Estimating catches with automatic identification system (AIS) data: a case study of single otter trawl in Zhoushan fishing ground, China



The sailing tracks of single otter trawl vessels were simulated using cubic hermite spline (cHs) interpolation method based on the automatic identification system (AIS) data of 6 sampled vessels that were fishing in the Zhoushan fishing grounds after the close of the fishing season from September 2012 to January 2013.  The vessels’ status (i.e. whether the vessels were fishing or not) were determined based on the integrated information of speed method and the simulated tracks.  Generalized additive model (GAM) was built according to the logbook data of the 6 vessels in 2010, and then the GAM was applied to the AIS data of the same 6 vessels from September 2012 to January 2013 to estimate the monthly catches of these vessels in this period.  The results show that the error of the simulated tracks increase with the increase in time interval, and the time interval of AIS data should be shorter than 30 minutes to prevent low accurate results. GAM can give viable estimates of catches when they do not greatly fluctuate over years.  The step-by-step GAM analyses indicate that the factors, which affect catch, are ordered by their importance as date, sea surface salinity (SSS), latitude, sea surface temperature (SST), sea surface height (SSH) and longitude.  This research is a new attempt for the study of fisheries resources in China using new data sources, which will be helpful for the improvement of fishery research in such data-poor countries as China.