谢漪云,王建捷. 2021. GRAPES千米尺度模式在西南复杂地形区降水预报偏差与成因初探[J]. 气象学报, (0):-, doi:10.11676/qxxb2021.053
GRAPES千米尺度模式在西南复杂地形区降水预报偏差与成因初探
Preliminary study on the precipitation prediction and its causes of deviation in complex terrain area of southwest China based on the GRAPES-Meso_3km model
投稿时间:2021-03-08  修订日期:2021-05-28
DOI:10.11676/qxxb2021.053
中文关键词:  西南复杂地形区,降水量和频率,偏差特征
英文关键词:complex terrain of southwest China  rainfall amount and frequency  deviation characteristics
基金项目:中国气象局GRAPES发展专项
作者单位邮编
谢漪云 中国气象科学研究院 100081
王建捷 国家气象中心 100081
摘要点击次数: 156
全文下载次数: 155
中文摘要:
      [资料和方法] 利用2019年夏季(6-8月)西南复杂地形区地面观测站逐时和逐日降水量观测数据,从降水量和频率入手,对同期GRAPES-Meso_3km业务模式短期(36h以内)降水预报性能,特别是在不同典型地貌区——四川盆地(子区A)、云贵高原(北部子区B和南部子区C)和青藏高原东缘山地(子区D)的预报偏差进行细致评估与分析。[结果和结论]结果表明:(1)GRAPES-Meso_3km模式合理刻画出西南复杂地形区夏季平均日降水量(频率)的空间分布和不同地貌区的日内尺度单峰/双峰演变特征,以及小时降水频次-强度的基本关系。(2)模式日降水量(频率)预报平均表现为清晰的正偏差,偏差值在4个子区大小不同、为观测值的1.3-2.1倍(1.1-1.3倍),其中偏差在盆地(子区A)最突出;4个子区日降水量正偏差均主要是由强降水日降水量预报偏多引起,但频率正偏差仅在盆地和高原东缘山地两子区主要是强降水日数预报偏多导致、而在云贵高原南北两子区则是中小雨日数预报偏多的贡献;强降水(中小雨)落区预报存在明显(轻微)偏大倾向,4个子区强降水(中小雨)预报落区偏大频率在53.6% - 82.8%(22.6%-37.6%),降水落区预报偏小频率普遍在10%以下。(3)日循环上,各时次小时降水量(频率)预报整体偏大、且主要正偏差出现在观测的夜雨峰值时段,其中海拔1200m以下区域的降水频率正偏差从夜间峰值区延展到中午,这与在小时降水频次-强度关系中看到的模式对强雨强预报偏弱的特征是吻合的,暗示模式偏强的日降水量预报往往表现为日内偏长的降水时长(或小时降水空报)。(4)诊断分析显示,模式在四川盆地区突出的夏季日降水预报正偏差是由模式偏强的对流层低层(云贵高原南-东南侧)西南风预报与西南地区特殊地形结合所致。
英文摘要:
      Based on the daily (08:00-08:00 BT) and hourly precipitation observation data in complex terrain area of southwest China in the summer of 2019, the performance and deviation of the short-term precipitation (within 36h) forecast of GRAPES-Meso_3km model has been carefully evaluated in term of the rainfall amount and frequency,in view of different typical landform areas such as the Sichuan Basin (area A), the Yunnan-Guizhou Plateau (northern area B and southern area C) and the highland over the eastern edge of Tibet Plateau (area D). Results show that: (1) In the study area, model reproduced the spatial distribution of summer average daily rainfall amount (frequency) and their evolution features in diurnal cycle, as well as the key relationship between hourly rainfall frequency and intensity reasonably well. (2) The daily rainfall amount (frequency) forecast performs a positive deviation on the whole, and the deviation shows differences in 4 sub-regions: the average forecast daily amount (frequency) is about 1.3-2.1 (1.1-1.3) times of observation, which?most significant is the deviation of rainfall amount in?basin (area A). The positive deviations of daily rainfall amount in each sub-region are mainly due to bigger amount forecast in heavy rain day, while the positive frequency deviations in area A/D (B/C) are mostly caused by more heavy (moderate and light ) rain day frequency in model. The forecast zone of daily rainfall amount has a tendency to be large, the larger days frequency of heavy (moderate and light) rain in 4 sub-regions are 53.6%-82.8% (22.6%-37.6%). (3) As for daily cycle, the forecast of hourly rainfall amount (frequency) at each time is generally larger, the main positive deviation appears over the observed peak period of night rain. At altitude down of 1200m, the frequency positive deviation stretchs from the observation night peak to noon, this fits the weaker forecast of heavy rain intensity in the relationship between frequency and intensity of hourly precipitation, it suggests that the strong daily rainfall amount forecast is usually manifested as a longer precipitation duration within the day (or hourly precipitation vacant forecast). (4) Diagnostic analysis shows that higher forecast of summer average daily rainfall amount over Sichuan Basin is caused by the combination of stronger southwest wind forecast in lower troposphere (south-southeast of Yunnan-Guizhou Plateau) and the special terrain in southwest China.
查看全文   查看/发表评论  下载PDF阅读器
分享按钮