Estimation of the capture of New Zealand sea lions (Phocarctos hookeri) in trawl fisheries, from 1995–96 to 2006–07

In this report, the number of New Zealand sea lion (Phocarctos hookeri) captures in New Zealand’s trawl fisheries are estimated for the 1995—96 to 2006—07 fishing years. Over this period, Ministry of Fisheries observers recorded the capture of between 5 and 39 sea lions within the New Zealand Exclusive Economic Zone, within each fishing year. During the 2006—07 fishing year, 15 sea lions were observed killed on trawls. For the purposes of this report, three captures were included in this total that were observed on the first day of the 2007—08 fishing year, in the last days of the southern blue whiting fishery. All of the captured sea lions were retrieved dead. Of the 15 sea lion captures, 7 were observed caught in the squid fishery around the Auckland Islands. This was the lowest number of observed captures in this fishery since the 1998—99 fishing year. Six sea lion captures were observed in the southern blue whiting fishery east of Campbell Island, continuing a trend of increasing captures in that fishery since 2002. Two other captures occurred, one in the scampi fishery near the Auckland Islands, and one in the squid fishery south of The Snares.

From these observations, estimates of total captures were made for four different trawl fisheries: the squid fishery near the Auckland Islands; the southern blue whiting fishery east of Campbell Island; other (non-squid) fisheries near the Auckland Islands; and all trawl fisheries on the southern end of the Stewart- Snares shelf. Bayesian generalised linear models were fitted to data from the first two of these fisheries. A previous model, used for estimating sea lion captures in the 2004—05 fishing year, was re-implemented for the Auckland Islands squid fishery. A new model was developed for the southern blue whiting fishery. Ratio estimates were calculated for the other (non-squid) Auckland Islands trawl fishery, and for all trawl fisheries on the southern end of the Stewart-Snares shelf.

Over the 12 years of data, fishing effort in the squid fishery near the Auckland Islands was highest in the 1995—96 fishing year, with 4467 trawls being made. Effort then decreased to fewer than 2000 trawls in each year between 1997—98 and 2002—03. In each year between 2003—04 and 2005—06, over 2400 trawls were made. The trawl effort fell to 1320 trawls in 2006—07.

Since 2001, squid trawl nets in the Auckland Islands squid fishery have increasingly been fitted with sea lion exclusion devices (SLEDs) that allow animals to escape from the net. The estimated interactions (captures and escapes via SLEDs) follow the patterns in effort. In 1995—96 and 1996—97 there were an estimated 141 and 140 interactions, respectively. In each year between 1997—98 and 2002—03 there were fewer than 75 estimated interactions. For the three years 2003—04 to 2005—06 there were 140 or more interactions. In 2006—07, the model estimated that there were 74 (95% c.i.: 32 to 136) sea lion interactions.

The model estimated that the probability that a sea lion failed to escape from a net fitted with a SLED (i.e., the retention probability) was 0.24 (95% c.i,: 0.13 to 0.39). This was similar to previous estimates, and correspondingly the predicted strike rate in 2006—07 of 5.6 sea lions per 100 trawls (95% c.i.: 2.7 to 10) was similar to estimates of the strike rate made previously. As SLEDs have been used more widely, the number of captures has fallen relative to the number of interactions, and in 2006—07 there were an estimated 20 (95% c.i.: 11 to 33) sea lion captures in the Auckland Islands squid fishery. The number of attributed mortalities in this fishery in 2006—07 was 60 (95% c.i.: 26 to 110), assuming a discount rate of 20%, which is less than the limit of 93 mortalities set by the Ministry of Fisheries.

There were 518 trawls made east of Campbell Island in the 2007 southern blue whiting season, of which 33% were observed. The observed sea lion capture rate was 3.49 animals per 100 trawls, and has increased over the last four years, from an observed captures rate of 0.37 animals per 100 trawls in 2002. Our model estimated 14 captures for 2007 (95% c.i.: 7 to 27), an estimated strike rate of 2.39 captures per hundred trawls (95% c.i.: 1.5 to 24.6).

In 2006—07, the bootstrapped ratio method provided estimates of 12 captures (95% c.i.: 1 to 27) in other (non-squid) Auckland Islands trawl fisheries, and 5 captures (95% c.i.: 2 to 9) for all trawl fisheries on the southern Stewart-Snares shelf. Estimates were also made for the 2004—05 and 2005—06 fishing years, because of the small number of captures no clear trends were discernible.

The four estimates were combined for the three fishing years 2004—05 to 2006—07. The total estimates for 2006—07 were 51 sea lion captures (95% c.i.: 33 to 74), and 105 sea lion interactions (95% c.i.: 60 to 169). This compared with a total estimate for 2005—06 of 60 captures (95% c.i.: 39 to 87) and 170 interactions (95% c.i.: 92 to 286). The reduction between the two years was largely due to a decrease in effort in the Auckland Islands squid fishery.