Towards a Generic Framework for the Initialization and the Observation of Socio-environmental Models

Researchers have sought to deal with the complexity of socio-ecosystems including biophysical and social dynamics, and their interactions. In order to cope with this complexity, they need increasingly complex models, whose initialization, and observation are becoming very difficult to implement. However, no generic framework has yet been developed to address this issue. Our objective is a generic framework for specifying and implementing the initialization from numerous heterogeneous data, and the observation producing the desired indicators. The contribution of the paper is the formulation of the initialization and observation as transformations among data and data structures. This formulation allows to use the Model Driven Engineering concepts in order to implement the generic framework and the corresponding domain specific languages. The expected result will be a set of domain specific languages and their tools for specifying and automatizing the whole model exploration process from the initialization to the production of the indicators.