An iteratively reweighted least-squares approach to adaptive robust adjustment of parameters in linear regression models with autoregressive and <Emphasis Type="Italic">t</Emphasis>-distributed deviations

In this paper, we investigate a linear regression time series model of possibly outlier-afflicted observations and autocorrelated random deviations. This colored noise is represented by a covariance-stationary autoregressive (AR) process, in which the independent error components follow a scaled (Student’s) t-distribution. This error model allows for the stochastic modeling of multiple outliers and for an adaptive robust maximum likelihood (ML) estimation of the unknown regression and AR coefficients, the scale parameter, and the degree of freedom of the t-distribution. This approach is meant to be an extension of known estimators, which tend to focus only on the regression model, or on the AR error model, or on normally distributed errors. For the purpose of ML estimation, we derive an expectation conditional maximization either algorithm, which leads to an easy-to-implement version of iteratively reweighted least squares. The estimation performance of the algorithm is evaluated via Monte Carlo simulations for a Fourier as well as a spline model in connection with AR colored noise models of different orders and with three different sampling distributions generating the white noise components. We apply the algorithm to a vibration dataset recorded by a high-accuracy, single-axis accelerometer, focusing on the evaluation of the estimated AR colored noise model.     

Spatial variation in fluid flow and geochemical fluxes across the sediment-seawater interface at the Carlos Ribeiro mud volcano (Gulf of Cadiz)

Submarine mud volcanism is an important pathway for transfer of deep-sourced fluids enriched in hydrocarbons and other elements into the ocean. Numerous mud volcanoes (MVs) have been discovered along oceanic plate margins, and integrated elemental fluxes are potentially significant for oceanic chemical budgets. Here, we present the first detailed study of the spatial variation in fluid and chemical fluxes at the Carlos Ribeiro MV in the Gulf of Cadiz. To this end, we combine analyses of the chemical composition of pore fluids with a 1-D transport-reaction model to quantify fluid fluxes, and fluxes of boron, lithium and methane, across the sediment-seawater interface. The pore fluids are significantly depleted in chloride, but enriched in lithium, boron and hydrocarbons, relative to seawater. Pore water profiles of sulphate, hydrogen sulphide and total alkalinity indicate that anaerobic oxidation of methane occurs at 34-180 cm depth below seafloor. Clay mineral dehydration, and in particular the transformation of smectite to illite, produces pore fluids that are depleted in chloride and potassium. Profiles of boron, lithium and potassium are closely related, which suggests that lithium and boron are released from the sediments during this transformation. Pore fluids are expelled into the water column by advection; fluid flow velocities are 4 cm yr⁻¹ at the apex of the MV but they rapidly decrease to 0.4 cm yr⁻¹ at the periphery. The associated fluxes of boron, lithium and methane vary between 7-301, 0.5-6 and 0-806 mmol m⁻² yr⁻¹, respectively. We demonstrate that fluxes of Li and B due to mud volcanism may be important on a global scale, however, release of methane into the overlying water column is suppressed by microbial methanotrophy.     

Probabilistic elasto-plasticity: solution and verification in 1D

In this paper, a solution is presented for evolution of probability density function (PDF) of elastic–plastic stress–strain relationship for material models with uncertain parameters. Developments in this paper are based on already derived general formulation presented in the companion paper. The solution presented is then specialized to a specific Drucker–Prager elastic–plastic material model. Three numerical problems are used to illustrate the developed solution. The stress–strain response (1D) is given as a PDF of stress as a function of strain. The presentation of the stress–strain response through the PDF differs significantly from the traditional presentation of such results, which are represented by a single, unique curve in stress–strain space. In addition to that the numerical solutions are verified against closed form solutions where available (elastic). In cases where the closed form solution does not exist (elastic–plastic), Monte Carlo simulations are used for verification.     

