Sm–Nd and U–Pb Isotope Geochemistry of the Palaeoproterozoic Mafic Magmatism in Eastern and Northern Finland

Home , Eastern Lapland, Juuka, Timanide Orogen

2018 Monograph • Geological Survey of Finland Survey Geological Bulletin 405 in eastern and northern Finland and northern in eastern and Yann Lahaye Jouni Vuollo, Irmeli Mänttäri Eero Hanski, Asko Kontinen, Hannu Huhma, of the Palaeoproterozoic mafic magmatism magmatism mafic Palaeoproterozoic of the Sm–Nd and U–Pb isotope geochemistry geochemistry isotope and U–Pb Sm–Nd Geological Survey of Finland Survey Geological

Geological Survey of Finland Geological Survey of Finland, Bulletin

The Bulletin of the Geological Survey of Finland publishes the results of scientific research that is thematically or geographically connected to Finnish or Fennoscandian geology, or otherwise related to research and innovation at GTK. Articles by researchers outside GTK are also welcome. All manuscripts are peer reviewed.

Editorial Board Prof. Pekka Nurmi, GTK, Chair Dr Stefan Bergman, SGU Dr Asko Käpyaho, GTK Dr Antti Ojala, GTK Dr Timo Tarvainen, GTK, Scientific Editor

Instructions for authors available from the Scientific Editor. GEOLOGICAL SURVEY OF FINLAND

Bulletin 405

Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Unless otherwise indicated, the figures have been prepared by the authors of the publication.

Layout: Elvi Turtiainen Oy

Espoo 2018 Huhma, H.1) , Hanski, E.2), Kontinen, A.3), Vuollo, J.4), Mänttäri, I.1) & Lahaye, Y.1) 2018. Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland. Geological Survey of Finland, Bulletin 405, 150 pages, 128 figures, 1 table and 11 appendices.

The extensive isotopic studies performed at the Geological Survey of Finland (GTK) since the early 1970s have shown that mafic magmas in the Karelia province of the Fennoscandian Shield were emplaced in several stages, including ca. 2.5 Ga, 2.44 Ga, 2.3 Ga, 2.22 Ga, 2.15 Ga, 2.12 Ga, 2.05 Ga, 2.0 Ga, 1.88 Ga and 1.78 Ga. Most of the rock associations formed during these events are related to episodes of shield-wide extension and rifting of the Archaean lithosphere and may be regarded as examples of ancient large igneous provinces. The Sm–Nd whole-rock and mineral data produced by GTK on Palaeoproterozoic mafic rocks in the Karelia province consist of ca. 800 analyses from ca. 100 rock units. The Sm–Nd mineral ages from well-preserved samples are mostly consistent with the available U–Pb zircon ages and provide reliable estimates for the initial isotope composition of the rocks in question. These data, together with geochemical and other geological information, are used to constrain the age and origin of mafic magmas and the evolution of the lithosphere.

The initial εNd values in the studied mafic rocks range from very positive to strongly negative and suggest that some of them were derived from a depleted mantle source, whereas others record a large contribution from old enriched lithosphere. Long-term mantle heterogeneity is evident from the isotopic data on high-REE

mantle-derived rocks. Nearly chondritic initial εNd values were obtained from the 2.6 Ga Siilinjärvi carbonatite, the 2.0 Ga Jormua OIB and 1.8 Ga lamprophyres, whereas the 2.0 Ga Laivajoki and Kortejärvi carbonatites have yielded clearly posi-

tive initial εNd values of +2.5. Further evidence for a depleted mantle source (by

εNd = +4) is provided, for example, by the 2.0–2.1 Ga komatiites and some basalts

in Lapland and 2.3 Ga dykes in eastern Finland. The positive εNd(T), particularly at 2.0–2.1 Ga, may indicate major attenuation of the lithosphere, which eventually allowed material from convective mantle to ascend uncontaminated to the surface.

Deep-crustal contamination of ultramafic magma may explain many features of the studied mafic-ultramafic rocks, such as the 2.44 Ga layered intrusions with

an εNd value of –2. Occasionally, contamination of country rock material at the final emplacement site of an intrusion may have been important, for example in

the case of the 2.06 Ga Kevitsa mafic intrusion, showing an initialε Nd value from –3.4 to –6.4.

The age and initial Nd isotope composition, together with other relevant information, provide tools for correlating dykes, intrusions and mafic extrusive units within the Fennoscandian Shield. The data are useful in constraining the ages of the Karelian lithostratigraphic units and their correlation. The results can also be used in correlating events between different cratons, particularly across the Atlantic to the Canadian Shield.

Electronic appendices are available at http://tupa.gtk.fi/julkaisu/liiteaineisto/bt_405_ appendix_1_11.xlsx Electronic table is available at http://tupa.gtk.fi/julkaisu/liiteaineisto/bt_405_table_1. xlsx

Keywords: Karelia Province, mafic magmas, gabbros, dykes, isotopes, absolute age, U/Pb, Sm/Nd, Proterozoic, Finland, Fennoscandian Shield

1) Geological Survey of Finland, P.O. Box 96, FI-02151 Espoo, Finland 2) Oulu Mining School, P.O. Box 3000, FI-90014 University of Oulu, Finland 3) Geological Survey of Finland, P.O. Box 1237, FI-70211 Kuopio, Finland 4) Geological Survey of Finland, P.O. Box 77, FI-96101 Rovaniemi, Finland

E-mail: [email protected]

ISBN 978-952-217-394-2 (PDF) ISSN 0367-522X (print) ISSN 2489-639X (online)

2 CONTENTS

DEDICATION...... 6

PREFACE ...... 7

1 INTRODUCTION...... 8

2 ANALYTICAL METHODS...... 8 2.1 Nd isotope analysis...... 8 2.2 U–Pb isotope analysis...... 10

3 LAPLAND...... 12 3.1 Geological background...... 12 3.2 The 2.4–2.5 Ga intrusions in Lapland...... 16 3.2.1 Tshokkoaivi intrusion ...... 16 3.2.2 Koitelainen intrusion and associated felsic volcanic rocks...... 17 3.2.3 Peuratunturi and Koulumaoiva intrusions...... 19 3.2.4 Lehtomaa intrusion...... 21 3.2.5 Onkamonlehto dyke ...... 22 3.3 The 2.22 Ga Palovaara intrusion...... 24 3.4 The 2.15 Ga intrusions...... 25 3.4.1 Rantavaara intrusion...... 25 3.4.2 Tanhua intrusions...... 26 3.5 The 2.05 Ga intrusions...... 29 3.5.1 The Kevitsa intrusion...... 29 3.5.2 Kevitsa dykes...... 32 3.5.3 The Moskuvaara intrusion...... 34 3.5.4 The Puijärvi and Satovaara intrusions...... 35 3.6 The 2.0 Ga intrusions in Kittilä...... 36 3.7 The 1.8 Ga Tainio and Lotto intrusions...... 40 3.7.1 Tainio intrusion...... 40 3.7.2 Lotto dyke...... 41 3.8 Intrusions with unknown age...... 41 3.8.1 Väkkärävaara intrusion...... 41 3.8.2 Värriö intrusion...... 42 3.9 Volcanic rocks...... 43

4 TAIVALKOSKI BLOCK IN THE LENTUA COMPLEX AND KUUSAMO SCHIST BELT...... 45 4.1 Geological background...... 45 4.2 The 2.44 Ga Koillismaa layered intrusion suite...... 47 4.3 Dykes in the Lake Pääjärvi and Suoperä areas, Russia...... 51 4.3.1. Gabbro-norite dyke A1412 Pääjärvi ...... 51 4.3.2 “Older Fe-tholeiitic dyke” A1414 Pääjärvi ...... 51

