Structural damage detection using the optimal weights of the approximating artificial neural networks

This work presents a novel neural network‐based approach to detect structural damage. The proposed approach comprises two steps. The first step, system identification, involves using neural system identification networks (NSINs) to identify the undamaged and damaged states of a structural system. The partial derivatives of the outputs with respect to the inputs of the NSIN, which identifies the system in a certain undamaged or damaged state, have a negligible variation with different system errors. This loosely defined unique property enables these partial derivatives to quantitatively indicate system damage from the model parameters. The second step, structural damage detection, involves using the neural damage detection network (NDDN) to detect the location and extent of the structural damage. The input to the NDDN is taken as the aforementioned partial derivatives of NSIN, and the output of the NDDN identifies the damage level for each member in the structure. Moreover, SDOF and MDOF examples are presented to demonstrate the feasibility of using the proposed method for damage detection of linear structures. Copyright © 2001 John Wiley & Sons, Ltd.     

A komatiite component in Apollo 16 highland breccias: implications for the nickel-cobalt systematics and bulk composition of the moon

The lunar crust at the Apollo 16 landing site contains substantial amounts of a “primitive component” in which the ferromagnesian group of elements is concentrated. The composition of this component can be retrieved via an analysis of mixing relationships displayed by lunar breccias. It is found to be a komatiite which is compositionally similar to terrestrial komatiites both in major and minor elements. The komatiite component of the lunar crust is believed to have formed by extensive degrees of melting of the lunar interior at depths greater than were involved in the formation of the lunar magma ocean which was parental to the crust. After formation of the anorthositic crust, it was invaded by extensive flows and intrusions of komatiite magma from these deeper source regions. The komatiites became intimately mixed with the anorthosite by intensive meteoroid impacts about 4.5 b.y. ago, thereby accounting for the observed mixing relationships displayed by the crust. The compositional similarity between lunar and terrestrial komatiites strongly implies a corresponding similarity between the compositions of their source regions in the lunar interior and the Earth's upper mantle. The composition of the lunar interior can be modelled more specifically by combining the komatiite composition with its liquidus olivine composition (as determined experimentally) in proportions chosen so as to produce a cosmochemically acceptable range of Mg/Si ratios for the bulk Moon. Except for higher FeO and lower Na₂O, the range of compositions thereby obtained for the bulk moon is very similar to the composition of the Earth's upper mantle.The effects of meteoritic contamination on the abundances of cobalt and nickel in lunar highland breccias were subtracted on the assumption that the contaminating projectiles were chondritic. The cobalt and nickel residuals thereby obtained were found to correlate strongly with the (Mg + Fe) content of the breccias, demonstrating that the Co and Ni are associated with the ferromagnesian component of the breccias and are genuinely indigenous to the Moon. The lunar highland Co and Ni residuals also display striking Ni/Co versus Ni correlations which follow a similar trend to those displayed by terrestrial basalts, picrites and komatiites. The lunar trends provide further decisive evidence of the indigenous nature of the Co and Ni residuals and suggest the operation of extensive fractionation controlled by olivine-liquid equilibria in producing the primitive component of the lunar breccias. Indigenous nickel abundances at the Apollo 14, 15 and 17 sites are much lower than at the Apollo 16 site, although rocks from all sites follow the same Ni/Co versus Ni trends. It is suggested that the primitive component at the Apollo 14, 15 and 17 sites was generally of basaltic composition, in contrast to the komatiitic nature of the Apollo 16 primitive component.     

Glacial flood pulse effects on benthic fauna in equatorial high‐Andean streams

Equatorial glacier‐fed streams present unique hydraulic patterns when compared to glacier‐fed observed in temperate regions as the main variability in discharge occurs on a daily basis. To assess how benthic fauna respond to these specific hydraulic conditions, we investigated the relationships between flow regime, hydraulic conditions (boundary Reynolds number, Re*), and macroinvertebrate communities (taxon richness and abundance) in a tropical glacier‐fed stream located in the high Ecuadorian Andes (> 4000 m). Both physical and biotic variables were measured under four discharge conditions (base‐flow and glacial flood pulses of various intensities), at 30 random points, in two sites whose hydraulic conditions were representative to those found in other streams of the study catchment. While daily glacial flood pulses significantly increased hydraulic stress in the benthic habitats (appearance of Re* > 2000), low stress areas still persisted even during extreme flood events (Re* < 500). In contrast to previous research in temperate glacier‐fed streams, taxon richness and abundance were not significantly affected by changes in hydraulic conditions induced by daily glacial flood pulses. However, we found that a few rare taxa, in particular rare ones, preferentially occurred in highly stressed hydraulic habitats. Monte‐Carlo simulations of benthic communities under glacial flood reduction scenarios predicted that taxon richness would be significantly reduced by the loss of high hydraulic stress habitats following glacier shrinking. This pioneer study on the relationship between hydraulic conditions and benthic diversity in an equatorial glacial stream evidenced unknown effects of climate change on singular yet endangered aquatic systems. Copyright © 2013 John Wiley & Sons, Ltd.     

