Introduction

The value of inland waters changes in the goods and services they provide have a strong impact on human welfare. In recent years, a raising awareness of the economic and ecologic costs caused by invasions in fresh waters has encouraged more proactive research and this has increased our understanding of invasive (biological) processes in aquatic systems (Gherardi 2007). First and foremost, predicting the likelihood of the success of a freshwater invader or predicting the invisibility of an aquatic system depends on a detailed understanding of the characteristics of the invader and of the system that is being invaded (Gherardi 2007). Together with other anthropogenic sources of disturbance, such as the impoundment of rivers (e.g. dams and weirs, water removal), water quality deterioration (e.g. pollution, eutrophication, acidification), habitat degradation and fragmentation (e.g. channelization and land use change), and overexploitation, the introduction of non-indigenous species (NIS) into fresh waters is today regarded as the main driver of biodiversity change (Millennium Ecosystem Assessment 2005; Cowx et al. 2010).

The Shadegan Wetland is the largest Ramsar convention site in Iran (designated since 1975), and has been recognized as an internationally important wetland that supports significantly rich biodiversity (Lotfi et al. 2003; Kaffashi et al. 2013). The designated area of the wetland in Ramsar Convention is 400,000 ha, excluding marine and littoral areas in Persian Gulf. Part of the wetland with the area of 327,000 ha is designated as Wildlife Refuge since 1974 and is protected as a no hunting area (Lotfi et al. 2003).

Cypriniformes with six families, 321 genera and some 3268 species (Nelson 2006) is one of the most widespread and of fishes all over the world (Moyson et al. 2016; Lauriano et al. 2016; Aliko et al. 2018). The family is found in North America, Eurasia and Africa, as well, a major element in Iran’s ichthyofauna, found in all its major drainage basins (171 species, 59.40%). The 288 species in 107 genera, 28 families, 22 orders and 3 classes reported from different Iranian basins (Esmaeili et al. 2017). Three species of Carassius, including C. carassius, C. auratus and C. gibelio, are found in Iran. The goldfish C. auratus is a cyprinid native to Eastern Asia; as a consequence of human introduction, it is now widespread throughout Europe, North and South America, New Zealand and Australia (Esmaeili et al. 2014, 2017). The biomass of this species in Shadegan Wetland from 21.64 kg/ha (9.73% of fish total biomass) in 1996 (Maramazi 1997), was increased to 67 kg/ha (16.03% of fish total biomass) in 2015 (Hashemi 2015). It seems, with the changes in the physical and ecological conditions of the wetland, species of wetland are changing.

Several studies were conducted on this species from different local, Iran (Sayadborani et al. 2013; Patimar 2009; Bagheri and Hedayati 2013). Maramazi (1997), Ansari (2001), Ansari et al. (2009), Hashemi (2010) and Hashemi et al. (2011, 2012) studied fishing status, biomass estimation of Shadegan Wetland. Lotfi et al. (2003), Kaffashi et al. (2012) and Malekmohammadi and Blouchi (2014) presented human activity and their effects on Shadegan Wetland. The objective of this study was to provide information pertaining to reproductive biology of this species in Shadegan Wetland and is the first to present complete of reproductive characteristics and population dynamics based on observations and information analysis. These data can be used for improved fish management for this wetland.

Materials and Methods

Experimental Fish and Areas

A total of 526 fishes of Carassius auratus (goldfish) were collected using fixed gill net with 15 mm mesh size from five stations, Mahshar (48°45′E, 30°33′N), Rogbe (48°33′E, 30°41′N), Khorosy (48°40′E, 30°39′N), Salmane (48°28′E, 30°40′N) and Ateish (48°40′E, 30°54′N) in the Shadegan Wetland in Khuzestan province from April 2014 to March 2015 and transported to lab with dry ice (Fig. 1). Total length of captured fish was measured to the nearest 0.01 cm and weighed to the nearest 0.01 g.

