Kyanite-grade metamorphism in the and Bogen Groups, , North

MARK G. STELTENPOHL & JOHN M. BARTLEY

Steltenpohl, M. G. & Bartley, J. M. : Kyanite-grade metamorphism in the Evenes and Bogen Groups, Ofoten, North Norway. Norsk Geologisk Tidsskrift, Vol. 64, pp. 21-26. 1984. ISSN 0029-196X. Directly north of Ofotfjorden in , pelitic schists within the Evenes and Bogen Groups contain the mineral assemblage + biotite ± kyanite ± staurolite + white mica + quartz ± plagio­ clase. This assemblage implies metamorphic P-T minima of- 540"C and - 4. 8 kb. The rocks are thus at a higher grade than suggested by previous reports, which placed them in the greenschist facies. This indicates that several metamorphic allochthons in Ofoten, including rocks of the , Evenes, Bogen, and Niingen Groups, are all at kyanite grade, supporting recent interpretations which on structural grounds concluded that the metamorphic peak outlasted stacking of these allochthons. A proposed correlation of the Evenes Group with the Middle Ordovician-Lower Silurian Supergroup implies that this stacking and associated kyanite-grade metamorphism are post-early Silurian and are related to the Scandian phase of the Caledonian orogeny.

M. G. Steltenpohl & J. M. Bartley, Department of Geology, University of North Carolina, Chapel Hill, North Carolina 27514, USA.

Our mapping, structural analysis, and petro­ ultramafic rocks of the Narvik Group, which are graphic studies in Ofoten have concentrated on presently at kyanite grade (Foslie 1941, 1949, the structural and metamorphic development of Gustavson 1966, 1972, Hodges 1982a, Tull et al. the Caledonian nappe stack. Figure l shows a in press). The thrust beneath the Narvik Group preliminary interpretation of the tectonostrati­ (Gustavson 1966, 1974, Hodges 1982a) cuts graphic sequence in the Skånland area. Details structurally upward passing northward so that of the lithologic sequence and petrography will the entire Narvik group and the Elvenes con­ be presented in a later paper (Steltenpohl in glomerate are missing north of . In this prep.). In this note we present field and petro­ northern area, the thrust places Evenes Group graphic observations regarding metamorphism of marbles upon 100-200 m of mica schist and roar­ the Evenes and Bogen Groups, which may have ble, which we will call the 'basal' allochthon. The major implications for the regional geology. 'basal' allochthon directly overlies Precambrian granitic basement. Gustavson (1974) correlated the 'basal' allochthon with either the Narvik or Rombak Groups; we doubt that these rocks cor­ Tectonostratigraphic context relate with the Narvik Group, but are uncertain The Evenes Group and overlying Bogen Group of their affinities pending more detailed study. contain a series of marbles and mica schists, with The contact between the Evenes and Bogen subordinate quartzites and metaconglomerates Groups is characterized by: l) a lithologic break, (Strand 1960). Gustavson (1966) included both from the marble-dominated Evenes Group which the Evenes and Bogen Groups in his contains rare and_minor mafic intrusions, to the Group. However, two distinct lithologic assem­ calc-schist dominated Bogen Group which was blages, which correspond well to the Evenes and intruded by numerous concordant granite sheets Bogen Groups of Strand (1960), are clearly pre­ prior to most of the penetrative deformation; sent within the Salangen Group in the Skånland and 2) a zone of concentrated strain in which area. We thus use the earlier nomenclature here. units in the underlying Evenes Group are trans­ South of Ofotfjorden (Fig. 1), the base of the posed and locally truncated. Evenes Group is marked by the Elvenes Con­ The contact could be either a thrust or a modi­ glomerate, which was deposited unconformably fied unconformity. A conglomerate present at or upon the schists, calc-schists, amphibolites, and near the structural top of the Evenes Group 22 M. G. Steltenpohl & J. M. Bartley NORSK GE OLOGISK TIDSSKRIFT l (1984)