Vertical amplitude reduction of Rayleigh waves by a row of piles in a poroelastic half‐space

A row of rigid piles is addressed as the countermeasures for isolating Rayleigh waves in a poroelastic half‐space. The complex characteristic equations for Rayleigh waves are derived via Biot's theory and their existence conditions are given. The piles are modeled as Euler–Bernoulli beams with longitudinal displacements and the diffracted field by each pile is constructed only with Rayleigh waves. Six infinite linear systems of algebraic equations are obtained in terms of the equilibrium of forces and continuity of displacements at the pile–soil interfaces. The systems are subsequently solved in the complex least‐squares sense. The influence of certain pile and soil characteristics such as the permeability of poroelastic soil, spacing between the piles and length of the piles on the isolating performance of a pile barrier is investigated. Computed results show that the permeability of poroelastic soil displays a significant effect on the vertical amplitude reduction of Rayleigh waves. Copyright © 2009 John Wiley & Sons, Ltd.     

Including spatial distribution in a data‐driven rainfall‐runoff model to improve reservoir inflow forecasting in Taiwan

Multi‐step ahead inflow forecasting has a critical role to play in reservoir operation and management in Taiwan during typhoons as statutory legislation requires a minimum of 3‐h warning to be issued before any reservoir releases are made. However, the complex spatial and temporal heterogeneity of typhoon rainfall, coupled with a remote and mountainous physiographic context, makes the development of real‐time rainfall‐runoff models that can accurately predict reservoir inflow several hours ahead of time challenging. Consequently, there is an urgent, operational requirement for models that can enhance reservoir inflow prediction at forecast horizons of more than 3 h. In this paper, we develop a novel semi‐distributed, data‐driven, rainfall‐runoff model for the Shihmen catchment, north Taiwan. A suite of Adaptive Network‐based Fuzzy Inference System solutions is created using various combinations of autoregressive, spatially lumped radar and point‐based rain gauge predictors. Different levels of spatially aggregated radar‐derived rainfall data are used to generate 4, 8 and 12 sub‐catchment input drivers. In general, the semi‐distributed radar rainfall models outperform their less complex counterparts in predictions of reservoir inflow at lead times greater than 3 h. Performance is found to be optimal when spatial aggregation is restricted to four sub‐catchments, with up to 30% improvements in the performance over lumped and point‐based models being evident at 5‐h lead times. The potential benefits of applying semi‐distributed, data‐driven models in reservoir inflow modelling specifically, and hydrological modelling more generally, are thus demonstrated. Copyright © 2012 John Wiley & Sons, Ltd.     

Oxygen isotope speedometry in granulite facies garnet recording fluid/melt–rock interaction (Sør Rondane Mountains,East Antarctica)

In situ analysis of a garnet porphyroblast from a granulite facies gneiss from Sør Rondane Mountains, East Antarctica, reveals discontinuous step‐wise zoning in phosphorus and large δ¹⁸O variations from the phosphorus‐rich core to the phosphorus‐poor rim. The gradually decreasing profile of oxygen isotope from the core (δ¹⁸O = ~15‰) to the rim (δ¹⁸O = ~11‰) suggests that the ¹⁸O/¹⁶O zoning was originally step‐wise, and modified by diffusion after the garnet rim formation at ~800°C and 0.8 GPa. Fitting of the ¹⁸O/¹⁶O data to the diffusion equation constrains a duration of the high‐T event (~800°C) to c. 0.5–40 Ma after the garnet rim formation. The low δ¹⁸O value of the garnet rim, together with the previously reported low δ¹⁸O values in metacarbonates, indicates regional infiltration, probably along a detachment fault, of low δ¹⁸O fluid/melt possibly derived from meta‐mafic to ultramafic rocks.     

