Sampling Designs for the Estimation of Longline Bycatch

Sampling Designs for the Estimation of Longline Bycatch

Muktha M. Menon, Vincent F. Gallucci, and Loveday L. Conquest

Sampling Designs for the Estimation of Longline BycatchThis is part of Fisheries Assessment and Management in Data-Limited Situations
To download the free PDF [472.9 KB], please enter:


Unbiased estimates of longline bycatch on data-limited species are critical, especially when bycatch data may be the main source of information on such species. Estimates might be affected both by the spatial distribution of animals along the longline as well as by the sampling design used. We compared bycatch estimates from four sampling designs over six spatial distributions of bycatch using simulated data. The four sampling designs were simple random sampling, systematic sampling, partial sampling, and systematic sampling of longline sections. Spatial distributions included random distribution of bycatch and distribution of bycatch with periodicity. The methods were also used to estimate shark bycatch from a subset of the International Pacific Halibut Commission's 2002 Pacific Halibut Stock Assessment Survey data. Both spatial pattern and sampling design were found to affect the estimates of bycatch from longlines. All sampling methods gave unbiased estimates when bycatch was random, aggregated, or rare. The estimates of longline bycatch were most affected by the presence of periodicity in bycatch distribution along the line. Systematic sampling and partial sampling provided unbiased and the most precise bycatch estimates in most cases, and thus are the preferred sampling designs. Of these two, partial sampling is logistically the easiest in the field.

Item details