2014大气科学进展:英文版Requires access

An Approach for Improving Short-Term Prediction of Summer Rainfall over North China by Decomposing Interannual and Decadal Variability

Han, Leqiong, LI -, Shuanglin, Liu

Open publisher page 0 citations

Abstract

一条统计 downscaling 途径被开发由基于观察数据集和动态模型产量考虑十的可变性的效果在诺思中国上改进夏天降雨的 seasonal-to-interannual 预言。predictands 和预言者首先被分解成 interannual 和十的部件。二个预兆的方程然后被使用 multivariate 为二不同 timescales 独立造线性回归基于独立样品确认。为 interannual timescale, 850-hPa 从多重动态模型 hindcasts 的南方的风和 500-hPa geopotential 高度和从观察数据集的 SST 被用来构造预言者。为十的 timescale,二著名盆规模 SST 十的摆动(大西洋 Multidecadal 摆动和和平的十的摆动) 索引被用作预言者。然后, downscaled predictands 被联合代表 predicted/hindcasted 总数降雨。预言从一条类似的途径,但是没有 timescale 分解与模型相比是未加工的 hindcasts 和那些。与从单个模型或他们的多模型整体平均数的 hindcasts 相比,现在的计划的技巧被发现显著地更高,与异例关联系数增加从对超过 0.4 并且与由多达 0.6 公里 d1 减少的 RMSE 将近中立。改进也在与观察降雨的预言的基于车站的时间的关联被看见,与从 0.1 ~ 0.87 的系数,显然,高,比模型,未加工的 hindcasted 降雨结果。因此,现在的途径展出一个伟人有利并且可能为在运作的预言的使用适当。

About this research paper

What this paper is about

一条统计 downscaling 途径被开发由基于观察数据集和动态模型产量考虑十的可变性的效果在诺思中国上改进夏天降雨的 seasonal-to-interannual 预言。predictands 和预言者首先被分解成 interannual 和十的部件。二个预兆的方程然后被使用 multivariate 为二不同 timescales 独立造线性回归基于独立样品确认。为 interannual timescale, 850-hPa 从多重动态模型 hindcasts 的南方的风和 500-hPa geopotential 高度和从观察数据集的 SST 被用来构造预言者。为十的 timescale,二著名盆规模 SST 十的摆动(大西洋 Multidecadal 摆动和和平的十的摆动) 索引被用作预言者。然后, downscaled predictands 被联合代表 predicted/hindcasted 总数降雨。预言从一条类似的途径,但是没有 timescale 分解与模型相比是未加工的 hindcasts 和那些。与从单个模型或他们的多模型整体平均数的 hindcasts 相比,现在的计划的技巧被发现显著地更高,与异例关联系数增加从对超过 0.4 并且与由多达 0.6 公里 d1 减少的 RMSE 将近中立。改进也在与观察降雨的预言的基于车站的时间的关联被看见,与从 0.1 ~ 0.87 的系数,显然,高,比模型,未加工的 hindcasted 降雨结果。因此,现在的途径展出一个伟人有利并且可能为在运作的预言的使用适当。

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

一条统计 downscaling 途径被开发由基于观察数据集和动态模型产量考虑十的可变性的效果在诺思中国上改进夏天降雨的 seasonal-to-interannual 预言。predictands 和预言者首先被分解成 interannual 和十的部件。二个预兆的方程然后被使用 multivariate 为二不同 timescales 独立造线性回归基于独立样品确认。为 interannual timescale, 850-hPa 从多重动态模型 hindcasts 的南方的风和 500-hPa geopotential 高度和从观察数据集的 SST 被用来构造预言者。为十的 timescale,二著名盆规模 SST 十的摆动(大西洋 Multidecadal 摆动和和平的十的摆动) 索引被用作预言者。然后, downscaled predictands 被联合代表 predicted/hindcasted 总数降雨。预言从一条类似的途径,但是没有 timescale 分解与模型相比是未加工的 hindcasts 和那些。与从单个模型或他们的多模型整体平均数的 hindcasts 相比,现在的计划的技巧被发现显著地更高,与异例关联系数增加从对超过 0.4 并且与由多达 0.6 公里 d1 减少的 RMSE 将近中立。改进也在与观察降雨的预言的基于车站的时间的关联被看见,与从 0.1 ~ 0.87 的系数,显然,高,比模型,未加工的 hindcasted 降雨结果。因此,现在的途径展出一个伟人有利并且可能为在运作的预言的使用适当。

Key concepts: Climatology, Downscaling, Environmental science, Geopotential height, Term (time), Teleconnection, Geopotential, Precipitation

Related papers

Back to paper searchBrowse research topicsOriginal source
An Approach for Improving Short-Term Prediction of Summer Rainfall over North China by Decomposing Interannual and Decadal Variability — Research Paper | ScholarLens