Forecasting Monsoon Precipitation Using Artificial Neural Networks

This paper explores the application of Artificial Intelligent (AI) techniques for climate forecast. It pres ents a study on modelling the monsoon precipitation forecast by means of Artificial Neural Networks (ANNs). Using the historical data of the total amount of summer rainfall over the Delta Area of Yangtze River in China, three ANNs models have been developed to forecast the monsoon precipitation in the corre sponding area one year, five-year, and ten-year forward respectively. Performances of the models have been validated using a ＇new＇ data set that has not been exposed to the models during the processes of model development and test. The experiment results are promising, indicating that the proposed ANNs models have good quality in terms of the accuracy, stability and generalisation ability.     

Desalinization of running waters: III. Changes in the structure of diatom assemblages caused by a decreasing salt load and changing ion spectra in the river Wipper (Thuringia, Germany)

Salt-loaded effluents were introduced into the river Wipper during the mining period for almost a century. Beginning with the year 1990, the waste water load was strongly reduced due to the termination of the potash industry. Prior to 1990, monthly means of the chloride concentrations at times exceeded 6,000 mg l⁻¹ in the strongly polluted sections. Maximum concentrations reached twice these values. Up to 1998, mean annual chloride concentrations decreased to values below 2,000 mg l⁻¹. This led to more balanced fluctuations in salinity which had been pronounced before, depending on discharge and short-term changes in production. Similarly, the physiologically adverse ion conditions improved due to decrasing potassium and increasing calcium proportions.

In 1963/64, 1986 and 1998, samples of epilithic, epiphytic and epipsammic diatoms were taken at locations of different salinities along the river and examined for the effects of the salinization on the structure of the diatom assemblages. These structures changed in dependence on salinity. Increasing salt concentrations coincided with decreasing oligohalophilic and increasing mesohalophilic and polyhalophilic species numbers. Above a chloride concentration of about 3,000 mg l⁻¹, the proportion of the latter exceed that of the former (halobion index > 50). Corresponding to different conditions of salinization along the river, characteristic diatom assemblages occur differring from each other and which are specific for the river section. Spring and autumn aspects of the diatom assemblages show also salt-dependent differences. The assemblages found in 1998 after decrease of salinization have changed markedly in comparison to those from 1963/64 and 1986. Halobiontic species predominating formerly occurred only occasionally or not at all. They were replaced by oligohalobic-indifferent forms.

An ecological assessment of the changes was performed based on the halobion index calculated from all the samples. For the strongly salinized section of the river Wipper, a shift from -mesohalobic/polyhalobic conditions in 1963/64 and 1986 to -oligohalobic/β-mesohalobic conditions in 1998 was found. However, constant -oligohalobic conditions are still not given. With regard to the transition from -oligohalobic (limnetic) to β-mesohalobic (brackish) conditions, a maximum chloride concentration of 600 mg l⁻¹ was found. To guarantee -oligohalobic conditions, a maximum chloride concentration of 400 mg l⁻¹ should not be exceeded.     

八家子银多金融矿矿田成矿规律与成矿预测

八家子矿田由NW、NE向两组断裂带与孤形推覆构造交织，控制构成银铅锌、银铁矿床成矿系列，基成矿组份具有组合分带特征。预测东山－冰沟银、铅、锌、炉沟钼和冰沟里银多金属找矿区，经验证，获得找矿成效。     

柴达木盆地CXQ构造油气化探特征及找矿预测

通过在CXQ构造上进行的油气化探结果表明深部拥有油气活动，结合地质构造综合评价认为它具有形成构造圈闭油气藏的条件。为下一步勘探提供了地球化学依据。     

SPECTRUM期间目标台风移动的数值试验

在目标台风Ｙａｎｃｙ的移动过程中，选择４个时段进行预报数值试验。模式初始场系根据ＳＰＥＣＴＲＵＭ和ＴＣＭ－９０加强观测资料确定。结果表明：实际大气中目标台风移动的若干特征在模式大气中能够显示出来，如Ｙａｎｃｙ在１９９０年８月２０日００时移向的转折现象等；台风非对称结构的低频振荡是产生移向转折现象的一个重要原因。     

城乡火险预报模式探讨

文章探讨了建立火险预报模式的必要性和可行性，分析了1981～1990年辽宁省火灾次数与风速、月降水量和相对湿度的关系。并从概念模式出发，讨论了与火险有关的逐因子，建立了定量的火险预报模式，试用效果较好。     

形变,应变短临前兆标志体系的应用研究

文献「３」中给出了标志体系的整体框架结构和各标志的内容，本文讨论标志体系的进一步应用问题。首先讨论了异常信息合成的几种形式，然后根据标志体系的特点，给出了前兆追踪分析的方案；最后将专家系统的有关理论瑟形变应变前兆标志体系相结合，给出了一个地震预报专家系统的逻辑结构。     

宏观地震忽略吸收确定震源深度问题的通解及其精确度

分析G-R-闵公式后得出,它是从点源幅射场导出的,与地震宏观场不相应。因在近场震源不能视作点源。在分析过程中,导出相当于点源、线源或面源以及复杂源的幅射场的深度公式。对实际的地震,它的源类型一无所知,所以不预作假定,将源指标几何扩散率n作为待定参数,导出了一个物理意义明确而又普遍化的震源深度公式。此公式的诸解法中,以计算方法准确、精度高,作图法有直观的优点,但准确性差、精度低,图算法只作获取粗略值和考察数据均匀性用。计算了9个8级以上巨震、1个71/2大震,和两个M_L=3的有感地震的震源深度,经对比结果很好,利用计算得出震源类型,结合宏观场研究了这些地震的震源几何学。并且首次在国际上给出宏观地震震源深度值的标准误差。     

关于辽宁地区b值方法定量预测的研究

本文利用地震学分析预报方法程式指南和“八五”攻关成果中提出的地震b值定量预测方法,对辽宁及邻区1969年以来发生的8次地震震例作了回顾性检验研究。在此基础上,根据震前b值的变化特征,总结出了可能适用于辽宁地区半年左右尺度的b值预测方法的异常羊别指标及预报规则。     

中国半湿润/半干旱类型及区域划分指标的研究

半湿润与半干旱地区之间的差异是客观存在的,但长期以来由于各种区划所采用的原则、方法和指标不同, 对它们之间的划分存在着较多的分歧和争议。本文对我国从综合自然地理学角度进行半湿润与半干旱区划分的指标进行了对比分析, 指出了其中存在的不足,并提出了在指标的选取上应着重考虑的几个方面。试图为中国生态地理区域系统的建立提供有益的参考。