Fig. 1
figure 1

The map of Iran, location of five stations in Shadegan Wetland (Khuzestan province, South West of Iran)

Full size image

Length–Weight Relationship

The following formula was used to calculate relationship between the fish total length and weight: Wi = a × L b i (Zar 2010), where, Wi was fish weight (g), Li was fish length (mm), “a” was constant coefficient, and “b” was the equation power. The following formula was used to find significant difference between the calculated “b” and B = 3 for a fish with similar growth: t = [(s.dx)/(s.dy)] × [(lb − 3l)/(√ (l − r2)] × [√(n − 2)], where s.dx is standard deviation of fork length natural log, s.dy is standard deviation of weight natural log, “b” is slope, r2 is coefficient of determination and “n” is sample size (Zar 2010).

Condition factor or quality factor (K) was calculated to investigate trend of the fish condition over the year, using the following formula (Beckman 1984): K = W × 102/L3, where “W” was the fish weight (g) and “L” was total length (cm).

Sexual Maturation and Length at the First Maturation (L M)

Sexual maturation stages were determined both macroscopically and microscopically (7-stage key) and stages higher than 4 were considered as matured (King 2007; Sparre and Venema 1998). Logistic model and Y = 1/1 + exp (− a − bX) equation were used to estimate length at the first maturation; where, “Y” is the proportion of matured males and females to whole fish within a same length class, “x” is total length (cm) and “a” and “b” are constant coefficients (King 2007; Sparre and Venema 1998).

Estimation of Growth Parameters, Mortality Rate

The data was then pooled monthly from different stations and subsequently grouped into classes of 2 cm interval. The data were analyses using FiSAT II (FAO-ICLARM Stock Assessment Tools, http://www.fao.org/fi) as explained in details by Gayanilo et al. (2003). Growth coefficient was estimated by fitting the von Bertalanffy growth function to length frequencies data. The von Bertalanffy growth equation is defined as follows (Sparre and Venema 1998): Lt = L∞ [(1 − exp (− K (t − t0))], where Lt is length at time t, L∞ the asymptotic length, K the growth coefficient and t0 is the hypothetical time at which length is equal to zero. The t0 value estimated using the empirical equation (Pauly 1979):

$${\text{Log}}_{10} ( - t_{0} ) = - 0.3922 - 0.2752{\text{Log}}_{10} L_{\infty } - 1.038{\text{Log}}_{10} K.$$

The fitting of the best growth curve was based on the ELEFAN I program (Pauly and David 1981), which allows the fitted curve through the maximum number of peaks of the length-frequency distribution.

Total mortality (Z) was calculated from a linearized length-converted catch-curve analysis (Gayanilo and Pauly 1997) using: Ln (Ci/∆t) = a + b·ti, where Ci is the number of fish in various length classes I, \(\Delta t_{i}\) is the time needed to grow through length class i. The natural mortality rate (M) was estimated using Pauly’s empirical relationship (Pauly 1980): Log10M = 0.0066 − 0.279Log10L∞ + 0.6543Log10K + 0.4634Log10T, where L∞ is expressed in cm and T, the mean annual environmental water temperature equals to 17 °C (Hashemi 2015). Fishing mortality (F) was obtained by subtracting M from Z and exploitation rate (E) was obtained from F/Z.

Production/Biomass Ratio or Population Production (P/B)

Production/biomass or population production (specific production) is calculated using this formula P/B = 2.64W −0.35mat , where Wmat is fish weight at maturation (Randall and Minns 2000).

Comparison of L–W relationship values during season and its temporal variation carried out by T test. Statistical analyses were performed with Ms Excel, SPSS 22 software package and a significance level of p > 0.05 was adopted.

Results

During this project, from April 2014 to March 2015, was conducted in Shadegan International wetland and a total of 526 fishes of this species were caught and biometry. A total fish, 47 were male (9.81%) and 479 were female (90.18%) and their sex ratio was about one to ten and there was a significant difference between observed amounts with expected value (3.84) (p < 0.05, X2 = 177). The mean of total length and total weight for male and female was 164 ± 35 (123–264) mm, 198 ± 40 (106–322) mm and 81 ± 68 (23–308) g, 148 ± 96 (23–650) g were obtained, respectively (Table 1) and the difference in total length between two sexes (t = 6.18, p ˂ 0.05), and the total weight between the two sexes (t = 4.25, p ˂ 0.05) was significant at the level of 0.05.

