杨浩,江志红,李肈新. 2017. 分位数调整法在动力降尺度模拟订正中的适用性评估—以北京为例[J]. 气象学报, ():-, doi:10.11676/qxxb2017.028
分位数调整法在动力降尺度模拟订正中的适用性评估—以北京为例
Applicability of a quantile-quantile (Q-Q) bias-correction method used after climate dynamical downscaling——a case study at Beijing
投稿时间:2016-09-09  最后修改时间:2016-12-08
DOI:10.11676/qxxb2017.028
中文关键词:  分位数调整,偏差订正,动力降尺度,气温,降水
英文关键词:quantile-quantile (Q-Q) adjustment, bias correction, dynamical downscaling, temperature, precipitation
基金项目:国家自然科学基金重点项目(41230528,91637211);湖北省气象局科技发展基金(2017C05)
作者单位E-mail
杨浩 中国气象局武汉暴雨研究所 yanghao0202@126.com 
江志红 南京信息工程大学 zhjiang@nuist.edu.cn 
李肈新 法国科研中心动力气象实验室 lilaurent@yahoo.fr 
摘要点击次数: 1021
全文下载次数: 612
中文摘要:
      分位数调整法对变网格模式LMDZ4在中国区域进行动力降尺度模拟的北京日平均气温和降水结果进行了统计误差订正。订正后的日平均气温在年循环、平均值和频率等方面均十分接近观测值,全年平均气温偏差由-1.2℃减小到-0.4℃。降水的订正过程较气温更加复杂,首先对降水日数进行订正,以消除模式产生的虚假性微小值降水,订正后降水日数误差由61.5%减小到3.7%。此外,分位数调整法可有效订正中小型与极端降水的频率和强度,订正后全年降水误差由0.28mm/d减小到0.07mm/d。订正后最大降水月份为7月,与观测一致,消除了冬季的虚假极端降水。分位数调整法无论是对气温还是降水,其订正效果都存在明显的季节性差异。日平均气温的订正在冬季和夏季要优于春季和秋季,对极端高、低温的订正更加显著。该统计误差订正方法不仅有效消除气候平均值的漂移,同时对极值也有一定改善,是一种相对较完善的订正方案。分位数调整法也存在一定的不确定性,订正效果受观测资料和模式模拟能力影响较大。
英文摘要:
      A statistical bias correction based on quantile-quantile (Q-Q) adjustment is applied to daily temperature and precipitation at Beijing simulated by the variable resolution model LMDZ4. After bias correction, the annual cycle, average and frequency of temperature are all closer to observation, the deviation of annual mean temperature decreases from -1.2 ℃ to -0.4 ℃. The bias correction can remove most of the spurious drizzle generated by the LMDZ4 model. Biases of rainy days decrease to 3.7% from 61.5%. The Q-Q adjustment shows good performance of correction on precipitation intensity and frequency, the deviation of annual mean precipitation decreases to 0.07mm/d from 0.28mm/d. After correction, precipitation peaks in July, consistent with observation, and the false extreme precipitation in winter is removed. The Q-Q adjustment is separately operated for different seasons for both temperature and precipitation. The correction effect for daily temperature is superior in winter and summer, compared to spring and autumn And significant improvements are obtained for extreme high and low temperatures. This statistical bias-correction method not only effectively eliminates drift on simulated climatological mean, but also increases the capability of reproducing extreme climate values. It is a relatively satisfying correction scheme. Meanwhile, there are still some uncertainties in Q-Q adjustment, the correction effect is influenced by observational data and model performance.
查看全文   查看/发表评论  下载PDF阅读器
分享按钮