Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Size, fecundity and condition factor changes in endangered delta smelt Hypomesus transpacificus over 10 generations in captivity

Size, fecundity and condition factor changes in endangered delta smelt Hypomesus transpacificus... INTRODUCTIONThe Delta smelt Hypomesus transpacificus is a type of small fish that is found exclusively in the San Francisco Estuary (Moyle et al., 2018). These fish live in both freshwater areas of the Delta and in brackish water areas such as Suisun Bay (Moyle, 2002). Once abundant, their numbers began declining in the mid‐1980s potentially because of extended drought, increased exports of water from the south Delta, introduction of invasive species and increases in contaminants and the distribution and abundance of aquatic weeds (Hobbs et al., 2017; Moyle et al., 2018). These factors led delta smelt to be federally listed as threatened under the Endangered Species Act in 1993 (U.S. Office of the Federal Register, 1993) and as endangered under the California Endangered Species Act (California Department of Fish and Wildlife [CDFW], 2021). In response to the listing of delta smelt, the Fish Conservation and Culture Laboratory (FCCL) at the University of California, Davis (UC Davis) was established in 1996 to develop culture techniques and to create a captive population of delta smelt since 2008 (Lindberg et al., 2013).In collaboration with the UC Davis Genomic Variation Laboratory (GVL), the FCCL functions as a conservation hatchery with substantial resources devoted to rearing and maintaining a refuge population of delta smelt. This population is managed to minimize inbreeding and loss of genetic variation (Fisch et al., 2013) and to replicate the wild population as closely as possible (Lindberg et al., 2013). The refuge population was developed with three main goals: (1) prevent the global extinction of the species, (2) serve as a source population for the supplementation of cultured fish to the wild population and (3) provide fish for research. This is achieved in part through the crossing of wild‐caught fish with cultured ones (fish produced at the facility; Lindberg et al., 2013), minimizing relatedness between cultured parents and equalizing family sizes at the egg stage (Fisch et al., 2013). Wild delta smelt are incorporated into the broodstock to produce the next generation. During the rearing process, every effort is made to minimize the potential effects of domestication, as a way of protecting this population that is in significant risk of extinction (Sousa‐Santos et al., 2014). The broodstock used in the FCCL rearing process mainly consists of cultured fish, with only wild fish being incorporated into the stock every year that may influence the source of parentage.The wild delta smelt population has diminished to the point where the scientific community has proposed supplementation with cultured fish, and that this approach should be experimentally evaluated as a management tool (Lessard et al., 2018). Bork et al. (2020) has stated doing nothing increases the likelihood of extinction of the single extant delta smelt population. Further declines in the wild population make recovery more difficult by reducing successful natural production, genetic diversity in the wild and numbers of wild broodstock available for the refuge population at the FCCL. Should the wild population decline to the point where adequate numbers are no longer available for broodstock, the small refuge population will inevitably lose genetic diversity, leading to increased genetic adaptation to captivity (i.e. inadvertent domestication), which may make the cultured population less suited for the supplementation (Lessard et al., 2018). For these reasons, experimental releases of cultured delta smelt are currently being conducted and evaluated (Tempel, 2022).Even though there are several captive breeding programmes designed for different species, the supplementation of cultured fish to wild populations can still pose risks through genetic or phenotypic alterations (Christie et al., 2014; Laikre et al., 2010). Domestication is an evolutionary process that involves adaptation to a captive environment, which is very different from the natural environment experienced by wild conspecifics (Harvey et al., 2016). Simply rearing wild fish in a captive environment may cause their adaptation by altering different phenotypic traits (Milla et al., 2021). Domesticated fish may differ from their wild ancestors mainly because of different genetic processes such as inbreeding, genetic drift, natural selection in captivity, relaxation of natural selection to wild conditions, artificial/active selection and epigenetics (Konstantinidis et al., 2020; Teletchea & Fontaine, 2014). Finger et al. (2018) found evidence of genetic adaptation in delta smelt cultured at the FCCL, even with the addition of wild broodstock every year. They found the crosses of delta smelt with higher levels of hatchery ancestry tend to produce a greater number of offspring. The phenotypic modification in some traits can occur because domestication causes a trade‐off between some traits to balance the allocation of resources for adaptation to the rearing environment (Lasne et al., 2018; Ruzzante, 1994). Reproduction is costly which may cause a trade‐off among reproductive investment, survival, growth and other phenotypic traits (Roff & Fairbairn, 2007). Domestication occurs after the consistent control of reproduction year after year in successive generations of fish maintained and bred in captivity (Bilio, 2007; Liao & Huang, 2000; Mylonas et al., 2010; Welcomme et al., 2010). This control of reproduction includes both broodstock management (i.e. genetic management of parents, induction of maturation and spawning, collection of gametes and eggs and incubation of eggs; Liao et al., 2001) and fish grow‐out.The goals of this study were (1) to evaluate whether any differences exist in fork length (FL), weight (BW) and fecundity (F) across all years between cultured and captive wild delta smelt, (2) identify if any difference in condition factor (K) exists between cultured and captive wild delta smelt populations by years and (3) to identify if there are any trade‐offs between measured traits (i.e. a trade‐off between length and fecundity; Roff, 1992). The findings will play a vital role in the assessment of potential impacts of introducing cultured delta smelt to wild populations, the estimation of recruitment and the development of successful management strategies.MATERIAL AND METHODSData collectionAround November of each year (seasons of 2008–2009 to 2017–2018), up to 100 adult wild delta smelt were collected (usually around 50 mm, Federal 10(a)1(A) permit #TE027742, State Scientific Collecting permit #D‐0012297859‐3 and D‐0021521915‐6 and CESA Memorandum of Understanding 2081a) from the lower Sacramento River using a lampara net (Lindberg et al., 2013; Moyle, 2002). The location of fish collected throughout the study period is shown in Figure 1 with a base map adapted from OpenStreetMap and processed with QGIS software (https://qgis.org/en/site/). Once wild delta smelt were collected, they were housed in 1100‐L circular black‐interior insulated fiberglass tanks at the FCCL. The wild fish were quarantined and received a 3‐day prophylactic treatment upon arrival. After the treatment, the fish holding tank was connected to recirculating aquaculture systems holding the laboratory‐bred broodstock. Detailed system set up, maintenance and fish care operations are described in Lindberg et al. (2013). The wild fish were weaned to newly hatched brine shrimp Artemia franciscana (Artemia International, Fairview, TX, USA) and a commercial dry feed diet (details shown later) upon arrival.1FIGURELocations of wild broodstock collection. The triangle dots represent the locations of field sites. Different colours indicate samples collected in different years, and the sizes of the points are proportional to the number of fishes caught each timeLaboratory‐bred delta smelt broodstock (cultured) was held in identical tanks as the wild delta smelt (captive). Fish were fed two types of diets at different periods of time: (i) from 2009 to early‐2016, fish were fed commercial diet consisted of 2:1 ration of 4/6 NRD (INVE Aquaculture, Salt Lake City, UT, USA) and 370 Hikari (By‐Rite Pet Supply, Hayward, CA) and (ii) from mid‐2016 to 2018, fish were fed commercial diet of BioVita starter (Bio‐Oregon, Longview, WA, USA). To implement the genetically managed breeding plan, it was necessary to provide broodfish with both an individual identification tag (as outlined by Sandford et al., 2019) and a fin clip. The tagging and fin‐clipping processes were initiated in January each year and throughout the spawning season. During the spawning season from February through May, broodstock was sorted twice a week for ripe females. As delta smelt can spawn multiple times in a spawning season (refractory time between clutches is about 40–50 days; Hung et al., 2019), only the first‐time spawners were used in this study. The ripeness of fish was determined via the egg development by gently squeezing out a very small number of eggs (less than 20) during the sorting process. When a fish was selected for spawning, eggs were extracted by applying mild pressure to the abdomen (Lindberg et al., 2013). The total numbers of eggs from individual females were volumetrically estimated (Baskerville‐Bridges et al., 2005) and recorded along with body weights (BW) and FL before the strip‐spawning process.Data of 10 generations (10 years, 2009–2018) of captive wild and cultured female delta smelt were collected and analysed (Table 1). Five main variables were used in the analyses: FL (millimeter), BW (gram), F (number), year (2009–2018) and population (cultured or wild).1TABLENumber of delta smelt used during this study over the 10 years (2009–2018)YearWild femaleCultured femaleYearWild femaleCultured female200925222201426264201010236201525254201123225201620236201225283201746235201326226201815257Length, weight and fecundity between cultured and wild delta smeltTwo‐way analyses of variance (ANOVA) were performed to explore the variation in FL and BW, whereas two‐way analysis of covariance (ANCOVA) was applied for F difference between cultured and wild populations in different years for delta smelt broodstock. In all analyses, FL, BW and F were transformed using logarithm. In the models, Year, Population and their interaction were included as categorical fixed factors, and FL was incorporated as a covariate in the ANCOVA analyses. Pairwise tests were conducted using Tukey adjustment for multiple comparisons between years to identify differences between specific years.Relationships among length, weight and fecundityThe following general linear models were used to investigate the relationship between FL and BW, FL and F and BW and F (Afshartous & Preston, 2011):FL and BW:1logFL=a0+a1logBW+a2Year+a3Population+a4Year×Population$$\begin{eqnarray}{\rm{lo}}{{\rm{g}}}\left( {FL} \right) &&= {a_{\rm{0}}} + {a_{\rm{1}}}{\rm{log}}\left( {BW} \right) + {a_{\rm{2}}}\left( {Year} \right) \nonumber\\ &&+ {a_{\rm{3}}}\left( {Population} \right) + {a_{\rm{4}}}\left( {Year \times Population} \right)\end{eqnarray}$$FL and F:2logF=a0+a1logFL+a2Year+a3Population+a4Year×Population$$\begin{eqnarray}{\rm{log}}\left( F \right) &&= {a_{\rm{0}}}{\rm{ + }}{a_{\rm{1}}}{\rm{log}}\left( {FL} \right) + {a_{\rm{2}}}\left( {Year} \right) \nonumber\\ &&+ {a_{\rm{3}}}\left( {Population} \right) + {a_{\rm{4}}}\left( {Year \times Population} \right)\end{eqnarray}$$BW and F:3logF=a0+a1logBW+a2Year+a3Population+a4Year×Population$$\begin{eqnarray}{\rm{log}}\left( F \right) &&= {a_{\rm{0}}} + {a_{\rm{1}}}{\rm{log}}\left( {BW} \right) + {a_{\rm{2}}}\left( {Year} \right) \nonumber\\ &&+ {a_{\rm{3}}}\left( {Population} \right) + {a_{\rm{4}}}\left( {Year \times Population} \right)\end{eqnarray}$$where Year (2009–2018) is a discrete variable, Population (wild or cultured) is a categorical variable, a0 is the intercept, a1–a4 are slope or effect coefficients and ‘×’ is the interaction term. As the above models incorporated most of the variables in all possible combinations and provided the highest R2 value with almost similar AIC, they were selected as the final models for the regression analyses to explore the relationships among different traits in this study.Condition factors between cultured and wild delta smeltFulton's condition factor, K (Fulton, 1902), which assumes that the shape of fish does not change with size (i.e. isometric growth), was calculated, and the mean of K was evaluated for each of the year and FL‐class intervals:4K=103×BWFL3$$\begin{equation}K{\rm{ = 1}}{{\rm{0}}^{\rm{3}}}{\rm{\; \times }}\frac{{BW}}{{F{L^{\rm{3}}}}}\end{equation}$$The following general linear model was fitted for K to explore the variation between cultured and wild populations among the 10 years:5K=a0+a1Year+a2Population+a3Year×Population$$\begin{equation}K = {a_{\rm{0}}} + {a_{\rm{1}}}\left( {Year} \right) + {a_{\rm{2}}}\left( {Population} \right) + {a_{\rm{3}}}\left( {Year \times Population} \right)\end{equation}$$The ANOVA was performed where Year, Population and their interaction were included as categorical fixed factors. ANOVA was used to explore the overall differences in K between cultured and wild populations. Tukey adjustment for multiple comparisons was used for pairwise tests between any of the two years.To find out which FL interval has the highest K, five groups of FL‐interval classes (50–59, 60–69, 70–79, 80–89 and 90–99 mm) were categorized, and the following general linear model was fitted:6K=a0+a1logFL−interval+a2Population+a3FL−interval×Population$$\begin{eqnarray}K &=& {a_{\rm{0}}}{\rm{ + }}{a_{\rm{1}}}{\rm{log}}\left( {FL{\rm{ - interval}}} \right) + {a_{\rm{2}}}\left( {Population} \right)\nonumber\\ && + {a_{\rm{3}}}\left( {FL{\rm{ - interval}} \times Population} \right)\end{eqnarray}$$The ANOVA was performed with Population and FL‐interval class, and their interaction was included as categorical fixed factors. ANOVA was used to explore the overall differences in K between cultured and wild population related to the FL‐interval class. Pairwise tests were conducted between any of the two FL‐interval classes.Statistical analysesAll analyses were performed using ‘R’ version 3.6.1 (R Development Core Team, 2021). The regression analyses, ANOVA and ANCOVA, were performed using the ‘car’ package (Fox & Weisberg, 2019). The pairwise tests were conducted using the ‘emmeans’ package (Lenth, 2020). To find the positive and negative associations (i.e. trade‐offs) among different traits, the Pearson correlation test was performed using the ‘PerformanceAnalytics’ package (Peterson & Carl, 2020). All plots were made by using the ‘ggplot2’ package (Wickham, 2016).RESULTSLength, weight and fecundity between cultured and wild delta smeltThe results show that cultured fish were significantly larger (Figure 2a, 2b) and more fecund (Figure 2c) than the captive wild fish. The ANOVA results indicated both FL and BW were significantly variable according to Year, Population and their interaction (p < 0.001). The ANCOVA results also show significant effects of Year (F9,2702 = 8.94, p < 0.001), Population (F1,2702 = 3.91, p < 0.05) and their interaction (F9,2702 = 3.34, p < 0.001) on the fecundity. The detailed findings are graphically presented in Figures 2a, 2b, 2c. The significant interactions between years and population suggest the difference between populations differed among years. However, FL in 2009, 2011 and 2017 (Figure 2a) and BW and F of wild fish in 2011 and 2017 (Figure 2b, 2c) were higher than the cultured fish.2FIGUREDifferences in (a) fork length, (b) body weight and (c) fecundity among different years between cultured and wild populations of delta smelt. Each coloured line represents the mean value of population for each year. The large square dark blue points represent the overall mean score within each population across different yearsRelationships among length, weight and fecundityThe regression models among FL‐BW, BW‐F and FL‐F showed significant relationships between variables across all years for cultured and wild populations (Tables 2–4). The regression models showed that FL (intercept = −4.7, slope = 0.003, p < 0.001), BW (intercept = −12.0, slope = 0.006, p < 0.001) and F (intercept = −51.0, slope = 0.03, p < 0.001) of the cultured population increased significantly with the years. On the other hand, FL (intercept = −0.03, slope = 0.001, p = 0.14), BW (intercept = −6.4, slope = 0.003, p = 0.18) and F (intercept = −23.0, slope = 0.01, p = 0.06) of the wild population were not significantly changed with time. The results showed that FL was dependent on population (p < 0.001) and the interaction of Year × Population (p < 0.001 and Figure 3a). Similarly, BW was significantly influenced by Population (p < 0.001) and the interaction of Year × Population (p < 0.001 and Figure 3b). No significant effect of Population (p = 0.99) and the interaction of Year × Population (p = 0.99) on F was observed.2TABLEOutputs of best fitting model investigating length–weight relationship between wild and cultured populations in different years for delta smeltVariablesEstimateSEt‐ValuepIntercept−3.030.46−6.64<0.001Body weight0.270.002131.17<0.001Year 2009 (reference)Year 20100.0060.0022.08<0.05Year 20110.010.0033.59<0.001Year 20120.020.0036.31<0.001Year 20130.020.0037.10<0.001Year 20140.030.0039.42<0.001Year 20150.040.00314.18<0.001Year 20160.040.00312.81<0.001Year 20170.020.0037.61<0.001Year 20180.030.0039.67<0.001Cultured population (reference)Wild population0.020.0062.71<0.01Year 2009 × cultured population (reference)Year 2010 × wild population−0.010.01−1.150.25Year 2011 × wild population0.0080.0090.880.38Year 2012 × wild population−0.030.009−3.40<0.001Year 2013 × wild population−0.0050.009−0.590.55Year 2014 × wild population−0.020.009−2.28<0.05Year 2015 × wild population−0.040.009−4.34<0.001Year 2016 × wild population−0.030.009−2.97<0.01Year 2017 × wild population−0.020.008−2.67<0.01Year 2018 × wild population−0.010.01−1.650.09Note: Here SE is the standard errors, t‐value is the statistical t‐value and p is the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.3TABLEOutputs of best fitting model investigating length‐fecundity relationship between wild and cultured populations in different years for delta smeltVariablesEstimateSEt‐ValuepIntercept−0.223.25−6.69<0.001Fork length2.690.0553.63<0.001Year 2009 (reference)Year 2010−0.030.02−1.390.16Year 2011−0.50.02−2.65<0.01Year 20120.0080.020.400.69Year 20130.070.023.35<0.001Year 20140.040.022.05<0.05Year 2015−0.020.02−1.280.20Year 20160.050.022.27<0.05Year 20170.030.021.550.12Year 20180.070.023.44<0.001Cultured population (reference)Wild population−0.110.05−2.33<0.05Year 2009 × cultured population (reference)Year 2010 × wild population0.090.081.100.27Year 2011 × wild population0.170.072.56<0.05Year 2012 × wild population0.140.062.19<0.05Year 2013 × wild population0.0020.060.040.96Year 2014 × wild population0.070.061.050.29Year 2015 × wild population0.040.060.600.54Year 2016 × wild population0.070.071.000.32Year 2017 × wild population0.180.063.02<0.01Year 2018 × wild population−0.110.07−1.460.15Note: Here SE is the standard errors, t‐value is the statistical t‐value and p s the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.4TABLEOutputs of best fitting model investigating weight–fecundity relationship between wild and cultured populations in different years for delta smeltVariablesEstimateSEt‐ValuepIntercept−26.612.95−9.03<0.001Weight0.860.0163.35<0.001Year 2009 (reference)Year 2010−0.010.02−0.670.50Year 2011−0.010.02−0.550.58Year 20120.050.023.04<0.01Year 20130.140.027.60<0.001Year 20140.120.026.55<0.001Year 20150.080.024.67<0.001Year 20160.160.028.69<0.001Year 20170.090.024.92<0.001Year 20180.110.026.02<0.001Cultured population (reference)Wild population−0.060.04−1.420.16Year 2009 × cultured population (reference)Year 2010 × wild population0.080.081.030.30Year 2011 × wild population0.180.063.05<0.01Year 2012 × wild population0.090.061.590.11Year 2013 × wild population−0.0040.06−0.070.94Year 2014 × wild population0.040.060.690.48Year 2015 × wild population−0.020.06−0.370.71Year 2016 × wild population0.0040.060.060.95Year 2017 × wild population0.110.051.99<0.05Year 2018 × wild population−0.120.07−1.720.08Note: Here SE is the standard errors, t‐value is the statistical t‐value and p is the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.3FIGUREThe relationship between (a) fork length and year and (b) body weight and year between cultured and wild populations of delta smelt. The solid line with 95% confidence interval (translucent band area) represents the linear regression model fit. The regression model parameters are presented in the equation where FL is the fork length (mm), BW is the body weight (g), Y is the year, P is the population and Y × P is the interaction of year and populationCondition factors between cultured and wild delta smeltA change in K of the cultured population was found between early‐ and mid‐2016. The condition factor of the cultured population was significantly decreased with years from 2009 to early‐2016 (intercept = 33.9, slope = −0.016, p < 0.001), and then it sharply increased during mid‐2016 and 2018 (intercept = 77.4, slope = 0.039, p < 0.001). No significant change of K was found in the wild population throughout the study period (intercept = 0.78, slope = 0.00004, p = 0.99). The overall regression model showed that K was significantly dependent on the population (p < 0.001) and the interaction of Year × Population (p < 0.001, Table 5 and Figure 4a). The ANOVA results indicate that the cultured population had significantly higher K (F1,2703 = 5.91, p < 0.05) than the wild population. The model also showed significant effects of year (F9,2703 = 41.59, p < 0.001) and the interaction of Year × Population (F9,2703 = 2.39, p < 0.05) on K (Figure 4b). The significant interaction between years and population suggests the difference of K between populations differed among years, and the cultured population had higher K than the wild population except for the years 2012, 2015, 2016 and 2017 (Figure 4b).5TABLEOutputs of regression model investigating condition factor and year relationship between wild and cultured populations for delta smeltVariablesEstimateSEt‐ValuepIntercept0.141.3410.65<0.001Year 2009 (reference)Year 2010−0.020.008−2.06<0.05Year 2011−0.050.008−5.80<0.001Year 2012−0.050.008−5.95<0.001Year 2013−0.080.008−9.57<0.001Year 2014−0.080.008−9.58<0.001Year 2015−0.110.008−13.25<0.001Year 2016−0.110.008−13.81<0.001Year 2017−0.060.008−7.02<0.001Year 2018−0.040.008−4.49<0.001Cultured population (reference)Wild population−0.040.02−2.45<0.05Year 2009 × cultured population (reference)Year 2010 × wild population0.010.030.360.72Year 2011 × wild population−0.010.03−0.450.65Year 2012 × wild population0.050.031.950.051Year 2013 × wild population0.0080.070.320.74Year 2014 × wild population0.030.031.110.27Year 2015 × wild population0.050.031.880.06Year 2016 × wild population0.060.032.12<0.05Year 2017 × wild population0.070.022.85<0.01Year 2018 × wild population0.0020.030.080.93Note: Here SE is the standard errors, t‐value is the statistical t‐value and p is the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.4FIGUREDifferences in condition factor (K) between cultured and wild populations of delta smelt in relation to years (a). The solid line with 95% confidence interval (translucent band area) represents the linear regression model fit. The regression model parameters are presented in the equation where K: condition factor, Y: year, P: population and Y×P: interaction of year and population. Differences in condition factor between populations are shown in (b) across different years and (c) different fork length interval classes (FL‐class). Each coloured line represents the mean value of population for each year for plot b and each FL‐class for plot c. The large square dark blue points represent the overall mean score within each population across different years for plot b and different FL‐classes for plot c.In the cultured population, K was significantly increased with the increase in FL (intercept = 0.53, slope = 0.019, p < 0.001), whereas slight but not significant variation was observed in the FL relationship in the wild population (intercept = 0.50, slope = 0.019, p = 0.35). The model showed no significant effect of Population (p = 0.94) and the interaction of FL × Population (p = 0.96) on K. The ANOVA results indicate significant effects of FL‐interval class (F4,2713 = 7.13, p < 0.001) and the interaction of FL‐interval class × Population (F4,2713 = 2.61, p < 0.05) on K, whereas no significant variation was found in the K value between the cultured and wild populations (F1,2713 = 3.42, p = 0.06 and Figure 4c). The significant interaction between FL‐interval class and Population suggests that the difference of K between populations differed among FL‐interval classes. The FL‐interval classes of 90–99 and 80–89 mm in the cultured population had significantly higher K values than other size classes of the wild population except for the 50–59 mm (Figure 4c). For the cultured population, the FL‐interval class of 90–99 mm had significantly higher K than that of 80–89 mm FL‐class (Figure 4c).Correlations among different traitsCorrelation tests showed significantly positive relationships between all traits except for FL‐K in the wild population (Table 6). No evidence of a significant trade‐off has been found between traits in any population.6TABLECorrelations (Pearson's test) among different variables of the delta smeltPopulationVariableFLBWFKCulturedFL–BW0.92 (0.001)–F0.75 (0.001)0.81 (0.001)–K0.084 (0.001)0.73 (0.001)0.31 (0.001)–WildFL–BW0.88 (0.001)–F0.68 (0.001)0.78 (0.001)–K0.059 (0.36)0.50 (0.001)0.42 (0.001)–Note: Values shown are correlation coefficients (r) and p‐values (in brackets). Significant values are given in bold font.DISCUSSIONSignificant differences in length, weight and fecundity were found between the cultured and wild populations of delta smelt at the FCCL over the last 10 generations. Cultured fish were longer, heavier and were more fecund than the wild fish. Similar results in other species have also shown cultured fish grew faster than the wild ones, such as western ruivaco Achondrostoma occidentale (Mameri et al., 2018), Atka mackerel Pleurogrammus monopterygius (McDermott et al., 2011), steelhead Oncorhynchus mykiss (Kostow, 2004), Atlantic salmon (Blanchet et al., 2008), gilthead sea bream Sparus aurata (Grigorakis et al., 2002) and brown trout (Hedenskog et al., 2002). O'Sullivan et al. (2020) showed that the fecundity of cultured Atlantic salmon was approximately 1.4 times higher than wild ones. The differences in body size and fecundity between the cultured and wild populations may be attributed to various environmental factors such as water quality, temperature, density, food availability and quality, predators, stress exposure and fish mobility and swimming scopes (Basaran et al., 2007; Favaloro & Mazzola, 2003; Patiyal et al., 2014).The observed differences in length, weight and fecundity between the cultured population at the FCCL and wild populations are unlikely to be caused by genetic differences, given the lack of evidence for significant genetic divergence between the two populations. Fisch et al. (2013) and Finger et al. (2018) did not detect any significant divergence using 12 neutral microsatellite markers. However, it is important to note that these markers may not