Abstract: Based on the multi-radar mosaic and minute-by-minute gauge observations in five provinces including Jiangxi, Zhejiang, Fujian, Anhui and Hunan, the cloud grouping Z-R  relationships fitting method was used to estimate precipitation quantitatively along Zhejiang-Jiangxi railway lines, and five rainfall cases in history resulting in railway disaster were selected to test the effect of precipitation estimation. The results are as follows: (1) The method of quantitative precipitation estimation (QPE) based on the radar-gauge mergence information along the railway lines could effectively capture the difference of the type and area of rainfall, and the prediction result provided better calibration effects than the simple grouping Z-R relationships fitting scheme. (2) The effect test of radar QPE indicated that the system error and local error were improved effectively.

Key words: three-dimensional mosaic, Z-R relationships, QPE (quantitative precipitation estimation)

摘要： 利用雷达组网数据和自动站逐分钟雨量数据，选取浙赣铁路灾害降水个例，运用云团分型的动态Z-R关系拟合算法进行定量降水估测。结果表明：（1）基于铁路沿线的雷达-雨量计降水估测方案可以有效识别降水类型和降水区域的差异性特征，其估测精度优于简单分组Z-R关系拟合算法；（2）对降水估测结果进行效果检验，评估结果可以有效改善雷达QPE的系统误差和局部误差。

关键词: 三维组网, Z-R关系, 定量降水估测