Structure of the classification

A framework for the classification (EUNIS levels 2 and 3)

Whilst the classification has been developed for nature conservation purposes and hence needed to be biologically driven, the dynamic nature of certain populations of species, and sometimes whole communities, meant it was essential to identify the habitat within which the community (of potentially varying composition) occurs to ensure types defined would be robust over time. Full use is also made of the habitat attributes to provide a structure to the classification which is both logical and easy to use. In this way much more significant use of habitat characteristics is made than for many terrestrial classifications, where vegetation alone is often the prime determinant of the classification's structure. The classification is presented in such a way as to allow access via either the habitat attributes through a series of habitat matrices or the biological community in a hierarchical classification of biotopes and higher types.
 
Each of the environmental gradients outlined in Table 1 can be considered to form an axis within a multi-dimensional matrix. Each community develops according to a suite of environmental conditions (and biological influences) which lie within such a multi-dimensional matrix, reflecting varying biological character according to its position along each particular gradient. Although the degree of importance of each habitat attribute varies for differing communities, the first two, namely substratum and the vertical gradient or zonation, appear to play a highly significant role in all communities. They are also the most easily and reliably recorded attributes in the field and are readily mapped. These factors combine to make the attributes of substratum and zonation the most appropriate for structuring the upper end of the classification.
 
The primary habitat matrix of substrata versus zonation (Table 3) illustrates the framework adopted for the classification. It represents EUNIS levels 2 and 3 in the hierarchical classification and has been developed to reflect the most significant changes in biology at a scale appropriate to an internationally applicable classification. Table 4 outlines the rationale behind the divisions adopted for these two levels in the classification.
 

Development of a hierarchical classification

It was considered essential to develop a hierarchical classification structure in which broader, higher types in the classification could be more finely divided to support more detailed use. The development of the hierarchy comes from both a top-down and a bottom-up approach:
 

Top-down classification

Taking the marine environment as a whole, it can be sub-divided into a series of broad habitat categories, based largely on their physical character as described here. At the very broadest level, differentiation can be made between rock and sediment habitats, and between those on the shore (intertidal) and those in the subtidal or deep ocean. These high-level divisions can be further subdivided on the basis of different types of sediment (e.g. gravel, mud), different degrees of wave exposure on rocky coasts (exposed, sheltered) and varying depth bands below the low water mark (e.g. shallow water where light penetrates, deeper water with little light). Such broad-scale differences in habitat character are readily understood by non-specialists and provide classification types that are easily mapped; however, they also have ecological relevance as they reflect major changes in habitat character upon which species depend (see above).
 
The top-level types depicted in the primary habitat matrix (Table 3) show levels 2 and 3 in the hierarchical classification. It is important to note that these top-level categories were developed after consideration of how best to classify biological data at the lower end of the classification.
 

Bottom-up classification

Field survey, whether on the shore or in the subtidal, reveals that different places support different communities of species. The precise combination of species and their relative abundance varies from place to place and is dependent both on environmental characteristics and upon interactions between species. Visits to different sites that have similar environmental characteristics, such as sediment type and depth, show certain levels of similarity in their species composition. Multivariate analysis of the data from field surveys (e.g. grabs, diver observations) groups these data into clusters that have similar character – this forms the basis of defining the types at the lower end of the classification (levels 5 and 6). These can themselves be grouped into higher types with similar character (level 4), thus forming the basis for the bottom-up approach to development of the classification based on real field sample data.
 
The two approaches have been merged together into a single hierarchy, thus catering for broad-scale application in management and mapping and fine-scale application for detailed survey, monitoring and scientific study. The levels can be differentiated in relation to their degree of biological distinctiveness, to the ability to discriminate types by various methods of remote and in situ sampling, to the ease of recognition by workers with differing skill levels and to the end use of the classification for conservation management at various scales.
 
Six levels in the hierarchy have thus been developed, equating directly to the levels in the EUNIS classification:
 
Level 1: Environment (marine) – A single category is defined within EUNIS to distinguish the marine environment from terrestrial and freshwater habitats.
 
Level 2: Broad habitats - These are extremely broad divisions of national and international application for which EC Habitats Directive Annex I habitats (e.g. reefs, mudflats and sandflats not covered by seawater at low tide) are the approximate equivalent.
 
Level 3: Main habitats - These serve to provide very broad divisions of national and international application which reflect major differences in biological character. They are equivalent to the intertidal Sites of Special Scientific Interest (SSSI) selection units (for designation of shores in the UK) (Joint Nature Conservation Committee 1996) and can be used as national mapping units.
 
Level 4: Biotope complexes - These are groups of biotopes with similar overall physical and biological character. Where biotopes consistently occur together and are relatively restricted in their extent, such as rocky shores and very near-shore subtidal rocky habitats, they provide better units for mapping than the component biotopes, better units for management and for assessing sensitivity than the individual biotopes. They are relatively easy to identify, either by non-specialists or by coarser methods of survey (such as video or rapid shore surveys), thereby offering opportunities for data collection by a wide range of people and without recourse to specialist species identification skills.
 
