Abstract: In recent years, the importance of four-dimensional (4D) trajectory prediction in air traffic management systems has been gradually increasing. On this foundational technology, the demand for intelligent solutions of tasks, such as conflict detection and resolution, aircraft anomaly monitoring, and management of congested flight paths, has also been rising year by year. The dynamics in airport terminal zones and crowded airspaces are intricate and ever-changing. However, current methodologies cannot fully capture the interactions among aircrafts in these two scenarios. To tackle these challenges, a dual-attention based spatiotemporal graph convolutional network (DA-STGCN) model is proposed. This model reconstructs the adjacency matrix through a self-attention approach, enhancing the capture of node correlations, and employs graph attention to distill spatiotemporal characteristics, thereby generating a probabilistic distribution of predicted trajectories. The experimental results indicate that, compared with the existing main algorithms, the adjacency matrix reconstructed using self-attention mechanism and the graph attention network can be continuously optimized during the network training, thereby better reflecting the potential correlations between nodes, significantly improving the accuracy of 4D trajectory prediction results.