Enriquecimento semântico de modelos ecológicos baseados em raciocínio qualitativo : um estudo de caso sobre a dinâmica de comunidades de vegetação lenhosa no cerrado brasileiro

The use of models has become increasingly common, especially in scientific studies, but also in society in general as an important planning tool and scenario analysis among other objectives. However, the structures used in the modeling do not incorporate meanings and contents. These limitations led to the development of tools that allow modelers to explore semantic relations in the models. Ontologies and other semantic technologies are increasingly being used to support the modeling systems natural activities and to enhance its accuracy and consistency. The present work aims to semantically enrich models based on Qualitative Reasoning in the area of Cerrado Ecology. Thus, we seek to relate data bases of woody plants in the Brazilian Cerrado communities obtained from scientific studies, meteorology, environmental variables, geographic information (maps) and conceptual models of Qualitative Reasoning. Two domain ontologies, Cerrado Concepts and Wood Plant Dynamics Ontology (CCON) and Fire Ontology (Fire), were developed to represent scientific knowledge about the Cerrado vegetation ecology focusing on dynamic aspects under different fire regimes. Data sets, made available by agencies and institutions of the Brazilian government, and the data obtained from the scientific literature were transformed to RDF (Resource Framework Descrition) format using LOD Refine (Linked Open Data Refine). Moreover, qualitative models were developed to represent the structure en dinamics of Cerrado vegetation communities which the disequilibrium hypothesis of tree-grass coexistence in that biome is exploited. The models highlight the role of fire and competition relationships over resources between wood and grass layers, that allows their coexistence in Cerrado sensu stricto environments. Geographical links have been established between the occurrence of species and environmental variables with qualitative reasoning ecological models, that can be used for simulations and support predictions and explanations in the behavior of specific communities of woody plants of the Cerrado. Ongoing work aims at validation of qualitative models, improvement of the current and construction of new ontologies, expansion of databases to explore different hypotheses about the structure and functioning of Cerrado communities.