Multi-scale support vector algorithms for hot spot detection and modelling

The algorithmic approach to data modelling has developed rapidly these last years, in particular methods based on data mining and machine learning have been used in a growing number of applications. These methods follow a data-driven methodology, aiming at providing the best possible generalization and predictive abilities instead of concentrating on the properties of the data model. One of the most successful groups of such methods is known as Support Vector algorithms. Following the fruitful developments in applying Support Vector algorithms to spatial data, this paper introduces a new extension of the traditional support vector regression (SVR) algorithm. This extension allows for the simultaneous modelling of environmental data at several spatial scales. The joint influence of environmental processes presenting different patterns at different scales is here learned automatically from data, providing the optimum mixture of short and large-scale models. The method is adaptive to the spatial scale of the data. With this advantage, it can provide efficient means to model local anomalies that may typically arise in situations at an early phase of an environmental emergency. However, the proposed approach still requires some prior knowledge on the possible existence of such short-scale patterns. This is a possible limitation of the method for its implementation in early warning systems. The purpose of this paper is to present the multi-scale SVR model and to illustrate its use with an application to the mapping of Cs¹³⁷ activity given the measurements taken in the region of Briansk following the Chernobyl accident.     

Coastal erosion vs riverine sediment discharge in the Arctic Shelf seas

This article presents a comparison of sediment input by rivers and by coastal erosion into both the Laptev Sea and the Canadian Beaufort Sea (CBS). New data on coastal erosion in the Laptev Sea, which are based on field measurements and remote sensing information, and existing data on coastal erosion in the CBS as well as riverine sediment discharge into both the Laptev Sea and the CBS are included. Strong regional differences in the percentages of coastal erosion and riverine sediment supply are observed. The CBS is dominated by the riverine sediment discharge (64.45᎒⁶ t a^-1) mainly of the Mackenzie River, which is the largest single source of sediments in the Arctic. Riverine sediment discharge into the Laptev Sea amounts to 24.10᎒⁶ t a^-1, more than 70% of which are related to the Lena River. In comparison with the CBS, the Laptev Sea coast on average delivers approximately twice as much sediment mass per kilometer, a result of higher erosion rates due to higher cliffs and seasonal ice melting. In the Laptev Sea sediment input by coastal erosion (58.4᎒⁶ t a^-1) is therefore more important than in the CBS and the ratio between riverine and coastal sediment input amounts to 0.4. Coastal erosion supplying 5.6᎒⁶ t a^-1 is less significant for the sediment budget of the CBS where riverine sediment discharge exceeds coastal sediment input by a factor of ca. 10.     

Magnetosheath Structure in an Anisotropic Plasma Model

Variations of the magnetic field and plasma parameters across the Earth's magnetosheath are calculated for an anisotropic plasma in the Chew-Goldberger-Low approximation. Additionally, proton pitch-angle diffusion is taken into account as the energy transfer mechanism between the direction perpendicular and parallel to the magnetic field. We discuss the main characteristics of the magnetic barrier for different temperature relaxation times and their dependence on the interplanetary magnetic field orientation. This revised version was published online in July 2006 with corrections to the Cover Date.     

Conditions of pocket formation in the Oktyabrskaya tourmaline-rich gem pegmatite (the Malkhan field, Central Transbaikalia, Russia)

Coexisting melt (MI), fluid-melt (FMI) and fluid (FI) inclusions in quartz from the Oktaybrskaya pegmatite, central Transbaikalia, have been studied and the thermodynamic modeling of PVTX-properties of aqueous orthoboric-acid fluids has been carried out to define the conditions of pocket formation. At room temperature, FMI in early pocket quartz and in quartz from the coarse-grained quartz–oligoclase host pegmatite contain crystalline aggregates and an orthoboric-acid fluid. The portion of FMI in inclusion assemblages decreases and the volume of fluid in inclusions increases from the early to the late growth zones in the pocket quartz. No FMI have been found in the late growth zones. Significant variations of solid/fluid ratios in the neighboring FMI result from heterogeneous entrapment of coexisting melts and fluids by a host mineral. Raman spectroscopy, SEM EDS and EMPA indicate that the crystalline aggregates in FMI are dominated by mica minerals of the boron-rich muscovite–nanpingite CsAl₂[AlSi₃O₁₀](OH,F)₂ series as well as lepidolite. Topaz, quartz, potassium feldspar and several unidentified minerals occur in much lower amounts. Fluid isolations in FMI and FI have similar total salinity (4–8 wt.% NaCl eq.) and H₃BO₃ contents (12–16 wt.%). The melt inclusions in host-pegmatite quartz homogenize at 570–600 °C. The silicate crystalline aggregates in large inclusions in pocket quartz completely melt at 615 °C. However, even after those inclusions were significantly overheated at 650±10 °C and 2.5 kbar during 24 h they remained non-homogeneous and displayed two types: (i) glass+unmelted crystals and (ii) fluid+glass. The FMI glasses contain 1.94–2.73 wt.% F, 2.51 wt.% B₂O₃, 3.64–5.20 wt.% Cs₂O, 0.54 wt.% Li₂O, 0.57 wt.% Ta₂O₅, 0.10 wt.% Nb₂O₅, 0.12 wt.% BeO. The H₂O content of the glass could exceed 12 wt.%. Such compositions suggest that the residual melts of the latest magmatic stage were strongly enriched in H₂O, B, F, Cs and contained elevated concentrations of Li, Be, Ta, and Nb. FMI microthermometry showed that those melts could have crystallized at 615–550 °C.

