Why do we need to classify habitats?

 
There are two main reasons for using habitat classes: firstly, habitat categories are a way of reducing the complexity of the natural world to make it more understandable. Multivariate data (multiple species together with environmental variables) are synthesised into a manageable number of classes that contain biologically relevant information. A good classification helps interpret such data to produce information and add to our knowledge of the environment. Secondly, habitat classes enable the comparison of like-with-like. For example, we often ‘classify’ bottles by the colour of their glass for recycling purposes; comparing the colour of your bottle with the ‘classes’ makes it easy to deposit it in the correct container. In a marine habitat context, there may be little justification in comparing the species diversity of a kelp forest with a sand plain, but comparing the relative richness of an observation from a kelp forest in an area impacted by human activities with records of a kelp forest from a non-impacted area might be useful in a management context.
 
Typically, habitat types are arranged into a habitat classification scheme that can be defined as a structured system of habitat types (classes), often arranged in a hierarchy, where the types are clearly defined and recur in different geographical places.
 
Classification of samples into habitat types can be undertaken as part of the analysis of data collected within a single study and the resulting classification scheme might only be relevant to that study. Alternatively, data can be fitted to an established classification system that would then enable the local distribution of habitats to be set into a broader geographic context when compared with other maps. In practice, the two approaches are not exclusive since international or national classification systems are derived from survey data collected at local level and should be flexible enough to accommodate new data. Historically, it was the default approach to define a new locally-based classification scheme using the data from a geographically limited area of a single study. This is becoming increasingly inappropriate for a number of reasons:
 
  • The mapped classes may not be relevant to a broader description of ecosystems;
  • The resultant maps are not compatible with those from other studies, especially if the classes are derived in very different ways;
  • The derived datasets (maps) cannot be aggregated with other datasets to make maps covering larger areas without translation of the local classification schemes to a common system;
  • The derived data cannot be used to assess the relative importance of a ‘local’ habitat within a international, national or regional context, which reduces the value of that study to the wider scientific and management community;
  • The adoption of a unified classification scheme appropriate to a broader geographical area provides a context in which to place the results of a particular study and a standard by which the data are interpreted. It ensures that similar habitats can be compared with one another across a broad geographic range. Such standardised interpretation of data is increasingly demanded by end-users, who require knowledge of the relevance of the mapped data within a local study area in a wider regional, national or international context.
 
Standardising existing habitat map data to a single classification scheme and preparing guidelines to promote more standardised interpretation of the new map data were two of the key aims of the MESH project.
 
 

All material variously copyrighted by MESH project partners 2004-2010

| List Access Keys | Mapping European Seabed Habitats (MESH) - Home | Overview | News | Work areas in the MESH Project | Partners | Contact Us | Search | Site Map | Partner Extranet |