Table 1 The number of sample and average values of length and weight of Carassius auratus in different months of 2014–2015 Shadegan Wetland
Full size table

The relationship of length–weight was for male genus W = 0.000008L3.13 (R2 = 0.96) and female W = 0.00001L3.10 (R2 = 0.93) and for the total fish W = 0.000009L3.11 (R2 = 0.94), and length and weight relationship was significant at the level of 0.05 (Fig. 2). The amount of b-relationship between length and weight of total fish and female sex showed positive allometric growth, male sex was observed isometric growth.

Fig. 2
figure 2

The length–weight relationship curve for Carassius auratus from the Shadegan Wetland

Full size image

The mean K values for male and female sexes were 1.57 ± 0.18 and 1.64 ± 0.64, respectively (Table 1). The distribution of the various stages of gonadal development based on the 7-step is shown in Fig. 3. This figure shows that the spawning time of April until November and the spawning period is very large.

Fig. 3
figure 3

Sexual maturity stages for Carassius auratus from the Shadegan Wetland

Full size image

The LM50 curve (length that 50% of the fish are mature) was obtained according to the length of fish and the percentage of sexual maturity in each group (Fig. 4). The maturity length and maturity weight were 163 mm [Confidence interval (CI) 162.95–163.04] and 37 g, respectively.

Fig. 4
figure 4

Length at the first maturation (LM) values Carassius auratus from the Shadegan Wetland

Full size image

The L∞, K and t0 were estimated to be 346 mm, 0.36 (year−1) and − 0.23 respectively (Fig. 5). Natural mortality (M), fishing mortality (F), total mortality (Z), and exploitation rate (E) were 0.75 (year−1), 0.77 year−1, 1.52 year−1, and 0.51 year−1 respectively. Production per biomass (P/B) ratio was 0.74 per year.

Fig. 5
figure 5

Von Bertalanffy growth curve of Carassius auratus superimposed on the restricted length frequency histogram (L∞ = 346 mm and K = 0.36 year−1)

Full size image

Discussion

There was a significant difference between the male and female sex ratio of one to one ratio, which in other studies also reported similar results with different percentages, which is due to gynogenesis pattern of this species. It is mentioned in various sources (Sayadborani et al. 2013; Lorenzoni et al. 2007; Patimar 2009; Bagheri and Hedayati 2013). In Shadegan Wetland has been reported as gynogenesis pattern of Carasobarbus luteus species with a ratio sex of 16% male and 84% female (Ghorbani et al. 2015; Hashemi 2015). The length and the weight of the fish sampled indicate that usually female sexes with weight and length greater than male sex. Various reports of C. auratus species length and weight relationship from different parts of the world are presented in Table 2 and their differences may be due to different conditions in each region.

The C. auratus fishes in Anzali wetland have a higher length and weight (mean 196 ± 137 g and 195 ± 5 mm) than Shadegan Wetlands and Gorgan river estuaries (90 ± 14 mm and 14 ± 2 g). Also, this species in Shadegan Wetland have more weight in similar lengths than Alagol and Amagol wetlands and Gorgan river estuarine (Patimar 2009; Bagheri and Hedayati 2013), which probably indicates its nutritional status. The slope of length–weight relationship implies the positive growth of this species (total fish), which is consistent with other reports (Lorenzoni et al. 2007; Patimar 2009; Bagheri and Hedayati 2013). Differences in the length–weight relationship can be due to seasonal fluctuations with environmental parameters, physiological conditions of fish at collection time, gender, gonad development, and nutritional conditions in different ecosystems (Yıldırım et al. 2002).