Crystallization of quartz–feldspar pegmatite matrix leads to the formation of H₂O-, B- and F-enriched residual melts and associated fluids (prototypes of pockets). Fluids of different compositions and residual melts of different liquidus–solidus P–T-conditions would form pockets with various internal fluid pressures. During crystallization, those melts release more aqueous fluids resulting in a further increase of the fluid pressure in pockets. A significant overpressure and a possible pressure gradient between the neighboring pockets would induce fracturing of pockets and “fluid explosions”. The fracturing commonly results in the crushing of pocket walls, formation of new fractures connecting adjacent pockets, heterogenization and mixing of pocket fluids. Such newly formed fluids would interact with a primary pegmatite matrix along the fractures and cause autometasomatic alteration, recrystallization, leaching and formation of “primary–secondary” pockets.     

Contribution of the tully groups to the soft X-ray background

We examine a model for the expected contributions to the soft X-ray background (SXRB) due to bremsstrahlung and re-combination radiation of gas, enriched with heavy elements, located in groups of galaxies taken from Tully'sCatalog of Nearby Groups of Galaxies. It is shown that the contribution to the SXRB of groups of galaxies can explain up to 20% of the observed flux depending on the chemical composition and ratio of emitting mass to assumed virial mass.     

Benthic nutrient regeneration in South Texas coastal waters

Because of the variable, unpredictable nature of many potential nutrient sources for coastal phytoplankton primary production needs in the north-western Gulf of Mexico, benthic regeneration was investigated as a more constant source of nutrients to this ecosystem. Water column ammonia profiles taken at several locations on the south Texas inner-shelf in the last seven years showed peak concentrations in bottom waters. Benthic chamber measurements verified the presence of ammonia fluxes at the mud-water interface in south Texas coastal waters. These fluxes were related to benthic faunal activity as measured by sediment metabolism. Laboratory experiments, designed to test the effect of benthic faunal removal on nutrient regeneration, suggested that the fauna play a role in regulating this process. How this regulation may occur is discussed. It was estimated that annual benthic regeneration rates can supply 69% of the nitrogen required to support phytoplankton primary production in these coastal waters.     

New insights into the crustal structure of the North German Basin from reprocessing of seismic reflection data using the Common Reflection Surface stack

The influence of deep crustal processes on basin formation and evolution and its relation to current morphology is not well understood yet. A key feature to unravel these issues is a detailed seismic image of the crust. A part of the data recorded by the hydrocarbon industry in the late 1970s and 1980s in the North German Basin were released to the public recently. The seismic reflection data were recorded down to 15 s two-way travel time. The mean Common Midpoint fold of about 20 is relatively low compared to contemporary seismic acquisitions. The processing of the 1980s focussed on the sedimentary structures to explore the hydrocarbon potential of this area. We applied the Common Reflection Surface stack technique to the data sets, which is well suited for low-fold data. The reprocessing was focussed on the imaging of the subsedimentary crustal range. The reprocessed images show enhanced reflections, especially in the mid and lower crustal part. Also, the image of the salt structures in the graben area was improved. Furthermore, the reprocessed images indicate an almost flat Moho topography in the area of the Glückstadt Graben and an additional lower crustal structure, which can be correlated with a high-density body found in recent gravity modeling studies.     