capture genetic divergence at loci that are under selection or linked to loci under selection. Therefore, other factors such as environmental conditions and selective pressures may be contributing to the observed differences. However, we cannot rule out genetic adaptation to captivity, and some studies are ongoing to answer this question (Claussen & Philipp, 2022; Yagound et al., 2022). Nutritional and environmental differences, especially during the early life stages of wild delta smelt before they were brought into the hatchery, clearly affect the size and fecundity of them (Baxter et al., 2015; Hammock et al., 2015) and can potentially lead to the low recruitment and eventual decline of the wild population.Cultured delta smelt population was spawned and raised at the FCCL under consistent conditions for all life stages. Live prey and/or dry feed were provided consistently to ensure that fish were not food limited, there were no predators or competitors, and water quality was always managed properly. Studies have shown that cultured fish grew significantly faster than their wild counterparts due to stable food availability and quality, higher consumption rates, less exercise and energy expenditure (Islam et al., 2020; Teletchea, 2019; Wringe et al., 2016). Other researchers have also shown that many other factors can influence fecundity, such as environmental conditions (Salvanes & Braithwaite, 2005), stress (Schreck et al., 2001), fish size (Kjesbu et al., 1996) and the duration of the spawning period (Bennett, 2005; Brown et al., 2013; Hunter et al., 1985). Fish size and fecundity of delta smelt are likely to have also been influenced by these factors.The changes in size (FL and BW) of cultured delta smelt in 2011 and 2017 were possibly due to different feeding strategies implemented at the FCCL. Prior to 2011, fish were fed newly hatched brine shrimp until 160 days post hatch (DPH). In July 2010, the FCCL tried to improve culture methods to increase the survival of fish, and juvenile delta smelt were co‐fed with a dry diet together with the newly hatched brine shrimp. In 2013, the discontinuation of the diet used and the process of finding the replacement may have caused the slower growth than the subsequent years (2014 and 2015). Since 2017, fish were co‐fed newly hatched brine shrimp and the new dry feed at 80 DPH. The weaning process started at 115 DPH, and the fish were fully weaned to the dry diet at only 120 DPH (Hung et al., 2022). This alteration may have significantly improved the growth of cultured delta smelt since 2018. However, the abnormal lower growth and fecundity of the cultured population and the relatively higher growth and fecundity of the wild population during 2011 and 2017 require further investigations.This study showed that the condition of cultured delta smelt was significantly higher in most years than the population in the wild. This finding aligns with some other studies where K of cultured fish were significantly higher than their wild origins, including Atlantic cod (Grant et al., 1998), European sea bass (Handelsman et al., 2010), Atlantic halibut (Haug et al., 1989), Atlantic salmon (Glover et al., 2009) and brown trout (Serrano et al., 2009). The K of the cultured population decreased in this study from 2013 to 2016. One plausible reason could be the weaning of cultured juveniles in 2011 or the feed used might reduce their overall condition rather than the growth, that is reducing body fatness or other indices (Schloesser & Fabrizio, 2017; Vandersteen et al., 2019). The sharp increase in K from 2017 might be due to the change to a new feeding strategy. The relationship between the K and FL‐interval class was significantly strong in the case of the cultured population, which was consistent with other fishes indicating that K has been increased with the increase of body size (Blackwell et al., 2000; Nahdi et al., 2016). However, the reason of the 80–89 mm group wild fish having a significantly reduced K was unclear. Due to variable rearing conditions and other unidentified factors (i.e. stress and climate change) across years, the findings showed that K of delta smelt not only depended on population status but also years, FL‐interval class and the interaction of Year, Population and FL‐interval class.Some studies suggested trade‐offs (inverse relationship between two traits) happened in fishes when they adapted to new environments. For example, trade‐offs were found between size and reproduction in 70 Amazonian floodplain fish species (Röpke et al., 2021). Röpke et al. (2021) showed that some fish have low values of somatic investment before sexual maturity (SIBSM), indicating their early sexual maturity in life with early reduction in somatic investment, whereas some other fish have higher values of SIBSM, indicating late sexual maturity in life with longer somatic investment. According to Roff (1983), fish prioritize investing energy towards gonadal development rather than somatic growth, indicating a trade‐off between the two processes. Thus, trade‐offs between fecundity and fish size can occur depending on species, life stage, types of traits and environmental conditions (Chigbu & Sibley, 1994; Garland, 2014; Roff, 1983). In this study, findings showed no trade‐off in both populations, even though the cultured population continued to grow faster.The goals of the FCCL were to create a genetically managed refuge population that stayed as close as possible to the wild population, to provide a source population for the supplementation of cultured fish to the wild population or in the case of wild extinction, and to propagate fish for research activities (Lessard et al., 2018; Lindberg et al., 2013). The FCCL has been trying to produce a cultured population that is closer to the wild population, but it was unavoidable that the fish became domesticated. Finger et al. (2018) found that increasing levels of hatchery ancestry in a 1:1 cross led that cross to having a greater probability of producing offspring that survive to maturity the following year, signalling adaptation to captivity. Cultured delta smelt were raised in a controlled environment that might provide better environmental conditions needed for survival from eggs to adults. Cultivated parents hold an advantageous position right from the start as they ensure the successful development of their eggs, whereas wild delta smelt must rely on their survival instincts to endure in an estuary that no longer provides them with an ideal habitat (Hobbs et al., 2017; Moyle et al., 2019). The delta smelt population continues to exhibit a precipitous decline with concerns of extinction should the decline in suitable habitat continue (Hobbs et al., 2017). Releasing fish into an unsuitable or rapidly declining environment will mostly likely lead to little success of repopulation.Scientists concluded rapid progress towards the development of a viable and testable supplementation programme must be a priority for the management and conservation of delta smelt (Hobbs et al., 2017; Lessard et al., 2018). The FCCL was not trying to raise fish specifically capable to survive in the wild, but due to more recent trends and information, the supplementation of the wild delta smelt population using cultured delta smelt is planned (FWS BiOp 2019), and experimental releases are currently underway using the cultured fish spawned and reared at the FCCL.As cultured delta smelt are not subjected to the stressors of their wild counterparts, there are concerns that when they are suddenly thrust into a new environment, they might fall victim to those stressors and not fare as well, but Hung et al. (2019) found that captive‐reared sub‐adult delta smelt could survive in a semi‐natural environment with uncontrolled water quality and naturally produced wild prey through spawning and into their post spawning phase. This gives hope that these fish may survive and be able to spawn if released in the estuary. Given that cultured delta smelt tend to grow larger and produce a greater number of eggs compared to wild fish (Anderson, 1988; Cargnelli & Gross, 1996; Vincenzi et al., 2010), they may be better suited to cope with stressors. With cultured delta smelt showing a higher fecundity than wild‐origin delta smelt, it must be looked at how this could affect the wild population if the cultured fish were released. Cultured fish might be able to produce more eggs and contribute a higher percentage of individuals to the next generation than the wild ones. The success of cultured delta smelt may have two potential outcomes; on one hand, they may push the wild population towards extinction, but on the other hand, if they interbreed with wild fish and increase the reproductive success of the wild population, even though this may also raise the risk of outbreeding depression in the offspring. To try and get the best possible results with supplementation trials, it seems natural to want to release the fish most likely to survive and reproduce successfully. There is hope cultured delta smelt will meet these criteria.CONCLUSIONThe results of this study provide information on captive rearing, growth and reproductive performance of delta smelt that may benefit future culture and management efforts focused on conserving the species, including potential implications of releasing cultured fish from the FCCL into the wild. Delta smelt cultured at the FCCL have been produced with all considerations to reduce domestication, but it still happened. The cultured fish used for supplementation to the wild have differences from their wild counterparts, and this paper set out to describe those differences so managers can better understand what may happen upon the release of the fish into the wild. Given the continued decline of the species and the on‐going experimental release efforts, further study on how cultured delta smelt can benefit the species recovery is highly warranted.AUTHOR CONTRIBUTIONSConceptualization; data curation; investigation; methodology; supervision; writing – review and editing: Luke Ellison. Formal analysis; writing – review and editing: Md Moshiur Rahman. Conceptualization; methodology; supervision; writing – review and editing: Amanda J. Finger. Writing – original draft: Marade Sandford. Formal analysis; writing – review and editing: Chih‐Hsin Hsueh. Conceptualization; resources; writing – review and editing: Andrew A. Schultz. Conceptualization; data curation; funding acquisition; investigation; methodology; project administration; resources; supervision; writing – original draft; writing – review and editing: Tien‐Chieh Hung.ACKNOWLEDGEMENTSThe authors would like to thank all current and former FCCL and GVL staff, especially to Lindberg J and May B, for their hard work on the development and maintaining the delta smelt refuge population and data collection. The authors also thank USBR biologists led by Reyes R for the assistance on the fish collection and Yang W‐R for the assistance on the map image production. The study was supported by the California State Department of Fish and Game (#P0730201), CALFED Bay‐Delta Program (#1048), the California Department of Water Resources (#4600007604), the Interagency Ecological Program and the U.S. Bureau of Reclamation (#R10AC20014 and #R15AC00030). Any opinions, findings and conclusions or recommendations expressed in this article are those of the authors and do not necessarily reflect the views of the supporting agencies.CONFLICT OF INTEREST STATEMENTThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.DATA AVAILABILITY STATEMENTData that support the findings of this study are available from the corresponding author upon reasonable requests.ETHICS STATEMENTAll the animal handling protocols and rearing operations were reviewed and approved by the Institutional Animal Care and Use Committee (#18081 and 19747) at the University of California, Davis.REFERENCESAfshartous, D. & Preston, R.A. (2011) Key results of interaction models with centering. Journal of Statistics Education, 19, 1–24.Anderson, J.T. (1988) A review of size dependent survival during pre‐recruit stages of fishes in relation to recruitment. Journal of Northwest Atlantic Fishery Science, 8, 55–66.Basaran, F., Ozbilgin, H. & Ozbilgin, Y.D. (2007) Comparison of the swimming performance of farmed and wild gilthead sea bream, Sparus aurata. Aquaculture Research, 38, 452–456.Baskerville‐Bridges, B., Lindberg, J.C. & Doroshov, S.I. (2005) Manual for the intensive culture of delta smelt (Hypomesus transpacificus). Sacramento: University of California. Davis report to CALFED Bay−Delta Program, ERP‐02‐P31.Baxter, R., Brown, L., Castillo, G., Conrad, L., Culberson, S., Dekar, M. et al. (2015) An updated conceptual model of delta smelt biology: our evolving understanding of an estuarine fish. Interagency Ecological Program for the San Francisco Bay/Delta Estuary, pp. 32–68. Technical report 90. Available at: https://pubs.er.usgs.gov/publication/70141018Bennett, W.A. (2005) Critical assessment of the delta smelt population in the San Francisco estuary, California. San Francisco Estuary and Watershed Science, 3, 2–72.Bilio, M. (2007) Controlled reproduction and domestication in aquaculture‐ the current state of the art, Part I. Aquaculture Europe, 32, 5–14.Blackwell, B.G., Brown, M.L. & Willis, D.W. (2000) Relative weight (Wr) status and current use in fisheries assessment and management. Reviews in Fisheries Science, 8, 1–44.Blanchet, S., Páez, D.J., Bernatchez, L. & Dodson, J.J. (2008) An integrated comparison of captive‐bred and wild Atlantic salmon (Salmo salar): implications for supportive breeding programs. Biological Conservation, 141, 1989–1999.Bork, K., Moyle, P., Durand, J., Hung, T.‐C. & Rypel, A. (2020) Small populations in jeopardy: a delta smelt case study (December 15, 2020). 