地面气压在纵向峡谷区大风预报中的应用探讨

doi:10.11755/j.issn.1006-7639-2024-05-0804

摘要/Abstract

摘要：

为减少大风天气对金沙江下游水电站建设和电力输送的影响，通过分析干季频繁发生的大风事件，探索峡谷区风速变化与气压相关参数之间的关系，寻找预警大风的有效指标。该地区因风速预报难度大，数值预报在峡谷区受到限制，因而地面观测尤为重要，尤其是气压变化。本文以金沙江下游四川和云南交界的南北纵向峡谷为研究区，利用峡谷区自建观测站数据，结合水电站坝区的经验预报，研究峡谷区大风事件中风速与气压相关参数的关系。结果表明，大风发生前，站点气压通常明显下降，下降值超过7.7 hPa并持续3~6 h后，风速增加的可能性增大；若风速稳定升高持续1~6 h，能达到7级大风。测站的变压值对风速预警极为重要，尤其是3 h变压，当3 h负变压超过2.0 hPa时，地面风速开始增加。通过格兰杰因果关系检验方法发现，峡谷上下游风速变化存在明显滞后性，葫芦口大桥站通常先于下游1~2 h出现大风。气压差可作为重要大风预警指标，当荒田与葫芦口大桥站气压差达18.1 hPa时，大风发生的可能性明显增大。

关键词: 大风预报, 气压变化, 纵向峡谷, 格兰杰算法, 金沙江

Abstract:

To minimize the impact of windy weather on the construction of hydropower stations and power transmission in the lower reaches of the Jinsha River, this study aims to explore the relationship between wind speed variations and barometric pressure parameters in the canyon area, and to find effective indicators about early warning of windy weather through in-depth analysis of frequent windy events in the dry season. Numerical forecasting is limited in canyon areas due to the difficulty of wind speed forecasting, thus ground-based observations are particularly important, especially for barometric pressure variations. This paper takes the north-south longitudinal canyon at the junction of Sichuan and Yunnan in the lower reaches of the Jinsha River as the research object, and uses the data from the self-built observatory in the canyon area, combining with the empirical forecasts from the hydropower station dam area, to study the relationship between wind speed and barometric pressure related parameters in the canyon area during the gale events. The results show that, before the occurrence of gale, the barometric pressure at the station usually decreases significantly, and after the decrease exceeds 7.7 hPa and lasts for 3-6 hours, the possibility of wind speed increasing increases; if wind speed increases steadily for 1-6 hours, the wind can reach the level 7 gale. The pressure variation values of the stations are very important for wind speed warning, especially the 3-hour pressure change values. When the 3-hour negative pressure change value exceeds 2.0 hPa, the surface wind speed starts to increase. The Granger causality test method reveals that there is an obvious lag in the wind speed change in the upstream and downstream of the canyon, for example, Hulukou Bridge Station usually experiences gale 1-2 hours before the downstream. The difference in barometric pressure can be used as an important early warning indicator of gales, and when the difference in barometric pressure between Huangtian and Hulukou Bridge Station reaches 18.1 hPa, the possibility of gales increases significantly.

Key words: gale forecast, pressure variation, north-south canyon, Granger causality algorithm, Jinsha River

中图分类号:

P457

靳双龙, 刘晓琳, 王勃, 白锦丰, 白爱娟, 张强. 地面气压在纵向峡谷区大风预报中的应用探讨[J]. 干旱气象, 2024, 42(5): 804-812.

JIN Shuanglong, LIU Xiaolin, WANG Bo, BAI Jinfeng, BAI Aijuan, ZHANG Qiang. Exploring the application of surface pressure in gale forecasting over north-south canyon areas[J]. Journal of Arid Meteorology, 2024, 42(5): 804-812.

