Approaches to modeling take many forms. The mathematical, computational, and encapsulated components of models can be diverse in terms of complexity and scale, as well as in published implementation (mathematics, source code, and executable files). Many of these systems are attempting to solve real-world problems in isolation. However an increasing trend in big data science is in the long-term interest in allowing greater access to and preservation of models and their data, and to enable simulations to be combined in order to address ever more complex issues. Model-driven approaches, markup languages, metadata specifications, and ontologies have emerged as pathways to greater interoperability. Domain-specific modeling languages allow for a declarative development process to be achieved. Metadata specifications enable coupling while ontologies allow cross platform integration of data. The goal of this workshop is to bring together researchers from across scientific disciplines whose computational models require interoperability. This may arise through interactions between different domains, systems being modelled, connecting model repositories, or coupling models themselves, for instance in multi-scale or hybrid simulations. These interactions requires the ability to interoperate, and is characteristic of big data applications that are becoming more prevalent, even in the sciences. The outcomes of this workshop will be to better understand the nature of multidisciplinary computational modeling and data handling. Moreover we hope to identify common abstractions and crosscutting themes in future interoperability research applied to the broader domain of scientific computing.