https://cpw.cvlcollections.org/files/original/63427ce4d137f32e7a466040d2756e18.pdf d271115d7a9d22d9171be3b4c56cf4a5 PDF Text Text The research in this publication was partially or fully funded by Colorado Parks and Wildlife. Dan Prenzlow, Director, Colorado Parks and Wildlife • Parks and Wildlife Commission: Marvin McDaniel, Chair • Carrie Besnette Hauser, Vice-Chair Marie Haskett, Secretary • Taishya Adams • Betsy Blecha • Charles Garcia • Dallas May • Duke Phillips, IV • Luke B. Schafer • James Jay Tutchton • Eden Vardy �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) 1 ARTICLE Induced triploidy reduces mercury bioaccumulation in a piscivorous ﬁsh Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Collin J. Farrell, Brett M. Johnson, Adam G. Hansen, and Christopher A. Myrick Abstract: We compared mercury bioaccumulation in triploid and diploid walleye (Sander vitreus) in Narraguinnep Reservoir, Colorado, USA, and made several hypotheses that sex- and ploidy-speciﬁc differences in the allocation of energy towards reproductive development would affect mercury bioaccumulation. We tested our hypotheses with linear regression and a bioenergetics model informed by ﬁeld data. We found diploid walleye had 28%–31% higher mercury concentrations on average than triploids, but there were no differences between sexes of the same ploidy. Triploids of mature age exhibited minimal gonadal development when compared to diploids. After accounting for reproductive investment, the bioenergetics model accounted for most of the observed difference in average mercury concentration between ploidies for females. Conversely, the energetic cost of producing testes was low, and gonadal development could not explain observed patterns for males. Costs associated with elevated swimming activity and metabolism by diploid males relative to other groups could explain the difference but requires further investigation. The use of triploid ﬁsh in stocking programs could prove useful for reducing mercury in ﬁsh destined for human consumption. Résumé : Nous comparons la bioaccumulation de mercure dans des dorés jaunes (Sander vitreus) triploïdes et diploïdes dans le � réservoir Narraguinnep (Colorado, Etats-Unis) et formulons plusieurs hypothèses à l’effet que des différences selon le sexe et la ploïdie sur le plan de l’affectation de ressources énergétiques au développement des organes reproducteurs auraient une incidence sur la bioaccumulation de mercure. Nous validons ces hypothèses à l’aide de la régression linéaire et d’un modèle bioénergétique alimenté de données de terrain. Nous constatons que les dorés diploïdes présentent des concentrations de mercure de 28 % à 31 % plus importantes, en moyenne, que les triploïdes, mais aucune différence n’est relevée entre les sexes de même ploïdie. Les triploïdes d’âge mature présentent un développement gonadique minimal comparativement aux diploïdes. Une fois pris en compte l’investissement dans le système reproducteur, le modèle bioénergétique explique la majeure partie des différences observées des concentrations de mercure moyennes entre femelles de ploïdies différentes. À l’inverse, le coût énergétique de la production de testicules est faible, et le développement des gonades ne peut expliquer les motifs observés chez les mâles. Les coûts associés à une activité de nage et un métabolisme plus élevés chez les mâles diploïdes par rapport aux autres groupes pourraient expliquer la différence, mais cela nécessite un examen plus approfondi. L’utilisation de poissons triploïdes dans les programmes d’empoissonnement pourrait s’avérer utile pour réduire les concentrations de mercure dans les poissons destinés à la consommation humaine. [Traduit par la Rédaction] Introduction Induced triploidy is commonly used in aquaculture operations (Piferrer et al. 2009), for biocontrol of aquatic vegetation (Allen and Wattendorf 1987), and increasingly as a stocking option for recreational ﬁsheries (Cassinelli et al. 2019; Koch et al. 2018; Teuscher et al. 2003), primarily because triploid ﬁsh are reproductively sterile (Benfey 1999). Triploid females typically have small ovaries incapable of producing viable ova, while triploid males in many species develop normally but produce aneuploid spermatozoa (Benfey 1999; Thorgaard 1983). Yet, triploid males may still attempt to spawn with diploid females (Benfey 1999) and can act as a control on natural reproduction (Piferrer et al. 2009). Despite increased interest in triploidy as a stocking option for recreational ﬁsheries, studies examining the ecology of triploids and diploids in sympatry are rare. Sterility confers several potential advantages for the use of triploids as a stocking option in recreational ﬁsheries, most notably for reproductive containment and the potential for increased growth rates (Piferrer et al. 2009). It is commonly hypothesized that triploids should grow faster and reach larger body sizes relative to their diploid counterparts because of the reduced energetic requirements associated with gonadal development, particularly for females (Leary et al. 1985; Maxime 2008; Tiwary et al. 2004). However, potential differences in growth performance between ploidies remain inconclusive (Benfey 1999; Maxime 2008). Although growth has not been fully reconciled, differential energetic requirements associated with spawning could modify consumption dynamics and the bioaccumulation of contaminants between diploid and triploid ﬁsh — an overlooked element that may have important implications for the use of triploids in recreational ﬁsheries. Mercury (Hg) is a globally pervasive contaminant and potent neurotoxin. Once released into the atmosphere, Hg can disperse far from its source (Wentz et al. 2014) before being deposited in terrestrial, freshwater, and marine ecosystems (Chen et al. 2008; Driscoll et al. 2013; Morel et al. 1998). Thus, controlling atmospheric deposition of Hg is challenging. Under certain conditions Received 16 February 2021. Accepted 9 June 2021. C.J. Farrell, B.M. Johnson, and C.A. Myrick. Department of Fish, Wildlife, and Conservation Biology, Colorado State University, 1474 Campus Delivery, Fort Collins, CO 80523, USA. A.G. Hansen. Colorado Parks and Wildlife, Aquatic Research Section, 317 West Prospect Road, Fort Collins, CO 80526, USA. Corresponding author: Collin J. Farrell (email: collin.farrell@colostate.edu). © 2021 The Author(s). Permission for reuse (free in most cases) can be obtained from copyright.com. Can. J. Fish. Aquat. Sci. 00: 1–13 (0000) dx.doi.org/10.1139/cjfas-2021-0037 Published at www.cdnsciencepub.com/cjfas on 17 June 2021. �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. 2 (Paranjape and Hall 2017), microbial processes transform Hg into methylmercury (MeHg), the most toxic and bioavailable form of Hg, which bioaccumulates in organisms and biomagniﬁes in food webs. Concentrations can reach levels that are hazardous or lethal to ﬁsh, wildlife, and humans (Boening 2000; Morel et al. 1998; Scheuhammer et al. 2007). Mercury is fully or partially responsible for the vast majority (>80%) of ﬁsh consumption advisories in the US and Canada (Eagles-Smith et al. 2016b). A better understanding of factors that affect bioaccumulation of Hg in ﬁsh could help identify new mitigation strategies that protect the health of humans and piscivorous ﬁsh and wildlife species. Several studies have investigated how MeHg bioaccumulation affects reproduction (Crump and Trudeau 2009), but the effects of reproductive investment (i.e., the cumulative energetic investment into gamete production over a ﬁsh’s lifetime) on MeHg bioaccumulation are not well understood. Spawning is energetically costly for both males and females and requires ﬁsh to meet this energetic demand by consuming prey (Diana 1983; Trudel et al. 2000). Since food consumption is the primary pathway of Hg uptake in predators (Hall et al. 1997), the energetic costs associated with spawning should at least partially regulate MeHg bioaccumulation (Nicoletto and Hendricks 1988). Therefore, we would expect a positive correlation between MeHg concentrations (i.e., [MeHg]) and reproductive investment. Reproductive investment was correlated with higher [MeHg] in sharks (Coelho et al. 2010; Pethybridge et al. 2010) and ray-ﬁnned ﬁshes (Nicoletto and Hendricks 1988; Son et al. 2014). However, these studies only examined the relationship between gonadosomatic index (GSI) and [MeHg] or compared concentrations between mature and immature ﬁsh. As GSI, maturation, and [MeHg] are collinear with age (Grieb et al. 1990; Shatunovskii and Ruban 2009), the effects of reproductive investment could not be isolated. Given that the development of ovaries is typically more energetically costly than the development of testes (McBride et al. 2015), one would expect that reproductive females would consume more food, and therefore bioaccumulate more MeHg than reproductive males. However, several studies demonstrated that reproductive males have similar or higher [MeHg] relative to reproductive females of the same age (Bastos et al. 2016; Gewurtz et al. 2011; Madenjian et al. 2016). Differences in resource use (Lepak et al. 2012a), activity levels (Henderson et al. 2003), and standard metabolic, Hg elimination, or growth rates (Madenjian et al. 2014, 2016; Trudel and Rasmussen 2006) between males and females complicate studies using betweensex comparisons to examine the effects of reproductive investment on [MeHg]. No study has effectively accounted for age or other potential sex-dependent variables to discern the relative importance of reproductive investment in governing the bioaccumulation of Hg. Differential gonad development arising from ploidy manipulation gives us the ability to isolate the effects of reproductive investment on Hg bioaccumulation and overcome previous limitations. In this study, we used ﬁeld sampling to compare Hg dynamics in diploid and triploid walleye (Sander vitreus; but see Bruner 2021) co-occurring in the wild, and bioenergetics modeling to elucidate potential explanations for observed patterns in [MeHg] by quantifying prey consumption and dietary Hg exposure among the four sex-by-ploidy groups. Reservoir ﬁsheries management in the upper Colorado River Basin, USA, emphasizes a suite of measures to reconcile nonnative sport ﬁshing with the recovery of endangered ﬁsh endemic to the Colorado River and its tributaries. This includes stocking triploid walleye to diversify recreational angling opportunities in sensitive locations, deter illegal ﬁsh stocking, and possibly interfere with diploid reproduction in unwanted populations (Fetherman et al. 2015; Johnson et al. 2009). The relatively long history of triploid walleye stocking by the state of Colorado offered a unique opportunity to evaluate the ecology of diploid and triploid ﬁsh in sympatry, including Hg dynamics. Although originally stocked with diploid walleye as early as 1972, management of Narraguinnep Reservoir in southwest Colorado shifted Can. J. Fish. Aquat. Sci. Vol. 00, 0000 to triploid walleye stocking in 2008 to support native ﬁsh conservation efforts downstream. Colorado Parks and Wildlife has stocked triploid walleye in Narraguinnep Reservoir every year since 2008, except for 2009 and 2020, when walleye were not stocked. This has resulted in a mixed population of diploid and