图/表 10

图1 金沙江下游地形（a）和峡谷水电站坝区观测站位置（b）注:基于自然资源部国家地理信息公共服务平台(天地图)网站标准地图[审图号:GS(2024)0650号]制作，底图边界无修改（黑色粗线为四川省界；填色为地形高度，单位：m）

Fig.1 Downstream topography of the Jinsha River （a） and the locations of observation stations in the canyon region hydropower dam area （b）（The area enclosed by the black thick lines is the Sichuan Province； the color shaded indicates topographic height， Unit： m）

图2 新田站2020年1月7日16:00—9日16:00（a）和12月21日08:00—23日08:00（b）风速与气压随时间的变化（红色虚线为7级大风阈值）

Fig.2 Changes of wind speed and air pressure from 16:00 on 7 to 16:00 on 9 January 2020 (a) and from 08:00 on 21 to 08:00 on 23 December 2020 (b) of Xintian Station (The red dashed line is the threshold for a Force 7 gale)

图3 新田站和葫芦口大桥站大风事件中最大气压差（a）、风速增加时间与气压峰值时间差（b）及风速增加到7级所需时间（c）的箱线图

Fig.3 The box plots of maximum air pressure difference （a）， difference between wind speed increasing time and air pressure peak time （b） and duration to reach Force 7 winds （c） in strong wind events at Xintian Station and Hulukou Bridge Station

表1 Percentages of positive and negative pressure changes prior to wind speed increase in 51 strong wind events during the dry season at Xintian Station in 2020 单位：%

Tab.1

变压	风速增加前时间
变压	1 h	2 h	3 h	4 h	5 h	6 h	7 h	8 h	9 h
负变压	71	73	69	73	75	78	78	82	82
正变压	29	27	31	27	25	22	22	18	18

图4 2020年新田站干季51次大风事件中风速与前1~9 h负变压的相关系数（a）以及3、6、9 h负变压分布（b）

Fig.4 Correlation coefficients between wind speed and negative pressure changes prior to 1-9 hours （a） and the distributions of negative pressure changes prior to 3， 6 and 9 hours （b） in 51 strong wind events during the dry season of 2020 at Xintian Station

图5 葫芦口大桥站、骑骡沟站和荒田站2022年2月9日16：00—13日00：00的10 min平均风速（红色虚线为7级大风阈值）

Fig.5 The 10-minute average wind speed at Hulukou Bridge， Qiluogou， and Huangtian stations from 16：00 on 9 to 00：00 on 13 February， 2022 （The red dashed line is the threshold for a Force 7 gale）

图6 葫芦口大桥站、骑骡沟站和坝顶站2022年11、12月大风事件的滞后时间关系

Fig.6 Time lag relationship of strong wind events at Hulukou Bridge and Qiluogou with Dam Top stations in November and December 2022

图7 新田站2 min平均风速与荒田站与骑骡沟站气压差及荒田站与葫芦口大桥站气压差的散点图

Fig.7 Scatter plots of 2-minute average wind speed at Xintian Station versus pressure differences between Huangtian and Qiluogou stations， and between Huangtian and Hulukou Bridge stations

图8 新田站发生大风时峡谷区大风发生概率随荒田站与骑骡沟站气压差（a）及荒田站与葫芦口大桥站气压差（b）的变化

Fig.8 Variation in the probability of strong winds in the canyon area with changes in air pressure differences between Huangtian Station and Qiluogou Station (a), and between Huangtian Station and Hulukou Bridge Station (b) when strong wind occurring at Xintian Station

表2 新田站2023年1月大风事件参数检验

Tab.2 Parameter tests for strong wind events at Xintian Station in January 2023

大风日	最大气压差/hPa	风速增加时间与气压峰值时间差/h	风速增加到7级所需时间/h	3 h 变压/hPa	骑骡沟站和新田站风速峰值时间差/h	葫芦口大桥站和新田站风速峰值时间差/h	荒田站和葫芦口大桥站气压差/hPa	满足条件的参数个数
1日	5.4	6*	2*	-1.5	-2*	-2*	17.8	4
3日	8.7*	3*	4*	-2.7*	0	1	18.5*	5
4日	7.3	5*	3*	-4.7*	-1*	-2*	20.1*	6
11日	9.5*	0	4*	0.9	1	1	14.1	2
12日	7.2	10	3*	0.6	1	-1*	14.7	2
18日	12.5*	6*	2*	-6.7*	1	2	18.2*	5
19日	8.6*	3*	1*	-3.2*	-3*	-3*	18.6*	7
20日	7.5	4*	1*	-5.0*	-2*	-2*	19.6*	6
21日	14.9*	10	7	-0.1	2	1	20.8*	2

