MULTI-HIERARCHICAL STRUCTURE AND JUMP OF TEMPERATURE FOR THE GLOBE, CHINA AND YUNNAN OVER THE PAST 100 YEARS

An analysis has been conducted of the multi-hierarchical structure and jump of temperature variation for the globe, China and Yunnan Province over the past 100 years using an auto-adaptive, multi-resolution data filter set up in You, Lin and Deng (1997). The result is shown below in three aspects. (11 The variation of global temperature in this period is marked by warming on a large scale and can be divided into three stages of being cold (prior to 1919), warm (between 1920 and 1978) and warmer (since 1979). Well-defined jumps are with the variation in correspondence with the hierarchical evolution on such scale, occurring in 1920 and 1979 when there is the most substantial jump towards warming. For the evolution on smaller scales, however, the variation has shown more of alternations of cold and warm temperatures. The preceding hierarchical structure and warming jump are added with new ones. (2) The trend in which temperature varies is much the same for China and the Yunnan Province, but it is not consistent with that globally, the largest difference being that a weak period of cold temperature in 1955-1978 across the globe was suspended in 1979 when it jumped to a significant warming,while a period of very cold temperature in 1955-1986 in China and Yunnan was not followed by warming in similar extent until 1987. (3) Though there are consistent hierarchical structure and jumping features throughout the year in Yunnan, significant changes with season are also present and the most striking difference is that temperature tends to vary consistently with China in winter and spring but with the globe in summer and fall.     

气象场序列几种插补方案的对比试验

采用基于 Ｅ Ｏ Ｆ Ｓ的主分量回归（ Ｐ Ｃ Ｒ）、 Ｅ Ｏ Ｆ Ｓ迭代法（ Ｉ Ｅ Ｏ Ｆ）和基于主分量典型相关的典型变量回归（ Ｃ Ｖ Ｒ）３ 种不同的统计插补计算方案,对同一区域同一种气象要素序列进行缺测资料的插补试验。结果表明,各种方案插补精度都与参数选择有关,无论缺测站点空间分布类型如何,当缺测点数小于 ６０ ％ 时,３ 种方案均有较好效果,以 Ｃ Ｖ Ｒ 最佳,且随缺测年数增长, Ｃ Ｖ Ｒ 优势更显著。     

中国近50年冬夏季极端气温场的年代际 空间型态及其演变特征研究

利用我国１９５３～１９９５年冬夏季平均最低气温（Ｔｍ）和平均最高气温（ＴＭ）资料,分析冬夏季ＴＭ、Ｔｍ距平场的年代际空间型态及其演变特征．指出冬夏季Ｔｍ、ＴＭ场的空间型态基本都在７０年代末至８０年代初发生显著变化,但空间型态变化的冬夏季差异十分明显,且冬季Ｔｍ、夏季ＴＭ空间型态的年代际变化在相应Ｔｍ、ＴＭ场的变化中占有明显优势．     

遥感与GIS的青海湖岸线分维研究

讨论了TM图像以及分形几何学对青海湖岸线问题的求解算法,同时估算了不同像元尺度下,青海湖的岸线长度得出青海湖岸线的分形维数为1.0237,并探讨了其地学意义,研究表明,分形维数不仅是一种可行的表达岸线特征的方法,而且岸线分维数的计算可大大减少工作量和提高工作效率,可见遥感与GIS技术在岸线分维数的计算中,具有十分良好的应用前景。     

基于EOF和小波分析的福建近四十年旱涝时空变化特征研究

采用经验正交函数分解(EOF)、快速富里叶变换(FFT)、连续 (CWT)和正交小波变换(OWT),对近41 a来福建省25个代表站不同雨季旱涝指数(Z指数)序列进行了分析.结果表明:(1) 福建旱涝在1965-1975及1990年代具有显著的2-3 a周期;(2) 在1980年代中期以后南北反向变化具有显著的1 a和3-4 a周期;(3) 中西部与其它区域的反向变化,在1985-1998年间有显著的1-2 a周期,在1980年代以后9-13 a周期较强;(4) 近40年来有干旱趋势,南部(东部)比北部(西部)明显,其中1990年代这种趋势显著.(5) 1960年代和1980年代相对湿润,而1970年代和1990年代则相对干旱.     

El Ni?o区域SST的空间结构

本文利用MHAT小波对1970～1990年El Ni?o 区域的SST进行多尺度分析，研究结果发现:(1) 在洲洋尺度(4×10 3～13×10 3 km)里，El Ni?o区域存在着特征波长为6000 km的纬向长波#Aψ*T/2@#a和波长为6500 km的经向长波ψ1；(2) 在区域尺度(1×10 3～4×10 3 km)里，El Ni?o区域的SST存在着特征波长为4500 km的经向中波ψ3和波长为3500 km的纬向中波ψ4；(3) 大尺度范围的海温异常及4波共振机制是发生El Ni?o现象的两个重要原因。     

亚洲夏季风与福建汛期降水的关系分析

采用EOF（Emprical Orthogonal Function）、小波分析和交叉小波分析等方法研究了福建省汛期降水的不同空间分布型在不同的时间尺度下与西北太平洋夏季风、印度洋夏季风的关系.结果表明：福建汛期降水空间分布以全省一致型和东南—西北反相型两种模态为主;西北太平洋夏季风主要影响福建整个汛期降水总量的多寡,当西太平洋夏季风较强（弱）时,福建整个汛期全省降水偏多（少）;而印度洋夏季风则主要影响福建汛期降水量东南-西北向差异,即ISM强（弱）,福建东南-西北部降水差异增大（减小）.     

SPATIAL/TEMPORAL FEATURES OF DROUGHT/FLOOD IN FUJIAN FOR THE PAST FOUR DECADES

41a (1961 - 2001) seasonal Z index series of 25 representative weather stations are investigated by virtue of EOF, FFT, continuous wavelet transformation (CWT) and orthogonal wavelet transformation (OWT). It shows that: (1) Fujian drought/flood (DF) has a significant 2 - 3a cycle for the periods 1965 - 1975 and 1990's; (2) the pattern, which represents the opposite DF trend between the southern and northern parts, has 1a and 3 - 4a cycles since the middle of 1980's; (3) EOF3, which denotes the reverse change between the middle-west region and other areas, has significant 1 - 2a cycle for the period from 1985 to 1998 and 9 - 13a cycle since 1980s; (4) there is an obvious drought trend for the last 40a (especially in the 1990's), which is more outstanding in the south (east) than in the north (west); (5) the 1960's and 1980's are in relatively wet phases and the 1970's and 1990's are in drought spells.