triploid walleye with sufﬁcient diversity in age classes to quantify Hg dynamics and account for potential differences in resource use, growth, or other factors between sexes and ploidies. Thus, Narraguinnep Reservoir served as an ideal study system to test various hypotheses of how reproductive investment inﬂuences Hg bioaccumulation. We hypothesized that (1) diploid female walleye would have signiﬁcantly higher [MeHg] than triploid females, driven by greater prey consumption associated with supporting the production of energetically costly eggs, (2) diploid and triploid males would have similar [MeHg], since triploid males of other species often develop normal testes (Benfey 1999), and (3) males of both ploidies would have lower [MeHg] relative to diploid females because the development of testes is less energetically costly than the development of ovaries (McBride et al. 2015). We tested our hypotheses by pairing empirical data on [MeHg], energy density of somatic and gonadal tissue, diet composition, and growth with bioenergetics model simulations to discern the role of sex-dependent reproductive investment in MeHg bioaccumulation. Methods Study site Narraguinnep Reservoir is a 215-ha irrigation water storage reservoir in southwest Colorado, USA (Fig. 1). The primary water supply is the Dolores River. The reservoir has a maximum depth of 25 m at full pool (surface elevation of 2037 m above mean sea level). Total dissolved solids average 240 mg·L�1. The reservoir is polymictic and oxygen is present throughout the water column. As is typical of reservoirs in the region, large annual water-level drawdowns during the irrigation season, and rising water levels during reﬁlling periodically inundate terrestrial vegetation (Gray et al. 2005) and contribute organic matter that may stimulate MeHg production (Selch et al. 2007; Sorensen et al. 2005). The ﬁsh community is comprised of diploid and triploid walleye, northern pike (Esox lucius), channel catﬁsh (Ictalurus punctatus), smallmouth bass (Micropterus dolomieu), white sucker (Catostomus commersonii), brown trout (Salmo trutta), rainbow trout (Oncorhynchus mykiss), black crappie (Pomoxis nigromaculatus), and yellow perch (Perca ﬂavescens). Elevated [MeHg] were found in several ﬁsh species from the reservoir, and ﬁsh consumption advisories for MeHg have been in place at Narraguinnep Reservoir since 1991 (Butler et al. 1995). Fish sampling Walleye were collected with standard fall walleye index netting gillnets (Morgan 2002) during spring and summer 2018–2019, and spring 2020. Spring sampling coincided with the spawning period for walleye. We recorded total length (TL, mm), total wet weight (WW, g), and gonad WW (g) of each ﬁsh. Sex, maturity, and gonad condition were classiﬁed according to Duffy et al. (2000). We characterized gonadal development for each ﬁsh using GSI: ð1Þ GSI ¼ gonad WW � 100 total WW Sagittal otoliths were collected for age determination and growth estimation. To determine ploidy, blood samples were collected via cardiac puncture (Duman et al. 2019), stored in tubes coated with lithium heparin (anticoagulant), and chilled until ploidy analysis could be performed at the Genomic Variation Laboratory at the University of California-Davis. Ploidy was determined for each ﬁsh captured in 2019 and 2020 from blood samples using a Coulter Counter with methods described by Fiske et al. (2019). A skinless ﬁllet, a 1-cm3 epaxial muscle sample with skin removed, and one gonad from each ﬁsh was collected, frozen, and held at �20 °C Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Farrell et al. 3 Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Fig. 1. Map of the Upper Colorado River Basin (UCRB) in Arizona, Colorado (CO), New Mexico, Utah (UT), and Wyoming (WY). Stars show reservoirs within the UCRB with walleye present. CO1 = Narraguinnep Reservoir, CO2 = McPhee Reservoir, CO3 = Puett Reservoir, CO4 = Riﬂe Gap Reservoir, CO5 = Stagecoach Reservoir, CO6 = Totten Lake, CO7 = Vallecito Lake, UT1 = Big Sand Wash Reservoir, UT2 = Lake Powell, UT3 = Midview Reservoir, UT4 = Red Fleet Reservoir, UT5 = Starvation Reservoir, WY1 = Jim Bridger Pond. Letters of the index code indicate the US state in which the waterbody resides. Map created in ArcMap 10.8 (Environmental Systems Research Institute 2020) with data from the National Elevation Dataset (US Geological Survey 2012), National Hydrography Dataset (US Geological Survey 2020a, 2020b, 2020c, 2020d), Geographic Names Information System (US Geological Survey 2017), and state boundaries (US Census Bureau 2020). Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) 4 Can. J. Fish. Aquat. Sci. Vol. 00, 0000 Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. until subsequent analyses could be completed. Stomachs were removed and either frozen or preserved in 10% formalin for later analysis. Potential prey items were collected opportunistically for stable isotope analysis and food web characterization. Sampling procedures were approved by the Institutional Animal Care and Use Committee (Protocol # 18-7822A) at Colorado State University. Observed somatic [T-Hg] Skinless ﬁllets from captured ﬁsh were analyzed for total mercury (T-Hg) as a surrogate for MeHg, as T-Hg analyses are less expensive and MeHg typically comprises approximately 95% of T-Hg in ﬁsh (Bloom 1992). Somatic [T-Hg] (lg·g�1 WW) were estimated for 56 diploid and 34 triploid walleyes captured during spring and summer 2019 by the Colorado Department of Public Health and Environment using EPA method 7473. Samples included nearly equal numbers of males and females for each ploidy. Walleye ranged from 248 to 680 mm TL. Triploids ranged in age from two to eight years, and diploids from two to 21 years. Linear models were used to test for differences in [T-Hg] between each sex and ploidy after accounting for age. We used Akaike information criterion with small sample size correction (AICc) to identify plausible models that best described patterns in [T-Hg] (Burnham and Anderson 2002). The full model was parameterized with a three-way interaction among age, sex, and ploidy (i.e., AGE � SEX � PLOIDY), and all models ranging in complexity from the full to the null model were tested (n = 19). Regressions were restricted to age classes represented by both ploidies in the catch (i.e., ≤age-8), which reduced the number of diploids included in the analysis to 34. Due to model selection uncertainty, comparisons of [T-Hg] among sex-by-ploidy groups were made using model-averaged estimates of regression coefﬁcients (Burnham and Anderson 2002). The 95% conﬁdence set of best models (i.e., cumulative AICc weight, v i ≤ 0.95) was used for model averaging (Symonds and Moussalli 2011). One outlier, an age-3 diploid female (0.467 lg T-Hg·g�1), had 107%–123% higher [T-Hg] than the other age-3 diploid females and was removed from this analysis. All statistical analyses were performed using R 4.0.3 (R Core Team 2020) and ﬁgures were created using package ggplot2 (Wickham 2016). Package MuMIn was used for AICc model selec� 2020). Package emmeans was tion and model averaging (Barton used to obtain estimated marginal means, and to compute contrasts and pairwise differences (Lenth 2020). We used a = 0.05 to determine statistical signiﬁcance, and Tukey-adjusted p values for pairwise comparisons. Energy density of somatic and gonadal tissue Epaxial muscle and gonadal tissue samples were weighed and then dried to remove water at 60 °C to a constant weight and homogenized with a mortar and pestle. We computed dry matter content (DM, %) for each tissue type as follows: ð2Þ DM ð%Þ ¼ dry mass ðgÞ � 100 WW Since muscle comprises the majority of soma for zander (Sander lucioperca; Jankowska et al. 2003), we used DM to estimate somatic energy densities (ED, J·g�1 WW) for each individual with the model of Hartman and Brandt (1995): ð3Þ ED ¼ 45:29 � DM1:507 We used AICc to identify plausible models (i.e., DAICc < 2) that best described patterns in somatic ED. The full model was parameterized with a three-way interaction among age, sex, and ploidy. Unlike somatic ED, we used a semimicro bomb calorimeter to directly measure gonadal ED, given uncertainty in the appropriateness of using existing ED–DM models that were developed by homogenizing whole ﬁsh (Johnson et al. 2017). Gonadal ED (J·g�1 dry weight) was measured with a Parr Instrument Company Model 6752 semimicro bomb calorimeter. A subsample of dried gonadal tissue (0.25 6 0.05 g) was pelletized and placed into the combustion chamber, which was charged to 30 atmospheres with pure oxygen (Parr Instrument Company 2013). Three subsamples were combusted from each gonad, and we only used data from subsamples that underwent complete combustion. Replicate measurements were averaged for each ﬁsh. A benzoic acid standard (Gundry et al. 1969) was used to verify accuracy and precision of the calorimeter. Standards were combusted at the beginning and end of each session, as well as after every tenth combustion cycle. Estimates of ED on a dry weight basis were converted to WW using corresponding estimates of percent water content. Linear regression was used to test for and characterize age dependency in male and female gonadal ED for diploids. Diet and stable isotope analysis We estimated the diet composition of walleye to inform prey proportions as input into the bioenergetics model. Stomachs preserved in the ﬁeld were dissected and prey items were identiﬁed to the lowest possible taxonomic classiﬁcation. The lengths of ingested prey were measured directly if intact or reconstructed from diagnostic bones. Diet composition was quantiﬁed using the displacement volume method of Hazzard and Madsen (1933). Each prey item was blotted to remove excess liquid and displacement volume was measured to the nearest 0.1 mL. Diet composition was expressed as the mean proportion of each prey taxon by volume. We used quantile regression to test whether maximum (95th percentile) and median (50th percentile) prey size increased with predator size (Chipps and Garvey 2007), as this could inﬂuence the ingestion of Hg by different age classes of walleye. Stable isotope analysis was used to generate more timeintegrated measures of resource use by each sex-by-ploidy group and to compliment stomach content analyses. Dried and homogenized epaxial muscle tissue samples from walleye and potential prey items were sent to the Cornell University Stable Isotope Laboratory for determination of d13C, d15N and C:N ratios. Carbon signatures were corrected for lipid content if necessary (Post et al. 2007; Skinner et al. 2016). Trophic position and dietary sources for different groups of walleye were inferred based on expected isotopic fractionation and corresponding increases in d15N (3.4%) and d13C (≤1.0%) values when moving from prey to predator (Post 2002). Growth estimation We estimated the age and growth of walleye from sagittal otoliths. Otoliths were sectioned transversely through the core and examined through a compound microscope using reﬂected light at 40–100� magniﬁcation. Ages were assigned, blind to ﬁsh length, independently by two readers. When assigned ages disagreed, both readers examined the structure together to reach a consensus-based age. Annulus radius measurements were made on magniﬁed images with the RFishBC package in R (Ogle 2019). Measurements were from the nucleus of the otolith to each annulus along the distal growth axis adjacent to the sulcus. Individual lifetime growth trajectories were estimated by backcalculation of length-at-age using otolith annulus radius measurements. Heidinger and Clodfelter (1987) found that walleye have an asymptotic otolith radius – total length (OR–TL) relationship, and Schirripa (2002) demonstrated that a Weibull cumulative function was the best model formulation for characterizing this relationship: �Rc �b � � � i ð4Þ Lci ¼ g 1 � e a where Lci is the TL at capture for ﬁsh i, Rci is otolith radius at capture for ﬁsh i, and a , b , and g are estimated parameters. We ﬁt the Weibull function to data separately for each sex using Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Farrell et al. 5 Table 1. Parameter estimates for weight–length regressions (WLR; ln WW = a + b � ln TL) and von Bertalanffy growth functions (VBGF) estimated for each sex-by-ploidy group of walleye in Narraguinnep Reservoir. WLR parameters von Bertalanffy parameters Ploidy Sex a b R2 L1 K t0 2N Male Female Male Female 1.48 � 10–6 4.12 � 10–6 3.68 � 10–6 9.55 � 10–7 3.28 3.10 3.12 3.35 0.976 0.979 0.965 0.987 495 (11.43) 573 (10.76) 514 (19.75) 534 (15.00) 0.373 (0.026) 0.285 (0.024) 0.322 (0.043) 0.305 (0.032) –0.074 (0.109) –0.210 (0.101) –0.294 (0.171) –0.192 (0.131) 3N Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Note: Corresponding standard errors for the VBGF parameter estimates are listed in parentheses. 