Adaptive mesh fluid simulations on GPU

We describe an implementation of compressible inviscid fluid solvers with block-structured adaptive mesh refinement on Graphics Processing Units using NVIDIA’s CUDA. We show that a class of high resolution shock capturing schemes can be mapped naturally on this architecture. Using the method of lines approach with the second order total variation diminishing Runge–Kutta time integration scheme, piecewise linear reconstruction, and a Harten–Lax–van Leer Riemann solver, we achieve an overall speedup of approximately 10 times faster execution on one graphics card as compared to a single core on the host computer. We attain this speedup in uniform grid runs as well as in problems with deep AMR hierarchies. Our framework can readily be applied to more general systems of conservation laws and extended to higher order shock capturing schemes. This is shown directly by an implementation of a magneto-hydrodynamic solver and comparing its performance to the pure hydrodynamic case. Finally, we also combined our CUDA parallel scheme with MPI to make the code run on GPU clusters. Close to ideal speedup is observed on up to four GPUs.     

Orientation Distribution Within a Single Hematite Crystal

While crystallography conventionally presumes that a single crystal carries a unique crystallographic orientation, modern experimental techniques reveal that a single crystal may exhibit an orientation distribution. However, this distribution is largely concentrated; it is extremely concentrated when compared with orientation distributions of polycrystalline specimen. A case study of a deformation experiment with a single hematite crystal is presented, where the experimental deformation induced twining, which in turn changed a largely concentrated unimodal “parent” orientation distribution into a multimodal orientation distribution with a major mode resembling the parent mode and three minor modes corresponding to the progressive twining. The free and open source software MTEX for texture analysis was used to compute and visualize orientations density functions from both integral orientation measurements, i.e. neutron diffraction pole intensity data, and individual orientation measurements, i.e. electron back scatter diffraction data. Thus it is exemplified that MTEX is capable of analysing orientation data from largely concentrated orientation distributions.     

Influence of ice rheology and dust content on the dynamics of the north-polar cap of Mars

The evolution and dynamics of the north-polar cap (residual-ice-cap/layered-deposits complex) of Mars is simulated with a thermomechanical ice-sheet model. We consider a scenario with ice-free initial conditions at 5 Ma before present due to the large obliquities which prevailed prior to this time. The north-polar cap is then built up to its present shape, driven by a parameterized climate forcing (surface temperature, surface mass balance) based on the obliquity and eccentricity history. The effects of different ice rheologies and different dust contents are investigated. It is found that the build-up scenarios require an accumulation rate of approximately 0.15-0.2 mm a⁻¹ at present. The topography evolution is essentially independent of the ice dynamics due to the slow ice flow. Owing to the uncertainties associated with the ice rheology and the dust content, flow velocities can only be predicted within a range of two orders of magnitude. Likely present values are of the order of 0.1-1 mm a⁻¹, and a strong variation over the climatic cycles is found. For all cases, computed basal temperatures are far below pressure melting.     

Line-of-sight velocity distributions of low-luminosity elliptical galaxies

The shape of the line-of-sight velocity distribution (LOSVD) is measured for a sample of 14 elliptical galaxies, predominantly low-luminosity ellipticals. The sample is dominated by galaxies in the Virgo cluster but also contains ellipticals in nearby groups and low-density environments. The parametrization of the LOSVD given by Gerhard and van der Marel & Franx is adopted, which measures the asymmetrical and symmetrical deviations of the LOSVD from a Gaussian by the amplitudes h ₃ and h ₄ of the Gauss–Hermite series. Rotation, velocity dispersion, h ₃ and h ₄ are determined as a function of radius for both major and minor axes. Non-Gaussian LOSVDs are found for all galaxies along the major axes. Deviations from a Gaussian LOSVD along the minor axis are of much lower amplitude if present at all. Central decreases in velocity dispersion are found for three galaxies. Two galaxies have kinematically decoupled cores: NGC 4458 and the well-known case of NGC 3608.     

Kinematics and stellar populations of 17 dwarf early-type galaxies

We present kinematics and stellar population properties of 17 dwarf early-type galaxies in the luminosity range -14 ≥ M _B ≥ -19. Our sample fills the gap between the intensively studied giant elliptical and Local Group dwarf spheroidal galaxies. The dwarf ellipticals of the present sample have constant velocity dispersion profiles within their effective radii and do not show significant rotation, hence are clearly anisotropic. The dwarf lenticulars, instead, rotate faster and are, at least partially, supported by rotation. From optical Lick absorption indices, we derive metallicities and element abundances. Combining our sample with literature data of the Local Group dwarf spheroidals and giant ellipticals, we find a surprisingly tight linear correlation between metallicity and luminosity over a wide range: -8 ≥ M _B ≥ -22. The α/Fe ratios of our dwarf ellipticals are significantly lower than the ones of giant elliptical galaxies, which is in agreement with spectroscopy of individual stars in Local Group dwarf spheroidals. Our results suggest the existence of a clear kinematic and stellar population dichotomy between dwarf and giant elliptical galaxies. This result is important for theories of galaxy formation, because it implies that present-day dwarf ellipticals are not the fossiled building blocks of giant ellipticals. This revised version was published online in August 2006 with corrections to the Cover Date.     

Estimating a Lagrangian Length Scale Using Measurements of CO2 in a Plant Canopy

Analytical Lagrangian equations capable of predicting concentration profiles from known source distributions offer the opportunity to calculate source/sink distributions through inverted forms of these equations. Inverse analytical Lagrangian equations provide a practical means of estimating source profiles using concentration and turbulence measurements. Uncertainty concerning estimates of the essentially immeasurable Lagrangian length scale ( ${\mathcal{L}}$ ), a key input, impedes the operational practicality of this method. The present study evaluates ${\mathcal{L}}$ within a corn canopy by using field measurements to constrain an analytical Lagrangian equation. Measurements of net CO₂ flux, soil-to-atmosphere CO₂ flux, and in-canopy profiles of CO₂ concentration provided the information required to solve for ${\mathcal{L}}$ in a global optimization algorithm for 30-min time intervals. For days when the canopy was a strong CO₂ sink, the optimization frequently located ${\mathcal{L}}$ profiles that follow a convex shape. A constrained optimization then fit the profile shape to a smooth sigmoidal equation. Inputting the optimized ${\mathcal{L}}$ profiles in the forward and inverse Lagrangian equations leads to strong correlations between measured and calculated concentrations and fluxes. Coefficients of the sigmoidal equation were specific to each 30-min period and did not scale with any measured variable. Plausible looking ${\mathcal{L}}$ profiles were associated with negative bulk Richardson number values. Once the canopy senesced, a simple eddy diffusivity profile sufficed to relate concentrations and sources in the analytical Lagrangian equations.     

Geochemical studies of impact breccias and country rocks from the El'gygytgyn impact structure,Russia

The complex impact structure El'gygytgyn (age 3.6 Ma, diameter 18 km) in northeastern Russia was formed in ~88 Ma old volcanic target rocks of the Ochotsk‐Chukotsky Volcanic Belt (OCVB). In 2009, El'gygytgyn was the target of a drilling project of the International Continental Scientific Drilling Program (ICDP), and in summer 2011 it was investigated further by a Russian–German expedition. Drill core material and surface samples, including volcanic target rocks and impactites, have been investigated by various geochemical techniques in order to improve the record of trace element characteristics for these lithologies and to attempt to detect and constrain a possible meteoritic component. The bedrock units of the ICDP drill core reflect the felsic volcanics that are predominant in the crater vicinity. The overlying suevites comprise a mixture of all currently known target lithologies, dominated by felsic rocks but lacking a discernable meteoritic component based on platinum group element abundances. The reworked suevite, directly overlain by lake sediments, is not only comparatively enriched in shocked minerals and impact glass spherules, but also contains the highest concentrations of Os, Ir, Ru, and Rh compared to other El'gygytgyn impactites. This is—to a lesser extent—the result of admixture of a mafic component, but more likely the signature of a chondritic meteoritic component. However, the highly siderophile element contribution from target material akin to the mafic blocks of the ICDP drill core to the impactites remains poorly constrained.     

Collisional processes at the junction of the Aleutian–Kamchatka arcs: new evidence from fission track analysis and field observations

The Aleutian island arc collides with the Kuril–Kamchatka arc in the area of the Cape Kamchatka peninsula. Field studies of neotectonic structures and apatite fission track analysis provide evidence for crustal plate shortening onshore the Cape Kamchatka peninsula. Tectonic blocks show differential mean exhumation rates varying from 0.18 ± 0.04 mm yr⁻¹ in the north up to 1.2 ± 0.18 mm yr⁻¹ in the south of the peninsula. A few of the fission track length data point to an unsteady exhumation rate. The blocks are separated by major dextral fault zones splaying off from Aleutian island arc fault zones. Across the western segment of the North American–Pacific Plate boundary the strain is partitioned along the fault zones and increases from north to south. Results from this study suggest that indentation and accretion of island arc fragments has recently occurred in the southeastern part of the Cape Kamchatka peninsula.