一种基于二次灰关联分析的数列预测法

数列预测一直是评价决策中最常见而又信于把握的问题之一。本文在总结现有预测方法的基础上，结合前铛预报思想和灰色预测理论，依据数据求同原则，提出了运用二次灰关联分析，建立组合模型的预测方法，该方法易于计算机实现，并已实际应用于油气储量预测研究中。     

Estimation of groundwater evaporation and salt flux from Owens Lake, California, USA

Groundwater evaporation and subsequent precipitation of soluble salts at Owens Lake in eastern California have created one of the single largest sources of airborne dust in the USA, yet the evaporation and salt flux have not been fully quantified. In this study, we compare eddy correlation, microlysimeters and solute profiling methods to determine their validity and sensitivity in playa environments. These techniques are often used to estimate evaporative losses, yet have not been critically compared at one field site to judge their relative effectiveness and accuracy. Results suggest that eddy correlation methods are the most widely applicable for the variety of conditions found on large playa lakes. Chloride profiling is shown to be highly sensitive to thermal and density-driven fluxes in the near surface and, as a result, appears to underestimate yearly groundwater evaporation. Yearly mean groundwater evaporation from the playa surface estimated from the three study areas was found to range from 88 to 104 mm year⁻¹, whereas mean evaporation from the brine-covered areas was 872 mm year⁻¹. Uncertainties on these mean rates were estimated to be ±25%, based on comparisons between eddy correlation and lysimeter estimates. On a yearly basis, evaporation accounts for approximately 47 × 10⁶ m³ of water loss from the playa surface and open-water areas of the lake. Over the playa area, as much as 7.5 × 10⁸ kg (7.5 × 10⁵ t) of salt are annually concentrated by evaporation at or near the playa surface, much of which appears to be lost during dust storms in area.     

A bivariate meta-Gaussian density for use in hydrology

Convenient bivariate densities found in the literature are often unsuitable for modeling hydrologic variates. They either constrain the range of association between variates, or fix the form of the marginal distributions. The bivariate meta-Gaussian density is constructed by embedding the normal quantile transform of each variate into the Gaussian law. The density can represent a full range of association between variates and admits arbitrarily specified marginal distributions. Modeling and estimation can be decomposed into i) independent analyses of the marginal distributions, and ii) investigation of the dependence structure. Both statistical and judgmental estimation procedures are possible. Some comparisons to recent applications of bivariate densities in the hydrologic literature motivate and illustrate the model.     

Organism relics or kerogens in oils as oil-source rock correlation indicator

Organism relics or kerogens in oils are first obsenred by a confocal laser scanning microscope (CLSM) and a transmission electron microscope (TEM). The complexes of kerogens in oils are characterized by the presence of a great variety of microfossils or macerals. These kerogens in oils are mainly the residues of the original organic substances from which oil formed, and minor kerogens from enclosing rocks enter the oils, therefore, the components and types of the kerogens in crude oils can serve as an indirect indicator of oil-source rock correlation. This method was applied to Jurassic oils in the Junggar Basin and the Turpan-Hami Basin, and there are two types of the kerogens in oils: one containing a lot of macerals from terrestrial plants may derive from coals, and the other, characterized by a high content of microscopic algae, fungus spores and Acritarch, may originate from Permian organic matter. In addition, the reflectance of the vitrodetrinites in oils can be used as an indicator of oil-source rock correlation.     

Fundamental study on the response characteristics of drivers during an earthquake based on driving simulator experiments

After the 1995 Kobe earthquake, the expressway structures in Japan were retrofitted and they will not now be seriously damaged under a certain level of strong earthquake motion. However, the stability of a moving vehicle has not been investigated yet. It has been reported that drivers feel seismically induced vibrations, especially in the transverse direction of vehicles. Owing to this phenomenon, drivers have some difficulty in controlling the vehicles during strong shaking. For further safety promotion of the expressway networks, it is important to understand the drivers' reactions to seismic motion. The present authors have performed a series of seismic response analyses of a moving vehicle to investigate its response characteristics based on numerical simulation. However, the responses of the driver were not considered in the simulation process. In order to investigate the drivers' reactions during an earthquake, a series of virtual tests were conducted using a driving simulator. This driving simulator has six servomotor‐powered electric actuators that control its motions. Several types of tests were carried out for different examinees to investigate drivers' responses while controlling the simulator under seismic motion. The results of this study showed that a larger response time lag to strong shaking and over turning of the steering wheel may shift the vehicle into the next lane. According to this finding, trafficaccidents could possibly occur under strong ground shaking in the case of heavy traffic. Copyright © 2004 John Wiley & Sons, Ltd.     

