Short-term mortality in post-emergent larval capelin Mallotus villosus. I. Analysis of multiple in situ estimates

Abstract

Forty short-term mortality estimates of post-emergent larval capelin Mallotus villosus in a small embayment in eastern Newfoundland during 1981 to 1983 ranged from 3 to 100% d-l, and averaged 60% d-' Mortality estimates derived under 5 sets of different assumptions regarding immigration and emigration showed that failure to account for variation in immigration rates (production of newly-emerged larvae) greatly biased the mortality estimates. Larval immigration alone explained between 32 and 66 % of the apparent mortality. The contagious distribution of pre-emergent larvae was a major contributor to imprecise estimates of immigration. The effect of advective transport on population dispersion was consistent within and between years and dominated over diffusion effects. Ignoring variations in advective losses resulted in artificially elevated mortality estimates. Obtaining accurate estimates of larval mortality in the small system we studied was more than lust a trivial problem despite the many simplifying advantages. Obtaining accurate estimates of larval mortality in larger ocean systems with more dispersed populations is doubtful.