Abstract
In order to achieve conservation targets while minimising the economic and societal impacts of the conservation area, systematic conservation planning algorithms, such as MARXAN, are now a preferred solution for designing Marine Protected Areas (MPAs). While MARXAN is widely used for marine conservation planning, there has been little research into the effects that the quality of benthic habitat data have on the spatial outcome of the process. This study investigates the effects that data sets at different thematic and spatial resolutions have on the design and efficiency of a MPA at Rottnest Island, Western Australia. Rottnest Island is surrounded by relatively shallow water and seagrass, Ecklonia, Sargassum, algal turf, rocky platforms and sand comprise the dominant habitat types. Benthic habitat data derived from three sensors, namely airborne Geoscan multispectral, Quickbird satellite multispectral and airborne Hyvista hyperspectral were used. Spatial resolution varied from 2.4 m to 5 m pixels. In combination with bathymetry, prominent biodiversity features and human usage patterns, the influence on planning outcomes on the three habitat datasets are compared. This work is of relevance as much of Australia’s coastal waters require conservation efforts, and the necessary procurement of benthic habitat data sets are an expensive component of MPA planning.