Meeting the objectives of fisheries observer programs through electronic monitoring
Data from fisheries observer programs support fundamental research and management applications, ranging from conducting stock assessments to assessing the performance of ecosystem-based harvest strategies. Electronic monitoring (EM) systems are increasingly being used to augment conventional coverage by human onboard observers, as well as to provide at-sea coverage where none previously existed. Here we review findings from studies that compared the precision and accuracy of data fields collected by EM and onboard observers. EM data have relatively high precision to data collected by onboard observers, but with some areas identified where EM systems are in need of improvement, such as detection of some species of non-retained catch, life status of the catch and whether crew used specific fishing methods and equipment. We describe how data from EM systems, unlike at-sea observer program data, are not susceptible to intimidation, corruption, and observer placement that is not randomized and balanced (sampled proportionately across ports and vessel categories), so that fleet-wide extrapolations accurately characterize the entire fleet, and avoid statistical sampling bias. We discuss how a fisheries monitoring programs designed so that all vessels in a fleet are outfitted with onboard EM systems and where random and balanced samples of the raw data are analyzed would eliminate an ‘observer effect’, where the alternative of 100% human observer coverage to eliminate this source of bias is likely more costly. Fields collected by at-sea human observers of regional observer programs of the western and central Pacific Ocean for pelagic longline and tuna purse seine fisheries were analyzed, identifying 101 fields that are not possible to collect with contemporary EM systems. Of these, 45 fields could be collected by dockside monitoring, however, 7 of these fields might not remain static during the course of a trip, in particular for distant-water vessels. Innovations in EM technology could be feasible to collect 70 of the fields. Many of the identified changes to contemporary EM technology are likely feasible through simple changes, such as positioning existing or additional EM cameras to ensure that certain fishing activities are within the field of view. Others, however, are highly uncertain, require research and development, such as a probe that analyzes sex-specific genetic markers or hormones to determine the sex of the catch, and require trials to determine their accuracy, such as the use of thermal cameras to determine life status, and the use of underwater cameras and EM digital length software to estimate the dimensions of submerged gear. Tissue samples and information on gonad stage of the catch, which are not possible to collect by dockside inspections or EM, could potentially be collected by fishers. Having observer coverage of vessels receiving at-sea transshipped catch would be necessary to collect information on the receiving vessel. It may only be possible for human onboard observers to collect information on the fishers of distant-water vessels that exchange their captain and crew multiple times during a trip. Achieving the political will and capacity to transition to EM may be strengthened given the support of catch sector and other seafood companies. Examples are provided of industry-desired information that could be delivered by EM systems. Improvements in EM technology promise to enable EM systems to collect most fields collected by contemporary human onboard observer programs, to improve data quality for some fields, and to broaden data collection fields to meet the expanding data requirements of fisheries monitoring programs as management authorities continue to transition to implementing elements of ecosystem-based fisheries management.