Bayesian model selection for ARX models and its application to structural health monitoring

A Bayesian framework for model order selection of auto‐regressive exogenous (ARX) models is developed and applied to actual earthquake response data obtained by the structural health monitoring system of a high‐rise building. The model orders of ARX models are selected appropriately by the Bayesian framework, and differ significantly from the optimal order estimated by AIC; in fact, in many cases AIC does not even give an optimal order. A method is also proposed for consistently selecting the same ‘genuine’ modes of interest from the whole set of modes corresponding to each of the identified models from a sequence of earthquake records. In the identification analysis based on building response records from 43 earthquakes over 9 years, the modal parameters of the first four modes in each horizontal direction are estimated appropriately in all cases, showing that the developed methods are effective and robust. As the estimates of natural frequency depend significantly on the response amplitude, they are compensated by an empirical correction so that the influence of the response amplitude is removed. The compensated natural frequencies are much more stable over the nine‐year period studied, indicating that the building had no significant change in its global dynamic characteristics during this period. Copyright © 2010 John Wiley & Sons, Ltd.     

A comparative simulation study of coupled THM processes and their effect on fractured rock permeability around nuclear waste repositories

This paper presents an international, multiple-code, simulation study of coupled thermal, hydrological, and mechanical (THM) processes and their effect on permeability and fluid flow in fractured rock around heated underground nuclear waste emplacement drifts. Simulations were conducted considering two types of repository settings (1) open emplacement drifts in relatively shallow unsaturated volcanic rock, and (2) backfilled emplacement drifts in deeper saturated crystalline rock. The results showed that for the two assumed repository settings, the dominant mechanism of changes in rock permeability was thermal–mechanically induced closure (reduced aperture) of vertical fractures, caused by thermal stress resulting from repository-wide heating of the rock mass. The magnitude of thermal–mechanically induced changes in permeability was more substantial in the case of an emplacement drift located in a relatively shallow, low-stress environment where the rock is more compliant, allowing more substantial fracture closure during thermal stressing. However, in both of the assumed repository settings in this study, the thermal–mechanically induced changes in permeability caused relatively small changes in the flow field, with most changes occurring in the vicinity of the emplacement drifts.     

The large volume multi-anvil press as a highP-T deformation apparatus

The rheological properties of mantle materials are being investigated up to pressures of 16 GPa and temperatures of 1600°C for times up to 24 h, using a new sample assembly for the 6–8 multi-anvil apparatus. Al₂O₃ pistons, together with a liquid confining medium, are used to generate deviatoric stress in the specimen. Strain rates are estimated by monitoring the relative displacement of the guide blocks of the multi-anvil apparatus, scaled to the total axial strain of the sample. The applied stress on the sample is estimated using grain size piezometry. Strain rates and flow stresses of approximately 10^–4 to 10^–6 s^–1 and 50 to 250 MPa respectively, are presently attainable.Preliminary results on San Carlos olivine single crystals, partially dynamically recrystallized to a grain size of 10 to 300 m, indicate that the effective viscosity of polycrystalline olivine is consistent with values obtained from olivine single crystal creep laws. Assuming a dislocation creep mechanism (n3.5) with (010)[001] as the dominant slip system, the data are best fit using a creep activation volume of 5 to 10×10^–6 m³ mol^–1.     

Tourmaline breakdown in a pelitic system: implications for boron cycling through subduction zones

Pressure–temperature conditions of tourmaline breakdown in a metapelite were determined by high-pressure experiments at 700–900°C and 4–6 GPa. These experiments produced an eclogite–facies assemblage of garnet, clinopyroxene, phengite, coesite, kyanite and rare rutile. The modal proportions of tourmaline clearly decreased between 4.5 and 5 GPa at 700°C, between 4 and 4.5 GPa at 800°C, and between 800 and 850°C at 4 GPa, with tourmaline that survived the higher temperature conditions appearing corroded and thus metastable. Decreases in the modal abundance of tourmaline are accompanied by decreasing modal abundance of coesite, and increasing that of clinopyroxene, garnet and kyanite; the boron content of phengite increases significantly. These changes suggest that, with increasing pressure and temperature, tourmaline reacts with coesite to produce clinopyroxene, garnet, kyanite, and boron-bearing phengite and fluid. Our results suggest that: (1) tourmaline breakdown occurs at lower pressures and temperatures in SiO₂-saturated systems than in SiO₂-undersaturated systems. (2) In even cold subduction zones, subducting sediments should release boron-rich fluids by tourmaline breakdown before reaching depths of 150 km, and (3) even after tourmaline breakdown, a significant amount of boron partitioned into phengite could be stored in deeply subducted sediments.     

