Time series of metals in mosses and their correlation with selected sampling site-specific and ecoregional characteristics in Germany.

GOAL, SCOPE AND
BACKGROUND: The UNECE Heavy Metals in Mosses Surveys provide exposure data (Predicted Environmental Concentrations--PEC) for ecotoxicological risk assessments by measuring the accumulation of several metals in naturally growing mosses throughout Europe. Germany took part in the monitoring campaigns 1990, 1995 and 2000. The article deals with the description and application of the extensive methodical design developed to investigate the following three hypotheses: 1. The metal accumulation in mosses measured at up to 1000 sites may be geostatistically extrapolated from the sampling sites to ecoregions in order to transform the site-specific PEC values to surface PEC values. 2. The metal specific measurement values may be aggregated to metals integrating accumulation indices. 3. The ecoregional situation as well as sampling site-specific topographical features such as altitude, slope gradient or direction influence the accumulation of metals in mosses.
METHODS: The methodical design integrates several data sources as well as statistical analysis and GIS techniques: The site-specific data on metal accumulation are geostatistically transformed to valid surface data on metal accumulation. The sampling site-specific measurement data and the estimated surface data on the accumulation of As, Cd, Cr, Cu, Fe, Hg, Ni, Pb, Sb, Ti, V and Zn are aggregated to integrative metal accumulation indices by means of percentile statistics. The metal-specific estimated data and the metals integrating accumulation indices are correlated with site-specific data on altitude, slope gradient and direction as well as with the ecoregional conditions derived from a multivariate ecoregionalisation. The mean multi-metal accumulation index for the whole of Germany over the ten year period from 1990 to 2000 was related to the accumulation indices within each of the ecoregions and each of the monitoring campaigns 1990, 1995 and 2000. In addition to this ecoregionalisation of the temporal trends of metal accumulation, the hot spots of accumulation are mapped. RESULTS AND DISCUSSION: The Heavy Metals in Mosses Survey 2000 reveals low metal concentrations in most European countries. In Germany, all metals decreased about 22% (Cu) to 64% (Pb) from 1990 to 2000. Mapping concentrations as dot maps deliver the most unbiased and detailed picture of the spatial structure of the metal accumulation in mosses. This information, detailed with respect to metal species and sampling site, is spatially generalized by means of geostatistical estimation. According to the cross-validation, the precision of the geostatistical estimation is quite good so that the extrapolation does not bias the spatial structure, but helps to clarify it. By percentile statistics, an ordinally scaled multi-metal accumulation index is calculated and spatially differentiated over time in terms of ecoregions which were calculated by means of Classification and Regression Trees (CART). The integrative statistical analysis reveals declinations of up to 80% of the multi-metal accumulation in some of Germany's ecoregions from 1990 to 2000.
CONCLUSION: The monitoring of heavy metal bioaccumulation by means of mosses is an effective and cheap method for the analysis of the environmental concentrations of metals accumulated in terrestrial ecosystems. Geostatistics, percentile statistics, cross-tabulations and ecoregionalisation serve well to clarify the spatial and temporal trends in the large data sets coming out of the UNECE Heavy Metals in Mosses Surveys: By combining these statistical methods and integrating them into a geographical information system (GIS), they allow to detect and map the spatial-temporal trends of metal accumulation, to calculate metal-integrating accumulation indices, to describe temporal trends of metal accumulation within ecoregions, and to detect and map hot spots. RECOMMENDATION AND PERSPECTIVE: The spatial and temporal trends of metal accumulation in mosses should be linked with deposition data and data on the vitality of forest ecosystems, as well as with data on human health. Statistical valid interspecies calibration is needed. The integrated methodology of data evaluation presented in the article at hand should be implemented in the future UNECE Heavy Metals in Mosses Surveys. The hot spot areas should be investigated with special care in the 2005 survey to prove if the PEC values of the metals exceed the Predicted no Effect Concentrations (PNEC values). In addition to the metals, the 2005 survey should monitor the nitrogen accumulation in mosses.