PHYSICAL AND CHEMICAL processes in A LARGE magma CHAMBER:
THE BJERKREIM-SOKNDAL INTRUSION
Brian Robins
Steeply-dipping, modally-layered, plagioclase-hypersthene-magnetite-ilmenite cumulate (norite) from MCU III in the Bjerkreim lobe of the Bjerkreim -Sokndal Intrusion. Stratigraphic up is to the left. The blue A4 map case lying on the outcrop provides a scale. Layered sequences such as this provide a tangible record of the physical and chemical processes that took place a large magma chamber. |
Magma
chambers are large pockets of molten rock that are emplaced, slowly cool,
crystallise and differentiate within the crust of the Earth, commonly beneath
active central volcanoes. The processes that take place in magma chambers are
important as they govern the frequency, nature and products of volcanic
eruptions as well as the compositional evolution of the enclosed magma(s). They
also lead to the concentration of elements such as Cr, Fe, Ti, V, Pt, Pd, Au, P
and Al into economic deposits. There is exploitation or test mining of large
deposits of these important elements within or adjacent to fossil magma
chambers at several places in Norway, Finland, Greenland (and many other
locations in the world), and active exploration for others.
Cooling and
crystallisation of low-viscosity magma may lead to various forms of magma
circulation in chambers, such as thermal, compositional and two-phase
convection and possibly also double-diffusive convection. Cooling of magma
chambers is commonly interrupted by periodic replenishment and the incursions
of fresh magma may interact and mix in various ways with the resident melts,
depending principally on viscosity and density differences and the momentum of
the inflowing melts. Melting of rocks forming the walls and roof of chambers,
or occurring as included blocks (xenoliths), can result in a significant degree
of contamination of the resident magmas. Repeated replenishment and/or melting
of the roof or buoyant xenoliths may result in the development of a persistent
compositional stratification within chambers. Many of these processes are,
however, contentious among petrologists.
Minerals that
crystallise in magma chambers coat cooling surfaces, such as the roof and
walls, but mainly they accumulate successively on the floor, resulting in layered
series. They can subsequently be compacted, recrystallised and react to
varying degrees with percolating melts and other fluids as solidification
continues to completion, forming rocks referred to as cumulates. The
individual mineral species in cumulates generally preserve evidence of the
composition, temperature, oxygen fugacity and isotopic ratios of the magmas
from which they crystallised. Thus liquidus minerals in cumulates that formed
successively on the floor of chambers contain a more or less continuous record
of variations in melt composition, temperature etc. during the evolution of the
magma chamber.
Figure 1. Location (A), sketch map of the
Bjerkreim-Sokndal Intrusion (B), and large-scale stratigraphic subdivision
of the Bjerkreim lobe (C). Note the southern and smaller Sokndal lobe of the
intrusion in B.
The study of
layered series in solidified magma chambers can provide a detailed picture of
the processes that have taken place during their evolution, and complements the
data that can be obtained from volcanic sequences that erupted from individual
magma chambers. Such studies have contributed to an appreciation of the complex
nature of the interactions that occur in magma chambers between different
physical and chemical processes. However, despite considerable research effort
there is currently much dispute among petrologists regarding the
nature and
relative importance of the processes that take place in magma chambers. The
significance of many of the processes that have been proposed is still
uncertain and much remains to be revealed about the behaviour and evolution of
magma chambers. It is conceivable that important physical and chemical
processes still remain to be discovered and it is therefore important that
research into this topic continues to receive financial support.
The project aims to throw further light on the evolution of a specific magma chamber as revealed by the Layered Series of the Proterozoic Bjerkreim-Sokndal Intrusion, part of the Rogaland Anorthosite Province, SW Norway (Fig. 1). The Bjerkreim-Sokndal Intrusion is the largest Norwegian layered intrusion, contains a thick Layered Series (>7 km) of unusual composition (Fig. 2), and is well exposed and very accessible. It has been studied intensively over recent years (see e.g. Wilson et al. 1996, Robins et al 1997, Robins & Wilson 2001a & b, Jensen et al. 2003). The northern and largest part of the intrusion, the Bjerkreim lobe, has been systematically mapped at the scale of 1:5,000 by the applicant, his colleagues, and masters and Ph.d. students. A map compilation at the scale of 1:25,000 will shortly be published by the Geological Survey of Norway in connection with a comprehensive account of our current knowledge of this part of the intrusion (Robins et al, in prep.). Despite the volume of this earlier work, several important aspects of the evolution of the Bjerkreim-Sokndal Intrusion and magma chamber remain unresolved.