Electrical conductivity of dry and hydrous NaAlSi<Subscript>3</Subscript>O<Subscript>8</Subscript> glasses and liquids at high pressures

We report the first study of electrical conductivities of silicate melts at very high pressures (up to 10 GPa) and temperatures (up to 2,173 K). Impedance spectroscopy was applied to dry and hydrous albite (NaAlSi₃O₈) glasses and liquids (with 0.02–5.7 wt% H₂O) at 473–1,773 K and 0.9–1.8 GPa in a piston-cylinder apparatus, using a coaxial cylindrical setup. Measurements were also taken at 473–2,173 K and 6–10 GPa in two multianvil presses, using simple plate geometry. The electrical conductivity of albite melts is found to increase with temperature and water content but to decrease with pressure. However, at 6 GPa, conductivity increases rapidly with temperature above 1,773 K, so that at temperatures beyond 2,200 K, conductivity may actually increase with pressure. Moreover, the effect of water in enhancing conductivity appears to be more pronounced at 6 GPa than at 1.8 GPa. These observations suggest that smaller fractions of partial melt than previously assumed may be sufficient to explain anomalously high conductivities, such as in the asthenosphere. For dry melt at 1.8 GPa, the activation energy at T > 1,073 K is higher than that at T < 1,073 K, and the inflection point coincides with the rheological glass transition. Upon heating at 6–10 GPa, dry albite glass often shows a conductivity depression starting from ~1,173 K (due to crystallization), followed by rapid conductivity enhancement when temperature approaches the albite liquidus. For hydrous melts at 0.9–1.8 GPa, the activation energies for conductivity at ≥1,373 K are lower than those at <973 K, with a complex transition pattern in between. Electrical conductivity and previously reported Na diffusivity in albite melt are consistent with the Nernst–Einstein relation, suggesting the dominance of Na transport for electrical conduction in albite melts.     

Systematic study of hydrogen incorporation into Fe-free wadsleyite

The H₂O content of wadsleyite were measured in a wide pressure (13–20 GPa) and temperature range (1,200–1,900°C) using FTIR method. We confirmed significant decrease of the H₂O content of wadsleyite with increasing temperature and reported first systematic data for temperature interval of 1,400–1,900°C. Wadsleyite contains 0.37–0.55 wt% H₂O at 1,600°C, which may be close to its water storage capacity along average mantle geotherm in the transition zone. Accordingly, water storage capacity of the average mantle in the transition zone may be estimated as 0.2–0.3 wt% H₂O. The H₂O contents of wadsleyite at 1,800–1,900°C are 0.22–0.39 wt%, indicating that it can store significant amount of water even under the hot mantle environments. Temperature dependence of the H₂O content of wadsleyite can be described by exponential equation C_{\textH2 \textO} = 6 3 7.0 7 \texte ^{- 0.00 4 8T} , C_{{{\text{H}}_{2} {\text{O}}}} = 6 3 7.0 7 {\text{e}}^{ - 0.00 4 8T} , where T is in °C. This equation is valid for temperature range 1,200–2,100°C with the coefficient of determination R ² = 0.954. Temperature dependence of H₂O partition coefficient between wadsleyite and forsterite (D _wd/fo) is complex. According to our data apparent D_wd/fo decreases with increasing temperature from D _wd/fo = 4–5 at 1,200°C, reaches a minimum of D _wd/fo = 2.0 at 1,400–1,500°C, and then again increases to D _wd/fo = 4–6 at 1,700–1,900°C.     

Importance of instantaneous radiative forcing for rapid tropospheric adjustment

To better understand CFMIP/CMIP inter-model differences in rapid low cloud responses to CO₂ increases and their associated effective radiative forcings, we examined the tropospheric adjustment of the lower tropospheric stability (LTS) in three general circulation models (GCMs): HadGEM2-A, MIROC3.2 medres, and MIROC5. MIROC3.2 medres showed a reduction in LTS over the sub-tropical ocean, in contrast to the other two models. This reduction was consistent with a temperature decrease in the mid-troposphere. The temperature decrease was mainly driven by instantaneous radiative forcing (RF) caused by an increase in CO₂. Reductions in radiative and latent heating, due to clouds, and in adiabatic and advective heating, also contribute to the temperature decrease. The instantaneous RF in the mid-troposphere in MIROC3.2 medres is inconsistent with the results of line-by-line (LBL) calculations, and thus it is considered questionable. These results illustrate the importance of evaluating the vertical profile of instantaneous RF with LBL calculations; improved future model performance in this regard should help to increase our confidence in the tropospheric adjustment in GCMs.     