2N = diploid, 3N = triploid. maximum likelihood estimation with lognormal error structure to account for increasing variation in Rc as age increased. Statistical model ﬁtting was performed using the bbmle package in R (Bolker and R Core Team 2020). We then applied the corresponding back-calculation model Schirripa (2002) developed for ﬁsh with an asymptotic OR–TL relationship: �Rc �b � � � �� Lc � ai 1�e ð5Þ Lit ¼ g i Lpi where Lit is the back-calculated TL for ﬁsh i at age t, and Lpi is the theoretical length of ﬁsh i according to its otolith radius as predicted by the ﬁtted OR–TL relationship. Back-calculated lengthat-age estimates are repeated measurements of size over time for individual ﬁsh, which have a hierarchical structure. Therefore, we ﬁt the parameters of the von Bertalanffy growth model to these data for each sex-by-ploidy group using hierarchical nonlinear maximum likelihood estimations following the procedures recommended by Ogle et al. (2017): ð6Þ Li ¼ L1h ð1 � eKh ½t�t0h � Þ þ « hi where Li is the TL for ﬁsh i, h is an integer that identiﬁes the population (i.e., sex-by-ploidy group), L1 is the asymptotic mean length, K is the Brody growth coefﬁcient, t0 is the theoretical age at zero length, and « hi are within-population random errors, assumed normally distributed with a mean of 0 and standard deviation of s . Lastly, length-at-age estimates from the ﬁtted von Bertalanffy growth curves were linked to WW–TL regressions (Table 1) derived for each sex-by-ploidy group during early summer (when gonadal tissue was minimal) to characterize somatic WW-at-age for incorporation into the bioenergetics model. Bioenergetics simulations We used a bioenergetics model informed by ﬁeld data and implemented in Fish Bioenergetics 4.0 (Deslauriers et al. 2017) to estimate prey consumption, dietary T-Hg exposure (i.e., the amount of Hg ingested), and resulting [T-Hg] in somatic tissue for triploid and diploid walleyes of each sex for comparison to empirical [T-Hg] observations. Consumption and respiration parameters were taken from Kitchell et al. (1977), and egestion and excretion parameters from Stewart et al. (1983). The walleye bioenergetics model assumed that standard metabolic rate (SMR) and activity did not vary between the sexes. Simulations were structured to align with the phenology of gonadal development from immediately post-spawn (somatic WW only) to immediately pre-spawn (somatic + full gonad WW) the following year. Simulations, ﬁt to observed annual growth in total WW, were conducted for an average individual from each age class (age-2 to age-8), sex, and ploidy to estimate daily prey consumption rates and resulting dietary T-Hg exposure accumulated over one year. Walleye typically spawn when surface water temperatures reach 5–10 °C in spring (Barton and Barry 2011). Therefore, we deﬁned day one of each simulation as the day after presumed spawning on April 1, when surface temperatures reached 7.0 °C in Narraguinnep Reservoir. The ﬁnal day was set to that of presumed spawning on March 31 the following year. Daily thermal experience was estimated from surface temperatures recorded with Onset HOBO Pendant UA-002-08 temperature loggers placed in Narraguinnep Reservoir during 2020. Simulations assumed a diet of 100% crayﬁsh (Orconectes spp.) based on results from diet and stable isotope analyses (see below). The indigestible proportion of crayﬁsh was set to 0.19 (Stein and Murphy 1976). Crayﬁsh ED (3706 J·g�1 WW) was derived from a study on another reservoir in southwestern Colorado (Pate et al. 2014). Dietary T-Hg exposure was calculated from the mean T-Hg value estimated for crayﬁsh (0.061 lg·g�1) from previous work on Narraguinnep and other nearby reservoirs (Colorado Department of Public Health and Environment 2021) multiplied by the total estimated annual consumption. Annual somatic gross growth efﬁciency (GGES) was calculated for each simulation as the change in predator somatic WW divided by the WW of prey consumed. Simulating differential reproductive investment We accounted for differential reproductive investment in estimates of prey consumption and dietary T-Hg exposure from the bioenergetics model by manipulating daily inputs for the somatic and gonadal ED of walleye to reﬂect observed differences in seasonal gonadal development and energy content among sexes and ploidies. Linear models developed above provided agespeciﬁc estimates of mean somatic and gonadal ED for each sexby-ploidy group. Next, we used a local polynomial regression model ﬁt to monthly GSI values estimated for male and female walleye from Henderson et al. (1996) to characterize the daily progression of gonad development and corresponding changes in gonad WW for each age class of mature diploid walleye over the one-year simulation period. Age-at-maturity was deﬁned as the ﬁrst age class where >50% of diploid ﬁsh sampled during the spawning season were deemed reproductive (≥age-3 for males and ≥age-5 from females). Conversely, for male and female triploid walleye of mature age — based on observations from their diploid counterparts — we used the mean GSI observed in spring to estimate their maximum gonad WW and mean GSI observed in summer to estimate their minimum gonad WW. These values were linearly interpolated to estimate daily GSIs and gonad WW over the simulation interval. Daily gonad and somatic WW were summed to generate daily total WW, and to then compute a mean composite whole-body ED weighted by the relative proportion of each tissue type. For diploid ﬁsh, simulations were conducted with (i.e., spawning diploids) and without gonadal development (i.e., non-spawning diploids) to separate the relative effect of reproductive investment from somatic growth and ED on model-predicted [T-Hg]. Translating dietary T-Hg exposure into somatic [T-Hg] Converting the bioenergetics-based estimates of dietary T-Hg exposure to expected [T-Hg] in somatic tissue enabled relative comparisons to the empirical data. Annual T-Hg accumulation was calculated by multiplying T-Hg exposure by the T-Hg assimilation efﬁciency of ﬁsh consuming contaminated crayﬁsh (0.94; Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) 6 Can. J. Fish. Aquat. Sci. Vol. 00, 0000 Table 2. Models retained (cumulative v i ≤ 0.95) for predicting [T-Hg] based on sex, ploidy, and ﬁsh age for walleyes captured at Narraguinnep Reservoir (March and June 2019). Model structure R2 AGE � PLOIDY + SEX AGE � PLOIDY AGE � SEX � PLOIDY AGE � SEX + AGE � PLOIDY 0.833 6 0.825 5 0.846 9 0.833 7 k loglik AICc DAICc vi 92.1 90.7 94.7 92.1 –170.6 –170.4 –168.2 –168.1 0.000 0.197 2.381 2.492 0.350 0.317 0.107 0.101 Fig. 2. Observed somatic mercury concentrations [T-Hg] by age for walleye from Narraguinnep Reservoir (points). Lines represent model-averaged estimates of regression coefﬁcients for retained models (cumulative v i ≤ 0.95). 2N = diploid, 3N = triploid, F = female, M = male. [Colour online.] Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Note: R2 = coefﬁcient of determination, k = number of parameters, loglik = model log-likelihood, AICc = Akaike’s information criterion adjusted for small sample sizes, DAICc = difference in AICc between the given model and the most parsimonious model, v i = Akaike weights. The full model was SEX � AGE � PLOIDY. Bowling et al. 2011) and subtracting the amount of T-Hg lost to spawning. Spawning losses were assumed zero for triploids and immature diploids, 0.043 lg Hg·g�1 gonadal tissue for mature diploid females, and 0.055 lg Hg·g�1 gonadal tissue for mature diploid males (Mauk and Brown 2001). Since Mauk and Brown (2001) measured whole testes rather than milt, we assumed that males lost 100% of the T-Hg in their testes due to spawning. Cumulative somatic T-Hg burdens adjusted for losses from spawning were then divided by ending somatic WW to obtain expected [T-Hg] for each age class. We used the average observed [T-Hg] for age-2 walleye from Narraguinnep Reservoir as the starting value in our calculation of predicted [T-Hg] from the bioenergetics model. Because the metabolic elimination rate of MeHg is variable and poorly understood (Yao and Drouillard 2019), our conversions between dietary T-Hg exposure and [T-Hg] did not include metabolic elimination and therefore should be viewed as overestimates. Results A total of 473 walleyes were sampled from Narraguinnep Reservoir between 20 March 2018 and 20 March 2020. Diploid females, diploid males, triploid females, and triploid males made up 11.0%, 67.4%, 10.2%, 11.4% of the catch in spring samples and 26.4%, 31.3%, 23.6%, 18.8% in summer samples, respectively. We sampled diploid walleye from the 1998–2017 cohorts, and triploids from the 2011–2017 cohorts. While triploid stocking at Narraguinnep Reservoir began in 2008, it appears the ﬁrst two cohorts (2008 and 2010) failed to recruit. Table 3. Relative pairwise contrasts of [T-Hg] at age-8 derived from the average model. Observed somatic [T-Hg] Somatic [T-Hg] ranged from 0.13 to 0.95 lg T-Hg·g�1 across age classes and sex-by-ploidy groups. The results of our linear modelling revealed that triploids had lower [T-Hg] than diploids after accounting for age. The top two models, AGE � PLOIDY + SEX (F[3,57] = 69.79; R2 = 0.833; p < 0.0001; v i = 0.350) and AGE � PLOIDY (F[2,57] = 89.7; R2 = 0.825; p < 0.0001; v i = 0.317), were the only models with DAICc < 2, and accounted for 66.8% of the Akaike weights (Table 2). Two other models, AGE � SEX � PLOIDY (F[7,53] = 42.1; R2 = 0.848; p < 0.0001; v i = 0.107) and AGE � SEX + AGE � PLOIDY (F[5,55] = 54.9; R2 = 0.833; p < 0.0001; v i = 0.101), were included in the average model (Table 2; Fig. 2). At age-8, diploid females had 30.8% (95% CI = 24.1%–37.5%, p < 0.0001) higher [T-Hg] than triploid females, and diploid males had 28.3% (95% CI = 20.3%–36.2%, p = 0.0002) higher [T-Hg] than triploid males (Table 3). Within each ploidy, males had higher [T-Hg] than females, but these contrasts were not signiﬁcantly different (Table 3). Note: p values are Tukey-adjusted. Bolded sex-by-ploidy group indicates the group the percent difference is relative to. 2N = diploid, 3N = triploid, F = female, M = male. Gonadosomatic index Gonadal development as characterized by seasonal GSI values varied by sex and ploidy. Diploids of both sexes exhibited much higher GSI values than their triploid counterparts during the spawning period (Fig. 3). Mean GSI for spawning diploid females was 12.6% (n = 17, SD = 4.77%) compared to 0.4% (n = 11, SD = 0.20%) for similarly aged triploid females. Mean GSI for spawning diploid males was 2.3% (n = 157, SD = 0.76%) compared to 0.3% (n = 11, Contrast Percent difference (95% CI) t ratio p value 2NF � 3NF 2NF � 2NM 2NF � 3NM 3NF − 2NM 3NF − 3NM 2NM � 3NM 30.8% (24.1%�37.5%) 0.7% (–5.7%�7.1%) 27.8% (20.9%�34.7%) 31.2% (23.9%�38.6%) 4.2% (–1.3%�9.7%) 28.3% (20.3%�36.2%) 5.769 –0.137 5.048 –5.399 –0.686 4.491 <0.0001 0.9991 <0.0001 <0.0001 0.9019 0.0002 SD = 0.03%) for triploid males. Mean GSI values observed for mature diploids decreased by more than 80% from spring to summer. In the summer, mean GSI for mature diploid females was 1.5% (n = 20; SD = 0.35%), and 0.3% (n = 25, SD = 0.25%) for mature diploid males. Mean GSI values observed for mature aged triploids during summer were only slightly lower than those during spring for both females (0.2%, SD = 0.14%, n = 2) and males (0.1%, SD = 0.005%, n = 2). Energy density of somatic and gonadal tissue The best model as determined by AICc model selection demonstrated a signiﬁcant effect of sex and ploidy on somatic ED after accounting for age, and this difference was driven by diploid females. The full model (i.e., AGE � SEX � PLOIDY) was the only model with a DAICc value < 2 (F[7,165] = 9.71; R2 = 0.292; p < 0.001). Estimated marginal mean somatic EDs were similar among the sex-by-ploidy groups (diploid males = 3983 J·g�1 WW, triploid males = 3951 J·g�1 WW, triploid females = 3890 J·g�1 WW, diploid females = 3811 J·g�1 WW). However, the effect of age on somatic ED was not the same for all groups, ED increased signiﬁcantly with age for all except diploid females. The slope for diploid Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Farrell et al. 7 Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Fig. 3. Summarized distributions of gonadosomatic index values (%) by sex and ploidy for adult walleye captured during the spawning period in March 2019 and March 2020 in Narraguinnep Reservoir. Boxes show the median (thick line), ﬁrst and third quartiles (lower and upper hinges). Box whiskers extend from the hinges to the most extreme value no further than 1.5 � the interquartile range from the hinge. 2N = diploid, 3N = triploid, F = female, M = male. females ( b age = �40.8 J·year�1; SE = 41.7 J·year�1) was signiﬁcantly different from triploid females (b age = 111.5 J·year�1; SE = 27.6 J·year�1; p = 0.014) and diploid males ( b age = 132.1 J·year�1; SE = 24.9 J·year�1; p = 0.003), but not from triploid males ( b age = 86.1 J·year�1; SE = 37.0 J·year�1; p = 0.107). All other pairwise contrasts of slopes among sex-by-ploidy groups for somatic ED were not signiﬁcant (p ≥ 0.73). Furthermore, the ED of fully developed diploid testes increased signiﬁcantly with age ( b age = 88.9 J·year�1; SE = 26.3 J·year�1; R2 = 0.289; p = 0.0022), and averaged 3875 J·g�1 WW (SD = 750 J·g�1 WW; n = 30) across age classes. No age dependency was observed for the ED of fully developed diploid ovaries (F[1,8] = 0.251; p = 0.63), which averaged 10 048 J·g�1 WW (SD = 973 J·g�1 WW; n = 10). Gonadal energy densities estimated for triploids averaged 3680 J·g�1 WW (SD = 624 J·g�1 WW; n = 7) for females and 2991 J·g�1 WW (SD = 892 J·g�1 WW; n = 3) for males. Diet and stable isotope analysis Stomach content analysis demonstrated minimal variation in diet composition among sexes and sizes of walleye. Overall, we analyzed 60 stomachs, of which 16 were empty, from walleye ranging in length from 233–593 mm TL captured during summer 2018. By volume, crayﬁsh comprised 94.4% of the diets, yellow perch accounted for 5.3%, and ﬁsh of unknown species comprised 0.2%. Diet composition was similar for males (96% crayﬁsh by volume; n = 24) and females (94% crayﬁsh by volume; n = 36). Carapace lengths of crayﬁsh found in walleye stomachs ranged in size from 13–54 mm and averaged 34 mm (SD = 9.48 mm; n = 38). Yellow perch found in walleye stomachs ranged in size from 33– 81 mm TL and averaged 52 mm (SD = 13.5 mm; n = 12). There was no evidence of a relationship between predator TL and maximum (F[1,52] = 0.91; p = 0.541) or median (F[1,52] = 0.03; p = 0.878) prey size. Unforeseen issues regarding laboratory sample analyses precluded ploidy determination for ﬁsh captured in 2018. However, the consistent and strong presence of crayﬁsh indicated minimal differences in diet between ploidies. Stable isotope analyses conﬁrmed minimal differences in diet among age classes of walleye (n = 173) and sex-by-ploidy groups and supported the notion that walleye in Narraguinnep Reservoir prey primarily on crayﬁsh (Fig. 4). Mean d15N and d13C values for each sex-by-ploidy group were indistinguishable (Table 4). In addition, the linear model with age only best described variation in d15N (F[1,171] = 168; R2 = 0.495; p < 0.001) based on AICc. However, the estimated rate of change in d15N with age was low ( b age = 0.165 %·year�1), reﬂecting a negligible shift in trophic position from age-2 to age-8. Conversely, the top model for d13C was SEX + AGE � PLOIDY (F[5,167] = 19.9; R2 = 0.36; p < 0.001). Only the pairwise contrasts between marginal means for diploid males and triploid females (p = 0.001), and diploid and triploid males (p = 0.031) were signiﬁcantly different. These differences were <0.312% and not ecologically meaningful given the relatively large range in d13C present in the food web. The position of walleye relative to potential prey and isotopic baselines in d-space aligned with a diet predominated by crayﬁsh based on the expected enrichment of d15N and d13C when moving from prey to predator (Fig. 4). Great pond snails (Lymnaea stagnalis) (n = 51; d15N = 5.30% 6 0.87%; d13C = �32.0% 6 1.38%; mean 6 SD) and bulk zooplankton samples (n = 2; d15N = 5.1% 6 0.07%; d13C = �36.5% 6 0.22%) represented the benthic and pelagic baselines, respectively. Crayﬁsh sampled (n = 69; d15N = 8.79% 6 0.95%; d13C = �25.2% 6 1.20%) ranged in size from 22–65 mm carapace length (mean = 49 mm), which aligned well with the range of sizes observed in walleye diets. Potential ﬁsh prey, represented by yellow perch (n = 6; d15N = 11.1% 6 1.27%; d13C = �26.5% 6 1.73%), ranged in size from 50–265 mm total length (mean 6 SD = 131 6 89.2 mm). The trophic positions of walleye, yellow perch, and crayﬁsh relative to the benthic baseline were 3.09, 2.71, and 2.03, respectively. Thus, walleye and crayﬁsh approximately differed by one full trophic level and remained within the expected range for d13C, whereas the difference in trophic position between walleye and yellow perch was too small to suggest that yellow perch were common prey for walleye in Narraguinnep Reservoir. Predicted somatic [T-Hg] from bioenergetics model After integrating ﬁeld data on ED, GSI, diet and growth, the bioenergetics simulations showed that the energetic costs of spawning increased [T-Hg] for females, but not males, due to differences in prey consumption required to meet somatic and gonadal growth. On a cumulative basis from age-2 to age-8, Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) 8 Can. J. Fish. Aquat. Sci. Vol. 00, 0000 Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Fig. 4. (a) Bi-plot of stable isotope values from individual walleye (age ≤ 8) and potential prey items in Narraguinnep Reservoir. Lines represent 95% conﬁdence ellipses for walleye by sex and ploidy, and error bars for mean values from prey items represent 95% conﬁdence intervals. (b) Magniﬁed view for walleye only from panel (a). CRA = crayﬁsh, GPS = great pond snail, YPE = yellow perch, ZOO = zooplankton, 2NF = diploid female walleye, 2NM = diploid male walleye, 3NF = triploid female walleye, 3NM = triploid male walleye. [Colour online.] Table 4. Mean d15N and d13C (%) values and corresponding standard deviations estimated for each sex-by-ploidy group of walleye (age ≤ 8) captured at Narraguinnep Reservoir (March and June 2019). Ploidy Diploid Sex Male Female Triploid Male Female Total length (mm) d15N (%) d13C (%) n Range Mean SD Mean SD Mean SD 57 26 31 59 310–519 258–537 248–479 269–506 425 428 357 392 12.3 12.4 12.2 12.3 –25.3 –25.2 –25.4 –25.2 49.6 70.6 63.1 71.3 0.44 0.37 0.45 0.39 Fig. 5. Somatic total mercury concentrations ([T-Hg]; lg·g�1) by age for walleye estimated by the bioenergetics model. 2N = diploid, 3N = triploid, F = female, M = male. For diploid walleye, solid lines represent spawning ﬁsh (full gonadal development observed in the ﬁeld), while dashed lines represent hypothetical non-spawning ﬁsh (no gonadal development). [Colour online.] 0.53 0.45 0.54 0.57 spawning diploid females consumed the most prey (22 648 g) and T-Hg (1382 lg), followed by non-spawning diploid females (17 182 g; 1408 lg), triploid females (16 442 g; 1003 lg), spawning diploid males (15 499 g; 945 lg), nons-pawning diploid males (15 049 g; 918 lg), and triploid males (14 649 g; 894 lg). From age-at-maturity to age-8, diploid males required 405.4 kJ of energy to develop testes, while diploid females required 7525.9 kJ to develop ovaries. Predicted [T-Hg] for spawning diploid females diverged from all other groups beginning at age-5, when they began devoting energy toward egg development (Fig. 5). At age-8, [T-Hg] for spawning diploid females (0.900 lg·g�1) was 25% higher than non-spawning diploid females (0.719 lg·g�1), 20% higher than triploid females (0.748 lg·g�1), and 8% higher than spawning diploid males (0.836 lg·g�1). Spawning diploid males had 2% higher [T-Hg] than non-spawning diploid males (0.820 lg·g�1 T-Hg), and 4% higher [T-Hg] than triploid males (0.806 lg·g�1 T-Hg). Triploid males had 7% higher [T-Hg] than triploid females. Annual patterns in total WW-at-age and corresponding modelderived GGES for each sex-by-ploidy group helped demonstrate the effects of allocating energy to gamete production on somatic growth and why [T-Hg] for spawning diploid females diverged from all other groups. While spawning diploid females had higher GGES prior to maturity relative to their triploid counterparts, their GGES decreased following maturation (Fig. 6). By age-8, the energetic costs of spawning decreased GGES for diploids of both sexes, but more so for females than males. Non-spawning diploid females had 46% higher GGES (0.347) than spawning diploid females (0.238), and non-spawning diploid males (0.321) had 4% higher GGES than spawning diploid males (0.310). Triploid females (0.329) had 38% higher GGES than spawning diploid females, and triploid males (0.321) had 4% higher GGES than spawning diploid males. Discussion This is the ﬁrst study to investigate the role of reproductive investment in Hg bioaccumulation using triploid ﬁsh as reproductive controls. As hypothesized, triploid females had signiﬁcantly lower observed somatic [T-Hg] than diploid females. This difference could not be explained by diet and somatic growth alone. Relative comparisons of predicted [T-Hg] among spawning diploid, non-spawning diploid, and triploid females from the bioenergetics model indicated that the observed difference in somatic Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Farrell et al. 9 Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Fig. 6. (a) Total wet weight-at-age (somatic + gonadal tissue) and (b) annual somatic gross growth efﬁciency (GGES) estimated from the bioenergetics model for walleye in Narraguinnep Reservoir. For diploid walleye, solid lines represent spawning ﬁsh (full gonadal development observed in the ﬁeld), while dashed lines represent hypothetical non-spawning ﬁsh (no gonadal development). Total wet weight-at-age was used to ﬁt the bioenergetics models. The dotted and dashed vertical lines show observed age-at-maturity for diploid males and females, respectively. [Colour online.] [T-Hg] was mostly driven by increased prey consumption required by diploid females to meet the energetic demand of ovarian development. Contrary to the premise of our second hypothesis, triploid males exhibited reduced testicular development. In addition, our second hypothesis was not supported, as triploid males showed lower [T-Hg] than diploid males despite similar diet and somatic growth rates. However, unanticipated differential testicular development could not explain observed differences in [T-Hg] between ploidies for males in the bioenergetics model. Lastly, our third hypothesis was only partially supported. We expected that males of both ploidies would exhibit lower [T-Hg] than diploid females because testes are typically less energetically costly to produce than ovaries. As anticipated, our bioenergetics modelling conﬁrmed that testes are less energetically costly to produce than ovaries. Differences in the energetic costs of reproductive development likely drove the disparity in observed [T-Hg] between diploid females and triploid males. However, observed [T-Hg] for diploid males did not differ from diploid females despite reduced consumption requirements. Thus, processes unrelated to investment of energy into gonadal tissue was driving elevated [T-Hg] in diploid males relative to a priori expectations. We hypothesized that reproductive investment was an important process driving differential [T-Hg] between diploid and triploid females because the energetic demand of ovarian development would require additional prey consumption. Since ﬁsh primarily uptake Hg via food (Hall et al. 1997), the need for additional prey should lead to increased dietary Hg exposure and decreased GGES for diploid females. Essentially, reproductive investment should “distill” Hg from food that accumulates in somatic tissue because the biochemical composition of eggs is not conducive to the binding of MeHg and resulting [MeHg] in eggs are typically low (Niimi 1983). Methylmercury primarily binds to sulfhydryl groups in muscle protein (Wiener and Spry 1996), whereas teleost eggs are primarily composed of lipids and yolk proteins (Brooks et al. 1997). For example, mean [T-Hg] measured in eggs from seven walleye populations across the United States and Canada was only Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. 10 2.1% of that in muscle (Johnston et al. 2001). Thus, eggs are not a meaningful route of elimination of Hg in walleye. Results from the bioenergetics simulations supported the “distillation” hypothesis for females. The relative magnitudes of the difference in model-predicted [T-Hg] at age-8 between spawning diploids and triploids (20%) and between spawning and nonspawning diploids (25%) were similar to that observed between diploid and triploid females in the ﬁeld (31%). Walleye eggs were more than 2.5 times as energy-dense as somatic tissue. Since ovaries comprised �15% of the total body WW when fully developed, diploids had higher annual energetic requirements relative to triploids and ingested more T-Hg. On a cumulative basis, spawning diploid females consumed 38% and 32% more prey by WW and their GGES was 27.7% and 31.4% lower than triploid and nonspawning diploid females, respectively. Assuming the bioenergetics model characterized Hg dynamics perfectly, the inclusion of differential reproductive investment explained 53%–104% of the observed difference in [T-Hg] for females. The premise of our second hypothesis, namely that triploid males in Narraguinnep Reservoir would exhibit normal testicular development, was not supported. We did not capture any triploid males that expressed milt, nor ﬁnd strong evidence of sexual maturation, during the spawning season in 2019 or 2020. Triploid males from a diverse set of species typically show normal testicular development (Benfey 1999), but there are exceptions (Tiwary et al. 2004). We are unsure why triploid walleye may be another exception to the norm, but Tiwary et al. (2000) suggested that low levels of sex steroids may impair testicular development in triploid stinging catﬁsh (Heteropneustes fossilis). Alternatively, it is possible that testicular development is severely delayed for triploid male walleye, and the population of triploid males in Narraguinnep Reservoir has not yet reached sexual maturity. Furthermore, we did not ﬁnd support for our second hypothesis that diploid and triploid males would have similar [T-Hg]. Rather, diploid males had 28.3% higher observed [T-Hg] than triploid males at age-8. However, differences in testicular development likely did not contribute to the differences in T-Hg bioaccumulation between triploids and diploids, as the investment of energy into reproductive tissue had little inﬂuence on estimated consumption rates for males. At age-8, the bioenergetics model predicted relatively small differences in [T-Hg] among triploid, spawning diploid, and nonspawning diploid males; all pairwise comparisons of predicted [T-Hg] among these groups were less than 4%. Thus, while triploid males did not develop testes to the same extent as mature diploid males, similarities in diet, body composition, and somatic growth resulted in similar consumption requirements according to the bioenergetics model. Our third hypothesis postulated that males of both ploidies would have lower [T-Hg] than diploid females due to differences in the energetic costs of developing testes versus ovaries. As expected, the energetic cost of producing testes was low relative to the cost of producing ovaries. On a cumulative basis from age-2 to age-8, ovarian development required 18.6-times more energy intake than testicular development, despite males reaching maturity two years earlier. As a result, the bioenergetics model predicted that [T-Hg] for all male groups (i.e., spawning diploids, non-spawning diploids, and triploids) would fall below diploid females by 7%–10%. However, only the triploid males had signiﬁcantly lower observed [T-Hg] than diploid females at age-8, while diploid males exhibited similar observed [T-Hg] as diploid females of the same age. Differences in swimming activity among the sex-by-ploidy groups could explain why diploid male walleye exhibited elevated [T-Hg] relative to a priori expectations and corresponding predictions from the bioenergetics model. As in many species of ﬁsh (Acolas et al. 2004; Altimiras et al. 1996; Bruch and Binkowski 2002; Dean et al. 2014; Lucas 1992), male walleyes are more active during the spawning period than females (Becker 1983). Female walleye Can. J. Fish. Aquat. Sci. Vol. 00, 0000 participate in spawning for a few days at most, while males roam the spawning grounds for up to four weeks (Becker 1983). In addition, scramble competition among male walleyes for mating opportunities substantially increases activity and associated metabolic costs (Henderson et al. 2003). Catch rates for diploid males were much higher than those for any other sex-by-ploidy group during the spawning season in Narraguinnep Reservoir; diploid males represented 67% of the catch, whereas all other groups only represented 10%–11% of the catch. During summer however, when spawning should not inﬂuence behavior or distribution, our catch was more evenly distributed among the sex-by-ploidy groups, and diploid males comprised 26% of the catch. The relative proportions of diploid males caught with passive gear during and outside of the spawning period suggest that diploid males are likely more active than the other sex-byploidy groups. Likewise, differences in standard metabolic rates (SMRs), in addition to higher swimming activity, could also explain why diploid males exhibited elevated [T-Hg] relative to a priori expectations and corresponding predictions from the bioenergetics model. Standard metabolic rates are generally higher for male ﬁsh than for female ﬁsh, and this pattern is evident for a wide range of vertebrates (Madenjian et al. 2016). Malison et al. (1985) found that administered doses of 17a-methyltestosterone proportionally inhibited growth in yellow perch, which suggests that testosterone may increase SMRs in percids. The minimal testicular development we observed for triploid male walleye could have resulted in lower testosterone, and therefore lower SMR, leading to higher growth efﬁciencies and lower [T-Hg] relative to diploid males. While previous laboratory studies have not found differences in SMRs between triploid and diploid ﬁsh (see Maxime 2008), these studies used immature ﬁsh, and we would not expect ploidy-speciﬁc differences in SMRs to emerge until diploids reach sexual maturity. It is also important to note that observed patterns in [T-Hg] among diploid males, triploid males and diploid females did not solely reﬂect differences in energy intake or expenditure, as diploid males likely eliminated Hg at a faster rate than any other sex-by-ploidy group (Madenjian et al. 2016). Although unexpected, reduced testicular development and [T-Hg] in triploid males offered more direct insight than previously possible into potential behavioral and physiological mechanisms for why similarly aged diploid males exhibited similar [T-Hg] as larger-bodied diploid females in this study and elsewhere (Henderson et al. 2003; Selch et al. 2019). Further comparisons between diploid and triploid walleye could help elucidate the role of testosterone in Hg elimination (Madenjian et al. 2014), quantify sex-dependent standard metabolic rates and activity costs associated with spawning, and improve energeticsbased Hg bioaccumulation models. Trophic position affects predator [T-Hg] because Hg biomagniﬁes in food webs (Lavoie et al. 2013). For example, [T-Hg] in lake trout (Salvelinus namaycush) was positively correlated to d15N, a suitable indicator of trophic position, in seven Canadian shield lakes (Cabana and Rasmussen 1994). Similarly, Taylor et al. (2020) found that an increase of trophic position for lake trout following a dietary shift from invertebrates to prey ﬁsh increased their [T-Hg]. In this study, stable isotope analyses indicated that the trophic position of walleye did not differ among sex-by-ploidy groups. A difference in d15N of 3.4% denotes a full shift in trophic position (Post 2002). The maximum difference in mean d15N among sex-by-ploidy groups for walleye in Narraguinnep Reservoir was only 0.3% and did not contribute to