3 4.3.3 “Younger Fe-tholeiitic dyke” A1492 Pääjärvi ...... 53 4.3.4 Orthopyroxene-phyric dyke A1465 Pääjärvi ...... 53 4.3.5 Fe-tholeiitic Oulanka dyke...... 54 4.3.6 Gabbro-norite dyke, A1415 Suoperä...... 54 4.4 The 2.3 Ga Karkuvaara intrusion ...... 55 4.5 Dykes in the Taivalkoski town area...... 57 4.5.1 A1466 Taivalkoski ...... 57 4.5.2 A1471 Taivalkoski ...... 57 4.5.3 A1797 Törninkuru...... 58 4.5.4 A1796 Kallioniemi...... 58 4.5.5 A1800 Murhiniemi...... 60 4.5.6 A1798 Kontioluoma...... 60 4.5.7 A1794, A1795 Tilsanvaara...... 61 4.5.8 A1802 Koivuvaara...... 61 4.5.9 A1801 Hirsikangas...... 61 4.6 Volcanic rocks in the Kuusamo schist belt ...... 62

5 PUDASJÄRVI COMPLEX AND THE PERÄPOHJA SCHIST BELT...... 63 5.1 Geological background...... 63 5.2 The 2.44 Ga Kemi, Penikat, Kilvenvaara and Siikakämä intrusions ...... 65 5.3 Loljunmaa gabbro-noritic dyke ...... 68 5.4 Tholeiitic dykes, A1410 Uolevinlehto, Pudasjärvi...... 69 5.5 The 2.44 Ga Vengasvaara intrusion...... 70 5.6 Palomaa dyke, A1743...... 71 5.7 Tervonkangas dyke A1808 ...... 71 5.8 The 2.13–2.14 Ga dykes in the Peräpohja schist belt, A1214 Koppakumpu and A2087 Kuusivaara ...... 73 5.9 Rytijänkkä dyke A854...... 74 5.10 Volcanic rocks...... 75

6 KUHMO BLOCK IN THE LENTUA COMPLEX...... 76 6.1 Geological background...... 76 6.2 The 2.4 Ga boninite-norite Viianki dyke, A1356 ...... 78 6.3 The 2.3 Ga dykes, Lohisärkkä A1914, Kovavaara A1361, Karhuvaara A1672...... 79 6.4 The 2.2 Ga Rasiaho dyke A261...... 80 6.5 The 2.15 Ga Petronjärvi dyke A1363...... 83 6.6 The 2.1 Ga Kapea-aho dyke A1212...... 83 6.7 The 2.0–1.95 Ga dykes, Kivikevätti A1409, Puuropuro A1673, Peräaho A1519, Kivimäki A1460.84 6.8 Dykes in the Veitsivaara area, A1489b & A1489c...... 89 6.9 Dykes in the Romuvaara area...... 90

7 KAINUU SCHIST BELT...... 90 7.1 Geological background ...... 90 7.2 The 2.44 Ga Junttilanniemi plutonic-volcanic complex (A1595-6)...... 93 7.3 The Kapustakangas intrusive suite (A1373)...... 96 7.4 The 2.22 Ga intrusions in the Kainuu schist belt...... 97 7.5 The 1.95 Ga Jormua ophiolite...... 101 7.6 Volcanic rocks...... 105

8 IISALMI COMPLEX...... 105 8.1 Geological background...... 105 8.2 The 2.3 Ga dykes, Humppi A135, Siunaussalmi A1369, Petäiskangas A1362...... 106

4 8.3 The 2.13 Ga Nieminen dyke, A1223 and A1368...... 110 8.4 The 2.06 Ga Otanmäki intrusion, A1381...... 111 8.5 The ca. 2.0 Ga dykes Koirakoski A1875, Jäkäläkangas A1838...... 112 8.6 The 1.89 Ga Lapinlahti intrusion...... 114 8.7 Volcanic rocks in the Siilinjärvi area...... 115

9 THE OUTOKUMPU AREA...... 116

10 TOHMAJÄRVI VOLCANIC COMPLEX AND A BASEMENT DYKE...... 118 10.1 Oravaara gabbro A398...... 118 10.2 Purola dyke A1231...... 119

11 CARBONATITES AND LAMPROPHYRES...... 120 11.1 Geological background...... 120 11.2 The 2.61 Ga Siilinjärvi carbonatite...... 121 11.3 The Sokli carbonatite...... 121 11.4 The 2.0 Ga Laivajoki and Kortejärvi carbonatites...... 122 11.5 Lamprophyres...... 123

12 DISCUSSION...... 124 12.1 Episodic rifting stages of the Archaean lithosphere...... 124

12.2 Range of εNd – evidence for heterogeneous mantle and crustal contamination...... 127 12.3 The 2.44–2.50 Ga intrusions and dykes...... 136 12.4 The 2.3 Ga mafic rocks...... 138 12.5 The 2.22 Ga intrusions...... 139 12.6 The 2.1–2.15 Ga mafic rocks...... 140 12.7 The 1.95–2.06 Ga mafic rocks...... 140

13 CLOSING REMARKS ...... 141

ACKNOWLEDGEMENTS...... 142

REFERENCES ...... 142

5 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

DEDICATION

We want to dedicate this work to Dr Olavi Kouvo, the founder of isotope geology in Finland, who recently passed away at the age of 96. He started his isotope geological career at Princeton University in the United States in the mid-1950s. There, he had an opportunity to work with eminent pioneers of geochronology, including George Tilton, George Wetherill and Paul Gast. His doctoral thesis in 1958 was based on U–Pb, Rb–Sr and K–Ar isotope analyses conducted on minerals from Finland. His results, particularly the U–Pb data on zircon, revolutionised our understanding of Finnish Precambrian geology. In the early 1960s, Olavi Kouvo established the isotope laboratory at the Geological Survey of Finland, Espoo, applying the know-how he had acquired during his visit to the Carnegie Institution of Washington. Since then, active collaboration of the isotope laboratory with field geologists at the survey and in universities and mining companies has been the driving force in collecting systematic high-quality sample materials and isotope data from Finland. His interest was not only in dating rocks, but in the overall understanding of geological processes and evolution. Related to this, his enthu- siasm, contacts and the respect he had within the We are confident that the current paper would scientific community at home and abroad greatly have been of special interest to Olavi Kouvo, noting helped in realizing the initiatives to set up facilities that he was among the first geochronologists to use and personnel for Sm–Nd and stable isotope studies the U–Pb zircon method on zircon to date coarse- at the Geological Survey of Finland. grained mafic igneous rocks.

6 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

PREFACE

The isotope laboratory at the Geological Survey of reasonably good results have been obtained by using Finland (GTK) was established in the early 1960s by relict primary igneous phases when clean mineral Dr Olavi Kouvo. From the beginning, the principal separates could be produced. The emphasis of this method has been U–Pb dating, mainly applied to volume is on presenting Sm–Nd isotope data on zircon. The age data produced at the laboratory have well-preserved samples from a large number of been fundamental in establishing the age relation- mafic to ultramafic intrusions and dykes occur- ships of the main geological units in the Finnish ring in the Karelia province, and on providing part of the Fennoscandian Shield. tools for constraining the age and origin of the Subsequently, contributions from other meth- Palaeoproterozoic mafic magmatism and for mod- ods have also been important. The Pb–Pb isotope elling the geological evolution of the lithosphere analyses on sulphides by Matti Vaasjoki since the of the shield. 1970s focused on ore genetic studies. Sm–Nd and In these studies, co-operation with several geol- stable isotope methods were set up in the 1980s by ogists from GTK and Finnish universities has been Hannu Huhma and Juha Karhu, respectively. The very important. Without systematic mapping and Sm–Nd data have contributed significantly to the research performed by geologists with good local understanding of crustal genesis, and C-isotopes geological knowledge, especially in Lapland, many in sedimentary carbonates have been used in pal- rocks with primary igneous phases would not have aeoenvironmental studies. been found and studied. In this context, it is also In addition to zircon, important results have been appropriate to acknowledge the staff of the iso- obtained by applying the U–Pb method to other tope laboratory at GTK, including Matti Karhunen, minerals, e.g., monazite, which can be used to con- Leena Järvinen, Tuula Hokkanen, Lasse Heikkinen strain the timing of metamorphism. As a whole, our and Arto Pulkkinen, who have made a major con- understanding of the geological evolution of Finland tribution to sample processing and isotope meas- greatly relies on the results produced by the GTK urements. Special thanks go to Tuula Hokkanen, isotope laboratory. who has performed most of the final purification The main objective of this volume is to pub- of minerals and chemical processing, and to Arto lish isotope data on Palaeoproterozoic mafic rocks Pulkkinen for most of the recent mass spectrom- from the Karelia province in eastern and northern etry. Discussions with Hugh O’Brien and the late Finland. The Sm–Nd method has mainly been used Matti Vaasjoki at the isotope laboratory, as well as in studying the genesis of crustal rocks, but also in many other colleagues over the years, are greatly obtaining geochronological information on igne- acknowledged. ous rocks. Alteration has been a frequent problem in the Sm–Nd dating of ancient metamorphosed Hannu Huhma mafic igneous rocks from the Finnish bedrock, but

