This chapter presents a comprehensive approach to mining geo-referenced frequent patterns and geo-referenced periodic-frequent patterns by integrating spatial and temporal dimensions in transactional databases. Geo-referenced frequent patterns focus on identifying sets of spatially proximate items that occur frequently, while geo-referenced periodic-frequent patterns extend this by considering periodicity in their occurrence. Efficient search techniques such as the anti-monotonic property and neighborhood-aware depth-first search are utilized to manage the large search space inherent in these tasks. The chapter also introduces algorithms from the PAMI library, including FSP-growth and GPFPMiner, which facilitate the discovery of these patterns. Real-world applications, such as environmental sensor networks, can benefit from the insights gained through these mining techniques, enabling a better understanding of spatial-temporal dynamics. Practical Python implementations are provided to demonstrate how to mine, save, and analyze geo-referenced patterns in large datasets. This chapter highlights the importance of combining spatial and temporal analyses for improving decision-making and system optimization in various domains.