Laurence T. Kell

A potential method for the differentiation between haddock fish stocks by computer vision using canonical discriminant analysis

Morphometric measures of shape have been used successfully in evolutionary biology to distinguish between different types of fish (Bookstein et ab, 1985) and to identify fish species Tayama et al. (1982). Both colour and shape differences of cod caught from different fishing grounds have been noted by Love (1970). Attempts have been made to measure differences in the shape of haddock between the North Sea and Rockall fishing grounds by measuring various characteristics such as head and body length using a ruler on about 3@40 fish from each fishing ground (Mater, 1993). However, no significant differences were obtained. A machine using computer vision to sort fish by species and size using shape and colour descriptors has been described by Strachan (1993a, b). The machine carries out these measurements quickly, easily, and accurately on a large number of fish. Colour images of haddock from the North Sea and Rockall Plateau were collected and measurements on colour and shape were then made using image-processing techniques. This allowed canonical discriminant analysis to be employed to determine whether there were any significant differences in the colour and shape of the stocks between the two different sea areas.

Chapter 15 Operational management procedures: An introduction to the use of evaluation frameworks*

Publisher Summary A management procedure is a simulation-tested set of rules used to determine management actions, in which the data, assessment methods, and the harvest control rules for implementing management actions are prespecified. An operational management procedure (OMP) is defined as a management procedure that is currently being used to determine management actions or has been intensively tested by a competent management body to a level where it could be used in practice. An important benefit of management procedures is they can be designed to satisfy a variety of pre-agreed management objectives, including both biological and economic factors, making them more likely to be acceptable to a wide range of user groups. However, the modelling philosophy may be difficult to understand for non-experts, which means that extra effort is needed to explain them and how they are designed. This chapter describes various aspects of management procedures, including the philosophy behind them, how they are constructed based on available data and knowledge, their objectives, the methods used to evaluate them, and their differences and advantages over traditional stock assessment and management regimes. In addition, some examples of management procedures are also reviewed.

Mixed and Multi-Stock FisheriesIntroduction

The September 2003 ICES Annual Science Conference was held in Talinn, Estonia, and Theme Session V was ‘‘Mixed and multi-stock fisheries e challenges and tools for assessments, prediction, and management.’’ The theme session brought together researchers from the salmonid and marine fishery fields to address common problems faced when multiple fleets harvest a common stock, or multiple stocks are harvested by a common fleet. In all, 23 papers were presented during the theme session, prefaced by an invited plenary lecture by Randall Peterman that focused on challenges facing fisheries scientists and managers. The ten papers that follow represent the diversity of topics in the Theme Session, but all focus on approaches to overcome the uncertainty inherent in these complex situations. Fleets and fisheries that harvest multiple stocks, and stocks harvested by multiple fisheries, face a number of complexities. Less productive stocks in a mix of stocks may suffer unsustainable mortality, while more-productive stocks continue to support sustainable catches. Stocks harvested together may be at a very different status relative to safe biological limits, which would make different harvest strategies necessary in the same fisheries. Where fisheries take multiple stocks of the same species, uncertainties about allocating catches to stocks present special assessment problems. Projections to guide advice on effort levels appropriate for different stocks in the harvested complex can also thwart traditional approaches, and can produce results that are either incompatible across stocks, or not straightforward for managers to apply. Therefore, the objectives for Theme Session V were to describe developments in assessment and projection methods for multi-stock and multi-fleet fisheries, and to discuss opportunities for applying new concepts more widely. Management of complex systems is not necessarily facilitated by making stock assessment and management more complex. Ideally, simple rules should be devised that are robust to the pervasive uncertainties in such systems. However, the process by which these simple rules could be found would incorporate as much complexity as possible.

Predicting global tuna vulnerabilities with spatial, economic, biological and climatic considerations

Overfishing impacts the three pillars of sustainability: social, ecological and economic. Tuna represent a significant part of the global seafood market with an annual value exceeding USD

FLR: an open-source framework for the evaluation and development of management strategies

42B and are vulnerable to overfishing. Our understanding of how social and economic drivers contribute to overexploitation is not well developed. We address this problem by integrating social, ecological and economic indicators to help predict changes in exploitation status, namely fishing mortality relative to the level that would support the maximum sustainable yield (F/FMSY). To do this we examined F/FMSY for 23 stocks exploited by more than 80 states across the world’s oceans. Low-HDI countries were most at risk of overexploitation of the tuna stocks we examined and increases in economic and social development were not always associated with improved stock status. In the short-term frozen price was a dominant predictor of F/FMSY providing a positive link between the market dynamics and the quantity of fish landed. Given the dependence on seafood in low-income regions, improved measures to safeguard against fisheries overexploitation in the face of global change and uncertainty are needed.