Slow earthquakes and great earthquakes along the Nankai trough

We have reexamined reports indicating that slow deformation occurred before the great Japan earthquakes of 1944 (Tonankai) and 1946 (Nankaido) and find that the observations are well founded. Although no quantitative models have previously been proposed to explain all of the relevant data we show that they are satisfied by a simple model for both earthquakes. The model, based on known properties of subduction zones, has slow slip on the subduction interface in an area deeper than the seismic rupture. If this model is correct and a similar physical situation holds for an anticipated Tokai earthquake, existing instruments will be able to reveal the pre-slip in real time. While differences among the deformation time series at different sites will provide strong constraints on the slow rupture propagation, these differences could result in delaying the recognition of a coherent event.     

Functional series TARMA modelling and simulation of earthquake ground motion

The non‐stationary Functional Series time‐dependent autoregressive moving average (TARMA) modelling and simulation of earthquake ground motion is considered. Full Functional Series TARMA models, capable of modelling both resonances and antiresonances, are examined for the first time via a novel mixed parametric/non‐parametric estimation scheme, and critical comparisons with pure TAR and recursive ARMA (RARMA)‐recursive maximum likelihood (RML) adaptive filtering type modelling are made. The study is based upon two California ground motion signals: a 1979 El Centro accelerogram and a 1994 Pacoima Dam accelerogram. A systematic analysis, employing various functional subspaces and model orders, leads to two Haar function based models: a TARMA(2,4)₈ model for the El Centro case and a TARMA(6,2)₁₀ model for the Pacoima Dam case. Both models are formally validated and their simulation (synthesis) capabilities are demonstrated via Monte Carlo experiments focusing on important time domain signal characteristics. The Functional Series TAR/TARMA models are shown to achieve parsimony, as well as superior accuracy and simulation capabilities, over their RARMA counterparts. Copyright © 2001 John Wiley & Sons Ltd.     

基于GIS的地下水模拟可视化系统开发的初步探讨

系统开发的目的是克服GIS时空分析能力的不足，实现地下水模型与MAPGIS的耦合集成，在集成的基础上构建系统的可视化功能，在GIS平台上为地下水模拟的全过程提供一个统一高效的计算机环境。系统开发采用面向对象的软件开发方法，对各种地下水模型进行分解、抽象，建立基本的C＋＋类类型，每个模型都是基本类及其派生类对象的聚集类。系统实现的技术途径是以MAPGIS的二次开发函数作为“后台”支持，以VC＋＋作为编程工具，在充分利用MAPGIS的可视化功能的基础上，借助于OpenGL并应用科学可视化技术开发更能满足地下水模拟需要的可视化功能。     

岩溶塌陷的地质概化模型

对于岩溶塌陷机制的分析，应以其所处的地质条件为基础，在一定的水文地质条件的基础上，同样的土-岩体，可能具有完全不同的失稳机制。以工程实例为基础，提出了几种岩溶塌陷的盖层地质概化模型，并对每一种概化模型的致塌机制进行了分析和讨论，同时，还提出了水压差场与气压差场的概念。     

Short-Term Ozone Forecasting with a Network Model System during Summer 1999

Within the framework of the Tropospheric Research Programme (TFS) four working groups in Germany (University of Stuttgart, University of Cologne, Research Center Karlsruhe, German Weather Service) formed a network developing a model system for chemistry transport calculations on different scales. The network was optimized, based on each module and combined for the different interfaces in order to perform daily preoperational ozone forecasts and its precursors at the German Weather Service in summer 1999 (May through September). The setup of the model system, the interfaces and changes of each module are described. The results of the whole preoperational episode are displayed and discussed. Main efforts were done analyzing the model climate statistics and the verification of the predictions with an extensive data base of observations at stations in Germany. The results of this extensive verification demonstrate the relatively good performance of the entire forecast system.     

1690年以来全球非偶极子磁场的变化特征

根据J Bloxham和A Jackson地磁场模型，计算和分析了1690-1990年期间非偶极子磁异常的变化特片。结构表明，1690年以来一直存在的变化磁异常有7个，在一定时期内出现的磁异常有3个，它们是1830年以后出现的北太平洋负磁异常、1850年以后出现的东南太平洋负磁异常和1690-1900年期间在南太平洋地区出现的负磁异常。各个磁异常的强度和中心位置随时间变化，变化趋势各不相同。与根据IGRF模型计算的20世纪的磁异常变化比较，除非洲负磁异常外，两种模型得出的其它磁异常在同一时段内的变化趋势基本一致。     

Intelligent technology-based control of motion and vibration using MR dampers

Due to their intrinsically nonlinear characteristics, development of control strategies that are implementable and can fully utilize the capabilities of semiactive control devices is an important and challenging task. In this study, two control strategies are proposed for protecting buildings against dynamic hazards, such as severe earthquakes and strong winds, using one of the most promising semiactive control devices, the magnetorheological (MR) damper. The first control strategy is implemented by introducing an inverse neural network (NN) model of the MR damper. These NN models provide direct estimation of the voltage that is required to produce a target control force calculated from some optimal control algorithms. The major objective of this research is to provide an effective means for implementation of the MR damper with existing control algorithms. The second control strategy involves the design of a fuzzy controller and an adaptation law. The control objective is to minimize the difference between some desirable responses and the response of the combined system by adaptively adjusting the MR damper. The use of the adaptation law eliminates the need to acquire characteristics of the combined system in advance. Because the control strategy based on the combination of the fuzzy controller and the adaptation law doesn’t require a prior knowledge of the combined building-damper system, this approach provides a robust control strategy that can be used to protect nonlinear or uncertain structures subjected to random loads. Supported by: Hong Kong Research Grant Council Competitive Earmarked Research Grant HKUST 6218 / 99E and by the National Science Foundation under grant CMS 99-00234.     

THE MODELING IMPROVEMENT OF HURRICANE ANDREW BY THE TECHNIQUE OF IRREVERSIBLE THERMODYNAMIC OPERATORS

Although incorporating the horizontal diffusion into the atmospheric governing equations is apart of the dynamics of a model,it is based on the computational consideration originally,which isto overcome the nonlinear aliasing or computational instability and smooth the physical fields at thesame time.The generally used forms of the fourth-order diffusion,either linear or nonlinear,however,can not properly describe the real physical diffusion processes and thus affect theaccuracy of forecasting and modeling.In this paper,based on the nature of the physical diffusion.the scheme of horizontal diffusion of the MM5 Version 3 is improved according to the second law ofthermodynamics which controls the irreversable evolutionary direction of a many-body system andthe potential ability of the improvement of the new scheme in modeling accuracy is revealed byintroducing the technique suggested in this paper,via a theoretical case and the computational caseof the hurricane Andrew that landed on Florida in 1992.