The Ensemble Kalman Filter: theoretical formulation and practical implementation

The purpose of this paper is to provide a comprehensive presentation and interpretation of the Ensemble Kalman Filter (EnKF) and its numerical implementation. The EnKF has a large user group, and numerous publications have discussed applications and theoretical aspects of it. This paper reviews the important results from these studies and also presents new ideas and alternative interpretations which further explain the success of the EnKF. In addition to providing the theoretical framework needed for using the EnKF, there is also a focus on the algorithmic formulation and optimal numerical implementation. A program listing is given for some of the key subroutines. The paper also touches upon specific issues such as the use of nonlinear measurements, in situ profiles of temperature and salinity, and data which are available with high frequency in time. An ensemble based optimal interpolation (EnOI) scheme is presented as a cost-effective approach which may serve as an alternative to the EnKF in some applications. A fairly extensive discussion is devoted to the use of time correlated model errors and the estimation of model bias.Responsible Editor: Jörg-Olaf Wolff     

A consistent derivation of the wave-energy equation from basic hydrodynamic principles

Based on a decomposition of the velocity into mean flow, turbulent and wave components, momentum and hereafter a wave-energy equation is derived. It contains a turbulent energy dissipation term which is closed by applying a wave-related mixing length model and linear wave theory solutions. This closure produces a non-linear turbulent wave-energy dissipation including the wave energy in a 5/2 power law. The theory is able to predict correctly the shape of deep-water wave spectra according to Phillips' similarity law.Responsible Editor: Hans Burchard     

Development and sensitivity analysis of a model for assessing stratification and safety of Lake Nyos during artificial degassing

To prevent the recurrence of a disastrous eruption of carbon dioxide (CO₂) from Lake Nyos, a degassing plan has been set up for the lake. Since there are concerns that the degassing of the lake may reduce the stability of the density stratification, there is an urgent need for a simulation tool to predict the evolution of the lake stratification in different scenarios. This paper describes the development of a numerical model to predict the CO₂ and dissolved solids concentrations, and the temperature structure as well as the stability of the water column of Lake Nyos. The model is tested with profiles of CO₂ concentrations and temperature taken in the years 1986 to 1996. It reproduces well the general mixing patterns observed in the lake. However, the intensity of the mixing tends to be overestimated in the epilimnion and underestimated in the monimolimnion. The overestimation of the mixing depth in the epilimnion is caused either by the parameterization of the k-epsilon model, or by the uncertainty in the calculation of the surface heat fluxes. The simulated mixing depth is highly sensitive to the surface heat fluxes, and errors in the mixing depth propagate from one year to the following. A precise simulation of the mixolimnion deepening therefore requires high accuracy in the meteorological forcing and the parameterization of the heat fluxes. Neither the meteorological data nor the formulae for the calculation of the heat fluxes are available with the necessary precision. Consequently, it will be indispensable to consider different forcing scenarios in the safety analysis in order to obtain robust boundary conditions for safe degassing. The input of temperature and CO₂ to the lake bottom can be adequately simulated for the years 1986 to 1996 with a constant sublacustrine source of 18 l s^–1 with a CO₂ concentration of 0.395 mol l^–1 and a temperature of 26 °C. The results of this study indicate that the model needs to be calibrated with more detailed field data before using it for its final purpose: the prediction of the stability and the safety of Lake Nyos during the degassing process.Responsible Editor: Hans Burchard     

Mineral-scale Sr-isotope constraints on magma evolution and chamber dynamics in the Rum layered intrusion,Scotland