Level 5: Biotopes - These are typically distinguished by their different dominant species or suites of conspicuous species. On rocky substrata, most should be readily recognised by workers with a basic knowledge of marine species, although quantitative sampling will be necessary in many of the sediment types. The vast majority of available biological sample data are attributable to this level (or the sub-biotope level), which is equivalent to the communities defined in terrestrial classifications such as the UK National Vegetation Classification (e.g. Rodwell ed. 1995). Intertidal and subtidal sediment biotopes may cover very extensive areas of shore or seabed.
 
Level 6: Sub-biotopes - These are typically defined on the basis of less obvious differences in species composition (e.g. less conspicuous species), minor geographical and temporal variations, more subtle variations in the habitat or disturbed and polluted variations of a natural biotope. They will often require greater expertise or survey effort to identify.
 
The primary habitat matrix (Table 3) provides an overview of levels 2 and 3 in the classification. Matrices also been created for each broad habitat, showing the relationship of biotopes and sub-biotopes to key environmental factors (click here for more). For each broad habitat, a hierarchy structure diagram showing the relationship between units at the higher and lower hierarchical levels has been created in Excel™ (click here for more). A fully expandable hierarchical list of the whole classification system, from broad habitats to sub-biotope level, is available here. To make navigation through the hierarchical classification structure easier, a standard colour scheme is used throughout this website, with a single colour for each hierarchical level.
 

Distinguishing and defining types

To ensure consistency across the classification in how types are defined, a working definition as to what constitutes a biotope, enabling its distinction from closely-related types, has been developed. The following criteria are applied:
 
1.         The entity can be distinguished on the basis of a consistent difference in species composition based on:
 
  • different dominant species, some of which (e.g. mussels and kelps) may be structurally important; and

 

  • the co-occurrence of several species characteristic of the particular habitat conditions (even though some of these may occur more widely in other combinations).

 

A combination of both the presence and abundance of the most 'obvious' species in a community is used. Sub-biotopes are often defined using less conspicuous species.
 
2.         It occurs in a recognisably different habitat (but acknowledging that distinct communities may develop in the same habitat through change with time). Sub-biotopes are often defined on the basis of more subtle habitat differences. Some highly subtle differences may be critical in determining community structure (e.g. water circulation/exchange patterns in sealoch basins, oxygenation levels in the water column/sediment, sediment structure other than grain size composition). The separate divisions of habitat factors used in field recording are not necessarily be reflected in the end division of types.
 
3.         It is a recognisable entity in the field, i.e. it is not an artefact of data analysis.
 
4.         The assemblage of species recurs under similar habitat conditions in (at least several) widely-separate geographical locations. Associations of species confined to a small geographical area are considered unlikely to represent a recurrent community (unless the habitat is considered unique), but should rather be treated as a variation of a more widely occurring type.
 
5.         As a working guide the biotope extends over an area at least 5 m x 5 m, but can also cover many square kilometres, such as for extensive offshore sediment plains. For minor habitats, such as rockpools and overhangs on the shore, this 'minimum size' can be split into several discrete patches at a site. Small features, such as crevices in rock or the biota on kelp stipes, are described as features of the main biotope rather than biotopes in their own right. Some entities, by virtue of their extent around the coast, may warrant description despite showing only minor differences in species composition; such types are often treated as sub-biotopes.
 
6.         It is a single entity in the field, although there may be some spatial variation or patchiness from one square metre to the next. Therefore each area of shore or seabed should correlate to only one biotope defined in classification (a 1:1 relationship of field units to classification units). Whenever possible, the surface species characteristics of sediment habitats (their epibiota) are described in association with the sediment infauna as a single entity, rather than treated as separate communities. Note however that the nature of available data has severely restricted the clear association of these two aspects in the classification as they are typically derived from differing survey techniques. Thus in the present classification there remain units defined primarily on the basis of their epibiota or their infauna but which, given further research, will be shown to be the same biotope. Epibiota-derived biotopes may also 'overlay' a number of infaunal biotopes, which are differentiated by more subtle environmental differences, and thus need to be referred to a higher unit in the classification.
 
The following considerations are also taken into account in deciding whether to establish a biotope:
 
  • There is a need to recognise that it is commonplace to have no distinct boundary between two different 'types', but a gradual transition, such that distinction of types is somewhat arbitrary at particular reference points or nodes along a continuum. Additionally, some communities may be largely transitional (in a temporal sense) in nature and whilst recognisable in the field represent a stage between two or more 'stable' biotopes. In some areas, e.g. due to periodic disturbance, a community may be held in a transitional or sub-climactic state for prolonged periods and certain habitats may be so variable that the position of a biotope along a gradient cannot be accurately defined. These factors are of critical importance when assessing typicality of a site to a particular type or its quality or conservation importance.

 

  • Where different associations are shown to occur within the same habitat, they may be spatial or temporal mosaics caused by factors such as grazing, disturbance or chance recruitment. These should be linked together in the classification as, for conservation purposes, it is important to manage or protect the habitat in which several communities may occur over time.

 

  • To produce a practicable working classification it has been necessary at times to be general rather than specific in splitting different types, so that an excessively and unnecessarily complex classification is not developed (bearing in mind the end units that are necessary for practical use).

 

  • Separation of communities can be related to conservation value - does the type add variety (of habitat or species) to a particular stretch of coast. This relates to natural habitats and excludes artificial, polluted or disturbed habitats which should not be considered of high conservation value although they may support distinct communities.