7 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

1 INTRODUCTION

Abundant isotope data have been obtained from work also involved ca. 200 analyses of a control mafic rocks in Finland since the pioneering studies sample (A382). by Olavi Kouvo in the 1960s, when he established The Sm–Nd results, together with published the U–Pb zircon dating laboratory at GTK in Espoo. data, cover approximately 90 intrusions and 20 The Sm–Nd method was set up in the early 1980s, mafic volcanic formations within the Karelia prov- and from the beginning has been widely used to ince in Finland (Fig. 1). In addition, new U–Pb zir- date and study the origin of mantle-derived mafic con or baddeleyite ages are given for ca. 30 samples rocks. Numerous studies have utilised these results, (no Nd available for these). In presenting the iso- but in many cases, the primary data have remained tope results for mafic rocks, we group the studied unpublished or incompletely published (e.g., Vuollo rocks largely based on their areal distribution in & Huhma 2005, Huhma et al. 1995). One of the aims different geographical units, such as Lapland and of this paper is to publish all these data. An impor- the Outokumpu and Tohmajärvi areas, or geologi- tant contribution is also provided by recent laser cal units, such as Palaeoproterozoic schist belts, ablation ICP-MS analyses conducted on many of the including the Peräpohja, Kuusamo and Kainuu previously studied samples, from which multigrain schist belts, and Archaean basement areas, such as ID-TIMS analyses had yielded discordant and het- the Taivalkoski and Kuhmo blocks (Vuollo & Huhma erogeneous results. In addition to the data produced 2005) in the Lentua complex and Pudasjärvi and at GTK, some U–Pb data on baddeleyite analysed at Iisalmi complexes. Carbonatites and lamprophyres the Royal Ontario Museum, University of Toronto, are considered in a chapter of their own. The study are also included in this volume. sites are located within the borders of Finland, This paper reports ca. 400 previously unpublished with the exception of six mafic dykes that occur in Sm–Nd analyses on mafic rocks from eastern and the Pääjärvi and Suoperä areas in Russia, east and northern Finland (Appendix 1). These are combined southeast of Kuusamo (Fig. 38). with ca. 200 previously published Sm–Nd analyses After reporting the data, with possible comments to constrain the origin of mafic magmatism and on their significance in the interpretation of local related geological evolution. The previously unpub- geological relationships, the results are briefly lished U–Pb data include ca. 220 analyses by TIMS, summarised and discussed in order to constrain ca. 240 analyses by SIMS (NORDSIM and SHRIMP) the age and origin of the Palaeoproterozoic mafic and ca.1200 analyses by LA-MC-ICPMS. The ICPMS magmatism within the Karelia province.

2 ANALYTICAL METHODS

2.1 Nd isotope analysis

Most Sm–Nd studies in this paper were performed was achieved. Samarium and Nd were separated in at the Geological Survey of Finland using the fresh- two stages using a conventional cation exchange est samples that were available. Standard proce- procedure (7 ml of AG50Wx8 ion exchange resin dures were used for crushing and the separation of in a bed of 12 cm length) and a modified version of plagioclase and pyroxene, and the final purifica- the Teflon-HDEHP (hydrogen diethylhexyl phos- tion often involved hand-picking. For the Sm–Nd phate) method developed by Richard et al. (1976). analyses, mineral concentrates were washed ultra- The measurements were carried out in a dynamic sonically in warm 6 M HCl for 30 min and rinsed mode on a VG SECTOR 54 mass spectrometer using several times in water. The samples (150–200 mg) Ta-Re triple filaments. Nd ratios were normalised were dissolved in HF-HNO3 using Savillex screw- to 146Nd/144Nd = 0.7219. The long-term average cap Teflon beakers or sealed Teflon bombs (felsic 143Nd/144Nd ratio for the La Jolla standard is 0.511850 whole rocks) for 48 h. A mixed 149Sm–150Nd spike ± 0.000010 (standard deviation for 220 measure- was added to the sample prior to the dissolution. ments during 1996–2010). The Sm/Nd ratio of the After careful evaporation of fluorides, the residue spike was calibrated against the Caltech mixed was dissolved in 6 M HCl so that a clear solution Sm/Nd standard (Wasserburg et al. 1981). Based on

8 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland ¢

^_ ^_Tshokkoaivi �o�o ^_

Intrusions and volcanic rocks ^_ XW ^_ ^_ XW ^_ ^_ ^_ ^_^_XW Sm-Nd (±U-Pb) ^_^_^_^_ ^_ XW^_^_^_^_ ^_^_^_ ^_^_^_ _ ^_ ^_^_ ^_^_^_ ^_ ^_ ^_ ^ ^_ ^_ ^_^_^_ ^_ ^_^_ ^_^_ Volcanic rocks ^_ ^_^_^_ ^_ ^_ ^_ Lapland ^_ U-Pb ( in this paper ) ^_ ^_ ^_ ^_ ^_XW ^_ Tainio ^_ ^_ U-Pb ^_ ^_ (age published elsewhere) ^_^_ Sm-Nd (age ?) ^_ ^_ XW ^_ ^_ 2.4 - 2.5 Ga felsic rocks ^_ ^_ ^_^_ Pääjärvi ( in this paper ) ^_^_^_^_^_^_ ^_ ^_ ^_ ^_^_ ^_ ^_ ^_ ^_^_ ^_^_ Age ^_ ^_^_ ^_ ^_ Tornio ^_^_^_ ^_ Taivalkoski <1930 Ma ^_^_ ^_ ^_ ^_ ^_^_ ^_^_^_ 1931 - 2080 Pudasjärvi^_ ^_^_ ^_^_ Näränkävaara ^_ 2081 - 2180 ^_ ^_^_ ^_ ^_^_ ^_ ^_ 2181 - 2280 ^_ ^_ ^_ 2281 - 2380 ^_ ^_ ^_^_ ^_ ^_ >2380 ^_^_ ^_ ^_^_^_ ^_ 50 km ^_ ^_^_ ^_ ^_ ^_ ^_ ^_ ^_^_^_^_ ^_ ^_ ^_ ^_^_ ^_ ^_^_ Kuhmo ^_ Iisalmi ^_^_ ^_ ^_ ^_ ^_ ^_^_ ^_ ^_ ^_^_^_ ^_ ^_^_ ^_ ^_^_ ^_ ^_ ^_^_

Fig. 1. Map showing the localities of the studied Palaeoproterozoic mafic rocks in the Karelia province. Stars and dots – intrusions; Red star – Sm–Nd (±U–Pb, 86 targets), green star – volcanic rocks (Sm–Nd, 22 targets), red dot – Sm–Nd in this paper, but age not well constrained (6), blue star – U–Pb in this paper (no Nd data available, 29), black open star – U–Pb age published elsewhere (78). Symbol size correlates with age: >2380 Ma (large symbol); other break values 2280, 2180, 2080 and 1930 (<1930 Ma small symbol). Base map – geological map of the Fennoscandian Shield (Koistinen et al. 2001).