Mineralogical and geochemical investigation of clay-rich mine tailings from a closed phosphate mine, Bartow Florida, USA

Clay-rich mine tailings from phosphate mine operations in Florida are a major environmental and economic problem. Options for reclamation and restoration for these tailings are very limited and are fundamentally restricted by poor physical properties such as low mechanical strength, low hydraulic conductivity, and heavy metal content. The major control on these bulk physical properties is the mineralogy of the materials. Eight continuous push borings were obtained to investigate stratigraphy, mineralogy, aspects of geochemistry, and bulk properties of a deposit of clay-rich mine tailings from a phosphate mine near Bartow, Florida that ceased operations in the early 1970s. Stratigraphy is dominated by laminated clay-rich sediment with minor units of silt and sand. An intact kaolinite liner occurs near the impoundment walls and the impoundment floor has approximately 4 m of relief. Moisture content varies from 4.35 to 57.40 wt% and organic content varies from 0.41 to 9.53 wt%. Bulk XRF investigation indicates that the P₂O₅ concentrations vary from approximately 4 to 21 wt%. A very strong correlation (r ² = 0.92) between CaO and P₂O₅ indicates that apatite is a major control on the phosphate. The strong correlation (r ² = 0.77) of Al₂O₃ and TiO₂ suggests that the source materials for this deposit are comparatively uniform. A number of heavy metal elements and trace elements occur. Cr, V, Ni, Cu are interpreted to be in phosphate minerals, largely apatite. Sr and Pb are interpreted to be in both phyllosilicates and phosphate minerals. Two populations of apatite were observed in the clay-sized fraction, one that was Fe and Si- bearing and another that was only Si-bearing. Fe-bearing apatite had Fe₂O₃ contents that varied from 0.38 to 5.32 wt% and SiO₂ contents that varied from 0.90 to 3.32 wt%. The other apatite population had a wider range of SiO₂ contents that varied from 0.77 to 8.80 wt%. TEM imaging shows that apatite grains are dominantly single crystals with lesser amounts of aggregates. Wavellite commonly occurs as individual or clusters of lath-like crystals and the chemical composition differs from the pure aluminium phosphate end member with average concentrations of components being that of CaO (1.57 wt%), Fe₂O₃ (1.98 wt%), SiO₂ (5.94 wt%). In the clay-sized phosphate minerals investigated no fluorine was found above detection limit (approximately 0.15 wt%), nor was any uranium, radium, heavy metal, or REE element detected. The phyllosilicate mineralogy of the deposit is dominated by smectite (montmorillonite with lesser amounts of nontronite), palygorskite, illite and kaolinite. No systematic variation in the relative proportions of phyllosilicates was observed in the clay deposit. Energy dispersive spectroscopy EDS analysis indicates that chemical compositions of phyllosilicates are somewhat typical but overall are enriched with respect to Fe compared to theoretical end members. The relative enrichment of Fe is interpreted to be a primary sedimentary feature. Ca content in smectite minerals is high and may inhibit stabilization using lime or similar methods. The high percentages of montmorillonite and palygorskite explain the high bulk water contents observed. This investigation provides fundamentally new details regarding clay tailing deposits from closed phosphate mines in central Florida which can be used in restoration and reclamation efforts.     

Recognition criteria,characteristics and implications of the fluvial to marine transition zone in ancient deltaic deposits (Lajas Formation,Argentina)

The seaward end of modern rivers is characterized by the interactions of marine and fluvial processes, a tract known as the fluvial to marine transition zone, which varies between systems due to the relative strength of these processes. To understand how fluvial and tidal process interactions and the fluvial to marine transition zone are preserved in the rock record, large‐scale outcrops of deltaic deposits of the Middle Jurassic Lajas Formation (Neuquén Basin, Argentina) have been investigated. Fluvial–tidal indicators consist of cyclically distributed carbonaceous drapes in unidirectional, seaward‐oriented cross‐stratifications, which are interpreted as the result of tidal modulation of the fluvial current in the inner part of the fluvial to marine transition zone. Heterolithic deposits with decimetre‐scale interbedding of coarser‐grained and finer‐grained facies with mixed fluvial and tidal affinities are interpreted to indicate fluvial discharge fluctuations (seasonality) and subordinate tidal influence. Many other potential tidal indicators are argued to be the result of fluvial–tidal interactions with overall fluvial dominance or of purely fluvial processes. No purely tidal or tide‐dominated facies were recognized in the studied deposits. Moreover, fluvial–tidal features are found mainly in deposits interpreted as interflood (forming during low river stage) in distal (delta front) or off‐axis (interdistributary) parts of the system. Along major channel axes, the interpreted fluvial to marine transition zone is mainly represented by the fluvial‐dominated section, whereas little or no tide‐dominated section is identified. The system is interpreted to have been hyposynchronous with a poorly developed turbidity maximum. These conditions and the architectural elements described, including major and minor distributary channels, terminal distributary channels, mouth bars and crevasse mouth bars, are consistent with an interpretation of a fluvial‐dominated, tide‐influenced delta system and with an estimated short backwater length and inferred microtidal conditions. The improved identification of process interactions, and their preservation in ancient fluvial to marine transition zones, is fundamental to refining interpretations of ancient deltaic successions.