Statistical forecast of the marine surge

Natural Hazards - This paper studies different machine learning methods for solving the regression problem of estimating the marine surge value given meteorological data. The marine surge is...     

Drained triaxial testing of shales: insight from the Opalinus Clay

Acta Geotechnica - The investigation of the mechanical behaviour of swelling claystones and shales is challenging because of their very low permeability and of their high sensitivity to changes in...     

Geological and archaeological chronology of a late Holocene coastal enclosure: The Sanguinet lake (SW France)

Sanguinet lake is separated from the Atlantic Ocean by a wide Holocene coastal dunes system in SW France. The present day lake level is 21 m above mean sea level (msl). It formed when aeolian sand closed the mouth of the small La Gourgue river which gradually became a lagoon and then a lake. Dated sub‐lacustrine archaeological remains (human settlements, canoes, and wooden architectural structures), as well as paleoenvironmental evidence (drowned tree stumps and lagoonal deposits exposed on the beach) are used to interpret the formation and chronology of lake level rise during the past 4000 years. Around 2000–1650 B.C., the river flowed into a lagoon or an estuary which connected with the ocean west of the present Sanguinet Lake. Its level was affected by the tide, which ranged between 2 m below and 3 m above msl. The accumulation of aeolian sand before 1500–1000 B.C. began to close the connection with the sea. At this time, the elevation of the surface of the lake water was approximately 5 m above msl, but it still remained connected to the ocean. Around 1000 B.C., the lake level rose quickly by 1 to 2 m during a period of renewed mobility of the coastal aeolian sand, and continued to rise slowly until about 100 A.D. when there was a gradual closure of the lake outlet. This rise forced people who were living on the lake shore and along the rivers to move to higher land along the valley. The nearby Gallo‐Roman site of Losa was settled at the end of the 1st century B.C.; then the final blocking of the outlet occurred because of spit growth as a result of north‐south littoral drift accompanied by the deposit of aeolian sand. This led to the lake level rising rapidly. Consequently, Losa was abandoned in the 3rd century A.D. and ruins of its temple (at 17 m above msl) were submerged in the 6th century. Further oscillations of the lake level probably correspond to water table fluctuations before it became stable at around 1000 A.D. The highest lake level (23.35 m) was reached during the 18th century as a consequence of modern dune formation, and thus was artificially reduced to 21 m in 1840 by construction of an overflow channel. © 2008 Wiley Periodicals, Inc.     

Enhanced Oxygen Isotope Determination in Uranium Oxides Using BrF5 Fluorination

A new method for accurate determination of oxygen isotopes in uranium oxides encountered in the nuclear fuel cycle was developed using the conventional BrF₅ fluorination technique. Laser‐assisted fluorination was tested for comparison. We focused on fine powders of triuranium octoxide (U₃O₈), uranium dioxide (UO_2±x with 0 ≤ x ≤ 0.25), uranium trioxide (UO₃.nH₂O, with 0.8 ≤ n ≤ 2) and diuranates (M₂U₂O₇.nH₂O, with M = NH₄, Na or Mg_0.5 and 0 ≤ n ≤ 6). Fluorination at room temperature and heating under vacuum at 150 °C are shown to eliminate both adsorbed and structural water from the powder samples. Precision fit for purpose of δ¹⁸O values (± 0.3‰, 1s) and oxygen yields (close to 100%) were obtained for U₃O₈ and UO₂ where oxygen is only bound to uranium. A lower precision was observed for UO₃.nH₂O and M₂U₂O₇.nH₂O where oxygen is both present in the structural H₂O and bonded to uranium and where the extracted O_2(g) can be contaminated by NF₃ and NO_x compounds. Laser‐assisted fluorination gave shifted δ¹⁸O values between +0.8 and +1.4‰ for U₃O₈, around ?0.8‰ for UO₃.nH₂O and between ?3.9 and ?4.5‰ for M₂U₂O₇.nH₂O (± 0.3‰, 1s) compared with the conventional method.     

Les laves du mont Bangou : une première manifestation volcanique éocène, à affinité transitionnelle,de la Ligne du Cameroun

Mount Bangou, an Eocene volcano (⁴⁰K–⁴⁰Ar ages between 44.7 and 43.1 ± 1 Ma) is the oldest dated volcano of the Cameroon Line. In this region, two magmatic series, evolving by fractional crystallization, show transitional affinities that are exceptionally known in this sector. Mineral compositions of basaltic rocks (scarce modal olivine and occurrence of normative hypersthene) as well as geochemical characteristics (low Ba, La, Ta contents and high Y/Nb ratios) are in agreement with this trend. The succession of magmas evolving in time from transitional to more typical alkaline compositions is evidenced in a continental setting. To cite this article: J. Fosso et al., C. R. Geoscience 337 (2005).     

