Novel Mapping Of The Shallow Water INFOMAR Data Set: Towards Irelands First Shallow Water Atlas (NoMans_TIF)

UCC PI’s: Aaron Lim, Andy Wheeler

It is estimated that less than 5% of the seafloor is mapped at a resolution to that of similar studies on land (Wölfl et al., 2019). Despite this, national seabed mapping efforts are not ubiquitous with countries such as Ireland and Norway having mapped significantly more of their seabed territory than other countries. Seabed 2030, a collaborative project between the Nippon Foundation of Japan and the General Bathymetric Chart of the Oceans (GEBCO), aims centralise all available bathymetric data producing a publicly available map of the world ocean floor by 2030 (Mayer et al., 2018). As such, by 2030, there will be extremely large volumes of bathymetric data available. There is a ‘bottleneck’ in the time it takes to analyse this data into policy-ready information as well as data analyses completed in geographic isolation meaning results from area to area and not comparable (Lim et al., in review). Here, we propose a first-order geomorphological-based seabed classification method that can be applied to large volumes of bathymetric data which can reduce time and subjectivity through a standardised approach. Specifically, this report applies various first-order geomorphological-based classifications to segments of the INFOMAR dataset to assess their advantages and disadvantages.

The report shows that while all methods require further testing on other more segments, specifically those with gradational boundaries, they all present specific advantages and disadvantages. Nevertheless, an ensemble approach seems most advantageous. However, it is strongly advised to further test the pure object-based method with more suitable learn-applied machine learning classifiers which have shown promising results in similar studies (Summers et al., in prep.).

The project has 3 main aims (from the original proposal):

Aim 1) collate the datasets already collected by INFOMAR, combine this with data from existing studies and create a geomorphological map of the features on the Irish shelf (<200m bsl)

Aim 2) The development of techniques to map large datasets, especially in areas where ground-truthing is scarce. This will involve automated feature detection and the designation of ‘potential’ habitats

Aim 3) To produce an atlas of seabed features that can be used by a range of stakeholders to benefit the marine sector in Ireland, as well as on an international scale.