Neural networks and the separation of cosmic microwave background and astrophysical signals in sky maps

We implement an independent component analysis (ICA) algorithm to separate signals of different origin in sky maps at several frequencies. Owing to its self-organizing capability, it works without prior assumptions on either the frequency dependence or the angular power spectrum of the various signals; rather, it learns directly from the input data how to identify the statistically independent components, on the assumption that all but, at most, one of the components have non-Gaussian distributions.
We have applied the ICA algorithm to simulated patches of the sky at the four frequencies (30, 44, 70 and 100 GHz) used by the Low Frequency Instrument of the European Space Agency's Planck satellite. Simulations include the cosmic microwave background (CMB), the synchrotron and thermal dust emissions, and extragalactic radio sources. The effects of the angular response functions of the detectors and of instrumental noise have been ignored in this first exploratory study. The ICA algorithm reconstructs the spatial distribution of each component with rms errors of about 1 per cent for the CMB, and 10 per cent for the much weaker Galactic components. Radio sources are almost completely recovered down to a flux limit corresponding to ≃0.7 σ _CMB, where σ _CMB is the rms level of the CMB fluctuations. The signal recovered has equal quality on all scales larger than the pixel size. In addition, we show that for the strongest components (CMB and radio sources) the frequency scaling is recovered with per cent precision. Thus, algorithms of the type presented here appear to be very promising tools for component separation. On the other hand, we have been dealing here with a highly idealized situation. Work to include instrumental noise, the effect of different resolving powers at different frequencies and a more complete and realistic characterization of astrophysical foregrounds is in progress.     

The Ninole Basalt — Implications for the structural evolution of Mauna Loa volcano, Hawaii

Lava flows of the Ninole Basalt, the oldest rocks exposed on the south side of the island of Hawaii, provide age and compositional constraints on the evolution of Mauna Loa volcano and the southeastward age progression of Hawaiian volcanism. Although the tholeiitic Ninole Basalt differs from historic lavas of Mauna Loa volcano in most major-element contents (e.g., variably lower K, Na, Si; higher Al, Fe, Ti, Ca), REE and other relatively immobile minor elements are similar to historic and prehistoric Mauna Loa lavas, and the present major-element differences are mainly due to incipient weathering in the tropical environment. New K-Ar whole-rock ages, from relatively fresh roadcut samples, suggest that the age of the Ninole Basalt is approximately 0.1–0.2 Ma, although resolution is poor because of low contents of K and radiogenic Ar. Originally considered the remnants of a separate volcano, the Ninole Hills are here interpreted as faulted remnants of the old south flank of Mauna Loa. Deep canyons in the Ninole Hills, eroded after massive landslide failure of flanks of the southwest rift zone, have been preserved from burial by younger lava due to westward migration of the rift zone. Landslide-induced depressurization of the southwest rift zone may also have induced phreatomagmatic eruptions that could have deposited widespread Basaltic ash that overlies the Ninole Basalt. Subaerial presence of the Ninole Basalt documents that the southern part of Hawaii Island had grown to much of its present size above sea level by 0.1–0.2 Ma, and places significant limits on subsequent enlargement of the south flank of Mauna Loa.     

Computer modeling and simulation of coalbed methane resources

Coal seam gas reservoirs are complex both geologically and in the mechanism of gas production. Understanding these naturally fractured reservoirs for two-phase (gas–water) flow conditions is often limited by a lack of data. This paper illustrates that reservoir simulation is a powerful tool which can be used to determine key data requirements, and how variability in reservoir properties and operating practices affect performance at the field level. The paper presents examples of how reservoir simulation can be used to assess the efficiency of well completions (fracturing or cavitation), identify candidate wells for remedial treatment, examine methane drainage in advance of mining, and assess the impact of errors in measured data on long-term gas production forecasts.     