Stellar contents and star formation in the young star cluster Be 59

We present   UBV  I _c   CCD photometry of the young open cluster Be 59 with the aim to study the star formation scenario in the cluster. The radial extent of the cluster is found to be ∼10 arcmin (2.9 pc). The interstellar extinction in the cluster region varies between   E ( B − V ) ≃ 1.4  to 1.8 mag. The ratio of total-to-selective extinction in the cluster region is estimated as  3.7 ± 0.3  . The distance of the cluster is found to be  1.00 ± 0.05 kpc  . Using near-infrared (NIR) colours and slitless spectroscopy, we have identified young stellar objects (YSOs) in the open cluster Be 59 region. The ages of these YSOs range between <1 and ∼2 Myr, whereas the mean age of the massive stars in the cluster region is found to be ∼2 Myr. There is evidence for second-generation star formation outside the boundary of the cluster, which may be triggered by massive stars in the cluster. The slope of the initial mass function, Γ, in the mass range  2.5 < M /M_⊙≤ 28  is found to be  −1.01 ± 0.11  which is shallower than the Salpeter value (−1.35), whereas in the mass range  1.5 < M /M_⊙≤ 2.5  the slope is almost flat. The slope of the K -band luminosity function is estimated as  0.27 ± 0.02  , which is smaller than the average value (∼0.4) reported for young embedded clusters. Approximately 32 per cent of Hα emission stars of Be 59 exhibit NIR excess indicating that inner discs of the T Tauri star (TTS) population have not dissipated. The Midcourse Space Experiment (MSX) and IRAS-HIRES images around the cluster region are also used to study the emission from unidentified infrared bands and to estimate the spatial distribution of optical depth of warm and cold interstellar dust.     

Triggered star formation and evolution of T-Tauri stars in and around bright-rimmed clouds

The aim of this paper is to quantitatively testify the ' small-scale sequential star formation ' hypothesis in and around bright-rimmed clouds (BRCs). As a continuation of the recent attempt by Ogura et al., we have carried out   BVI_c   photometry of four more BRC aggregates along with deeper re-observations of two previously observed BRCs. Again, quantitative age gradients are found in almost all the BRCs studied in the present work. Archival Spitzer /Infrared Array Camera data also support this result. The global distribution of near-infrared excess stars in each H  ii region studied here clearly shows evidence that a series of radiation-driven implosion processes proceeded in the past from near the central O star(s) towards the peripheries of the H  ii region. We found that in general weak-line T-Tauri stars (WTTSs) are somewhat older than classical T-Tauri stars (CTTSs). Also the fraction of CTTSs among the T-Tauri stars (TTSs) associated with the BRCs is found to decrease with age. These facts are in accordance with the recent conclusion by Bertout, Siess & Cabrit that CTTSs evolve into WTTSs. It seems that in general the equivalent width of Hα emission in TTSs associated with the BRCs decreases with age. The mass function (MF) of the aggregates associated with the BRCs of the morphological type 'A' seems to follow that found in young open clusters, whereas 'B/C'-type BRCs show significantly steeper MF.     

Fast and slow timescales in the tropical low-cloud response to increasing CO2 in two climate models

To obtain physical insights into the response and feedback of low clouds (C _l ) to global warming, ensemble 4?×?CO₂ experiments were carried out with two climate models, the Model for Interdisciplinary Research on Climate (MIROC) versions 3.2 and 5. For quadrupling CO₂, tropical-mean C _l decreases, and hence, acts as positive feedback in MIROC3, whereas it increases and serves as negative feedback in MIROC5. Three time scales of tropical-mean C _l change were identified—an initial adjustment without change in the global-mean surface air temperature, a slow response emerging after 10–20?years, and a fast response in between. The two models share common features for the former two changes in which C _l decreases. The slow response reflects the variability of C _l associated with the El Ni?o-Southern Oscillation in the control integration, and may therefore be constrained by observations. However, the fast response is opposite in the two models and dominates the total response of C _l . Its sign is determined by a subtle residual of the C _l increase and decrease over the ascending and subsidence regions, respectively. The regional C _l increase is consistent with a more frequent occurrence of a stable condition, and vice versa, as measured by lower-tropospheric stability (LTS). The above frequency change in LTS is similarly found in six other climate models despite a large difference in both the mean and the changes in the low-cloud fraction for a given LTS. This suggests that the response of the thermodynamic constraint for C _l to increasing CO₂ concentrations is a robust part of the climate change.