L..of­ lslands

• kyanlfe locality O staurollte loctJiity

locallty ,. ky8staur lll:�:::sc:���� � LONER ALLOCHTHON(dotted pottem) NARVIK GROUP(Iined pattom) BASEMENT o-.._....;5:..-....-'o;;..._,.:15kmo

Figure l. Tectonic map and simplified tectonostratigraphic column of the Skånland area, showing kyanite and staurolite localities. Tectonic units modified from Gustavson (1974).

suggests that the contact may be an inverted view that the Evenes and Bogen Groups are unconformity. The absence of granitic intrusions separate nappes which were thrust together. in the structurally lower Evenes Group could be The structurally highest nappe preserved in interpreted to indicate that it was deposited upon Ofoten comprises rocks of the Niingen Group, a the Bogen Group after intrusion of the granite scarcely studied lithologic assemblage broadly sheets, and then the sequence was tectonically similar to the Narvik Group (Gustavson 1966). inverted. This would imply that the conglomer­ The contact between the Bogen Group and over­ ate at the top of the Evenes Group is equivalent lying Niingen Group is a low-dipping zone of to the Elvenes conglomerate at its base, and that intense foliation which separates contrasting a major structural repetition occurs within the suites of metasedimentary rocks. lntrusions in Evenes Group. The lithologies of these two con­ the Niingen Group are generally pegmatitic, glomerates are dissimilar, and we see no evi­ while those in the Bogen Group are fine grained dence for such a repetition. Further, interpreta­ (Gustavson 1966, Steltenpohl in prep.). We rec­ tion of the contact as an unconformity implies ognize no retrograde metamorphism along this that the Bogen Group is structurally equivalent contact; mineral assemblages in the highly to the Narvik Group, which it does not lithologi­ strained rocks at the contact are the same as cally resemble. Combined with the presence of those away from the thrust. No evidence of trun­ concentrated strain and local truncations of rock cation or transposition of the principal metamor­ units at the contact, this leads us to favour the phic fabrics across this or the other probable NORSK GEOLOGISK TIDSSKRIFT l (1984) Kyanite-grade metamorphism in North Norway 23 nappe boundaries has been found in the Skån­ sion is strongest in rocks affected by late fold­ land area. If this contact reflects major tectonic phases which post-date the metamorphic peak juxtaposition, it would appear to predate the (F2, F3 of Gustavson 1972, F3, F4 of Bartley metamorphic peak. 1981, in press, and Steltenpohl 1983, in prep.). Retrograde metamorphism does not appear to be restricted to or concentrated at any particular Prograde metamorphism structural leve!. In some retrograded samples, greenschist-facies mineral assemblages replace Nine occurrences of kyanite and/or staurolite amphibolite-facies minerals along late-stage have been located in the Evenes and Bogen spaced cleavage planes (Fig. 2b ). The typical Groups (see Fig. l for geographic and tectono­ retrograde mineral assemblage in pelites is stratigraphic locations). The mineral assem­ blages are (Fig. 2a): chlorite + biotite ± white mica ± epidote ± quartz. l) Kyanite+ garnet+ biotite 2) Staurolite+ garnet+ biotite Other samples contain these retrograde phases 3) Kyanite+ staurolite+ garnet+ biotite without development of a late-stage cleavage. It is unclear whether garnet was stable during The rocks also contain quartz, white mica, and the retrograde metamorphism. In many samples, plagioclase. In the idealized pelitic system garnet is partly replaced by chlorite; however, (Thompson 1957), assemblages l and 2 are divar­ this may only reflect a shift in the stable garnet iant, while assemblage 3 is univariant. It is likely composition during cooling. Some show that the actual variances are higher due to signifi­ inclusion-free overgrowths over 'snowball' cores cant amounts of Ca O and MnO, although assem­ (Gustavson 1966, Juve 1967, Steltenpohl 1983). blage 3 alternatively may indicate internal buffer­ In similar garnets from the Narvik Group, elec­ ing of fH o (Rumble 1978, Hodges & Spear 2 tron microprobe analyses suggest that rims grew 1982). The absence of staurolite in assemblage l during cooling and may relate to retrogression implies metamorphic P and T above the intersec­ (Hodges 1982b). tion of the quasi-univariant reactions: staurolite+ muscovite+ quartz

