Effects of resource spatial distribution and tow overlap in the accuracy and precision of common methods used to estimate dredge efficiency for bottom trawl fisheries

KITTLEIN, M.; ALBERTI, J.

Douglas

Workshop; 23rd International Pectinid Workshop; 2024

Effective fisheries management relies heavily on accurate and unbiased estimates of stock status, primarily derived from measurements of relative abundance. Inaccuracies in this complex assessment process can lead to unsustainable management practices. A crucial element in stock assessment is the precise estimation of gear efficiency, which is the fraction of the target species in the gear path that are captured and retained. This is particularly important for species that are difficult to sample by other means. The evolution of methodology in evaluating fishing gear efficiency includes initial approaches that postulate a simple relationship between catch per unit effort (CPUE) and cumulative catch or effort. Successive studies have built upon these approaches, introducing new adaptations and extensions to enhance accuracy, despite the inherent limitations such as the need for a defined population limit and assumptions about individual mixing after each extraction. The advent of high-resolution geopositioning technologies has facilitated the development of novel methods and strategies, such as the patch model. This spatially explicit approach, successfully applied in various fisheries, combines tow positions and their temporal sequence to assess spatial depletion accurately. It identifies areas within each tow that have been previously swept at different frequencies, quantifying the proportion of each tow's area that has been swept multiple times. Here we evaluate various methods, including the Leslie-Davis and DeLury models and the k-pass depletion method and the patch model, for their precision and bias in estimating gear efficiency. These methods have their own set of assumptions and limitations, from requiring a closed population to assuming constant capture probability for all individuals. The patch Model stands out for its ability to incorporate the spatial distribution of fishing effort, offering a more nuanced correction of expected catch based on an area's fishing history. We simulated different scenarios populated with shellfish exhibiting different spatial patterns to assess the performance of these methods. The patch model demonstrated superior precision and minimal bias in estimating gear efficiency across a variety of challenging conditions, including irregular resource distribution and varying degrees of experimental trawl overlap. In contrast, the other methods often underestimated or overestimated efficiency depending on the trawl overlap level, highlighting the importance of considering spatial heterogeneity and fishing effort distribution (Figure A8.1). Despite its reliability, the patch model, like all methods, is subject to uncertainties that impact its variability and bias. The study emphasizes the critical nature of cautious interpretation of findings and consideration of real-world scenarios and their implications for stock management. Advances in technology and the adoption of sophisticated models may mitigate the challenges of positional errors, enhancing the application of such models in estimating efficiency in depletion experiments. This comparative analysis underscores the strengths and limitations of different estimation methods, providing valuable insights for informed decision-making in fisheries management.