Sample collection
Specimens of whole individuals (> 300 mm) of skate species R. brachyura and R. clavata were collected by seagoing observers. Specimens were hauled with a beam trawl aboard commercial beam trawlers, and during two scientific surveys aboard Research Vessel (RV) Belgica, and RV Celtic Explorer in the Southern North Sea (ICES area 4.c) and Eastern English Channel (ICES area 7.d) from November 2020 to August 2021 (Fig 1).
Laboratory procedures
Detailed morphometric measurements of total length (TL), disc width (DW), disc length (DL), caudal length (CL), and total weight (TW) were recorded using the ‘Smartfish’ digital measuring board (ILVO, Clemaco Trading NV; d=1 mm), a calliper (Stanley; d=0.1mm) and a scale (Sartorius Combics 1; d=0.5 g) (Fig 2).
Individuals were dissected by opening of the ventral body cavity, after which the liver and the digestive tract were removed. The sexual maturity of the individuals was assessed by the macroscopic morphology of the reproductive tract, and on the presence of external maturity indicators in male individuals (i.e., clasper length, clasper calcification, presence of multiple rows of alar and/or malar thorns) following the ICES Maturity Scale for Oviparous Elasmobranchs from the ICES Workshop on Sexual Maturity Staging of Elasmobranchs (ICES 2020). The clasper length was measured from the forward rim of the pelvic girdle to the tip of the clasper following Collenot (1969).
The age of all specimens was assessed by visual inspection of stained growth bands on the vertebral centra following the ILVO age reading protocol developed within the SUMARiS project (Interreg 2 Seas programme 2014-2020) (Fig. 3) (SUMARiS 2021). The vertebral column was removed by lateral sections along the dorsal surface of each individual. The vertebral column was cleaned by cooking in a water bath at 90°C for 30 minutes. A vertebral segment containing the ten most anterior vertebrae was extracted from each vertebral column, the neural arches were removed by scalpel to reveal the vertebral centrum. The vertebral centra were immersed in a 47°C pepsin-hydrochloric acid solution (500 ml water, 8 ml pepsin, 4 ml 25% hydrochloric acid) for 30 to 45 minutes depending on vertebral size, using a stir heating plate (IKA RT 10 Magnetic Stirrer, Labinco L80 Hotplate). The pepsin solution was drained, and the centra were immersed in an 80°C water bath for 30 to 45 minutes. The centra were air dried after immersion in ethanol (99.8%) for 5 to 10 minutes. For visualisation of the growth bands, the centra were etched in 5% EDTA solution (5g EDTA, 100 ml distilled water) for 5 to 10 minutes, stained in 0.01% crystal violet solution (2ml 0.5% crystal violet, 100 ml distilled water), and air dried. Using a stereomicroscope (light source, Zeiss KL 1500 LCD; microscope, Zeiss Stemi 2000-C) and a camera (Leica MC 171 HD), photomicrographs of ten vertebral centra per individual were taken and optimised for focus artefacts in SmartLab v1.2. The three most discernible centra per individual were independently read in SmartDots v2.3 through indication of the birth mark and growth bands by two operators, without prior knowledge of species, sex, size, maturity, and date of capture. Ages were verified by plotting against total length, outliers were re-examined through the previously mentioned age-reading procedure.
Data analysis
The least squares method was used to adjust the linear relations between morphometric measurements for the officially reported size measurements per species: DW ~ a TL + b, with a, the slope of the regression line, and b, the intercept. Morphometric ratios (i.e., body shape descriptors) for disc shape (DL:DW), whole body shape (DW:TL), relative disc size (DL:TL), and relative tail size (CL:TL) were calculated per individual. Morphometric measurements of size and body shape descriptors were compared among species and sexes (unpaired t-test). Principal Component Analysis (PCA) was applied for: (1) morphometric measurements of size, with the variation of these measurements summarised in the alternative variable ‘morphometry’; (2) body shape descriptors, with the variation in these descriptors summarised in the alternative variable ‘shape’. Alternative variables were compared among species (unpaired t-tests). A flexible discriminant analysis (FDA) (Hastie et al. 1994) was applied to investigate which combination of morphometric ratios was adequate to discriminate species for: (1) all individuals; (2) landing-sized individuals (TL > 500 mm) and discards (TL < 500 mm); (3) immature versus mature individuals.
The LWR parameters were established per species by sex for the relationship between total weight (TW, in g) and total length (TL, in mm) using the log (base 10) transformation formula of le Cren (1951).
log (TW) = log (a) + b log (TL)
TW = a TLb
Where a is the initial growth coefficient, and b is the growth rate. Likelihood ratio tests (LRTs, χ2 statistic) of the log transformed linear model were applied to investigate significance of predictors species and sex. Multiple R-squared was used to assess model fit.
Growth curves were fitted using the von Bertalanffy (1938) growth model per species by nonlinear regression:
With Lt, the estimated length at age t (mm), L∞, the theoretical maximum length (mm), K, the Brody growth rate coefficient (yr-1), and t0, the theoretical age at which total length equals zero (yrs). Model fit was visually assessed. Length-at-age data was compared among species and sexes by ANCOVAs (F-statistic).
Maturity ogives for length (proportion of mature individuals at any total length) and age (proportion of mature individuals at any age) were adjusted per species and sex by logistic regression in the Generalized Linear Models framework (McCullagh and Nelder 1989) with binomial maturity data (0, immature; 1, mature).
With p, the proportion of individuals in mature condition for total length (L, in mm) and age (A, in years). The size at which 50% of individuals are mature was defined as the mean size of maturity (L50), and the age at which 50% of individuals are mature was defined as the mean age of maturity (A50). LRTs (χ2 statistic) were used to test for differences between species and sexes. To ascertain the relationship between body size and sexual development in male individuals, clasper lengths (in mm) were plotted to total length per species by maturity stage.
All analyses were performed in R STUDIO v1.4.1106 (RStudio Team, 2020) using the following main packages: car, dplyr, DescTools, emmeans, factoextra, FactoMineR, FSA, ggplot2, ggpubr, ggspatial, ggsignif, grid, mda, nlstools, rnaturalearth, stats, sf, vegan. Outliers were inspected by boxplots. Assumptions of normality, homogeneity, and linearity were investigated by graphical examination of the residuals and variables. Autocorrelation was checked using a correlogram with Pearson correlation coefficients.
Table 1. Number of male (M) and female (F) specimens per species caught in the Eastern English Channel (7.d) and Southern North Sea (4.c) per trip for the period November 2020 to August 2021.
Trip
|
Area
|
R. brachyura
|
R. clavata
|
Total
|
F
|
M
|
F
|
M
|
November 2020
|
7.d
|
1
|
-
|
3
|
4
|
8
|
December 2020
|
7.d
4.c
|
23
7
|
23
2
|
39
-
|
31
-
|
116
9
|
August 2021
|
4.c
|
2
|
1
|
8
|
8
|
19
|
Total
|
33
|
26
|
50
|
43
|
152
|
59
|
93
|