An attempt to evaluate seismic hazard in Central-Southern Italy

A method for the evaluation of seismic hazard in a given zone, taking into account both the spreading of macroseismic effects and seismic catalogue information, is applied. A data-bank of some 500 digitized isoseisms of earthquakes having occurred in Italy between 1542 and 1986 is used. The isoseismical maps are digitized considering for each degree of intensity the length of 24 spreading rays starting from the macroseismic epicenter or barycentre of the megaseismic area. These rays are separated from each other by the same angle, i.e. every isoseism is divided into 24 equal circular sectors. The year 1542 is taken as the beginning of the time span, since this is when the first seismic event occurred for which reliable isoseismal maps are available. The epicentral intensities considered lie between theVI andXI degrees of the Mercalli-Cancani-Sieberg scale (MCS). This digitized data-bank is analyzed to achieve, for each homogeneous seismogenetic zone that has been recognized, the mean azimuthal spreadings of effects for each degree of intensity as a function of the epicentral intensity. Once a mean propagation model is obtained for each zone, this is applied to seismic events of the same zone, the isoseismal maps of which are not available. A geographic grid is defined to cover the analyzed area, and for each cell of this grid it is then possible to count the number of felt events and their degree. These effects have been evaluated either on the basis of the isoseismal maps (when available) or on the basis of the mean propagations of the zone in which the single event occurred. Moreover, an index summarizing the seismic information was computed for each cell of the previous grid. All the events producing effects and their provenance are stored on files, allowing the main seismogenic zones influencing this cell to be identified. This methodology has been applied to central and southern Italy in an area between the latitudes 40.6 and 43.3 N. In particular, attention is focussed on the sample areas of Rome (given the historical and political importance of the city) and of the Sannio-Matese and Irpinia zone (in which some of the strongest earthquakes of the Apennine chain have occurred). Finally, in order to evaluate the maximum expected magnitude, extreme value statistics (Gumbel III-type) are applied to the Colli Albani area, which represents the seismogenic zone nearest to Rome. For the Sannio-Matese and Irpinia area, considering the more dangerous zone as a ‘unicum’, theWeibull distribution has been hypothesized to determine the mean return time for events with an intensity greater than or equal to IX.     

Ab initio study of MgSiO3 C2/c enstatite

We have used a newly developed ab initio constant-pressure molecular dynamics with variable cell shape technique to investigate the zero temperature behaviour of high pressure clinoenstatite (MgSiO₃-C2/c) from 0 up to 30 GPa. The optimum structure at 8 GPa, as well as structural trends under pressure, compare very well with experimental data. At this pressure, we find noticeable “fluctuations” in the chain configuration which suggests the structure is on the verge of a mechanical instability. Two distinct compressive behaviours then appear: one below and another above 8 GPa. This phenomenon may be related to the observed transition to a lower symmetry P2₁/c phase which involves a reconfiguration of the silicate chains, and suggests that the C2/c structure at low pressures found here, may be an artifact of the dynamical algorithm which preserves space group in the absence of symmetry breaking fluctuations. Comparison with calculations in other magnesium silicate phases, indicates that the size and shape of the silicate units (isolated and/or linked tetrahedra and octahedra) are generally well described by the local density approximation; however, the weaker linkages provided by the O-Mg-O bonds, are not as well described. This trend suggests that, as in the recently studied case of H₂O-ice, the structural properties of more inhomogeneous systems, like enstatite, may be improved by using gradientcorrected density functionals.     

Measurement of manganese, zinc and cadmium complexation in seawater using Chelex ion exchange equilibria

Equilibria between Chelex 100* and manganese, zinc and cadmium ions were used to determine the complexation of these trace metals in 36‰ Gulf Stream seawater at 25°C and pH 8.2. The method utilized radiotracers (⁵⁴Mn, ⁶⁵Zn, and ¹⁰⁹Cd) to quantify trace metal adsorption from trace metal-amended seawater and from seawater containing a series of ethylenediaminetetracetate (EDTA)—metal ion buffers. Results were consistent with Chelex adsorption of both trace metal ions and trace metal—EDTA chelates. Equilibrium models fitted to the data were used to establish conditional stability constants for Chelex adsorption of manganese, zinc and cadmium ions and for adsorption of EDTA-chelates. These models also yielded ratios of free metal ions to total dissolved trace metal concentrations in seawater: 10^−0.1 for manganese, 10^−0.2 for zinc, and 10^−1.5 for cadmium. Independent measurements with a cadmium ion-selective electrode also yielded a free: total cadmium ratio of 10^−1.5.     

Sulfur biogeochemical cycling and novel Fe-S mineralization pathways in a tidally re-flooded wetland