高密度电法中的"异常扩展效应"及归位计算方法

对高密度电法勘探的异常扩展效应问题进行了较深人的分析；提出了利用相关分析进行归位计算的方法。通过对球体及板状体的Pole-Pole、Wanner-alpha装置的异常归位计算，表明该方法对复杂异常的解释具有效果。     

鲁西南深埋粘性土物理力学指标多元统计研究

运用R因子分析和R型聚类两种多元统计分析方法,对鲁西南地区深埋粘性土的20组样的8个物理力学指标数据进行了分析。利用R因子分析方法,揭示了各指标对鲁西南地区深埋粘性土状态和力学性质的影响关系。利用R型聚类分析方法,研究了深部粘性土各物理力学指标之间的相关性,为进一步从物理力学指标对深埋粘性土的特殊状态和力学特性的研究打下基础。     

顺义地裂缝成因与顺义-良乡断裂北段第四纪活动性讨论

本文通过钻孔地层对比方法研究了顺义一良乡断裂北段的第四纪活动性，资料显示该断裂在第四纪期间呈现强弱交替的分期活动特征，距今315万年以来有3个较强活动期和3个较弱活动期，前三者分别距今266～315、171～228、73～147万年，后三者距今分别为228．266、147～171、0～73万年。中更新世晚期以来断裂活动不明显。超量开采地下水导致地面不均匀沉降是造成顺义地裂缝现今活动的主要原因。     

Spatial character of the gaseous and particulate state compound correlation of urban atmospheric pollution in winter and summer

The spatial/temporal variation information of atmospheric dynamic-chemical processes at observation site points of the "canopy" boundary of Beijing urban building ensemble and over urban area "surface", as well as the seasonal correlation structure of the gaseous and particulate states of urban atmospheric pollution (UAP) and its seasonal conversion feature at observation points are investigated, using the comprehensive observation data of the Beijing City Air Pollution Observation Experiment (BECAPEX) in winter and summer 2003 with a "point-surface" combined research approach. By using "one dimension spatial empirical orthogonal function (EOF)" principal component analysis (PCA) mode, the seasonal change of gaseous and particulate states of atmospheric aerosols and the association feature of pollutant species under the background of the complicated structure of urban boundary layer (UBL) are analyzed. The comprehensive analyses of the principal components of particle concentrations,gaseous pollutant species, and meteorological conditions reveal the seasonal changes of the complex constituent and structure features of the gaseous and particulate states of UAP to further trace the impact feature of urban aerosol pollution surface sources and the seasonal difference of the component structure of UAP. Research results suggest that in the temporal evolution of the gaseous and particulate states of winter/summer UAP, NOx, CO, and SO2 showed an "in-phase" evolution feature, however, O3 showed an "inverse-phase" relation with other species,all possessing distinctive dependent feature. On the whole, summer concentrations of gaseous pollutants CO, SO2, and NOx were obviously lower than winter ones, especially, the reduction in CO concentration was most distinctive, and ones in SO2 and NOx were next. However, the summer O3 concentration was more than twice winter one. Winter/summer differences in PM10and PM2.5 particle concentrations were relatively not obvious, which indicates that responses of PM10 and PM2.5 particle concentrations to the difference of winter/summer heating period emission sources are far less distinctive than those of NOx, SO2, and CO. The correlation feature of winter/summer gaseous and particulate states depicts that both PM10 and PM2.5 particles were significantly correlated with NOx, and their correlations with NOx are more significant than those with other pollutants. Through PCA, it is found that there was a distinctive difference in the principal component combination structure of winter/summer PM10 and PM2.5 particles: SO2 and NOx dominated in the principal component of winter PM10 and PM2.5 particles; while CO and NOx played the major role in the principal component of summer PM10 and PM2.5 particles. For winter/summer PM10 and PM2.5 particles, there might exist the gaseous and particulate states correlation structures of different "combinations" of such dependent pollutant species. Research results also uncover that the interaction processes of gaseous and particulate states were also related with the vertical structure of UBL, that is to say, the low value layer of UBL O3 concentration was associated with the collocation of atmospheric vertical structures of the low level inversion,inverse humidity, and small wind, which depicts summer boundary layer atmospheric character, i.e.the compound impact of the dependent factor "combination" of wind, temperature, and humidity elements and their collocation structure on the variations of different gaseous pollutant concentrations. Such a depth structure of the extremely low value of O3 concentration in the UBL accords with its "inverse-phase" relation with other gaseous pollutant species. The PCA of meteorological factors associated with PM10 and PM2.5 concentrations also reveals the sensitivity of PM10 and PM2.5 concentration to the combinatory feature of local meteorological conditions.