Data aggregations from Sensors to a sink in wireless sensor networks (WSNs) are typically characterized by correlation along the spatial, semantic, and temporal dimensions. Exploiting such correlation when performing data aggregation can result in considerable improvements in the bandwidth and energy performance of WSNs. For the sensors-to-sink data delivery, we first explore two theoretical solutions: the shortest path tree (SPT) and the minimum spanning tree (MST) approaches. To approximate the optimal solution (MST) in case of perfect correlation among data, we propose a new aggregation which combines the minimum dominating set (MDS) with the shortest path tree (SPT) in order to aggregate correlated data. To reduce the redundancy among correlated data and simplify the synchronization among transmission, the proposed aggregation takes two stages: local aggregation among sensors around a node in the MDS and global aggregation among sensors in the MDS. Finally, using discrete event simulations, we show that the proposed aggregation outperforms the SPT and closely approximates the centralized optimal solution, the MST, with less amount of overhead and in a decentralized fashion.