| 摘要: |
| 【目的】自2008年我国南方遭遇历史罕见冰冻灾害以来，结冰预报逐渐成为气象服务的焦点。1961—2020年威宁县年均结冰日达61 d，是贵州省同期年均结冰日的4.3倍，为全省之冠。为加深认识本地结冰现象的发生、发展规律，本文从气候角度分析了近60 a来威宁县结冰日时序演变特征并对其趋势进行预测。【方法】选取1961—2020年威宁县国家基准气候站逐月结冰日和气温观测资料，采用M—K突变检验、R/S分析、Morlet小波分析等数理统计方法对结冰日进行气候特征分析，得出结冰日与气温的对应关系，并进一步对年结冰日做出趋势预测分析。【结果】（1）99.1%的结冰日集中在11月—翌年3月，逐月平均结冰日呈“单峰型”分布，1月结冰日最多。不同气候背景下的年结冰日表现为冷期平均年结冰日明显多于暖期，冷期年结冰日减少率几乎与整个研究期保持一致，而暖期明显强于冷期（整个研究期）。（2）年结冰日演变过程中的主周期主要为26、12、5、3 a，分别为第一、二、三、四主周期，4种时间尺度共同影响着威宁县年结冰日的年际和年代际周期变化。（3）对第一和第二主周期下的小波系数进行拟合，重构小波系数与年结冰日具有很好的一致性，两者的阶段顶底偏差集中在1—3 a。【结论】总的来看，威宁县年结冰日与同期平均气温表现为反相关关系且总体呈减少趋势，这种趋势也会持续影响未来。预计2021—2030年平均结冰日约为50.6d，年结冰日在2027年附近达到最多但大概率不会创历史新高。 |
| 关键词: 结冰日；R/S分析；Morlet小波分析；趋势预测 |
| DOI： |
| 投稿时间：2023-12-06修订日期：2024-08-16 |
| 基金项目:贵州省气象局科研业务登记制项目（黔气科登[2024]4—13号） |
|
| The spatiotemporal distribution characteristics and trend prediction analysis of freezing days in Weining County from 1961 to 2020 |
| caijun,linxuefei |
| (Weining Meteorological Bureau;Zunyi Meteorological Bureau) |
| Abstract: |
| Since the rare historical freezing disaster in southern China in 2008, ice forecast has gradually become the focus of meteorological services. From 1961 to 2020, the average annual freezing day in Weining County reached 61 days, which is 4.3 times that of Guizhou Province during the same period and the highest in the province. To deepen our understanding of the occurrence and development patterns of local icing phenomena, this article analyzes the temporal evolution characteristics of icing in Weining County over the past 60 years from a climate perspective and predicts its trends. This article selects monthly icing days and temperature observation data from the national benchmark climate station in Weining County from 1961 to 2020. Mathematical statistical methods such as M—K mutation test, R/S analysis, and Morlet wavelet analysis are used to analyze the climate characteristics of icing days. The corresponding relationship between icing days and temperature is obtained, and further trend prediction analysis is made for annual icing days. (1) 99.1% of the freezing days are concentrated from November to March of the following year, and the monthly average freezing days show a "unimodal" distribution, with January having the most freezing days. The average annual icing days under different climate backgrounds are significantly higher in the cold season than in the warm season. The reduction rate of annual icing days during the cold season is almost consistent with the entire study period, while the warm season is significantly stronger than the cold season (the entire study period). (2) The main periods in the evolution process of annual icing days are 26, 12, 5, and 3a, which are the first, second, third, and fourth main periods, respectively. The four time scales together affect the interannual and interdecadal periodic changes of annual icing days in Weining County. (3) Fit the wavelet coefficients under the first and second main periods, and the restructure wavelet coefficients have good consistency with the annual icing day. The top and bottom deviations of the two stages are concentrated in 1—3a. Overall, there is an inverse correlation between the annual freezing day and the average temperature during the same period in Weining County, and the overall trend is decreasing. This trend will continue to affect the future. It is expected that the average freezing day from 2021 to 2030 will be about 50.6 days, and the annual freezing day will reach its maximum near 2027, but it is highly likely that it will not reach a historical high. |
| Key words: icing day; R/S analysis; Morlet wavelet analysis; trend prediction |
