Population biology and distribution of the portunid crab Callinectes ornatus ( Decapoda : Brachyura ) in an estuary-bay complex of southern Brazil

Trawl fisheries are associated with catches of swimming crabs, which are an important economic resource for commercial as well for small-scale fisheries. This study evaluated the population biology and distribution of the swimming crab Callinectes ornatus (Ordway, 1863) in the Estuary-Bay of São Vicente, state of São Paulo, Brazil. Crabs were collected from a shrimp fishing boat equipped with a semi-balloon otter-trawl net, on eight transects (four in the estuary and four in the bay) from March 2007 through February 2008. Specimens caught were identified, sexed and measured. Samples of bottom water were collected and the temperature and salinity measured. A total of 618 crabs were captured (332 males, 267 females and 19 ovigerous females), with a sex ratio close to 1:1. A large number of juveniles were captured (77.67%). Crab spatial distributions were positively correlated with salinity (Rs = 0.73, p = 0.0395) and temperature (Rs = 0.71, p = 0.0092). Two peaks of recruitment occurred, in summer and autumn, and ovigerous females were mostly captured during summer, showing a seasonal reproductive pattern. The results showed that C. ornatus uses the bay as a nursery area for juvenile development. Callinectes ornatus is not yet a legally protected species, and the minimum allowed size of crabs caught in the area, although already restricted, should be carefully evaluated since the removal of large numbers of juveniles could negatively impact the local population.

In Brazil, most crustacean fisheries target shrimps, and crabs caught in the trawls have secondary economic importance (COSTA & NEGREIROS-FRANSOZO 1998) or are mostly consumed by small fishing communities (SFORZA et al. 2010).The most economically important crab worldwide is Callinectes sapidus (Rathbun, 1896), but Callinectes danae (Smith, 1869) and Callinectes ornatus (Ordway, 1863) are also caught on the Brazilian coast, which has motivated scientific studies (SANTOS & BUENO 2002, KEUNECKE et al. 2009, SANT'ANNA et al. 2012a).In the Brazilian trawl fishery for the shrimp Xiphopenaeus kroyeri (Heller, 1862), C. ornatus is one of the most abundant brachyurans discarded in the bycatch (CARVALHO et al. 2011).
Callinectes ornatus is widely distributed in the western Atlantic, usually on sand or mud bottoms near rivers, and usually in brackish waters (MELO 1996).Most studies on C. ornatus were conducted along the coast with rivers nearby, but the distribution of this species on the coast as well as in the estuaries was not investigated.This study analyzed the population biology and the seasonal and spatial distribution in the Estuary-Bay Complex of São Vicente, São Paulo, Brazil.

MATERIAL AND METHODS
The swimming crabs were collected monthly in the São Vicente Estuary-Bay Complex, state of São Paulo, Brazil, from March 2007 through February 2008.A shrimp fishing boat equipped with an otter-trawl net 4.0 m wide, 2.0 m high and 9.5 long with a 15 mm-mesh body and 10 mm-mesh cod liner was used to capture the crabs during a 20-min sampling on eight transects, four in the estuary and four in the bay (Fig. 1).On each transect, after the trawl, the bottom water was collected with a Nansen bottle and the temperature (°C) and salinity (‰) were measured.
ses (SOKAL & ROHLF 1995).The reproductive period was determined by the monthly incidence of ovigerous females during the course of the sampling period.

