Bathymetry data used for benthic habitat mapping and marine geomorphology come from a variety of
sources, offering information on seabed terrain at a variety of data resolutions. Multibeam surveys
have largely become the preferred means for acquiring bathymetry data, where funds permit.
However, as mapping continues, people are increasingly making use of compiled datasets either by
combining several neighbouring multibeam surveys or by including other sources of bathymetry data.
The compilation process may be local, national or even regional (e.g. EMODNET Hydrography Portal)
and global (e.g. GEBCO) and typically the broader the area the more diverse sources of bathymetry
have gone into creating the compiled bathymetry product.

These compiled datasets are a fantastic resource, providing ready-gridded bathymetry data, either at
a single, or multiple resolutions. This meets a demand for bathymetry information which can be used
for many applications, including benthic habitat mapping. However, compiled data resources, by their
very nature, mean that the data user is increasingly distant from the original data source. Even if
quality data are supplied with the data, the user no longer has the same contact with the data
acquisition and processing pipeline as they did with discrete area surveys. This can make it all too
easy to ignore issues of data quality and/or uncertainty which are inherent to the use of gridded
bathymetry data.

Focussing on application of such data to geomorphology and benthic habitat mapping we examine
those issues that remain particularly important to consider when using bathymetry data from several
sources, and compiled datasets. Using slope as an example we focus on the implications of data
resolution, quality and data analysis scale in deriving terrain variables which are quantitative
measures of geomorphic properties relevant to habitat mapping. We also present a practical method
for computation of a confidence index for ready-gridded bathymetry data which is based on a Monte-
Carlo simulation.