基于PC-CCA方法的气象场资料插补试验

引入主成分典型相关分析（ＰＣ－ＣＣＡ）方法建立缺测气象要素场序列的插补模式,对区域性气象场序列（以长江流域月气温距平场为例）各种时空缺测分布型态作插补试验。结果表明,当插补场站网分布型取包含子场在内的混合站网时,建模样本量达２５年就可有最优插补精度,且性能稳定,效果优良。尤其当距平符号单一且大尺度分量占优势时,插补精度随缺测场变化很小     

基于PC-CCA方法的气象场资料插补试验

引入主成分典型相关分析(PC-CCA)方法建立缺测气象要素场序列的插补模式,对区域性气象场序列（以长江流域月气温距平场为例）各种时空缺测分布型态作插补试验。结果表明,当插补场站网分布型取包含子场在内的混合站网时,建模样本量达25年就可有最优插补精度,且性能稳定,效果优良。尤其当距平符号单一且大尺度分量占优势时,插补精度随缺测场变化很小。     

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。     

影响山西省降水的物理因子分析

本文采用滑动相关普查及多信度统计检验方法,利用近30多年来山西30个代表站点5月至10月总降水量与前期北半球500hPa月平均高度场(5°×10°格点),北太平洋月平均海温场(5°×5°格点)以及74个环流特征量作逐点、逐年、逐月滑动相关普查分析,找出了显著影响山西省降水稳定可靠的关键区因子,并对这些物理因子作了分析。     

一种基于SVD的迭代方法及其用于气候资料场的插补试验

提出一种基于SVD的迭代对气象场序列缺测记录插补延长的技术方法,对长江流域20个测站1月份气温做插补试验,平均均方误差为0.25,插补精度明显优于迭代EOF,插补效果良好且性能稳定;而且插补站数所占比例越小效果越好.此研究表明,基于SVD迭代的插补方法是一种非常有效的插补途径.     

气象格点资料预测压缩方法初探

通过二维线性预测方法来降低气象格点资料间的冗余度,解决气象格点资料在存储时所占内存空间太大和在网络传输过程中的带宽问题。从原理及500hPa高度场分析得出:预测误差序列的方差比原序列方差小一个数量级,这充分说明和原序列相比,误差序列的相关性已有很大的降低,证实了格点资料预测压缩是完全可以实现的。     

西安观测站长序列气温资料缺测记录插补和非均一性检验

以西安观测站1971—2013年日平均气温、最高气温和最低气温序列为研究对象,利用标准序列法和多元线性回归法进行插补实验,计算插补值与实测值的平均误差、平均绝对误差、均方根误差和插补值与实测值误差在0.5℃以内的样本比例,对比分析两种插值方法的相对优劣。结果表明：多元线性回归法插补得到的气温序列效果好于标准序列法,并且气候趋势特征与实际观测值序列更具一致性。采用t检验法、惩罚最大T检验（PMTT）、惩罚最大F检验（PMFT）对西安站1951—2020年平均气温序列的均一性进行检验。依据台站历史沿革数据进行的t检验,在6次台站历史沿革变化中,只有2次造成了年平均气温和年平均最高气温序列间断,分别由观测时次增加和仪器换型导致;年平均最低气温有4次出现间断,分别由台站站址迁移、观测时次增加、仪器换型、缺测值插补造成。PMTT和PMFT检测中发现的4次间断点因无元数据支持,认为属于合理间断点,这2种方法均未检测出因缺测值插补引起的间断点,一定程度上说明采用多元线性回归法对缺测值插补得到的西安站1951—2020年气温序列相对合理,气温序列的均一性较好。     

气象格点资料预测压缩方法初探

通过二维线性预测方法来降低气象格点资料间的冗余度，解决气象格点资料在存储时所占内存空间太大和在网络传输过程中的带宽问题。从原理及500hPa高度场分析得出：预测误差序列的方差比原序列方差小一个数量级，这充分说明和原序列相比，误差序列的相关性已有很大的降低，证实了格点资料预测压缩是完全可以实现的。     

