We present sampling designs for estimating total areas of habitat types and total fish numbers in small streams. Designs are applied independently within strata constructed on the basis of habitat unit type and stream reach. Visual methods for estimating habitat areas and fish numbers are used to increase sample sizes and thereby reduce errors of estimation. Visual estimates of area are made for all habitat units, and visual estimates of fish numbers are made for systematic samples of units within given habitat types. Use of systematic sampling circumvents the requirement for a preexisting map of habitat unit locations and simplifies selection of units. We adjust for possible proportional bias of visual estimation methods by calibrating visual estimates against more accurate estimates made in subsamples of those units for which visual estimates are made. In a test application of these sampling designs, correlations between visual estimates and more accurate estimates were generally high, r > 0.90. Calculated 95% confidence bounds on errors of estimation were 13 and 16% for total areas of pools and riffles, respectively, and were 17 and 22% for total numbers of 1 + steelhead trout (Salmo gairdneri) and juvenile coho salmon (Oncorhynchus kisutch), respectively. Our methods appear to offer a cost-effective alternative to more traditional methods for estimating fish abundance in small streams. In addition, visual estimation surveys can produce detailed maps of the areas and locations of all stream habitat units.

We report results of age-specific mating experiments carried out with chinook salmon (Oncorhynchus tshawytscha) at Elk River Hatchery, Oregon. Our analysis of returns from these experiments includes assessment of the marine growth of progeny, and we also account for the negative bias on mean age of returning mature progeny that is a consequence of troll fishery harvest of immature salmon. Results suggest that (a) heritability of age of maturity is relatively high in this species (calculated h2 were 0.49–0.57 and 0.39–0.41 for males and females, respectively), (b) inheritance of age of maturity of females appears to be independent of age of male parent, and (c) for a given parental age, "faster-growing" progeny generally mature at younger ages, but (d) progeny from older parents are not generally smaller at age than progeny from younger parents. Inheritance of age of maturity therefore cannot be a simple reflection of inheritance of growth rate. We tentatively propose the existence of heritable minimum threshold lengths that differentially trigger maturation according to age and sex of parents. We also consider the significance of these experiments for artificial propagation of this species.

No abstract

A common, although generally unrecognized, use of multistage sampling designs in freshwater fisheries research is for estimation of the total number of fish in small streams. Here there are two stages of sampling. At the first stage one selects a sample of stream sections, usually of equal length, and at the second stage one estimates the total number of fish present in each selected section. This paper argues that the conventional practice of selecting stream sections of equal length is ill-advised on both biological and statistical grounds, and that errors of estimation of fish numbers within selected sections will usually be small compared with errors of estimation resulting from expansion of sampled sections to an entire stream. If stream sections are instead allowed to vary in size according to natural habitat units, then alternative two-stage sampling designs may take advantage of the probable strong correlation between habitat unit sizes and fish numbers. When stream sections of unequal sizes are selected with probabilities proportional to their size (PPS), or measures of the sizes of selected sections are incorporated into estimators, one may substantially increase precision of estimation of the total number of fish in small streams. Relative performances of four alternative two-stage designs are contrasted in terms of precision, relative cost, and overall cost-effectiveness. Choice among alternative designs depends primarily on the correlation between fish numbers and habitat unit sizes, on the total number of stream sections, and on sample size. Recommendations for choices among the designs are presented based on these criteria.

Because of their similar appearance and frequent hybridization, juvenile steelhead Oncorhynchus mykiss and coastal cutthroat trout O. clarkii clarkii are difficult to distinguish visually. Nevertheless, field biologists often use visual methods to classify juvenile individuals. This study investigated hybridization between these species and determined the accuracy of field identification where hybridization occurred. Using a five-point classification system, two evaluators identified 500 fish collected from three watersheds in Humboldt County, California. Individuals were then genotyped at seven single-copy nuclear DNA genes and one mitochondrial gene, all assumed to be diagnostic for each species. Single-locus Hardy-Weinberg equilibrium, pairwise genotypic disequilibrium, and cytonuclear disequilibrium calculations revealed that subpopulations of these species were mating assortatively. Presumptive F 1 hybrid individuals were rare, whereas introgressed individuals were more common. These presumptive later-generation backcross hybrids were produced with both parental species but were more frequently produced with coastal cutthroat trout. Interspecific matings appeared to be bidirectional. Conditional classification probabilities between evaluator identifications and genotypes showed that both evaluators had moderate to substantial success identifying individuals less than 85 mm total length, whereas individuals 85 mm and larger were identified less successfully. Evaluators successfully identified coastal cutthroat trout but had moderate difficulty identifying steelhead (sometimes misidentified as hybrids) and always misidentified hybrids as coastal cutthroat trout. Although visual identifications are not without error, approximately unbiased estimates of the percentage of hybrids may be generated from a combination of visual assignments and supplementary genetic analyses.

