Inter-annual variability in the supply of young fish to the fishery remains a significant problem, if not an impossible problem. Hjort’s (1914) foundational hypothesis concerning a critical period for larval feeding stimulated various bio-physical hypotheses, some of which include the interaction of growth (G) and mortality (M) during the larval stage. Both M and G require estimates of age, arduously obtained from daily growth increments, at fine spatial and temporal scales. To compound the problem, predation tends to remove the starving and the slow growers, highlighting the interdependence of growth and mortality, which is sometimes described as the “single process”.
Building on the theory of size structured populations and the size-frequency distribution of zooplankton, we show that the descending slope of the biomass size-frequency distribution of a particular species of larval fish is the ratio of M to G (known as the recruitment potential). Numerical simulations show that a size-based estimate of M/G performs as well as various traditional (and newer) age-based methods, and with smaller sample sizes (>50 larvae). Additional metrics to the size spectrum slope, could include the abundance or elevation of the spectrum, and the geometric mean size. Our method can eminently be applied to many long term surveys of commercially important larval fish, where larval lengths are routinely collected, where otoliths have degraded, and where recruitment outcomes are known. We report on some new methods to convert length to larval biomass, consistent with optical technologies to potentially quantify larval fish in situ. Our method has relevance for assessing oceanographic features, by explicitly and routinely incorporating the competing effects of growth and mortality for the first time. Our new view of the recruitment problem will re-invigorate fisheries oceanography, particularly with regard to the ephemeral frontal eddies of boundary currents, adjacent to the substantial coastal fisheries.