Patterns in the landed catches and determinants of catch composition in the South African inshore trawl fishery

Besseling, Natasha

Patterns in the landed catches and determinants of catch composition in the South African inshore trawl fishery

Thesis / Dissertation

2023

Abstract

The inshore trawl fishery, which nominally targets Merluccius capensis (shallow water hake) and Austroglossus pectoralis (east coast sole), has the second highest bycatch rate of all South African fisheries. For over one hundred years there has been concern about the sustainability of the harvests in this fishery. To enable effective management of the many species harvested, I aim to describe patterns in the effort, landed catches, catch composition, and bycatch rate in the inshore trawl fishery from 1990 to 2019. In a second analysis, I aim to determine the driving factors that influence landed catch composition in the fishery, with a specific focus on area and vessel effects, which can be used to improve management strategies. The total size of the fishery has decreased from 40 vessels in 1990 to 15 vessels in 2019 and has a total bycatch rate of 36%. A total of 48 nominal species were recorded among the landings, but only thirteen made up 99% of the total landed catch composition by weight. Total effort and landed catch decreased by 70% and 62% respectively from 1990 to 2019, but bycatch rate increased by 22.7%. The landed catch per unit effort (cpue) of the landed target and bycatch species showed a cycle of increasing periodicity of five, six, eight, and nine years. Bycatch rates were 3% higher in trawls targeting A. pectoralis than those targeting M. capensis (t(df=284077) = 26.5, p <0.001). Bycatch rates were 3% higher during the day than at night (t(df= 86239) = 30.33, p <0.001). Bycatch rates were 12% higher in the depth group with the highest bycatch rates (90 to 100 m) than in the group with the lowest bycatch rates (110 to 120 m) (H = 4955.2, DFgroups= 5, P<0.001). I tested whether the variation in landed catch composition among trips was due to the vessel or environmental influences, including year and season, using PERMANOVA models. Based on landed catch compositions, the years clustered into three groups at an 85% similarity (1990 to 1999, 2012 to 2015, and the remaining years in the period 2000 to 2019). The months clustered into three groups at a 94% similarity splitting the year into two almost equal halves, and then splitting the second half of the year into winter and spring. The vessels clustered into five groups at a 78% similarity. The inshore trawl grounds were divided into 20' by 20' grid cells, which were used to aggregate the trawls and clustered into six groups at an 87% similarity. 7 | P a g e The landed catch composition per trip was influenced by the vessel (F(1, 21) = 32.7, p < 0.001, R2 = 0.31), year (F(1, 29) = 16.0, p < 0.001, R2 = 0.21), and season (F(1, 3) = 30.0, p < 0.001, R2 = 0.04). To test whether the variation in landed catch composition that was explained by the vessel was due to differences in vessels or differences in the areas trawled by the vessels, direct comparisons were made between landed catches of vessels fishing in the same area and the same year and season. The landed catch composition of five vessels was compared for two years, while ‘simultaneously' operating in ten grid cells. The landed catch composition per trawl was influenced by the grid cell (F(1, 9) = 15.5623, p < 0.001, R2 = 0.12), vessel (F(1, 4) = 25.2, p < 0.001, R2 = 0.08) and year (F(1, 1) = 22.9, p < 0.001, R2 = 0.02). The grid cell has the largest measured impact on the landed catch composition. The grid cells in all clusters were grouped spatially, except for one grid cell in group four. Each group showed distinct species assemblages that could be used to tailor management strategies for bycatch in the fishery. Vessel is an important factor in determining the landed catch composition of a trawl and has a greater effect than year. The ability to control the catch composition is posited to be a function of the fishers' skill level, the fishing strategy employed, the vessel specifications, and decisions regarding discarding and retention of bycatch species. Knowledge of the effect of the vessel on the landed catch composition implies that there is scope for targeting fisher behaviour when designing new mechanisms to control bycatch.

Keywords

Marine Biology

Reference:

Collections

Masters

Full item page