Recently launched Earth observation satellites, which deliver between 8 and 10TB of image data per day, open emerging opportunities in domains ranging from environmental monitoring to urban planning and climate studies. However, domain-oriented applications require image metadata to be enriched with data coming from various sources (either static or dynamic), in order to support decision making on the observed areas. The integration of heterogeneous data highly relying on spatio-temporal representations raises a major challenge. We present a semantic approach to support data integration thannks to spatio-temporal relations between image metadata and various open data sets. We propose a vocabulary that specializes standards (like SOSA, GeoSPARQL) as well as a process to map and integrate heterogeneous geo-spatial data sets. This process relies on image tiles, representing a fixed area of a grid associated with the Earth's surface, to handle data with a fixed spatial component. The temporal relationships are calculated on the fly based on temporal topology.