A web service for the dynamic linkage and visualisation of multivariate spatiotemporal information

In spatial health research, it is necessary to not only consider the spatial and temporal distributions of diseases, but also external factors that influence the disease, such as environmental and socio-economic factors. In this paper, we propose a method for dynamically linking, and subsequently visualising multivariate spatiotemporal data. The aim of the method is to enable a user to create new analysis data sets by combining the output of multiple web services to facilitate the exploration of data relating to factors that may influence, or are related to, a disease or group of disease under consideration. By creating a generic layer store, termed a visualisation object, the approach utilises semantic web concepts in order to populate the layer store with vector data from map data layers, and the results of processing web services. This approach represents a dynamic, query based approach to web GIS that focuses on enabling access to data and data exploration. A number of visualisations, leveraging both the multivariate and temporal nature of the visualisation object, were developed to explore the information visualisation aspect of the visualisation object. While complex visualisations are possible using this approach, thematic maps for a single epoch can be generated and viewed within a standard web GIS using a previously proposed dynamic web map server, thus enabling the caching of the results of complex processing results for visualisation and data access.