9 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye several duplicate analyses, the error of the (TDM) were calculated after DePaolo (1981). Linear 147Sm/144Nd ratio was estimated to be better than regression calculations and plotting of isotope data 0.4%. on isochron diagrams were performed using the Measurements performed on the rock stand- Isoplot/Ex 3.0 programme of Ludwig (2003). ard BCR-1 provided the following values: Sm = A few older Sm–Nd analyses carried out using an 6.58 ppm, Nd = 28.8 ppm, 147Sm/144Nd = 0.1380, old technique and a non-commercial mass spec- 143Nd/144Nd = 0.51264 ± 0.00002. The blank was trometer (Huhma 1986) are included in this paper. 30–100 pg for Sm and 100–300 pg for Nd. The Compared with more recent analyses, they tend to 147 143 144 εNd values were calculated using λ Sm = 6.54 · 10-12 yield slightly larger errors in Nd/ Nd, but based a-1, 147Sm/144Nd = 0.1966, and 143Nd/144Nd = 0.512640 on duplicated newer analyses, are consistent within for the present CHUR. Depleted-mantle model ages error.

2.2 U–Pb isotope analysis

Sampling for the isotope studies was mostly car- 207Pb/204Pb. The measured Pb blank was 10– ried out in conjunction with extensive mapping 50 pg, but probably higher during the few ancient projects, and the samples should thus be well analyses included. The U–Pb age calculations were chosen and relevant in solving major geological performed using the PbDat and the Isoplot/Ex pro- problems. Samples for the isotope studies were grammes (Ludwig 1991, 2003). Some U–Pb analy- washed, crushed, cleaned from light minerals using ses on baddeleyite were obtained from the Royal a Wilfley table, and treated with methylene iodide Ontario Museum, University of Toronto, following and Clerici® solutions for separation of the heavy the methods of Krogh et al. (1987). minerals. Non-magnetic heavy fractions were sep- For SIMS and LA-ICP-MS analyses, zircon grains arated with a Frantz isodynamic separator. Zircon were hand-picked under a binocular microscope, grains were selected for analyses by hand-pick- mounted in epoxy resin, sectioned approximately ing. Some of the fractions were air abraded (Krogh in half and polished. To target the analysis spots, 1982) for ID-TIMS (isotope dilution thermal ion- back-scattered electron (BSE) and cathodolumi- isation mass spectrometry) and U–Pb analyses, nescence (CL) images of the zircon grains were and for some more recent analyses, zircons were taken using SEM. Half of the SIMS U–Pb analy- treated using the chemical abrasion (CA) method ses were performed using the Nordic Cameca IMS by Mattinson (2005). When applying the CA-TIMS 1270 at the Swedish Museum of Natural History, technique, we largely followed the steps described Stockholm (NORDSIM facility). The spot diameter by Schoene et al. (2006), in which zircon was placed for the 4 nA primary O2- ion beam was 25 μm, in a furnace at 900 oC for 60 hours in beakers before and oxygen flooding in the sample chamber was being transferred to Teflon microcapsules, placed used to increase the production of Pb+ ions. Three in high-pressure vessels, and leached in 29M HF counting blocks, each including four cycles of the for 12 hours. The decomposition of minerals and Zr, Pb, Th and U species of interest, were measured extraction of U and Pb for multigrain ID-TIMS anal- from each spot. The mass resolution (M/ΔM) was yses mainly followed the procedure described by 5400 (10%). The raw data were calibrated against a Krogh (1973). 235U-208Pb-spiked and -unspiked zircon standard (91500; Wiedenbeck et al. 1995) and isotope ratios were measured using a VG Sector corrected for modern common lead (T = 0; Stacey 54 or non-commercial mass spectrometers at the & Kramers 1975). For the detailed analytical proce- Geological Survey of Finland in Espoo. The mea- dure, see Whitehouse et al. (1997). All the errors in sured lead and uranium isotope ratios were nor- age reported in the text and figures are given at the malised to the accepted values of the NBS 981 and 2σ level. For some samples, the dating was carried U500 standards. The performance of the ion counter out at VGESEI in St Petersburg using SHRIMP II was checked by repeated measurements of the NBS and the analytical methods described by Larionov 983 standard. et al. (2004). Common-lead corrections were carried out The measurements by LA-MC-ICPMS were per- using the age-related Pb isotope composition of formed utilising the Nu Plasma HR multicollector the Stacey and Kramers (1975) model and errors ICPMS at the Geological Survey of Finland in Espoo. of 0.2 for 206Pb/204Pb and 208Pb/204Pb and 0.1 for A technique very similar to that described by Rosa

10 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland et al. (2009) was applied, with the exception that a The calculations were performed offline using New Wave UP193 Nd:YAG laser microprobe was used. an interactive spreadsheet programme written in Samples were ablated in He gas (gas flow = 0.2– Microsoft Excel/VBA by T. Andersen (Rosa et al. 0.3 l/min) using a low-volume teardrop-shaped 2009). To minimise the effects of laser-induced (<2.5 cm3) laser ablation cell (Horstwood et al. elemental fractionation, the depth-to-diameter 2003). The He aerosol was mixed with Ar (gas flow = ratio of the ablation pit was kept low, and isotopi- 1.2 l/min) in a Teflon mixing cell prior to entry into cally homogeneous segments of the time-resolved the plasma. The gas mixture was optimised daily for traces were calibrated against the corresponding maximum sensitivity. All analyses were conducted time interval for each mass in the reference zir- in static ablation mode. The ablation conditions con. To compensate for drift in instrument sen- were the following: beam diameter generally 25 μm, sitivity and Faraday vs. electron multiplier gain pulse frequency 10 Hz and beam energy density during an analytical session, a correlation of sig- 1.4 J/cm2. A single U–Pb measurement included 30 s nal vs. time was assumed for the reference zircons. of on-mass background measurement, followed by A description of the algorithms used is provided 60 s of ablation with a stationary beam. Masses 204, in Rosa et al. (2009). Plotting of the U–Pb isotope 206 and 207 were measured in secondary electron data and age calculations were performed using the multipliers and 238 in the extra high-mass Faraday Isoplot/Ex 3.0 programme (Ludwig 2003). All ages collector. The geometry of the collector block does were calculated with 2σ errors and without decay not allow simultaneous measurement of 208Pb and constant errors. 232Th. Ion counts were converted and reported as A few recent U–Pb analyses were performed volts by the Nu Plasma time-resolved analysis soft- using a Nu Plasma AttoM single collector ICP-MS ware. 235U was calculated from the signal at mass at the Geological Survey of Finland in Espoo, con- 238 using a natural 238U/235U ratio of 137.88. Mass nected to a Photon Machine Excite laser ablation number 204 was used for monitoring the amount of system. Samples were ablated in He gas (gas flows common 204Pb. In ICP-MS analyses, 204Hg mainly = 0.4 and 0.1 l/min) within a HelEx ablation cell originates from the He supply. The observed back- (Müller et al. 2009). The He aerosol was mixed with ground counting rate at mass 204 was ca. 1200 (ca. Ar (gas flow = 0.8 l/min) prior to entry into the 1.3 × 10−5 V), and had been stable at that level dur- plasma. The gas mixture was optimised daily for ing the year prior to the measurements. The con- maximum sensitivity. Typical ablation conditions tribution of 204Hg from the plasma was eliminated were the following: beam diameter 25 μm, pulse by on-mass background measurement prior to each frequency 5 Hz and beam energy density 2 J/cm2. analysis. A single U–Pb measurement included a short pre- Age-related common-lead (Stacey & Kramers ablation, 10 s of on-mass background measure- 1975) correction was used if the analysis showed ment, followed by 30 s of ablation with a stationary common-lead contents above the detection limit. beam. 235U was calculated from the signal at mass Signal strengths at mass 206 were typically >10−3 V, 238 using a natural 238U/235U = 137.88. Mass num- depending on the uranium content and age of the ber 204 was used as a monitor for common 204Pb. zircon. Two calibration standards were run in dupli- The observed background counting rate at mass 204 cate at the beginning and end of each analytical ses- was 150–200 cps, and had been stable at that level sion and at regular intervals during sessions. Raw over the previous two years. The contribution of data were corrected for background, laser-induced 204Hg from the plasma was eliminated by on-mass elemental fractionation, mass discrimination and background measurement prior to each analysis. drift in ion counter gains, and reduced to U–Pb iso- Age-related common lead (Stacey & Kramers 1975) tope ratios by calibration to concordant reference correction was used when the analysis showed com- zircons of known age, using protocols adapted from mon lead contents significantly above the detection Andersen et al. (2004) and Jackson et al. (2004). limit (i.e., >50 cps). Signal strengths at mass 206 Standard zircons GJ-01 (609 ± 1 Ma; Belousova et were typically 100 000 cps, depending on the ura- al. 2006) and an in-house standard, A1772 (2711 ± nium content and age of the zircon. 3 Ma/TIMS; 2712 ± 1 Ma/SIMS, Huhma et al. 2012a), were used for calibration.