We explored the long-term consequences of three mating regimes ((1) completely random, (2) completely random but excluding jacks (age 2 males), and (3) male length ≥ female length) on age and sex structure of wild and hatchery populations of Chinook salmon ( Oncorhynchus tshawytscha ). Regimes similar to regimes 1 or 2 are used at most salmon hatcheries, whereas regime 3 emulates the outcomes of natural spawning behaviors that favor larger males. Inheritance of age at maturity is captured in age- and sex-structured models via matrices of age- and sex-specific conditional maturation probabilities that depend on age and sex of parents. In unexploited populations, regime 1 leads to substantial long-term selection for younger age at maturity, an effect that is somewhat reduced by regime 2, but greatly reduced under regime 3. Equilibrium age and sex structures for wild and hatchery populations under regime 3 are similar to those of natural populations, whereas mating regime 1 generates age structure that is greatly shifted toward younger ages and jacks. To prevent unintentional selection for younger age at maturity, we recommend that large-scale hatcheries replace unnatural completely random mating regimes with mating regimes that emulate the outcomes of natural spawning behaviors.

We used multiple resightings (live recoveries) of passive integrated transponder (PIT) tagged fish to estimate overwinter (October‐March) survival rates of juvenile coho salmon Oncorhynchus kisutch in a near‐pristine northern California headwater stream. Cormack‐Jolly‐Seber models were used to estimate period‐specific survival rates and recapture probabilities from three tag groups released during October and November 1999 and March 2000. Parameter estimates from the best‐supported models provided strong evidence of initial tagging mortality and also suggested that subsequent survival rates were affected by fish size at tagging. We found statistical evidence for initial PIT tagging mortality, suggested by the lower estimated period‐specific survival rates for tagged fish in their first period at large compared with estimated survival rates for tagged fish that had survived through a previous period. Estimated overwinter survival rates (November‐March) for fish tagged in October that were uncontaminated by initial PIT tagging mortality ranged from about 21% for 55‐mm fish (fork length at tagging) to about 84% for 106‐mm fish (overall average = 45.5%). Lower apparent survival rates for smaller juveniles may reflect (1) size‐dependent movement out of the study reach, (2) chronic size‐dependent mortality related to PIT tagging, or (3) size‐dependent winter mortality unrelated to PIT tagging. We believe that development of successful recovery strategies for coho salmon stocks will require similar stock‐specific (ideal) or region‐specific (minimally adequate) estimates of survival rates through preadult life stages.

Of 11072 adult female Dungeness crabs (Cancer magister) tagged and released in northern California, 463 were recovered with useful location data that could be used for analyses of crab movement patterns. Although qualitative analyses of movement data suggested possible directed northward movement during winter months, application of two nonparametric tests of movement directionality (the Rayleigh test and Moore's test) failed to support significant directed movement during winter. Large numbers of tagged crabs were recovered inshore of release in shallow sandy areas during spring months, but valid statistical analyses of spring movement data were ruled out by concentration of fishing effort in shallow waters during spring. Nevertheless, recovery of large numbers of tagged females in inshore areas during spring is entirely consistent with an hypothesis of spring inshore movement of females for molting, mating, and later extrusion of egg masses. This hypothesis can be constructed on the basis of information independent of tag recovery data. The most striking finding was that 46% of all recovered crabs were recaptured within 2 km of original release sites; many of these crabs had been at large more than 1 yr. Adult female Dungeness crabs appear to constitute extremely localized stocks in northern California.