On the integration of database systems and computational support for high-level modelling of spatio-temporal phenomana1
Terence R. Smith · Innovations in GIS · 2024
A modelling and data-base system (MDBS) provides computational support that facilitates the iterative development of scientific models in data-intensive and computationally intensive applications in various areas of scientific application. We describe and exemplify core elements of an MDBS that is based both upon a simple characterization of scientific activity and upon the results of joint research with a group of EOS earth scientists. The MDBS that we describe is focused in particular on the representation and manipulation of large numbers of complex, spatio-temporal entities that are modelled in terms of an advanced and comprehensive data model and a high-level database and modelling language (MDBL). The data model is based on a simple characterization of scientific activity that captures the essence of scientific modelling and database activities in terms of four fundamental sets of activities. These activities are defined in relation to large lattices of representational domains (R-domains) that are employed to organize scientific knowledge, data and algorithms. The data model separates the specification and representation of R-domains of elements, and allows multiple representations that are explicitly controlled by the user. Datasets and transformations are organized in terms of an extensible lattice of abstract and concrete R-domains. In particular, the data model allows the easy definition and manipulation of large numbers of domains of spatio-temporal elements. MDBL is a high-level, largely declarative language that is also based on our characterization of science and provides easy expression of the four classes of scientific activities in relation to R-domains. MDBL permits the easy integration of modules that support scientific modelling and database activities. We are implementing a system in which support for the data model and language is integrated with support for a distributed database of large, spatio-temporal dataseis.