Cape Verde coastal ecosystem: a study of community structure, trophic interactions and exploitation pattern

Tese de dout., Ciências e Tecnologia das Pescas, Faculdade de Ciências do Mar e do Ambiente, Universidade do Algarve, 2005

A mass-balance trophic model was created to describe the coastal ecosystem of the Cape Verde Archipelago for the time period from 1981 to 1985, using available estimates on biomass and catches. This time period was characterised by predominantly artisanal fisheries and a low level of motorisation; the initial phase of a more intensive fishery development. Subsequently, a dynamic simulation model, using Ecopath with Ecosim, was used to simulate from 1986 to 2000, incorporating time series information on biomass, catches and catch per unit of effort (CPUE). Using this approach, a number of problems were detected in model specification such as incompatible biomass estimates for small pelagics and various demersal fish groups. Measures of fishing effort in fishing days or trips were found to be unreliable, in particular for the artisanal fishery. Available biomass estimates for small pelagic resources from acoustic surveys were questionable. Conventional biomass dynamic modelling was therefore applied to estimate biomass of small pelagics and provide indications for management purposes, including parameter estimation and risk assessment using the frequentist and Bayesian approaches. After a number of adjustments applied to the initial model, the approach used in simulation was to fit the model to observed catch estimates by adjusting effort, placing less emphasis on fitting to CPUE and biomass estimates. A reasonable overall fit to time series of catches was obtained for 18 fish groups, using only two overall trends for artisanal and industrial effort and three specific trends for small pelagics, yellowfin, and skipjack, which are the main targets of the indutrial fisheries. The observed decrease in abundance of important predators such as yellowfin and skipjack tuna resulted in decreased predation on neritic pelagic species and some demersal fish groups, but this was compensated by higher fishing mortality over the study period. Consequently, the model estimated an almost constant biomass of neritic fish species from 1986 to 2000. Overall fish biomass decreased by 10 percent, including pelagic migratory species. Relative fishing effort was assumed to have almost tripled over the time period from 1986 to 2000 (effort directed towards small pelagics was assumed to have increased by a factor 5), but this resulted in only a 19 per cent increase in catches. Thus, previous assessments of potential fish harvest, ranging from 25 000 to 58 000 tonnes, appear to have been overly optimistic. Alternative methodologies were applied to assess the fishery resources in Cape Verde in order to gain further understanding on the dynamics of the system and the effects of fishing over time as well as to validate the results of ecosystem modelling. This included nonparametric multidimensional scaling (MDS), generalized linear models (GLM), size spectra analysis, and a type of time series analysis (MAFA). The changes observed over time in the size structure and species composition/abundance of demersal fish communities were slight or even negligible. There was a tendency for lower abundance of demersal fish in recent years, but a decreasing trend was not apparent. It is however important to note that these iv results were based on trawl survey data, which concern a limited area of trawlable grounds in Cape Verde. In contrast, time series analysis of catch data indicates that a shift has occurred with decreasing catches of important pelagic species such as yellowfin and skipjack tuna and increasing catches of small pelagics and neritic tuna as well as some demersal species. Simulation with Ecosim gave similar results as they were based on the same catch data, but this is nevertheless reassuring as alternative methods gave consistent results. We believe that this study has been successful in taking the first steps towards an ecosystem approach to assess the effects of fishing in Cape Verde, but further research is necessary to resolve some crucial issues such as the conflicting results on demersal fish biomass and the possible over-exploitation of demersal predators and moray eels. Priority should also be given to the definition of fleet components, effort and CPUE estimation, and the study of increasing efficiency, using the available catch and effort data.

Fundação para a Ciência e a Tecnologia