Phlogopite in the generation of olivine-melilitites from Namaqualand, South Africa and implications for element fractionation processes in the upper mantle

Major and trace element and Sr, Nd and Pb isotope analyses are presented for thirteen olivine-melilitites from Namaqualand, South Africa. Major element variations are consistent with derivation from carbonated garnet-peridotite at depths of at least 100 km and trace element abundances indicate melt fractions of 4%. Ubiquitous negative K anomalies and low, buffered K₂O concentrations are interpreted to reflect the effect of residual phlogopite during melting. It is suggested that phlogopite stability and low melt potassium saturation concentrations are enhanced by high CO₂/(CO₂ + H₂O) conditions. Residual phlogopite can also account for low measured Rb/Sr, Ba/Sr and Th/U ratios in the melilitites. REE abundances are controlled by residual garnet and hence Sm/Nd ratios are low (0.13–0.18). U/Pb ratios vary from 0.05 to 5 and are a function of Pb concentration which is in turn controlled by residual Pb-rich phase (probably sulphide). Nd and Sr isotopes are comparable with OIB from St. Helena, although two samples extend to higher ⁸⁷Sr/⁸⁶Sr ratios. Present day Pb isotopes are much more variable and partly reflect radiogenic growth since emplacement as a result of the highly variable U/Pb ratios.

Many of the trace element characteristics of the melilitites are distinct from those of within-plate potassic magmas despite both being derived from phlogopite-bearing, enriched mantle source regions. This can be attributed to the depth at which source enrichment occurred and the subsequent control exerted by phlogopite and carbonate during melting. In contrast to melilitites, potassic magmas are derived from shallower depths under low CO₂/(CO₂ + H₂O) conditions and at higher temperatures at which phlogopite melts more readily.

The incompatible element ratios of the melilitites are also similar to those both observed in HIMU ocean island basalts (OIB) and inferred for HIMU OIB source regions from isotope variations (viz, low Sm/Nd, Rb/Sr, K/Nb, Th/U and high U/Pb and Ce/Pb). It is suggested that HIMU OIB's may be derived from sources that have been subject to enrichment by a melt generated in the presence of residual phlogopite.     

Turbulent dissipation of cold air lake in a basin

Summary Problems of turbulent dissipation of a cold air lake (CAL) and the inversion layer bordering CAL on the upper boundary are presented and studied with the compound model. In wintertime such cold air lakes can persist for days even if rather strong winds are blowing above them. The required conditions for CAL dissipation are removed processes of its formation or maintenance, as well as a sufficiently strong invasion of turbulence in the inversion layer from above down-wards. By this, the inversion layer at first becomes stronger and dissipation is stopped, until the increase of turbulent kinetic energy of the upper flow enables further dissipation. Such turbulent dissipation process is shown by the model for typical conditions and for different initial values of the relevent variables.With 6 Figures     

Experimental determination of U and Th partitioning between clinopyroxene and natural and synthetic basaltic liquid

Clinopyroxene-silicate liquid partition coefficients for U and Th have been determined by particle track radiography from 1 atm crystallization experiments at controlledfO₂. Two natural basaltic and one synthetic composition were studied atfO₂ values from the NiNiO oxygen buffer to 1 log unit more oxidizing than FeFeO (IW+ 1). Over the range offO₂ values and compositions studied,D_U^cpx/liq = 0.0034–0.015,D_Th^cpx/liq = 0.008–0.036, andD_ThD_U= 3.4–1.1. With decreasingfO₂,D_ThD_U can decrease by up to a factor of 3 for a given composition, primarily from an increase inD_U^cpx/liq, which we interpret as resulting from an increase in the proportion of tetravalent U in the system with decreasingfO₂. This demonstrates that crystal-liquid UTh fractionation isfO₂ dependent and that U in terrestrial magmas is not entirely tetravalent.D_Th^cpx/liq appears to decrease in the two basalts at the lowestfO₂, possibly as a result of changes in composition withfO₂.

Our data show the sense of UTh fractionation by clinopyroxene-liquid partitioning is consistent with previous experimental determinations, in thatD_Th^cpxD_U^cpx> 1 in all cases. This indicates that, during partial melting, the liquid will have aTh/U ratio less than the clinopyroxene in the source. The observed²³⁸U²³⁰Th disequilibrium in MORB requires that the partial melt should have aTh/U ratio greater than the bulk source, and, therefore, cannot result from clinopyroxene-liquid partitioning. Further, the magnitudes of the measured partition coefficients are too small to generate significant UTh fractionation in either direction. Assuming that clinopyroxene contains the bulk of the U and Th in MORB source, our results indicate that²³⁸U²³⁰Th disequilibrium in MORB may not be caused by partial melting at all.     

