Contemporary Multidisciplinary Approaches to Coastal Classification and Environmental Risk Analysis
Keywords:Coastal classification, typology, GIS, coastal management, vulnerability/sensitivity
Coastal classification or typology based on multidisciplinary data and multivariate analysis has recently emerged as a tool in coastal management. In this paper, eighteen published accounts of coastal classification procedures are reviewed in order to determine the reasons for such an increase, the variability between different approaches and the utility of each approach. The increase in use of such approaches to coastal classification may be linked to technological advances and widespread use of Geographic Information Systems (GIS). The main differences identified between the indices are in terms of scale of application, variables included, mode of analysis, mode of presentation and the nature of the risks being assessed. While many authors drew attention to limitations imposed by lack of availability of data, in general it was concluded that few indices adequately considered the physical basis for interaction between variables used in the classification procedure. In particular, while most indices recognise the need for socio-economic data, few were able to adequately incorporate such information. Those indices with the highest utility in risk assessment are considered to be those in which (a) the nature of potential perturbation and (b) the issues of management concern were clearly defined. Those in which neither is adequately defined are likely to be of use mainly as databases. A potential stepwise approach to development of specific coastal classification indices is outlined in which user needs and interrelationships between variables are examined in the planning stage. We recommend development of a GIS based hierarchy of coastal classifications on varying spatial scales in which resolution may be adapted and variables combined differently according to specific aspects of management concern at different spatial management levels.