1996年我国南方洪涝的海气环境场分析

选用 1 95 1～ 1 996年中国 1 6 0站逐月降水资料 ,太平洋逐月平均网格点海温资料( 1 0°S～ 5 0°N,1 2 0～ 80°E) ,逐月大尺度环流指数资料 ,分析并探讨造成 1 996年南方洪涝的环流特征、气候背景及海温影响机制。结果表明 :大气环流指数与前期太平洋海温场有明显的遥相关关系。冬、春季海温的冷暖变化程度 ,可预示后期 7月各环流系统的趋势变化 ,环流系统指数自身演变存在较明显的年代际变化规律 ,1 996年南方洪涝就是在这种气候背景下形成的。海温冷暖变化影响大气环流系统的这种演变关系 ,对南方洪涝的短期气候预测具有一定的指示意义。     

应用500hPa高度场、太平洋海温格点资料建立统计预报方程的试验

气候变化与大气环流、海面温度的变化关系密切。因此,应用５００ｈＰａ月平均位势高度场、太平洋海温格点资料,通过因子筛选建立了广西单站及区域气候要素的预报方程。１　初选预报因子　　采用５～３６０°Ｅ、１０～８５°Ｎ区域,１０°×５°经纬度,共５７６个格点的５００ｈＰａ月平均位势高度场资料。资料年代为１９５１～１９９６年,其数值为实际值减去５００位势什米。北太平洋海温资料为１０°。～５０°Ｎ,５°×５°经纬度,共２８６个格点,资料年代１９５１～１９９６年。初选预报因子分两步进行：第一步,进行网格点普查…     

由最高最低气温求算的平均气温对我国年平均气温序列影响

针对研究全国近百年平均气温长期变化的实际需要，利用603个测站1961—2002年气温观测资料，比较分析了最高最低平均气温距平序列和4次观测记录平均气温距平序列的差异，讨论了最高、最低气温变化趋势。结果表明:两种统计方法得到的平均气温距平序列及增温速率的差异均不明显，在一定条件下两者可以互相替换。此外，最高、最低气温变化普遍存在不对称现象，且可分为4种类型，这种不对称性对平均气温变化速率并没有明确一致的影响。     

Multidecadal variability in local growing season during 1901–2009

Global warming exerts a lengthening effect on the growing season, with observational evidences emerging from different regions over the world. However, the difficulty for a global overview of this effect for the last century arises from limited availability of the long-term daily observations. In this study, we find a good linear relationship between the start (end) date of local growing season (LGS) and the monthly mean temperature in April (October) using the global gridded daily temperature dataset for 1960–1999. Using homogenized daily temperature records from nine stations where the time series go back to the beginning of the twentieth century, we find that the rate of change in the start (end) date of the LGS for per degree warming in April (October) mean temperature keeps nearly constant throughout the time. This enables us to study LGS changes during the last century using global gridded monthly mean temperature data. The results show that during the period 1901–2009, averaged over the observation areas, the LGS length has increased by a rate of 0.89 days decade^?1, mainly due to an earlier start (?0.58 days decade^?1). This is smaller than those estimates for the late half of the twentieth century, because of multidecadal climate variability (MDV). A MDV component of the LGS index series is extracted by using Ensemble Empirical Mode Decomposition method. The MDV exhibits significant positive correlation with the Atlantic Multi–decadal Oscillation (AMO) over most of the Northern Hemisphere lands, but negative in parts of North America and Western Asia for start date of LGS. These are explained by analyzing differences in atmospheric circulation expressed by sea level pressure departures between the warm and cool phases of AMO. It is suggested that MDV in association with AMO accelerates the lengthening of LGS in Northern Hemisphere by 53 % for the period 1980–2009.     