= kyanite+ garnet+biotite + H20 (l) kyanite = sillimanite (2) Discussion This intersection was placed by Carmichael The metamorphic map of southert;� and (1978) at 540°C and 4.8 kb, based on experimen­ Ofoten by Gustavson (1966) shows the b),Ilk of tal calibrations and field observations. If we as­ the Salangen Group (i.e., the Evenes and Bogen surne that additional components have stabilized Groups) to be within the garnet zone of the staurolite in assemblages 2 and 3, then assem­ greenschist facies. Part of the Salangen Group on blage l is probably the most reliable indicator of Rolla (see Fig. l for location) is shown to be grade because it more closely approaches the within the staurolite and kyanite zones (amphi­ simplified systems used in experimental calibra­ bolite facies), as is the overlying Niingen Group. tion. In this case, Carmichael's (1978) P-T values Gustavson (1966) shows a 'thrust plane (?)' for the intersection of the univariant equilibria which places kyanite-grade Niingen Group rocks are minima for the metamorphic conditions of upon garnet-grade Salangen Group rocks, which the rocks. The presence of staurolite in some implies a metamorphic inversion at the contact. assemblages probably indicates that this tem­ Gustavson (1966, p. 131) noted that the absence perature is not greatly exceeded, but pressures as of kyanite in the Salangen Group might reflect high as 9-10 kb are not excluded. Petrological high CaO content rather than a difference in studies in progress will better quantify the meta­ metamorphic grade, but he placed the Salangen morphic conditions. Group rocks in the garnet zone for the purposes of the map. The present work extends the kyan­ Retrograde metamorphism ite/staurolite zones from Rolla southward throughout the Evenes and Bogen group rocks of Retrograde metamorphism is common but errati­ the western limb of the Ofoten synform (Fig. 1). cally developed in the Skånland area. Retrogres- There appears to be no difference in metamor- 24 M. G. Steltenpohl & J. M. Bartley NORSK GEOLOGISK TIDSSKR!Ff l (1984) NORSK GEOLOGISK TIDSSKRIFT 1 (1984) Kyanite-grade metamorphism in North Norway 25 phic grade between the Niingen Group and un­ lain by the greenschist facies Balsfjord Super­ derlying rocks. group (Minsaas 1981, Minsaas & Sturt 1981, All of the non-retrograde mineral assemblages Binns & Matthews 1981), which contains Middle we have observed in the Evenes and Bogen Ordovician to Lower Silurian fossils. Parts of the Groups are consistent with kyanite-grade meta­ Balsfjord Supergroup can be directly correlated morphism. The usual pelitic assemblage is southward with the Lower Silurian Sagelvvatn garnet + biotite + white mica + quartz ± plagio­ Group (Bjørlykke & Olaussen 1981). Boyd clase, which is a stable assemblage at metamor­ (1983) recently recognized low-grade metasedi­ phic grades from the garnet zone up through the mentary rocks at (see Fig. l for loca­ sillimanite zone. Recent studies in western Ofo­ tion), which are generally similar to the Sa­ ten have failed to find field or