Numerical instabilities of a local transmitting boundary

A local transmitting boundary is presented in a compact form, which can be directly incorporated into finite elements. The accuracy of the boundary is studied thoroughly for a one-dimensional model in order to clarify numerical instabilities introduced by the boundary. Discretization of the model and reflection from the boundary are rigorously considered in the study, and the mechanism of the instability is then illuminated in the frequency domain by the amplification of reflection from the boundary and the multi-reflection of wave motion in a finite computational region. Typical characteristics of the instability in the time domain are illustrated by numerical results of the simple model and explained completely by the mechanism. On the basis of this understanding of the mechanism, a modified transmitting boundary is devised and its stability criterion is given for the one-dimensional model.     

A nonlinear deformation modulus based on rock mass classification

Summary This paper describes the development of an empirical nonlinear stress dependent expression for the deformation modulus of rock masses based on rock mass classification. The expression defines the deformation modulus as the ratio of the deviator stress at failure to the major principal strain at failure. The Hoek and Brown failure criterion is used to predict the deviator stress at failure. Research was directed toward developing a failure criterion defining the major principal strain at failure. The expression for the deformation modulus was extended to rock mass conditions through correlations with observed deformations from case history studies and predicted deformations from finite element analyses.     

Spiral filaments — The signature of singular points in a large delta spot

A large delta spot (active region NOAA 6891, October 23 – November 4, 1991) is analysed, and it is found that some spiral filaments across the spot can be regarded as signatures of a singular point entity (SPE) which lies near a separator of this complicated magnetic field region. Near such an entity, 86% of the flares in the region were produced, including two white-light flares, one of which, being the largest flare of the region, was accompanied by a powerful mass ejection. In an island delta spot, a SPE could be recognized very close to the usual U-shaped inversion lines. Together with the other characteristics (bright H emission, highly sheared magnetic field, umbra obscured by H, magnetic flux imbalance in the range 21–31), the SPE can help us to predict effectively the sites for great flares to occur (Zirin and Liggett, 1987).On leaving from Purple Mountain Observatory, Nanjing, China.     

Uranus and Neptune interior models

This paper is concerned with the interior structure of Uranus and Neptune. Our approach is three-fold. First, a set of three-layer models for both Uranus and Neptune are constructed using a method similar to that used in the study of the terrestrial planets. The variations of the mass density (s) and flattening e(s) with fractional mean radius s for two representative models of Uranus and Neptune are calculated. The results are tabulated. A comparison of these models shows that these two planets are probably very similar to each other in their basic dynamical features. Such similarity is very seldom seen in our solar system. Secondly, we check the conformance between the theoretical results and observational data for the two planets. And thirdly, the 6th degree Stokes zonal parameters for Uranus and for Neptune are predicted, based on the interior models put forward in this paper.     

Analysis of the Marshall Islands Fireball of February 1, 1994

On February 1, 1994, a large meteoroid impacted over the Pacific Ocean at 2.6° N, 164.1° E. The impact was observed by space based IR sensors operated by the US Department of Defense and by visible wavelength sensors operated by the US Department of Energy. During entry the object broke into several pieces, one of which detonated at 34 km and another at 21 km altitude. The entry velocity of the object is estimated to be 24–25 km/sec. Based on the visible wavelength data, the integrated intensity of the radiated energy of the fireball was approximately 1.3 × 10¹³ joules. Assuming a 6000 K black body and a 30% efficiency for the conversion of the kinetic energy of the body into visible light, we estimate the mass of the body to be between 1.6×10⁵ kg and 4.4×10⁶ kg, and to have a diameter of between 4.4 and 13.5 meters. The object entered at a 45° angle, traveling on a heading of approximately 300°, i.e. from the southeast to the northwest. Calculations using a gross-fragmentation model indicate that the body was most likely a stony object larger than 10 m with an Apollo orbit prior to impact.E T Space SystemsSandia National LaboratoriesOndejov Observatory     

Physical conditions in the compact double star HD 187399

We present the results of a study of the spectra of the compact double system HD 187399. We have measured and identified lines in the spectrum of the main star and computed their equivalent widths, the central residual intensities, and the electron densities in the atmosphere of the main star and the common envelope of the system. We have constructed the curve of radial velocities of the main star and determined the expansion velocity of the common envelope of the system: v_r =90 km/sec. We have measured the velocity of the interstellar calcium: v_r =–3.5 km/sec. We have established the agreement of the radial velocity curves for the main star obtained in the present paper and in that of Merrill [1]. The comparison enabled us to make the following conjecture: there has been no noticeable loss of mass in the system HD 187399 over the past 35 years, and possibly all that has occurred is a flow of matter from one component toward the other. We note the strong similarity with the Lyrae system.Translated fromAstrofizika, Vol. 38, No. 2, 1995.     

Close Binary System Go Cygnus: Standard Ubv Observations and Light Curve Analysis

Standard UBV light curves of eclipsing binary star GO CYGNUS havebeen presented. These light curves indicate a Lyrae typevariations with circular orbit. The Light curve analysis have beencarried out on three colours U, B and V. The results obtained fromthese analysis represent an occultation solution with r r =1.24.