<p>In a dynamic underwater environment, reliable data transmission and energy conservation of the sensor nodes remain a challenging task. With this aim, an energy-efficient dynamic cluster-based routing algorithm for an underwater acoustic sensor network is proposed in this article. Unlike traditional cluster-based routing protocols, the proposed method employs an unequal clustering methodology and divides the deployed network into multiple groups based on node communication range and current position. Each node evaluates and compares its fitness index to compete for the cluster head. From each group, the most suitable cluster head is elected in a non-probabilistic way to ensure optimal performance of the network. Due to the underwater node mobility and changing network configuration, a flexible cluster formation methodology is proposed. A robust multi-hop routing methodology, along with adaptive data aggregation, is also adopted to relay the data packets through the most energy-efficient routing paths. This unique coordination of non-uniform and flexible clustering, robust multi-hop routing, and dynamic aggregation shows increased network lifetime and throughput. The performance of the proposed routing protocol has been comprehensively evaluated through multiple simulations and benchmarked against existing underwater clustering schemes. Relative to the traditional methods, the results show that the proposed routing scheme extends the network lifetime in terms of the total number of alive nodes and achieves a higher packet delivery ratio. Moreover, the protocol demonstrates robustness under varying network conditions with overall balanced energy consumption.</p>