Size- and age-dependent natural mortality in fish populations: Biology, models, implications, and a generalized length-inverse mortality paradigm

Natural mortality rates (M) in fish populations vary with body size and age, often by orders of magnitude over the life cycle. Traditionally, fisheries models and stock assessment methods have treated M as constant in the recruited stock, but that axiom has been challenged on empirical and theoretical grounds, and by practical assessment needs. Reviewing biological considerations, empirical generalizations, and theoretical models of size- and age-dependent natural mortality in fish populations, I show how multiple strands of evidence lead to a coherent new M paradigm best described as ‘generalized length-inverse mortality’ (GLIM). GLIM holds that mortality declines inversely with body length throughout much of the juvenile and adult phases of the fish lifecycle. Deviations from the length-inverse pattern may occur in older ages due to senescence and in early juveniles due to density-dependence. GLIM is strongly supported by empirical meta-analyses of mortality-size relationships and is also broadly consistent with multi-species and ecosystem models of predation mortality. Whether operationalized in closed functional form or through multi-species modeling of predation and residual mortality, GLIM provides a new ‘standard M′ for fish population modeling and stock assessment applications. Consequences of mis-specifying size- and age-dependent M in stock assessment applications vary from moderate in many cases to severe under certain conditions, but even moderate consequences can be quantitatively significant in stock assessment and management. Further research is indicated with regards to senescence and to the representation of residual or non-predation mortality (M1) in multi-species and ecosystem models.