petrographic evi­ gelvvatn Group and may be a southward conti­ dence for post-metamorphic structural breaks nuation into Ofoten of the low-grade Silurian between the high-grade cover units. The struc­ rocks. tural histories of the nappes from the metamor­ Detailed study of the lithostratigraphy of the phic peak onward are indistinguishable (Bartley Evenes Group (Steltenpohl & Tull in prep.) sup­ 1981, in press, Tull et al. in press, Hodges, ports the correlation proposed by Binns (1978) of 1982a, in prep., Steltenpohl in prep.). Prior to the Evenes Group with Silurian rocks of the the metamorphic peak, different cover units may Balsfjord Supergroup. The correlation cannot be have bad different histories. However, later de­ proven until the area between Sagelvvatn and formation and kyanite-grade metamorphism Ofoten has been mapped in more detail. Howev­ have made evidence for earl y deformation sparse er, if this correlation is correct, then the stacking and ambiguous. and metamorphism of the kyanite-grade allochth­ We thus believe that the Narvik, Evenes, Bo­ ons in Ofoten must be post-early Silurian, rather gen, and Niingen Groups were metamorphosed than Finnmarkian. This requires major post-ear­ together at kyanite grade, during or directly fol­ ly Silurian transport of this composite allochthon lowing any tectonic stacking. The new kyanite after the metamorphic peak, in order to juxta­ localities strongly support this interpretation. We pose the composite kyanite-grade allochthon support the suggestion made by Gustavson with low-grade Silurian rocks. Pre-Silurian, per­ (1966) that the common absence of kyanite and haps Finnmarkian, deformation may have affect­ staurolite in the Evenes and Bogen Groups re­ ed the Narvik Group, which contains dismem­ flects bulk composition rather than metamorphic bered ophiolites (Hodges 1982a, in prep., Boyd grade. 1983, P. Crowley, pers. commun.) possibly A further implication of this interpretation re­ lated to the ophiolite. However, it lolP­ gards the timing of kyanite-grade regional meta­ pears that in Ofoten, Finnmarkian structures, if morphism of the allochthons. The Silurian final present, have been virtually obliterated by post­ phase of the Scandinavian Caledonide orogen early Silurian events. has been termed the Scandian phase by Gee Acknow/edgements. - This note benefited from the comments (1975). However, the metamorphic peak in the of J. Robert Butler. We acknowledge support from U.S. Na­ high-grade Caledonian nappes of oc­ tional Science Foundation grant no. EAR-8107525 to JMB, curred during .the Cambro-Ordovician Finnmar­ and field support to MGS from Norges geologiske under­ kian phase of the Caledonian orogeny (Sturt et søkelse and the Martin-McCarthy trust funds administered by the Department of Geology at the University of North Caroli­ al. 1978). The Lyngen ophiolite nappe, located na. approximately 100 km north of Ofoten, was em­ Manuscript received August 1983, revised January 1984 placed and metamorphosed during the Finnmar­ kian. The Lyngen nappe is unconformably over-

