Reproductive potential of Pagrus pagrus ( Perciformes : Sparidae ) in coastal waters of Buenos Aires Province ( Argentina ) and Uruguay ( 34 o-39 o S )

The reproductive potential of red porgy in coastal waters of Buenos Aires Province (Argentina) and Uruguay (34o-39oS) was studied by means of a macroscopic and histological analysis of the gonads. Length and age at first maturity were determined, and fecundity, spawning frequency and egg quality were estimated. The spawning season extended from spring through summer between October and January with a peak of spawning in December. Length and age at maturity for sexes combined was 24.5 cm TL and 1.54 years, respectively. Batch fecundity ranged from 6,974 (25 cm TL) to 110,725 (39 cm TL) hydrated oocytes and showed significant linear relationships with total length and ovary-free weight. Relative fecundity ranged from 16 to 172 oocytes per female gram (ovary free). Spawning frequency was 58.5% during January 2011, indicating that females spawned once every 1-2 days at the peak of the spawning season. The life history of red porgy in Argentina and Uruguay was characterized by a young age at first maturity and partial, almost daily, spawning in a bounded time period, a life history strategy that would facilitate population expansion or colonization of new areas, and high population recovery rates or growth (r).


Introduction
The red porgy Pagrus pagrus (Linnaeus, 1758) (Perciformes: Sparidae) inhabits warm temperate to subtropical waters on both sides of the North Atlantic, including the northern Gulf of Mexico and Mediterranean Sea, and in the southwest Atlantic from Venezuela to Argentina (Manooch, Hassler, 1978;Menezes, Figueiredo, 1985;Cotrina, 1989, Bauchot, Hureau, 1990;Vaughan et al., 1992;Haimovici, 1998;Labroupoulou et al., 1999;Galván et al., 2009;Menezes et al., 2006).Adults are typically associated with low-profile hard (live) bottom, rocky, or gravel habitats (Manooch, Hassler, 1978;Grimes et al., 1982;Alekseev, 1982).In Argentina and Uruguay, P. pagrus can be found in coastal waters no deeper than 75 m, mainly from 34° to 41° 25'S.This species also occurs seasonally, mainly November to March, in the San Matías Gulf (Rio Negro) on Playa Dorada shores (41°39'S), and in the San Jose Gulf (42°15'S) and Nuevo Gulf (42°57'S) in Península Valdés (Chubut) (Galván et al., 2009;García, Molinari, 2013).Ninety-four marine and freshwater fish species belonging to 49 families cohabit in the Buenos Aires Coastal Zone, also known as the Bonaerense 2 e160127 [2] Coastal Ecosystem (BCE).The BCE supports a multispecies demersal fishery, known as "variado costero", with different commercial fleet types (craft, bay or creek, coastal and offshore).During 2015, red porgy ranked 7th in total total landings among 35 species included in the "variado costero" (Prosdocimi et al., 2016).The current scheme of extraction includes an important participation of the fleet with greater fishing power, operating with non-selective bottom trawls in the months and areas where the resource is concentrated during the reproductive activity (Lagos et al., 2013).This situation, along with an increasing trend in catches from 2003 onwards, together with the economic and fishing importance of the species, determined the need for a fishery diagnosis.For this reason from 2006, biological and fishing studies were taken up and prioritized, aiming to obtain the necessary information to do recommendations for a sustainable management of the resource.In this context, between 2009 and 2012, was established a maximum allowable catch of 6,500 t per year applicable to the Argentine -Uruguayan Common Fishing Zone (ZCPAU), which was suggested with a precautionary approach.From 2003 increased landings of this species to reach peak levels in 2008 and 2009 (close to 6500 t) followed by a decline, with fluctuating around 3,000t in 2011-2015.However, biomass estimates show a growing trend with a risk less than 10% declining below 40% virgin biomass in the medium term (10 years), maintaining current levels of exploitation (Lagos et al., 2015).
