高级检索

    FY-4A与FY-4B定量降水估计产品在四川的精度评估

    Accuracy assessment of FY-4A and FY-4B Quantitative Precipitation Estimation in Sichuan

    • 摘要: 利用2022年9月至2023年8月四川省地面站逐小时降水数据,通过统计指标和评价指标对风云四号A星(FY-4A)与风云四号B星(FY-4B)定量降水估计产品(Quantitative Precipitation Estimation,QPE)在四川的精度进行对比评估。结果表明:FY-4A与FY-4B QPE产品在降水较少的高原地区表现优于降水较多的川东北和高原盆地过渡带。FY-4A倾向于低估降水,而FY-4B容易高估降水,FY-4A整体表现优于FY-4B。年尺度下,FY-4A和FY-4B反演的降水量与地面观测的降水量相关系数分别为0.69和0.59,FY-4A更接近地面观测降水。季、月尺度下,春季、10月两种卫星QPE产品与地面观测的降水量相关性高,冬季较低。春季和9—10月 FY-4A性能表现较好,7—8月FY-4B性能表现更好。除FY-4B略高估2.0 mm·h-1以下的降水外,其他降水量级下,两种QPE产品主要表现为低估降水,FY-4A低估更多。随雨强增高,低估越明显,均方根误差也越大。在地面无降水时,FY-4A表现略优于FY-4B。对2023年7月两次暴雨事件分析表明,两种QPE产品能够指示降水落区的大致范围,对降水出现时段的反映能力较好,但低估了降水量。

       

      Abstract: Using hourly precipitation data from ground stations in Sichuan Province from September 2022 to August 2023,the accuracy of Quantitative Precipitation Estimation (QPE) products from FengYun-4A (FY-4A) and FengYun-4B (FY-4B) satellites in Sichuan was compared by statistical and evaluation indicators.The results show that FY-4A and FY-4B QPE products perform better in regions with less precipitation,such as the plateau region,than in regions with more precipitation,like northeastern Sichuan and the plateau-basin transition zone.FY-4A tends to underestimate precipitation,while FY-4B tends to overestimate it,with FY-4A performing better overall than FY-4B.At the annual scale,the correlation coefficients between precipitation from FY-4A and FY-4B and that observed on the ground are 0.69 and 0.59,respectively.This indicates that FY-4A is closer to ground-based precipitation observations.At the seasonal and monthly scales,the correlation coefficients between the two satellite QPE products and the ground-observed precipitation are high in spring and October,and lower in winter.FY-4A performs better in spring and September-October,while FY-4B performs better in July-August.Except for the slight overestimation by FY-4B for precipitation below 2.0 mm·h-1,both QPE products mainly underestimate precipitation at other precipitation levels,with FY-4A underestimating more.As rainfall intensity increases,the underestimation worsens,and the root mean square error increases.FY-4A shows slightly better performance than FY-4B when there is no ground precipitation.Analysis of two heavy rainfall events in July 2023 indicates that both QPE products can indicate the approximate range of precipitation areas and reflect the timing of precipitation occurrence well,but both underestimate the amount of precipitation.

       

    /

    返回文章
    返回