Risk Management at the Science–Policy Interface: Two Contrasting Cases in the Field of Flood Protection in Germany

Abstract

This paper concerns the way in which the scientific debate on climate change and new risks is being adopted as a basis for political decision making. How does the crucial risk issue ‘diffuse’ into policy, which in turn has to give the public account of the risks of climate change? In discussing the science–policy interface, reference is made to the debate on blurring boundaries between science and policy making. Here, heterogeneous and often competing discourses come into play. This makes discourse analysis an appropriate and well-accepted tool in both conceptual and theoretical aspect. However, which of the competing discourses wins the day can hardly be explained in a discourse analytical way, due to its constructivist bias. Case studies provide some evidence that complex understanding can be obtained when discourse analysis is framed by a more realistic approach, such as Kingdon's policy window approach.

Two cases are presented. Although representing overlapping policy domains, the risk management differs considerably. In both the state proves to be the pivotal actor. In the first case, on coastal protection in northern Germany, the administrative officers in charge try to transform and to curtail the risk issue and its emphasis on uncertainty in a way that makes it compatible with their own safety discourse, thus generating a scientific–administrative hybrid. The second case, a newly enacted political strategy on riparian flood protection, draws explicitly on uncertainty and risk, thus transferring and integrating the issue firsthand into the political–administrative system. Taking a governance perspective, the explanation refers to different steering contexts in terms of institutional settings, actor constellations, political framings and natural extreme events.

Keywords:

• Climate change
• science-policy interface
• governance
• policy windows
• risk management
• discourse analysis
• Germany

Notes

1. For further information relating to the two cases see www.krim.uni-bremen.de and www.innig.uni-bremen.de.

2. Their work is based on their experience in the context of the FAO and refers to the global nitrogen cycle.

3. The empirical basis of the finding consists of 50 interviews conducted in 2003 and 2006. They included officers in charge of different subtasks in the realm of coastal and flood protection at all levels (local, district, federal state) and the involved political entities in terms of geography: the federal states, so-called Länder, Niedersachsen, Bremen, Hamburg. In addition, political actors (parties represented in parliaments, NGOs) were interviewed as well as further non-state actors, such as dyke associations. The interviewees—and further analysed documents— were selected in a way that ensures a complete coverage of the political–administrative system of coastal and flood protection in the area. In analysing these sources, Keller (1998, 2004) was referred to.

4. http://www.artec.uni-bremen.de/files/projekte/Endbericht_KRIM.pdf; subsequently quoted as Lange et al. (2005). The area under investigation comprises the coastline of the mainland around the Jade Bay and the respective administrative districts (Landkreise) Friesland, Wesermarsch and Cuxhaven, including the island of Wangerooge and the estuary of the River Weser up to the city of Bremen. Large parts of the area are situated at up to 2 m below the mean high tide margin. Because of this topology and because it consists largely of supple marine sediments, the coastal area is prone to erosion and flooding.

5. Coastal protection in Germany is planned and executed at three administrative levels (nation state, federal states, local/municipality level) covering four different federal states—each of them following more or less different planning concepts including different forms of providing information and allotting responsibility to citizens living in flood-prone areas (Lange et al., 2005, pp. 22 ff.).

6. In detail, the procedure is actually more sophisticated (see Lange et al., 2005, p. 31.)

7. In the last decade similar circulation patterns caused the Odra flood in 1997 and the Vistula flood in 2001 (Nachtnebel, 2003).

8. See note 5.

9. Exceptions are possible if nine closely defined requirements are met. All of them have to be fulfilled completely in every individual case. They include that the municipality concerned has no alternative for human settlement development, that no lives are at risk and no significant property damage is to be expected, and that the structure of new buildings is adapted to flood events (BMU, 2005).