Origin of the magmas and the hydrothermal systems associated with the Skye Central Complex

From Earthwise
Jump to navigation Jump to search
From: Bell, B.R. and Harris, J.W. An excursion guide to the geology of the Isle of Skye : Geological Society of Glasgow, 1986. © 1986 B.R. Bell & J.W. Harris. All rights reserved.

Chapter 12 Origin of the magmas and the hydrothermal systems associated with the Skye Centre[edit]

(A) Introduction[edit]

The genesis and evolution of the magmas involved in the development of the intrusive centre on Skye and its associated lava field have been the subject of numerous investigations over the last thirty years. In the available space it is only possible to give brief summaries of the published studies. More detailed discussions will be found in the references cited.

Our knowledge of the compositions of primary magmas involved in the development of the Skye Centre is based largely on the work of Thompson and his co-workers (see Section (12B), below). In addition, their investigations have shown that crustal contamination has significantly modified these magmas.

The composition and evolution of the magma(s) involved in the development of the Cuillin Complex are considered in Section (12C), whilst the competing hypotheses of crustal melting and crystal-liquid fractionation of basic melts to produce the granites are dealt with in Section (12D). Section (12E) considers the role of magma-mixing to produce the hybrid rocks which occur throughout the centre.

The nature and form of the hydrothermal convective systems associated with the Skye Centre are discussed in Section (12F).

(B) Mantle melting events and crustal contamination of magmas[edit]

Experimental melting studies (Thompson 1974a) have shown that the most magnesian basalts of the Skye Main Lava Series (SMLS, see Section (3D) of Chapter 3) represent primary magmas generated by partial fusion of mantle spinel lherzolite at depths of approximately 60km. Thompson et al. (1980a) found that members of the SMLS have several chemical characteristics very similar to basic rocks recorded from ocean islands such as Iceland and Hawaii. Members of the other main group of basic lavas, the distinctly tholeiitic Preshal Mhor type, which are intercalated with members of the SMLS near to the top of the preserved sequence (see Section (3D) of Chapter 3), are more akin to Mid-Ocean Ridge Basalts (MORB). To account for these compositional differences, Thompson et al. (1980a) proposed a single mantle source and explained their temporal appearances in terms of progressive partial melting of the mantle spinel lherzolite. Initially, SMLS magnesian basalts are generated by small amounts of partial melting, leaving a lherzolitic residuum. Further melting of this depleted mantle generates the Preshal Mhor type of basalts. Similar changes are recorded in the regional dyke swarm (see Section (9C) of Chapter 9).

With this working hypothesis it is still necessary to explain several geochemical 'anomalies' which have come to light during other studies of Lower Tertiary volcanic rocks throughout the province. To explain the very low concentrations of several incompatible trace-elements recorded from the lavas, relative to other transitional and alkali basalt provinces, Morrison et al. (1980) suggested that, prior to the Lower Tertiary magmatic events, some form of depletion of the mantle source-rocks took place. They concluded that this extraction event resulted in the formation of a suite of highly alkaline lamprophyre dykes of Permian age, which occur throughout the province.

Another important aspect of the evolution of the Skye Centre and its associated lava field is the influence that crustal contamination has had in modifying primary magma compositions (Moorbath and Thompson 1980; Dickin 1981; Thompson 1982; Thompson et al. 1982). By investigating whole-rock concentrations of isotopes of Sr, Pb and Nd, Dickin (1981) concluded: (1) Fe-rich members of the SMLS have been selectively contaminated by both lower crustal granulite-facies and upper crustal amphibolite-facies Lewisian Gneiss; (2) Fe-poor members of the SMLS have been selectively contaminated by only the granulitefacies gneiss; (3) members of the Preshal Mhor type of lavas show only a small degree of contamination, by upper crustal amphibolite-facies gneiss; and, (4) basic magmas which evolved to produce granites (see Section (12D), below) were contaminated both in the lower crust by interaction with granulite-facies gneiss, and in the upper crust by the incorporation of small batches of partial melt from amphibolite-facies gneiss.

