Long-term surface elevation changes in German North Sea salt marshes

We analyzed long-term (19–21 years) surface elevation changes of salt marshes along the German mainland North Sea coast in a large and consistent data set (423 plots). Our aim was to determine the influence of 12 predictor variables, e.g. elevation, micro-topography and management data. For each plot, we measured elevation above sea level, once 1988–1990 and again 2009. The calculated surface elevation change per year (target variable SEC) was +0.6 cm yr−1 on average. 85.8% of the plots showed positive SEC values. Although 11 predictors were significantly correlated to SEC, only five driving on different scales were included by the tree fitting algorithm when using regression tree analysis. The resulting tree model with 11 terminal nodes explained 45.7% proportion of the variance in SEC values. Plots tended to high SEC values when a) lying low in relation to mean high tide, b) being located in the northern regions of the study area, and c) lying in drainage ditches. d) Grazing management did account for SEC variance only in northern regions. e) SEC was enhanced on non-grazed plots lying high in relation to mean high tide. We concluded that most of the studied salt marshes may outpace a predicted sea-level rise of about 0.1–0.2 cm yr−1, but barely half of the marsh areas might follow a sea-level rise of 0.6 cm yr−1. Regression tree analysis is a useful approach for studying changes in elevation of salt marshes when considering various variables.

Figure optionsDownload high-quality image (133 K)Download as PowerPoint slideHighlights
► Regression tree analysis is useful for studying elevation changes of salt marshes.
► Surface elevation changes (SECs) tended to high values especially in northern regions.
► SEC tended to high values on plots lying low in relation to mean high tide.
► In some regions, surface elevation was enhanced on non-grazed plots.
► Most of the studied salt marshes may outpace minor predicted sea-level rise.