Transparency of Environmental Computer Models

Environmental computer models are considered essential tools in supporting environmental decision making, but their main value is that they allow a better understanding of our complex environment. Despite numerous attempts to promote good modelling practice, transparency of current environmental computer models is limited, which hinders progress in both science and policy making. An important cause is that the structure, meaning and context of environmental computer models is often not clear for other people than the model developers. In the proposed research project we would like to find out whether it is possible to increase the transparency of environmental computer models by making their underlying conceptual model explicit. In preliminary research we identi ed the following challenges: 1) many model developers are mainly focused on the computational instead of the descriptive aspects of computer models 2) many environmental modellers may not consider the lack of transparency a big problem nor do they see computer scientists as natural partners in cooperation. However, we think that both environmental and computer science could bene t from an interdisciplinary or even a totally integrated approach. We expect that ex- perimenting with tools and methods from computer science could teach us important lessons on the practice of environmental modelling and hopefully guide us to this novel, integrated way of performing e-science.