Thompson et al. (1982) have shown that such contamination also affects the abundances and ratios of Ba, K, Rb, Sr, Th and the light rareearth-elements and that the generation and assimilation of small quantities of silicate partial melt is the dominant contamination mechanism. This process is best facilitated if the main magma conduits are dykes and sills, with high surface area-to-volume ratios, thereby allowing maximum interaction of magma and crust to take place.

(C) The parental magmas of the Cuillin Complex[edit]

Assessing the composition(s) of the parental magma(s) of the Cuillin Complex is difficult because many of these rocks are cumulates. In his original study, Harker (1904) concluded that the layering developed within the complex was the result of the streaking out of heterogeneous magmas, the compositions of which approximated to the rocks which subsequently crystallised (for example, peridotites from peridotitic magma). In a general sense, the presence of orthopyroxene, together with interstitial patches of alkali feldspar (+ quartz) within the cumulates, suggests that the magmas were predominantly tholeiitic.

Wager and Brown (1968), in a general review of the geology of the complex, suggested that the magmas were basic and that extensive crystal-liquid fractionation and convective processes within the chamber had produced the various layered rocks.

Other studies (for example, Dreyer and Johnston 1966; Gibb 1976; Hutchison and Bevan 1977) have favoured parental magmas which tend towards ultrabasic compositions. Studies of ultrabasic minor intrusions associated with the complex (see Section (9I) of Chapter 9) led Gibb (1976) to suggest that an eucritic magma, containing olivine in suspension, would be capable of producing the cumulates present. However, this hypothesis does not account for the calcic plagioclases present within the Outer Layered Eucrite Series or the Inner Layered Gabbro Series (see Sections (4F) and (4K) of Chapter 4, respectively). Hutchison and Bevan (1977) concluded that the parental magmas approximated in composition to that of the associated cumulates. In the case of the ultrabasic cumulates, a parental picritic magma is inferred. Specifically, they note that a 'space problem' exists if basic magmas are involved in the formation of the ultrabasic rocks. Essentially, the magma chamber would have been, for basic magmas, too large, when field relationships of the complex are considered.

In the future, it is likely that mineral chemistry data, combined with accurate determinations of partition coefficients between crystals and liquids, will allow original magma compositions to be determined more accurately. At present, it may be noted that the early magmas were most likely ultrabasic, and that with time they tended towards more basic compositions.

(D) Generation of the acid magmas[edit]

Studies of acid rocks associated with the Skye Centre undertaken prior to 1975 were hampered by a lack of high quality trace-element data. Unfortunately, this led to a dogmatic polarisation of views on the competing hypotheses of partial melting of crustal rocks versus crystal-liquid fractionation of basic magmas, in the production of acid magmas.

The generation of acid magmas by crustal melting has been advocated by many as the dominant mechanism of granite production in the Skye Centre (see reviews by J.D. Bell 1976, 1982, Gass and Thorpe 1976, Meighan 1979, and Thompson 1982). For example, the Coire Uaigneich Granite has been the subject of numerous investigations in order to ascertain its origin (Wager et al. 1953; Brown 1963; Meighan 1976, 1979; Dickin and Exley 1981). Details of these studies are presented in Section (4L) of Chapter 4, where it is noted that often strongly divergent views exist, depending upon the evidence used.

Thompson (1982) has argued strongly that trace-element data may be used to show that crystal-liquid fractionation of tholeiitic basalt magmas is the dominant mechanism of acid magma production. He identifies three groups of granites within the British Tertiary Volcanic Province, on the basis of trace-element characteristics:

(1) Primitive compositions (typically adamellites), with: Ba 850- 1700ppm; Rb 90–135; Sr 120–250; Y ~ 60; Zr 350–400; (Ce/Yb)N 5–6; Eu/Eu* 0.6–0.9. For example: the Glamaig Granite of the Western Red Hills Centre (see Section (6D) of Chapter 6) and the Coire Uaigneich Granite (see Section (4L) of Chapter 4).
(2) Peralkaline leucogranites, with: Ba 30–150ppm; Rb 140–400; Sr 510; Y 130–180; Zr 650–850; (Ce/Yb)N 1.5–4.5; Eu/Eu* 0.1–0.3. For example: the Southern Porphyritic Granite of the Western Red Hills Centre (see Section (6G) of Chapter 6).
(3) Subalkaline leucogranites, with: Ba < 50ppm; Rb 450–600; Sr 5–20; Zr 50–120; (Ce/Yb)N 4–10; Eu/Eu* ~ 0.1.

