The Prediction Project aims to develop models that predict patterns of marine biodiversity, with specifiable precision from more broadly available physical surrogates, at regional and national scales around Australia. It will collate suitable biological data, examine the performance of a variety of statistical models (precision, sensitivity, scale-dependency, etc.), and examine which biodiversity measures are consistent surrogates for other measures (eg. phyla, diversity indices, habitats). The project will encompass depths from temperate and tropical shallow reefs, through continental shelf, outer shelf, to the continental slope (1000m is maximum depth for which biological data are available in Australia), using the most data rich areas to develop the statistical approaches. There are strong links with the Surrogates and Biodiversity projects.