Goldfish fish spawning season, from February to July (peak spawning in the spring) in the various sources is given in Table 2 and it seems that the Shadegan Wetland in according to the extent of the warm months a wider period than other areas. Spawning time and duration of spawning fish, goldfish in different areas has a lot of differences with each other, which is likely due to changes in their environment. Difference in spawning time depends on the environmental characteristics also water temperature (Dutta et al. 2012). Most species of fish have spring spawning in the Shadegan Wetland, in some species have several months (Hashemi et al. 2012). Typically, fish species spawn when nutrient and environmental conditions are provided for the survival of their larvae (Hashemi 2010).

Table 2 Comparison of length and weight, maturity and spawning time of Carassius auratus with other parts of the world
Full size table

The length of maturity of this fish in Shadegan Wetland is higher than that of Trasimeno (Table 2). The time to reach sexual maturity varies between different species, and even the age or maturity may be different between the sexes or the populations or species of the same species (King 2007).

The average total length, mean of total weight, infinite length, maturing length decreased in the Shadegan Wetland region, and the growth coefficient shows an increase (Hashemi 2015). It seems that changing the physico-chemical conditions, for example, increased salinity, increasing nitrate and reducing dissolved oxygen (Hashemi 2015) of the wetland will lead to more environmental stress. One of the most important factors affecting this issue is the entry of various industrial, agricultural and human wastes into the wetland and, as a result, increasing environmental stress. In addition to the construction of dams in Khuzestan (Iran) since 1980 has also altered the hydrological regime dramatically (Hashemi 2015).

By comparing the trend of K, we can say that average female sex is more than male sex. There was no clear trend in its values, and it is expected that a relative increase in K at the time of non-reproduction of this species (winter season). The K coefficient is a well-being indicator or a relative condition factor for fish (Hashemi et al. 2012). Changes in K, different regions depend on various factors such as population density, fish diseases, nutrition, spawning time, as well as age and type of water source. Condition factor or quality factor coefficient is a useful indicator in the species biology and nutrition cycle and is another way of expressing the relation between weight and weight in a given fish (Abdul et al. 2016; Anani and Nunoo 2016). It can also be used to measure seasonal variations in fish size throughout the year (Abdul et al. 2016).

By comparing these species growth parameters in different regions, it can be said that goldfish in the river has infinite length and growth coefficient is less than wetland and this species in Anzali wetland have an infinitely length longer than Shadegan wetland and Alagol and Amagol wetland, and goldfish in Shadegan wetland has a higher growth coefficient than Anzali wetland and Alagol and Amagol wetland (Patimar 2009; Bagheri and Hedayati 2013). Differences in the infinity length and the growth coefficient from one region to another can be due to the quantity and quality of food and weather conditions (Bartulovic et al. 2004), as well as various factors that can effects include age, sex, season, year, type of nutrition, physiological conditions, differences in food availability and reproductive period (Laleyeè 2006).

The longevity of goldfish in Shadegan Wetland was estimated at 8 years, according to the formula tmax = t0 + 3/K. Regarding the short life span of this species in Shadegan Wetland, it seems that the ecological strategies of this fish in Shadegan Wetland are due to opportunistic strategy.

This species in Shadegan Wetland have natural mortality, fishing, total, as well as exploitation rate over than Anzali wetland (Sayadborani et al. 2013; Esmaeili et al. 2017) (Table 2), these results indicate different environmental condition, including water physiochemical parameters, nutritional quality and exploitation status of the fish living place.

The production of biomass or the production of this species in the Shadegan Wetland seems to be high. In the past study in this wetland, P/B rate for goldfish biomass was 0.54 per year (Hashemi 2010) and P/B ratio to fish biomass in inland waters fluctuate within the range of 0.2–3 and in tropical areas higher values than colder waters. The P/B (specific production) can be indicative of the population growth potential of fish corresponding to the production capacity of the habitat (Jenning et al. 2000; Bradford et al. 2014). These values are useful for fishery biologists to determine the estimation of fish production in different ecosystems.