干旱气象 ›› 2012, Vol. 30 ›› Issue (3): 465-471.

• 业务技术应用 • 上一篇    下一篇

应用24 h 变量相关系数检验数值预报产品

周甘霖尚可政王式功程一帆   

  1. 兰州大学大气科学学院,甘肃省干旱气候变化与减灾重点实验室,甘肃兰州730000
  • 收稿日期:2012-04-09 出版日期:2012-09-30 发布日期:2012-06-04
  • 作者简介:周甘霖( 1986 - ) ,男,硕士研究生,主要从事干旱气候和现代天气预报技术和方法研究. E - mail: zhougl09@ lzu. cn
  • 基金资助:

    国家科技支撑计划( 2009BAC53B02) 、国家公益性行业专项项目( GYHY201106034) 和国家自然科学基金项目( 41075103) 共同资助

Evaluation of Numerical Predicting Products by Using24 Hours Variable Correlation Coefficient

ZHOU Ganlin,SHANG Kezheng,WANG Shigong,CHEN Yifan   

  1. College of Atmospheric Sciences,Lanzhou University,Lanzhou 730000,China; Key Laboratory of Arid Climatic Change and Reducing Disaster of Gansu Province,Lanzhou 730020,China
  • Received:2012-04-09 Online:2012-09-30 Published:2012-06-04

摘要:

气象要素的变量可反应天气系统的变化趋势,所以,在数值预报产品释用业务中,气象要素场及一些物理量的24 h 变量经常作为重要的预报因子,因此,有必要以24 h 变量相关系数为指标,对数值预报产品进行检验。利用国家气象中心T213 L31 数值预报产品,对东亚范围内2006 ~ 2011 年对流层各等压面的高度场、温度场、风场、垂直速度场和水汽场的预报,以24 h 变量相关系数为指标,进行了检验。结果表明: 1) 高度场预报,通过相关显著性水平α = 0.001 检验的时效为10 d,高可信度( 24 h 变量相关系数> 0.4) 的时效,对流层中上部6 ~ 7 d,对流层中下部5 d; 2) 温度场预报,通过α =0.001 水平相关显著性检验的时效为9 d,高可信度的时效,对流层中上部4 d,对流层中下部5 d; 3)风场预报,通过α = 0.001 水平相关显著性检验的时效为8 d,高可信度的时效,对流层上部、中部和下部分别为5 d、4 d 和3 d; 4) 垂直速度场预报,通过α = 0.001 水平相关显著性检验的时效为5 d,高可信度的时效仅为1 ~ 2 d; 5) 水汽场预报,通过α = 0.001 水平相关显著性检验的时效为6 d,高可信度的时效,对流层中部2 d,下部3 ~ 4 d。这些将为T213 L31 数值预报产品的合理有效利用提供参考依据。

关键词: T213 L31 数值预报产品, 24 h 变量相关系数, 检验

Abstract:

Based on the T213 L31 numerical prediction products,the height field,temperature field,wind field,vertical velocity field, and humidity field on every pressure level in range of east Asia have been evaluated by using 24 hours variable correlation coefficient. The results are as follows: 1) For the height field forecast,the predicting length that could pass the significance test at the level of α = 0. 001 was 10 days,and the higher credible ( 24 h variable correlation coefficient more than 0. 4) predicting length could reach 6 - 7 days at the middle and upper layer of troposphere and 5 days at the middle and lower layer of troposphere; 2) For the temperature field forecast,the predicting length that could pass the significance test at the level of α = 0. 001 was 9 days,and the higher credible predicting length could reach 4 days at the middle and upper layer of troposphere,and 5 days at the middle and lower layer of troposphere; 3) For the wind field forecast,the predicting length that could pass the significance test at the level of α = 0. 001 was 8 days,and the higher credible predicting length could reach 5 days at the upper layer of troposphere,and 4 days at the middle layer of troposphere, and 3 days at the lower layer of troposphere; 4) For the vertical velocity field forecast,the predicting length that could pass the significance test at the level of α = 0. 001 was 5 days,and the higher credible predicting length could only reach 1 - 2 days at troposphere; 5) For the specific humidity field forecast,the predicting length that could pass the significance test at the level of α = 0. 001 was 6 days,and the higher credible predicting length could reach 2 days at the middle layer of troposphere,and 3 - 4 days at the lower layer of troposphere.

Key words: T213 L31 numerical prediction products, 24 h variable correlation coefficient, verification

中图分类号: