Research Article: Bayesian Belief Network-based assessment of nutrient regulating ecosystem services in Northern Germany

Date Published: April 30, 2019

Publisher: Public Library of Science

Author(s): Sabine Bicking, Benjamin Burkhard, Marion Kruse, Felix Müller, Stoyan Nedkov.


This study aims to assess the potential supply of the ecosystem service (ES) nutrient regulation on two spatial scales, the federal German state of Schleswig-Holstein (regional) and the Bornhöved Lakes District (local), exemplarily for the nutrient nitrogen. The methodology was developed using the ES matrix approach, which can be applied to evaluate and map ES at different geospatial units such as land use/land cover classes. A Bayesian Belief Network (BBN) was constructed in order to include additional spatial information on environmental characteristics in the assessment. The integration of additional data, which describes site-specific characteristics such as soil type and slope, resulted in shifted probability distributions for the nutrient regulation ES potential. The focal objective of the study was of methodological nature: to test the application of a BBN as an integrative modelling approach combining the information from the ES matrix with additional data sets. In the process, both study areas were assessed with a regional differentiation with regard to the predominant landscape types. For both study areas, regional differences could be detected. Furthermore, the results indicate a spatial mismatch between ES demand and supply of the nutrient regulation potential. Land management and agricultural practices seem not to be in harmony with the spatial patterns of the environmental characteristics in the study areas. The assessment on the local scale, which comprised higher resolution input data, emphasized these circumstances even more clearly.

Partial Text

Research on the interrelations between human activities and the environment is of key importance for our society. Increased understanding of the environment and of the effects of our behaviour on the environment will support sustainable policy and decision making. In particular, decision making with regard to spatial planning, land and resources management and agricultural practices has to be founded on precise information in order to support sustainable land use. In this regard, the concept of ecosystem services (ES) is a valid assessment approach. The concept commonly differentiates between three ES categories: regulating, provisioning and cultural ES, all of which contribute to human well-being (i.e. [1–6]) ES analyses can be used to support spatial planning and attaining sustainable land management [7,8]. The relevance of this approach is consolidated also in European policy. Through the Biodiversity Strategy 2020, the European Union called upon its member states to map and assess the states of their ecosystems and the services they provide (Target 2 Action 5 of the Biodiversity Strategy; see [9]). This demand emphasizes the need for simple and applicable methods for ES assessment and mapping. In several previous studies, the assessment and mapping of the ES supply and demand has been based on the ES matrix approach by Burkhard et al. ([2,10] also in [11–13]). The ES matrix has been used to distinguish, amongst others, ES supply for different land cover types. CORINE land cover data [14] from the European Union has often been selected as underlying data set [10,15].

This section comprises information on the study areas followed by some general knowledge about nutrients focusing on nitrogen and the ES nutrient regulation.