One such aspect concerns the origin of the voluminous mangerites and quartz mangerites (and minor charnockites) that form the uppermost part of the intrusion (Duchesne & Wilmart 1997). These have been variously regarded as: 1. The ultimate differentiates of the Bjerkreim-Sokndal magma chamber; 2. Forming a separate, later intrusion; 3. The result of melting of the roof of the magma-chamber; or combinations thereof. In addition, Philpotts (1981) suggested that field relationships and plagioclase compositions in massif-type anorthosite complexes are consistent with crystallisation of jotunite (equivalent to hypersthene monzonorite) and quartz mangerite from conjugate immiscible magmas, and presented preliminary experimental evidence supporting this hypothesis. In charges consisting of a mixture of jotunite and quartz mangerite he observed the co-existence of two immiscible melts, one relatively Si-poor and Fe-rich and the other Si- and alkali-rich, at temperatures 25ºC below the liquidus at atmospheric pressure and at an oxygen fugacity of the Ni-NiO buffer. Experiments on a glass prepared from the jotunite itself also demonstrated immiscible melts and Philpotts (1981) concluded that these observations strongly suggest that liquid immiscibility played a role in the formation of jotunite and quartz mangerite. However, a recent experimental study of a jotunite chill from the Bjerkreim-Sokndal Intrusion by Vander Auwera & Longi (1994) yielded no evidence of unmixing. Some of these
experiments were carried out at relevant
pressures (~5kb) but in graphite capsules, resulting in unrealistically-low
oxygen fugacities.
Figure 2. Generalised stratigraphy and cryptic layering
of the northern part of the Bjerkreim-Sokndal Layered Intrusion. Revised
after Wilson et al. (1996). Note that there is a single cumulus feldspar
that varies from plagioclase to antiperthite and finally mesoperthite as
the or component increases. The roman numerals refer to Macrocyclic units (MCU) 0, IA,
IB, II, III and IV, reflecting important episodes of magma-chamber
replenishment.
The parent magma of the Bjerkreim-Sokndal
Intrusion is inferred to have been jotunitic (i.e. hypersthene monzonorite)
(Robins et al 1997). If fractional crystallisation of this magma was
accompanied by separation of an immiscible Si- and alkali-rich melt, then two
magma layers may have developed with a differentiated jotunitic magma
underlying a smaller volume of less-dense quartz mangerite magma. On
separation, and subsequently (on the condition that thermal and chemical
equilibrium was maintained between them), these magmas could crystallise
identical minerals, but in very different relative amounts. Hence, in cumulates
forming at the base of the magma chamber the existence of two such immiscible
magmas would not be obvious. Should one of the magmas, for instance the basal
Si-poor melt, be consumed before the upper Si- and alkali-rich melt then the
latter could continue to crystallise fractionally as the temperature declined
further. This could result in a marked change in the mineral proportions of the
cumulates, while the compositions of the individual constituent minerals
changed continuously. This is the case across the Layered Series/mangerite
contact in the Bjerkreim-Sokndal Intrusion, and it is speculated that this
contact may mark the elimination of a lower layer of immiscible magma and the
continued crystallisation of the remaining Si- and alkali-rich upper magma
layer (Philpotts, pers. com.). Additional field, geochemical and experimental
data is clearly required to test the immiscibilty hypothesis and further
clarify the origin of the mangerites, quartz mangerites and charnockites in the
Bjerkreim-Sokndal Intrusion.
An additional unresolved aspect of the
Bjerkreim-Sokndal Intrusion is the origin of particularly ilmenite-rich
sequences. In the Bjerkreim Layered Series ilmenite-rich layers and sequences
are generally restricted to the regressive zones at the bases of Macrocyclic
Units III and IV that are believed to be related to magma-chamber recharge.
