Diurnal variation in bottom trawl survey catches: does it pay to adjust?

Abstract

We have studied the diurnal bias of bottom trawl survey catches with the purpose of adjusting for it and thereby improving the accuracy of abundance estimates. The correction term is estimated with uncertainty, and thus increases the variance of the resulting abundance estimate. To investigate this adequately we use a stochastic model describing diurnal fluctuations, and we examine the annual variation of the diurnal amplitude as a function of species and length. The diurnal amplitude is fairly stable for large fish, and for these the bias corrected estimate leads to a moderate increase in variance. For small fish the diurnal amplitude is unstable, however, and the correction of diurnal bias occurs at the expense of a large increase in variance. This unstable amplitude also leads to a large year-to-year variation in catchability for small fish. For haddock the diurnal amplitude depends heavily on fish length, indicating a strong decrease in catchability with decreasing fish length.