To optimize the total energy consumption for event-driven sensor networks, an adaptive routing algorithm based on data aggregation is proposed, which clusters nodes within event areas in a distributed manner with low control overhead, and builds an approximation Steiner Tree based on events in the help of the routing aggregation center to decrease the amount of transmission for data packets and control packets effectively. Theory analysis and experiments show that the proposed algorithm has better overhead on routing structure building and maintenance, optimizes the data aggregation ratio, achieves energy-efficient data collection, and promotes network performance eventually.