11 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

3 LAPLAND

3.1 Geological background

The main units of the geology of Finnish Lap Finnish–Norwegian border (Fig. 1). In the south, land are the Archaean basement complexes, it is bordered by the Central Lapland Granitoid Palaeoproterozoic sedimentary and volcanic rocks of Complex and Vuojärvi suite (paragneisses between the Central Lapland greenstone belt, granitoids, and the southern granite complex and northern CLGB) the Lapland granulite belt. The largest basement and in the northeast by the Lapland granulite belt. terrain, the Kemihaara granitoid complex, occurs In the division and terminology recently proposed in eastern Lapland, south of the Lapland granulite by Nironen et al. (2016), the Vuojärvi group/suite belt (Fig. 1). It is composed of typical Archaean TTG is renamed as the Central Lapland supersuite. gneisses, granitic intrusions, small greenstone belts Descriptions of the stratigraphic successions of the and a 15- to 30-km-wide and 100-km-long mica CLGB are found in Lehtonen et al. (1998) and Hanski gneiss area called the Tuntsa suite (Fig. 2). The lat- & Huhma (2005), and a large amount of geochrono- ter hosts some of the Palaeoproterozoic intrusions logical data was published in a volume of the Special dated in this study (Chapter 3.2.3 and 3.8.2). Paper of the Geological Survey of Finland edited The geology of Finnish Lapland is dominated by by Vaasjoki (2001). The stratigraphy is divided into the Palaeoproterozoic Central Lapland greenstone several lithostratigraphic groups (Fig. 3) bearing belt (CLGB). The greenstone belt extends from witness to a sedimentary and volcanic evolution of the Finnish–Russian border at Salla through the hundreds of millions of years. In this paper, we fol- Sodankylä and Kittilä areas to the north until the low the approach of Hanski & Huhma (2005) with

Kuotko ^_ A1671XW ¢ ^_A1273 Nu�o Selkäsenvuoma ^_ A0157XW A0206 ^_A0580 Tuntsa suite ^_ ^_ Ki�lä suite ^_ A1432 ^_Pi�arova A0659XWXW ^_ ^_ XW ^_ ^_^_ ^_ ^_ ^_ ^_ Palovaara Puijärvi Koitelainen ^_ ^_ Keivitsa^_^_ Peuratunturi ^_ ^_ ^_Satovaara Mosk^_ uvaara ^_ ^_ ^_A1565 ^_^_ ^_ ^_A0817 Silmäsvaara ^_ ^_ ^_ ^_^_ ^_ ^_Kannusvaara Rantavaara Värriö Intrusions and ^_ ^_ volcanic rocks ^_Haaskalehto ^_A0418 ^_ ^_ Koulumaoiva Sm-Nd (±U-Pb) ^_ ^_ Rovasvaara Volcanic rocks ^_ ^_Akanvaara U-Pb ( in this paper ) ^_ ^_ ^_ ^_ Norway Tainio A1524^_XW U-Pb (age published elsewhere) ^_ Sm-Nd (age ?) Russia ^_ XW ^_ ^_ 2.4 - 2.5 Ga felsic rocks ^_Ahvenvaara ( in this paper ) ^_ Sweden Age ^_^_ ^_ <1930 Ma ^_ 1931 - 2080 Finland ^_ 2081 - 2180 2181 - 2280 ^_ ^_ ^_ Onkamonlehto ^_ 2281 - 2380 ^_ >2380 ^_ 0 30 km Lehtomaa ^_

Fig. 2. Geological map of Lapland showing sample localities. For symbols, see Figure 1. The Sm–Nd results from the 2.44 Ga Koitelainen and Akanvaara layered intrusions were published by Hanski et al. (2001a) and from the 2.2 Ga Haaskalehto, Ahvenvaara and Silmäsvaara intrusions by Hanski et al. (2010).

12 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Fig. 3. Group-level lithostratigraphy of the Central Lapland greenstone belt after Lehtonen et al. (1998) with modifications: the Kumpu and Lainio Groups have been united with the Kumpu Group, the name of the Onkamo Group has been changed to the Kuusamo Group and the Vuojärvi suite has been included. Numbers indicate ages in billions of years. remarks on the updated terminology (Nironen et the 2.05 Ga Kevitsa layered intrusion (Mutanen & al. 2016, 2017). Huhma 2001) into black schists of the Savukoski The development of the CLGB started around the Group constrains the age of these rocks. The ca. Archaean–Proterozoic boundary, with rifting of the 2.0 Ga Kittilä Group (Kittilä suite in Finnstrati, Nironen Archaean basement and subaereal eruption of inter- et al. 2016), which comprises various basaltic vol- mediate to felsic volcanic rocks of the Salla Group. canic rocks (NMORB, OIB, IAT), chemical sedi- These were followed by the crustally contaminated ments and pieces of ophiolitic mantle rocks, is komatiites and mafic lavas of the Kuusamo Group thought to be an allochthonous sliver of oceanic (corresponding to the Onkamo Group of Lehtonen crust (Hanski 1997) and thus in tectonic con- et al. (1998) and Hanski & Huhma (2005)). The 2.44 tact with the sedimentary-volcanic associations Ga layered intrusions (Koitelainen and Akanvaara) described above. Quartzites and conglomerates of cut Salla Group rocks, but not those of the Kuusamo the Kumpu Group and minor volcanic rocks cap Group (Manninen & Huhma 2001). The early rifting- earlier rocks with angular unconformity. These related volcanism was followed by the deposition of molasse-type sediments contain pebbles of ca. siliciclastic sediments, basaltic flood basalt-type 1.88 Ga synorogenic granitoids. lavas, carbonate rocks and pelitic sediments of the In the Central Lapland greenstone belt, differen- Sodankylä Group. The quartzites of the Sodankylä tiated mafic-ultramafic bodies and mafic dykes were Group are cut by 2.22 Ga mafic-ultramafic sills, giv- emplaced in several stages, including ca. 2.44 Ga, ing a minimum age for these sedimentary rocks. 2.22 Ga, 2.15 Ga, 2.05 Ga and 2.0 Ga (Hanski et al. Subsequent deepening of the sedimentary basin 2001a). Small mafic intrusions are also associated resulted in the accumulation of fine-grained sedi- with generally felsic syn- to post-orogenic granitic ments, phyllites and black schists, assigned to the magmatism, as discussed later on. The ca. 2.44 Ga Savukoski Group. The upper part of this group is age group is represented by the Koitelainen and composed of komatiitic and picritic volcanic rocks Akanvaara intrusions (Fig. 4). Zircon U–Pb dat- and their derivative basalts. The emplacement of ing results obtained at GTK for these intrusions

