Statewide assessment of Victorian marine protected areas using existing data April

The main findings of this report are briefly summarised below. Victorian MPAs represented the major habitat types and biotopes found within their biounits. However, subtidal rocky reef habitats and their associated biotopes tended to be over-represented in MPAs across the state, while sediment habitats and their associated biotopes tended to be under-represented. This was not always the case, though, with some MPAs over-representing sediment habitats, especially those encompassing deeper depth ranges like Wilsons Promontory Marine National Park (MNP). Some rarer biotope classes like Rhodolith beds and Non-reef epibiota were not well captured in MPAs. The oceanographic and environmental conditions of the broader biological regions were well represented in MPAs, but marine sanctuaries tended to capture areas of higher energy (higher wave orbital velocities and currents speeds) due to their shallower depths. Overall, the larger marine national parks more adequately captured the oceanographic and environmental conditions in each region. Connectivity models showed strong geographic patterns across the state, and patterns of connectivity were similar within the MPAs and their associated biounits. A higher total amount of larvae settled in the west and east of the state, and self-recruitment was higher in the central part of the state and in MPAs in the far west. Central and eastern parts of the state had a higher number of connections between habitats. Machine learning approaches were effective at identifying the relationships between species and groups of species with environmental and habitat variables. Stronger relationships were found with communities, followed by the abundance of species and functional groups. Overall, cooler temperatures, lower wave energy, higher current speeds and more complex seafloor habitat supported the greatest abundance and diversity of species. The relationships identified through this approach can be used to inform MPA management by providing information about the types of conditions that will best support a given species or group of species. These conditions can then be targeted for management interventions (e.g., compliance, enforcement or restoration) in an MPA if conservation of that species or community is a goal. The combination of BRUVS and habitat mapping data allowed for effective species distribution models to be produced for the whole state. Models for species richness and diversity performed well and had good predictive power, showing that richness and diversity hotspots occurred around reef habitat. This suggests that the over-representation of reef found in MPAs is beneficial for preserving biodiversity. Models for individual species had mixed performance and predictive power. Models for highly site-attached species tended to perform better than those for less site-attached species, but all could be used to gain an understanding of the distribution of species inside and outside MPAs. The assessment of MPA effectiveness found that fish species richness was higher inside MPAs than outside. Larger MPAs with fewer habitat and depth barriers supported a higher richness of fish, while fish richness declined in MPAs closer to human populations and ports. Invertebrate richness was also higher inside MPAs than outside, but MPA size and reef area had complex effects. High richness is found in both small and large MPAs, while there is an initial increase in invertebrate richness as reef area increases followed by a decrease and/or flattening off.