On improving operation and use of the elefan programs. part 3, correcting length-frequency data for the effects of gear selection and/or incomplete recruitment

Abstract

One important feature of length€”frequency data that is usually ignored when data are collected for subsequent estimation of growth and various related parameter estimates is that length€”frequency data are necessarily collected with gears that are always selective for certain sizes. This feature often results in length€”frequency samples that cannot be used to extract growth and mortality parameter estimates, for example in the case of data sampled by means of a gillnet. On the other hand, it is usually possible to extract growth and mortality parameter estimates from length€”frequency data obtained using trawls, Danish seines and similar gears. These estimates will usually be biased, however, because of some residual selection and other effects. A method is presented here which allows for a partial correction of such biases, as related to the estimation of growth parameters (L. and K) using the ELEFAN I program.