As regards the reproductive characteristics, Pagrus pagrus is a protogynous species and gonadal development proceeds through three different ways.In the first way, immature fish (less than 3 years-old) develop the testicular tissue and the ovaries degenerate before sexual maturity ("primary males" type).In the second way, the development of the gonads is completed with maturation of the ovarian zone, and the fish function as females.In the last way, after a single, or possibly repeated spawnings, females change sex and function as males ("secondary males" type) (Fostier et al., 2000).Both in captivity and in the wild, red porgy first mature at 2 years old (Cotrina, Christiansen, 1994;Kokokiris et al., 1999).Macroscopic staging and gonadosomatic index data suggest that the main spawning season for red porgy in the BCE occurs between late spring and summer (Ciechomski, Cassia, 1974).Fecundity of this species has been shown to be indeterminate (Daniel, 2003).Estimation of fecundity is often difficult since most fish produce large numbers of small eggs; but it can be particularly challenging for indeterminate spawners like red porgy because fecundity is not fixed prior to the start of spawning, so oocyte counts for individual females do not accurately reflect annual fecundity.To estimate this variable in indeterminate species, the number of batches of eggs produced by a female during the reproductive season is multiplied by batch fecundity, which is the number of oocytes spawned in a single batch (Kjesbu, 2009).Typically, the number of batches is calculated as the product of spawning season duration in days and the spawning fraction (number of days between spawning events), which is estimated as the proportion of mature females that are found to possess spawning indicators, such as hydrated oocytes or postovulatory follicles, when caught during the spawning season (Murua et al., 2003).
Information on reproductive biology of exploited fish stocks is essential for several reasons.Age or size at sexual maturity is a parameter used in most stock assessment models, directly affecting the estimation of spawning biomass and productivity.Also, the reproductive effort measured by the size or age of a species is essential in Stock-Recruitment models to understand the variability in the strength of recruitment.Other reproductive traits such as fecundity, spawning frequency and duration of the spawning season are necessary to properly estimate egg and larvae production, especially in species with indeterminate fecundity (Hunter, Macewicz, 1985).Annual changes in these variables could affect the stock productivity and produce variability in fish recruitment (Macchi et al., 2004).On the other hand, some reproductive variables may be associated with changes in the population structure, such as size or age at first maturity, which may decline over the years as a consequence of over-exploitation (Hubold, 1978;Beacham, 1983;Trippel, 1995).
We provide data on the reproductive potential of red porgy at the southern edge of their range in the western Atlantic based on macroscopic and histological gonad analysis.Because red porgy is intensely exploited in Argentina and Uruguay, the need is urgent to better understand its biology.The aim of this study was to determine size and age at sexual maturity in order to compare with previously estimations obtained in the 80'.Moreover, fecundity, spawning frequency and egg quality were estimated for first time for BCE.This information is critical for assessment and management of red porgy, and the estimated reproductive parameters may have direct application in models used to assess the status and dynamics of their populations.In some years additional samples were obtained from commercial landings from Mar del Plata port.Tab. 1 shows dates, sample sizes, and size ranges of red porgy from each trawl survey and annual summaries of the same data from commercial landings samples.Total length (TL) in centimeters and total weight in grams (TW) were recorded for each sampled fish.Individuals were sexed (males, females and hermaphrodites) and maturity stage was determined macroscopically only for males and females, using the following key: (1) immature; (2) developing; (3) spawning capable; (4) regressing; and (5) regenerating (Macchi, Pájaro, 2003;Brown-Peterson et al., 2011).In addition, during 2011 research survey, sagittal otoliths were collected for age determination following (García, Déspos, 2015).

e160127[3]
For histological examinations, gonads were preserved in 10% neutral buffered formalin (n=138 males, n=821 females and n=141 hermaphrodites).Fixed gonads were weighed (GW) to the nearest 0.1 g and a portion of tissue was removed from the centre of each gonad, dehydrated in ethanol, cleared in xylol and embedded in paraffin.Tissues were sectioned at approximately 4 μm thick and stained with Harris's haematoxylin followed by eosin counterstain (García del Moral, 1993).
To determine the seasonal pattern of the reproductive cycle, mean monthly gonadosomatic index (GSI), calculated as GW/TW*100 for each individual, and the mean estimated by month and sex, in order to analyze the general pattern of the reproductive cycle from commercial landings samples during 2011-2012.