13 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

(Mutanen & Huhma 2001) and other similar lay- lower zone of the Burakovka intrusion seem to be ered intrusions from the Tornio-Näränkävaara slightly higher, from –1 to 0. The Sm–Nd results intrusion belt (Alapieti 1982, Perttunen & Vaasjoki for most mafic dykes of this age group also give

2001, Maier et al. in press, this work) are consist- initial εNd values of –1 to –2, although some dykes ent with each other and also with ages obtained have yielded higher values of up to +1.7 (this paper). for mafic layered intrusions in Russia, including The Sodankylä Group quartzites are cut by 2.22 the Oulanga complex and the Burakovka, Imandra, Ga, strongly differentiated mafic-ultramafic sills Fedorovo-Pansky and Mt. Generalskaya intrusions that are called karjalites or assigned to the gabbro- (e.g., Amelin et al. 1995). In contrast, slightly older wehrlite association (GWA), based on their common ages close to ca. 2.50 Ga have been reported for the occurrence in Karelian schist belts or their typi- Monche, Fedorovo-Pansky and Mt. Generalskaya cal lithology, respectively (Vuollo & Huhma 2005, intrusions (Balashov et al. 1993, Amelin et al. 1995, Hanski et al. 2010). The presence of these differ- Bayanova et al. 1999). There has been no evidence entiated sills, together with positive δ13C values in for the presence of intrusions of this age group in dolomitic carbonate rocks (Karhu 2005), allows the Finland until this study (see Chapter 3.2.1). correlation of the Sodankylä Group quartzites with Sm–Nd studies have demonstrated that the Jatulian-system quartzites in other Karelian schist Koitelainen and Akanvaara intrusions (Hanski belts, such as the Peräpohja, Kuusamo, Kainuu et al. 2001c) and the bulk of the other ca. 2.44– and North Karelia belts. Using the TIMS and SIMS 2.50 Ga mafic intrusions (Huhma et al. 1990, methods and an electron microprobe, Hanski et al. Tolstikhin et al. 1992, Balashov et al. 1993, Amelin (2010) performed a detailed geochronological and & Semenov 1996, Karinen 2010) are characterised mineralogical study on zircon in the GWA sills from by negative initial εNd values from –1 to –2, thus several schist belts in order to find reasons for the having a clear lithospheric signature in their Nd commonly observed strong discordancy of the iso- isotope composition. The initial εNd values from tope compositions and to clarify the real magmatic the Tsipringa intrusion (Oulanga complex) and the ages of the rocks. One of the most well-preserved

Fig. 4. Geological map showing major intrusions related to the several generations of mafic-ultramafic magmatism in Central Lapland.

14 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland and precisely dated examples of the 2.22 Ga sills is lower crust (“magmatic underplating”) (Neilson et the Haaskalehto intrusion (Fig. 4), which is located al. 2009). The results of seismic reflection sound- ~20 km west of Sodankylä (Hanski et al. 2010). It ings in Finnish Lapland support the idea of under- seems that these sills represent a short-lived but plated magmas. widely distributed magmatic phase in Karelian In Chapters 3.2–3.9, we report geochronologi- formations in eastern and northern Finland, with cal data and Nd isotope compositions mainly for similar sills also being recently recognised in the mafic intrusive and volcanic rocks from the Central Kautokeino belt, northern Norway (Bingen et al. Lapland greenstone belt (Fig. 2). In this context, we 2015). also publish new data suggesting that some felsic The following intrusive event took place at ca. volcanism took place at ca. 2.50 Ga. The other stud- 2.15 Ga, producing a group of differentiated intru- ied intrusive rocks include a granophyre from the sions in the Tanhua area and the Rantavaara sill 2.44 Ga Koitelainen layered intrusion (Fig. 4), two in an area between the Koitelainen intrusion and ca. 2.4–2.43 Ga dyke-like intrusions (Lehtomaa, Sodankylä town (Fig. 4). The magmatism that Onkamonlehto) (Fig. 2), four intrusive bodies of occurred at 2.05 Ga in Lapland has special signifi- the ca. 2.15 Ga age group (Tanhua, Rantavaara) cance, as it included voluminous komatiitic volcanic (Fig. 4), four intrusions and one dyke having an eruptions (Hanski et al. 2001b) and differentiated age of ca. 2.05 Ga (Kevitsa, Satovaara, Moskuvaara, mafic–ultramafic intrusions with a high ore poten- Puijärvi) (Fig. 4), and two gabbroic intrusions tial, as testified by the Ni–Cu–PGE-bearing Kevitsa (Selkäsenvuoma, Pittarova) and felsic porphyries intrusion (Mutanen 1997, Mutanen & Huhma 2001, from the Kittilä ca. 2.0 Ga allochthon (Fig. 2). Santaguida et al. 2015) located just south of the Among the studied volcanic rocks from the Koitelainen intrusion (Fig. 4). Mafic intrusions CLGB are Savukoski Group komatiites from the with an age of ca. 2.0 Ga occur within the Kittilä Sattasvaara area and Savukoski Group mafic pil- allochthon (Kittilä Group) and, together with felsic low lavas from the Linkupalo area (Fig. 5). Some porphyries, serve as good targets for obtaining geo- felsic rocks were also included due to problems chronological information for this aerially extensive related to the previous geochronological data. geological unit (Rastas et al. 2001). One problematic unit, the Honkavaara Formation Palaeoproterozoic felsic plutonic rocks in north- of Lehtonen et al. (1998), contains rocks that are ern Finland are divided into two main broad groups: interpreted as intermediate and felsic volcanic rocks the ca. 1.89–1.88 Ga synorogenic plutons, includ- (Fig. 5). Even though these rocks are regarded as ing the Haaparanta Suite, and postorogenic, ca. Palaeoproterozoic (Lehtonen et al. 1998, Lahtinen 1.80–1.77 Ga granites (Nironen 2005). Both of these et al. 2015a), U–Pb zircon ages acquired for them magmatic stages include a small amount of mafic are Archaean (ca. 2.72, 2.75 Ga; Rastas et al. 2001). rocks. An example of synorogenic mafic intrusions These types of rocks are found close to the south- is the Fe–Ti oxide-bearing Karhujupukka intrusion ern and southeastern margin of the CLGB, not far (Karvinen et al. 1988). Postorogenic mafic rocks from the Central Lapland Granitoid Complex, and are represented by appinitic intrusions, which are have been assigned to the Vuojärvi suite or Central melanocratic hornblende-rich syenite, monzonite Lapland supersuite (Nironen et al. 2016). A charac- or diorite rocks rich in volatites. Currently, more teristic feature that they share is strong albitization than twenty appinitic pipes and dykes have been (Eilu 1994). recognised in Finnish Lapland (Mutanen 2003, Outside the CLGB, two intrusions (Palovaara, Mutanen & Väänänen 2004, Mutanen 2011). The Tshokkoaivi, Fig. 1), ca. 2.22 and 2.50 Ga in age, were intrusions occur in swarms, with each swarm being studied from the NW corner of Finland and two ca. located above a positive Bouguer gravity anomaly. 2.45 Ga intrusions (Peuratunturi and Koulumaoiva) This type of relationship between the occurrence from the Tuntsa suite in eastern Lapland (Fig. 2). of dense deep-crustal bodies and the location of Data are also presented for two other intrusions appinitic intrusions seems to be the case practi- from the Tuntsa suite, which have an uncertain age. cally everywhere, in southern Finland, eastern The youngest (1.80 Ga) mafic magmatism of this Norrbotten in Sweden (Bergman et al. 2001), and work is represented by the appinitic Tainio intru- in the wider world, suggesting that the sources of sion in the northern part of the Central Lapland appinitic rocks were huge masses of mafic magma Granitoid Complex and the Lotto mafic dyke within emplaced at the mantle–crust boundary or into the the Lapland granulite belt (Fig. 1).