Cumulates elsewhere in the Layered Series contain limited amounts of ilmenite,
as expected during the cotectic crystallisation of ilmenite and plagioclase or
ilmenite, plagioclase and orthopyroxene. It is speculated that the unusually
high concentration of ilmenite in the regressive sequences is related to mixing
of compositionally-different, but ilmenite-saturated, magmas during
magma-chamber recharge. The most compelling evidence for the hybridisation of
inflowing magma and resident, evolved magma during episodes of magma-chamber
recharge is the variation of initial 87Sr/86Sr ratios
across the lower contact of MCU IV in the Bjerkreim lobe of the Bjerkreim-Sokndal
Intrusion. A section across this contact has been described in detail by Nielsen
and Wilson (1991), Jensen et al. (1993) and Nielsen et al. (1996), but the
ilmenite- and pyroxene-rich layers present in the sequence have hitherto
received scant attention and similar investigations in equivalent parts of the
Layered Series elsewhere in the intrusion are lacking. The regressive base of
MCU III has some unique stratigraphic features. The MCU II/III boundary is
characterised by a sulphide-enriched subzone associated with either a layer of
ilmenite orthopyroxenite, unique in the intrusion, or ilmenite-rich
melanorites. This is succeeded by a regression in mineral compositions that
generally culminates in troctolitic cumulates. The stratigraphic relations
appear to be consistent with prolonged magma-chamber recharge associated with
progressive mixing of inflowing jotunitic magma and a resident, stratified
magma. The sulphide-enriched orthopyroxenite and related melanocratic ilmenite
norite represent the initial response to the replenishment event and have been
explained by crystallisation of hybrid magma residing within the orthopyroxene
phase volume. This is considered to be a consequence of mixing of resident
magma saturated with plagioclase, orthopyroxene and ilmenite and inflowing
magma with plagioclase and olivine on the liquidus and the curvature of the
natural plagioclase-orthopyroxene-ilmenite cotectic (Jensen et al. 2003). The
stratigraphic study of the MCU II/III boundary by Jensen et al. (2003) focussed
on the origin of the sulphide-enrichment present in the orthopyroxenite and
melanorites and neither included data on the composition of ilmenite nor
discussed specifically the origin of the modal enrichment in ilmenite.
Additional detailed sampling and analytical investigations of the ilmenite-rich
rocks, and in particular the major- and trace-element composition of the
ilmenite itself, are required to clarify the relationship between their
enrichment in ilmenite and mixing of magmas during replenishment of the magma
chamber. The results of such a study can be significant for the interpretation
of ilmenite ores elsewhere.
Detailed mapping and sampling of the Bjerkreim lobe of the
Bjerkreim-Sokndal Intrusion is almost completed. However, the smaller Sokndal
lobe of the intrusion has hitherto only been the subject of reconnaissance-type
studies. The types and distribution of the cumulates have neither been
identified and mapped in the same detail, nor have an adequate number of
samples been collected and analysed. As a result, phase contacts cannot be
drawn with any confidence on a geological map and the cryptic variation is
imperfectly documented. The studies already carried out suggest that the main
features of the cumulates present in the Sokndal lobe are similar to those of
the uppermost part of the Layered Series in the Bjerkreim lobe (i.e.
Macrocyclic Unit IV). The cumulus minerals appear in the same sequence and the
minerals show generally similar compositional variations. Equivalents to the
plagioclase-rich lowermost cumulates in MCU IV in the Bjerkreim lobe appear to
be present only locally in the Sokndal lobe and the most primitive, troctolitic
cumulates in the latter differ from those in the Bjerkreim lobe in the presence
of apatite. It is not known for certain whether any older cumulates are
represented in the Sokndal lobe. The development of a comprehensive model for
the evolution of the Bjerkreim-Sokndal magma chamber as a whole will require
time-consuming mapping and sampling of the Sokndal lobe in the same detail as
in the Bjerkreim lobe.
Resolution of the issues outlined above and
further progress in our understanding of the evolution of the Bjerkreim-Sokndal
magma chamber will require additional detailed mapping and sampling, mineral
analyses, isotopic analyses and experimental investigations. During the course
of the project the following specific topics will be investigated:
q The
structure, mineralogy and geochemistry of the uppermost units in the intrusion
(mangerite, quartz mangerite and charnockite), particularly in the northernmost
part of the intrusion, and their relationship with underlying Layered Series.
An evaluation of the possible role played by liquid immiscibility in the
genesis of the mangerite, quartz mangerite and charnockite will be included.
