Evaluation of bioelectrical impedance analysis as an estimator of moose body composition

Estimation of body composition of wild ungulates yields important information regarding nutritional status of individuals and populations; yet, there are few suitable field techniques that are nondestructive, unbiased, precise, and quick to perform. We tested the suitability of bioelectrical impedance analysis (BIA) as an estimator of body composition of moose (Alces alces) for use in the field. A derived BIA variable, impedance volume, was a significant predictor of body fat (mass and percentage) and body water (mass and percentage) when sex was added to models as an indicator variable but explained only 48-57% of variation in composition. Best predictive models included impedance volume, sex, body mass, and a body mass x sex interaction. Due to difficulty measuring body mass of moose in the field, we also generated predictive models when body mass was replaced with a proxy (lengthxgirth(2)). Predictive equations for body water were more precise than were those for body fat. Impedance estimates decreased as the subject's hind leg was straightened, indicating that animal positioning must be standardized to minimize bias. Lack of precision made BIA unsuitable for estimating moose body fat in the field. BIA was a precise and quick estimator of body water in moose, but its limitations make it more suitable for the laboratory than the field.