Understanding Maximum Size in the Catch: Atlantic Blue Marlin as an Example

Abstract

AbstractMean and maximum sizes in the catch are readily understood by both fishermen and scientists as important indicators of population health, but the maximum is more complicated to interpret because it increases with the number of observations. Size distributions of catches were computed for stable age distributions of hypothetical Atlantic populations of Blue Marlin Makaira nigricans. Fishing mortality rates were selected that reduced the population per age‐1 recruit by 25, 50, 67, 75, and 90% by number from the unfished state. A new metric, NZ50, was applied to evaluate the performance of maximum size as an indicator of population status. Sample sizes required to include large marlin in a set of observations increased to disproportionately large numbers with increased fishing. With the best estimates of growth and a natural mortality of 0.1, the number of random observations required for a 50% probability of including a 350‐cm individual increased by about 43% when the population was depressed by 25%. This value rose to more than 50‐fold when the population was reduced below 10%. In contrast, mean lengths were reduced by only 1.6% to 14% and mean weights by 5% to 45% for the same range of fishing. These results provide quantitative evidence confirming the view that a diminished number of large fish is a sensitive indicator of excessive fishing, even when mean size of the catch shows only a small change. Increased catches above large size thresholds or nearing historical maxima could be sensitive indicators of stock recovery. Measures of the frequency of occurrence of individuals above some threshold defined for large fish will generally be superior to the maximuma observed in a sample (or set of samples) as a biological reference criterion because of the stochastic nature of individual observations.Received July 31, 2014; accepted October 31, 2014