Toward an exploratory spatial and temporal data analysis for Time Geography

International audienceThese last 20 years, a large variety of sources of individual based spatio-temporal data have been developed. Classical surveys in the mobility field are completed by GPS tracking techniques dedicated to a better understanding of trips behavior. More recently, a lot of new ICT data collected by cell phone or internet providers are available for the scientific community. Then, the micro-simulation models (LUTI, activity based model) produce an impressive quantity of information to manage and understand. The theoretical expansions of time geographic framework offer interesting perspectives to analyze these new sources. In this context, we need to develop specific software in order to give operational framework to the theoretical expansions of the time geographical concepts. At this end, we could improve the knowledge of travel behavior, accessibility analysis or activity pattern. During the last 3 years, we develop an exploratory spatial and temporal data analysis tool dedicated to the time geographical dataset. Named Geographer, this geovisualization tool is based on 3 principles. First, the function permits to link statistics maps and graphs according to the work of J W Tukey and later L Anceline. Secondly, space-time interactions are considered with the means of voxels. Thirdly, we can generate actogrammes in order to observe chronicles for individual or groups available in a dataset. Two case studies will illustrate the potential of Geographer. The first example considers a specific survey with de combination of daily and residential mobility, whereas the second case will present the results of simulated data