Fig 1.
Diagram showing the two slipping methods: a. modified (MS) and b. standard (SS).
See main text for detailed descriptions.
Table 1.
Summary information for the field and monitoring phases with mean values (and ranges) for: Transport density at sea, holding temperature, number of fish, length, weight, condition factor (CF, adapted from Fulton’s “K”) and dead sardines at the end of the study.
“Reps” stands for replicates, “Days” stands for days of monitoring.
Table 2.
Summary of biological data (mean and standard deviation) for sardines sampled for blood analysis in the three treatments.
Fig 2.
Survival estimates of sardines by Kaplan-Meier survival estimator (dashed line).
Overlay with the predictions from the selected mixture-distribution survival model M3 (solid lines with 95% confidence bands) for the three treatments: Control (Red), Standard slipping (SS-Blue) and Modified slipping (MS-Green).
Fig 3.
Estimated model parameters for control (C), modified slipping (MS) and standard slipping (SS) treatments.
Numbers indicating individual replicates and the absence of numbers indicating values from the selected candidate model fit to data pooled across replicates. The dots are the maximum-likelihood estimates and the bars are 95% confidence intervals. The individual panels are for a) alpha and b) gamma, respectively the rate and shape parameters of the Weibull distribution, as well as c) the estimated survival rate at the asymptote and d) the time required to reach the asymptote (day).
Fig 4.
Effect of study duration on the accuracy of the estimated survival rates at asymptote for the control (black circles), MS (grey squares) and SS (open triangles) treatments.
The symbols are the maximum-likelihood estimates and the bars are the 95% confidence intervals for the estimated survival rate at asymptote. Lines indicate the survival asymptote estimates for each respective treatment based on the original data.
Table 3.
Summary statistics of GLM fitted to fish scale loss as a function of time in captivity (days) and state (dead or alive) for each sampled treatment (control; modified slipping; standard slipping).
The exponent of the intercept corresponds to the mean percentage of scale loss on the first day in captivity (Day 0), while the slope indicates the mean scale loss (percentage). Standard error (S.E.) given in parentheses.
Fig 5.
Sardine scale loss (%) during the first week in captivity, separately for fish sampled dead (O) and alive (x) in treatments a. control, b. modified slipping and c. standard slipping.
Sampling day for live fish is shifted by 0.5 days to improve clarity. Lines indicate fitted GLM (shown in Table 3; solid line, live fish; dashed line, dead fish). d. Distribution (boxplots) of scale loss for all sardines that died during capture (day-1) up to the first 2 days in captivity in the three sampled treatments (0-control; 1—modified slipping; 2—standard slipping). ** symbol indicates significance of SS treatment over the other two treatments (Kruskal-Wallis: H = 73.3, d.f. = 2, P<0.001).
Table 4.
ANOVA values for treatments affecting blood parameters analyzed (cortisol, glucose, lactose, protein, lipids and osmolality) at sea (day -1) and during 28 days monitoring in captivity.
Fig 6.
Evolution of physiological parameters in sardine blood plasma for each treatment (C – control; MS- modified slipping; SS – standard slipping) during the monitoring period are shown.
Black lines for cortisol and glucose are the mean for observations at the beginning (solid) and at the end of purse seine fishing operations (dashed) calculated from Marçalo et al. (2006); Red lines are the mean of observations for sardines sampled at day 49 for “low stress” (dotted) and “high stress” (dotdash).