Estimation of reproductive variables.Duration of the breeding season was estimated based on diagnosis of the maturity stages obtained by visual information and histological analysis.The variation in the monthly composition of the maturity stages was used to describe the reproductive cycle.Females were considered as reproductively active when they were capable of spawning at the time of capture or in the near future (stages 2 or 3 of the macroscopic scale) (Hunter et al., 1992).
Description of the stages of postovulatory follicles (POF) degeneration was adapted from that reported for Micropogonias furnieri and Macrodon ancylodon of the Río de la Plata area (Macchi et al., 2003;Militelli, Macchi, 2004).In these species the degenerative process of the POFs was faster than that observed for other species (Hunter, Goldberg, 1980) because of high water temperatures in the Río de la Plata area during summer (Macchi et al., 2003), so it was considered the most adequate scale for the purposes of the present study (new POFs: between 0 and 6 h after spawning; POF1: 12-h-old).
The size and age at first maturity (L 50 and A 50 , respectively) were estimated by a logistic model by length/age class using the maximum likelihood method: S(L)=1/(1+exp(-c*(L-L 50 ))) (Kendall, Stuart, 1967).All individuals with gonad stages other than immature (1) were considered mature or adult (based on macroscopic and microscopic examination), and their frequency was used as a response variable and total length was the explanatory variable.This analysis was performed during the reproductive season, to avoid biases in maturity determination, and only for those years containing samples of both immature and mature individuals (Tab.2).Coefficients of the regressions obtained for from the different sampled years were compared using a Chi-square test (Aubone, Wöhler, 2000).Due to the low number of juvenile male and hermaphrodite individuals in samples and to avoid problems in the maturation stage assignment in hermaphrodites, L 50 and A 50 were estimated with combined males and females and the hermaphrodites were not included.Spawning frequency was estimated from the incidence of females with postovulatory follicles (POFs 1), following Hunter, Goldberg (1980), and only from samples collected during the January 2011 research survey, the only cruise that covered the entire distribution area of the species.To calculate the mean and coefficient of variation of this variable we used the equations developed by Picquelle, Stauffer (1985).
Batch fecundity (BF, number of oocytes released per spawning) was estimated by using the hydrated oocyte method (Hunter et al., 1985).Samples were examined histologically to determine the presence of postovulatory follicles (POFs) and hydrated oocytes.To avoid biases when estimating batch fecundity, only ovaries with hydrated oocytes and without resent POF (less than 6h) were used, because of this only samples collected during January 2011 and December 2013 can used (Tab.2).Three pieces of ovary of approximately 0.1 g each were sampled from the anterior, middle and posterior section of each gonad, weighed (0.1 mg) and the number of hydrated oocytes counted.Batch fecundity was the product of the mean number of hydrated oocytes per unit ovarian weight and total ovarian weight (GW).
Relative fecundity (RF, number of hydrated oocytes per gram of ovary-free body weight) was estimated as the batch fecundity divided by female weight (without ovaries).The nutritional condition was obtained for each female using the Fulton´s condition factor (K= (TW/TL3)*100).The relationships between BF and the variables TL, TW (ovary-free) and K, were described using simple standard regression (Draper, Smith, 1981).Comparisons between the relationships BF versus TL obtained from different years were based on the overlapping length ranges of females applying an analysis of covariance (Draper, Smith, 1981).
The size and weight of unfertilized eggs can tentatively be used to evaluate or estimate the overall developmental potential of eggs after fertilization (Hinckley, 1990;Wootton, 1994;Trippel, 1998).For instance, the size of egg and oil drop was sometimes considered to be beneficial for the future development of the embryo.To obtain an estimate of the quality of spawning in different years, the diameter (OD), oil droplet diameter (OdD) and dry weight (DW) of hydrated oocytes from P. pagrus collected in 2010, 2011, 2012 and 2013 (Tab.2).Samples of 100 hydrated oocytes were removed from each individual, rinsed in distilled water, dried for 24h at 60ºC and weighed (0.1 mg).The oocyte diameter and oil droplet diameter were measured only for fully hydrated oocytes using an optical microscope at 4x magnification and an image analyzer program.

