Modeling typical domestic water users in the upper Danube catchment area (GLOWA-Danube/Environmental Psychology)


Factors like global environmental and climate change, changes in land use, pollution of freshwater resources or a changing daily per-capita water use affect the overall freshwater availability as well as the fresh water quality. As a response, Germany’s Federal Ministry of Education and Research (BMBF) launched a pilot program known as GLOWA (Global Change and the Hydrological Cycle), which contains a total of five GLOWA-projects (

One of the GLOWA-projects is GLOWA-Danube, which started in 2000 and is now in its ongoing third phase (2007 – 2010). The aim of GLOWA-Danube is to develop the decision support system DANUBIA, and to use this integrative tool to examine the sustainability of future water use along the upper Danube. GLOWA-Danube consists of an interdisciplinary team of 11 disciplines: hydrology, remote sensing, water management, meteorology, economy, agricultural economy, ecosystem research, glaciology, tourism research, computer science, and environmental psychology. The environmental psychology team - located at the Center for Environmental Systems Research at the University of Kassel - has developed the agent-based model of water use behavior of households and the service sector. Its aim is to calculate the drinking water use under changing climatic and societal conditions.

The objective of the environmental psychology part of the project is to model in a spatially explicit way all households in terms of their water use decisions, their subjective perception of water-related risks, and their acceptance of different water-saving innovations as well as water-related political decisions.

Based on this analysis, social science scenarios are created under conditions of global climate change to assess the potentials of conflict and risks of acceptance.
The Investigation area is the Upper Danube Basin, covering an area of about 76,000 km² at gauge Passau. There are approximately 10 million people living in the area.


The agent-based DeepHousehold model used in the environmental psychology part has been implemented in JAVA. The model contains both current theories about environmental behavior (consumption and conservation behavior) and empirical data, based on extensive research.

The DeepHousehold model is a multi-agent based simulation model, which reconstructs habitual water use as well as deliberate decision making on the basis of structural and process aspects of individual decision making. This means that the DeepHousehold model contains decision-making entities called actors.

There are 5 different types of actors, based on the Sinus-Milieus®. Each actor (milieu) has its own preferences, i.e. the actors perceive the variables upon which they base their decisions as varying in importance: Whereas it might be very important for one actor to behave in an environmentally friendly way, another actor bases his decision mostly upon financial aspects.

The DeepHousehold model implements theories of cognitive and social psychology that describe behavior under different conditions. If there are no exceptional events, the actors behave like they always do (habit). Otherwise (e.g. an exceptionally high temperature or water scarcity), they evaluate their behavioral options based on bounded rational choice.

Systematic analyses of scenario runs under different climatic and societal conditions have shown that the DeepHousehold model produces logical and plausible results. Moreover, it has been shown that the model can be employed to assess the likely consequences of policy measures such as information campaigns or subsidization policies. The model has been validated successfully against statistical domestic water demand data on the municipal level.

Homepage of GLOWA-Danube: GLOWA-Danube

GLOWA homepage: GLOWA

GLOWA brochure of the German Federal Ministry of Education and Research (BMBF): Brochure

Online-Atlas of GLOWA-Danube: Online-Atlas