The topic will involve mapping and systematic sampling of the uppermost part of
the intrusion, whole-rock major- and trace-element chemical analyses,
microprobe analysis of mineral compositions, and mass-spectrometric
determination of initial Sr87/86 and Nd143/144 ratios in
selected samples. In addition, a series of high-pressure crystallisation
experiments will be carried out to investigate the feasibility of liquid
immiscibility in jotunitic magmas with relevant compositions, and its possible
role in the origin of the mangerite, quartz mangerite and charnockite. In contrast
to earlier attempts, the experiments will be carried out at more realistic
oxygen fugacities.
q The
Layered Series in the southern, Sokndal lobe of the Bjerkreim-Sokndal Intrusion
and its relationship with that developed in the northern, Bjerkreim lobe of the
intrusion. This topic will require detailed mapping and sampling to supplement
earlier, preliminary work in the Sokndal area, whole-rock major- and
trace-element chemical analyses, electron microprobe analysis of mineral
compositions and mass-spectrometric determination of initial Sr87/86
and Nd143/144 ratios in selected samples. This topic will involve a
comprehensive study of the petrology of an important part of the
Bjerkreim-Sokndal Intrusion. It is regarded as suitable for and dependent on
granting of a Ph.D. stipend. The topic will also include a more detailed study
of the origin of exceptionally Fe-Ti oxide-rich zones and layers in the Layered
Series of the Sokndal lobe, some of which were formerly mined as iron ore, as
well as the delineation and characterisation of potential
magnetite-ilmenite-apatite ores in the area.
q The
origin of ilmenite-rich layers and sequences in regressive intervals in the
Layered Series in the Bjerkreim lobe of the intrusion in relation to magma
inflow and mixing in the magma chamber. This topic will require detailed
logging and sampling of ultramafic and adjacent layers at several localities in
specific, previously recognised parts of the Bjerkreim Layered Series, electron
microprobe and laser-ablation ICPMS analysis of the constituent minerals,
especially ilmenite, and mass-spectrometric determination of initial Sr87/86
and Nd143/144 ratios in selected samples and mineral separates.
Stratigraphic variations in the trace-element composition of liquidus ilmenite
are expected to be particularly informative as a monitor of changes in magma
composition, temperature and oxygen fugacity as a function of magma mixing
during chamber replenishment.
The request for a Ph.D. stipend is based primarily on the comprehensive aims of the proposed project. In addition, the Sokndal lobe of the Bjerkreim-Sokndal Intrusion is regarded as a suitable training for a Ph.D. stipend in field-based petrology. It is self contained, sufficiently challenging and has the potential for an important contribution to our understanding of the evolution of magma chambers.
The number of postgraduates in igneous petrology in Norway and the other Nordic countries has decreased quite dramatically over recent years and is now extremely limited. This is disturbing in view of the need for a new generation of university staff in the near future. At the University of Bergen alone, 3 professors in petrology will retire during the next decade. The availability of a Ph.D. stipend could stimulate interest in igneous petrology among prospective students and contribute to the recruitment of future university scholars in a field in which there will be a future need for recruitment.
Dr. J. R.
Wilson, Department of Earth Sciences, Aarhus University, Denmark (Mapping and
sampling, mineralogy, petrology).
Professor
Sven Maaløe, Department of Earth Science, University of Bergen, Norway
(Experimental petrology).
Professor
J. C. Duchesne, Laboratory for Geology, Petrology and Geochemistry, University
of Liége, Belgium (Geochemistry).
Dr. G.
Meyer, The Geological Survey of Norway, Trondheim (Fe-Ti oxide chemistry).
Dr. H.
Schiellerup, The Geological Survey of Norway, Trondheim (Field relations).
Siv.
ing. A. Kornelliussen, The Geological Survey of Norway, Trondheim (Fe-Ti-P
resources)
Dr.
Peter Robinson, The Geological Survey of Norway, Trondheim (Ilmenite
concentrations in relation to magma mixing).
Budget
|
Specification |
2005 |
2006 |
2007 |
Total |
|
Salary-related costs: |
|
|
|
|
|
Ph.D. stipend |
555000 |
555000 |
555000 |
1665000 |
|
|
|
|
|
|
|
Running costs: |
|
|
|
|
|
Field work1 |
60000 |
60000 |
30000 |
150000 |
|
Thin & polished sections2 |
10000 |
10000 |
8000 |
28000 |
|
Major- and trace-element chemical analyses3 |
5000 |
15000 |
10000 |
30000 |
|
Electron microprobe analyses4 |
10000 |
20000 |
20000 |
50000 |
|
Isotopic analyses5 |
14000 |
25000 |
15000 |
54000 |
|
Laser-ablation ICPMS analyses6 |
0 |
20000 |
25000 |
45000 |
|
Experimental investigations7 |
70000 |
0 |
0 |
70000 |
|
Diverse8 |
5000 |
15000 |
10000 |
30000 |
|
Meetings with collaborators in Trondheim and Aarhus |
10000 |
15000 |
15000 |
40000 |
|
Presentation of results at national and international conferences |
0 |
15000 |
20000 |
35000 |
|
Total running costs |
184000 |
195000 |
153000 |
532000 |
|
|
|
|
|
|
|
Total |
739000 |
750000 |
708000 |
2197000 |
All
amounts in NOK
1.
