基于新一代多卫星遥感联合反演的西南河流源区地形雨监测及径流模拟

Precipitation is a critical variable for runoff simulation and prediction. Due to the complex mountainous terrains, it is mostly infeasible to fully monitor precipitation with conventional rain gauge networks or weather radars over the ungauged source areas of major rivers in Southwest China. Precisely measuring precipitation has become the challenging task for the hydrologic simulation over these regions. The new-generation integrated multi-satellite precipitation retrieval has become an effective approach to quickly obtain the continuous surface precipitation information at a broader coverage. Presently, the most important objective is to improve the data accuracy and spatiotemporal resolution of the satellite precipitation estimates in the source areas. Specifically, the great hydrologic potential of the Global Precipitation Measurement (GPM) should be further developed and applied over the typical ungauged basins. .The overarching goal of this proposal is to improve the satellite-derived estimates of orographic rainfall and investigate the nonlinear error propagation of satellite precipitation in the hydrologic processes. We intend to install some in-situ instruments (e.g., tipping-bucket rain gauges and laser optical disdrometers) to observe the orographic rainfall in the upstream areas of Yalong River and Yarlung Tsangpo River in order to reveal the dependence of precipitation on orography. Then, we will improve the accuracy of satellite estimates for orographic rainfall by using the meteorological outputs of mesoscale atmospheric model and ground observations. Finally, we will trace the propagation of the systematic and random errors of satellite-based precipitation estimates in the rainfall-runoff processes and furtherly identify the error sources of runoff simulation associated with the satellite-derived precipitation inputs and hydrologic model parameters. Results drawn from this study may provide a novel method for measuring the orographic rainfall in mountainous areas. This project can serve for the primary scientific objectives of the NSFC Major Program entitled with “Runoff changes and adaptive utilization over the source areas of major rivers in Southwest China”.

降水是径流变化模拟和预测的关键。在缺资料的西南河流源区，由于复杂的高山地形，地面降水获取困难成为该地区径流变化模拟的瓶颈。新一代多卫星遥感联合反演技术为大范围快速获取连续面降水信息提供了一种新方法。目前，河源区的卫星降水研究急需实现高精度、高分辨率的突破，特别是全球降水计划GPM的水文应用潜力亟待挖掘。本项目拟围绕“卫星降水反演中的地形雨问题”和“卫星降水误差的非线性水文传递”两个关键科学问题开展研究。项目组将在雅砻江流域和雅鲁藏布江流域的上游山区进行地形雨的野外观测工作，揭示高山区降水对地形的依赖度；结合高精度气候模式模拟的气象要素及地面观测，改进卫星降水系统中地形降水的反演精度；解析卫星降水的系统误差和随机误差在径流模拟中的传递过程，实现河源区缺资料流域径流模拟的误差溯源。研究成果可为高山区降水要素的监测提供新技术和新方法，为重大计划科学目标“径流变化与预测”提供基础理论和科学支撑。

降水是径流变化模拟和预测的关键。在缺资料的西南河流源区，由于复杂的高山地形，地面降水获取困难成为该地区径流变化模拟的瓶颈。新一代多卫星遥感联合反演技术为大范围快速获取连续面降水信息提供了一种新方法。.多卫星遥感降水联合反演的地形雨监测及误差溯源。项目组在雅鲁藏布江中游古觉村流域建立了GPM验证格网，在滁州水文山建立了GPM地面对比验证基地。河源区地面降雨观测总体偏低（原因：雨量站位于低平地区，而降雨主要发生在山体的迎风坡面）；可用于高山野外观测的便携式翻斗雨量计HOBO平均低估降雨3.96%，最大偏差达到-12.87%。基于星载双频雷达及地面观测，研究发现：河源区2800m高程以上区域地形雨显著增强，在3500m左右达到峰值，之后地形雨快速衰减；最新的GPM-DPR星载双频雷达能够捕捉到3000-3500m之间的降水峰值，而TRMM-PR降水反演整体偏高且探测不到地形雨的峰值。对GPM的10种微波多传感器误差溯源，结果显示微波imager总体上优于sounder；GMI和TMI在强降雨的监测上具有明显优势。考虑地形因子，采用病态最小二乘法对主流GPM卫星降水产品进行小时尺度订正，生产出一套河源区高精度卫星降水数据集（1h，0.1°×0.1°），可用于河源区各流域的径流模拟与计算。.河源区径流变化分析及模拟。过去半个世纪，我国西南河源区径流总体呈增加趋势（24个径流站中17个增加、7个减少），2000年后径流增加更明显；空间上呈现中部增加、西部和东部减少的分布特征，但不显著；径流量显著变化的有2个：金沙江上游沱沱河站显著增加（原因：降水增加），澜沧江下游允景洪站显著减少（原因：水库大坝）。河源区1990年开始温度急剧上升，2000年出现了一个明显的气候转型，青藏高原局地大气循环开始加速，风速加快、辐射增强、湿度减少，河源区冰川消融加速，气候趋向更干、更热。.本项目利用GPM星载双频雷达对地形雨监测的优势，揭示河源区典型流域对流性降水对地形的依赖特性，根据不同的传感器特性并考虑地形因子对卫星降水产品进行校正，产生一套河源区高精度卫星降水数据集。.发表SCI论文8篇、EI论文1篇、中文3篇，另外有4篇SCI论文在修订中；获2018年度教育部自然科学二等奖1项（第2）；培养博士研究生7人、硕士研究生15人（毕业9人）。