“…The design of field surveys thus needs to be clear which habitats, body sizes and taxa it has focused on, and what has been excluded; i.e., how it has 'stratified' sampling. Perhaps the most effective way to place the data into an environmental context is to map the geographic distribution of environmental variables (e.g., depth, salinity, temperature, substratum, topography) and habitats (Costello 1992;Costello and Emblow 2005;Costello et al , 2010aHiscock 2014). These environmental variables can be mapped through 'remote sensing' from satellites, aircraft and ships (Andréfouët et al 2008b(Andréfouët et al , 2011 and can include: seabed depth, topography, and roughness; surface water colour (an estimate of phytoplankton biomass and dominance) and temperature; depth-profiles of density (salinity) and temperature; acoustic signatures of zooplankton and pelagic megafauna; and the distribution and extent of intertidal and shallow-water habitats such as coral reefs, kelp and seagrass beds, mangrove forests, and salt-marshes.…”