To address the challenges of searching for and handling dynamic targets in vast, unknown waters, existing cooperative strategies for unmanned systems still exhibit deficiencies in managing platform heterogeneity, as well as the global and dynamic nature of task allocation. To solve this problem, this paper proposes an efficient cooperative framework for Unmanned Aerial Vehicles (UAVs) and Unmanned Surface Vehicles (USVs) based on hierarchical decision-making. The core of this framework is an asymmetric search strategy based on a dynamic probability heatmap, which fully leverages the high mobility of UAVs for wide-area exploration and the persistence of USVs for regional control. Concurrently, a periodic global replanning mechanism for task allocation is designed. Its cost function incorporates the future task burden of the platforms and can intelligently combine and optimize adjacent tasks to overcome the myopic shortcomings of traditional methods. Test results across various simulation scenarios demonstrate that the proposed cooperative framework achieves significant improvements in target discovery efficiency, task disposal speed, and system robustness, providing an effective solution for the application of heterogeneous unmanned systems in complex environments.