Research Document 1998/147

The analysis of 1997 small salmon returns to Newfoundland rivers using distributional testing

By M. Power

Abstract

Data available of small salmon returns to Newfoundland rivers were used to determine the appropriate statistical distribution for describing those same returns. Only those data series which were judged to be statistically adequate random samples were used in distributional testing procedures. Anderson-Darling tests were used to test the acceptability of the Normal, Lognormal and Weibull distributional models as descriptions of small salmon returns. P-values, allowing selection between the candidate models were also computed and reported. Distributional models selected as the best descriptors of the returns data were then parameterized using available historical data and used to compute the probabilities of observing returns as low as those reported for 1997. Computed probabilities indicate some rivers experienced abnormally low returns in 1997 (N.E. Trepassey and Salmon Brook), while others did not (Rocky and Exploits). Although data were not adjusted for losses to the commercial fishery priory to 1992, conclusions reached here apply to observations about trends in the actual size of potential small salmon spawning stocks. On balance there appears to be little evidence of a large scale synchronous decline in small salmon returns to Newfoundland rivers.

Complete document (PDF, 797 KB)

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.