Growth Models with Corrections for the Retardative Effects of Tagging

Abstract

Length increment data from mark–recapture experiments are commonly used to obtain information on animal growth, assuming that tagging does not affect the growth of marked animals. The assumption is violated in many studies, but the effects of tagging on growth and estimates of growth parameters have not been and cannot be examined without appropriate models. This paper describes a model allowing quantification and estimation of the retarding effects of tagging on animal growth simultaneously with growth parameters in all existing growth models, reduction or elimination of biases in growth parameters induced by tagging, and relaxation of a key assumption in growth analysis using length increment data. A special case of this model was applied to simulated data and to tagging data from a centropomid perch (Lates calcarifer) to demonstrate its general utility. Tagging was inferred to have stopped the fish growth for 36.44 d (ASE = 12.70 d) if von Bertalanffy growth is assumed, but the period of recovery from tagging seemed size or age independent within the size range studied. If tagging retards animal growth, L∞ is slightly overestimated and K underestimated for unbiased data. Potential applications and limitations of the model are also discussed.