engleski

Tracing compositional variability and source rocks of beach sands along the Croatian coast using cluster analysis

The eastern Adriatic coast of Croatia is characterized by an extremely indented hard rock coastline made up predominantly of carbonate rocks. However, where sufficient amounts of clastic material are available sandy pocket beaches occur. These beaches represent a delicate coastal environment, one which is easily subject to destructive processes if imbalance is created between the supply of detritus and erosion. Studies have shown that considerable variability in composition may exist among closely spaced beaches (e.g. Lužar-Oberiter et al., 2008), advocating the necessity of a localized approach when considering management issues such as beach nourishment. We have compiled previously published and new unpublished quantitative heavy mineral data from beach sands and potential source rocks along the Croatian coast. In order to demonstrate the variability in beach sand composition, identify and explain regional similarities and link beach material to source rocks, hierarchical cluster analysis was applied on the compositional variables after having transformed them using a logistic transformation (Hartigan, 1975). The number of clusters was determined using Pseudo-t2 statistic and cubic clustering criterion (Milligan & Cooper, 1985). Profiling of the clusters and identification of key heavy minerals is facilitated by a decision tree. Analysis was performed using SAS v9.2 and JMP v8 software. The derived clusters distinctly mirror geographic regions along the coast from which the samples come from. In many cases there is a close relation between beaches and rocks in the nearby hinterland, making it possible to identify specific source rocks supplying individual beaches, but also raises questions on mixing of material from more than one source. The quantitative tools offer a better understanding of how beaches are supplied with sandy material along the Croatian coast, which is essential if their aesthetic, recreational and economic values are to be preserved. It is our future goal to further build and expand the database of Croatia’s beach environments to include and integrate other quantitative data regarding composition, grain size, as well as relevant environmental data. Information derived from such a database should benefit planning of sustainable development along the Croatian coastline. Hartigan JA (1975) Clustering Algorithms. John Wiley & Sons, New York. Lužar-Oberiter B, Mikulčić Pavlaković S, Crnjaković M and Babić Lj (2008) Variable sources of beach sands of north Adriatic islands: examples from Rab and Susak. Geologia Croatica 61, 379-384. Milligan GW and Cooper MC (1985) An Examination of Procedures for Determining the Number of Clusters in a Data Set. Psychometrika 50, 159–179.

Croatia; modern beaches; sand; provenance; heavy minerals; cluster analysis

nije evidentirano