On the basis of mass balance constraints imposed by trace-element concentrations, Thompson (1982) concludes that granites from any of these three groups cannot represent near-total melts of basement Lewisian Gneiss (see Section (2A) of Chapter 2). Furthermore, the Ba data indicate that the peralkaline granites (Type (2), above) are not low-fraction partial melts of either the basement gneisses, or of an earlier, primitive intrusion (such as Type (1), above). Experimental melting data suggests that any partial melt of basement rocks (either Lewisian Gneiss or Torridonian sedimentary rocks) is always peraluminous (Thompson 1981).

According to Thompson (1982), the basic parental magmas of the granites were derived by the mixing of approximately 10% each of granulite- and amphibolite-facies Lewisian Gneiss with basalt magma of the Preshal Mhor type (see Section (3D) of Chapter 3). It is possible, however, that other basic magma types, such as the Skye Main Lava Series (SMLS) or Fairy Bridge type (see Section (3D) of Chapter 3 and Section (9C) of Chapter 9, respectively), were also involved (Thorpe et al. 1977; Thorpe 1978; B.R. Bell 1984a,b).

(E) Magma-mixing[edit]

On Skye, evidence for the co-existence of acid and basic magmas is plentiful. The role of magma-mixing in the development of several of the subvolcanic intrusions has been discussed in previous chapters (Chapter 4, Section (4L); Chapter 6, Sections (6H) and (6K); Chapter 7, Sections (7B) and (7H)).

In general, fluid dynamic processes within the magma reservoir(s) involved appear to have controlled the interactions of these contrasting magma compositions (B.R. Bell 1983). For example, the removal or addition of magma from a subvolcanic reservoir would create a fluid dynamic situation whereby magma-mixing takes place.

Consider the case of acid magma ponded above an evolving, compositionally-stratified, basic magma. The introduction of a new batch of less-dense, more primitive, basic magma into the bottom of the reservoir will cause both mixing and an overall increase in temperature and pressure. In turn, this will cause the basic magma to convect rapidly and interact along the mutual interface with the acid magma ponded above, forming various hybrid compositions (B.R. Bell 1983).

(F) Hydrothermal systems associated with the Skye Centre[edit]

In a detailed study using the stable isotopes of oxygen and hydrogen, Forester and Taylor (1977) showed that within 4km of the Skye Centre extensive convective flow of heated, meteoric groundwaters caused extreme depletion of the heavier isotopes of these elements in the volcanic and subvolcanic units and the surrounding country-rocks. This depletion decreases towards the margin of the intrusive centre and resulted from sub-solidus interactions between the heated ground-waters and the still-hot igneous rocks. Studies of mineral separates, however, suggests that some of these interactions took place at magmatic temperatures. For. example, data from some of the oldest granites indicate that only 80% of the oxygen-exchange took place at sub-solidus temperatures. Furthermore, they conclude that both the Coire Uaigneich Granite (see Section (4L) of Chapter 4) and the Southern Porphyritic Granite (see Section (6G) of Chapter 6) are composed, in part, of partially-melted, hydrothermally-altered country-rocks.

It is suggested that the hydrothermal system had an average integrated water/rock ratio of one, with at least 2000km3 of heated, meteoric groundwater passing through the intrusive centre. Obvious manifestations of these fluid-rock interactions include: the secondary mineral assemblages found within many of the igneous rocks associated with the Skye Centre; and, the presence of numerous veins containing calcite, epidote, chlorite, and other low-temperature 'hydrothermal' minerals, also found within these igneous rocks, as well as in the surrounding country-rocks (see Section (4M) of Chapter 4 and Section (7I) of Chapter 7).


Appendix 1: Glossary of petrological names and terms[edit]

Appendix 2: Glossary of fossil names[edit]

Appendix 3: Glossary of place names and grid references[edit]

At all times follow: The Scottish Access Codeand Code of conduct for geological field work