The school shark stock is subject to a recovery plan so commercial targeting cannot resume until adequate recovery is demonstrated. The consequent avoidance of school shark means CPUE is no longer a valid index of relative abundance. The existing stock assessment is effectively being used to project the population forward in time (from 1997), given known catches, and is thus predicting the rate and level of recovery of the stock. No information is currently available to compare the predicted recovery rate with the actual rate. Anecdotal reports from the fishing industry suggest that recovery is more rapid than predicted, suggesting lost revenue from current low bycatch TACs.
Avoidance of school shark is reducing the economic efficiency of the gummy shark industry. Economic losses are most keenly felt in South Australia where sea-lion and dolphin protection has closed large areas of traditional fishing grounds. The 20% school:gummy ratio imposes additional, effective, closures of areas of high school shark abundance. The need to protect school shark has also lead to a conservative response from SharkRAG when stock assessments have suggested increases in the gummy shark TAC (SharkRAG 2013).
School shark are highly mobile and the current stock assessment model relies heavily on a model of movement that hasn’t been tested against data. The true movements and underlying stock structure of school sharks are key uncertainties for understanding and effectively managing this stock, complicating the interpretation of any CPUE based abundance index. This project is likely to better understanding of stock structure and movement, and may facilitate use of the existing Pittwater pup index, which stretches back to the 1940s.
This research proposal is in response to the updated ComFRAB call for research on 23 May 2013, which asked for proposals relating to "Develop better measures for School Shark abundance".
We found 65 half sibling pairs (HSPs), 3 parent-offspring pairs (POPs) and 34 full sibling pairs (FSPs); sufficient for close kin modelling. Our model estimates a School Shark stock in the region of 50,000 mature individuals during 2000. Although the coefficient of variation (CV) for our abundance estimate ranges from 0.23 to 0.28 over 2000 to 2011 (most precise in 2002-2003, at 0.23) the standard error on the trend in mature abundance is large relative to the trend itself so that although the median trend is slightly upwards, a downward trend cannot be ruled out.
Future projections assuming varying levels of future close kin sampling for up to four years showed that standard errors on trend and abundance should greatly reduce. SharkRAG have recognised that CKMR provides a viable alternative to conventional stock assessment for School Shark and have recommended that CKMR continue to be used as a monitoring tool for School Shark and we scoped such continuing work.We developed two, very simple, models that provided similar abundance estimates to those of our more sophisticated close kin model, giving us confidence that the close kin model correctly interpreted the close kin data. Our estimate of abundance is three to four times lower than that of the most recent stock assessment model, when that was projected forwards assuming similar levels of catch to those that have occurred (Thomson, 2012). Our model was not able to sustain the catches that occurred during the 1990s, even under optimal survival conditions for juvenile School Shark. This suggests that the School Shark population consists of more than one reproductively isolated stock, and that the population that we measured is likely to be a remnant of what was present in the 1990s.
It is possible that environmental degradation of School Shark nursery areas (DEWR, 2008) is the explanation for our finding. As there has been little recovery of those areas, School Shark might not have the capability to recover to their previous stock size. In this case, management reference points that rely on the assumption that stocks will recover to their pristine abundance in the absence of fishing, are not useful for School Shark. Conventional stock assessment models give more precise estimates of relative, than of absolute, abundance but CKMR gives reliable estimates of absolute abundance. This provides managers the opportunity to manage School Shark according to a more relevant quantity than abundance relative to a no longer attainable pristine state last seen in the 1920s.
Our work has advanced close kin methodology through the refinement of software developed for quality control of genetic sequencing data, and for kin finding. Our work represents the first application of CKMR to a commercially exploited shark population.