Identification of physical processes inherent in artificial neural network rainfall runoff models

The emergence of artificial neural network (ANN) technology has provided many promising results in the field of hydrology and water resources simulation. However, one of the major criticisms of ANN hydrologic models is that they do not consider/explain the underlying physical processes in a watershed, resulting in them being labelled as black‐box models. This paper discusses a research study conducted in order to examine whether or not the physical processes in a watershed are inherent in a trained ANN rainfall‐runoff model. The investigation is based on analysing definite statistical measures of strength of relationship between the disintegrated hidden neuron responses of an ANN model and its input variables, as well as various deterministic components of a conceptual rainfall‐runoff model. The approach is illustrated by presenting a case study for the Kentucky River watershed. The results suggest that the distributed structure of the ANN is able to capture certain physical behaviour of the rainfall‐runoff process. The results demonstrate that the hidden neurons in the ANN rainfall‐runoff model approximate various components of the hydrologic system, such as infiltration, base flow, and delayed and quick surface flow, etc., and represent the rising limb and different portions of the falling limb of a flow hydrograph. Copyright © 2004 John Wiley & Sons, Ltd.     

Automatic Determination of Stellar Atmospheric Parameters Using Neural Networks and Instance-Based Machine Learning

In this article we show how machine learning methods can beeffectively applied to the problem of automatically predictingstellar atmospheric parameters from spectral information, a veryimportant problem in stellar astronomy. We apply feedforwardneural networks, Kohonen's self-organizing maps andlocally-weighted regression to predict the stellar atmosphericparameters effective temperature, surface gravity and metallicityfrom spectral indices. Our experimental results show that thethree methods are capable of predicting the parameters with verygood accuracy. Locally weighted regression gives slightly betterresults than the other methods using the original dataset asinput, while self-organizing maps outperform the other methods when significant amounts of noise are added. We also implemented a heterogeneous ensemble of predictors, combining the results given by the three algorithms. This ensemble yields better results than any of the three algorithms alone, using both the original and the noisy data.     

Reliability assessment of serviceability performance of braced retaining walls using a neural network approach

In urban environments, one major concern with deep excavations in soft clay is the potentially large ground deformations in and around the excavation. Excessive movements can damage adjacent buildings and utilities. There are many uncertainties associated with the calculation of the ultimate or serviceability performance of a braced excavation system. These include the variabilities of the loadings, geotechnical soil properties, and engineering and geometrical properties of the wall. A risk‐based approach to serviceability performance failure is necessary to incorporate systematically the uncertainties associated with the various design parameters. This paper demonstrates the use of an integrated neural network–reliability method to assess the risk of serviceability failure through the calculation of the reliability index. By first performing a series of parametric studies using the finite element method and then approximating the non‐linear limit state surface (the boundary separating the safe and ‘failure’ domains) through a neural network model, the reliability index can be determined with the aid of a spreadsheet. Two illustrative examples are presented to show how the serviceability performance for braced excavation problems can be assessed using the reliability index. Copyright © 2005 John Wiley & Sons, Ltd.     

密度和压缩系数的散射层析成像法

本文在速度成像的基础上研究了同时对密度和压缩系数成像的散射波层析成像法.对不同散射角度的计算可以得到一系列反演图像,拟合这些图像,从而可以有效地达到对密度和压缩系数（或速度）成像的目的.与单纯的速度成像相比,增加了反演的难度.首先是对资料的方位性要求增加;其次是对资料的利用率下降.即便如此,从对较少量的炮点和检波点资料的数值计算来看,仍取得了满意的成像结果.我们对组成字母“A”的散射体结构进行了成像计算,结果能够同时再现密度和压缩系数,成像清晰,表明了方法的可行性,并能应用于复杂结构的成像问题.     

Predicting ground-motion variations at the Turkey-flat test site,California

Results of a single group participating in an international experiment are analyzed. The experiment served to verify computational predictions of the ground-motion variations due to near-surface geological effects at a site established for that purpose by the California Department of Conservation. Based on an acceleration record at a rock location, and geotechnical model of medium, records at the other locations of a nearby sedimentary deposit were predicted. A 2-D finite-difference sensitivity analysis suggested that the lateral wave-propagation effects are negligibly small, and locally 1-D computations are sufficient for the present site. Those computations are compared with observations not available to the authors during the blind prediction. Peak accelerations, peak velocities and RMS accelerations were predicted with errors less than 159%, 114% and 62%, respectively. Maxima of the response spectra were fitted within a factor of 2. The predicted and observed Husid's plots (i.e., the normalized cumulative plots of the acceleration squared) have the correlation coefficients 0.98. The detected misfits do not show any simple relation to the instrument location, component, frequency, or time.     

