【摘要】 农业水利管理是保障全球粮食安全与水资源可持续利用的关键环节，亟需高效、精准的信息感知与调控手段。近年来，“天-空-地”一体化的多源遥感观测体系，为农业水利信息的动态监测，尤其是在区域和田块尺度上的应用，提供了新的发展动力。本文系统综述了多源遥感数据在农业水利信息感知中的最新研究成果，涵盖遥感数据获取与处理、建模方法及典型应用。在数据获取方面，卫星、无人机与地面平台多类型传感器的协同工作，显著提升了数据的空间分辨率和观测维度；在数据处理方面，遥感数据处理逐步从本地化处理向云平台协同处理转型，从而提高了数据融合的效率和时空一致性；在建模方面，融合物理机制与数据驱动的混合模型正成为主流，显著提高了预测精度与模型的泛化能力。上述进展推动遥感技术广泛应用于农业水旱监测、作物生长状态评估和环境监测等领域。尽管多源遥感技术已取得显著进展，其在农业水利信息感知中的应用仍面临诸多挑战，主要包括平台间信息整合的困难、数据处理标准化的不足、模型性能的提升空间以及成果转化与服务能力需进一步加强等问题。未来研究将聚焦于构建高时空协同的观测体系，发展平台化与智能化的数据处理流程，推动机理与智能融合的建模方法，以及实现遥感服务与实际应用场景的深度融合，为智慧农业的实现和可持续发展目标的达成提供更强有力的支撑。更多还原

【Abstract】 Agricultural water management is a critical component in ensuring global food security and the sustainable use of water resources, necessitating efficient and precise information sensing and regulation methods. In recent years, the integrated “sky-space-ground” multi-source remote sensing observation system has provided opportunities for the dynamic monitoring of agricultural water resources, particularly at the regional and field scales. The latest research advancements in the application of multi-source remote sensing data for agricultural water resources perception, covering data acquisition and processing, modeling methods, and typical applications were systematically reviewed. In terms of data acquisition, the collaboration between satellite, drone, and ground-based platform sensors significantly enhanced data spatial resolution and observation dimensions. Regarding data processing, remote sensing data processing was transitioning from localized approaches to cloud-based collaborative processing, thereby improving data fusion efficiency and spatiotemporal consistency. In modeling, hybrid models combining physical mechanisms and data-driven approaches were becoming mainstream, significantly improving predictive accuracy and model generalization. These advancements drove the widespread application of remote sensing technologies in agricultural drought and flood monitoring, crop growth status assessment, and environmental monitoring. However, despite significant progress in multi-source remote sensing technology, several challenges remained in its application for agricultural water resources perception. These challenges included difficulties in information integration between platforms, lack of standardization in data processing, room for improvement in model performance, and the need to enhance the conversion and service capabilities of research outcomes. Looking ahead, future research should focus on building high spatiotemporal collaborative observation systems, developing platform-based and intelligent data processing workflows, promoting modeling methods that integrated mechanisms with intelligence, and deepening the fusion of remote sensing services with practical application scenarios, aiming to provide stronger support for the realization of smart agriculture and the achievement of sustainable development goals.更多还原