Inversion of time-dependent nuclear well-logging data using neural networks

The purpose of this work was to investigate a new and fast inversion methodology for the prediction of subsurface formation properties such as porosity, salinity and oil saturation, using time‐dependent nuclear well logging data. Although the ultimate aim is to apply the technique to real‐field data, an initial investigation as described in this paper, was first required; this has been carried out using simulation results from the time‐dependent radiation transport problem within a borehole. Simulated neutron and γ‐ray fluxes at two sodium iodide (NaI) detectors, one near and one far from a pulsed neutron source emitting at ～14 MeV, were used for the investigation. A total of 67 energy groups from the BUGLE96 cross section library together with 567 property combinations were employed for the original flux response generation, achieved by solving numerically the time‐dependent Boltzmann radiation transport equation in its even parity form. Material property combinations (scenarios) and their correspondent teaching outputs (flux response at detectors) are used to train the Artificial Neural Networks (ANNs) and test data is used to assess the accuracy of the ANNs. The trained networks are then used to produce a surrogate model of the expensive, in terms of computational time and resources, forward model with which a simple inversion method is applied to calculate material properties from the time evolution of flux responses at the two detectors. The inversion technique uses a fast surrogate model comprising 8026 artificial neural networks, which consist of an input layer with three input units (neurons) for porosity, salinity and oil saturation; and two hidden layers and one output neuron representing the scalar photon or neutron flux prediction at the detector. This is the first time this technique has been applied to invert pulsed neutron logging tool information and the results produced are very promising. The next step in the procedure is to apply the methodology to real data.     

A radar transparent layer in a temperate valley glacier: Bench Glacier,Alaska

Radar surveys of Bench Glacier, Alaska, collected over five field seasons between 2002 and 2006 reveal a surface layer of radar transparent ice in this temperate valley glacier. The transparent layer covers the up‐glacier half of the ablation zone and is defined by a distinct lack of the radar scattering events considered typical of temperate ice. Radar scattering ice underlies the transparent zone, and extends to the surface elsewhere on the glacier. We observed the layering in constant offset radar surveys conducted with characteristic frequencies ranging from 5 MHz to 100 MHz. The radar transparent layer extends from the surface to 20 m depth on average, but up to 50 m in some places. Bench Glacier's transparent layer appears similar to the cold surface layer of polythermal glaciers, however, observations in over 50 boreholes on Bench Glacier suggest there is no cold ice corresponding to the radar transparent layer. We conclude that spatially extensive radar‐transparent layers normally used to identify cold ice in polythermal glaciers are present in some temperate glaciers. Copyright © 2009 John Wiley & Sons, Ltd.     

Modelling the impact of a hypothetical sub-Plinian eruption at La Soufrière of Guadeloupe (Lesser Antilles)

This paper describes the development and application of an impact model for a future hypothetical sub-Plinian eruption of La Soufrière of Guadeloupe. The model was designed to assess the impact from either a single or multiple eruption scenarios, each defined in terms of a map of the intensity of three volcanic hazards; volcanogenic earthquake, tephra fallout and pyroclastic density currents. The impact from the three hazards can be assessed independently or alternatively the joint impact of the three hazards can be assessed. The outputs that are produced from the model are; the number of buildings with collapsed roofs, and the number of fatal and non-fatal casualties.     

Rapid rotation,active nests of convection and global-scale flows in solar-likestars

In the solar convection zone, rotation couples with intensely turbulent convection to build global-scale flows of differential rotation and meridional circulation. Our sun must have rotated more rapidly in its past, as is suggested by observations of many rapidly rotating young solar-type stars. Here we explore the effects of more rapid rotation on the patterns of convection in such stars and the global-scale flows which are self-consistently established. The convection in these systems is richly time dependent and in our most rapidly rotating suns a striking pattern of spatially localized convection emerges. Convection near the equator in these systems is dominated by one or two patches of locally enhanced convection, with nearly quiescent streaming flow in between at the highest rotation rates. These active nests of convection maintain a strong differential rotation despite their small size. The structure of differential rotation is similar in all of our more rapidly rotating suns, with fast equators and slower poles. We find that the total shear in differential rotation, as measured by latitudinal angular velocity contrast, ΔΩ, increases with more rapid rotation while the relative shear, ΔΩ/Ω, decreases. In contrast, at more rapid rotation the meridional circulations decrease in both energy and peak velocities and break into multiple cells of circulation in both radius and latitude. (© 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Effects of sediment contaminants and environmental gradients on macrobenthic community trophic structure in Gulf of Mexico estuaries

