MISA - Marine Innovation SA

• Seafood Product Quality & Value-adding
• Aquaculture Innovation
• Ecosystem Services
• Biosecurity

Foraging ecology and diet analysis of Australian sea lions

Executive Summary

Recent Commonwealth Department of the Environment and the Heritage (DEH) Ecological Sustainable Development (ESD) assessments of the South Australian (SA) rock lobster (SARLF) and southern and eastern scalefish and shark fishery (SESSF) identified interactions with protected species (particularly seals), as one of the key bycatch issues. The issues are most relevant to SA waters where threatened Australian sea lion (ASL) populations are located, and where un-quantified interactions between seals and the SARLF and gillnet sector of the SESSF fisheries are known to occur. Recommendations from fishery ESD assessments, fishery Bycatch Action Plans, and a recently drafted Recovery Plan for the ASL, have all identified the importance of assessing and mitigating interactions between seals and commercial fisheries. This study provides a desk-top risk-assessment of seal fisheries interactions in the SARLF and gillnet sector SESSF in SA and adjacent waters, and makes recommendations on future research and management responses. A review of the PIRSA and AFMA fishery logbooks identified the major constraint to the assessment of bycatch risk to seal subpopulations was the absence of quantitative data on bycatch rates in both the gillnet sector SESSF and SARLF. Anecdotal evidence and entanglement data suggest there has been significant underreporting of seal interactions in these fisheries.

In SA there are 38 ASL subpopulations that produce around 2,674 pups, with the total population size estimated at about 10,900. However, most pup production (67%) occurs at 6 sites, hence the median pup production is very low (25.5 pups), with the majority of sites producing small numbers of pups (60% produce <30 pups per season). Population viability analysis (PVA) on ASL subpopulations reinforced the recent listing of the ASL as a threatened species, by confirming that large numbers of subpopulations with low pup production are vulnerable to extinction. PVA simulations suggested that in absence of anthropogenic mortality, a number of ASL subpopulations will go quasi-extinct (ie the number of adult females is too low to ensure population persistence; <10 females), but in the face of small (1-2 additional females/year) but sustained anthropogenic mortality (eg. from fishery bycatch), most other small subpopulations will become quasi-extinct and negative growth will become a feature of even the largest subpopulations. There is apparent depletion (ie. very low pup production) of a large number of subpopulations that may be indicative of widespread subpopulation declines in the species. That such declines may be ongoing and attributable to anthropogenic mortality (ie. fishery bycatch) is a hypothesis that requires urgent attention.

The risk of bycatch in the gillnet SESSF and SARLF were assessed based on estimates of interaction probabilities. These were a function of the extent to which historic fishing effort and seal foraging effort (based on foraging distribution and population models) overlap in space and time. ASL demonstrated a high risk of significant depletion and quasi-extinction as a result of fishery bycatch. By combining PVA outcomes with bycatch scenarios based on interaction probabilities, this study identified the subpopulations, regions and marine fishing areas (MFAs) most at-risk from seal bycatch.

Bycatch from the gillnet SESSF is most likely to provide the greatest risk to ASL, because of almost complete spatial overlap in fishing effort with ASL foraging effort, it is a year-round fishery with relatively high fishing effort that can potentially interact with all ASL age-classes. The impact from SARLF is likely to be less because there is less overlap in fishing effort with ASL foraging effort, fishing is restricted to seven months of the year (November-May) and bycatch is likely to be restricted to pups and juvenile seals. However, the potential additive and interactive impacts posed by combined bycatch in these fisheries could be significant. Results from this study suggest the two fisheries investigated lend themselves to different mitigation approaches to addressing seal bycatch issues. In the gillnet SESSF, gear modification options are limited, but spatial management of fishing effort may provide a range of risk-reduction options, but would need to be coupled with independent observer bycatch data to demonstrate and justify the benefits from different closure options. In contrast, there are significant options for gear modification in the SARLF, with pot-protection devices already used in some parts of the fishery. Quantitative testing of these and alternate protection measures (as is taking place in the WA WRLF), and industry wide adoption of best-mitigation practices may eliminate seal bycatch in this fishery, without the need for an expansive and costly independent observer program. Recommendations for future research are made, that should result in the successful mitigation of seal bycatch issues and as a consequence address the recommendations of the fishery ESD, Bycatch Action Plan, ASL Recovery Plan and assist in the recovery of the threatened ASL.

Enhanced spatial tools for risk assessment will be required if spatial management of fishing effort is to become a management strategy for mitigating ASL bycatch in the demersal gillnet fishery. Such tools would provide a simple mechanism for policy makers and managers to evaluate the benefits and costs of different spatial allocations of fishing effort, in terms of increasing or decreasing: 1) risk to sea lion subpopulations and 2) fishery catches. However, further development of such tools are required, because current models are limited by the absence of data on the foraging movements of sea lions in some high-risk regions, as well as the absence of accurate fishing effort data.

Further satellite tracking of ASLs at subpopulations identified as high-risk was undertaken as part of this study, to improve the accuracy of spatial foraging models. This pilot study demonstrated an approach to refine assessments of sea lion interactions with commercial fisheries, where quantitative data on interactions are not available. This approach may enhance the spatial information on which mitigation options and decisions about spatial management of fisheries are based. Importantly, the pilot study determined that colony specific information on sea lion foraging effort could be used to refine the spatial management of fisheries when using fishing effort data that was summarised into 1 x 1 degree boxes (Marine Fishing Areas). In 2006, demersal gillnet fishers were required to record the latitude/longitude positions of each net-set, and from July 2007, all vessels will be fitted with satellite-linked vessel monitoring systems that will significantly improve the resolution of fishing effort. Following these improvements to fishing effort data sets, it is recommended that bycatch probabilities be re-estimated with colony-specific information on seal lion foraging effort and used to model the benefits of different spatial-management scenarios that could include area-closures, and reductions or redistributions of fishing effort.

The diet of the ASL is currently poorly understood, which hampers our understanding of their key prey species, habitats and trophic interactions with fisheries. Traditional faecal analysis techniques have proven ineffective in ASL because most prey remains are completely digested. We conducted a feeding trial on captive ASL to determine whether faecal DNA analysis could be used to quantify sea lion diet. This study demonstrated that analysis of faecal DNA can detect the presence of sea lion prey DNA, despite no identifiable prey hard parts being recovered from the same scats. The results from our experiments also demonstrate that DNA extracted from ASL scats is highly degraded, because we could not detect prey DNA using molecular primers more than 100 base pairs in length. Quantitative PCR indicated that the amount of mtDNA amplified from scats was related to the amount of novel prey ingested. Quantitative estimates determined the difference between periods of high consumption of novel prey and low consumption of a particular prey type, but did not pick up smaller differences in consumption rates. Interestingly, mtDNA quantitative estimates were significantly higher for squid than shark from periods when the seal was fed in equal proportions. These differences may be related to the concentration of DNA in the different tissue types of each species.