Adrian Hordyk

A cost-effective alternative for assessing the size of deep-water fish aggregations

Acoustic methodologies are important tools for monitoring deep-water fish and have the potential to provide high-precision estimates of aggregation size. However, they can be costly to design and implement for monitoring fish. Data from 2 years of scientific surveys of the spawning aggregations of orange roughy (Hoplostethus atlanticus, Collett, 1889) on the Cascade Plateau, Tasmania, collected using commercial fishing vessels and echosounders, were used to develop a cost-effective approach for estimating the size of deep-water aggregations. Criteria were developed to standardise the identification of orange roughy echo-traces from acoustic data from 23 surveys in 2001 and 19 in 2005. The spawning condition of the fish was monitored simultaneously with the acoustics in each year (n = 29 trawls each year). The volumes of the aggregations were estimated throughout the survey period. Although the precision of the estimated aggregation size is low, large amounts of data can be collected over extended periods by using this approach and the equipment on standard commercial fishing echosounders. Aggregation volumes varied markedly during each spawning season and changes in volume appear to be linked to the spawning biology. Monitoring the spawning biology, therefore, provides crucial complementary information for interpreting estimates of aggregation size from acoustic surveys.

A quantitative evaluation of a qualitative risk assessment framework: Examining the assumptions and predictions of the Productivity Susceptibility Analysis (PSA)

Qualitative risk assessment frameworks, such as the Productivity Susceptibility Analysis (PSA), have been developed to rapidly evaluate the risks of fishing to marine populations and prioritize management and research among species. Despite being applied to over 1,000 fish populations, and an ongoing debate about the most appropriate method to convert biological and fishery characteristics into an overall measure of risk, the assumptions and predictive capacity of these approaches have not been evaluated. Several interpretations of the PSA were mapped to a conventional age-structured fisheries dynamics model to evaluate the performance of the approach under a range of assumptions regarding exploitation rates and measures of biological risk. The results demonstrate that the underlying assumptions of these qualitative risk-based approaches are inappropriate, and the expected performance is poor for a wide range of conditions. The information required to score a fishery using a PSA-type approach is comparable to that required to populate an operating model and evaluating the population dynamics within a simulation framework. In addition to providing a more credible characterization of complex system dynamics, the operating model approach is transparent, reproducible and can evaluate alternative management strategies over a range of plausible hypotheses for the system.