Macrobenthic communities from estuaries throughout the northern Gulf of Mexico were studied to assess the influence of sediment contaminants and natural environmental factors on macrobenthic community trophic structure. Community trophic data were also used to evaluate whether results from laboratory sediment toxicity tests were effective indicators of site-specific differences in benthic trophic structure. A multiple regression model consisting of five composite factors (principal components) was used to distinguish the effects of sediment contaminants and environmental variables on benthic community trophic structure. This model explained 33.5% of the variation in macrobenthic trophic diversity (p<0.001), a variable derived from the distribution of taxas among nine original trophic categories. A significant negative relatinship was found between principal components reflecting concentrations of sediment contaminants and macrobenthic trophic diversity. Detritivores including surface deposit-feeders (SDF), subsurface deposit-feeders (SSDF), and filter feeders (FF) were numerically dominant at 201 random sites, each group accounting for 25–30% of total macrobenthic abundance. The relative abundance of SDFs was considerably lower (12.1±2.9% to 17.1±4.4%) at sites where sediment contaminant concentrations exceeded minimum biological effects thresholds (ER-L values from Long and Morgan 1990 than at sites sampled at random (29.3±5.7%). SSDFs were proportionally more abundant at contaminated sites (42.0±7.7% to 63.6±10.3%) versus random sites (27.5±5.7%), and the relative abundance of SSDFs was positively correlated with concentrations of particular contaminants. Benthic trophic structure was also found to be a function of salinity, where the proportion of SSDFs was negatively correlated with salinity (p=0.035, r=−0.223, n=326). Silt-clay content loaded fairly strongly on the first principal component, but trophic structure parameters were not significantly correlated with sediment grain size or dissolved oxygen (perhaps due, in part, to covariation). Results from laboratory sediment toxicity tests with mysids were predictive of differences in macrobenthic trophic structure in situ (i.e., mysid survival was negatively correlated with %SSDF; p<0.001, r=−0.292, n=326). Results from laboratory sediment toxicity tests with ampeliscid amphipods were not indicative of site-specific differences in benthic trophic structure.     

The Deep Space 1 Encounter with Comet 19p/Borrelly

NASA's Deep Space 1 (DS1) spacecraft successfully encountered comet 19P/Borrelly near perihelion and the Miniature Integrated Camera and Spectrometer (MICAS) imaging system onboard DS1 returned the first high-resolution images of a Jupiter-family comet nucleus and surrounding environment. The images span solar phase angles from 88° to 52°, providing stereoscopic coverage of the dust coma and nucleus. Numerous surface features are revealed on the 8-km long nucleus in the highest resolution images(47–58 m pixel). A smooth, broad basin containing brighter regions and mesa-likestructures is present in the central part of the nucleus that seems to be the source ofjet-like dust features seen in the coma. High ridges seen along the jagged terminator lead to rugged terrain on both ends of the nucleus containing dark patches and smaller series of parallel grooves. No evidence of impact craters with diameters larger thanabout 200-m are present, indicating a young and active surface. The nucleus is very dark with albedo variations from 0.007 to 0.035. Short-wavelength, infrared spectra from 1.3 to 2.6 μm revealed a hot, dry surface consistent with less than about10% actively sublimating. Two types of dust features are seen: broad fans and highlycollimated “jets” in the sunward hemisphere that can be traced to the surface. The source region of the main jet feature, which resolved into at least three smaller “jets” near the surface, is consistent with an area around the rotation pole that is constantly illuminated by the sun during the encounter. Within a few nuclear radii, entrained dustis rapidly accelerated and fragmented and geometrical effects caused from extended source regions are present, as evidenced in radial intensity profiles centered on the jet features that show an increase in source strength with increasing cometocentric distance. Asymmetries in the dust from dayside to nightside are pronounced and may show evidence of lateral flow transporting dust to structures observed in the nightside coma. A summary of the initial results of the Deep Space 1 Mission is provided, highlighting the new knowledge that has been gained thus far.     

The clustering of colour-selected galaxies

We present measurements of the angular correlation function of galaxies selected from a B _J ∼23.5 multicolour survey of two 5°×5° fields located at high galactic latitudes. The galaxy catalogue of ∼4×10⁵ galaxies is comparable in size to catalogues used to determine the galaxy correlation function at low redshift. Measurements of the z ∼0.4 correlation function at large angular scales show no evidence for a break from a power law, although our results are not inconsistent with a break at ≳15 h⁻¹ Mpc. Despite the large fields-of-view, there are large discrepancies between the measurements of the correlation function in each field, possibly caused by dwarf galaxies within z ∼0.11 clusters near the South Galactic Pole.
Colour selection is used to study the clustering of galaxies from z ∼0 to z ∼0.4. The galaxy correlation function is found to depend strongly on colour, with red galaxies more strongly clustered than blue galaxies by a factor of ≳5 at small scales. The slope of the correlation function is also found to vary with colour, with γ∼1.8 for red galaxies and γ∼1.5 for blue galaxies. The clustering of red galaxies is consistently strong over the entire magnitude range studied, although there are large variations between the two fields. The clustering of blue galaxies is extremely weak over the observed magnitude range, with clustering consistent with r ₀∼2 h⁻¹ Mpc. This is weaker than the clustering of late-type galaxies in the local Universe, and suggests that galaxy clustering is more strongly correlated with colour than morphology. This may also be the first detection of a substantial low-redshift galaxy population with clustering properties similar to faint blue galaxies.     

Twelve years of tracking 52-Hz whale calls from a unique source in the North Pacific

A unique whale call with 50–52 Hz emphasis from a single source has been tracked over 12 years in the central and eastern North Pacific. These calls, referred to as 52-Hz calls, were monitored and analyzed from acoustic data recorded by hydrophones of the US Navy Sound Surveillance System (SOSUS) and other arrays. The calls were noticed first in 1989, and have been detected and tracked since 1992. No other calls with similar characteristics have been identified in the acoustic data from any hydrophone system in the North Pacific basin. Only one series of these 52-Hz calls has been recorded at a time, with no call overlap, suggesting that a single whale produced the calls. The calls were recorded from August to February with most in December and January. The species producing these calls is unknown. The tracks of the 52-Hz whale were different each year, and varied in length from 708 to 11,062 km with travel speeds ranging from 0.7 to 3.8 km/h. Tracks included (A) meandering over short ranges, (B) predominantly west-to-east movement, and (C) mostly north-to-south travel. These tracks consistently appeared to be unrelated to the presence or movement of other whale species (blue, fin and humpback) monitored year-round with the same hydrophones.