Mineralogy and geochemical behaviour during weathering of greenstone belt under tropical dry conditions in the extreme North Cameroon (Central Africa)

Mineralogy, major, trace and rare earth elements of a weathering profile developed on tertiary greenstone belt in the extreme North Cameroon are reported. The aim of which was to investigate mineralogical evolution and element mobilization and redistribution during weathering under dry tropical climate. The weathering profile consists of four main horizons: (1) a spheroidal weathering zone constituted by a corestone–shell complex, (2) a C horizon, (3) a Bw horizon and an Ah horizon. The results indicate that nontronite, a Fe-rich smectite, is the exclusive clay mineral formed in the exfoliated shells and the C horizon. It is associated with kaolinite in the upper horizons. The coexistence of these two clay minerals induced a decrease of CEC and pH which becomes neutral. The weathering index (WI) values reveal that weathering becomes more and more intensive from the corestone up to Bw horizon, which is the most weathered horizon in the weathering profile. Mass balance calculations, using Th as immobile element, indicate that Ti is quite mobile and that Al and Fe are relatively enriched at the bottom and strongly leached at the top of the profile. Alkalis and alkaline earth elements are strong leached through out the profile, except Ca which displays similar trend as Al and Fe. The same goes for LILE (Cs, Sr), TTE (Cr, Co, Ni) and HSFE (Y, Nb, Hf). In opposite, REE are depleted at the bottom and enriched in the upper horizons, with more enrichment for LREE than for HREE. It appears that weathering of greenstone belt causes a fractionation of HREE and induces a concentration of LREEs. Ce and Eu anomalies display opposite behaviour.     

Luna 20: mineralogy and petrology of fragments less than 125 μm size

Two milligrams of less than 125 μm size particles from the Luna 20 soil sample (22001,17) have been examined. The results of the mineral identification and analysis verify earlier assumptions that the nature of the lunar highlands is predominantly anorthositic in composition. The presence of highly magnesian clinopyroxene, orthopyroxene and olivine suggests crystallization of Highland rocks in an ultrabasic environment. No fragments were observed that could be unequivocally assigned to mare basalt types.     

Cumulate gabbros from the Southwest Indian Ridge, 54°S-7° 16′ E: implications for magmatic processes at a slow spreading ridge

A diverse volcanic and plutonic rock suite was recovered from the center of the 80 km long ridge segment of the Southwest Indian Ridge (54°S, 7°16 E) between the Islas Orcadas and Shaka Fracture Zones. The cumulus nature of the gabbroic rocks in the suite is indicated by phase, modal and cryptic layering, igneous lamination, and low incompatible element abundances. We present a mass-balance model for calculating the proportions and compositions of cumulus phases and crystallized intercumulus liquid from bulk-rock major element compositions. The model is based on the ability to define a compositional array of basaltic liquids and on the assumption that cumulus minerals are initially in equilibrium with trapped liquid. Calculated proportions of trapped liquid range from 3%–15%; values that are characteristic of adcumulates to mesocumulates. Models of postcumulus crystallization indicate significant enrichments of incompatible elements and buffering of compatible elements in residual trapped liquids, thus explaining the high TiO₂ contents observed in magnesian clinopyroxenes. Cumulus phase assemblages and compositions suggest solidification in shallow level magma chambers, but disequilibrium plagioclase compositions suggest some crystallization at greater depth. Furthermore, basalt compositions projected onto the olivine-clinopyroxenequartz pseudoternary suggest magma generation over a range of pressures (from less than 10 to greater than 20 kb) as well as polybaric fractional crystallization. We suggest that the Southwest Indian Ridge is characterized by low magma supply with small batches of melt that either ascend directly to the surface having undergone limited polybaric crystallization or are trapped in shallow crustal magma chambers where they evolve and solidify to form cumulate gabbros. The adcumulus nature of the gabbros investigated here suggests slow cooling rates typical of large intrusions implying relatively large, but ephemeral magma chambers below segments of the Southwest Indian Ridge.     

Fitting the Linear Model of Coregionalization by Generalized Least Squares

In geostatistical studies, the fitting of the linear model of coregionalization (LMC) to direct and cross experimental semivariograms is usually performed with a weighted least-squares (WLS) procedure based on the number of pairs of observations at each lag. So far, no study has investigated the efficiency of other least-squares procedures, such as ordinary least squares (OLS), generalized least squares (GLS), and WLS with other weighing functions, in the context of the LMC. In this article, we compare the statistical properties of the sill estimators obtained with eight least-squares procedures for fitting the LMC: OLS, four WLS, and three GLS. The WLS procedures are based on approximations of the variance of semivariogram estimates at each distance lag. The GLS procedures use a variance–covariance matrix of semivariogram estimates that is (i) estimated using the fourth-order moments with sill estimates (GLS₁), (ii) calculated using the fourth-order moments with the theoretical sills (GLS₂), and (iii) based on an approximation using the correlation between semivariogram estimates in the case of spatial independence of the observations (GLS₃). The current algorithm for fitting the LMC by WLS while ensuring the positive semidefiniteness of sill matrix estimates is modified to include any least-squares procedure. A Monte Carlo study is performed for 16 scenarios corresponding to different combinations of the number of variables, number of spatial structures, values of ranges, and scale dependence of the correlations among variables. Simulation results show that the mean square error is accounted for mostly by the variance of the sill estimators instead of their squared bias. Overall, the estimated GLS₁ and theoretical GLS₂ are the most efficient, followed by the WLS procedure that is based on the number of pairs of observations and the average distance at each lag. On that basis, GLS₁ can be recommended for future studies using the LMC.