ModeleR: An enviromental model repository as knowledge base for experts

In this paper, we present the development of ModeleR, a repository of models accessible from the web, which enables the user to design, document, manage, and execute environmental models. The technique and features offered can be applied to any scientific context. Based on the development of its ontology, a metadata system has been established to document the modeling process. The set of models managed from ModeleR reflects the knowledge base of the experts of the system, allowing other experts to reuse, replicate, and delve deeper into the existing models in the repository. Different levels of integration have been included, from the conceptual description of the model to the process needed to execute a model from a remote server, acting as an execution engine through the use of workflow managers. In this paper, we present the problems encountered as well as the solutions reached on developing a prototype of ModeleR set up for ecosystem research and an environmental monitoring lab.

► We have developed a repository of environmental models. ► We have analyzed and assessed other different approaches to the problem. ► We have designed and implemented a metadata schema for environmental models. ► The ModeleR allows a different degrees of integration for each model. ► ModeleR is a knowledge base for modelers that want to document and execute workflows.