Corals in the Galápagos present diverse reef configurations, ranging from biogenic coral reefs to coral communities growing on rocks and sand. These corals have experienced decades of disturbances, including recurring El Niño events. However, traditional ecological methods have limited capacity to describe coral demographic trends across large spatial scales. 
We surveyed seven reef sites across the archipelago using underwater photogrammetry, developed new methods for 3D annotation and fractal dimension calculation, and resurveyed the same sites one year later. However, comparing 3D maps over time requires a complicated set of operations. In particular, aligning 3D models of dynamic reef sites is challenging because different components of the scene move in different directions and at different rates. Moreover, analyzing 3D maps for coral community structure requires laborious image annotation. To address these challenges, we developed methods for differential alignment and model-assisted labeling. Instead of aligning the complete reef models, we divided them into smaller sections and aligned each section separately. This enabled us to transfer the locations of 3D survey points between years and conduct direct pointwise comparisons. To annotate the same 3D points over time, we used classifications generated by a deep neural network and validated by a human annotator, resulting in substantial time savings. Our results show that differential alignment is effective for registering 3D reef models over time and that the classifier was useful in expediting the annotation process although it reached precision values of approximately 60% for key classes. Our findings reveal variation in coral cover, diversity, and reef structural complexity across the archipelago. Over time, structural complexity generally decreased, while coral cover showed site-specific trends. These patterns warrant further research to determine how spatial and temporal changes in reef structure and community composition may affect coral resilience and recovery. We released two unique datasets containing spatial and temporal photogrammetric surveys of Galápagos coral reefs, including 3D models and annotated images. Together, this work establishes an important baseline and a practical framework for long-term monitoring, research, and conservation in the Galápagos and beyond. It can support evidence-based management and advance our understanding of coral reef resilience and recovery.