Colin Farrow

RB-SR CLOSURE TEMPERATURES IN BI-MINERALIC ROCKS - A MODE EFFECT AND TEST FOR DIFFERENT DIFFUSION-MODELS

Abstract Various diffusion models can be applied to RbSr systems in cooling rocks. It is argued that a closed-system diffusion-controlled model, involving local equilibrium along grain boundaries, is most appropriate for describing the exchange of Sr isotopes between minerals in a rock sample during cooling. A finite-difference numerical method is used to solve the diffusion and mass-balance equations. Closure temperatures of mineral-pair RbSr isochrons are predicted to depend on the factors involved in Dodsons infinite reservoir closure temperature formulation (cooling and diffusion rate, grain size and shape) of both the minerals. In addition, the closure temperature for mineral pairs is also dependent on the proportion of Sr in each mineral, which is dominantly a function of rock mode. This implies that Rb-Sr cooling ages from interbedded rocks having the same cooling history, but distinct modes, should differ: at slow cooling rates age differences could be > 100 Ma. Such effects, if unrecognised could result in erroneous cooling curves, but if recognised could be utilised in estimating true cooling rates. Our closed-system model :may also apply to other isotope decay schemes, such as the SmNd system. A simple test is proposed which would allow the relevance of different diffusion models to be assessed.

Cool: A FORTRAN-77 computer program for modeling stable isotopes in cooling closed systems

Abstract COOL is a FORTRAN-77 program for modeling stable isotope ratios of minerals in cooling closed systems. The closure temperatures of the minerals in a rock are used to define the temperature at which each mineral ceases stable isotope exchange with the other minerals in the rock. The mass balance relationship between the minerals is used together with the closure temperatures to calculate the δ values of the individual minerals as the rock cools. As well as calculating the mineral δ values in a cooling rock, COOL also can calculate apparent isotope equilibrium temperatures and estimate the rate at which a rock has cooled from experimental data. COOL can be used to model mineral δ values through a limited range of temperatures, so that the input data can be obtained from, or the output data used in, programs which model stable isotope exchange in open systems under isothermal conditions.

The relationship of argon retentivity and chemical composition of hornblende

K-Ar ages of 26 hornblende separates from Connemara, western Ireland do not show any correlation with the chemical composition of the hornblendes, including their iron contents. It is suggested that this is the usual pattern as there are only rare reports of compositional control of K-Ar ages.

Geochemical Modelling of Igneous Processes – Principles And Recipes in R Language

Provides basics of R language syntax and its application to geochemical problems; gives a comprehensive introduction to the GCDkit system Explains fundamentals of numerical modelling of igneous processes (including not only formulae, but also showing the successful modelling strategies) Includes numerous worked examples indicating how geochemical modelling helps us to understand geological problems

K-POOR TITANIAN FLUOR-RICHTERITE FROM NEAR NULLAGINE, WESTERN AUSTRALIA

Abstract Strongly zoned amphibole formed oftitanian fluor-richterite cores with 2.24% TiO2 and margins oftitanian ferro-richterite with 6.65% TiO2 is described from a basaltic andesite. The cores are richer in F, AI, Mg, and Ca than the edges, which are richer in OH, Ti, Fe, Na, and K. Similar sodic-calcic amphiboles are widely known to appear in lamproites, but typically they contain 5% or more K2O, whereas the present sample reaches a maximum of 1.18% K20 at the margin. As is characteristic of titanian potassium richterite, there is Ti in tetrahedral coordination and a pink to yellow pleochroism. Ba, Sr, Li, B, CI, Be, Nb, La, Ce, Pr, Y, V, and Zr contents are given. Zoned augite with Ca42Mg42Fe16 cores and Ca43Mg36Fe22 rims is also present, plus chloritic material and feldspar.

Data Manipulation and Simple Calculations

This chapter will demonstrate the practical use of the R language (for overview of its syntax, see Appendix A) and GCDkit (Appendix B) to solve common problems in igneous geochemistry. We shall follow the basic procedure from loading the data into the system, through their subsetting, calculation of basic indexes (such as mg# or A/CNK values) or cationic parameters (after Niggli, Debon & Le Fort and De la Roche), to normative recalculations (e.g., CIPW norm). Briefly mentioned are also statistical applications of the R language, such as obtaining simple descriptive statistics and use of factors-based grouping to deal with complex geochemical data sets.

Constraining a Model

This chapter shows how information can be gleaned from various sources to con-strain the parameters needed to build a full model. Some of the parameters come from observations of the petrology or the geochemical patterns of the rocks stud-ied. In other cases, information must be sought from the literature. The remaining parameters can be calculated based on the previous information. The art of modelling consists in assembling this disparate information in a consistent set.

Progressive Melting of a Metasedimentary Sequence: the Saint-Malo Migmatitic Complex, France

This chapter shows a worked example, modelling the crustal anatexis to form a migmatitic complex. Starting with geological and petrological data, we describe the main geochemical features of the lavas and model their evolution. In this environment, the melts are not well extracted from their solid residue. They are poorly homogenised and their composition largely reflects the variations of the particular source lithologies. Fortunately, field relations allow to directly constrain the local melt amount. Increasing melt fractions correspond to successive melting reactions, and thus a residue with an evolving composition. We propose, therefore, a strategy based on describing the evolution of melt’s composition for a given source as a function of the melt amount (and therefore of the nature of the residue), and of the source’s composition. Finally, we bracket the possible range of melts between the compositions derived from two end-member sources.

Dilute Trace Elements: Partition Coefficients

This chapter presents the key concept of the partition coefficient (applied to igneous systems). The partition coefficient is the ratio between the composition in a mineral phase, and the concentration in the melt, for a given element. Whereas major-element composition of the mineral is known or it can be assumed (see Chap. 6), for trace elements the best constrained parameter is the partition coefficient. We explore some of the factors that control its value, as well as, in turn, the way this coefficient controls the distribution of elements between liquid and solid phases.

Differentiation of a Calc-Alkaline Volcanic Series: Example of the Atacazo-Ninahuilca Volcanoes, Ecuador

This chapter presents a worked example, based on the evolution of a calc-alkaline differentiation series, from recent volcanoes in Ecuador (Andes). Starting with geological and petrological data, we describe the main geochemical features of the lavas and model their evolution. We show that fractional crystallization was the dominant process, and that all the lavas in the volcano are related by fractionation from a common parent. The differentiation story is modelled here as a two-step process, with two successive cumulate compositions. We also explore some uncertainties of the modelling exercise and discuss the range of possible solutions permissible by geochemistry.