Figure 2. Photomicrographs of Evenes Group schists. a. Kyanite (ky) + staurolite (st) +garnet (ga)+ biotite (bi)+ muscovite (mu) in a pelitic schist of the Evenes Group. Lang dimension of photo is 5 mm. b. Retrograde features in Evenes Group pelitic schist. The prograde assemblage is garnet (ga)+biotite (bit)+muscovite (mut). The retrograde assemblage is chlorite (el)+biotite (bi2) +muscovite (mu2). Micas mut and bit define the prominent lepidoblastic foliation which is slightly inclined to the right of the photo. A late-stage spaced cleavage, steeply inclined in the photo, crenulates the earlier foliation and truncates the right-hand side of the garnet porphyroblast, with resulting formation of chlorite. Lang dimension of the photo is 5 mm. 26 M. G. Steltenpohl & J. M. Bartley NORSK GEOLOGISK TIDSSKRIFT l (1984)

References Hodges, K. V. 1982a: Tectonic evolution of the Aefjord-Sitas area, Norway-. Unpubl. Ph. D. dissertation, Massa­ Bartley, J. M. 1981: Structural geo/ogy, metamorphism, and chusetts Institute of Technology, Cambridge, USA, 192 pp. Rb/Sr geochronology of east Hinnøy, north Norway. Unpubl. Hodges, K. V. 1982b: The use of geothermometry and geobar­ Ph. D. dissertation, Massachusetts Institute of Technology, ometry to constrain the uplift history of a portion of the Cambridge, USA, 263 pp. Scandinavian Caledonides. Geo/. Soc. Am. Abstracts with Bartley, J. M. in press: Caledonian structural geology and Programs 14, 516. tectonics, east Hinnøy, north Norway. Nor. geo/. unders. Hodges, K. V. & Spear, F. S. 1982: Geothermometry, geobar­ Binns, R. E. 1978: Caledonian nappe correlation and orogenic ometry and the Al2Si05 triple point at Mt. Moosilauke, New history in Scandinavia north of lat 67°N. Geo/. Soc. Am. Hampshire. Am. Mineral. 67, 1118-1134. Bull. 89, 1475-1490. Juve, G. 1967: Zinc and lead deposits in the Håfjell syncline, Binns, R. E. & Matthews, D. W. 1981: Stratigraphy and Ofoten, northern Norway. Nor. geo/. unders. 244, 54 pp. structure of the Ordovician-Silurian Balsfjord Supergroup, Minsaas, O. 1981: Lyngenhalvøyas geologi, med spesiell vekt Troms, North Norway. Nor. geo/. unders. 365, 39-54. på den sedimentologiske utvikling av de Ordovisisk-Siluriske Bjørlykke, A. & Olaussen, S. 1981: Silurian sediments, volcan­ klastiske sekvenser som overligger Lyngen gabbro kompleks. ics, and mineral deposits in the Sagelvvatn area, Troms, Unpubl. cand. real. thesis. University of , 295 pp. North Norway: Nor. geo/. unders. 365, 1-38. Minsaas, O. & Sturt, B. A. 1981: The Ordovician clastic Boyd, R. 1983: The Lillevik dike complex, Narvik: geo­ sequence immediately overlying the Lyngen gabbro complex chemistry and tectonic implications of a probable ophiolite and its environmental significance. Terra Cognita l (Uppsala fragment in the Caledonides of the Ofoten region, N. Nor­ Caledonide Symposium abstracts), 59-60. way: Nor. Geo/. Tidsskr. 63, 39 -54. Rumble, D. 1111978: Mineralogy, petrology, and oxygen isoto­ Carmichael, D. 1978: Metamorphic bathozones and batho­ pic geochemistry of the Clough formation, Black Mountain, grads: a measure of the depth of post-metamorphic uplift western New Hampshire, U.S.A. J. Petro/. 19, 317-340. and erosion on a regional scale. Am. J. Sei. 278, 769-797. Steltenpohl, M. G. 1983: The structure and stratigraphy of the Foslie, S. 1941: Tysfjords geologi. Nor. geo/. unders. 149, 298 Ofoten synform, North Norway. Unpubl. M. S. thesis, Uni­ PP· versity of Alabama, Tuscaloosa, USA, 108 pp. Foslie, S. 1949: Håfjellsmulden i Ofoten og dens sedimentære Strand, T. 1960: Cambro-Silurian deposits outside the Oslo jern-mangan-malmer. Nor. geo/. unders. 174, 129 pp. region, Geology of Norway. Nor. geo/. unders. 208, 151-169. Gee, D. G. 1975: A tectonic for the central part of the Scandi­ Sturt, B. A., Pringle, l. R. & Ramsay, D. M. 1978: The navian Caledonides. Am. J. Sei. 275-A, 46&-515. Finnmarkian phase of the Caledonian orogeny. J. geo/. Soc. Gustavson, M. 1966: The Caledonian mountain chain of the London 135, 597-610. southern Troms and Ofoten areas. Part l. Basement rocks Thompson, J. B., Jr. 1957: The graphical analysis of mineral and Caledonian metasediments. Nor. geo/. unders. 239, 162 assemblages in pelitic schists. Am. Mineral. 42, 842-858. PP· Tull, J. F., Bartley, J. M., Hodges, K. V., Andresen, A., Gustavson, M. 1972: The Caledonjan mountain chain of the Steltenpohl, M. G., & White, J. M. in press: The Caledon­ southem Troms and Ofoten areas. Part Ill. Structures and ides in the Ofoten region (�-6go), North Norway: key structural history. Nor. geo/. unders. 283, 56 pp. aspects of tectonic evolution. In Gee, D. G., and Sturt, B. Gustavson, M. 1974: Bergrunnskart "Narvik", 1:250000. Nor. A., (eds.), The Ca/edonide Drogen- Scandinavia and Rela­ geo/. unders., Trondheim. ted Areas. New York, Wiley-lnterscience Publishers.