Estimating food consumption rates of fish using a mercury mass balance model

Abstract

We present a simple method for estimating food consumption rates of fish in the field based on a mercury (Hg) mass balance model. This method requires the determination of fish age, size, and growth and Hg concentration in fish and their food. The model was validated using data obtained from a previously published laboratory experiment. A field test of the model showed that food consumption rates determined with the Hg mass balance model differed from independent estimates obtained with the 137 Cs method by only 0.6-16.1%. The model was applied to fish from various lakes in Quebec and Ontario. Food consumption rates estimated with the Hg mass balance model varied significantly both among species and among populations. Furthermore, female fish tended to eat 30-40% more food than males, probably to meet the larger energy requirement associated with egg production. A sensitivity analysis indicated that the Hg mass balance model was mostly responsive to variables that can be easily measured in the field, such as fish size and Hg concentration in fish and their food. By providing a low-effort approach to quantifying food consumption rates of fish in the field, this method may help to refine our understanding of the environmental factors that influence the quantity of food consumed by fish. 428 Résumé : Nous présentons une méthode simple dérivée d'un bilan massique du mercure (Hg) pour estimer les taux de consommation des poissons en milieu naturel. Cette méthode nécessite une estimation de l'âge, la masse et la croissance des poissons, ainsi que de la teneur en Hg des poissons et de leurs proies et a été validée à partir d'expériences réalisées en laboratoire. Une évaluation effectuée in situ a démontré que les taux de consommation obtenus à partir de ce bilan massique ne différaient que par 0,6-16,1% des valeurs obtenues de façon indépendante à l'aide du 137 Cs. Ce modèle a été utilisé pour estimer les taux de consommation de poissons provenants de divers lacs du Québec et de l'Ontario. Ces estimations différaient significativement entre les espèces et les populations. De plus, les femelles consommaient 30-40% plus de nourriture que les mâles, probablement pour combler les coûts de production de gonades plus élevés. Une analyse de sensibilité a indiqué que le bilan massique du Hg était principalement influencé par les variables qui peuvent être facilement mesurées en milieu naturel, telles la masse des poissons et la teneur en Hg des poissons et des proies. Comme cette méthode requiert peu d'efforts en milieu naturel, elle pourrait nous aider à mieux comprendre les facteurs environnementaux qui influencent la quantité de nourriture consommée par les poissons. Trudel et al.

Prédation planctonique des deux principaux poissons fourrages pélagiques du Lac Saint-Jean :

“…The concentration of Hg in fish (C, ng-g 1 ) can be expressed by the Hg mass balance model as (Thomann 1981, Trudel et al 2000:…”

“…Sensitivity analysis was performed on each model parameter according to Trudel et al (2000) using a ± 20% variation.…”

“…Sensitivity analysis revealed that the model was sensitive to four parameters. Trudel et al (2000) showed that the most sensitive parameters for their model were the weight of the fish, the MeHg concentration in fish and in the prey and the assimilation efficiency. The same parameters were the most sensitive to a change of value in the present study on rainbow smelt.…”

“…Moreover, the evacuation rate-based method give the ingestion rate for only one day, and this estimate cannot be used for generalization if food ingestion varies between days (Forseth et al 1992). Other field-based methods are the mass balance contaminant models that provide food ingestion rates almost identical to those derived from stomach contents (Forseth et al 1992), with considerably less sampling effort (Trudel et al 2000). Those models do not estimate food ingestion rate during a single day but for a time interval that may vary from days to months.…”

“…Those models do not estimate food ingestion rate during a single day but for a time interval that may vary from days to months. Contaminants such as polychlorinated biphenyls (PCB), radioactive cesium or mercury (Hg) have been used to quantify consumption rate of adult fish (Borgmann and Whittle 1992, Forseth et al 1992, Tucker and Rasmussen 1999, Trudel et al 2000 but never during early life stages.…”