observed variation in [T-Hg] among groups. Diet composition inﬂuences Hg bioaccumulation because prey consumption is the dominant pathway of MeHg uptake in predators (Hall et al. 1997). Lepak et al. (2012a) found that prey selection differed by sex for walleye in two Colorado reservoirs: males consumed smaller, lower quality prey (i.e., resident ﬁsh and invertebrates), while females consumed larger, higher quality prey Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Farrell et al. (i.e., stocked rainbow trout). Males exhibited higher [T-Hg] than females because of their dietary differences. While typically piscivorous, walleye consume invertebrate prey in systems with depauperate prey ﬁsh assemblages (Chipps and Graeb 2011), which is the case for many Colorado reservoirs (Johnson et al. 2015; Wolff et al. 2017). While we acknowledge that our diet analysis was temporally limited and we were not able to include ploidy in our comparisons, stable isotope analyses conﬁrmed that walleye in Narraguinnep Reservoir were primarily invertivorous, and that all sex-by-ploidy groups were consuming similar prey items. Thus, dietary differences are not a plausible explanation for patterns of observed [T-Hg] among groups. Under equivalent assimilation efﬁciency and elimination, Hg bioaccumulation is largely controlled by the contamination level in prey, the amount of prey consumed, and the efﬁciency with which the predator allocates consumed energy to growth (Trudel and Rasmussen 2006). Faster growing ﬁsh may consume less food to reach a given size than slower growing ﬁsh if feeding on higher quality prey or occupying more suitable thermal habitat, reducing contaminant concentrations through growth dilution (Simoneau et al. 2005). Indeed, growth dilution can contribute to differences in [T-Hg] between sexes in diploid walleye (Madenjian et al. 2016). Neither diet composition nor thermal regime varied among the four combinations of sex and ploidy in our study. Thus, differences in [T-Hg] among these four groups were not attributable to differences in diet composition or thermal experience. Our results demonstrated that investment of energy into ovaries could account for a substantial portion of the difference in [T-Hg] between diploid and triploid female walleye. Further, our results suggested that higher energy expenditure by diploid male walleye, stemming from higher SMR and greater swimming activity, compared with triploid male walleye could account for the higher [T-Hg] observed for diploid male walleye. Regardless of the potential mechanisms, the signiﬁcantly lower mean [T-Hg] exhibited by triploid walleye in our study suggests that ploidy manipulations for stocked ﬁsh could help address widespread MeHg contamination problems in ﬁsh destined for human consumption. Reducing [MeHg] in ﬁsh and wildlife by regulatory actions is challenging. Atmospheric Hg levels have risen steadily with humanity’s reliance on fossil fuels (Eagles-Smith et al. 2016a) and are forecasted to increase further with the expansion of coal-ﬁred electricity generation in the developing world (Driscoll et al. 2013). The long occupancy and dispersal of Hg in the atmosphere (Boening 2000) decouples point-source pollution from loading to aquatic systems. Thus, deposition from distant sources may undermine local efforts to reduce Hg inputs. Furthermore, recycling of legacy Hg can prolong bioaccumulation in marine and freshwater ecosystems even when external sources are controlled (Amos et al. 2013; Driscoll et al. 2013). While global actions to reduce anthropogenic Hg loading to the environment are essential, alternative management interventions at the local scale are needed to protect ﬁsh, wildlife, and humans from elevated Hg exposure. Methylmercury concentrations in ﬁsh can be manipulated by several common ﬁsheries management strategies (Lepak et al. 2012a, 2012b; Sharma et al. 2008). The use of triploid ﬁsh in stocking programs alone or in conjunction with other strategies could prove useful for reducing [MeHg] in ﬁsh and subsequent exposure to wildlife and humans. Furthermore, since triploid ﬁsh grew more efﬁciently and consumed less prey, they may serve as a viable alternative stocking option in areas where the predatory impact of piscivorous sport ﬁsh is of concern. Acknowledgements Funding for this study was provided by Colorado Parks and Wildlife (CPW), and Colorado Department of Public Health and Environment. We appreciate the keen insight of two anonymous 11 referees, whose suggestions greatly improved the manuscript. We thank George Schisler, John Alves, Jeff Spohn, and Lori Martin for ﬁnancial support and Jim White, Ryan Lane, Tyler Kersey, and several other CPW regional personnel for logistical support. Grant Wilcox created the map. Noah Angell, James Grinolds, Adam Hundley, Lillian Legg, Michael Miller, Bill Pate, Joe Pitti, Lindsey Roberts, Jesse Stokes, and Matt Swerdfeger assisted in the ﬁeld and laboratory. Ryan Farrell assisted with computational programming. Montezuma Valley Irrigation Company provided access to the reservoir. References Acolas, M.L., Bégout Anras, M.L., Véron, V., Jourdan, H., Sabatié, M.R., and Baglinière, J.L. 2004. An assessment of the upstream migration and reproductive behaviour of allis shad (Alosa alosa L.) using acoustic tracking. ICES. J. Mar. Sci. 61(8): 1291–1304. doi:10.1016/j.icesjms.2004.07.023. Allen, S.K., and Wattendorf, R.J. 1987. Triploid grass carp: status and management implications. Fisheries, 12(4): 20–24. doi:10.1577/1548-8446(1987) 012<0020:TGCSAM>2.0.CO;2. Altimiras, J., Johnstone, A.D.F., Lucas, M.C., and Priede, I.G. 1996. Sex differences in the heart rate variability spectrum of free-swimming Atlantic salmon (Salmo salar L.) during the spawning season. Physiol. Zool. 69(4): 770–784. doi:10.1086/physzool.69.4.30164229. Amos, H.M., Jacob, D.J., Streets, D.G., and Sunderland, E.M. 2013. Legacy impacts of all-time anthropogenic emissions on the global mercury cycle. Global Biogeochem. Cycles, 27(2): 410–421. doi:10.1002/gbc.20040. Barton, B.A., and Barry, T.P. 2011. Reproduction and environmental biology. In Biology, management, and culture of walleye and sauger. Edited by B.A. Barton. American Fisheries Society, Bethesda, Md. pp. 233–302. � , K. 2020. MuMIn: multi-model inference R package version 1.43.17. Barton Available from https://CRAN.R-project.org/package=MuMIn. Bastos, W.R., Dorea, J.G., Bernardi, J.V.E., Manzatto, A.G., Mussy, M.H., Lauthartte, L.C., et al. 2016. Sex-related mercury bioaccumulation in ﬁsh from the Madeira River, Amazon. Environ. Res. 144(Pt. A): 73–80. doi:10.1016/j. envres.2015.11.001. PMID:26580025. Becker, G.C. 1983. Fishes of Wisconsin. University of Wisconsin Press, Madison. Benfey, T.J. 1999. The physiology and behavior of triploid ﬁshes. Rev. Fish. Sci. 7(1): 39–67. doi:10.1080/10641269991319162. Bloom, N.S. 1992. On the chemical form of mercury in edible ﬁsh and marine invertebrate tissue. Can. J. Fish. Aquat. Sci. 49(5): 1010–1017. doi:10.1139/f92-113. Boening, D.W. 2000. Ecological effects, transport, and fate of mercury: a general review. Chemosphere, 40(12): 1335–1351. doi:10.1016/S0045-6535(99)00283-0. PMID:10789973. Bolker, B., and R Core Team. 2020. bbmle: tools for general maximum likelihood estimation R package version 1.0.23.1. Available from https://CRAN. R-project.org/package=bbmle. Bowling, A.M., Hammerschmidt, C.R., and Oris, J.T. 2011. Necrophagy by a benthic omnivore inﬂuences biomagniﬁcation of methylmercury in ﬁsh. Aquat. Toxicol. 102(3–4): 134–141. doi:10.1016/j.aquatox.2011.01.006. PMID: 21356175. Brooks, S., Tyler, C.R., and Sumpter, J.P. 1997. Egg quality in ﬁsh: what makes a good egg? Rev. Fish Biol. Fish. 7(4): 387–416. doi:10.1023/A:1018400130692. Bruch, R.M., and Binkowski, F.P. 2002. Spawning behavior of lake sturgeon (Acipenser fulvescens). J. Appl. Ichthyol. 18(4–6): 570–579. doi:10.1046/j.14390426.2002.00421.x. Bruner, J.C. 2021. Stizostedion Raﬁnesque, 1820 (Percidae) is the Valid Generic Name for Walleye, Sauger, and Eurasian Pikeperch. Fisheries, 46(6): 298– 302. doi:10.1002/fsh.10582. Burnham, K., and Anderson, D. 2002. Model selection and multi-model inference. 2nd ed. Springer, New York. Butler, D.L., Krueger, R.P., Osmundson, B.C., and Jensen, E.G. 1995. Reconnaissance investigation of water quality, bottom sediment, and biota associated with irrigation drainage in the Dolores Project area, southwestern Colorado and southeastern Utah, 1990-91. 94-4041, O.-F.R.S. US Geological Survey: USGS Earth Science Information Center. Cabana, G., and Rasmussen, J.B. 1994. Modelling food chain structure and contaminant bioaccumulation using stable nitrogen isotopes. Nature, 372(6503): 255–257. doi:10.1038/372255a0. Cassinelli, J.D., Meyer, K.A., Koenig, M.K., Vu, N.V., and Campbell, M.R. 2019. Performance of diploid and triploid westslope cutthroat trout fry stocked into Idaho alpine lakes. N. Am. J. Fish. Manage. 39(1): 112–123. doi:10.1002/ nafm.10254. Chen, C.Y., Serrell, N., Evers, D.C., Fleishman, B.J., Lambert, K.F., Weiss, J., et al. 2008. Meeting report: methylmercury in marine ecosystems — from sources to seafood consumers. Environ. Health Perspect. 116(12): 1706– 1712. doi:10.1289/ehp.11211. PMID:19079724. Chipps, S.R., and Garvey, J.E. 2007. Assessment of diets and feeding patterns. In Analysis and interpretation of freshwater ﬁsheries data. Edited by C.S. Guy and M.L. Brown. American Fisheries Society, Bethesda, Md. pp. 473–514. Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. 12 Chipps, S.R., and Graeb, B.D.S. 2011. Feeding ecology and energetics. In Biology and management of Walleye and Sauger. Edited by B.A. Barton. American Fisheries Society, Bethesda, Md. pp. 303–320. Coelho, J.P., Santos, H., Reis, A.T., Falcão, J., Rodrigues, E.T., Pereira, M.E., et al. 2010. Mercury bioaccumulation in the spotted dogﬁsh (Scyliorhinus canicula) from the Atlantic Ocean. Mar. Pollut. Bull. 60(8): 1372–1375. doi:10. 1016/j.marpolbul.2010.05.008. PMID:20541227. Colorado Department of Public Health and Environment. 2021. Fish consumption: site-speciﬁc guidelines. Available from https://cdphe.colorado.gov/waterquality/clean-water/rivers-lakes-and-streams/ﬁsh-consumption [accessed 19 January 2021]. Crump, K.L., and Trudeau, V.L. 2009. Mercury-induced reproductive impairment in ﬁsh. Environ. Toxicol. Chem. 28(5): 895–907. doi:10.1897/08-151.1. PMID:19102578. Dean, M.J., Hoffman, W.S., Zemeckis, D.R., and Armstrong, M.P. 2014. Finescale diel and gender-based patterns in behaviour of Atlantic cod (Gadus morhua) on a spawning ground in the Western Gulf of Maine. ICES. J. Mar. Sci. 71(6): 1474–1489. doi:10.1093/icesjms/fsu040. Deslauriers, D., Chipps, S.R., Breck, J.E., Rice, J.A., and Madenjian, C.P. 2017. Fish Bioenergetics 4.0: An R-based modeling application. Fisheries, 42(11): 586–596. doi:10.1080/03632415.2017.1377558. Diana, J. 1983. An energy budget for northern pike (Esox lucius). Can. J. Zool. 61(9): 1968–1975. doi:10.1139/z83-259. Driscoll, C.T., Mason, R.P., Chan, H.M., Jacob, D.J., and Pirrone, N. 2013. Mercury as a global pollutant: sources, pathways, and effects. Environ. Sci. Technol. 