Comparison of Fishers' Logbooks to Observer Data in the Gulf of Mexico and U.S. South Atlantic Bottom Longline Fishery

Commercial fisheries logbooks, which contain self-reported data from fishers, are sources of catch and effort data that are used in stock assessments and to estimate bycatch in fisheries. Although the official landings are compiled from dealer-supplied reports, the fishers themselves record trip-specific at-sea catch and effort information on vessel logbooks. Another source of at-sea commercial fishery data is supplied by observers that document all animals caught, whether kept, released alive, or discarded dead. Observers are placed on a small fraction of fishing vessels as time and funding permit. Therefore, a small proportion of the commercial fishing data can be compared to that reported by the fisher.  The objective of this study is to compare at-sea observer data to the fishers’ logbooks when those records overlap. We have such data available from 2005 to 2010 covering 363 trips with 1565 hauls over 1232 sea days from observers and the corresponding reports from the logbooks (88.8% matched). By fitting a generalized additive model (GAM) to the data, sources of bias in reporting catch and bycatch, including protected species interactions, can be characterized. Spatio-temporal, environmental, and operational variables provided in the observer data are tested to determine if the logbook catch rates and the observer catch rates are comparable. The GAM results may be used to provide predictors for logbook records to serve in the place of observers on the large fraction of bottom longline trips that do not carry an observer (approximately 95%).  Finally, this study sheds light on the debate of whether there is a true under-reporting bias in logbook data and provides correction for it when the data are used.