50 Environmental Law Reporter 10714 (2020). Available at SSRN: https://ssrn.com/abstract=3749669Brown, L.R., Bennett, W.A., Wagner, R.W., Morgan‐King, T., Knowles, N., Feyrer, F. et al. (2013) Implications for future survival of delta smelt from four climate change scenarios for the Sacramento–San Joaquin Delta, California. Estuaries and Coasts, 36, 754–774.California Department of Fish and Wildlife. (2021) State and federally listed endangered and threatened animals of California. North Highlands: California Department of Fish and Wildlife, The Natural Resources Agency, pp. 1–26. Available online: file:///C:/Users/Moshiur/Downloads/CNDDB_Endangered_and_Threatened_Ani.Cargnelli, L.M. & Gross, M.R. (1996) The temporal dimension in fish recruitment: birth date, body size, and size‐dependent survival in a sunfish (bluegill: Lepomis macrochirus). Canadian Journal of Fisheries and Aquatic Sciences, 53, 360–367.Chigbu, P. & Sibley, T.H. (1994) Relationship between abundance, growth, egg size and fecundity in a landlocked population of longfin smelt, Spirinchus thaleichthys. Journal of Fish Biology, 45, 1–15.Christie, M.R., Ford, M.J. & Blouin, M.S. (2014) On the reproductive success of early‐generation hatchery fish in the wild. Evolutionary Applications, 7, 883–896.Claussen, J.E. & Philipp, D.P. (2022) Assessing the role of supplementation stocking: a perspective. Fisheries Management and Ecology, 1–9.Favaloro, E. & Mazzola, A. (2003) Shape change during the growth of sharpsnout seabream reared under different conditions in a fish farm of the southern Tyrrhenian Sea. Aquacultural Engineering, 29, 57–63.Finger, A.J., Mahardja, B., Fisch, K.M., Benjamin, A., Lindberg, J., Ellison, L. et al. (2018) A conservation hatchery population of delta smelt shows evidence of genetic adaptation to captivity after 9 generations. Journal of Heredity, 109, 689–699.Fisch, K.M., Ivy, J.A., Burton, R.S. & May, B. (2013) Evaluating the performance of captive breeding techniques for conservation hatcheries: a case study of the delta smelt captive breeding program. Journal of Heredity, 104, 92–104.Fox, J. & Weisberg, S. (2019) An R companion to applied regression, 3rd edition. Thousand Oaks, CA, Sage. https://socialsciences.mcmaster.ca/jfox/Books/Companion/Fulton, T.W. (1902) Rate of growth of sea fishes. Seattle: Neill & Company.Garland, T. (2014). Trade‐offs. Current Biology, 24, R60–R61.Glover, K.A., Otterå, H., Olsen, R.E., Slinde, E., Taranger, G.L. & Skaala, Ø. (2009) A comparison of farmed, wild and hybrid Atlantic salmon (Salmo salar L.) reared under farming conditions. Aquaculture, 286, 203–210.Grant, S.M., Arown, J.A. & Boyce, D.L. (1998) Enlarged fatty livers of small juvenile cod: a comparison of laboratory‐cultured and wild juveniles. Journal of Fish Biology, 52, 1105–1114.Grigorakis, K., Alexis, M.N., Taylor, K.D.A. & Hole, M. (2002) Comparison of wild and cultured gilthead sea bream (Sparus aurata); composition, appearance and seasonal variations. International Journal of Food Science and Technology, 37, 477–484.Hammock, B.G., Hobbs, J.A., Slater, S.B., Acuña, S. & Teh, S.J. (2015) Contaminant and food limitation stress in an endangered estuarine fish. Science of the Total Environment, 532, 316–326.Handelsman, C., Claireaux, G. & Nelson, J.A. (2010) Swimming ability and ecological performance of cultured and wild European sea bass (Dicentrarchus labrax) in coastal tidal ponds. Physiological and Biochemical Zoology, 83, 435–445.Harvey, A.C., Solberg, M.F., Troianou, E., Carvalho, G.R., Taylor, M.I., Creer, S. et al. (2016) Plasticity in growth of farmed and wild Atlantic salmon: is the increased growth rate of farmed salmon caused by evolutionary adaptations to the commercial diet? BMC Evolutionary Biology, 16, 264.Haug, T., Huse, I., Kjørsvik, E. & Rabben, H. (1989) Observations on the growth of juvenile Atlantic halibut (Hippoglossus hippoglossus L.) in captivity. Aquaculture, 80, 79–86.Hedenskog, M., Petersson, E. & Järvi, T. (2002) Agonistic behavior and growth in newly emerged brown trout (Salmo trutta L) of sea‐ranched and wild origin. Aggressive Behavior, 28, 145–153.Hobbs, J.A., Moyle, P.B., Fangue, N. & Connon, R.E. (2017) Is extinction inevitable for delta smelt and Longfin Smelt? An opinion and recommendations for recovery. San Francisco Estuary and Watershed Science, 15, 1–19.Hung, T., Ellison, L., Stevenson, T., Sandford, M., Schultz, A.A. & Eads, A.R. (2022) Early weaning in endangered delta smelt: effect of weaning time on growth and survival. North American Journal of Aquaculture, 84, 249–260.Hung, T., Rosales, M., Kurobe, T., Stevenson, T., Ellison, L., Tigan, G. et al. (2019) A pilot study of the performance of captive‐reared delta smelt Hypomesus transpacificus in a semi‐natural environment. Journal of Fish Biology, 95, 1517–1522.Hunter, J.R., Lo, N.C. & Leong, R.J.H. (1985) Batch fecundity in multiple spawning fishes. Pages 61–72 in Lasker R, editor. An egg production method of estimating spawning biomass of pelagic fish: application to the Northern anchovy, Engraulis mordax. US Department of Commerce, NOAA. Technical report.Islam, S.S., Yates, M.C. & Fraser, D.J. (2020) Single generation exposure to a captive diet: a primer for domestication selection in a salmonid fish? BioRxiv, 2020.01.24.919175. https://doi.org/10.1101/2020.01.24.919175Kjesbu, O.S., Solemdal, P., Bratland, P. & Fonn, M. (1996) Variation in annual egg production in individual captive Atlantic cod (Gadus morhua). Canadian Journal of Fisheries and Aquatic Sciences, 53, 610–620.Konstantinidis, I., Sætrom, P., Mjelle, R., Nedoluzhko, A.V., Robledo, D. & Fernandes, J.M.O. (2020) Major gene expression changes and epigenetic remodelling in Nile tilapia muscle after just one generation of domestication. Epigenetics, 15, 1052–1067.Kostow, K.E. (2004) Differences in juvenile phenotypes and survival between hatchery stocks and a natural population provide evidence for modified selection due to captive breeding. Canadian Journal of Fisheries and Aquatic Sciences, 61, 577–589.Laikre, L., Schwartz, M.K., Waples, R.S. & Ryman, N. (2010) Compromising genetic diversity in the wild: unmonitored large‐scale release of plants and animals. Trends in Ecology & Evolution, 25, 520–529.Lasne, E., Leblanc, C.A.‐L. & Gillet, C. (2018) Egg size versus number of offspring trade‐off: female age rather than size matters in a domesticated Arctic charr population. Evolutionary Biology, 45, 105–112.Lenth, R. (2020) emmeans: estimated marginal means, aka least‐squares means. R package version 1.5.2‐1. https://CRAN.R‐project.org/package=emmeansLessard, J., Cavallo, B., Anders, P., Sommer, T., Schreier, B., Gille, D. et al. (2018) Considerations for the use of captive‐reared delta smelt for species recovery and research. San Francisco Estuary and Watershed Science, 16, article 3.Liao, I.C. & Huang, Y.S. (2000) Methodological approach used for the domestication of potential candidates for aquaculture. Cahiers Options Méditerranéennes, 47, 97–107.Liao, I.C., Su, H.M. & Chang, E.Y. (2001) Techniques in finfish larviculture in Taiwan. Aquaculture, 200, 1–31.Lindberg, J.C., Tigan, G., Ellison, L., Rettinghouse, T., Nagel, M.M. & Fisch, K.M. (2013) Aquaculture methods for a genetically managed population of endangered delta smelt. North American Journal of Aquaculture, 75, 186–196.Mameri, D., Sousa‐Santos, C., Robalo, J.I. & Magalhães, M.F. (2018) Growth and age structure in captive and wild stocks of the endangered western Ruivaco Achondrostoma occidentale (cyprinidae). Limnetica, 37, 105–115.McDermott, S.F., Cooper, D.W., Guthridge, J.L., Spies, I.B., Canino, M.F., Woods, P. et al. (2011) Effects of maternal growth on fecundity and egg quality of wild and captive Atka mackerel. Marine and Coastal Fisheries, 3, 324–335.Milla, S., Pasquet, A., El Mohajer, L. & Fontaine, P. (2021) How domestication alters fish phenotypes. Reviews in Aquaculture, 13, 388–405.Moyle, P., Bork, K., Durand, J., Hung, T.‐C. & Rypel, A. (2019) Futures for delta smelt. California WaterBlog. Available at: https://californiawaterblog.com/2019/12/15/futures‐for‐delta‐smelt/Moyle, P.B. (2002) Inland fishes of California. Berkeley, CA: University of California Press.Moyle, P.B., Hobbs, J.A. & Durand, J.R. (2018) delta smelt and water politics in California. Fisheries, 43, 42–50.Mylonas, C.C., De La Gándara, F., Corriero, A. & Ríos, A.B. (2010) Atlantic bluefin tuna (Thunnus thynnus) farming and fattening in the Mediterranean sea. Reviews in Fisheries Science, 18, 266–280.Nahdi, A.A.L., Garcia de Leaniz, C. & King, A.J. (2016) Spatio‐temporal variation in length‐weight relationships and condition of the ribbonfish Trichiurus lepturus (Linnaeus, 1758): implications for fisheries management. PLoS One, 11, e0161989.O'Sullivan, R.J., Aykanat, T., Johnston, S.E., Rogan, G., Poole, R., Prodöhl, P.A. et al. (2020) Captive‐bred Atlantic salmon released into the wild have fewer offspring than wild‐bred fish and decrease population productivity. Proceedings of the Royal Society B: Biological Sciences, 287, 20201671.Patiyal, R.S., Mir, J.I., Sharma, R.C., Chandra, S. & Mahanta, P.C. (2014) Pattern of meristic and morphometric variations between wild and captive stocks of endangered Tor putitora (Hamilton 1822) using multivariate statistical analysis methods. Proceedings of the National Academy of Sciences, India, Section B: Biological Sciences, 84, 123–129.Peterson, B.G. & Carl, P. (2020) PerformanceAnalytics: econometric tools for performance and risk analysis. R package version 2.0.4. https://CRAN.R‐project.org/package=PerformanceAnalyticsR Development Core Team. (2021) R: a language and environment for statistical computing, version 4.0.5. Vienna, Austria: R Foundation for Statistical Computing. Available online: https://www.r‐project.org/ [Accessed 12 July 2022].Roff, D.A. (1983) An allocation model of growth and reproduction in fish. Canadian Journal of Fisheries and Aquatic Sciences, 40, 1395–1404.Roff, D.A. (1992) The evolution of life histories: theory and analysis. New York: Chapman & Hall, pp. 47–56.Roff, D.A. & Fairbairn, D.J. (2007) The evolution of trade‐offs: where are we? Journal of Evolutionary Biology, 20, 433–447.Röpke, C., Pires, T.H.S., Zuanon, J., Freitas, C.E.C., Hernandes, M.C., Souza, F. et al. (2021) Growth–reproduction trade‐off and fecundity regulate population stability in Amazon floodplain fishes. Freshwater Biology, 66, 1101–1109.Ruzzante, D.E. (1994) Domestication effects on aggressive and schooling behavior in fish. Aquaculture, 120, 1–24.Salvanes, A.G.V. & Braithwaite, V.A. (2005) Exposure to variable spatial information in the early rearing environment generates asymmetries in social interactions in cod (Gadus morhua). Behavioral Ecology and Sociobiology, 59, 250–257.Sandford, M., Castillo, G. & Hung, T. (2019) A review of fish identification methods applied on small fish. Reviews in Aquaculture, 12, 542–554.Schloesser, R.W. & Fabrizio, M.C. (2017) Condition indices as surrogates of energy density and lipid content in juveniles of three fish species. Transactions of the American Fisheries Society, 146, 1058–1069.Schreck, C.B., Contreras‐Sanchez, W. & Fitzpatrick, M.S. (2001) Effects of stress on fish reproduction, gamete quality, and progeny. In: Reproductive biotechnology in finfish aquaculture. Amsterdam: Elsevier, pp. 3–24.Serrano, I., Larsson, S. & Eriksson, L.‐O. (2009) Migration performance of wild and hatchery sea trout (Salmo trutta L.) smolts—implications for compensatory hatchery programs. Fisheries Research, 99, 210–215.Sousa‐Santos, C., Gil, F. & Almada, V.C. (2014) Ex situ reproduction of Portuguese endangered cyprinids in the context of their conservation. Ichthyological Research, 61, 193–198.Teletchea, F. (2019) Fish domestication: an overview. In: Teletchea F. (Ed.) Animal domestication. London, UK: IntechOpen, pp. 69–90.Teletchea, F. & Fontaine, P. (2014) Levels of domestication in fish: implications for the sustainable future of aquaculture. Fish and Fisheries, 15, 181–195.Tempel, T. (2022) History in the making: updates on the first experimental release of captive‐reared delta smelt. California Department of Water Resources. Available at: https://nrm.dfg.ca.gov/FileHandler.ashx?DocumentID=199263&inlineU.S. Office of the Federal Register. (1993) Endangered and threatened wildlife and plants: determination of threatened status for the delta smelt. Federal Register, 58, 12854–12864.Vandersteen, W.E., Leggatt, R., Sundström, L.F. & Devlin, R.H. (2019) Importance of experimental environmental conditions in estimating risks and associated uncertainty of transgenic fish prior to entry into nature. Scientific Reports, 9, 406.Vincenzi, S., Crivelli, A.J., Jesensek, D. & De Leo, G.A. (2010) Individual growth and its implications for the recruitment dynamics of stream‐dwelling marble trout (Salmo marmoratus). Ecology of Freshwater Fish, 19, 477–486.Welcomme, R.L., Cowx, I.G., Coates, D., Béné, C., Funge‐Smith, S., Halls, A. et al. (2010) Inland capture fisheries. Philosophical Transactions of the Royal Society B: Biological Sciences, 365, 2881–2896.Wickham, H. (2016) ggplot2: elegant graphics for data analysis. New York: Springer‐Verlag.Wringe, B.F., Purchase, C.F. & Fleming, I.A. (2016) In search of a “cultured fish phenotype”: a systematic review, meta‐analysis and vote‐counting analysis. Reviews in Fish Biology and Fisheries, 26, 351–373.Yagound, B., West, A.J., Richardson, M.F., Gruber, J., Reid, J.G., Whiting, M.J. et al. (2022) Captivity induces large and population‐dependent brain transcriptomic changes in wild‐caught cane toads (Rhinella marina). Molecular Ecology, 31, 4949–4961. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Aquaculture Fish and Fisheries Wiley

Size, fecundity and condition factor changes in endangered delta smelt Hypomesus transpacificus over 10 generations in captivity

Loading next page...