Gassmann Modeling of Acoustic Properties of Sand-clay Mixtures

—The feasibility of modeling elastic properties of a fluid-saturated sand-clay mixture rock is analyzed by assuming that the rock is composed of macroscopic regions of sand and clay. The elastic properties of such a composite rock are computed using two alternative schemes.¶The first scheme, which we call the composite Gassmann (CG) scheme, uses Gassmann equations to compute elastic moduli of the saturated sand and clay from their respective dry moduli. The effective elastic moduli of the fluid-saturated composite rock are then computed by applying one of the mixing laws commonly used to estimate elastic properties of composite materials.¶In the second scheme which we call the Berryman-Milton scheme, the elastic moduli of the dry composite rock matrix are computed from the moduli of dry sand and clay matrices using the same composite mixing law used in the first scheme. Next, the saturated composite rock moduli are computed using the equations of Brown and Korringa, which, together with the expressions for the coefficients derived by Berryman and Milton, provide an extension of Gassmann equations to rocks with a heterogeneous solid matrix.¶For both schemes, the moduli of the dry homogeneous sand and clay matrices are assumed to obey the Krief’s velocity-porosity relationship. As a mixing law we use the self-consistent coherent potential approximation proposed by Berryman.¶The calculated dependence of compressional and shear velocities on porosity and clay content for a given set of parameters using the two schemes depends on the distribution of total porosity between the sand and clay regions. If the distribution of total porosity between sand and clay is relatively uniform, the predictions of the two schemes in the porosity range up to 0.3 are very similar to each other. For higher porosities and medium-to-large clay content the elastic moduli predicted by CG scheme are significantly higher than those predicted by the BM scheme.¶This difference is explained by the fact that the BM model predicts the fully relaxed moduli, wherein the fluid can move freely between sand and clay regions. In contrast, the CG scheme predicts the no-flow or unrelaxed moduli. Our analysis reveals that due to the extremely low permeability of clays, at seismic and higher frequencies the fluid has no time to move between sand and clay regions. Thus, the CG scheme is more appropriate for clay-rich rocks.     

Influence of friction on the motion of quasigeostrophic vortices

Abstract

Who show through numerical experiments that friction may enhance westward motion of vortices on the beta-plane and establish a relation betwen the dissipation coefficients and westward aceleration. This suggests an explanation fot the dunamics of westward intensification.     

Quasi-isotropic approximation of ray theory for anisotropic media

Wave propagation in weakly anisotropic inhomogeneous media is studied by the quasi-isotropic approximation of ray theory. The approach is based on the ray-tracing and dynamic ray-tracing differential equations for an isotropic background medium. In addition, it requires the integration of a system of two complex coupled differential equations along the isotropic ray.
The interference of the qS waves is described by traveltime and polarization corrections of interacting isotropic S waves. For qP waves the approach leads to a correction of the traveltime of the P wave in the isotropic background medium.
Seismograms and particle-motion diagrams obtained from numerical computations are presented for models with different strengths of anisotropy.
The equivalence of the quasi-isotropic approximation and the quasi-shear-wave coupling theory is demonstrated. The quasi-isotropic approximation allows for a consideration of the limit from weak anisotropy to isotropy, especially in the case of qS waves, where the usual ray theory for anisotropic media fails.     

Magnetostratigraphy of the lower Triassic volcanics from deep drill SG6 in western Siberia: evidence for long-lasting Permo–Triassic volcanic activity

The deep drill hole SG6 in western Siberia (66°N, 78.5°E) penetrated 1.1  km of lower Triassic basalts, which are possibly an extension of the central Siberian Permo– Triassic flood basalt province. About 300 samples of these basalts were progressively demagnetized and measured. Principal component analysis often shows multiple magnetizations carried by haematite and magnetite. The corrected mean inclinations are +77° and −77° for the haematite component. A magnetostratigraphic scale was derived and showed a N–R–N–R–N succession. This is quite different from the Noril'sk and Taimyr typical polarity scale, R–N.
  The basalts found in the SG6 deep drill hole are slightly younger than those of central Siberia and Taimyr. They correspond to middle–upper Induan age, whereas the Noril'sk and Taimyr sections correspond to an uppermost Permian and lower Induan age. Altogether they indicate that, after a high output rate of volcanic material near the Permo–Triassic boundary, this activity slowed down drastically on the Siberian platform and Taimyr, but persisted for several million years in western Siberia.     

Eruptive history of the Barombi Mbo Maar, Cameroon Volcanic Line, Central Africa: Constraints from volcanic facies analysis