RESULTS
A total of 618 crabs (332 males, 267 females and 19 ovigerous females) were caught; juvenile crabs comprised 77.67% (480 individuals) of this total.Most of the crabs were caught in the bay (84.30%, 521 individuals) (Table I).The spatial distribution of the number of individuals (Fig. 2) was positively correlated with salinity (Table II); most individuals occurred in locations with higher salinity.There was also a significant positive correlation with temperature, where more crabs were found in the warmer months (Table II, Fig. 3).The crabs were sorted and identified according to MELO (1996), the sex was determined by inspection of abdominal morphology, and the sexual maturity was determined through inspection of the abdomen: adhered for juveniles or loose for adults.The carapace width (CW, measured between the bases of the last and penultimate lateral spines) was recorded using a caliper (0.01 mm).For the size-frequency distribution analysis, size-class intervals of 6 mm of carapace width were used.The size intervals for each class were determined according to the mathematical formula of STURGES (1926).The Kolmogorov-Smirnov (KS) was used to test the normality of the size-frequency distribution for each population category (sex).The size was compared between the sexes with the Kruskal-Wallis test.The monthly and overall sex ratios (M:F) were tested with the Chi-square test ( 2 ) and the data for abiotic factors (temperature and salinity) and number of individuals were analyzed by the Spearman's rank correlation coefficient.A significance level of p < 0.05 was adopted for all statistical analy- The high proportion of juveniles decreased the overall mean size, and therefore the ovigerous females were significantly larger than males and non-ovigerous females (Tables III  and IV).The overall size-frequency distribution of C. ornatus (Fig. 4) showed a bimodal distribution for males and a unimodal distribution for females, with the distribution differing from normality for males (KS = 0.1306, p < 0.01), and normal distributions for females and ovigerous females (KS = 0.0701, p = 0.08; KS = 0.1587, p = 0.30, respectively).The seasonal sizefrequency distribution indicated a relationship to the seasons (Fig. 5), with crabs present in larger numbers in the summer and autumn.
The sex ratio of the C. ornatus population did not differ significantly from the expected 1:1 (² = 3.576, p = 0.06).This pattern was the same in all months, except in March 2007, when males were more abundant than females (Table V).Ovigerous females were recorded in higher percentages in the warmer seasons, probably related to seasonal reproduction (Fig. 6).

DISCUSSION
Callinectes ornatus is more euryhaline than other seawater portunids such as Arenaeus cribrarius (Lamarck, 1818)      euryhalinity characteristic of the species.Further studies on competition for resources such as territory use or food are needed to test this hypothesis.However, one clue supporting this hypothesis is that the distribution and salinity were positively correlated, indicating that competition with other species could also explain the low number of the adults in both the estuary and bay.For instance, PINHEIRO et al. (1997) found that different swimming crabs can show niche overlap, but that the number of individuals of C. ornatus was lower when in the presence of greater numbers of larger crabs such as C. danae or A. cribrarius.
The high proportion of juveniles caught indicates that this species uses the bay for the early part of its life cycle.Our data, which revealed the presence of large numbers of juveniles, mostly in the bay (Table I), indicate that C. ornatus uses the bay as a nursery for the development of the juveniles.These data differ from the findings of TUDESCO et al. (2012), of lower numbers of juveniles on the coast near the estuary of Guanabara Bay, Rio de Janeiro, Brazil compared to the number of adults; although the presence of C. ornatus in the estuary was not checked.However, BAPTISTA et al. (2003) also found large numbers of juveniles during the summer near Paranaguá Bay in southern Brazil, outside the estuary, similarly to our data.Another factor that could explain the high proportion of juveniles is the calmer water in the bay; since the juveniles of Callinectes spp.usually prefer calm, shallow waters with more organic matter (NEGREIROS-FRANSOZO & FRANSOZO 1995, TUDESCO et al. 2012), it is possible that the conditions in the bay also favor the presence of juveniles.During the larval period, some crustaceans do not have a large capacity for osmoregulation and cannot tolerate sudden changes of salinity, which explains why the juveniles use these areas (HARTNOLL 1982).
Species of Callinectes have low tolerance for extreme temperature changes (LEFFLER 1972, ROME et al. 2005), and C. ornatus is usually found in tropical seas ranging from 18 to 31°C (WIL-LIAMS 1984).The low temperatures during colder seasons in São Vicente Bay could explain the low occurrence of C. ornatus during the study period.The migration pattern of this population showed that the swimming crabs were more abundant during warmer periods; according to COLTON et al. (2014), winter temperatures could influence the number of individuals and impact the fluctuation and dynamic synchronization of a population of the blue crab C. sapidus.Some species of Callinectes migrate, which is more evident when correlating their presence with the reproductive period (MILLIKIN & WILL-IAMS 1984, HINES et al. 1987, SANT'ANNA et al. 2012).
During the sampling period, there were high numbers of juveniles during the summer and autumn months, indicating a possible seasonal recruitment.Smaller recruitment peaks were not observed in the other seasons, although C. ornatus has multiple and continuous spawning with peaks in the autumn at Ubatuba (MANTELLATO & FRANSOZO 1997), and reproduction was more active during autumn and summer, similar to the present study.The differences between our study and others could be explained by the specific conditions in the different locations, the environment (SCHAFFNER & DIAZ 1988, FERNANDES et al. 2006), or the differences in sampling effort in each study.
Few crustacean species maintain a sex ratio of 1:1 (WENNER 1972), and C. ornatus shows a segregated distribution according to the local environmental features (NEGREIROS-FRANSOZO et al. 1999, TUDESCO et al. 2012).Thus, males and females showed different population concentrations as well as varying ratios in different size classes; this is also true for C. ornatus, where the habitats of adults and juveniles differ (NEGREIROS-FRANSOZO et al. 1999).In the present study, the data suggest that the São Vicente estuary-bay complex is the main habitat of juveniles, which explains the deviation toward lower individual sizes as well as the trend toward a bimodal distribution for males.The high proportion of juveniles helps to explain the 1:1 sex ratio, since most of the population was sexually immature, and females do not migrate as described for other Callinectes species (AGUILAR et al. 2005, SANT'ANNA et al. 2012a).
Usually, portunids reproduce throughout the year with peaks in certain seasons, generally winter and summer (PITA et al. 1985, BAPTISTA-METRI et al. 2005).The local population of C. ornatus showed a peak of reproduction only in summer, with the presence of ovigerous females indicating a seasonal rather than a continuous reproduction pattern.Ovigerous females of some Callinectes species prefer higher-salinity waters, which could help the larvae to float and maintain osmotic pressure (PITA et al. 1985, MANTELLATO 2000).This could explain the low number of females in the study area, which receives a large freshwater inflow from the estuary.
The results of this study revealed that juveniles of C. ornatus use the bay area intensively; and also indicated a regional plasticity in reproduction, since C. ornatus shows different reproductive patterns along the Brazilian southern coast.Although the minimum catch size for C. sapidus and C. danae is regulated (BRAZIL 1983), the minimum catch size for C. ornatus is not.Despite the similarities in external morphology of these species, C. ornatus is not yet legally protected.The management of C. ornatus should be carefully evaluated, since it is also an economically important species, and trawling in the area, although highly restricted (IBAMA 1994), could still affect the population since most of the individuals were juveniles and this is likely to be an important nursery ground for their development.