Sr isotopic zoning within single plagioclase crystals from rocks from Unit 9 of the Rum layered intrusion is used to infer events during crystal growth in a magma undergoing contamination. The ⁸⁷Sr/⁸⁶Sr diversity among minerals and between cores and rims of plagioclase crystals increases as the boundary between unit 9 and the overlying Unit 10 peridotite is approached. Models of near-solidus interaction of the cumulate with a fluid or melt, or large scale textural re-equilibration, cannot easily account for the systematic differences in ⁸⁷Sr/⁸⁶Sr between small crystals and the rims of larger crystals.We propose a simple interpretation in which crystal growth is concentrated along the cool margins of the reservoir. Crystals are subsequently advected to a site of accumulation at the base of the reservoir, probably by episodic plume-like dense downwellings allowing mixing of isotopically zoned and unzoned crystals.If the core-rim isotope variations are inherited from primary magmatic growth, then the small distances over which they are now preserved (1–2 mm) place constraints on the minimum cooling rate of the intrusion. Although the length scale of diffusive equilibration is influenced by a number of poorly-constrained variables (starting temperature, feldspar composition, temperature-time path) cooling was clearly very rapid with cooling to effective closure (~1,000 °C) within a few thousand years.Editorial responsibility: I. Parsons     

Clinopyroxenes from mantle-related xenocrysts in alkaline basalts from Hannuoba (China): augite–pigeonite exsolutions and their thermal significance

Optically homogeneous augite xenocrysts, closely associated with spinel–peridotite nodules, occur in alkali basalts from Hannuoba (Hebei province, China). They were studied by electron and X-ray diffraction to define the occurrence and significance of pigeonite exsolution microtextures. Sub-calcic augite (Wo₃₄) exsolved into En_62–62Fs_25–21Wo_13–17 pigeonite and En_46–45Fs_14–14Wo_40–42 augite, as revealed by TEM through diffuse coarser (001) lamellae (100–300 Å) and only incipient (100) thinner ones (<70 Å). C2/c augite and P2₁/c pigeonite lattices, measured by CCD-XRD, relate through a(Aug)?a(Pgt), b(Aug)?b(Pgt), c(Aug)≠c(Pgt) [5.278(1) vs 5.189(1)Å] and β(Aug)≠β(Pgt) [106.55(1) vs 108.55(2)°]. Cell and site volumes strongly support the hypothesis that the augite xenocrysts crystallised at mantle depth from alkaline melts. After the augite xenocrysts entered the magma, (001) lamellae first formed by spinodal decomposition at a T_min of about 1,100 °C, and coarsened during very rapid transport to the surface; in a later phase, possibly on cooling, incipient (100) lamellae then formed.     

Activity–composition relations for phases in petrological calculations: an asymmetric multicomponent formulation

For petrological calculations, including geothermobarometry and the calculation of phase diagrams (for example, P–T petrogenetic grids and pseudosections), it is necessary to be able to express the activity–composition (a–x) relations of minerals, melt and fluid in multicomponent systems. Although the symmetric formalism—a macroscopic regular model approach to a–x relations—is an easy-to-formulate, general way of doing this, the energetic relationships are a symmetric function of composition. We allow asymmetric energetics to be accommodated via a simple extension to the symmetric formalism which turns it into a macroscopic van Laar formulation. We term this the asymmetric formalism (ASF). In the symmetric formalism, the a–x relations are specified by an interaction energy for each of the constituent binaries amongst the independent set of end members used to represent the phase. In the asymmetric formalism, there is additionally a "size parameter" for each of the end members in the independent set, with size parameter differences between end members accounting for asymmetry. In the case of fluid mixtures, for example, H₂O–CO₂, the volumes of the end members as a function of pressure and temperature serve as the size parameters, providing an excellent fit to the a–x relations. In the case of minerals and silicate liquid, the size parameters are empirical parameters to be determined along with the interaction energies as part of the calibration of the a–x relations. In this way, we determine the a–x relations for feldspars in the systems KAlSi₃O₈–NaAlSi₃O₈ and KAlSi₃O₈–NaAlSi₃O₈–CaAl₂Si₂O₈, for carbonates in the system CaCO₃–MgCO₃, for melt in the melting relationships involving forsterite, protoenstatite and cristobalite in the system Mg₂SiO₄–SiO₂, as well as for fluids in the system H₂O–CO₂. In each case the a–x relations allow the corresponding, experimentally determined phase diagrams to be reproduced faithfully. The asymmetric formalism provides a powerful and flexible way of handling a–x relations of complex phases in multicomponent systems for petrological calculations.     