Sulfur biogeochemical cycling and associated Fe-S mineralization processes exert a major influence over acidity dynamics, electron flow and contaminant mobility in wetlands, benthic sediments and groundwater systems. While S biogeochemical cycling has been studied intensively in many environmental settings, relatively little direct information exists on S cycling in formerly drained wetlands that have been remediated via tidal re-flooding. This study focuses on a tidal wetland that was drained in the 1970s (causing severe soil and water acidification), and subsequently remediated by controlled re-flooding in 2002. We examine reduction rates and Fe-S mineralization at the tidal fringe, 7 years after the commencement of re-flooding. The initial drainage of the wetland examined here caused in-situ pyrite (FeS₂) oxidation, resulting in the drained soil layers being highly acidic and rich in -bearing Fe(III) minerals, including jarosite (KFe₃(SO₄)₂(OH)₆). Tidal re-flooding has neutralized much of the previous acidity, with the pore-water pH now mostly spanning pH 5-7. The fastest rates of in-situ reduction (up to ∼300 nmol cm⁻³ day⁻¹) occur within the inter-tidal zone in the near-surface soil layers (to ∼60 cm below ground surface). The reduction rates correlate with pore-water dissolved organic C concentrations, thereby suggesting that electron donor supply was the predominant rate determining factor. Elemental S was a major short-term product of reduction, comprising up to 69% of reduced inorganic S in the near-surface soil layers. This enrichment in elemental S can be partly attributed to interactions between biogenic H₂S and jarosite - a process that also contributed to enrichment in pore-water Fe²⁺ (up to 55 mM) and (up to 50 mM). The iron sulfide thiospinel, greigite (Fe₃S₄), was abundant in near-surface soil layers within the inter- to sub-tidal zone where tidal water level fluctuations created oscillatory redox conditions. There was evidence for relatively rapid pyrite re-formation within the re-flooded soil layers. However, the results indicate that pyrite re-formation has occurred mainly in the lower formerly drained soil layers, whereas the accumulation of elemental S and greigite has been confined towards the soil surface. The discovery that pyrite formation was spatially decoupled from that of elemental S and greigite challenges the concept that greigite is an essential precursor required for sedimentary pyrite formation. In fact, the results suggest that greigite and pyrite may represent distinct end-points of divergent Fe-S mineralization pathways. Overall, this study highlights novel aspects of Fe-S mineralization within tidal wetlands that have been drained and re-flooded, in contrast to normal, undisturbed tidal wetlands. As such, the long-term biogeochemical trajectory of drained and acidified wetlands that are remediated by tidal re-flooding cannot be predicted from the well-studied behaviour of normal tidal wetlands.     

Seismic study of the Mesozoic carbonate basement around Mt. Somma–Vesuvius, Italy

Fifteen seismic reflection lines from AGIP surveys, in and around the Campanian Plain and Mt. Somma–Vesuvius (south Italy) have been interpreted. The attention has been focused to the horizon pertinent to the top of the Mesozoic carbonate sequence and the Quaternary faults dissecting it. As a matter of fact, both are very important elements for understanding the origin of the volcanic activity in the area, that often in the past, has been the topic of debates not supported by reliable data. In the study area, referring to the depth of the carbonate basement, comparison between the result achieved by the seismic prospecting and previous gravity studies has been made. It shows coherence in some areas but large discrepancy within others. Near the town of S. Anastasia, the gravity and seismic depth estimates differ as much as 1000 m or more. Furthermore, the seismic data show that the source of the greatest volcanic eruption in the area (the so-called ‘Campanian Ignimbrite') is probably not located in the Acerra depression, as suggested by other authors. A main NE–SW fault directed toward Vesuvius, considered as playing a primary role on volcanogenetic processes and previously recognised only offshore by marine seismic survey, has been now identified also inland using this new seismic information. The results presented here strengthen the hypothesis that Mt. Vesuvius is located at the crossing point of two regional Quaternary sets of fault heading NW–SE and NE–SW.     

Centrifuge modeling of seismic behavior of a slope in liquefiable soil

Two centrifuge tests were designed to improve the understanding the response of liquefied sandy slopes beyond initial liquefaction. A distinctive dilative behavior of the soil was observed near the slope where static shear stresses are present. The corresponding drops in the piezometric records and simultaneous negative upslope spikes in the acceleration records were measured in the transducer raw data. This dilative response became stronger as the input acceleration increased and tends to limit the downslope accumulation and thus reducing the permanent lateral displacements. Therefore, the maximum permanent displacement was smaller in the model with the larger input motion, because it developed a stronger dilative response. The dilative response was not observed away from the slope, where no static shear stresses are present.     

A Parametric Approach for Dealing with Compositional Rounded Zeros

In this work, a parametric approach for replacing data below the detection limit, also known as rounded zeros, in compositional data sets is proposed. Compositional rounded zeros correspond to small proportions of some whole that cannot be reliably detected by the analytical instruments under given operating conditions. This kind of zeros appear frequently in the data collection process in geosciences. They must be treated in an adequate way before some multivariate analysis can be applied. Our procedure results from a modification of the Expectation-Maximization (EM) algorithm and is based on the additive log-ratio transformation. Its coherence with the nature of compositional data and with basic operations in the simplex sample space is checked. Using real data sets, we find that this approach improves other parametric and non-parametric techniques for compositional rounded zeros.