Recent trends in regional air temperature and precipitation and links to global climate change in the Maharlo watershed,Southwestern Iran

Trends in air temperature and precipitation data are investigated for linkages to global warming and climate change. After checking for serial correlation with trend-free pre-whitening procedure, the Mann–Kendall test is used to detect monotonic trends and the Mann–Whitney test is used for trend step change. The case study is Maharlo watershed, Southwestern Iran, representing a semi-arid environment. Data are for the 1951–2011 period, from four temperature sites and seven precipitation sites. A homogeneity test investigates regional similarity of the time series data. The results include mean annual, mean annual maximum and minimum and seasonal analysis of air temperature and precipitation data. Mean annual temperature results indicate an increasing trend, while a non-significant trend in precipitation is observed in all the stations. Furthermore, significant phase change was detected in mean annual air temperature trend of Shiraz station in 1977, indicating decreasing trend during 1951–1976 and increasing trend during 1977–2011. The annual precipitation analysis for Shiraz shows a non-significant decrease during 1951–1976 and 1977–2011. The result of homogeneity test reveals that the studied stations form one homogeneous region. While air temperature trends appear as regional linkage to global warming/global climate change, more definite outcome requires analysis of longer time series data on precipitation and air temperature.     

1961-2007年台安县气候变化特征分析

在全球变化背景下,台安县气候变化地域特征明显。本文利用1961—2007年的气候资料,通过气候倾向率和Mann-Kendall检测等方法,研究了台安县气候变化特征。结果表明:近47 a来,台安县气候明显变暖,平均气温以0.34℃/10 a的速度上升,积温净增加约260 ℃或以上,冬季增温对气候变暖的贡献最大,直接导致冬季最大冻土深度以3.6 cm/10 a速度变浅。年降水量出现减少趋势,春、夏季降水量减少比较明显,极端降水日数无变化规律。四季日照时数明显减少,不利于绿色植物的光合作用。近47 a间,除了年平均最高气温和降水量外,平均气温、年日照时数、平均最低气温、最大冻土深度、大于等于10 ℃积温和大于等于0 ℃积温等要素都发生了显著突变,但突变时间存在差异。     

Global warming mode of atmospheric circulation

Variations of global zonal-mean atmospheric circulation are studied using the National Center of Environment Prediction/National Center of Atmosphere Research (NCEP/NCAR) reanalysis data set from January 1948 to February 2001 and CCM3.6 (Community Climate Model Version 3.6) simulations for the period 1900–1999. Empirical orthogonal function (EOF) analysis indicates that variations of zonal-mean geopotential height in the tropics are usually opposite to those over the subpolar zone in both hemispheres. The first mode of EOF analysis shows that height is higher than normal in the tropics when it is lower over the subpolar zone in both hemispheres with much stronger westerlies over mid latitudes in both hemispheres, and vice versa. This mode explains about 50% variance and is predominant in the whole troposphere. The time series of EOF1 has a sharp transition near about 1977 and the polarity changes from negative to positive. This mode is closely related to the variations of global mean surface air temperature. The detrended correlation coefficient between EOF1 time series and the surface air temperature is 0.74 in the boreal winter. Furthermore, the lowest correlation coefficient among the other three seasons, annual mean, seasonal mean and monthly mean is higher than 0.42 which indicates the fairly good relationship between this mode and the surface air temperature. This result has been verified using CCM3.6 simulations.     

全球平均温度序列的比较

着重评述了目前国际上常用的3个全球平均温度序列：HadCRUT3、NCDC及GISS。后两个序列1982年之后引入了卫星观测资料。NCDC对资料作了插补,建立了1880-2009年覆盖面完整的序列。HadCRUT3序列所显示的近10年变暖的停滞可能主要是缺少极区资料造成的。NCDC与GISS两个序列依然保持约0.1℃/10a的增温。3个序列1910-2009年的变暖趋势十分接近,在0.70～0.75℃/100a之间。     

多窗谱分析方法及其在全球变暖研究中的应用

多窗谱分析方法是一种低方差、高分辨的谱分析方法 ,尤其适合于非线性气候系统中高噪声背景下弱信号、时频演变信号的诊断分析。文中简要地介绍了多窗谱分析方法的基本原理 ,并将其用于近 1 5 0 a来全球及南、北半球温度变化的研究。结果表明 :(1 )在温度呈线性变暖的背景趋势下 ,北半球、全球年平均温度还具有显著的 40～ 70 a的准周期低频振荡 ,它们与变暖过程中的波动性密切相关 ;(2 )年代际振荡信号在北半球温度序列中也相当明显 ,南半球则存在多种低于 5 a尺度的显著振荡周期 ,半球或全球 ENSO振荡分量、QBO周期信号的振幅都具有缓变包络的特性 ;(3 )与传统功率谱估计、最大熵谱估计结果的对比表明 ,多窗谱分析方法方法得到的谱估计分辨力高 ,稳定性强。     

