GEMOC ARC National Key Centre

Doone Wyborn,¹

1. GEMOC ANU

The great range of granitic bodies of the Lachlan Fold Belt (LFB) provides us with an ideal laboratory for understanding the processes that lead to diversity both within and between granite bodies. At least two large groups of granites of the LFB, the highly fractionated (Rb-enriched) felsic granites and the Boggy Plain Supersuite (BPS), have characteristics that are distinct from most others. These differences have been pointed out in many publications over the last two decades.

The range of compositions within these two specific groups has been ascribed to fractional crystallisation. That is, crystals precipitate from the melt, change the composition of the melt as they crystallise, and hence progressively change the composition of the solid cumulate rocks that are formed. Such solid rock products would comprise a mix of precipitated crystals, and a proportion of trapped melt that is residual from crystallisation. This principle is well-established in mafic and ultramafic intrusives, but has been more difficult to quantify in felsic rocks. Only after combining all the information derived from field relations, petrography, mineral chemistry, lithochemistry and isotopes, for the two distinctive groups, and making comparisons with the other more common granites of the LFB can models be developed that best satisfy all information.

The most compelling evidence that fractional crystallisation is taking place in the felsic granites lies in the behaviour of trace elements accommodated in feldspar. Strong depletions in Ba and Sr are coupled with strong increases in Rb and Cs, while the major element compositions change very little. These characteristics can only be imparted on a felsic liquid by progressive fractional crystallisation of feldspars and quartz close to the Tuttle and Bowen (1958) minimum. The problem is to delineate processes that start with a primary liquid and end with a separation of crystallised phases from a derivative liquid. These processes are taking place at essentially the same temperature and with virtually indistinguishable physical properties of the primary and derivative liquids. We must call upon field relations to provide some evidence of the processes, and the best example is the Lottah Granite in northeastern Tasmania studied by Mackenzie et al. (1988). In that pluton a north-central stock is the most primitive. Flat sheets or sill-like bodies spread to the south and southwest for several kilometres, and the rocks become progressively more fractionated away from the central stock. Clearly the fractionation process results from the magma flowing southwards along the sill leaving cumulates behind during its progression. It is that dynamic environment that facilitates the separation of derivative liquid from crystallising phases. For the Lottah Granite, notions of alteration producing the range of chemical compositions of the solid products are now discredited, from a large body of comparative data on similar rocks elsewhere, and from experimental data.

In the BPS, the fractional crystallisation that produced the great range of compositions shown by the group can be ascribed to different dynamical mechanisms to the felsic granites. However before discussing these, it should be pointed out that there are marked differences between mafic rocks of this group and those of other granite groups in the LFB. These differences are obvious, from a field comparison through to the most detailed mineral and chemical investigations. Mafic rocks of the BPS consist of high temperature cumulate crystals with crystallised trapped interstitial derivative melt that ranged in volume from as little as 1% to greater than 40% of the total. The absence of enclaves and crystal clots is in great contrast to mafic rocks from other LFB I-type granites, thought to contain abundant restite. These BPS cumulates are complemented by felsic rocks which include both intrusives and extrusives (high temperature rhyolites). This complementarity leads to a compositional gap with only small volumes of rock in the range 65-70% SiO2. In contrast, that compositional range dominates among the restite-bearing I-type granites. There is also a strong contrast of BPS extrusives and those thought to contain abundant restite. BPS rhyolites (SiO2 >70%) contain only sparse high-temperature (1000_C) phenocrysts (plagioclase, pyroxenes, magnetite and ilmenite), while the restite-rich dacites (I- and S-type, SiO2 <70%) contain abundant phenocrysts (up to 60%) of minerals equilibrated at lower temperatures (800-900_C).

Density differences between the primary and derivative liquids drove the dynamical processes leading to fractional crystallisation of BPS liquids. This was achieved by sidewall crystallisation in some plutons, by the process termed convective fractionation by Sparks et al. (1984), or liquid fractionation by McBirney and his co-workers in a series of publications in 1985 and 1997. In the BPS magma chambers, sidewall precipitation of high-temperature cumulate phases produced a derivative liquid with lower density than the primary liquid. This low density liquid escaped up the walls of the crystallising front. Initially only small amounts of the derivative liquid mixed back into the primary liquid in the core of the chamber, since the derivative liquid was hot, with low viscosity. Derivative liquid accumulated at the top of the chamber with a sharp interface between it and the primary liquid. Some derivative liquid back-mixed into the primary liquid, changing its composition progressively. As inwards solidification proceeded, a greater proportion of the derivative liquid, with progressively higher viscosity (lower temperature and more felsic), back-mixed, until eventually all the derivative liquid was captured into the main chamber. Final solidification proceeded in a closed system. Examples of the processes are found in the Yeoval Batholith where it can be demonstrated, in excellent exposures, that a zoned diorite unit is overlain by a felsic granite cap derived from the same primary melt as the diorite (James, 1997).

The ability of BPS melts to undergo convective fractionation producing complementary dioritic cumulates and felsic derivative liquids is largely related to the relative high proportion of minimum-melt fraction present in the primary melt, which is, in turn, related to the rather potassic character of the liquid in this case. During sidewall crystallisation of high temperature minerals, a relatively large volume of derivative liquid will locally form, resulting in an open crystalline structure and providing an easy escape for the liquid. Taking a tonalitic primary liquid at similar temperature (>1000_C), sidewall crystallisation of high temperature minerals will result in a high proportion of the melt solidifying. The volume of derivative liquid formed is much smaller than with potassic primary liquids. The crystalline structure will be more tightly packed and the derivative liquid will be trapped. No fractional crystallisation will take place. The magma chamber will progressively solidify with little change in composition being possible. A homogeneous body will result. Such bodies are common in the large tonalitic continental margin batholiths of the world.

References

Baker, B H, and McBirney, A H, 1985. Liquid Fractionation: Part 3: Geochemistry of zoned magmas and the compositional effects of liquid fractionation. Journal of Volcanology and Geothermal Research 24, 55-81.

James, M, 1997. The geology of the Gullengambel area, central New South Wales. BSc (Hons) thesis, Australian National University. 73pp.

Maaloe, S, and McBirney, A R, 1997. Liquid fractionation. Part 4: Scale models for liquid fractionation of calc-alkaline magmas. Journal of Volcanology and Geothermal Research 76, 111-125.

Mackenzie, D E, Black, L P, and Sun, S-S, 1988. Origin of alkali-feldspar granites: An example from the Poimena Granite, northeastern Tasmania, Australia. Geochimica et Cosmochimica Acta 52, 2507-2524.

McBirney, A R, Baker, B H, and Nilson, R H, 1985. Liquid Fractionation: Part 1: Basic principles and experimental simulations. Journal of Volcanology and Geothermal Research 24, 1-24.

Nilson, R H, McBirney, A R, and Baker, B H, 1985. Liquid Fractionation: Part 2: Fluid dynamics and quantitative implications for magma systems. Journal of Volcanology and Geothermal Research 24, 25-54.

Sparks, R S J, Huppert, H E, and Turner, J S, 1984. The fluid dynamics of evolving magma chambers. Philosophical Transactions of the Royal Society of London A310, 511-534.
Tuttle, O F, and Bowen, N L, 1958. Origin of granite in the light of experimental studies in the system NaAlSi3O8-KalSi3O8-SiO2-H2O. Memoir of the Geological Society of America 74, 153pp.