New density measurements on carbonate liquids and the partial molar volume of the CaCO<Subscript>3</Subscript> component

Density measurements on nine liquids in the CaCO₃–Li₂CO₃–Na₂CO₃–K₂CO₃ quaternary system were performed at 1 bar between 555 and 969 °C using the double-bob Archimedean method. Our density data on the end-member alkali carbonate liquids are in excellent agreement with the NIST standards compiled by Janz (1992). The results were fitted to a volume equation that is linear in composition and temperature; this model recovers the measured volumes within experimental error (±0.18% on average, with a maximum residual of ±0.50%). Our results indicate that the density of the CaCO₃ component in natrocarbonate liquids is 2.502 (±0.014) g/cm³ at 800 °C and 1 bar, which is within the range of silicate melts; its coefficient of thermal expansion is 1.8 (±0.5)×10^–4 K^–1 at 800 °C. Although the volumes of carbonate liquids mix linearly with respect to carbonate components, they do not mix linearly with silicate liquids. Our data are used with those in the literature to estimate the value of in alkaline silicate magmas (20 cm³/mol at 1400 °C and 20 kbar), where CO₂ is dissolved as carbonate in close association with Ca. Our volume measurements are combined with sound speed data in the literature to derive the compressibility of the end-member liquids Li₂CO₃, Na₂CO₃, and K₂CO₃. These results are combined with calorimetric data to calculate the fusion curves for Li₂CO₃, Na₂CO₃, and K₂CO₃ to 5 kbar; the calculations are in excellent agreement with experimental determinations of the respective melting reactions.Editorial responsibility: I Carmichael     

Argon retention properties of silicate glasses and implications for <Superscript>40</Superscript>Ar/<Superscript>39</Superscript>Ar age and noble gas diffusion studies

Sized aggregates of glasses (47–84 wt% SiO₂) were fused from igneous-derived cohesive fault rock and igneous rock, and step-heated from ~400 to >1,200 °C to obtain their ³⁹Ar diffusion properties (average E=33,400 cal mol^?1; D _o=4.63×10^?3 cm² s^?1). At T<~1,000 °C, glasses containing <~69 wt% SiO₂ and abundant network-forming cations (Ca, Fe, Mg) reveal moderate to strong non-linear increases in D and E, reflecting structural modifications as the solid transitions to melt. Extrapolation of these Arrhenius properties down to typical geologic T-t conditions could result in a 1.5 log₁₀ unit underestimation in the diffusion rate of Ar in similar materials. Numerical simulations based upon the diffusion results caution that some common geologic glasses will likely yield ⁴⁰Ar/³⁹Ar cooling ages rather than formation ages. However, if cooling rates are sufficiently high, ambient temperatures are sufficiently low (e.g., <65–175 °C), and coarse particles (e.g., radius (r) >~1 mm) are analyzed, glasses with compositions similar to ours may preserve their formation ages.     

Application of consequence-based design criteria in regions of moderate seismicity

Current design criteria and principles of earthquake engineering design are reviewed, including safety factors, probabilistic approach, and two-level and multi-level functional design ideas. The modern multi-functional idea is discussed in greater details. When designing a structure, its resistance to and the intensity of the earthquake action are considered. The consequence of failure of the structure is considered only through a rough and empirical factor of importance, ranging usually from 1.0 to 1.5. This paper suggests a method of “consequence-based design,” which considers the consequences of malfunctioning instead of simply an importance factor. The main argument for this method is that damage to a structure located in different types of societies may have very different consequences, which are dependant on its value and usefulness to the society and the seismicity in the region.