摘要:

Many fisheries data are commonly summarized by two statistics: mean and variance (or standard deviation). Because observed values are subject to various errors, which often are large and heterogeneous in fisheries studies, outliers commonly exist in the data. The existence of outliers biases estimation of the mean and variance if traditional estimation methods are used. Instead of assuming that errors in fisheries data follow a normal distribution with a constant variance, we propose that errors associated with observations for a variable may encompass a mixture of different levels of normally distributed errors. Based on concepts from a robust regression method, least median of squares, that is not sensitive to atypical observations in data, we develop a simple algorithm to estimate mean and standard deviation. We compare the proposed robust estimation approach with traditional methods and Tukey's biweight robust approach using simulated and field data. Based on simulations, we found little difference in estimated means and variances between the proposed and traditional methods when there were no outliers defined in simulated data. However, when outliers were defined in simulated data, the errors in estimation of the mean and its standard deviation were much smaller with the proposed method than were those estimated with traditional methods. Means and standard deviations estimated with the proposed method changed little, regardless of whether or not the simulated data were contaminated by atypical values. The proposed approach tended to have smaller estimation errors than did the robust biweight method. We demonstrate how the significance and interpretation of fisheries and ecological relationships may be adversely affected when outliers are present. We suggest using our proposed robust method to identify and down-weight outliers in estimating a mean and its standard deviation. One should justify deletion of the identified outliers using the knowledge about fish biology and environmental conditions independent of the variable assessed.

ISSN:0002-8487    

DOI:10.1577/1548-8659(1995)124<0401:REOMAV>2.3.CO;2

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

A comparison of diary and mail methodologies for a cohort of anglers who fished Lake Ontario was used to examine biases associated with 12-month recall from mail questionnaires. Significant differences in estimates were found between estimates reported by respondents in diaries (1992) versus mail questionnaires (1991) for number of days fished and fish consumption, but not for fishing expenditures and catch. After the data were adjusted for a decline in fishing on Lake Ontario between 1991 and 1992, it was found that angler-days were overestimated by 44–45% on the 12-month recall mail questionnaire. This percentage can serve as an initial estimate of a correction factor for future studies. Lower average annual fish consumption rates were reported in the diary year compared with the mail questionnaire year. However, because of the lower percentage of meals of sport-caught fish during the diary year and the knowledge that sportfishing declined in 1992 for Lake Ontario anglers, it is not clear what portion of the decline can be attributed to different factors. A rough estimate of 10% can be obtained by assuming that consumption of fish that were not sport caught was the same in both years and that anglers accurately reported the overall percentage of that consumption in 1991. Less avid anglers had a very small positive discrepancy between their mail (1991) and their diary (1992) estimates of fishing participation, whereas anglers who fished more frequently had a much larger positive discrepancy. With these data, the best mathematical procedure for describing that relationship involved regressing the square root of days fished in 1991 against days fished in 1992.

ISSN:0002-8487    

DOI:10.1577/1548-8659(1995)124<0413:UOADTE>2.3.CO;2

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

Dissolved gas data were collected downstream from a moderately sized midwestern reservoir over a period of 10 years. Among 281 observations from August 1983 through April 1994, the gas pressure differential (▵P) averaged 119 mm Hg (116.0% of saturation) and discharge averaged 245.0 m3/s. A maximum ▵P of 251 mm Hg (133.6% of saturation) was associated with record release rates during the flood of 1993. Three-fourths of the 281 observations exceeded the U.S. Environmental Protection Agency's criterion of 76 mm Hg as ▵P (110% of saturation). Periodic examinations of live fish and of fish collected from fish kills demonstrated sublethal and lethal gas bubble trauma in fish downstream from the dam. The occurrence of periodic gas supersaturation-induced fish kills was linked to continued high dissolved gas pressures during periods when the discharge from the reservoir was substantially decreased. Lower discharge rates decreased river depth and lowered compensating hydrostatic pressure. The occurrence of gas supersaturation at this reservoir is of interest because of the potential for gas supersaturation at other reservoirs where gas supersaturation might not be predicted.

ISSN:0002-8487    

DOI:10.1577/1548-8659(1995)124<0423:GSAGBT>2.3.CO;2

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

Population size and fishing mortality rate and their variances were estimated for the Spanish sardineSardinella auritaalong the southern Fujian coast, China, in 1971–1983 and 1986–1989 by Jones's length-based cohort analysis. Population abundance and biomass varied among different periods, whereas exploitation patterns were consistent within periods from 1971 to 1983 but differed significantly from those in 1986–1989. The approximate variance estimators for this method were developed in general form by the delta method. Errors in the estimates caused by the input parameters were examined, and we conclude that correlation between the estimates of asymptotic lengthL∞and the growth coefficientKtends to reduce errors in the population estimates from length-based cohort analysis. Risk of misestimating population size increases when length-based cohort analysis is used for stocks with high natural mortality. A comparison of the population biology and fisheries of the Spanish sardine around the world shows that the population along the southern Fujian coast is much smaller than that off northwestern Africa.

ISSN:0002-8487    

DOI:10.1577/1548-8659(1995)124<0437:PEWVOS>2.3.CO;2

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

摘要:

Downstream movements of age-0 Arctic graylingThymallus arcticusfrom an indigenous fluvial population (Big Hole River) and two inlet-spawning, lacustrine populations (Upper Red Rock Lake and Ennis Reservoir) were compared in a natural stream. All fish were incubated and reared together in a hatchery and acclimated together in the stream before being released in the stream. All fluvial test fish were F1progeny of parents originating as gametes from wild fish but reared in a natural lake (1993 trials) or in a hatchery (1994 trials). The F1fish in four 1993 trials were progeny of parents whose rheotaxis had also been tested as age-0 young. In all four trials of fish with 7–14 d stream acclimation, significantly fewer fluvial than lacustrine fish (P< 0.005) were recovered within 1 d in nets at a weir 1 km downstream from the release site. In two trials of fish with 1 d acclimation, significantly fewer fluvial than lacustrine fish were also recovered in one trial (P< 0.005) but not in the other (P> 0.500). Twice each year, 10–25 d after trials were concluded, the stream reach in which the test fish had been released was electrofished. Significantly more fluvial than lacustrine fish were captured in the stream in each of the four electrofishing surveys (P< 0.005). The results support earlier evidence from laboratory trials that fluvial Arctic grayling have innately stronger positive rheotaxis than those from lacustrine populations. Such evidence of fluvial-adaptive behavior in Arctic grayling of the Big Hole River reinforces the importance of conserving this small, fluvial population, which is the only known one remaining for the species south of Canada and Alaska.

ISSN:0002-8487    

DOI:10.1577/1548-8659(1995)124<0453:NROARB>2.3.CO;2

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor

ISSN:0002-8487    

DOI:10.1577/1548-8659-124.3.458

出版商:Taylor & Francis Group    

年代:1995

数据来源: Taylor