Synthesis of Sampling Site and Metal Specific Bioaccumulation Data by Classification and Regression Trees

Abstract

The UNECE Heavy Metals in Mosses Surveys provide exposure data (PEC) for ecotoxicological risk assessments by measuring the accumulation of metals in mosses. The article focuses on the synthesis of data on the bioaccumulation of metals in the German campaign 2000 by Classification and Regression Trees (CART). Within the framework of the European Metals in Mosses Surveys, moss samples were taken at 1028 sites. At each of these sites the mosses were analysed with respect to 20 metal elements. The sampling, the ecological characteristics of the sampling sites, and the chemical analysis are documented in a metadata base. An ecoregionalisation of Germany was calculated with data on natural vegetation, altitude, soil texture and climate by means of CART. The ecoregions are intersected with site specific multi-metal bioaccumulation indices calculated from the measurement data in a GIS environment. These indices are calculated by means of percentile statistics and integrate the concentrations of Cr, Cu, Fe, Ni, Pb, Ti, V and Zn in terms of ordinal scaled indices for atmospheric metal accumulation. To synthesize metadata, multi-metal bioaccumulation indices and the ecoregionalisation CART-models are calculated for each campaign. The presented approach helps to assess the multivariate correlations of metal bioaccumulation with site specific and ecoregional characteristics in a comprehensive and holistic manner over space and metal species. This is of importance for the environmental reporting within the framework of the national and international environmental information systems.