佳木斯地区冬、春季大-暴雪成因分析

冬、春季的天气影响系统既有共同点，又有不同之处。通过近20a大-暴雪个例的物理成因及卫星云图等特征的分析，总结出有预报意义的指标。     

用SSM/I微波遥感图像分析海上台风的螺旋云带

文中介绍了美国国防气象卫星专用微波成像仪 (SSM /I)上各通道的特性 ,分析了大气中各种粒子 (尤其是云滴和降水滴 )对各通道辐射的吸收和散射效应。通过对SSM /I图像上台风单通道剖面、双通道散点图的分析 ,揭示了台风在微波图像上表现形式的内在物理原因。在此基础上 ,设计了一个降水指数 ,方法是 :将 85 .5GHz的吸收段对称拉伸到散射段的延长线上 ,然后求归一化后的 19.35 ,37.0GHz和经拉伸处理的 85 .5GHz图像 3者的平均值。 3个通道合成降水指数克服了 37.0GHz对大雨滴不敏感 ,和 85 .5GHz对中等大小雨滴不敏感的缺点 ,比原始单通道微波图像更清楚地显示了台风的螺旋云带结构     

乌鲁木齐雷暴天气的云图分析

通过对1997－1999年乌鲁木齐发生的11次雷暴天气的分析表明，雷暴天气发生于850hPa上反气旋环流的东南边缘的高能量区，雷暴云图由孤立的对流云图和锋面及涡旋云系中的对流云图发展而成，其形状不规则，生命史4－8h，当雷暴云图成熟时产生雷暴天气现象。     

改进的切线性模式对一个边界层模式变分资料同化的改善

大气、海洋中边界层物理过程的强非线性会对切线性模式近似（尤其是当积分时间长时）有较大的影响，从而给相应的4维变分资料同化问题的求解造成困难。本构造了Mellor-Yamada湍流封闭模式（level 2.5)的一个改进的切线性模式，相比通常的切线性模式和简化的切线性模式可以提高对非线性扰动的逼近。利用这个改进的切线性模式的伴随模式分别进行了1到7天的变分资料同化试验，得到了满意的结果。而用通常的伴随模式和简化的伴随模式都无法得到令人满意的结果。     

北疆夏季两次冰雹天气过程的对比分析

北疆夏季两次冰雹天气过程的对比分析发现，充足的水汽、足够大的上升运动和不稳定的层结果冰雹天气出现的必备条件，而降雹的强弱则与对流云图的发展和强度有关。     

南疆西部山区一次特大暴雨特征分析

通过对2001年9月22－23日南疆西北部山区特大暴雨的环流背景、物理量场、卫星云图分析，揭示了这场特大暴雨的物理机制和各要素场特征。     

Soluble organic compounds in fog and cloud droplets: what have we learned over the past few years?

We have recently set up a new procedure for characterising the water soluble organic compounds (WSOC) in fog water, for which information is still rather limited. Fog samples collected during the 1998–1999 fall–winter season in the Po Valley (Italy) were analysed following this procedure, which allows a quantitative determination of three main classes of organic compounds (neutral species, mono- and di-carboxylic acids, polycarboxylic acids), together accounting for ca. 85% of the total WSOC. This procedure also provides information on the main chemical characteristics of these three classes of compounds (functional groups, aliphatic vs. aromatic character, etc.). The enhanced chemical knowledge on fog/cloud chemical composition opens new scenarios as far as chemical and microphysical processes in clouds and fogs are concerned.     

Seismic response of self‐centring hysteretic SDOF systems

The seismic response of single‐degree‐of‐freedom (SDOF) systems incorporating flag‐shaped hysteretic structural behaviour, with self‐centring capability, is investigated numerically. For a SDOF system with a given initial period and strength level, the flag‐shaped hysteretic behaviour is fully defined by a post‐yielding stiffness parameter and an energy‐dissipation parameter. A comprehensive parametric study was conducted to determine the influence of these parameters on SDOF structural response, in terms of displacement ductility, absolute acceleration and absorbed energy. This parametric study was conducted using an ensemble of 20 historical earthquake records corresponding to ordinary ground motions having a probability of exceedence of 10% in 50 years, in California. The responses of the flag‐shaped hysteretic SDOF systems are compared against the responses of similar bilinear elasto‐plastic hysteretic SDOF systems. In this study the elasto‐plastic hysteretic SDOF systems are assigned parameters representative of steel moment resisting frames (MRFs) with post‐Northridge welded beam‐to‐column connections. In turn, the flag‐shaped hysteretic SDOF systems are representative of steel MRFs with newly proposed post‐tensioned energy‐dissipating connections. Building structures with initial periods ranging from 0.1 to 2.0s and having various strength levels are considered. It is shown that a flag‐shaped hysteretic SDOF system of equal or lesser strength can always be found to match or better the response of an elasto‐plastic hysteretic SDOF system in terms of displacement ductility and without incurring any residual drift from the seismic event. Copyright © 2002 John Wiley & Sons, Ltd.     

有限元分析方法在黄土地基震陷预测中的应用

将有限元方法应用于黄土地基的震陷量预测，定量地研究了黄土地基地震沉陷特性与场地条件、基础形式、土动力特性等因素的相关关系，为黄土地基基础的抗震增稳设计与处理提供了科学依据。     

Fuzzy neural network models for liquefaction prediction

Integrated fuzzy neural network models are developed for the assessment of liquefaction potential of a site. The models are trained with large databases of liquefaction case histories. A two-stage training algorithm is used to develop a fuzzy neural network model. In the preliminary training stage, the training case histories are used to determine initial network parameters. In the final training stage, the training case histories are processed one by one to develop membership functions for the network parameters. During the testing phase, input variables are described in linguistic terms such as ‘high’ and ‘low’. The prediction is made in terms of a liquefaction index representing the degree of liquefaction described in fuzzy terms such as ‘highly likely’, ‘likely’, or ‘unlikely’. The results from the model are compared with actual field observations and misclassified cases are identified. The models are found to have good predictive ability and are expected to be very useful for a preliminary evaluation of liquefaction potential of a site for which the input parameters are not well defined.