宁夏香山-天景山弧形断裂带新活动特征及1709年中卫南71/2级地震形变带

本文根据野外地质填图和水系位移测量结果,论述了香山-天景山弧形断裂带新生代有两个不同活动性质的阶段,即早期阶段的强烈挤压和晚期阶段的左旋走滑兼挤压。分析、讨论了不同活动阶段的时间界限和转变原因。指出了1709年中卫南71/_2级地震形变带的表现形式、延伸范围     

Crustal structure at the easternmost termination of the Variscan belt based on CELEBRATION 2000 and ALP 2002 data

The eastern margin of the Variscan belt in Europe comprises plate boundaries between continental blocks and terranes formed during different tectonic events. The crustal structure of that complicated area was studied using the data of the international refraction experiments CELEBRATION 2000 and ALP 2002. The seismic data were acquired along SW–NE oriented refraction and wide-angle reflection profiles CEL10 and ALP04 starting in the Eastern Alps, passing through the Moravo-Silesian zone of the Bohemian Massif and the Fore-Sudetic Monocline, and terminating in the TESZ in Poland. The data were interpreted by seismic tomographic inversion and by 2-D trial-and-error forward modelling of the P waves. Velocity models determine different types of the crust–mantle transition, reflecting variable crustal thickness and delimiting contacts of tectonic units in depth. In the Alpine area, few km thick LVZ with the Vp of 5.1 km s^− 1 dipping to the SW and outcropping at the surface represents the Molasse and Helvetic Flysch sediments overthrust by the Northern Calcareous Alps with higher velocities. In the Bohemian Massif, lower velocities in the range of 5.0–5.6 km s^− 1 down to a depth of 5 km might represent the SE termination of the Elbe Fault Zone. The Fore-Sudetic Monocline and the TESZ are covered by sediments with the velocities in the range of 3.6–5.5 km s^− 1 to the maximum depth of 15 km beneath the Mid-Polish Trough. The Moho in the Eastern Alps is dipping to the SW reaching the depth of 43–45 km. The lower crust at the eastern margin of the Bohemian Massif is characterized by elevated velocities and high Vp gradient, which seems to be a characteristic feature of the Moravo-Silesian. Slightly different properties in the Moravian and Silesian units might be attributed to varying distances of the profile from the Moldanubian Thrust front as well as a different type of contact of the Brunia with the Moldanubian and its northern root sector. The Moho beneath the Fore-Sudetic Monocline is the most pronounced and is interpreted as the first-order discontinuity at a depth of 30 km.     

A seismic source zone model for the seismic hazard assessment of the Italian territory

We designed a new seismic source model for Italy to be used as an input for country-wide probabilistic seismic hazard assessment (PSHA) in the frame of the compilation of a new national reference map.

We started off by reviewing existing models available for Italy and for other European countries, then discussed the main open issues in the current practice of seismogenic zoning.

The new model, termed ZS9, is largely based on data collected in the past 10 years, including historical earthquakes and instrumental seismicity, active faults and their seismogenic potential, and seismotectonic evidence from recent earthquakes. This information allowed us to propose new interpretations for poorly understood areas where the new data are in conflict with assumptions made in designing the previous and widely used model ZS4.

ZS9 is made out of 36 zones where earthquakes with M_w > = 5 are expected. It also assumes that earthquakes with M_w up to 5 may occur anywhere outside the seismogenic zones, although the associated probability is rather low. Special care was taken to ensure that each zone sampled a large enough number of earthquakes so that we could compute reliable earthquake production rates.

Although it was drawn following criteria that are standard practice in PSHA, ZS9 is also innovative in that every zone is characterised also by its mean seismogenic depth (the depth of the crustal volume that will presumably release future earthquakes) and predominant focal mechanism (their most likely rupture mechanism). These properties were determined using instrumental data, and only in a limited number of cases we resorted to geologic constraints and expert judgment to cope with lack of data or conflicting indications. These attributes allow ZS9 to be used with more accurate regionalized depth-dependent attenuation relations, and are ultimately expected to increase significantly the reliability of seismic hazard estimates.     

Development and application of a GIS-based tool for earthquake-induced hazard prediction

An integrated GIS-based tool (GTIS) was constructed to estimate site effects related to the earthquake hazards in the Gyeongju area of Korea. To build the GTIS for the study area, intensive site investigations and geotechnical data collections were performed and a walk-over site survey was additionally carried out to acquire surface geo-knowledge data in accordance with the procedure developed to build the GTIS. For practical applications of the GTIS used to estimate the site effects associated with the amplification of ground motion, seismic microzoning maps of the characteristic site period and the mean shear wave velocity to a depth of 30 m were created and presented as a regional synthetic strategy addressing earthquake-induced hazards. Additionally, based on one-dimensional site response analyses, various seismic microzoning maps for short- and mid-period amplification potentials were created for the study area. Case studies of seismic microzonations in the Gyeongju area verified the usefulness of the GTIS for predicting seismic hazards in the region.     

Water and power networks and urban fragmentation in Los Angeles: Rethinking assumed mechanisms

Los Angeles is often described as the epitome of urban fragmentation, a notion which in this context is frequently connected to, or even conflated with urban sprawl. At the same time, the city features integrated water and power networks which have been under public ownership for over 70 years. We thus have an apparent paradox in the context of the debate on ‘splintering urbanism’, between socio-spatial fragmentation and the integration of networks. In discussing the idea that deregulation of infrastructural networks exacerbates urban fragmentation, the authors use the case of Los Angeles in order to highlight the central role of private interests in management decisions concerning infrastructure networks. The authors carry out their analysis in an historical perspective, revealing that network integration and universal access can often serve private interests more than the public good. Urban fragmentation in Los Angeles, they conclude, is the result of a complex process of instrumentalisation of network development and management.