his study presents the first and detail field investigations of exposed deposits at proximal sections of the Barombi Mbo Maar (BMM), NE Mt Cameroon, with the aim of documenting its past activity, providing insight on the stratigraphic distribution, depositional process, and evolution of the eruptive sequences during its formation. Field evidence reveals that the BMM deposit is about 126m thick, of which about 20m is buried lowermost under the lake level and covered by vegetation. Based on variation in pyroclastic facies within the deposit, it can be divided into three main stratigraphic units: U₁, U₂ and U₃. Interpretation of these features indicates that U₁ consists of alternating lapilli-ash-lapilli beds series, in which fallout derived individual lapilli-rich beds are demarcated by surges deposits made up of thin, fine-grained and consolidated ash-beds that are well-defined, well-sorted and laterally continuous in outcrop scale. U₂, a pyroclastic fall-derived unit, shows crudely lenticular stratified scoriaceous layers, in which many fluidal and spindle bombs-rich lapilli-beds are separated by very thin, coarse-vesiculatedash-beds, overlain by a mantle xenolith- and accidental lithic-rich explosive breccia, and massive lapilli tuff and lapillistone. U3 displays a series of surges and pyroclastic fall layers. Emplacement processes were largely controlled by fallout deposition and turbulent diluted pyroclastic density currents under “dry” and “wet” conditions. The eruptive activity evolved in a series of initial phreatic eruptions, which gradually became phreatomagmatic, followed by a phreato-Strombolian and a violent phreatomagmatic fragmentation. A relatively long-time break, demonstrated by a paleosol between U₂ and U₃, would have permitted the feeding of the root zone or the prominent crater by the water that sustained the next eruptive episode, dominated by subsequent phreatomagmatic eruptions. These preliminary results require complementary studies, such as geochemistry, for a better understanding of the changes in the eruptive styles, and to develop more constraints on the maar’s polygenetic origin.     

Reinterpreting the thermocline feedback in the western-central equatorial Pacific and its relationship with the ENSO modulation

Vertical stratification changes at low frequency over the last decades are the largest in the western-central Pacific and have the potential to modify the balance between ENSO feedback processes. Here we show evidence of an increase in thermocline feedback in the western-central equatorial Pacific over the last 50 years, and in particular after the climate shift of 1976. It is demonstrated that the thermocline feedback becomes more effective due to the increased stratification in the vicinity of the mean thermocline. This leads to an increase in vertical advection variability twice as large as the increase resulting from the stronger ENSO amplitude (positive asymmetry) in the eastern Pacific that connects to the thermocline in the western-central Pacific through the basin-scale ‘tilt’ mode. Although the zonal advective feedback is dominant over the western-central equatorial Pacific, the more effective thermocline feedback allows for counteracting its warming (cooling) effect during warm (cold) events, leading to the reduced covariability between SST and thermocline depth anomalies in the NINO4 (160°E–150°W; 5°S–5°N) region after the 1976 climate shift. This counter-intuitive relationship between thermocline feedback strength as derived from the linear relationship between SST and thermocline fluctuations and stratification changes is also investigated in a long-term general circulation coupled model simulation. It is suggested that an increase in ENSO amplitude may lead to the decoupling between eastern and central equatorial Pacific sea surface temperature anomalies through its effect on stratification and thermocline feedback in the central-western Pacific.     

Pseudotachylytes in the southern border fault of the Cenozoic intracontinental Teletsk basin (Altai, Russia)

The Cenozoic intracontinental Teletsk basin in the Central Asian Altai Mountains is composed of a complexly structured northern and a more simple southern sub-basin. These sub-basins formed in two distinct kinematic stages when first the NNW-striking Teletsk- and then the NE-striking West-Sayan shear zones became reactivated in the Cenozoic under dominant NS-oriented horizontal compression. Although the entire Teletsk basin strikes roughly NS, the southern sub-basin is parallel to the NNW-trending, amphibolite facies Teletsk ductile shear zone, while the northern sub-basin is NS-striking and flanked by differently structured, greenschist facies basement. Basement reactivation closely controlled the southern sub-basin formation, but this is less clear for the northern sub-basin. Contrasts between northern and southern basement and the exclusive occurrence of pseudotachylytes along the margins of the southern basin are explored for their contribution to the formation of the Teletsk basin with two distinct sub-basins.In the ductile shear fabric of the basement flanking the southern sub-basin, concordantly interleaved pseudotachylytes and isolated breccia lenses reflect local brittle deformation along the ductile fabric. The genetic link between breccia lenses and pseudotachylyte occurrences was defined by microstructural investigation. It allows to explore their possible development in a dextral strike–slip zone. These rocks occur in a large fault-bounded segment of the basement. The geometry of the structures in the segment is comparable with a dextral strike–slip sidewall-ripout structure along the Teletsk shear zone. Seismic slip related to pseudotachylytes is attributed to the sudden stress release on the NNW-striking Teletsk shear zone, when the latter became unconstrained by reactivation of the NE-trending West-Sayan fault zone at its northern boundary. The boundary of the sidewall-ripout structure was reactivated as a large listric fault in a later stage. The northern sub-basins roughly develop along an NS strike and are assumed to reflect reactivation of the ductile shear zone underneath the variably structured greenschist facies basement outcropping along the flanks of the sub-basin.