University of Patras
Προσδιορισμός κρίσιμων χαρακτηριστικών της ελληνικής αλιείας και ανάπτυξη μεθοδολογίας για τη βελτίωση της παρακολούθησης και διαχείρισής της
Abstract
dc:descriptionCommercial fisheries constitute a valuable source of information that can aid, either standalone or complementary to experimental surveys, in assessing fish stock status as well as providing management and monitoring insights. Within the European Union, member states collect commercial fisheries data within national monitoring surveys under the auspices of the Data Collection Framework (DCF) that supports the broader Common Fisheries Policy. Such routine monitoring surveys collect data at the haul level and typically record catch biomass and length composition data together with operational, technical and spatial information about the fishing gear deployment. Fisheries data, either commercial or experimental, need statistical adjustments in order to infer about the actual status of the fished species/stocks. These adjustments rise due to the sampling procedure, which is essentially the actual fishing activity, and refer to the magnitude of the fishing activity (fishing effort) and the ability of the fishing gear to capture individuals (selectivity). These adjustments constitute the so-called standardization of fisheries data and are routinely performed in fisheries science to correctly assess the status of a stock since extrapolating from the commercial catch to the actual population depends on actual effort and selectivity during fishing. Mixed fisheries is an umbrella term that refers to fisheries worldwide that cannot be characterized with a “single-fleet” and “single-stock” attribute. Mixed fisheries typically involve a variety of fleets and fishing gears (multi-gear) as well as a variety of stocks (multi-species) that are exploited by multiple fleets in an overlapping manner. This involvement of gears and species in mixed fisheries raises significant and non-trivial statistical concerns regarding the standardization of mixed fisheries data which can be both dependent on each specific fishery case as well as generic regarding the mixed fisheries global character. These concerns arise due to the variety of fishing gears and fishing gear technical characteristics deployed which translates to a lack of a common scale for effort and selectivity. The current thesis has focused on a) highlighting and overcoming the main peculiarities of mixed fisheries that hinder standardization of commercial data as well as b) presenting potential applications of the standardization procedure (using data routinely collected in observer surveys) in order to investigate the factors that shape the mixed fisheries character and to provide appropriate information for assessment and management potentially adopting an ecosystem-based approach. The data used were collected during observer surveys onboard fishing vessels in the Ionian Sea in Greece (FAO GFCM GSA 20) which constitutes a typical example of mixed fisheries in the broader Mediterranean Sea fisheries scene. The multi-gear and multi-species aspects of mixed fisheries were analyzed in a “multi-data” hierarchical (mixed) Generalized Additive Modeling (GAM) standardization framework that encompasses complementary information from different gears and species. In order to overcome the incomparability of the variety of fishing gears, the notion of Preferential Sampling was introduced which can account both for a) the sampling bias present in fisher behavior (since fishers opt for areas and seasons of high catches and hence high profit) and b) the lack of a common scale among the different fishing gears. The incorporation of many species in a single modeling framework was feasible by adopting a hierarchical modeling structure so that ecological information is potentially shared among the different species. The framework presented in the current work is shown to improve the statistical properties of inferring from single-gear and/or single-species models. Under this framework, some intrinsic applications and extensions are developed. The multi-species component of the model derived common ecological and distributional patterns (e.g. areas and seasons of persistently high abundance) among species, hence expanding the notion of developing species portfolios with regards to managing the available stocks. The multi-gear component was used to make the distinct gears and selectivities comparable and thus developing the notion of “multi-management” where gear-specific management measures can be put under a single test of effectiveness. Next, a straightforward way to add the ecosystem dimension in assessment and management was to utilize the size spectrum theory which can drop the taxonomic attribute of the data in favor of using size, which is a key biological trait, to represent the ecosystem status. Innate size-specific attributes such as abundance and diversity were studied in order to provide size-specific advice for potential management schemes. Additionally, the effect of climate change was put forth with regards to the disturbance of the relative abundance of the different sized individuals within the stocks. Within the size spectrum theory, an R package was developed in order to facilitate the exploration of length composition data (as collected during commercial monitoring programs or experimental surveys). This package uses size spectrum models in order to typify the workflow of analyzing factors affecting the compositional patterns of the relative sizes within a population utilizing the flexibilities of a GAM framework. Finally, management practices were proposed within a sustainability context. Three key sustainability indices were computed, based on the DCF collected data, and the sustainability of the fishing activity was examined by presenting hotspots of recruitment and reproduction as well as developing an effectiveness metric for the technical characteristics of the fishing gears with regards to optimizing sustainability. Some key aspects relevant to managing fisheries such as spatial/temporal heterogeneity within a single GSA as formally defined, skipper effect, and within-gear and between-gear variability have been measured and proven to be uncaptured by the current static management schemes; hence opening the way to introduce more adaptive management scenarios to be put to the test. The three-level standardization (catch standardization, size spectra, sustainability of multiple management measures) developed here can provide a useful inferential toolbox for the application of the Ecosystem-Based Fisheries Management in mixed fisheries worldwide.
Degree
thesis:*- Grantor dc:publisher
- University of Patras
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Lazaris, Alexios-Petros
- Λάζαρης, Αλέξιος-Πέτρος
Subjects
dc:subject × 10Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/60150