参考文献 25

[1]	曹鸿兴, 郑艳, 虞海燕, 等, 2008. 气候检测与归因的格兰杰检验法[J]. 气候变化研究进展, 4(1): 37-41.
[2]	曹辉, 张继顺, 陈翠华, 2018. 白鹤滩水电站水文要素特征分析[J]. 水力发电, 44(6): 35-37.
[3]	陈海涛, 林敏, 张运芝, 等, 2017. 营口地区冷锋过境引发的大风天气分析[J]. 现代农业科技(19): 226-227.
[4]	陈文龙, 2018. 白鹤滩水电站极端大风天气成因分析[J]. 农业与技术, 38(23): 138-141.
[5]	付桂琴, 赵春生, 张杏敏, 等, 2015. 1961—2010年河北省地面风变化特征及成因探讨[J]. 干旱气象, 33(5): 815-821. DOI
[6]	高梦竹, 刘松涛, 陈雪, 2019. 一次春季大风天气过程成因及预报着眼点分析[J]. 黑龙江气象, 36(4): 8-12.
[7]	李旭, 王式功, 尚可政, 等, 2015. 基于地面气象要素的延伸期相似预报方法初步研究[J]. 干旱气象, 33(2): 324-331. DOI
[8]	刘劲龙, 徐刚, 杨娟, 2013. 利用格兰杰因果检验法分析重庆极端气温事件及其与平均气温和降水量变化的关系[J]. 中国农业气象, 34(2): 236-242.
[9]	刘兰, 刘莉, 刘盛梅, 2021. 2020年6月20日14时至21日石河子大风天气分析[J]. 石河子科技(3): 3-5.
[10]	潘留杰, 梁绵, 祁春娟, 等, 2023. “十四运”关键场馆气象要素特征及客观预报检验[J]. 干旱气象, 41(3): 491-502. DOI
[11]	谭志强, 桑建人, 纪晓玲, 等, 2017. 宁夏一次大风扬沙天气过程机制分析[J]. 干旱区地理, 40(6): 1 134-1 142.
[12]	王黉, 李英, 吴哲红, 等, 2019. 我国大风机理研究和预报技术进展[J]. 气象科技, 47(4): 600-607.
[13]	王艳兰, 王娟, 伍静, 等, 2021. 桂林市2019年“3·21”极端大风与“4·24”致灾冰雹过程特征对比分析[J]. 热带气象学报, 37(2): 175-185.
[14]	吴琼, 徐卫民, 2019. 湖陆山地复杂地形下近地层风速预报研究[J]. 干旱气象, 37(3): 384-391.
[15]	胥良, 2004. 金沙江白鹤滩水电站金江滑坡成因机制及稳定性研究[D]. 成都: 成都理工大学.
[16]	姚望玲, 陈正洪, 向玉春, 2010. 武汉市气候变暖与极端天气事件变化的归因分析[J]. 气象, 36(11): 88-94.
[17]	尹晔, 王汉涛, 程志刚, 等, 2023. 大型水电站灾害性大风特征及预报:以白鹤滩水电站为例[M]. 北京: 气象出版社:31-46.
[18]	张贤亮, 尚尚, 魏海峰, 2019. 基于LOWESS回归的一阶海杂波识别算法[J]. 电子器件, 42(2): 458-462.
[19]	张晓芳, 李宇华, 2010. 春季大风的预报和分析[J]. 黑龙江气象, 27(2): 11-14.
[20]	周淑玲, 单宝臣, 盛春岩, 2014. 山东近海温带气旋强南向大风的特征分析[J]. 海洋通报, 33(2): 132-139.
[21]	朱男男, 刘彬贤, 孙密娜, 等, 2016. 引发渤海风暴潮一次江淮气旋北上过程诊断分析[J]. 气象与环境学报, 32(5): 10-17.
[22]	朱乾根, 林锦瑞, 寿绍文, 等, 2007. 天气学原理与方法[M]. 4版. 北京: 气象出版社.
[23]	CHARBA J, 1974. Application of gravity current model to analysis of squall-line gust front[J]. Monthly Weather Review, 102(2): 140-156.
[24]	DROEGEMEIER K K, WILHELMSON R B, 1987. Numerical simulation of thunderstorm outflow dynamics. part I: Outflow sensitivity experiments and turbulence dynamics[J]. Journal of the Atmospheric Sciences, 44(8): 1 180-1 210.
[25]	GRANGER C W J, 1969. Investigating causal relations by econometric models and cross-spectral methods[J]. Econometrica, 37(3): 424-438.