Classification problems of samples of finite mixtures of compositions

A discriminant technique based on mixture models is presented to be applied when observations are a sample of a mixture of compositions with each component following an additive logistic normal distribution on the d-dimensional simplex. The efficiency of this discriminant technique is compared empirically with the efficiency of the standard discriminant technique based on logcontrast. Simulated compositional data and a real dataset are used to carry out these comparisons.     

Monitoring Changes of Ice Streams Using Time Series of Satellite-Altimetry-Based Digital Terrain Models

The Antarctic Ice Sheet plays a major role in the global system, and the large ice streams discharging into the circumpolar sea represent its gateways to the world's oceans. Satellite radar altimeter data provide an opportunity for mapping surface elevation at kilometer-resolution with meter-accuracy. Geostatistical methods have been developed for the analysis of these data. Applications to Seasat data and data from the Geosat Exact Repeat Mission indicate that the grounding line of Lambert Glacier/Amery Ice Shelf, the largest ice stream in East Antarctica, has advanced 10–12 km between 1978 and 1987–89. The objectives of this paper are to explore possibilities and limitations of satellite-altimetry-based mapping to capture changes for shorter time windows and for smaller areas, and to investigate some methodological aspects of the data analysis. We establish that one season of radar altimeter data is sufficient for constructing a map. This allows study of interannual variation and is the key for a time-series analysis approach to study changes in ice streams. Maps of the lower Lambert Glacier and the entire Amery Ice Shelf are presented for austral winters 1978, 1987, 1988, and 1989. As a first step toward understanding the dynamics of the ice-stream/ice-shelf system, elevation changes are calculated for grounded ice, the grounding zone, and floating ice. In the absence of (sufficient) surface gravity control for the Lambert Glacier/Amery Ice Shelf area, altimetry-based maps may facilitate improvement of geoid models as they provide constraints on the terrain correction in the inverse gravimetric problem.     

Post-collisional extension of the East Greenland Caledonides: a geophysical perspective

Crustal extension during and following continental collision is well documented in the Arctic Caledonian fold belt. However, models for the post-collisional extension of the Caledonides are mainly based on geoscientific data from Scandinavia. For a more complete understanding of the evolution of the Caledonides, knowledge of the crustal structure of East Greenland is vital. Seismic and gravity studies have revealed a pronounced Moho topography and a west-dipping lower crustal reflector beneath the fjord region of East Greenland. These deep crustal structures are related to Late Caledonian extensional structures at the surface. The observations can be satisfactorily explained by applying simple shear or eduction models proposed for upper crustal extension in Scandinavia to the complementing lower crustal structures in East Greenland. However, exhumation of the Caledonian Northeast Greenland eclogite province cannot be accomplished by these models. Instead, a synthesis of geoscientific data has shown marked differences in the crustal structure of East Greenland north and south of about 76°N, indicating a different crustal evolution of the northern and southern parts of the East Greenland Caledonides.     

Fractionation of the noble metals by physical processes

During partial melting in the earth’s mantle, the noble metals become fractionated. Os, Ir, Ru, and Rh tend to remain in the mantle residue whereas Pt, Pd, and Re behave mildly incompatible and are sequestered to the silicate melt. There is consensus that sulfide plays a role in the fractionation process; the major noble metal repository in the mantle is sulfide, and most primitive mantle melts are sulfide-saturated when they leave their mantle sources. However, with sulfide–silicate partitioning, the fractionation cannot be modeled properly. All sulfide–silicate partition coefficients are so extremely high that a silicate melt segregating from a mantle source with residual sulfide should be largely platinum-group elements free. We offer a physical alternative to sulfide–silicate chemical partitioning and provide a mechanism of generating a noble metal-rich melt from a sulfide-saturated source: Because sulfide is at least partially molten at asthenospheric temperature, it will behave physically incompatible during melt segregation, and a silicate melt segregating from a mantle residue will entrain molten residual sulfide in suspension and incorporate it in the basaltic pool melt. The noble metal abundances of a basalt then become independent of sulfide–silicate chemical partitioning. They reflect the noble metal abundances in the drained sulfide fraction as well as the total amount of sulfide entrained. Contrary to convention, we suggest that a fertile, sulfide-rich mantle source has more potential to generate a noble metal-enriched basaltic melt than a refractory mantle source depleted by previous partial melting events.     

A heat flow and physical properties package for the surface of Mercury

European Space Agencies fifth cornerstone mission BepiColombo includes a ‘Surface Element’ to land a scientific payload on the surface of Mercury. The current strawman payload includes a heat flow and physical properties package (HP³), focussing on key thermal and mechanical properties of the near-surface material (down to a depth of 2–5 m) and the measurement of heat flow from Mercury's interior, an important constraining parameter for models of the planet's interior and evolution. We present here an overview of the HP³ experiment package and its possible accommodation in a self-inserting ‘mole’ device. A mole is considered to be the most appropriate deployment method for HP³, at least in the currently-assumed case of an airbag-assisted soft landing architecture for the Mercury Surface Element.