The influence of sampling method on detecting benthic biodiversity patterns at the ecoregion scale on the South African west coast

Brandt, Silke

The influence of sampling method on detecting benthic biodiversity patterns at the ecoregion scale on the South African west coast

Master Thesis

2022

Abstract

Long-term monitoring of marine benthic communities provides data which are essential for effective ocean management. However, long-term monitoring is limited by the difficulty and expense of sampling deep seafloor areas comprehensively enough to represent the whole benthic community. This has led to the development of a wide array of seafloor sampling methods. Consequently, the integration and prioritisation of data collected using different methods remains an area of concern. Demersal research trawling and grab sampling are two methods employed to sample the marine benthos in South Africa, targeting different habitats and fauna, at different scales and with different sampling efficiencies. Both datasets inform the national marine ecosystem classification, yet the consistency between biodiversity patterns detected by each sampling method has not yet been explored. The aim of this thesis is to determine the influence of sampling method on the detection of benthic biodiversity patterns. This was explored using demersal trawl and grab datasets collected from twenty-four pairs of stations within the Southern Benguela Shelf ecoregion on the west coast of South Africa (70 – 600 m) between the years 2009 and 2020. Differences in benthic structure, univariate diversity indices, and multivariate assemblage structure were compared between demersal trawl and grab datasets using both taxonomic and trait-based approaches. This study utilised the rarely applied co-correspondence analysis (CoCA) to test for congruency in multivariate assemblage patterns sampled by trawls and grabs. Furthermore, the use of Biological Traits Analysis (BTA) allowed for the assessment of functional diversity patterns which is often a missing link when measuring biodiversity relative to ecosystem functioning. The epifaunal community was dominated by Asteroidea, Decapoda and Gastropoda, whereas Polychaeta, Amphipoda and Bivalvia dominated infaunal communities. BTA found trawl samples to be dominated by epifauna exhibiting large sizes, dorso-ventrally flattened body forms, free-living and surface-crawling life habits, moderate mobility, predatory feeding, planktotrophic larval development and medium-long lifespans. Grab samples were dominated by infauna with small sizes, vermiform or laterally flattened body forms, low mobility, surface deposit feeding strategies, planktotrophic larval development and short lifespans. Demersal trawl and grab sampling detected significantly similar patterns in species abundance, species richness, species diversity, and functional richness values across the west coast. Species evenness, functional evenness, functional diversity, and functional redundancy gave no evidence of a significant relationship between the two sampling methods. CoCA found infaunal assemblage patterns to be highly correlated with epifaunal assemblage patterns across the study area using both taxonomic and trait-based approaches. Environmental and spatial gradients, including depth, longitude, and sediment characteristics, played key roles in structuring broad scale biodiversity patterns. The results of this thesis have implications for how biological datasets from trawl and grab surveys should be prioritised or weighted at different ecological scales when incorporated in the South African marine ecosystem classification and mapping process. This is the first step in transitioning the current ecosystem classification from a data informed, expert driven approach to an expert informed, data driven approach through the use of quantitative multivariate statistical techniques. Furthermore, multi-method biodiversity studies are crucial to represent the entire benthic community and understanding the extent to which the choice of sampling method affects the biodiversity patterns detected is an integral component of accurate ecosystem delineation. The findings from this thesis can be applied to future assessments of South Africa's marine ecosystem classification, increasing its accuracy, and therefore contributing to improved ecosystem-based sea-use management.

Keywords

Biological Sciences

Reference:

Collections

Masters

Full item page