Retrospective and geographical overview of the interaction between seabirds and the Spanish surface longline fishery targeting Swordfish in the Indian Ocean during the 1993-2017 period inferred from data provided by scientific observers at sea.

A total of 5.8 million hooks were scientifically observed at sea for seabird interactions in broad areas of the Indian Ocean between 1993 and 2017. Two types of information were obtained in the surface longline fleet targeting swordfish: (a) during regular commercial fishing and (b) during experimental surveys. 59.54% and 40.46% of the total effort was observed in each case, respectively. The geographical coverage of the study goes beyond the areas in which this commercial fleet has historically fished, because the information includes experimental surveys. Twenty years of regular commercial fishing data are included in the analysis and in thirteen of these years the interactions occurring were nil. Positive interaction occurred in twenty four of the one hundred and twelve 5ºx5º areas observed during regular commercial fishing and experimental surveys combined.
A total of 165 seabird interactions were recorded for both types of information in all areas and years combined, 77.58% of which occurred during experimental surveys and 22.42% during regular commercial fishing. 49.7% of the total interactions observed occurred in a single boat-survey in 2005-2006 in areas further east than 85ºE, where other boats involved in the same experimental survey at the same time recorded nil or very few interactions. No interaction occurred North of 20ºS and only a few sporadic interactions North of 30ºS. 70.3% of interactions occurred in areas well to the south and east, between 30º-45ºS/85º-115ºE, most of then related to experimental surveys in areas where commercial fishing by this fleet did not regularly take place. Procellariiformes was the order of seabird observed in encounters. Observations from commercial fishing alone generally suggest nil or low rates of interaction in most areas observed, while observations from surveys done in southern and eastern areas suggest relatively higher rates of interaction. The data provide a useful geographical overview of different areas of the Indian Ocean and suggest a possible latitudinal-longitudinal gradient related to higher interaction rates the further to the south and east fishing activity occurs, an aspect that should be verified by data from other fleets regularly fishing in those areas. The authors also include some comments on caution which should be exercised by readers when interpreting these results.