Mapping correlations between nitrogen concentrations in atmospheric deposition and mosses for natural landscapes in Europe

Abstract

Recent investigations proved that nitrogen (N) concentrations in mosses are primarily determined by atmospheric deposition. The correlations are country- and N compound-specific and agree well with spatial patterns and temporal trends across Europe as a whole and in single European countries. This study investigates whether correlations between the concentration of N in atmospheric deposition and mosses within the units of an ecological land classification of Europe can be established. To this end, N measurements from the 2005 European moss survey and modelled N atmospheric deposition in 2005 were intersected with a map of European landscapes. Then, considering minimum numbers of sampling sites required across Europe, in single European countries and within the landscapes of Europe and accounting for spatial auto-correlation, the correlations between the N concentration in mosses and corresponding deposition were calculated and mapped for each of those landscape units containing moss sampling sites. Using an example of one landscape with positive correlation and one landscape with no correlation between N concentrations in deposition and in mosses, influencing factors were ranked based on investigating the multivariate interactions between moss concentrations and, amongst others, atmospheric deposition, land use, elevation or moss species by classification and regression trees. From this study it could be concluded that the numbers of sampling sites within Europe and most participating countries as well as within most of the landscapes covering Europe are sufficient. Spatial patterns of correlations between the atmospheric N deposition and N concentration in mosses could be proven to vary across the landscapes of Europe. Where clear positive correlations between N concentrations in deposition and mosses exist in landscapes, multivariate ranking identifies the deposition as main influencing factor. In cases with no correlation between deposition and N concentrations in mosses, other factors such as e.g. moss species collected may be of importance. Therefore, mosses were proved to serve as biological indicators for atmospheric depositions and ecologically defined land classes could be identified as more complex indicators which allow relating exposure monitoring with effects assessment.