1950～1990年中国地磁剩余场冠谐分析

根据中国1950、1960、1970、1980年和1990年地磁三分量绝对测量资料，使用球冠谐和分析方法，分别计算1950～1990年各个年代中国地磁剩余场冠谐模型. 球冠极点位于36°N和104°E，球冠半角为30°，冠谐模型的截断阶数为8. 地磁剩余场冠谐模型的均方偏差分别为：对于X分量，1950年为93.1 nT，1960年为128.9 nT，1970年为107.2 nT，1980年为107.6 nT，1990年为95.2 nT；对于Y分量，1950～1990年依次为74.8 nT，98.1 nT，89.2 nT，89.9 nT和84.0 nT；对于Z分量，分别为122.2 nT(1950年)，135.0 nT(1960年)，137.7 nT(1970年)，110.1 nT(1980年)和107.5 nT(1990年). 根据中国地磁剩余场冠谐模型和全球地磁场DGRF模型，得到中国地磁场的冠谐模型，并对冠谐模型的边界效应进行了分析和讨论.     

桩端沉降测量对试桩成果分析的意义

桩端沉降测量和试桩资料的分析结果表明,桩在传递上部荷载与地基土之间产生的应力应变过程,表现为桩的弹性杆件和地基土的弹—塑性变形特征。从而对受荷桩,尤其是长柱和超长桩的应力应变特征有了更切合实际的认识。     

利用相关分析划分重力异常

本文从理论上对相关分析法区分重力异常的可行性进行了论述，并从一元线性回归方程着手，对大地水准面上的重力值与相应的地形高程作了相关分析，通过分析，认为在区域性小比例主均衡补偿基本完善的条件下，重力值与高程有极好的线性相关关系；当计算窗口较大时，地下的局部不均匀体，不足以破坏重力值与高程的相头发一，所以，在实际工作中，利用相关分析处理重力资料的叠加场，可以收到较好的效果，且方法简单易行。     

模拟退火剩余静校正的一步算法

当地震资料的剩余静校正量大、信噪比低时，利用传统的线性反演方法不能有效地拾取静校正量。本文介绍一种非线性反演方法（一步模拟退火法）来估算炮点和检波点的剩余静校正量。通过对合成地震记录的试算表明，该算法可获得近似于真实的剩余静校正量。     

河北省物理地层界线厘定

以现代地层学多重划分为理论依据,以岩石地层单元组的密度、磁化率、剩余磁化强度及厚度加权算术平均值等物性参数为研究对象,以有序地质量最优分割方法为数学手段,进行河北省全区综合地层柱物理地层界线的厘定,并对其地质意义进行了初步讨论。     

采空区上方修建大型建筑物地基稳定性评价

地表移动变形随时间的稳定性及剩余变形问题一直是采动覆岩沉陷研究的重要方面。笔者分析了采动地表移动变形的时间过程,探讨了地表沉陷的延续时间及地表剩余沉陷的预计方法,给出了采空区上方修建大型建筑物地基稳定性评价的指标,对采空区上建设建筑物提出了相应的技术措施。     

On Some Simplifications of Cokriging Neighborhood

Choosing the cokriging neighborhood is often difficult. A poor choice, ignoring influent data, can result in a loss of information as well as in artifacts in simulations based on cokriging. Then it is convenient to use if possible, or to refer to models that lead to simplified cokriging neighborhood. We essentially consider the case of two stationary variables, a target variable and an auxiliary one. By examining possible simplifications, we set up a list of models (essentially models with residuals) that, in general or under specific configurations, lead to simplifications of cokriging neighborhood. Collocated, dislocated, and other types of neighborhood are identified, that are optimal in some models and configurations. Possible extensions to cokriging with unknown means, and to more variables, are included.     

Fingerprinting the 8.2 ka event climate response in a coupled climate model

Using results from coupled climate model simulations of the 8.2 ka climate event that produced a cold period over Greenland in agreement with the reconstructed cooling from ice cores, we investigate the typical pattern of climate anomalies (fingerprint) to provide a framework for the interpretation of global proxy data for the 8.2 ka climate event. For this purpose we developed an analysis method that isolates the forced temperature response and provides information on spatial variations in magnitude, timing and duration that characterise the detectable climate event in proxy archives. Our analysis shows that delays in the temperature response to the freshwater forcing are present, mostly in the order of decades (30 a over central Greenland). The North Atlantic Ocean initially cools in response to the freshwater perturbation, followed in certain parts by a warm response. This delay, occurring more than 200 a after the freshwater pulse, hints at an overshoot in the recovery from the freshwater perturbation. The South Atlantic and the Southern Ocean show a warm response reflecting the bipolar seesaw effect. The duration of the simulated event varies for different areas, and the highest probability of recording the event in proxy archives is in the North Atlantic Ocean area north of 40° N. Our results may facilitate the interpretation of proxy archives recording the 8.2 ka event, as they show that timing and duration cannot be assumed to correspond with the timing and duration of the event as recorded in Greenland ice cores. Copyright © 2011 John Wiley & Sons, Ltd.