Results
Spawning period.Red porgy in coastal waters of the Buenos Aires Province (Argentina) and Uruguay were reproductively active between October and March, with a principal peak in December.Reproductively active females (stage 2 or 3) were observed between October and May (males October-April), with the highest percentages of spawning activity from October to January, coinciding with the highest values of GSI (Fig. 2).In addition, actively spawning females (those with hydrated oocytes and/or with post-ovulatory follicles) were observed between October and March, with the highest proportion in December (Fig. 3).Atretic oocytes were observed November through April; their incidence increased from December to a peak in March, toward the end of the reproductive season, indicating the end of spawning activity (Fig. 3).Length and age at first maturity.The estimates of length at first maturity (L 50 ) of P. pagrus collected during 2010 and 2013 didn't show significant differences (P<0.05) between years and was 24.9 cm TL (c=0.45,n=235) and 24.3 cm TL respectively (c=0.74; n=593).However, highly significant differences were observed in the logistic curve slope (P <0.01).Age at first maturity (A 50 ), only estimated for 2011, was 1.54 years (c=1.22,n = 419).The youngest mature individuals (both male and female) were estimated at 1 year old.

Tab
Spawning frequency.Females with POFs were found in the 91% of fishing hauls containing red porgy.The mean daily percentage of mature females with day 1 POFs was 58.5% (n=189; CV=0.6), equivalent to a spawning frequency of about 1-2 days for January 2011.
Fecundity.Batch fecundity (BF) estimated for P. pagrus collected in January 2011(n=30) ranged between 6974 (25 cm TL) and 110725 (39 cm TL) hydrated oocytes with a mean of 37315 (±8371).BF showed significant linear relationships (P<0.05) with total length (TL), total weight (TW, ovary free) and age but no relationship with condition factor (Fig. 4).In December 2013, BF showed no relationship with female size, weight or age but a significant linear relationship (P<0.05) with condition factor; however, sample size was small (n=7) (Fig. 4).2013 BF ranged from 7606 to 28552 hydrated oocytes, with a mean of 13071 (± 5360).Relative fecundity (RF) estimated with data of the two years ranged from 16 to 172 hydrated oocytes/g of ovary-free body weight, was highly variable, and showed no relationship to female size and age.The mean RF estimated for 2011 (90± 15 oocytes/g) differed significantly (P<0.05) from that for 2013 (32 ±12 oocytes/g).Size and weight of the oocytes.DW of 100 hydrated oocytes for the 221 red porgy we examined ranged between 1.2 and 3.6 mg.Mean DW was 2.55 ± 0.071 mg for 2010, 2.23 ± 0.075 mg for 2011, 1.51 ± 0.095 mg for 2012 and 2.02 ± 0.235 mg for 2013, and these values differed significantly (P<0.05) from each other.DW showed no significant relationship with size or condition (K) for 2011, 2012 and 2013; however in 2010 it did show a significant linear relationship (P<0.05) with total length (TL), age and weight (TW) (Fig. 5).A significant linear relationship (P<0.05) was also observed between DW and TW in 2011.
OD for the 43 red porgy with completely hydrated oocytes we examined ranged between 806 and 979 μm with a mean of 905 ± 26 μm.We found no significant relationship (P<0.05) between OD and TL, TW or K.The diameter of the oil droplet (OdD) ranged from 192 to 243 μm and averaged 215 ± 6 μm.As with OD there were no significant relationships (P<0.05) between OdD and TL, TW, or K.

Discussion
Our findings that the main spawning activity of red porgy in coastal waters of Buenos Aires Province (Argentina) and Uruguay (34º-39ºS) occurred between October and January, with a principal peak in December, are coincident to those reported by Ciechomski, Cassia (1974) and Cotrina, Christiansen (1994).Atretic oocytes were observed from November through April with a peak in March was quite similar to that of Fostier et al. (2000) for cultured P. pagrus in Greece.
Length at first maturity (L 50 ) estimated for P. pagrus in 2010 and 2013 not differed significantly between years.L 50 estimates for sexes combined of 24.9 and 24.3 cm TL in 2010 and 2013, respectively, were very similar to the only previous estimate for red porgy in the Argentine Sea, ~25 cm TL, reported by Cotrina, Christiansen (1994) and based on samples collected during 1976-1982.However, were lower than those obtained in the northern hemisphere by Daniel (2003) (28.9 cm TL) but similar to those estimated by Klibansky, Scharf (2013) (25.5 cm TL), and higher than obtained in the northeastern Gulf of Mexico by DeVries (2006) (21.1 cm TL).
