Our ability to map the seabed has been greatly enhanced by a suite of technologies.  Satellite technology can detect underwater mountains and trenches by measuring bumps and dips on the ocean surface.  Sonar technology determines depth by measuring the time it takes for sound to travel between a vessel and the seafloor, and back again, with the ‘strength’ of the echo providing information on whether the substrate is rock or sediment. Light Detection And Ranging (LIDAR) technologies, typically attached to aircraft, use infrared and blue-green pulsed laser beams pointed down towards the sea.  Whilst the infrared is reflected off the surface, the blue-green is able to penetrate the water column up to a depth of 30 meters (depending on water clarity).  The difference in the time it takes for the two lasers to return back to the aircraft indicates the depth.  Submersibles, remotely operated vehicles (ROVs) or autonomous underwater vehicles (AUVs) can be deployed with underwater cameras as well as designed to capture physical samples of the sea bed. 

Mapping marine benthic habitats is vital for many different purposes, not least for supporting marine spatial planning and ecosystem based management of the marine environment.  Benthic maps also support industrial endeavours such as offshore oil and gas developments, mining, and fisheries management, ensuring safety for mariners by identifying hazards and designing safe shipping channels, for conservation purposes such as designing marine protected areas, identifying critical habitats, and of course improving our knowledge of benthic ecosystems.

Broadly speaking marine benthic habitats are physically distinct areas of the seabed, with a typical assemblage of species (a community) that occur within it.  Habitats can be defined using a host of biotic (living) and abiotic (non-living) factors.  The occurrences of factors are not necessarily mutually inclusive. One, for example, may find two geologically identical areas with a different assemblage of species, or a species may favour one suite of abiotic factors over another at different stages of its life-cycle. 

Equally important is the species themselves.  A sedentary benthic organism’s habitat is much more spatially discrete than that of a migratory species which may utilize different parts of the ocean thousands of kilometres apart.  Furthermore it is often the case that whilst we can use technology to map an area of seabed, determining the full community makeup of the species that live in that area is not always possible.  In reality there are no sharp, well defined boundaries to tell us definitively where one habitat starts and where one ends, let alone know exactly which areas species are using.  The result is that benthic habitat maps tend to be derived from abiotic factors, creating “potential habitat maps” that indicate (to the best of our knowledge) where habitat favourable for a given species is, as oppose to definitive guides on where species are actually living. 

Benthic mapping tends to occur on a somewhat ad-hoc fashion, occurring at a specific place to meet some requirement.  The boundaries of those habitats – how they are classified - are defined to meet requirements.  Since there are many different requirements for benthic habitat maps, there are many different ways of categorising habitats. Some focus on specific regions, areas of the seabed, or feature.  Others may focus on a particular group of biota, or may be designed for a specific purpose such as the ‘Integrated Australian Classification Scheme’ developed by Dr Peter Last from CSIRO, Australia and colleagues to assist with the selection of a ‘nationally representative system of MPAs’ in Australian waters.  Whilst ‘purpose built’ classification schemes are arguably more suited to their intended purpose, the lack of standardisation means comparing habitat maps can be difficult, especially if there is a lack of quantitative data which demonstrates exactly what was in each habitat unit.   

Although varied, habitat classification systems tend to have a number of factors in common.  They tend to be hierarchical, based on different levels of scale.  The top level habitats are the ‘big picture’ in which other, lower levels of habitat classification are nested within.  Top level categories can be mapped at low resolution.  For example, “seascapes” -the equivalent of terrestrial landscapes, are heterogeneous areas created from mosaics of different habitat patches, very often with a ‘focal’ habitat patch embedded within it.  Conversely “microhabitats” tend to be single, small points with specific features and conditions, like a crevice in a rocky outcrop on a high energy shoreline that, because of the direction it faces, is a place of calm in an otherwise volatile environment. 

The level to which habitat can be classified in part depends on the resolution of the sampling technique chosen to map a given area.  Biotic and abiotic factors of an easily accessible coral reef in mapped with LIDAR and trained divers/snorkelers can be more easily mapped to a high resolution than can a deep sea ravine.  Classifications mustn’t just be identifiable units, discrete from each other, but mapable with the resources available.  Equally important is taking into account temporal variability.  Habitat factors that change over short timescales typically should be incorporated at the lower classification level, compared to those that change over long timescales.

With the proliferation of different benthic habitat mapping schemes, a number of Governmental departments have attempted to develop common platforms with the aim of providing comparable data as well as sufficient detail to meet a number of different purposes.  For example, US institution National Oceanic and Atmospheric Administration (NOAA) Coastal/Marine Ecological Classification Standard (CMECS) looks to categorise the whole marine environment, attempting to “offers an umbrella under which a national coastal and marine ecological classification can grow and evolve”.  CMECS breaks the marine environment into eight hierarchical levels, each of which contains a number of sub-components. 

At the broadest scale “ecological regions” are defined by variables such as ocean basins, ocean currents, and water temperature.  Half way down, “geoforms and hydroforms” include large features such as hydrothermal vents, sandbanks, or deep water coral reefs.  The smallest units – “habitats” and “biotopes” range from just 1 – 100 m2 in extent.  Under CMECS habitats are physically distinct, not dividable into further physical subcomponents.  Biotopes are similar in nature to habitats, but are divided into their subcomponents, based on both abiotic and biotic factors. 

The European Union takes a different approach to NOAA.  Set up to meet a number of European Union objectives and global strategies, the European Nature Information System (EUNIS) program is primarily a “reference information system for anyone working in ecology and conservation or those with an interest in the natural world”.  The EUNIS mandate is extremely broad, encompassing natural and artificial habitats in terrestrial/freshwater and marine environments.  There are six hierarchical levels that relate to benthic seabed mapping.  Level two (the first tells us if the area is in the marine environment, or other terrestrial/freshwater system) offers seven habitat types, such as littoral sediment, slope and rise benthic habitats, or abyssal benthic habitats. 

Similarly to CMECS, “biotopes” and “habitats” are considered separately, again with habitat categorisation focusing of the physical attributes whilst biotopes include biotic factors.  Whilst in CMECS habitat and biotope appear at the bottom of the hierarchy, EUNIS’ bottom level focuses on microhabitats, features less than 1m2 but nevertheless are important habitats.  In benthic ecosystems, this could include whale fall, which although spatial-temporal in nature provide habitat for a number of marine species, including specialist feeders. 

Mapping of the sea floor has come a long way since 1957 when cartographer Marie Thorp and geologist Bruce Heezen published the first physiographic map of the North Atlantic.  Today, we have mapped the entire ocean to a maximum resolution of 5 kilometres, which allows us to see the larger ocean features, like ridges, trenches, and seamounts.  Approximately 10 – 15% of the ocean is now mapped to 100 meter resolution.  With ever improving technology, our knowledge of the ocean is constantly increasing.  Equally, our requirements for benthic habitat mapping and associated classification is ever growing. 

Classification of any habitat is not a static process, rather requiring constant evolving to ensure classifications are optimally designed. The EUNIS classification scheme, which hasn’t been updated since 2004, is currently undergoing revision to include data from a number of EU initiatives such as Balance (Baltic Sea Management – Nature Conservation and Sustainable Development of the Ecosystem through Spatial Planning) and MESH (Mapping European Seabed Habitats).  Benthic habitat mapping and the classification schemes that allow us to make sense of the spatial data being produced has proved itself invaluable for conservation efforts and marine spatial planning, as well as improving our understanding the marine environment.  It is with great certainty that this need will only continue to grow.

This story was written for The Marine Professional, a publication of the Institute of Marine Engineering, Science & Technology (IMarEST).