place. The coefficient of determination (r²) is also shown. Averaged long-term overfishing thresholds (avF_msy, avFcoll, avB_msy) with 95% CI are also shown below.

Iterative Re-weighting Least Squares

Parameter Mean SE

________________________________________________________________________
r 0.59 0.09
K 1,396,000 fish 210,611 fish
c 0.88 0.21