15 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Kuotko ^_ A1671XW ¢

^_A1273 A0157 Nuttio ^_ XW Selkäsenvuoma ^_ ^_ ^_ Peuramaa/ ^_ Pittarova ^_ Savukoski Gr A0659XWXWA0206 ^_ Vesmajärvi/ ^_ ^_ ^_ ^_ ^_Kittilä suite Puijärvi ^_^_ ^_ Linkupalo Fm/ Jeesiörova/ Möykkelmä/ Koitelainen Savukoski Gr Kuusamo Gr ^_ ^_ ^_Savukoski Gr Keivitsa ^_^_ Palovaara Kautoselkä/ ^_ ^_ ^_Kittilä suite Sattasvaara Fm/ ^_ ^_Savukoski Gr ^_ ^_ ^_ ^_ Moskuvaara Norway A1565 ^_ ^_^_ Silmäsvaara^_ ^_ ^_ Väkkärävaara Rantavaara ^_ Russia Honkavaara Fm/^_ Sodankylä Gr Sweden ^_Haaskalehto ^_

Finland

Rovasvaara^_

0 30 km

Fig. 5. Geological map of Central Lapland showing the sample sites (symbols in Figure 1). ^_ ^_

3.2 The 2.4–2.5 Ga intrusions in Lapland

3.2.1 Tshokkoaivi intrusion cally the uppermost) part of the intrusion. Here, likewise, erratics of dark iron-rich gabbros with The Tshokkoaivi intrusion is located in the north- abundant disseminated magmatic Fe–Cu sulphides western part of Finnish Lapland (Fig. 1). It has indicate that such rocks, although unexposed, do previously been studied by Vaasjoki (1971) and occur amongst the uppermost cumulates. Kantti (2002). The intrusion is ca. 10 km long and Two samples of the Tshokkoaivi intrusion were has a total area of ca. 14 km2. In its northern part collected for Sm–Nd isotope studies from a field (Tshokkoaivi fell), the intrusion runs south, turn- of large in situ boulders ca. 2 km SW of the top of ing southwest at Kaamajoki. The rocks are mainly the Tshokkoaivi hill. The rock types represented plagioclase-pyroxene cumulates and plagioclase- are a pyroxene-bearing gabbro (A1317) and a poiki- pigeonite cumulates (augite norite in Kantti 2002), litic plagioclase- and olivine-bearing pyroxenite very similar to corresponding cumulates of the (pyroxene cumulate) (A1318). The primary mag- Koitelainen intrusion. Ultramafic rocks (olivine-rich matic minerals, including clinopyroxene, olivine cumulates and olivine-bearing pyroxene cumulates) and plagioclase, are well preserved in both samples. occur in the lower (eastern) part of the intrusion. The Sm–Nd data acquired from whole rock, Plagioclase-rich cumulates (anorthosites) are pyroxene and plagioclase are presented in Appendix known to occur near the eastern contact (Kantti 1 and in Figure 6. The analysed rocks have a rela- 2002), but judging from the common occurrence of tively low Sm/Nd ratio, i.e., a LREE-enriched chon- anorthosites as glacial erratics, anorthositic cumu- drite-normalised REE pattern, which is typical for lates should be common west of Lake Tshokkajavri, 2.4–2.5 Ga mafic intrusions in eastern and north- in the westernmost (and, supposedly, stratigraphi- ern Finland. Only three Sm–Nd analyses per sample

16 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5126 Tshokkoaivi intrusion

0.5122 Age = 2458 ± 81 Ma A1317cpx eps = -2.1 A1318 px MSWD = 1.9 n=6 0.5118 Nd

144 0.5114 A1317 A1317 wr#2 Age = 2463 ± 51 Ma Nd/ A1318 wr eps = -1.9

143 0.5110 MSWD = 1.4 n=3 A1318 A1317plag Age = 2441 ± 72 Ma 0.5106 A1318 plag eps= -2.5 MSWD = 1.9 n=3 0.5102 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 147 144 Sm/ Nd U-Pb zircon age 2499±11 Ma

Fig. 6. Sm–Nd isotope data for whole-rock samples and mineral separates from the Tshokkoaivi intrusion.

(plagioclase, pyroxene, whole rock) are available. 2.44–2.50 Ga intrusions (for a summary, see Hanski The data for gabbro sample A1317 provide an age of 2012).

2463 ± 51 Ma and an initial εNd value of –1.9 (MSWD = 1.4, Huhma et al. 1996). The Sm–Nd analyses for 3.2.2 Koitelainen intrusion and associated felsic the pyroxenite sample A1318 are slightly scattered volcanic rocks and give an age of 2441 ± 72 Ma (εNd = –2.5, MSWD = 1.9). Combining all the data, an age of 2458 ± 81 Ma The Koitelainen intrusion is located in central can be calculated. The initial εNd value is –2.1, and Finnish Lapland (Figs. 2, 4), and with an aerial slight scatter is indicated by the MSWD value of 1.9. extent of 26 km x 29 km is the largest layered intru- The age of the Tshokkoaivi intrusion was con- sion in Finland (Mutanen 1997). An age of ca. 2.44 Ga firmed by LA-MC-ICPMS analyses conducted on an was already obtained in the 1970s using U–Pb dating old zircon sample, A360. Twenty-five data points of zircon (Kouvo 1977). Mutanen & Huhma (2001) from zircon grains extracted from this sample define published U–Pb zircon ID-TIMS data and ages of a chord with intercepts at 2499 ± 11 Ma and 409 ± 2439 ± 3 Ma and 2439 ± 7 Ma for two gabbro pegma- 27 Ma (Fig. 7, Appendix 2). Many of these results toid samples from the intrusion. They also studied are discordant, but analyses performed utilising the a granophyre sample, A580 Kaitaselkä, which gave best-quality zircon grains gave concordant ages of a clearly younger date of 2405 ± 9 Ma. Recently, ca. 2.5 Ga. Five analyses of strongly altered zircon zircon grains from this sample were re-analysed grains with relatively high common lead contents by LA-ICP-MS. The data from the best-quality zir- suggested ages of 1.8–1.9 Ga. These spot analyses con yielded a concordia age of 2434 ± 5 Ma, which explain well the previously obtained discordant and is considered an igneous age (Fig. 8). A few data scattered multigrain TIMS data (Fig. 7). It is also values with elevated common lead tend to plot on likely that the lower intercept age of ca. 400 Ma the younger side (Appendix 2). These results pro- has geological significance, potentially related to vide an explanation for the slightly younger upper the development of the adjacent Caledonian Orogen. intercept age obtained from discordant ID-TIMS

Using an age of 2499 Ma, an initial εNd value of –1.8 data by Mutanen & Huhma (2001). can be calculated, which is a typical value for the ca.

17 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A360 Haukikuru/ Tshokkoaivi gabbro 0.6 data-point error ellipses are 2s Intercepts at 2499 ± 11 Ma & 409 ± 27 Ma 0.5 2600 MSWD = 0.92 n=25

2200

0.4 U 1800 238 0.3

Pb/ 1400 TIMS

206 0.2 1000

0.1 600

0.0 0 2 4 6 8 10 12 14 207Pb/235U

Fig. 7. Concordia plot of U–Pb zircon data from the sample A360, Tshokkoaivi intrusion. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses as red dots.

0.6 A580 Kaitaselkä granophyre Koitelainen intrusion

Concordia Age =

0.5

U 2434 ±5 Ma 2500

238 n=7

Pb/ 2300 206 0.4 2100

1900

1700 0.3 TIMS Intercepts at 1500 256±25 & 2408 ± 6 Ma MSWD = 10.8, n=10 1300

0.2 2 4 6 8 10 12 207Pb/235U data-point error ellipses are 2s

Fig. 8. Concordia plot of U–Pb zircon data for the granophyre sample A580, Koitelainen intrusion. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses as black dots.

