昆仑山北坡地形复杂，极端暴雨事件频发，但现有地面观测站点稀少且分布不均，难以捕捉复杂地形下极端暴雨的精细化分布和演变过程，高分辨率卫星数据为极端暴雨的监测提供了新的手段。本文基于昆仑山北坡383个气象台站逐小时降水数据和8种卫星降水产品，选取研究区内3次极端暴雨过程，综合评估了8种卫星降水产品在监测昆仑山北坡极端降水事件中的适用性，定量分析卫星降水产品对昆仑山北坡极端降水的监测能力。结果表明：（1） GPM IMERG Early、GPM IMERG Late、GPM IMERG Final降水产品在捕捉昆仑山北坡极端暴雨空间分布和降水量方面表现较好，命中率均大于99%。GPM IMERG Final产品表现最优，个例3中其与地面观测降水量的相关系数达0.64，FY2H和FY4A降水产品在极端降水过程定性和定量评估中均与站点实测有一定差距，相关系数最低在0.01以下。（2） 在极端降水过程的时间演变特征方面，8种卫星降水产品均存在一定误差，偏差幅度在-95%~250%，相较于其他卫星产品GPM IMERG Final在各时次降水量和变化趋势上表现最优，准确率和TS评分超过0.8。（3） 8种卫星降水产品再现3次极端暴雨空间范围较观测偏小，在降水强度上整体偏弱，存在明显的低估特征。漏报误差在偏差相对贡献率中普遍占比在50%以上，这是导致卫星降水产品对强降水事件低估的主要原因之一。总体来看，卫星降水产品能在一定程度上反映昆仑山北坡极端降水过程，但其监测精度需进一步提升，该研究结果可以为研究区不同应用选用最合适的卫星降水产品提供科学依据，同时为降水产品偏差校正、算法改进及该区域极端降水的监测提供一定的数据支撑。

The northern slope of the Kunlun Mountains is characterized by complex topography, extreme rainstorms, and an uneven distribution of meteorological stations, making it challenging to accurately capture the fine-scale spatial distribution and temporal evolution of extreme rainstorms using conventional, ground-based observations. Satellite precipitation products for monitoring rainstorms not only effectively fill the regional gap but also improve the capability of monitoring and early warning for severe catastrophic weather. On the basis of hourly precipitation data from 383 meteorological stations and eight sets of satellite precipitation products over the northern slope of the Kunlun Mountains, we selected three representative extreme rainstorm processes to comprehensively assess the applicability of eight satellite precipitation products for monitoring extreme precipitation events and quantitatively analyze their performance on the northern slope of the Kunlun Mountains. The results were as follows: (1) The GPM IMERG Early, GPM IMERG Late, and GPM IMERG Final products showed good performance in capturing the spatial distribution and magnitude of extreme precipitation, with hit rates exceeding 99%. Among them, the GPM IMERG Final product had the best performance, with a correlation coefficient of 0.64 between ground-observed precipitation and the GPM IMERG Final product during Event 3. The FY2H and FY4A precipitation products showed certain discrepancies in comparison with ground observations in both qualitative and quantitative evaluations, with low correlation coefficients (below 0.01). (2) In terms of temporal evolution characteristics of extreme precipitation processes, all satellite precipitation products showed certain biases, ranging from -95% to 250%. In comparison with the other satellite products, the GPM IMERG Final product outperformed in terms of precipitation amounts at each time step and temporal trends, with accuracy and threat score values above 0.8. (3) The spatial extent of the three extreme rainstorms reproduced by eight satellite precipitation products was smaller than that indicated by observations, and the intensity was weaker, with obvious underestimation characteristics. The missing detection bias generally accounted for more than 50% of the relative contribution rate of errors, which is one of the main reasons for the underestimation of extreme heavy precipitation events by satellite precipitation products. Overall, satellite precipitation products reflected the extreme precipitation processes on the northern slope of the Kunlun Mountains to a certain extent, but their accuracy in monitoring requires further improvement. These results provide a scientific basis for selecting appropriate satellite precipitation products for regional applications and offer data support for bias correction, algorithm optimization, and improved monitoring of extreme precipitation in this region.