Figures
Figures 2-3.(2) Number of individuals (bars) and the mean salinity (lines) of each transect collected through the year.(3) Monthly distribution of Callinectes ornatus (bars), with the mean water temperature (line) through the months.

Table I .
Number of individuals of Callinectes ornatus caught in the bay and estuary of São Vicente.(AM) Adult male, (JM) juvenile male, (AF) adult female, (JF) juvenile female, (OF) ovigerous female.

Table II
. Spearman's Rank test for salinity vs. number of individuals per transect and temperature (°C) vs. number of individuals per month.*p < 0.05.
or species of Portunus (Weber, 1795), but less euryhaline than other members of Callinectes such as C. sapidus, C. bocourti or C. danae (NORSE 1978).However, this may not in itself explain the low number of adults of C. ornatus in the bay area, where salinities are higher.This species is usually smaller than C. sapidus or C. danae (BUCHANAN & STONER 1988), and the large number of adults of C. danae in the region (SANT'ANNA et al. 2012a) may lead to agonistic relationships with C. ornatus.NORSE (1978) found that C. ornatus living in areas without other, larger species of Callinectes are more tolerant to hyposaline conditions than are

Table III .
Number of individuals and size range (carapace width) of

Table IV .
Kruskal-Wallis test results followed by Dunn test for size of males, females and ovigerous females of Callinectes ornatus.
(SANT'ANNA et al. 2012aCHMUTH et al. 2011)y toward conspecifics(CLARK et al. 1999, REICHMUTH et al. 2011), and might also be aggressive toward other species.We speculate that the presen-ce of the other, larger Callinectes species such as C. danae(SANT'ANNA et al. 2012a) in the estuary-bay complex may influence the distribution of C. ornatus, combined with the

Table V .
Sex ratio by month obtained for the population of Callinectes ornatus in the Estuary-Bay Complex of São Vicente.