 
/lp/wiley/size-fecundity-and-condition-factor-changes-in-endangered-delta-smelt-Rf02lXosHp

References (64)

Publisher
Wiley
Copyright
© 2023 The Authors. Aquaculture, Fish and Fisheries published by John Wiley & Sons Ltd.
eISSN
2693-8847
DOI
10.1002/aff2.116
Publisher site
See Article on Publisher Site

Abstract

INTRODUCTIONThe Delta smelt Hypomesus transpacificus is a type of small fish that is found exclusively in the San Francisco Estuary (Moyle et al., 2018). These fish live in both freshwater areas of the Delta and in brackish water areas such as Suisun Bay (Moyle, 2002). Once abundant, their numbers began declining in the mid‐1980s potentially because of extended drought, increased exports of water from the south Delta, introduction of invasive species and increases in contaminants and the distribution and abundance of aquatic weeds (Hobbs et al., 2017; Moyle et al., 2018). These factors led delta smelt to be federally listed as threatened under the Endangered Species Act in 1993 (U.S. Office of the Federal Register, 1993) and as endangered under the California Endangered Species Act (California Department of Fish and Wildlife [CDFW], 2021). In response to the listing of delta smelt, the Fish Conservation and Culture Laboratory (FCCL) at the University of California, Davis (UC Davis) was established in 1996 to develop culture techniques and to create a captive population of delta smelt since 2008 (Lindberg et al., 2013).In collaboration with the UC Davis Genomic Variation Laboratory (GVL), the FCCL functions as a conservation hatchery with substantial resources devoted to rearing and maintaining a refuge population of delta smelt. This population is managed to minimize inbreeding and loss of genetic variation (Fisch et al., 2013) and to replicate the wild population as closely as possible (Lindberg et al., 2013). The refuge population was developed with three main goals: (1) prevent the global extinction of the species, (2) serve as a source population for the supplementation of cultured fish to the wild population and (3) provide fish for research. This is achieved in part through the crossing of wild‐caught fish with cultured ones (fish produced at the facility; Lindberg et al., 2013), minimizing relatedness between cultured parents and equalizing family sizes at the egg stage (Fisch et al., 2013). Wild delta smelt are incorporated into the broodstock to produce the next generation. During the rearing process, every effort is made to minimize the potential effects of domestication, as a way of protecting this population that is in significant risk of extinction (Sousa‐Santos et al., 2014). The broodstock used in the FCCL rearing process mainly consists of cultured fish, with only wild fish being incorporated into the stock every year that may influence the source of parentage.The wild delta smelt population has diminished to the point where the scientific community has proposed supplementation with cultured fish, and that this approach should be experimentally evaluated as a management tool (Lessard et al., 2018). Bork et al. (2020) has stated doing nothing increases the likelihood of extinction of the single extant delta smelt population. Further declines in the wild population make recovery more difficult by reducing successful natural production, genetic diversity in the wild and numbers of wild broodstock available for the refuge population at the FCCL. Should the wild population decline to the point where adequate numbers are no longer available for broodstock, the small refuge population will inevitably lose genetic diversity, leading to increased genetic adaptation to captivity (i.e. inadvertent domestication), which may make the cultured population less suited for the supplementation (Lessard et al., 2018). For these reasons, experimental releases of cultured delta smelt are currently being conducted and evaluated (Tempel, 2022).Even though there are several captive breeding programmes designed for different species, the supplementation of cultured fish to wild populations can still pose risks through genetic or phenotypic alterations (Christie et al., 2014; Laikre et al., 2010). Domestication is an evolutionary process that involves adaptation to a captive environment, which is very different from the natural environment experienced by wild conspecifics (Harvey et al., 2016). Simply rearing wild fish in a captive environment may cause their adaptation by altering different phenotypic traits (Milla et al., 2021). Domesticated fish may differ from their wild ancestors mainly because of different genetic processes such as inbreeding, genetic drift, natural selection in captivity, relaxation of natural selection to wild conditions, artificial/active selection and epigenetics (Konstantinidis et al., 2020; Teletchea & Fontaine, 2014). Finger et al. (2018) found evidence of genetic adaptation in delta smelt cultured at the FCCL, even with the addition of wild broodstock every year. They found the crosses of delta smelt with higher levels of hatchery ancestry tend to produce a greater number of offspring. The phenotypic modification in some traits can occur because domestication causes a trade‐off between some traits to balance the allocation of resources for adaptation to the rearing environment (Lasne et al., 2018; Ruzzante, 1994). Reproduction is costly which may cause a trade‐off among reproductive investment, survival, growth and other phenotypic traits (Roff & Fairbairn, 2007). Domestication occurs after the consistent control of reproduction year after year in successive generations of fish maintained and bred in captivity (Bilio, 2007; Liao & Huang, 2000; Mylonas et al., 2010; Welcomme et al., 2010). This control of reproduction includes both broodstock management (i.e. genetic management of parents, induction of maturation and spawning, collection of gametes and eggs and incubation of eggs; Liao et al., 2001) and fish grow‐out.The goals of this study were (1) to evaluate whether any differences exist in fork length (FL), weight (BW) and fecundity (F) across all years between cultured and captive wild delta smelt, (2) identify if any difference in condition factor (K) exists between cultured and captive wild delta smelt populations by years and (3) to identify if there are any trade‐offs between measured traits (i.e. a trade‐off between length and fecundity; Roff, 1992). The findings will play a vital role in the assessment of potential impacts of introducing cultured delta smelt to wild populations, the estimation of recruitment and the development of successful management strategies.MATERIAL AND METHODSData collectionAround November of each year (seasons of 2008–2009 to 2017–2018), up to 100 adult wild delta smelt were collected (usually around 50 mm, Federal 10(a)1(A) permit #TE027742, State Scientific Collecting permit #D‐0012297859‐3 and D‐0021521915‐6 and CESA Memorandum of Understanding 2081a) from the lower Sacramento River using a lampara net (Lindberg et al., 2013; Moyle, 2002). The location of fish collected throughout the study period is shown in Figure 1 with a base map adapted from OpenStreetMap and processed with QGIS software (https://qgis.org/en/site/). Once wild delta smelt were collected, they were housed in 1100‐L circular black‐interior insulated fiberglass tanks at the FCCL. The wild fish were quarantined and received a 3‐day prophylactic treatment upon arrival. After the treatment, the fish holding tank was connected to recirculating aquaculture systems holding the laboratory‐bred broodstock. Detailed system set up, maintenance and fish care operations are described in Lindberg et al. (2013). The wild fish were weaned to newly hatched brine shrimp Artemia franciscana (Artemia International, Fairview, TX, USA) and a commercial dry feed diet (details shown later) upon arrival.1FIGURELocations of wild broodstock collection. The triangle dots represent the locations of field sites. Different colours indicate samples collected in different years, and the sizes of the points are proportional to the number of fishes caught each timeLaboratory‐bred delta smelt broodstock (cultured) was held in identical tanks as the wild delta smelt (captive). Fish were fed two types of diets at different periods of time: (i) from 2009 to early‐2016, fish were fed commercial diet consisted of 2:1 ration of 4/6 NRD (INVE Aquaculture, Salt Lake City, UT, USA) and 370 Hikari (By‐Rite Pet Supply, Hayward, CA) and (ii) from mid‐2016 to 2018, fish were fed commercial diet of BioVita starter (Bio‐Oregon, Longview, WA, USA). To implement the genetically managed breeding plan, it was necessary to provide broodfish with both an individual identification tag (as outlined by Sandford et al., 2019) and a fin clip. The tagging and fin‐clipping processes were initiated in January each year and throughout the spawning season. During the spawning season from February through May, broodstock was sorted twice a week for ripe females. As delta smelt can spawn multiple times in a spawning season (refractory time between clutches is about 40–50 days; Hung et al., 2019), only the first‐time spawners were used in this study. The ripeness of fish was determined via the egg development by gently squeezing out a very small number of eggs (less than 20) during the sorting process. When a fish was selected for spawning, eggs were extracted by applying mild pressure to the abdomen (Lindberg et al., 2013). The total numbers of eggs from individual females were volumetrically estimated (Baskerville‐Bridges et al., 2005) and recorded along with body weights (BW) and FL before the strip‐spawning process.Data of 10 generations (10 years, 2009–2018) of captive wild and cultured female delta smelt were collected and analysed (Table 1). Five main variables were used in the analyses: FL (millimeter), BW (gram), F (number), year (2009–2018) and population (cultured or wild).1TABLENumber of delta smelt used during this study over the 10 years (2009–2018)YearWild femaleCultured femaleYearWild femaleCultured female200925222201426264201010236201525254201123225201620236201225283201746235201326226201815257Length, weight and fecundity between cultured and wild delta smeltTwo‐way analyses of variance (ANOVA) were performed to explore the variation in FL and BW, whereas two‐way analysis of covariance (ANCOVA) was applied for F difference between cultured and wild populations in different years for delta smelt broodstock. In all analyses, FL, BW and F were transformed using logarithm. In the models, Year, Population and their interaction were included as categorical fixed factors, and FL was incorporated as a covariate in the ANCOVA analyses. Pairwise tests were conducted using Tukey adjustment for multiple comparisons between years to identify differences between specific years.Relationships among length, weight and fecundityThe following general linear models were used to investigate the relationship between FL and BW, FL and F and BW and F (Afshartous & Preston, 2011):FL and BW:1logFL=a0+a1logBW+a2Year+a3Population+a4Year×Population$$\begin{eqnarray}{\rm{lo}}{{\rm{g}}}\left( {FL} \right) &&= {a_{\rm{0}}} + {a_{\rm{1}}}{\rm{log}}\left( {BW} \right) + {a_{\rm{2}}}\left( {Year} \right) \nonumber\\ &&+ {a_{\rm{3}}}\left( {Population} \right) + {a_{\rm{4}}}\left( {Year \times Population} \right)\end{eqnarray}$$FL and F:2logF=a0+a1logFL+a2Year+a3Population+a4Year×Population$$\begin{eqnarray}{\rm{log}}\left( F \right) &&= {a_{\rm{0}}}{\rm{ + }}{a_{\rm{1}}}{\rm{log}}\left( {FL} \right) + {a_{\rm{2}}}\left( {Year} \right) \nonumber\\ &&+ {a_{\rm{3}}}\left( {Population} \right) + {a_{\rm{4}}}\left( {Year \times Population} \right)\end{eqnarray}$$BW and F:3logF=a0+a1logBW+a2Year+a3Population+a4Year×Population$$\begin{eqnarray}{\rm{log}}\left( F \right) &&= {a_{\rm{0}}} + {a_{\rm{1}}}{\rm{log}}\left( {BW} \right) + {a_{\rm{2}}}\left( {Year} \right) \nonumber\\ &&+ {a_{\rm{3}}}\left( {Population} \right) + {a_{\rm{4}}}\left( {Year \times Population} \right)\end{eqnarray}$$where Year (2009–2018) is a discrete variable, Population (wild or cultured) is a categorical variable, a0 is the intercept, a1–a4 are slope or effect coefficients and ‘×’ is the interaction term. As the above models incorporated most of the variables in all possible combinations and provided the highest R2 value with almost similar AIC, they were selected as the final models for the regression analyses to explore the relationships among different traits in this study.Condition factors between cultured and wild delta smeltFulton's condition factor, K (Fulton, 1902), which assumes that the shape of fish does not change with size (i.e. isometric growth), was calculated, and the mean of K was evaluated for each of the year and FL‐class intervals:4K=103×BWFL3$$\begin{equation}K{\rm{ = 1}}{{\rm{0}}^{\rm{3}}}{\rm{\; \times }}\frac{{BW}}{{F{L^{\rm{3}}}}}\end{equation}$$The following general linear model was fitted for K to explore the variation between cultured and wild populations among the 10 years:5K=a0+a1Year+a2Population+a3Year×Population$$\begin{equation}K = {a_{\rm{0}}} + {a_{\rm{1}}}\left( {Year} \right) + {a_{\rm{2}}}\left( {Population} \right) + {a_{\rm{3}}}\left( {Year \times Population} \right)\end{equation}$$The ANOVA was performed where Year, Population and their interaction were included as categorical fixed factors. ANOVA was used to explore the overall differences in K between cultured and wild populations. Tukey adjustment for multiple comparisons was used for pairwise tests between any of the two years.To find out which FL interval has the highest K, five groups of FL‐interval classes (50–59, 60–69, 70–79, 80–89 and 90–99 mm) were categorized, and the following general linear model was fitted:6K=a0+a1logFL−interval+a2Population+a3FL−interval×Population$$\begin{eqnarray}K &=& {a_{\rm{0}}}{\rm{ + }}{a_{\rm{1}}}{\rm{log}}\left( {FL{\rm{ - interval}}} \right) + {a_{\rm{2}}}\left( {Population} \right)\nonumber\\ && + {a_{\rm{3}}}\left( {FL{\rm{ - interval}} \times Population} \right)\end{eqnarray}$$The ANOVA was performed with Population and FL‐interval class, and their interaction was included as categorical fixed factors. ANOVA was used to explore the overall differences in K between cultured and wild population related to the FL‐interval class. Pairwise tests were conducted between any of the two FL‐interval classes.Statistical analysesAll analyses were performed using ‘R’ version 3.6.1 (R Development Core Team, 2021). The regression analyses, ANOVA and ANCOVA, were performed using the ‘car’ package (Fox & Weisberg, 2019). The pairwise tests were conducted using the ‘emmeans’ package (Lenth, 2020). To find the positive and negative associations (i.e. trade‐offs) among different traits, the Pearson correlation test was performed using the ‘PerformanceAnalytics’ package (Peterson & Carl, 2020). All plots were made by using the ‘ggplot2’ package (Wickham, 2016).RESULTSLength, weight and fecundity between cultured and wild delta smeltThe results show that cultured fish were significantly larger (Figure 2a, 2b) and more fecund (Figure 2c) than the captive wild fish. The ANOVA results indicated both FL and BW were significantly variable according to Year, Population and their interaction (p < 0.001). The ANCOVA results also show significant effects of Year (F9,2702 = 8.94, p < 0.001), Population (F1,2702 = 3.91, p < 0.05) and their interaction (F9,2702 = 3.34, p < 0.001) on the fecundity. The detailed findings are graphically presented in Figures 2a, 2b, 2c. The significant interactions between years and population suggest the difference between populations differed among years. However, FL in 2009, 2011 and 2017 (Figure 2a) and BW and F of wild fish in 2011 and 2017 (Figure 2b, 2c) were higher than the cultured fish.2FIGUREDifferences in (a) fork length, (b) body weight and (c) fecundity among different years between cultured and wild populations of delta smelt. Each coloured line represents the mean value of population for each year. The large square dark blue points represent the overall mean score within each population across different yearsRelationships among length, weight and fecundityThe regression models among FL‐BW, BW‐F and FL‐F showed significant relationships between variables across all years for cultured and wild populations (Tables 2–4). The regression models showed that FL (intercept = −4.7, slope = 0.003, p < 0.001), BW (intercept = −12.0, slope = 0.006, p < 0.001) and F (intercept = −51.0, slope = 0.03, p < 0.001) of the cultured population increased significantly with the years. On the other hand, FL (intercept = −0.03, slope = 0.001, p = 0.14), BW (intercept = −6.4, slope = 0.003, p = 0.18) and F (intercept = −23.0, slope = 0.01, p = 0.06) of the wild population were not significantly changed with time. The results showed that FL was dependent on population (p < 0.001) and the interaction of Year × Population (p < 0.001 and Figure 3a). Similarly, BW was significantly influenced by Population (p < 0.001) and the interaction of Year × Population (p < 0.001 and Figure 3b). No significant effect of Population (p = 0.99) and the interaction of Year × Population (p = 0.99) on F was observed.2TABLEOutputs of best fitting model investigating length–weight relationship between wild and cultured populations in different years for delta smeltVariablesEstimateSEt‐ValuepIntercept−3.030.46−6.64<0.001Body weight0.270.002131.17<0.001Year 2009 (reference)Year 20100.0060.0022.08<0.05Year 20110.010.0033.59<0.001Year 20120.020.0036.31<0.001Year 20130.020.0037.10<0.001Year 20140.030.0039.42<0.001Year 20150.040.00314.18<0.001Year 20160.040.00312.81<0.001Year 20170.020.0037.61<0.001Year 20180.030.0039.67<0.001Cultured population (reference)Wild population0.020.0062.71<0.01Year 2009 × cultured population (reference)Year 2010 × wild population−0.010.01−1.150.25Year 2011 × wild population0.0080.0090.880.38Year 2012 × wild population−0.030.009−3.40<0.001Year 2013 × wild population−0.0050.009−0.590.55Year 2014 × wild population−0.020.009−2.28<0.05Year 2015 × wild population−0.040.009−4.34<0.001Year 2016 × wild population−0.030.009−2.97<0.01Year 2017 × wild population−0.020.008−2.67<0.01Year 2018 × wild population−0.010.01−1.650.09Note: Here SE is the standard errors, t‐value is the statistical t‐value and p is the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.3TABLEOutputs of best fitting model investigating length‐fecundity relationship between wild and cultured populations in different years for delta smeltVariablesEstimateSEt‐ValuepIntercept−0.223.25−6.69<0.001Fork length2.690.0553.63<0.001Year 2009 (reference)Year 2010−0.030.02−1.390.16Year 2011−0.50.02−2.65<0.01Year 20120.0080.020.400.69Year 20130.070.023.35<0.001Year 20140.040.022.05<0.05Year 2015−0.020.02−1.280.20Year 20160.050.022.27<0.05Year 20170.030.021.550.12Year 20180.070.023.44<0.001Cultured population (reference)Wild population−0.110.05−2.33<0.05Year 2009 × cultured population (reference)Year 2010 × wild population0.090.081.100.27Year 2011 × wild population0.170.072.56<0.05Year 2012 × wild population0.140.062.19<0.05Year 2013 × wild population0.0020.060.040.96Year 2014 × wild population0.070.061.050.29Year 2015 × wild population0.040.060.600.54Year 2016 × wild population0.070.071.000.32Year 2017 × wild population0.180.063.02<0.01Year 2018 × wild population−0.110.07−1.460.15Note: Here SE is the standard errors, t‐value is the statistical t‐value and p s the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.4TABLEOutputs of best fitting model investigating weight–fecundity relationship between wild and cultured populations in different years for delta smeltVariablesEstimateSEt‐ValuepIntercept−26.612.95−9.03<0.001Weight0.860.0163.35<0.001Year 2009 (reference)Year 2010−0.010.02−0.670.50Year 2011−0.010.02−0.550.58Year 20120.050.023.04<0.01Year 20130.140.027.60<0.001Year 20140.120.026.55<0.001Year 20150.080.024.67<0.001Year 20160.160.028.69<0.001Year 20170.090.024.92<0.001Year 20180.110.026.02<0.001Cultured population (reference)Wild population−0.060.04−1.420.16Year 2009 × cultured population (reference)Year 2010 × wild population0.080.081.030.30Year 2011 × wild population0.180.063.05<0.01Year 2012 × wild population0.090.061.590.11Year 2013 × wild population−0.0040.06−0.070.94Year 2014 × wild population0.040.060.690.48Year 2015 × wild population−0.020.06−0.370.71Year 2016 × wild population0.0040.060.060.95Year 2017 × wild population0.110.051.99<0.05Year 2018 × wild population−0.120.07−1.720.08Note: Here SE is the standard errors, t‐value is the statistical t‐value and p is the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.3FIGUREThe relationship between (a) fork length and year and (b) body weight and year between cultured and wild populations of delta smelt. The solid line with 95% confidence interval (translucent band area) represents the linear regression model fit. The regression model parameters are presented in the equation where FL is the fork length (mm), BW is the body weight (g), Y is the year, P is the population and Y × P is the interaction of year and populationCondition factors between cultured and wild delta smeltA change in K of the cultured population was found between early‐ and mid‐2016. The condition factor of the cultured population was significantly decreased with years from 2009 to early‐2016 (intercept = 33.9, slope = −0.016, p < 0.001), and then it sharply increased during mid‐2016 and 2018 (intercept = 77.4, slope = 0.039, p < 0.001). No significant change of K was found in the wild population throughout the study period (intercept = 0.78, slope = 0.00004, p = 0.99). The overall regression model showed that K was significantly dependent on the population (p < 0.001) and the interaction of Year × Population (p < 0.001, Table 5 and Figure 4a). The ANOVA results indicate that the cultured population had significantly higher K (F1,2703 = 5.91, p < 0.05) than the wild population. The model also showed significant effects of year (F9,2703 = 41.59, p < 0.001) and the interaction of Year × Population (F9,2703 = 2.39, p < 0.05) on K (Figure 4b). The significant interaction between years and population suggests the difference of K between populations differed among years, and the cultured population had higher K than the wild population except for the years 2012, 2015, 2016 and 2017 (Figure 4b).5TABLEOutputs of regression model investigating condition factor and year relationship between wild and cultured populations for delta smeltVariablesEstimateSEt‐ValuepIntercept0.141.3410.65<0.001Year 2009 (reference)Year 2010−0.020.008−2.06<0.05Year 2011−0.050.008−5.80<0.001Year 2012−0.050.008−5.95<0.001Year 2013−0.080.008−9.57<0.001Year 2014−0.080.008−9.58<0.001Year 2015−0.110.008−13.25<0.001Year 2016−0.110.008−13.81<0.001Year 2017−0.060.008−7.02<0.001Year 2018−0.040.008−4.49<0.001Cultured population (reference)Wild population−0.040.02−2.45<0.05Year 2009 × cultured population (reference)Year 2010 × wild population0.010.030.360.72Year 2011 × wild population−0.010.03−0.450.65Year 2012 × wild population0.050.031.950.051Year 2013 × wild population0.0080.070.320.74Year 2014 × wild population0.030.031.110.27Year 2015 × wild population0.050.031.880.06Year 2016 × wild population0.060.032.12<0.05Year 2017 × wild population0.070.022.85<0.01Year 2018 × wild population0.0020.030.080.93Note: Here SE is the standard errors, t‐value is the statistical t‐value and p is the p‐value. Significance levels (p‐value) are denoted by <0.05; <0.01; <0.001 which are given in bold font.4FIGUREDifferences in condition factor (K) between cultured and wild populations of delta smelt in relation to years (a). The solid line with 95% confidence interval (translucent band area) represents the linear regression model fit. The regression model parameters are presented in the equation where K: condition factor, Y: year, P: population and Y×P: interaction of year and population. Differences in condition factor between populations are shown in (b) across different years and (c) different fork length interval classes (FL‐class). Each coloured line represents the mean value of population for each year for plot b and each FL‐class for plot c. The large square dark blue points represent the overall mean score within each population across different years for plot b and different FL‐classes for plot c.In the cultured population, K was significantly increased with the increase in FL (intercept = 0.53, slope = 0.019, p < 0.001), whereas slight but not significant variation was observed in the FL relationship in the wild population (intercept = 0.50, slope = 0.019, p = 0.35). The model showed no significant effect of Population (p = 0.94) and the interaction of FL × Population (p = 0.96) on K. The ANOVA results indicate significant effects of FL‐interval class (F4,2713 = 7.13, p < 0.001) and the interaction of FL‐interval class × Population (F4,2713 = 2.61, p < 0.05) on K, whereas no significant variation was found in the K value between the cultured and wild populations (F1,2713 = 3.42, p = 0.06 and Figure 4c). The significant interaction between FL‐interval class and Population suggests that the difference of K between populations differed among FL‐interval classes. The FL‐interval classes of 90–99 and 80–89 mm in the cultured population had significantly higher K values than other size classes of the wild population except for the 50–59 mm (Figure 4c). For the cultured population, the FL‐interval class of 90–99 mm had significantly higher K than that of 80–89 mm FL‐class (Figure 4c).Correlations among different traitsCorrelation tests showed significantly positive relationships between all traits except for FL‐K in the wild population (Table 6). No evidence of a significant trade‐off has been found between traits in any population.6TABLECorrelations (Pearson's test) among different variables of the delta smeltPopulationVariableFLBWFKCulturedFL–BW0.92 (0.001)–F0.75 (0.001)0.81 (0.001)–K0.084 (0.001)0.73 (0.001)0.31 (0.001)–WildFL–BW0.88 (0.001)–F0.68 (0.001)0.78 (0.001)–K0.059 (0.36)0.50 (0.001)0.42 (0.001)–Note: Values shown are correlation coefficients (r) and p‐values (in brackets). Significant values are given in bold font.DISCUSSIONSignificant differences in length, weight and fecundity were found between the cultured and wild populations of delta smelt at the FCCL over the last 10 generations. Cultured fish were longer, heavier and were more fecund than the wild fish. Similar results in other species have also shown cultured fish grew faster than the wild ones, such as western ruivaco Achondrostoma occidentale (Mameri et al., 2018), Atka mackerel Pleurogrammus monopterygius (McDermott et al., 2011), steelhead Oncorhynchus mykiss (Kostow, 2004), Atlantic salmon (Blanchet et al., 2008), gilthead sea bream Sparus aurata (Grigorakis et al., 2002) and brown trout (Hedenskog et al., 2002). O'Sullivan et al. (2020) showed that the fecundity of cultured Atlantic salmon was approximately 1.4 times higher than wild ones. The differences in body size and fecundity between the cultured and wild populations may be attributed to various environmental factors such as water quality, temperature, density, food availability and quality, predators, stress exposure and fish mobility and swimming scopes (Basaran et al., 2007; Favaloro & Mazzola, 2003; Patiyal et al., 2014).The observed differences in length, weight and fecundity between the cultured population at the FCCL and wild populations are unlikely to be caused by genetic differences, given the lack of evidence for significant genetic divergence between the two populations. Fisch et al. (2013) and Finger et al. (2018) did not detect any significant divergence using 12 neutral microsatellite