The goal of the iGreen project is to realize a distributed network of location-based services and knowledge sources, i.e. enabling the integration of decentralized, highly heterogeneous public/private information sources. iGreen also aims to provide a framework for the development of mobile decision support systems. These systems use the iGreen network to support and optimize energy efficient, ecological, environmentally friendly collaborative production processes. Possible domains of application that could benefit of the decision support processes provided by iGreen are for example agriculture, forestry, water management, landscaping, environmental and nature conservation. Since crop production is mostly driven by spatial-temporal information, it was chosen as the first showcase to demonstrate how iGreen can enable mobile decision support systems to access highly heterogeneous information sources. For example, the system can use spatial information, such as public geo-data, domain specific knowledge and private user data to deliver real-time, on site, personal and efficient support. In my talk I give insights into the application domain of knowledge management in agriculture, propose an approach to overcome the existing weaknesses and present our results.