2
investigators (project leader and Ph.D. student), 28 days each in 2005 and
2006, 14 days each in 2007.
2.
300
thin sections at NOK 60,-, 100 polished sections at NOK 100,-, Dept of Earth
Science, University of Bergen
3.
150
samples at NOK 200,-, Dept
of Earth Science, University of Bergen
4.
500 hours at NOK 100,- per hour, Dept of Earth Science,
University of Bergen
5.
60 samples at NOK 500,- per sample
& 60 samples at NOK 400,- per sample, Dept of Earth Science, University of
Bergen
6.
100 hours at NOK 450,- per hour, Dept of Earth Science,
University of Bergen
7.
Cost
of high-pressure crystallisation experiments.
8.
Field
equipment, copying, post & freight, film, topographic maps, data
consumables
Duchesne, J.-C.
& Wilmart, E. 1997: Igneous Charnockites and
related rocks from the Bjerkreim-Sokndal Layered Intrusion (southwest Norway): a
jotunite (hypersthene monzodiorite)-derived A-type granitoid suite. Journal of Petrology 38, 337-369.
Jensen, J. C., Nielsen, F. M., Duchesne, J.-C., Demaiffe, D. &
Wilson, J. R. 1993: Magma influx and mixing in the
Bjerkreim‑Sokndal layered intrusion, South Norway: evidence from the
boundary between two macrocyclic units at Storeknuten. Lithos 29, 311-325.
Jensen, K. K., Wilson, J. R., Robins, B.
& Chiodoni, F. 2003: A
sulphide-bearing
orthopyroxenite layer in the Bjerkreim-Sokndal Intrusion, Norway: Implications
for processes during magma-chamber replenishment. Lithos 67, 15-37.
Nielsen, F. M., Campbell, I. H., McCulloch, M. & Wilson, J. M.
1996: A
Strontium isotopic investigation of the Bjerkreim‑Sokndal Layered
Intrusion, southwest Norway. Jour. Petrology
37, 171-193.
Nielsen, F. M. & Wilson, J. R. 1991: Crystallization processes in
the Bjerkreim‑Sokndal layered intrusion, south Norway: evidence from the
boundary between two macrcyclic units. Contrib.
Mineral. Petrol. 107, 403‑414.
Philpotts, A. R. 1981: A model for the generation of massif-type
anorthosites. Canadian Mineralogist 19,
233-253.
Robins,
B. &
Wilson, J. R. 2001: The Bjerkreim-Sokndal Layered Intrusion. In
Duchesne, J.-C. The Rogaland Intrusive Massifs: An excursion Guide. NGU report 2001.29.
Geological Survey of Norway, pp.35-47.
Robins, B., Tumyr,
O., Tysseland, M. & Garmann, L.B. 1997: The Bjerkreim-Sokndal
Layered Intrusion, Rogaland, SW Norway: Evidence from marginal rocks for a
jotunite parent magma. Lithos 39, 121-133.
Robins,
B., Wilson,
J.R., Nielsen, F.M. & Schiellerup, H. In prep.: The Bjerkreim-Sokndal Layered Intrusion, Southwest
Norway: Field relations and petrography.
Vander Auwera,
J. & Longhi, J. 1994: Experimental study of a
jotunite (hypersthene monzodiorite): constraints on the parent magma composition
and crystallisation conditions (P, T, fO2) of the Bjerkreim-Sokndal
layered intrusion (Norway). Contrib.
Mineral. Petrol. 118, 60-78.
Wilson, J. R., Robins, B., Nielsen, F. M., Duchesne, J.-C. & Vander Auwera, J. 1996: The Berkreim-Sokndal Layered Intrusion, southwest Norway. In Cawthorn, R. G. (ed.): Layering in Igneous Rocks. Elsevier. pp. 231-255.
Bergen, 24th of
May, 2004