Good: The product meets more than 50% of the objectives specified by the challenge

It was possible to build bathymetry for all of the sample areas, although it wasn't possible to improve on GEBCO for areas not covered by Europe, USA and Canada. Sources met minimum requirements of 500m resolution for sample areas, but rarely the desired 100m. There were usability problems with some sources: E.g. EMODnet - Due to supply format, metadata was unusable for large areas or whole tiles at a time. Some supply formats were unexpected – GeoTiffs proved unusable as they didn’t include depth values – only a colour ramp (RGB values). Individual surveys could be downloaded from SeaDataNet, but this was not feasible for the ACP project, as such a large area was being considered. Source information was eventually extracted by way of screen capturing the online view, georeferencing and attributing regions. Supporting files could not be opened/analysed by standard GIS software (4.5GB CSV file too big for 4GB RAM and Excel/ArcMap/FME). Depth data supplied in EMO file formats were too big/complex to be read/analysed by standard GIS software, NetCDF and SD files not feasible given the software available. GeoTiffs unusable as they had no depth information. E.g. CHS - Data is in raster format, requires some pre-processing to extract depth band and convert to a format to use with other bathymetry layers, including merging into single layer. Confidence information hard to determine, not all data interpolated. Data was a mixture of continuous raster surfaces ranging to oddly scattered raster pixels, not a lot of consistency throughout the whole dataset. This was categorised into confidence values.