47(10): 4967–4983. doi:10.1021/es305071v. PMID:23590191. Duffy, M.J., McNulty, J.L., and Mosindy, T.E. 2000. Identiﬁcation of sex, maturity, and gonad condition of walleye (Stizostedion vitreum vitreum). Northwest Science and Technology Field Guide No. FG-05, Thunder Bay. Ontario Ministry of Natural Resources, Northwest Region Science and Technology Unit, Thunder Bay, Ont. Duman, M., Saticioglu, I.B., Suzer, B., and Altun, S. 2019. Practices for drawing blood samples from teleost ﬁsh. N. Am. J. Aquacult. 81(2): 119–125. doi:10.1002/ naaq.10077. Eagles-Smith, C.A., Ackerman, J.T., Willacker, J.J., Tate, M.T., Lutz, M.A., Fleck, J.A., et al. 2016a. Spatial and temporal patterns of mercury concentrations in freshwater ﬁsh across the Western United States and Canada. Sci. Total. Environ. 568: 1171–1184. doi:10.1016/j.scitotenv.2016.03.229. PMID: 27102274. Eagles-Smith, C.A., Wiener, J.G., Eckley, C.S., Willacker, J.J., Evers, D.C., Marvin-DiPasquale, M., et al. 2016b. Mercury in western North America: a synthesis of environmental contamination, ﬂuxes, bioaccumulation, and risk to ﬁsh and wildlife. Sci. Total. Environ. 568: 1213–1226. doi:10.1016/j. scitotenv.2016.05.094. PMID:27320732. Environmental Systems Research Institute. 2020. ArcMap version 10.8. Redlands, California. Fetherman, E.R., Lepak, J.M., Brown, B.L., and Harris, D.J. 2015. Optimizing time of initiation for triploid walleye production using pressure shock treatment. N. Am. J. Aquac. 77: 471–477. doi:10.1080/15222055.2015.1040568. Fiske, J.A., Van Eenennaam, J.P., Todgham, A.E., Young, S.P., Holem-Bell, C.E., Goodbla, A.M., and Schreier, A.D. 2019. A comparison of methods for determining ploidy in white sturgeon (Acipenser transmontanus). Aquaculture, 507: 435–442. doi:10.1016/j.aquaculture.2019.03.009. Gewurtz, S.B., Bhavsar, S.P., and Fletcher, R. 2011. Inﬂuence of ﬁsh size and sex on mercury/PCB concentration: Importance for ﬁsh consumption advisories. Environ. Int. 37(2): 425–434. doi:10.1016/j.envint.2010.11.005. PMID: 21168218. Gray, J.E., Fey, D.L., Holmes, C.W., and Lasorsa, B.K. 2005. Historical deposition and ﬂuxes of mercury in Narraguinnep Reservoir, southwestern Colorado, USA. Appl. Geochem. 20(1): 207–220. doi:10.1016/j.apgeochem.2004. 05.011. Grieb, T.M., Bowie, G.L., Driscoll, C.T., Gloss, S.P., Schoﬁeld, C.L., and Porcella, D.B. 1990. Factors affecting mercury accumulation in ﬁsh in the upper Michigan peninsula. Environ. Toxicol. Chem. 9(7): 919–930. doi:10.1002/ etc.5620090710. Gundry, H.A., Harrop, D., Head, A.J., and Lewis, G.B. 1969. Thermodynamic properties of organic oxygen compounds 21. Enthalpies of combustion of benzoic acid, pentan-1-ol, octan-1-ol, and hexadecan-1-ol. The J. Chem. Thermodyn. 1(3): 321–332. doi:10.1016/0021-9614(69)90051-2. Hall, B.D., Bodaly, R.A., Fudge, R.J.P., Rudd, J.W.M., and Rosenberg, D.M. 1997. Food as the dominant pathway of methylmercury uptake by ﬁsh. Water Air Soil Pollut. 100(1): 13–24. doi:10.1023/A:1018071406537. Hartman, K.J., and Brandt, S.B. 1995. Estimating energy density of ﬁsh. Trans. Am. Fish Soc. 124(3): 347–355. doi:10.1577/1548-8659(1995)124<0347:EEDOF>2.3. CO;2. Hazzard, A.S., and Madsen, M.J. 1933. Studies of the food of the cutthroat trout. Trans. Am. Fish Soc. 63(1): 198–203. doi:10.1577/1548-8659(1933)63 [198:SOTFOT]2.0.CO;2. Heidinger, R.C., and Clodfelter, K. 1987. Validity of the otolith for determining age and growth of walleye, striped bass, and smallmouth bass in power plant cooling ponds. In Age and growth of ﬁsh. Edited by R.C. Summerfelt and G.E. Hall. Iowa State University Press, Ames, Iowa. pp. 241–251. Henderson, B.A., Wong, J.L., and Nepszy, S.J. 1996. Reproduction of walleye in Lake Erie: allocation of energy. Can. J. Fish. Aquat. Sci. 53(1): 127–133. doi:10.1139/f95-162. Can. J. Fish. Aquat. Sci. Vol. 00, 0000 Henderson, B.A., Collins, N., Morgan, G.E., and Vaillancourt, A. 2003. Sexual size dimorphism of walleye (Stizostedion vitreum vitreum). Can. J. Fish. Aquat. Sci. 60(11): 1345–1352. doi:10.1139/f03-115. Jankowska, B., Zakes, Z., Zmijewski, T., and Szczepkowski, M. 2003. A comparison of selected quality features of the tissue and slaughter yield of wild and cultivated pikeperch Sander lucioperca (L.). Eur. Food Res. Technol. 217(5): 401–405. doi:10.1007/s00217-003-0757-5. Johnson, B.M., Arlinghaus, R., and Martinez, P.J. 2009. Are we doing all we can to stem the tide of illegal ﬁsh stocking? Fisheries, 34(8): 389–394. doi:10.1577/1548-8446-34.8.389. Johnson, B.M., Lepak, J.M., and Wolff, B.A. 2015. Effects of prey assemblage on mercury bioaccumulation in a piscivorous sport ﬁsh. Sci. Total. Environ. 506–507: 330–337. doi:10.1016/j.scitotenv.2014.10.101. PMID:25460967. Johnson, B.M., Pate, W.M., and Hansen, A.G. 2017. Energy density and dry matter content in ﬁsh: new observations and an evaluation of some empirical models. Trans. Am. Fish Soc. 146(6): 1262–1278. doi:10.1080/00028487.2017. 1360392. Johnston, T.A., Bodaly, R.A., Latif, M.A., Fudge, R.J., and Strange, N.E. 2001. Intra- and interpopulation variability in maternal transfer of mercury to eggs of walleye (Stizostedion vitreum). Aquat. Toxicol. 52(1): 73–85. doi:10.1016/ S0166-445X(00)00129-6. PMID:11163432. Kitchell, J.F., Stewart, D.J., and Weininger, D. 1977. Applications of a bioenergetics model to yellow perch (Perca ﬂavescens) and walleye (Stizostedion vitreum vitreum). J. Fish. Res. Bd. Can. 34(10): 1922–1935. doi:10.1139/f77-258. Koch, J., Steffen, C., Goeckler, J., Marteney, R., Jagels, J., and Brown, B. 2018. Comparison of diploid and triploid saugeye recruitment, growth, and condition in Kansas impoundments. North Am. J. Fish. Manage. 38(2): 446–453. doi:10.1002/nafm.10044. Lavoie, R.A., Jardine, T.D., Chumchal, M.M., Kidd, K.A., and Campbell, L.M. 2013. Biomagniﬁcation of mercury in aquatic food webs: a worldwide metaanalysis. Environ. Sci. Technol. 47(23): 13385–13394. doi:10.1021/es403103t. PMID:24151937. Leary, R.F., Allendorf, F.W., Knudsen, K.L., and Thorgaard, G.H. 1985. Heterozygosity and developmental stability in gynogenetic diploid and triploid rainbow trout. Heredity, 54(2): 219–225. doi:10.1038/hdy.1985.29. PMID:4008275. Lenth, R.V. 2020. emmeans: estimated marginal means, aka least-squares means. R package version 1.5.2-1. Available from https://CRAN.R-project. org/package=emmeans. Lepak, J.M., Hooten, M.B., and Johnson, B.M. 2012a. The inﬂuence of external subsidies on diet, growth and Hg concentrations of freshwater sport ﬁsh: implications for management and ﬁsh consumption advisories. Ecotoxicology, 21(7): 1878–1888. doi:10.1007/s10646-012-0921-4. PMID:22699411. Lepak, J.M., Kinzli, K.-D., Fetherman, E.R., Pate, W.M., Hansen, A.G., Gardunio, E.I., et al. 2012b. Manipulation of growth to reduce mercury concentrations in sport ﬁsh on a whole-system scale. Can. J. Fish. Aquat. Sci. 69(1): 122–135. doi:10.1139/f2011-136. Lucas, M.C. 1992. Spawning activity of male and female pike, Esox lucius L., determined by acoustic tracking. Can. J. Zool. 70(1): 191–196. doi:10.1139/ z92-029. Madenjian, C.P., Blanchﬁeld, P.J., Hrenchuk, L.E., and Van Walleghem, J.L.A. 2014. Mercury elimination rates for adult northern pike Esox lucius: evidence for a sex effect. Bull. Environ. Contam. Toxicol. 93(2): 144–148. doi:10.1007/ s00128-014-1256-z. PMID:24667854. Madenjian, C.P., Rediske, R.R., Krabbenhoft, D.P., Stapanian, M.A., Chernyak, S.M., and O’Keefe, J.P. 2016. Sex differences in contaminant concentrations of ﬁsh: a synthesis. Biol. Sex Differ. 7(1): 16. doi:10.1186/s13293-016-0069-7. PMID:26949510. Malison, J.A., Best, C.D., Kayes, T.B., Amundson, C.H., and Wentworth, B.C. 1985. Hormonal Growth Promotion and Evidence for a Size-Related Difference in Response to Estradiol-17b in Yellow Perch (Perca ﬂavescens). Can. J. Fish. Aquat. Sci. 42(10): 1627–1633. doi:10.1139/f85-203. Mauk, R.J., and Brown, M.L. 2001. Selenium and mercury concentrations in brood-stock walleye collected from three sites on Lake Oahe. Arch. Environ. Contam. Toxicol. 40(2): 257–263. doi:10.1007/s002440010170. PMID:11243328. Maxime, V. 2008. The physiology of triploid ﬁsh: current knowledge and comparisons with diploid ﬁsh. Fish Fish. 9(1): 67–78. doi:10.1111/j.14672979.2007.00269.x. McBride, R.S., Somarakis, S., Fitzhugh, G.R., Albert, A., Yaragina, N.A., Wuenschel, M.J., et al. 2015. Energy acquisition and allocation to egg production in relation to ﬁsh reproductive strategies. Fish Fish. 16(1): 23–57. doi:10.1111/faf.12043. Morel, F.M.M., Kraepiel, A.M.L., and Amyot, M. 1998. The chemical cycle and bioaccumulation of mercury. Annu. Rev. Ecol. Syst. 29(1): 543–566. doi:10. 1146/annurev.ecolsys.29.1.543. Morgan, G.E. 2002. Manual of instructions — fall walleye index netting (FWIN). Percid Community Synthesis, Diagnostics and Sampling Standards Working Group, Ontario Ministry of Natural Resources, Peterborough, Ont. Nicoletto, P.F., and Hendricks, A.C. 1988. Sexual differences in accumulation of mercury in four species of centrarchid ﬁshes. Can. J. Zool. 66(4): 944– 949. doi:10.1139/z88-140. Niimi, A.J. 1983. Biological and toxicologtcal effects of environmental contaminants in ﬁsh and their eggs. Can. J. Fish. Aquat. Sci. 40(3): 306–312. doi:10.1139/f83-045. Ogle, D.H. 2019. RFishBC. R package version 0.2.3. Available from https:// CRAN.R-project.org/package=RFishBC. Published by Canadian Science Publishing �Pagination not final (cite DOI) / Pagination provisoire (citer le DOI) Can. J. Fish. Aquat. Sci. Downloaded from cdnsciencepub.com by 67.174.106.68 on 11/15/21 For personal use only. Farrell et al. Ogle, D.H., Brenden, T.O., and McCormick, J.L. 2017. Growth estimation: growth models and statistical inference. In Age and growth of ﬁshes: principles and techniques. Edited by M.C. Quist and D.A. Isermann. American Fisheries Society, Bethesda, Md. pp. 265–359. Paranjape, A.R., and Hall, B.D. 2017. Recent advances in the study of mercury methylation in aquatic systems. FACETS, 2(1): 85–119. doi:10.1139/facets2016-0027. Parr Instrument Company. 2013. Introduction to bomb calorimetry. Available from https://www.parrinst.com/wp-content/uploads/downloads/2013/ 07/483M_Parr_Intro-to-Bomb-Calorimetry.pdf. Pate, W.M., Johnson, B.M., Lepak, J.M., and Brauch, D. 2014. Managing for coexistence of kokanee and trophy lake trout in a montane reservoir. N. Am. J. Fish Manage. 34(5): 908–922. doi:10.1080/02755947.2014.923072. Pethybridge, H., Cossa, D., and Butler, E.C.V. 2010. Mercury in 16 demersal sharks from southeast Australia: biotic and abiotic sources of variation and consumer health implications. Mar. Environ. Res. 69(1): 18–26. doi:10. 1016/j.marenvres.2009.07.006. PMID:19726079. Piferrer, F., Beaumont, A., Falguière, J.-C., Flajšhans, M., Haffray, P., and Colombo, L. 2009. Polyploid ﬁsh and shellﬁsh: production, biology and applications to aquaculture for performance improvement and genetic containment. Aquaculture, 293(3): 125–156. doi:10.1016/j.aquaculture.2009.04.036. Post, D.M. 2002. Using stable isotopes to estimate trophic position: models, methods, and assumptions. Ecology, 83: 703–718. doi:10.1890/0012-9658 (2002)083[0703:USITET]2.0.CO;2. Post, D.M., Layman, C.A., Arrington, D.A., Takimoto, G., Quattrochi, J., and Montaña, C.G. 2007. Getting to the fat of the matter: models, methods and assumptions for dealing with lipids in stable isotope analyses. Oecologia, 152(1): 179–189. doi:10.1007/s00442-006-0630-x. PMID:17225157. R Core Team. 2020. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available from https://www.R-project.org/. Scheuhammer, A.M., Meyer, M.W., Mark, B.S., and Michael, W.M. 2007. Effects of environmental methylmercury on the health of wild birds, mammals, and ﬁsh. Ambio, 36(1): 12–18. doi:10.1579/0044-7447(2007)36[12: EOEMOT]2.0.CO;2. PMID:17408187. Schirripa, M.J. 2002. An evaluation of back-calculation methodology using simulated otolith data. Fish. Bull. 100(4): 789–799. Selch, T.M., Hoagstrom, C.W., Weimer, E.J., Duehr, J.P., and Chipps, S.R. 2007. Inﬂuence of ﬂuctuating water levels on mercury concentrations in adult walleye. Bull. Environ. Contam. Toxicol. 79(1): 36–40. doi:10.1007/ s00128-007-9229-0. PMID:17618374. Selch, T.M., Chipps, S.R., Blackwell, B.G., and Hanten, R.P. 2019. Inﬂuence of season, sex, age and diet composition on mercury concentration in walleye Sander vitreus. Arch. Environ. Contam. Toxicol. 77(3): 336–343. doi:10. 1007/s00244-019-00658-1. PMID:31384978. Sharma, C.M., Borgstrøm, R., Huitfeldt, J.S., and Rosseland, B.O. 2008. Selective exploitation of large pike Esox lucius — Effects on mercury concentrations in ﬁsh populations. Sci. Total. Environ. 399(1): 33–40. doi:10.1016/j. scitotenv.2008.03.026. PMID:18485450. Shatunovskii, M.I., and Ruban, G.I. 2009. Ecological aspects of age-related dynamics of ﬁsh reproductive parameters. Russ. J. Ecol. 40(5): 320–327. doi:10.1134/S1067413609050038. Simoneau, M., Lucotte, M., Garceau, S., and Laliberté, D. 2005. Fish growth rates modulate mercury concentrations in walleye (Sander vitreus) from eastern Canadian lakes. Environ. Res. 98(1): 73–82. doi:10.1016/j.envres. 2004.08.002. PMID:15721886. Skinner, M.M., Martin, A.A., and Moore, B.C. 2016. Is lipid correction necessary in the stable isotope analysis of ﬁsh tissues? Rapid Commun. Mass Spectrom. 30(7): 881–889. doi:10.1002/rcm.7480. PMID:26969930. Son, K.-T., Kwon, J.-Y., Jo, M.-R., Yoon, M., Song, K.-C., Choi, W.-S., et al. 2014. Total mercury contents of Antarctic toothﬁsh Dissostichus mawsoni caught in the Antarctic Sea. Fish. Aquat. Sci. 17: 427–431. doi:10.5657/ FAS.2014.0427. Sorensen, J.A., Kallemeyn, L.W., and Sydor, M. 2005. Relationship between mercury accumulation in young-of-the-year yellow perch and water-level ﬂuctuations. Environ. Sci. Technol. 39(23): 9237–9243. doi:10.1021/es050471r. PMID:16382948. 13 Stein, R.A., and Murphy, M.L. 1976. Changes in proximate composition of the crayﬁsh Orconectes propinquus with size, sex, and life stage. J. Fish. Res. Bd. Can. 33(11): 2450–2458. doi:10.1139/f76-292. Stewart, D.J., Weininger, D., Rottiers, D.V., and Edsall, T.A. 1983. An energetics model for lake trout, Salvelinus namaycush: application to the Lake Michigan population. Can. J. Fish. Aquat. Sci. 40(6): 681–698. doi:10.1139/f83-091. Symonds, M.R., and Moussalli, A. 2011. A brief guide to model selection, multimodel inference and model averaging in behavioural ecology using Akaike’s information criterion. Behav. Ecol. Sociobiol. 65(1): 13–21. doi:10. 1007/s00265-010-1037-6. Taylor, M.S., Driscoll, C.T., Lepak, J.M., Josephson, D.C., Jirka, K.J., and Kraft, C.E. 2020. Temporal trends in ﬁsh mercury concentrations in an Adirondack lake managed with a continual predator removal program. Ecotoxicology, 29(10): 1762–1773. doi:10.1007/s10646-019-02156-5. PMID:31925620. Teuscher, D.M., Schill, D.J., Megargle, D.J., and Dillon, J.C. 2003. Relative Survival and Growth of Triploid and Diploid Rainbow Trout in Two Idaho Reservoirs. N. Am. J. Fish Manage. 23(3): 983–988. doi:10.1577/M02-026. Thorgaard, G.H. 1983. Chromosome set manipulation and sex control in ﬁsh. In Fish physiology. Edited by W.S. Hoar, D.J. Randall, and E.M. Donaldson. Academic Press, New York. pp. 405–434. Tiwary, B., Kirubagaran, R., and Ray, A. 2000. Gonadal development in triploid Heteropneustes fossilis. J. Fish Biol. 57(5): 1343–1348. doi:10.1111/j.10958649.2000.tb00493.x. Tiwary, B.K., Kirubagaran, R., and Ray, A.K. 2004. The biology of triploid ﬁsh. Rev. Fish Biol. Fish. 14(4): 391–402. doi:10.1007/s11160-004-8361-8. Trudel, M., and Rasmussen, J.B. 2006. Bioenergetics and mercury dynamics in ﬁsh: a modelling perspective. Can. J. Fish. Aquat. Sci. 63(8): 1890–1902. doi:10.1139/f06-081. Trudel, M., Tremblay, A., Schetagne, R., and Rasmussen, J.B. 2000. Estimating food consumption rates of ﬁsh using a mercury mass balance model. Can. J. Fish. Aquat. Sci. 57(2): 414–428. doi:10.1139/f99-262.0. US Census Bureau. 2020. Cartographic boudry ﬁles by geography: states. Available from https://www.census.gov/geographies/mapping-ﬁles/time-series/ geo/cartographic-boundary.html [accessed 02 October 2020]. US Geological Survey. 2012. National Elevation Dataset – NAVD88 Meters – 1/ 3rd-Arc-Second (Approx. 10m). Available from https://catalog.data.gov/dataset/ usgs-national-elevation-dataset-ned [accessed 02 October 2020]. US Geological Survey. 2017. GNIS for states, territories, associated areas of the United States. Available from https://apps.nationalmap.gov/downloader/#/ [accessed 02 October 2020]. US Geological Survey. 2020a. NHD 20200615 for Colorado state or territory shapeﬁle model version 2.2.1. Available from https://apps.nationalmap.gov/ downloader/#/ [accessed 02 October 2020]. US Geological Survey. 2020b. NHD 20200615 for New Mexico state or territory shapeﬁle model version 2.2.1. Available from https://apps.nationalmap. gov/downloader/#/ [accessed 02 October 2020]. US Geological Survey. 2020c. NHD 20200615 for Utah state or territory shapeﬁle model version 2.2.1. Available from https://apps.nationalmap. gov/downloader/#/ [accessed 02 October 2020]. US Geological Survey. 2020d. NHD 20200615 for Wyoming state or territory shapeﬁle model version 2.2.1. Available from https://apps.nationalmap. gov/downloader/#/ [accessed 02 October 2020]. Wentz, D.A., Brigham, M.E., Chasar, L.C., Lutz, M.A., and Krabbenhoft, D.P. 2014. Mercury in the nation’s streams — levels, trends, and implications. Circular 1395, US Geological Survey, Reston, Va. Wickham, H. 2016. ggplot2: elegant graphics for data analysis. Springer-Verlag, New York. Wiener, J.G., and Spry, D.J. 1996. Toxicological signiﬁcance of mercury in freshwater ﬁsh. In Environmental contaminants in wildlife — interpreting tissue concentrations. Edited by W.N. Beyer, G.H. Heinz, and A.W. Redmon-Norwood. Lewis Publishers, Boca Raton, Fla. pp. 297–339. Wolff, B.A., Johnson, B.M., and Lepak, J.M. 2017. Changes in sport ﬁsh mercury concentrations from food web shifts suggest partial decoupling from atmospheric deposition in two Colorado reservoirs. Arch. Environ. Contam. Toxicol. 72(2): 167–177. doi:10.1007/s00244-016-0353-x. PMID:28064370. Yao, S., and Drouillard, K.G. 2019. Prediction of mercury elimination rate coefﬁcients of ﬁsh is improved by incorporating ﬁsh temperature classiﬁcation into models. Bull. Environ. Contam. Toxicol. 103(5): 657–662. doi:10.1007/ s00128-019-02709-8. PMID:31492971. Published by Canadian Science Publishing � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Journal Articles Description An account of the resource CPW peer-reviewed journal publications Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Induced triploidy reduces mercury bioaccumulation in a piscivorous fish Description An account of the resource <div>We compared mercury bioaccumulation in triploid and diploid walleye (<i>Sander vitreus</i>) in Narraguinnep Reservoir, Colorado, USA, and made several hypotheses that sex- and ploidy-specific differences in the allocation of energy towards reproductive development would affect mercury bioaccumulation. We tested our hypotheses with linear regression and a bioenergetics model informed by field data. We found diploid walleye had 28%–31% higher mercury concentrations on average than triploids, but there were no differences between sexes of the same ploidy. Triploids of mature age exhibited minimal gonadal development when compared to diploids. After accounting for reproductive investment, the bioenergetics model accounted for most of the observed difference in average mercury concentration between ploidies for females. Conversely, the energetic cost of producing testes was low, and gonadal development could not explain observed patterns for males. Costs associated with elevated swimming activity and metabolism by diploid males relative to other groups could explain the difference but requires further investigation. The use of triploid fish in stocking programs could prove useful for reducing mercury in fish destined for human consumption.</div> <div><br />Nous comparons la bioaccumulation de mercure dans des dorés jaunes (<i>Sander vitreus</i>) triploïdes et diploïdes dans le réservoir Narraguinnep (Colorado, États-Unis) et formulons plusieurs hypothèses à l’effet que des différences selon le sexe et la ploïdie sur le plan de l’affectation de ressources énergétiques au développement des organes reproducteurs auraient une incidence sur la bioaccumulation de mercure. Nous validons ces hypothèses à l’aide de la régression linéaire et d’un modèle bioénergétique alimenté de données de terrain. Nous constatons que les dorés diploïdes présentent des concentrations de mercure de 28 % à 31 % plus importantes, en moyenne, que les triploïdes, mais aucune différence n’est relevée entre les sexes de même ploïdie. Les triploïdes d’âge mature présentent un développement gonadique minimal comparativement aux diploïdes. Une fois pris en compte l’investissement dans le système reproducteur, le modèle bioénergétique explique la majeure partie des différences observées des concentrations de mercure moyennes entre femelles de ploïdies différentes. À l’inverse, le coût énergétique de la production de testicules est faible, et le développement des gonades ne peut expliquer les motifs observés chez les mâles. Les coûts associés à une activité de nage et un métabolisme plus élevés chez les mâles diploïdes par rapport aux autres groupes pourraient expliquer la différence, mais cela nécessite un examen plus approfondi. L’utilisation de poissons triploïdes dans les programmes d’empoissonnement pourrait s’avérer utile pour réduire les concentrations de mercure dans les poissons destinés à la consommation humaine. [Traduit par la Rédaction]</div> Bibliographic Citation A bibliographic reference for the resource. Recommended practice is to include sufficient bibliographic detail to identify the resource as unambiguously as possible. <a href="https://doi.org/10.1139/cjfas-2021-0037" target="_blank" rel="noreferrer noopener">https://doi.org/10.1139/cjfas-2021-0037</a> Creator An entity primarily responsible for making the resource Farrell, Collin J. Johnson, Brett M. Hansen, Adam G. Myrick, Christopher A. Subject The topic of the resource Mercury bioaccumulation Triploid walleye Diploid walleye <em>Sander vitreus</em> Narraguinnep Reservoir, Colorado Extent The size or duration of the resource. 13 pages Date Created Date of creation of the resource. 2021-06-17 Rights Information about rights held in and over the resource <a href="http://rightsstatements.org/vocab/InC-NC/1.0/" target="_blank" rel="noreferrer noopener">In Copyright - Non-Commercial Use Permitted</a> Format The file format, physical medium, or dimensions of the resource application/pdf Language A language of the resource English; French (abstract only) Is Part Of A related resource in which the described resource is physically or logically included. Canadian Journal of Fisheries and Aquatic Sciences Type The nature or genre of the resource Article