markers. However, it is important to note that these markers may not capture genetic divergence at loci that are under selection or linked to loci under selection. Therefore, other factors such as environmental conditions and selective pressures may be contributing to the observed differences. However, we cannot rule out genetic adaptation to captivity, and some studies are ongoing to answer this question (Claussen & Philipp, 2022; Yagound et al., 2022). Nutritional and environmental differences, especially during the early life stages of wild delta smelt before they were brought into the hatchery, clearly affect the size and fecundity of them (Baxter et al., 2015; Hammock et al., 2015) and can potentially lead to the low recruitment and eventual decline of the wild population.Cultured delta smelt population was spawned and raised at the FCCL under consistent conditions for all life stages. Live prey and/or dry feed were provided consistently to ensure that fish were not food limited, there were no predators or competitors, and water quality was always managed properly. Studies have shown that cultured fish grew significantly faster than their wild counterparts due to stable food availability and quality, higher consumption rates, less exercise and energy expenditure (Islam et al., 2020; Teletchea, 2019; Wringe et al., 2016). Other researchers have also shown that many other factors can influence fecundity, such as environmental conditions (Salvanes & Braithwaite, 2005), stress (Schreck et al., 2001), fish size (Kjesbu et al., 1996) and the duration of the spawning period (Bennett, 2005; Brown et al., 2013; Hunter et al., 1985). Fish size and fecundity of delta smelt are likely to have also been influenced by these factors.The changes in size (FL and BW) of cultured delta smelt in 2011 and 2017 were possibly due to different feeding strategies implemented at the FCCL. Prior to 2011, fish were fed newly hatched brine shrimp until 160 days post hatch (DPH). In July 2010, the FCCL tried to improve culture methods to increase the survival of fish, and juvenile delta smelt were co‐fed with a dry diet together with the newly hatched brine shrimp. In 2013, the discontinuation of the diet used and the process of finding the replacement may have caused the slower growth than the subsequent years (2014 and 2015). Since 2017, fish were co‐fed newly hatched brine shrimp and the new dry feed at 80 DPH. The weaning process started at 115 DPH, and the fish were fully weaned to the dry diet at only 120 DPH (Hung et al., 2022). This alteration may have significantly improved the growth of cultured delta smelt since 2018. However, the abnormal lower growth and fecundity of the cultured population and the relatively higher growth and fecundity of the wild population during 2011 and 2017 require further investigations.This study showed that the condition of cultured delta smelt was significantly higher in most years than the population in the wild. This finding aligns with some other studies where K of cultured fish were significantly higher than their wild origins, including Atlantic cod (Grant et al., 1998), European sea bass (Handelsman et al., 2010), Atlantic halibut (Haug et al., 1989), Atlantic salmon (Glover et al., 2009) and brown trout (Serrano et al., 2009). The K of the cultured population decreased in this study from 2013 to 2016. One plausible reason could be the weaning of cultured juveniles in 2011 or the feed used might reduce their overall condition rather than the growth, that is reducing body fatness or other indices (Schloesser & Fabrizio, 2017; Vandersteen et al., 2019). The sharp increase in K from 2017 might be due to the change to a new feeding strategy. The relationship between the K and FL‐interval class was significantly strong in the case of the cultured population, which was consistent with other fishes indicating that K has been increased with the increase of body size (Blackwell et al., 2000; Nahdi et al., 2016). However, the reason of the 80–89 mm group wild fish having a significantly reduced K was unclear. Due to variable rearing conditions and other unidentified factors (i.e. stress and climate change) across years, the findings showed that K of delta smelt not only depended on population status but also years, FL‐interval class and the interaction of Year, Population and FL‐interval class.Some studies suggested trade‐offs (inverse relationship between two traits) happened in fishes when they adapted to new environments. For example, trade‐offs were found between size and reproduction in 70 Amazonian floodplain fish species (Röpke et al., 2021). Röpke et al. (2021) showed that some fish have low values of somatic investment before sexual maturity (SIBSM), indicating their early sexual maturity in life with early reduction in somatic investment, whereas some other fish have higher values of SIBSM, indicating late sexual maturity in life with longer somatic investment. According to Roff (1983), fish prioritize investing energy towards gonadal development rather than somatic growth, indicating a trade‐off between the two processes. Thus, trade‐offs between fecundity and fish size can occur depending on species, life stage, types of traits and environmental conditions (Chigbu & Sibley, 1994; Garland, 2014; Roff, 1983). In this study, findings showed no trade‐off in both populations, even though the cultured population continued to grow faster.The goals of the FCCL were to create a genetically managed refuge population that stayed as close as possible to the wild population, to provide a source population for the supplementation of cultured fish to the wild population or in the case of wild extinction, and to propagate fish for research activities (Lessard et al., 2018; Lindberg et al., 2013). The FCCL has been trying to produce a cultured population that is closer to the wild population, but it was unavoidable that the fish became domesticated. Finger et al. (2018) found that increasing levels of hatchery ancestry in a 1:1 cross led that cross to having a greater probability of producing offspring that survive to maturity the following year, signalling adaptation to captivity. Cultured delta smelt were raised in a controlled environment that might provide better environmental conditions needed for survival from eggs to adults. Cultivated parents hold an advantageous position right from the start as they ensure the successful development of their eggs, whereas wild delta smelt must rely on their survival instincts to endure in an estuary that no longer provides them with an ideal habitat (Hobbs et al., 2017; Moyle et al., 2019). The delta smelt population continues to exhibit a precipitous decline with concerns of extinction should the decline in suitable habitat continue (Hobbs et al., 2017). Releasing fish into an unsuitable or rapidly declining environment will mostly likely lead to little success of repopulation.Scientists concluded rapid progress towards the development of a viable and testable supplementation programme must be a priority for the management and conservation of delta smelt (Hobbs et al., 2017; Lessard et al., 2018). The FCCL was not trying to raise fish specifically capable to survive in the wild, but due to more recent trends and information, the supplementation of the wild delta smelt population using cultured delta smelt is planned (FWS BiOp 2019), and experimental releases are currently underway using the cultured fish spawned and reared at the FCCL.As cultured delta smelt are not subjected to the stressors of their wild counterparts, there are concerns that when they are suddenly thrust into a new environment, they might fall victim to those stressors and not fare as well, but Hung et al. (2019) found that captive‐reared sub‐adult delta smelt could survive in a semi‐natural environment with uncontrolled water quality and naturally produced wild prey through spawning and into their post spawning phase. This gives hope that these fish may survive and be able to spawn if released in the estuary. Given that cultured delta smelt tend to grow larger and produce a greater number of eggs compared to wild fish (Anderson, 1988; Cargnelli & Gross, 1996; Vincenzi et al., 2010), they may be better suited to cope with stressors. With cultured delta smelt showing a higher fecundity than wild‐origin delta smelt, it must be looked at how this could affect the wild population if the cultured fish were released. Cultured fish might be able to produce more eggs and contribute a higher percentage of individuals to the next generation than the wild ones. The success of cultured delta smelt may have two potential outcomes; on one hand, they may push the wild population towards extinction, but on the other hand, if they interbreed with wild fish and increase the reproductive success of the wild population, even though this may also raise the risk of outbreeding depression in the offspring. To try and get the best possible results with supplementation trials, it seems natural to want to release the fish most likely to survive and reproduce successfully. There is hope cultured delta smelt will meet these criteria.CONCLUSIONThe results of this study provide information on captive rearing, growth and reproductive performance of delta smelt that may benefit future culture and management efforts focused on conserving the species, including potential implications of releasing cultured fish from the FCCL into the wild. Delta smelt cultured at the FCCL have been produced with all considerations to reduce domestication, but it still happened. The cultured fish used for supplementation to the wild have differences from their wild counterparts, and this paper set out to describe those differences so managers can better understand what may happen upon the release of the fish into the wild. Given the continued decline of the species and the on‐going experimental release efforts, further study on how cultured delta smelt can benefit the species recovery is highly warranted.AUTHOR CONTRIBUTIONSConceptualization; data curation; investigation; methodology; supervision; writing – review and editing: Luke Ellison. Formal analysis; writing – review and editing: Md Moshiur Rahman. Conceptualization; methodology; supervision; writing – review and editing: Amanda J. Finger. Writing – original draft: Marade Sandford. Formal analysis; writing – review and editing: Chih‐Hsin Hsueh. Conceptualization; resources; writing – review and editing: Andrew A. Schultz. Conceptualization; data curation; funding acquisition; investigation; methodology; project administration; resources; supervision; writing – original draft; writing – review and editing: Tien‐Chieh Hung.ACKNOWLEDGEMENTSThe authors would like to thank all current and former FCCL and GVL staff, especially to Lindberg J and May B, for their hard work on the development and maintaining the delta smelt refuge population and data collection. The authors also thank USBR biologists led by Reyes R for the assistance on the fish collection and Yang W‐R for the assistance on the map image production. The study was supported by the California State Department of Fish and Game (#P0730201), CALFED Bay‐Delta Program (#1048), the California Department of Water Resources (#4600007604), the Interagency Ecological Program and the U.S. Bureau of Reclamation (#R10AC20014 and #R15AC00030). Any opinions, findings and conclusions or recommendations expressed in this article are those of the authors and do not necessarily reflect the views of the supporting agencies.CONFLICT OF INTEREST STATEMENTThe authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.DATA AVAILABILITY STATEMENTData that support the findings of this study are available from the corresponding author upon reasonable requests.ETHICS STATEMENTAll the animal handling protocols and rearing operations were reviewed and approved by the Institutional Animal Care and Use Committee (#18081 and 19747) at the University of California, Davis.REFERENCESAfshartous, D. & Preston, R.A. (2011) Key results of interaction models with centering. Journal of Statistics Education, 19, 1–24.Anderson, J.T. (1988) A review of size dependent survival during pre‐recruit stages of fishes in relation to recruitment. Journal of Northwest Atlantic Fishery Science, 8, 55–66.Basaran, F., Ozbilgin, H. & Ozbilgin, Y.D. (2007) Comparison of the swimming performance of farmed and wild gilthead sea bream, Sparus aurata. Aquaculture Research, 38, 452–456.Baskerville‐Bridges, B., Lindberg, J.C. & Doroshov, S.I. (2005) Manual for the intensive culture of delta smelt (Hypomesus transpacificus). Sacramento: University of California. Davis report to CALFED Bay−Delta Program, ERP‐02‐P31.Baxter, R., Brown, L., Castillo, G., Conrad, L., Culberson, S., Dekar, M. et al. (2015) An updated conceptual model of delta smelt biology: our evolving understanding of an estuarine fish. Interagency Ecological Program for the San Francisco Bay/Delta Estuary, pp. 32–68. Technical report 90. Available at: https://pubs.er.usgs.gov/publication/70141018Bennett, W.A. (2005) Critical assessment of the delta smelt population in the San Francisco estuary, California. San Francisco Estuary and Watershed Science, 3, 2–72.Bilio, M. (2007) Controlled reproduction and domestication in aquaculture‐ the current state of the art, Part I. Aquaculture Europe, 32, 5–14.Blackwell, B.G., Brown, M.L. & Willis, D.W. (2000) Relative weight (Wr) status and current use in fisheries assessment and management. Reviews in Fisheries Science, 8, 1–44.Blanchet, S., Páez, D.J., Bernatchez, L. & Dodson, J.J. (2008) An integrated comparison of captive‐bred and wild Atlantic salmon (Salmo salar): implications for supportive breeding programs. Biological Conservation, 141, 1989–1999.Bork, K., Moyle, P., Durand, J., Hung, T.