Batch fecundity (BF) obtained for 2011 was a linear function of total length and ovary-free body weight, contrary to that reported by Ciechomski, Cassia (1974), which observed a potential relationship in both cases.This may be due to the fact that our estimates only included females smaller than 40 cm TL.BF estimated for 2011 showed a linear function with age, this is the first description of this relationship for the BCE.Few studies have estimated batch fecundity of P. pagrus, primarily because hydrated females are rarely encountered and the diary spawning frequency makes difficult to encountered females without recent POF.Some previous studies on red porgy fecundity were made with females in advanced maturation stage (with yolked oocytes), prior to oocyte hydration (Ciechomski, Cassia, 1974;Manooch, 1976;Klibansky, Scharf, 2013).Ciechomski, Cassia (1974) and Manooch (1976) considered the estimation of the absolute or potential fecundity as the number of total yolked oocytes early in the spawning season; because of this the estimations obtained are not comparable with those of the present paper.However, Klibansky, Scharf (2013) estimated the BF by counting oocytes of the most advanced cohort, the results obtained during 2009-2010 were similar for the smaller females but lower for larger ones (~65000 for 40 cm TL).
Values of OD and OdD were coincident with those previously reported for both wild and captive red porgy (Machinandiarena et al., 2003;Radonic et al., 2005).Our results suggest that hydrated oocytes sampled in 2010 had a greater amount of reserves, suggesting an egg quality higher than in other years analyzed.Nonetheless, this conclusion is based on limited data; hence this possibility should be further investigated.In general, the size of hydrated oocytes in red porgy was similar to that found in other sparids that range 850-1090 µm (Glamuzina et al., 1989;Greco et al., 1993) and higher than Pagellus erythrinus (Güner et al., 2004).Pagrus pagrus OD and OdD were also similar to other species of BCE like M. furnieri, C. guatucupa, U. canosai (Militelli et al., 2013) and P. brasiliensis (Militelli, 1999), but the oocyte dry weight estimated for the last one was lower (1.70 ± 0.12 mg to 100 oocytes for 2003) (Rodrigues et al., 2007).
A species' life history characteristics are important factors in its population dynamics.Population growth rate depends directly upon fecundity, survivorship, and timing of reproduction.Averaged over many generations, the three parameters must balance, or the population eventually will decline to extinction or grow exponentially (Winemiller, 2006).Winemiller (2006) proposed three endpoint strategies that result from trade-offs among age of maturation (positively correlated with mean generation time), fecundity, and survivorship.The periodic strategy corresponds to high fecundity and older age at maturity (the latter a correlate of population turnover rate) and low juvenile survivorship.The opportunistic strategy of high r (= maximum population growth rate) via rapid maturation corresponds to low values of all three parameters.The equilibrium strategy corresponds to species with low fecundity, older age at maturity, and high juvenile survivorship.Red porgy's early age at first maturity (<2 years) and almost daily spawning, although only during a limited (spring -summer) period of time, most closely corresponds to an opportunistic life history strategy, which would explain its high recovery rates of population growth (r) as well as its expanding range and colonization of new areas (García, Molinari, 2013).
Our results provide new estimations of reproductive parameters of P. pagrus in the BCE after 30 years, which will help to fill a literature gap about these topics with the aim to collaborate in the specie assessment and management.
Pagrus pagrus and oceanographic data were obtained during six research trawl surveys of the Instituto Nacional de Investigación y Desarrollo Pesquero (INIDEP) carried out in coastal waters of Argentina and Uruguay, 2008-2013 (Fig. 1).

Fig. 1 .
Fig. 1.Locations where Pagrus pagrus were collected during the research surveys carried out in coastal waters of Argentina and Uruguay between 2008 and 2013.
. 1. Date, number and length range of Pagrus pagrus sampled during six research surveys and five years commercial landings carried out between 2008 and 2013.