Impact of Two "El Niño" Events of Different Intensity on the Hypoxic Soft Bottom Macrobenthos off the Central Peruvian wast

Abstract. Monthly changes in the community structure of hypoxic soft-bottom macrobenthos have been studied at a station at 34 m depth in Ancón Bay (Peru) before and during two El Niño (EN) events. Of these events, 1982-83 is considered the strongest, and 1991–93 one of the most prolonged in the 20th century. On the oceanographic scale, EN 1982–83 ranges as "very strong", whereas EN 1991–93 ranges as "moderate".
The thermal anomalies at the station during EN 1982–83 (+ 7.8 °C) were almost twice those of EN 1991–93 (+ 4.1 °C). However, the community changes were not in all cases proportional to the extent of warming. Species numbers increased in a similar way (up to 24 species from near zero in each of the two events), but maximum faunal density was five times higher, and mean biomass was twice as high in 1982–83 compared with the event a decade later. Species diversity was slightly higher during EN 1982–83, whereas successional and trophic changes occurred on a broader scale during EN 1991–93. On the whole, the impact of the long-lasting event on the small soft-bottom macrofauna was not much weaker than that of the exceptionally strong event.
The authors discuss the mechanisms which may be responsible for the differences and similarities encountered in the benthic community dynamics during these two events. Both in the unusually strong and the unusually long EN, the community revealed a pattern of early biological response several months ahead of the onset of local warming which marks the official begin of EN. The question is addressed to what extent the increase of certain parameters in the benthic community could be used for predictive purposes.     

Ozone production due to emissions from vegetation burning

Ozone has been observed in elevated concentrations by satellites over areas previously believed to be background. There is meteorological evidence, that these ozone plumes found over the Atlantic Ocean originate from vegetation fires on the African continent.In a previous study (DECAFE-88), we have investigated ozone and assumed precursor compounds over African tropical forest regions. Our measurements revealed large photosmog layers at altitudes from 1.5 to 4 km. Both chemical and meteorological evidence point to savanna fires up to several thousand km upwind as sources.Here we describe ozone mixing ratios observed over western Africa and compare ozone production ratios from different field measurement campaigns related to vegetation burning. We find that air masses containing photosmog ingredients require several days to develop their oxidation potential, similar to what is known from air polluted by emissions from fossil fuel burning. Finally, we estimate the global ozone production due to vegetation fires and conclude that this source is comparable in strength to the stratospheric input.     

Towards more systematic perceptual model development: a case study using 3 Luxembourgish catchments

The synthesis of experimental understanding of catchment behaviour and its translation into qualitative perceptual models is an important objective of hydrological sciences. We explore this challenge by examining the cumulative understanding of the hydrology of three experimental catchments and how it evolves through the application of different investigation techniques. The case study considers the Huewelerbach, Weierbach and Wollefsbach headwater catchments of the Attert basin in Luxembourg. Subsurface investigations including bore holes and pits, analysis of soil samples and Electrical Resistivity Tomography measurements are presented and discussed. Streamflow and tracer data are used to gain further insights into the streamflow dynamics of the catchments, using end‐member mixing analysis and hydrograph separation based on dissolved silica and electrical conductivity. We show that the streamflow generating processes in all three catchments are controlled primarily by the subsolum and underlying bedrock. In the Huewelerbach, the permeable sandstone formation supports a stable groundwater component with little seasonality, which reaches the stream through a series of sources at the contact zone with the impermeable marls formation. In the Weierbach, the schist formation is relatively impermeable and supports a ‘fill and spill’‐type of flow mechanism; during wet conditions, it produces a delayed response dominated by pre‐event water. In the Wollefsbach, the impermeable marls formation is responsible for a saturation‐excess runoff generating process, producing a fast and highly seasonal response dominated by event water. The distinct streamflow generating processes of the three catchments are represented qualitatively using perceptual models. The perceptual models are in turn translated into quantitative conceptual models, which simulate the hydrological processes using networks of connected reservoirs and transfer functions. More generally, the paper illustrates the evolution of perceptual models based on experimental fieldwork data, the translation of perceptual models into conceptual models and the value of different types of data for processes understanding and model representation. Copyright © 2014 John Wiley & Sons, Ltd.