Long-term trend in potential vorticity intrusion events over the Pacific Ocean: Role of global mean temperature rise

In this study, we examine a long-term increasing trend in subtropical potential vorticity (PV) intrusion events over the Pacific Ocean in relation to the global mean temperature rise, based on multiple reanalysis datasets. The frequency of the PV intrusions is closely related to the upper-tropospheric equatorial westerly duct and the subtropical jet (STJ). An overall strengthening of the westerly duct and weakening of the STJ are found to be driven by the warming-induced strengthening of Walker circulation and regional changes in Hadley circulation on multi-decadal timescale, leading to an increase in the PV intrusion frequency over the tropics. The results are robust in all datasets. The multi-decadal strengthening in the Pacific Walker circulation is consistent with the global mean temperature rise. In this way, the PV intrusions are correlated with the warming related global mean temperuate rise. When the interannual variability of ENSO is removed from the intrusion time series, the long-term trend in PV intrusions due to external forcing associated with anthropogenic warming (global mean temperature rise) becomes clearer. The link between the global mean temperature rise and intrusion frequency is further verified by performing a correlation analysis between the two. The significant (> 95%) correlation coefficient is 0.85, 0.94, 0.84, 0.83, and 0.84 for ERA-40, ERA-Interim, NCEP-NCAR, JRA-55, and JRA-25, respectively. This unequivocally indicates that the global mean temperature rise can explain around 69%–88% of the variance related to the long-term increase in PV intrusion frequency over the Pacific Ocean.     

Modeling fractionally integrated maximum temperature series in India in presence of structural break

In this study, the long memory behaviour of monthly maximum temperature of India for the period 1901 to 2007 is investigated. The correlogram of the series reveals a slow hyperbolic decay, a typical shape for time series having the long memory property. Wavelet transformation is applied to decompose the temperature series into time–frequency domain in order to study the local as well as global variation over different scale and time epochs. Significant increasing trend is found in the maximum temperature series in India. The rate of increase in maximum temperature accelerated after 1960s as compared to the earlier period. Here, an attempt is also made to detect the structural break for seasonally adjusted monthly maximum temperature series. It is found that there is a significant break in maximum temperature during July, 1963. Two-stage forecasting (TSF) approach to deal with the coexistence of long memory and structural change in temperature pattern is discussed thoroughly. The forecast performance of the fitted model is assessed on the basis of relative mean absolute prediction error (RMAPE), sum of squared errors (SSE) and mean squared errors (MSE) for different forecast horizons.     

气温的天气和气候记忆性特征分析:以济南和青岛为例

气温的天气和气候记忆性特征分析对于提高气候预测水平具有积极意义。利用济南和青岛1961—2020 年逐日、月和年平均气温资料,运用自相关性函数和标准化频率分布分析了上述时间序列的气温记忆性特征和概率分布特征,并利用结构函数法建立了月、年平均气温距平与日平均气温距平之间的分数阶导数关系。结果表明:(1)济南和青岛的月、年平均气温距平呈现不同程度的记忆性特征,其中年平均气温距平相比于月平均气温距平具有更好的记忆性。(2)济南和青岛的月、年平均气温距平与日平均气温距平之间存在分数阶导数关系,济南和青岛相应的月、年尺度阶数分别为0. 529、0. 665 和0. 553、0. 791,两地的月尺度阶数相近,但青岛略大,青岛的年尺度阶数大于济南,即青岛月和年平均气温距平的记忆性大于济南。(3)济南和青岛的月和年平均气温距平相比于日平均气温距平有不同程度的长尾特征,长尾特征反映了极值温度发生的概率。