Public Thesis Defense of Antoine SAINT-AMAND

Abstract:

The Great Barrier Reefs (GBR) is a treasure of biodiversity, with an immense ecological, economical and iconic value. However, this unique ecosystem is facing many threats, and is particularly suffering from global change. Management measures are therefore more than ever required to protect and restore the reefs. By reproducing both the ocean circulation and dispersal processes, biophysical models can help to inform and support such management measures. However, model outputs always come with a certain level of uncertainty, potentially impeding their reliability. Among others, the spatial resolution of ocean circulation models can affect the overall quality of their outputs. In this thesis, we use the 2D version of the multiscale coastal ocean model SLIM to assess the effect of spatial resolution on the whole modelling sequence. We hence compare the outputs of various model setups with five resolutions ranging from 250 m to 4 km. While the tidal signal is similar for all resolutions, strong discrepancies appear for simulated current velocities, especially over reefs and in their close vicinity. This in turn affects the dispersal pattern of virtual particles, leading to significant cumulative differences over multiple weeks. In a second stage, we also show that coral connectivity indicators inferred from larvae dispersal simulations vary with the resolution of the model. We hence underline the importance of using a resolution fine enough to represent the processes of interest when using models to support reef management. As a last step, we evaluate the footprint of sediments that would be released from a coal mine project adjacent to the GBR. We demonstrate that fine sediments could have far-reaching impacts on seagrass meadows, a dugong sanctuary, and important turtle nesting beaches. Throughout the thesis, we underline how processes happening at a small-scale influence in turn those happening at larger scales in coral reef environments.