Using Frequency Distributions of Catch per Unit Effort to Measure Fish-Stock Abundance

Abstract

Mean catch per unit effort often is used as an index of fish abundance, N. This index is often biased. Frequency distributions of catch per unit effort (C/f) and independent estimates of N from a headboat sport fishery for yellowtail snappers Ocyurus chrysurus off Islamorada, Florida, indicated that mean C/f underestimated changes in N. Frequency distributions of C/f were skewed and fit the negative binomial distribution. Some descriptors of the distributions (median, variance, negative binomial parameter k, and frequency of zero C/f) were systematically related to N. These relationships were used to derive an index of N less biased than mean C/f. The new index, square root of relative frequency of zero C/f, could be applicable to any recreational, scientific, or commercial fishing activity for which C/f distributions are available, skewed, and have fairly high frequencies of zero C/f. Skewness could be due to spatial dispersion of the fish population or to a gradient of skill among fishing units. Received September 28, 1982 Accepted June 3, 1983