9588

• UCREFRP Catalog Number: 9588
• Author: Bestgen, K. R. and e. al.
• Title: Population Status of Colorado Pikeminnow in the Green River Basin, Utah and Colorado.
• USFW Year: 2005.
• USFW - Doc Type: Fort Collins, CO.
• Copyright Material: NO

Standard errors for river-wide estimates were obtained from the variance-covariance matrix of
the likelihood from program MARK. Probability of annual transition for each reach (r&.) was
estimated; the low frequency of transitions between sampling passes conducted in a single year
did not warrant occasion-specific Vls.
Abundance of Colorado pikeminnow in each reach was estimated with the Huggins
model estimator (Huggins 1989, 1991; Alho 1990). Abundance estimates from the Huggins
model were derived by the equation K, llp', where M,+ , is the number of unique animals
captured over all short-term sampling occasions and:
P pi),
t
where p; is the probability of initial capture within the annual sampling season. Animals in the
population that were never captured have capture probability (1 -p)" but are conditioned out
(removed from) of the likelihood. The new multinomial distribution still sums to one, and
because only fish that were captured are included in the likelihood, individual covariates (here
TL or polynomials for such) were incorporated to estimate p, Or, and S, where appropriate.
Estimates ofp* are derived with information from both the closed-captures portion of the
likelihood used for abundance estimation and from the Cormack-Jolly-Seber (CJS) component of
the model used to estimate annual survival rates across all reaches. With the additional
information provided about p* from the CJS portion of the likelihood, the individualp's per pass
within the annual sampling period are identifiable based on the numbers of fish initially captured
during each sampling pass within a year. Thus, we could calculate abundance estimates for river
reaches and years where no animals were recaptured between passes within a single year.
21