A data description model for reusing, sharing and integrating geo-analysis models

2015/03/15

Journal:Environmental Earth Sciences

Cite as:Yue S, Wen Y, Chen M, et al. A data description model for reusing, sharing and integrating geo-analysis models[J]. Environmental Earth Sciences, 2015, 74(10):7081-7099.

Abstract:Geo-analysis models are necessary tools for understanding various geo-processes and phenomena in Earth environment. Studying, reusing, sharing and integrating geo-analysis models can help researchers solve complicated and synthetic geo-problems with interdisciplinary knowledge, especially for researchers who collaborate with each other to build virtual geographic environments (VGEs). While the integration frameworks of geo-analysis models could permit their practical use, it is essential for model users to prepare data according to the specific requirements of the different geo-analysis models. Model users should invest adequate effort and time into preparing such model data, particularly when employing multi-disciplinary geo-analysis models. This paper proposed a data description model, the Universal Data eXchange (UDX) model that can reduce the effort and difficulties of model data preparation and pre-processing for model users. With the UDX model, researchers from interdiscipline can build a collaborative workspace in VGEs more conveniently. A hierarchical structure was employed in the UDX model for the flexible description of heterogeneous model data, and a set of basic data node types was designed to provide a relatively stable organization method for the various data contents. In the UDX model, the structural format data (e.g., the Shapefile and NetCDF data) and the flexible plain text data content can be described in a uniform way. In addition, model data information can be completely and unambiguously described with the items in the attachment libraries (e.g., unit and dimension library, semantic library, spatial reference library, and data description template library). Furthermore, two different model integration case studies were conducted to prove that various data processing methods and efforts can be accumulated and organized with the designed UDX data processing library.