Advances in remote sensors and sensor networks have resulted in the generation of massive volumes of disparate, dynamic, and geographically distributed spatiotemporal data. This has recently been complemented by advances in social media that have also resulted in new types of spatiotemporal information that is contributed by the general public. At the same time, the interest for this information is expanding, as scientists from diverse disciplines and common citizens are interested in the information that can be extracted from such spatiotemporal datasets. However, one could argue that we find ourselves in a data-rich but information-poor environment. The rate at which geospatial data are being generated by diverse sensors and platforms clearly exceeds our ability to organize and analyze them to extract patterns that signify events of importance in our dynamically changing world. Computer science and geoinformatics are collaborating in order to address these scientific and computational challenges, and to provide innovative and effective solutions. More specifically, efficient and reliable data mining techniques are needed for extracting useful geoinformation from large heterogeneous, often multi-modal spatiotemporal datasets. Traditional data mining techniques are ineffective as they do not incorporate the idiosyncrasies of the spatial domain, which include (but are not limited to) spatial autocorrelation, spatial context, and spatial constraints. Extracting useful geoinformation from several terabytes of streaming multi-modal data per day also demands the use of modern computing in all its forms. Thus, we invite all researchers and practioners to participate in this event and share, contribute, and discuss the emerging challenges in spatial and spatiotemporal data mining.