Emergent vulnerabilities: exploring the role of drought for increasingly diverse groundwater conflicts in Germany

Historically, groundwater resources have been perceived as inexhaustible in Central Europe by policy-makers and the general public. However, recently increasing drought periods and user groups with competing interests caused conflicts about the usage of and access to groundwater resources. Groundwater-related conflicts, defined here as social issues resulting from divergent viewpoints among diverse stakeholders, have been extensively examined in regions with an extended history of water scarcity. Yet, there is limited research on the emergence of groundwater-related conflicts in Central Europe and the role of recent drought events in shaping these. Here, we study the emergence of groundwater-related conflicts in Germany since 2000 using a text-mining approach. Specifically, we investigate four research questions: (i) how are groundwater-related conflicts characterized, (ii) which influential stakeholders are shaping these conflicts, (iii) what are the spatio-temporal patterns of these conflicts and (iv) how do drought events and different socio-economic factors influence their occurrence? To address these questions, we use machine learning and text-mining techniques on more than one million newspaper articles to develop a spatio-temporal database of conflicts. We also extract and categorize involved stakeholders using a named entity recognition algorithm. Then, we use statistical modeling to link the occurrences of groundwater conflicts with drought indices and other additional explanatory variables. Our results reveal the growing diversity and geographical spread of groundwater-related conflicts in Germany. Also, our results shed light on the role of the recent drought events’ influence on conflicts. Our findings contribute to mapping the evolving landscape of groundwater-related conflicts in Germany and the effects of drought events. The proposed methods have the potential to enable large-scale studies of environmental conflicts using vastly available text data.