‐C. & Rypel, A. (2020) Small populations in jeopardy: a delta smelt case study (December 15, 2020). 50 Environmental Law Reporter 10714 (2020). Available at SSRN: https://ssrn.com/abstract=3749669Brown, L.R., Bennett, W.A., Wagner, R.W., Morgan‐King, T., Knowles, N., Feyrer, F. et al. (2013) Implications for future survival of delta smelt from four climate change scenarios for the Sacramento–San Joaquin Delta, California. Estuaries and Coasts, 36, 754–774.California Department of Fish and Wildlife. (2021) State and federally listed endangered and threatened animals of California. North Highlands: California Department of Fish and Wildlife, The Natural Resources Agency, pp. 1–26. Available online: file:///C:/Users/Moshiur/Downloads/CNDDB_Endangered_and_Threatened_Ani.Cargnelli, L.M. & Gross, M.R. (1996) The temporal dimension in fish recruitment: birth date, body size, and size‐dependent survival in a sunfish (bluegill: Lepomis macrochirus). Canadian Journal of Fisheries and Aquatic Sciences, 53, 360–367.Chigbu, P. & Sibley, T.H. (1994) Relationship between abundance, growth, egg size and fecundity in a landlocked population of longfin smelt, Spirinchus thaleichthys. Journal of Fish Biology, 45, 1–15.Christie, M.R., Ford, M.J. & Blouin, M.S. (2014) On the reproductive success of early‐generation hatchery fish in the wild. Evolutionary Applications, 7, 883–896.Claussen, J.E. & Philipp, D.P. (2022) Assessing the role of supplementation stocking: a perspective. Fisheries Management and Ecology, 1–9.Favaloro, E. & Mazzola, A. (2003) Shape change during the growth of sharpsnout seabream reared under different conditions in a fish farm of the southern Tyrrhenian Sea. Aquacultural Engineering, 29, 57–63.Finger, A.J., Mahardja, B., Fisch, K.M., Benjamin, A., Lindberg, J., Ellison, L. et al. (2018) A conservation hatchery population of delta smelt shows evidence of genetic adaptation to captivity after 9 generations. Journal of Heredity, 109, 689–699.Fisch, K.M., Ivy, J.A., Burton, R.S. & May, B. (2013) Evaluating the performance of captive breeding techniques for conservation hatcheries: a case study of the delta smelt captive breeding program. Journal of Heredity, 104, 92–104.Fox, J. & Weisberg, S. (2019) An R companion to applied regression, 3rd edition. Thousand Oaks, CA, Sage. https://socialsciences.mcmaster.ca/jfox/Books/Companion/Fulton, T.W. (1902) Rate of growth of sea fishes. Seattle: Neill & Company.Garland, T. (2014). Trade‐offs. Current Biology, 24, R60–R61.Glover, K.A., Otterå, H., Olsen, R.E., Slinde, E., Taranger, G.L. & Skaala, Ø. (2009) A comparison of farmed, wild and hybrid Atlantic salmon (Salmo salar L.) reared under farming conditions. Aquaculture, 286, 203–210.Grant, S.M., Arown, J.A. & Boyce, D.L. (1998) Enlarged fatty livers of small juvenile cod: a comparison of laboratory‐cultured and wild juveniles. Journal of Fish Biology, 52, 1105–1114.Grigorakis, K., Alexis, M.N., Taylor, K.D.A. & Hole, M. (2002) Comparison of wild and cultured gilthead sea bream (Sparus aurata); composition, appearance and seasonal variations. International Journal of Food Science and Technology, 37, 477–484.Hammock, B.G., Hobbs, J.A., Slater, S.B., Acuña, S. & Teh, S.J. (2015) Contaminant and food limitation stress in an endangered estuarine fish. Science of the Total Environment, 532, 316–326.Handelsman, C., Claireaux, G. & Nelson, J.A. (2010) Swimming ability and ecological performance of cultured and wild European sea bass (Dicentrarchus labrax) in coastal tidal ponds. Physiological and Biochemical Zoology, 83, 435–445.Harvey, A.C., Solberg, M.F., Troianou, E., Carvalho, G.R., Taylor, M.I., Creer, S. et al. (2016) Plasticity in growth of farmed and wild Atlantic salmon: is the increased growth rate of farmed salmon caused by evolutionary adaptations to the commercial diet? BMC Evolutionary Biology, 16, 264.Haug, T., Huse, I., Kjørsvik, E. & Rabben, H. (1989) Observations on the growth of juvenile Atlantic halibut (Hippoglossus hippoglossus L.) in captivity. Aquaculture, 80, 79–86.Hedenskog, M., Petersson, E. & Järvi, T. (2002) Agonistic behavior and growth in newly emerged brown trout (Salmo trutta L) of sea‐ranched and wild origin. Aggressive Behavior, 28, 145–153.Hobbs, J.A., Moyle, P.B., Fangue, N. & Connon, R.E. (2017) Is extinction inevitable for delta smelt and Longfin Smelt? An opinion and recommendations for recovery. San Francisco Estuary and Watershed Science, 15, 1–19.Hung, T., Ellison, L., Stevenson, T., Sandford, M., Schultz, A.A. & Eads, A.R. (2022) Early weaning in endangered delta smelt: effect of weaning time on growth and survival. North American Journal of Aquaculture, 84, 249–260.Hung, T., Rosales, M., Kurobe, T., Stevenson, T., Ellison, L., Tigan, G. et al. (2019) A pilot study of the performance of captive‐reared delta smelt Hypomesus transpacificus in a semi‐natural environment. Journal of Fish Biology, 95, 1517–1522.Hunter, J.R., Lo, N.C. & Leong, R.J.H. (1985) Batch fecundity in multiple spawning fishes. Pages 61–72 in Lasker R, editor. An egg production method of estimating spawning biomass of pelagic fish: application to the Northern anchovy, Engraulis mordax. US Department of Commerce, NOAA. Technical report.Islam, S.S., Yates, M.C. & Fraser, D.J. (2020) Single generation exposure to a captive diet: a primer for domestication selection in a salmonid fish? BioRxiv, 2020.01.24.919175. https://doi.org/10.1101/2020.01.24.919175Kjesbu, O.S., Solemdal, P., Bratland, P. & Fonn, M. (1996) Variation in annual egg production in individual captive Atlantic cod (Gadus morhua). Canadian Journal of Fisheries and Aquatic Sciences, 53, 610–620.Konstantinidis, I., Sætrom, P., Mjelle, R., Nedoluzhko, A.V., Robledo, D. & Fernandes, J.M.O. (2020) Major gene expression changes and epigenetic remodelling in Nile tilapia muscle after just one generation of domestication. Epigenetics, 15, 1052–1067.Kostow, K.E. (2004) Differences in juvenile phenotypes and survival between hatchery stocks and a natural population provide evidence for modified selection due to captive breeding. Canadian Journal of Fisheries and Aquatic Sciences, 61, 577–589.Laikre, L., Schwartz, M.K., Waples, R.S. & Ryman, N. (2010) Compromising genetic diversity in the wild: unmonitored large‐scale release of plants and animals. Trends in Ecology & Evolution, 25, 520–529.Lasne, E., Leblanc, C.A.‐L. & Gillet, C. (2018) Egg size versus number of offspring trade‐off: female age rather than size matters in a domesticated Arctic charr population. Evolutionary Biology, 45, 105–112.Lenth, R. (2020) emmeans: estimated marginal means, aka least‐squares means. R package version 1.5.2‐1. https://CRAN.R‐project.org/package=emmeansLessard, J., Cavallo, B., Anders, P., Sommer, T., Schreier, B., Gille, D. et al. (2018) Considerations for the use of captive‐reared delta smelt for species recovery and research. San Francisco Estuary and Watershed Science, 16, article 3.Liao, I.C. & Huang, Y.S. (2000) Methodological approach used for the domestication of potential candidates for aquaculture. Cahiers Options Méditerranéennes, 47, 97–107.Liao, I.C., Su, H.M. & Chang, E.Y. (2001) Techniques in finfish larviculture in Taiwan. Aquaculture, 200, 1–31.Lindberg, J.C., Tigan, G., Ellison, L., Rettinghouse, T., Nagel, M.M. & Fisch, K.M. (2013) Aquaculture methods for a genetically managed population of endangered delta smelt. North American Journal of Aquaculture, 75, 186–196.Mameri, D., Sousa‐Santos, C., Robalo, J.I. & Magalhães, M.F. (2018) Growth and age structure in captive and wild stocks of the endangered western Ruivaco Achondrostoma occidentale (cyprinidae). Limnetica, 37, 105–115.McDermott, S.F., Cooper, D.W., Guthridge, J.L., Spies, I.B., Canino, M.F., Woods, P. et al. (2011) Effects of maternal growth on fecundity and egg quality of wild and captive Atka mackerel. Marine and Coastal Fisheries, 3, 324–335.Milla, S., Pasquet, A., El Mohajer, L. & Fontaine, P. (2021) How domestication alters fish phenotypes. Reviews in Aquaculture, 13, 388–405.Moyle, P., Bork, K., Durand, J., Hung, T.‐C. & Rypel, A. (2019) Futures for delta smelt. California WaterBlog. Available at: https://californiawaterblog.com/2019/12/15/futures‐for‐delta‐smelt/Moyle, P.B. (2002) Inland fishes of California. Berkeley, CA: University of California Press.Moyle, P.B., Hobbs, J.A. & Durand, J.R. (2018) delta smelt and water politics in California. Fisheries, 43, 42–50.Mylonas, C.C., De La Gándara, F., Corriero, A. & Ríos, A.B. (2010) Atlantic bluefin tuna (Thunnus thynnus) farming and fattening in the Mediterranean sea. Reviews in Fisheries Science, 18, 266–280.Nahdi, A.A.L., Garcia de Leaniz, C. & King, A.J. (2016) Spatio‐temporal variation in length‐weight relationships and condition of the ribbonfish Trichiurus lepturus (Linnaeus, 1758): implications for fisheries management. PLoS One, 11, e0161989.O'Sullivan, R.J., Aykanat, T., Johnston, S.E., Rogan, G., Poole, R., Prodöhl, P.A. et al. (2020) Captive‐bred Atlantic salmon released into the wild have fewer offspring than wild‐bred fish and decrease population productivity. Proceedings of the Royal Society B: Biological Sciences, 287, 20201671.Patiyal, R.S., Mir, J.I., Sharma, R.C., Chandra, S. & Mahanta, P.C. (2014) Pattern of meristic and morphometric variations between wild and captive stocks of endangered Tor putitora (Hamilton 1822) using multivariate statistical analysis methods. Proceedings of the National Academy of Sciences, India, Section B: Biological Sciences, 84, 123–129.Peterson, B.G. & Carl, P. (2020) PerformanceAnalytics: econometric tools for performance and risk analysis. R package version 2.0.4. https://CRAN.R‐project.org/package=PerformanceAnalyticsR Development Core Team. (2021) R: a language and environment for statistical computing, version 4.0.5. Vienna, Austria: R Foundation for Statistical Computing. Available online: https://www.r‐project.org/ [Accessed 12 July 2022].Roff, D.A. (1983) An allocation model of growth and reproduction in fish. Canadian Journal of Fisheries and Aquatic Sciences, 40, 1395–1404.Roff, D.A. (1992) The evolution of life histories: theory and analysis. New York: Chapman & Hall, pp. 47–56.Roff, D.A. & Fairbairn, D.J. (2007) The evolution of trade‐offs: where are we? Journal of Evolutionary Biology, 20, 433–447.Röpke, C., Pires, T.H.S., Zuanon, J., Freitas, C.E.C., Hernandes, M.C., Souza, F. et al. (2021) Growth–reproduction trade‐off and fecundity regulate population stability in Amazon floodplain fishes. Freshwater Biology, 66, 1101–1109.Ruzzante, D.E. (1994) Domestication effects on aggressive and schooling behavior in fish. Aquaculture, 120, 1–24.Salvanes, A.G.V. & Braithwaite, V.A. (2005) Exposure to variable spatial information in the early rearing environment generates asymmetries in social interactions in cod (Gadus morhua). Behavioral Ecology and Sociobiology, 59, 250–257.Sandford, M., Castillo, G. & Hung, T. (2019) A review of fish identification methods applied on small fish. Reviews in Aquaculture, 12, 542–554.Schloesser, R.W. & Fabrizio, M.C. (2017) Condition indices as surrogates of energy density and lipid content in juveniles of three fish species. Transactions of the American Fisheries Society, 146, 1058–1069.Schreck, C.B., Contreras‐Sanchez, W. & Fitzpatrick, M.S. (2001) Effects of stress on fish reproduction, gamete quality, and progeny. In: Reproductive biotechnology in finfish aquaculture. Amsterdam: Elsevier, pp. 3–24.Serrano, I., Larsson, S. & Eriksson, L.‐O. (2009) Migration performance of wild and hatchery sea trout (Salmo trutta L.) smolts—implications for compensatory hatchery programs. Fisheries Research, 99, 210–215.Sousa‐Santos, C., Gil, F. & Almada, V.C. (2014) Ex situ reproduction of Portuguese endangered cyprinids in the context of their conservation. Ichthyological Research, 61, 193–198.Teletchea, F. (2019) Fish domestication: an overview. In: Teletchea F. (Ed.) Animal domestication. London, UK: IntechOpen, pp. 69–90.Teletchea, F. & Fontaine, P. (2014) Levels of domestication in fish: implications for the sustainable future of aquaculture. Fish and Fisheries, 15, 181–195.Tempel, T. (2022) History in the making: updates on the first experimental release of captive‐reared delta smelt. California Department of Water Resources. Available at: https://nrm.dfg.ca.gov/FileHandler.ashx?DocumentID=199263&inlineU.S. Office of the Federal Register. (1993) Endangered and threatened wildlife and plants: determination of threatened status for the delta smelt. Federal Register, 58, 12854–12864.Vandersteen, W.E., Leggatt, R., Sundström, L.F. & Devlin, R.H. (2019) Importance of experimental environmental conditions in estimating risks and associated uncertainty of transgenic fish prior to entry into nature. Scientific Reports, 9, 406.Vincenzi, S., Crivelli, A.J., Jesensek, D. & De Leo, G.A. (2010) Individual growth and its implications for the recruitment dynamics of stream‐dwelling marble trout (Salmo marmoratus). Ecology of Freshwater Fish, 19, 477–486.Welcomme, R.L., Cowx, I.G., Coates, D., Béné, C., Funge‐Smith, S., Halls, A. et al. (2010) Inland capture fisheries. Philosophical Transactions of the Royal Society B: Biological Sciences, 365, 2881–2896.Wickham, H. (2016) ggplot2: elegant graphics for data analysis. New York: Springer‐Verlag.Wringe, B.F., Purchase, C.F. & Fleming, I.A. (2016) In search of a “cultured fish phenotype”: a systematic review, meta‐analysis and vote‐counting analysis. Reviews in Fish Biology and Fisheries, 26, 351–373.Yagound, B., West, A.J., Richardson, M.F., Gruber, J., Reid, J.G., Whiting, M.J. et al. (2022) Captivity induces large and population‐dependent brain transcriptomic changes in wild‐caught cane toads (Rhinella marina). Molecular Ecology, 31, 4949–4961.

Journal

Aquaculture Fish and FisheriesWiley

Published: Aug 1, 2023

Keywords: California; domestication; endangered species; supplementation

There are no references for this article.