18 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

In this context, we report updated dating results the zircon U–Pb LA-MC-ICPMS data yielded a con- for the oldest Palaeoproterozoic felsic rocks in cordia age of 2501 ± 5 Ma (Fig. 9e, Appendix 3). A Central Lapland. These include rocks that are few analyses, mostly conducted on high U-domains, roughly coeval with the Koitelainen intrusion, but gave younger age indications, which explain well also rocks that are clearly older, with an age of ca. the 2.45 Ga age obtained with the multigrain TIMS 2.50 Ga (Appendix 3). The latter are represented method (Nironen & Mänttäri 2003). The initial by the Sadinoja volcanic breccia, studied earlier εNd(T) value for sample A157 is –1.0 (Appendix 6). by Manninen et al. (2001). It is located on the NW We also used LA-MC-ICPMS to analyse zircon side of the Archaean Tojottamanselkä dome in the from the Sakiamaa felsic rock A1432 located east of Koitelainen area (Fig. 5). Most of the LA-MC-ICPMS the Koitelainen intrusion (Räsänen & Huhma 2001). analyses of zircon from sample A206-Sadinoja The discordant TIMS data published by Räsänen & were concordant and yielded an age of 2505 ± 5 Ma Huhma (2001) suggested an age of ca. 2.44 Ga, but (Fig. 9a). Three grains are Archaean and well explain the concordant data acquired by MC-ICP-MS gave the heterogeneity in the conventional TIMS data by a younger age of 2411 ± 8 Ma (Fig. 9f, Appendix 3).

Manninen et al. (2001). Sm–Nd analysis of another The initial εNd(T) values for the Sakiamaa rocks are whole rock (A685) gave an initial εNd(T) value of very negative (average –7, Appendix 6), suggesting –3.2, which suggests a significant contribution a large contribution of very old crustal material. of older lithosphere in the genesis of these rocks Based on the recent data obtained using (Appendix 6). MC-ICP-MS, we may conclude that the rocks that The sample Yläliesijoki A659 was taken one kilo- have traditionally been assigned to the Rookkiaapa metre NW of the sample site Sadinoja A206 from a Formation show a span in their ages. The younger felsic tuff in a strongly tectonised zone close to the rocks are roughly coeval with the 2.44 Ga layered western contact of the Koitelainen layered intrusion intrusions, but the formation also contains 2.50 Ga (Manninen et al. 2001). The LA-MC-ICPMS data on felsic rocks that clearly predate these intrusions. zircon yielded an age of 2426 ± 6 Ma, confirming the previous age of 2438 ± 8 Ma obtained from dis- 3.2.3 Peuratunturi and Koulumaoiva intrusions cordant TIMS data (Manninen et al. 2001, Fig. 9b). Another felsic rock apparently coeval with the The Peuratunturi and Koulumaoiva mafic intrusions 2.44 Ga layered intrusions is represented by the are situated in the Archaean Tuntsa suite (Fig. 2). Akanvaara rhyolite A1524. The old TIMS data on On the aeromagnetic map, the Koulumaoiva intru- zircon are discordant and slightly heterogeneous sion is located within an anomaly area of ca. 2.5 km (Räsänen & Huhma 2001), whereas most of the U– in length and 0.5 km in width. The intrusion is not Pb data obtained by LA-MC-ICPMS are concordant exposed but is covered by about 30 m of soil. Based and yield an age of 2434 ± 8 Ma (Fig. 9c, Appendix on drill core data, the intrusion is mainly composed 3). A few analyses yielded clearly younger age of olivine cumulates, which contain some narrow indications, explaining the slightly heterogeneous layers of gabbro. The olivine cumulates locally pre- multigrain TIMS results. A concordant analysis giv- serve the primary magmatic minerals unaltered. ing an age of 2441 ± 2 Ma was also obtained using One such olivine cumulate (A1475) was chosen for chemical abrasion and TIMS (Appendix 5). Sm–Nd mineral separation. The main minerals in this rock analysis of whole rock gave an initial εNd(T) value are clinopyroxene, orthopyroxene and olivine, with of –2.3, which is most typical for rocks related to plagioclase occurring as an intercumulus phase. the 2.44 Ga magmatism (Appendix 6). The Peuratunturi intrusion is located ca. 20 km Support for the existence of slightly older, 2.50 Ga NE of the Koulumaoiva intrusion. On the low-alti- crust in the surroundings of the 2.44 Ga Koitelainen tude aeromagnetic map, this intrusion is associated intrusion is provided by new LA-MC-ICPMS data with a weak magnetic anomaly ca. 1 km in length on zircon from two felsic rocks NW of the intru- and 0.3 km in width. It consists of olivine-bearing sion, occurring in an area traditionally called the gabbros, which are characterised by surprisingly Vuotso complex. A TIMS age of ca. 2.49 Ga obtained well-preserved primary magmatic minerals. Several by Manninen et al. (2001) for the Kaunismännikkö similar small gabbro intrusions have been located gneiss A157 can now be refined to 2506 ± 6 Ma in the area between Peuratunturi and the Russian (Fig. 9d, Appendix 3). Another sample (A1671) was border. From the Peuratunturi intrusion, an olivine- obtained from Porttipahta, for which the bulk of bearing gabbro (A1474) was sampled for isotopic

19 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A206 Sadinoja breccia data-point error ellipses are 2s 0.75 0.60 Concordia Age = 2505 ± 5 Ma, n=27 A659 Yläliesijoki felsic tuff Concordia Age = 2426 ± 6 Ma (n=20) 0.56 0.65 3100

0.52

2900 U 2600 206 Pb 0.55 238 0.48 2500 238 2700 U Pb/ 2400

2500 206 0.44 0.45 2300 2300 TIMS Intercepts at 0.40 371 ± 94 & 2438 ± 8 Ma MSWD = 1.5 (n=3) X - A685_TIMS A) 0.35 B) 7 9 11 13 15 17 19 21 0.36 207Pb/235U 8 9 10 11 12 13 207Pb/235U data-point error ellipses are 2s A1524 Akanvaara rhyolite data-point error ellipses are 2s LA-MC-ICPMS 0.56 0.62 Concordia Age = 2434 ± 8 Ma (20/24) A157 Kaunismännikkö felsic gneiss Concordia Age = 2506 ± 6 Ma (n=20) 0.52 0.58 CA-TIMS concordant 2600 2441 ± 2 Ma 0.48 2800 0.54 U 2400 2700

238 0.44 206Pb 0.50 2600

Pb/ 2200 2500 0.40 238 U 0.46 206 2400 2000 0.36 2300 0.42 2200 0.32 TIMS upper intercept age 0.38 ca. 2.49 Ga (n=4/9) C) 0.28 5 7 9 11 13 D) 0.34 207 235 TIMS - diamonds Pb/ U grain 28: length 150µm 7 9 11 13 15 207Pb/235U

data-point error ellipses are 2s A1432 Sakiamaa felsic volcanic rock 0.54 data-point error ellipses are 2s A1671 Porttipahta gneiss 0.55 2501 5 Ma (n=26) LA-MC-ICPMS 0.50 Concordia Age = 2411 ± 8 Ma 2550 n=17 (/19) 2600

2450 U 2400 U 0.46 0.45 238 238 2350 2200

S-13-51 (20140331) Pb/ Pb/

0.42 2250 Concordia Age = 206 2000

206 2500 8 Ma n=14/16 0.35

S13-57 (20140306) TIMS Intercepts at 0.38 TIMS Intercepts at Concordia Age = 584 ± 58 & 2438 ± 11 Ma 445 130 & 2451 13 Ma 2501 6 Ma MSWD = 2.7 (4/5) A416 TIMS n=12/16 MSWD = 3.6 n=5 0.25 F) E) 5 7 9 11 13 0.34 207 235 7.5 8.5 9.5 10.5 11.5 12.5 Pb/ U 207Pb/235U two mounts, two sessions

Fig. 9. Concordia plot of U–Pb zircon data for 2.5–2.4 Ga felsic rocks in Central Lapland. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots. A) Sadinoja volcanic breccia A206. The three TIMS analyses of a similar rock sample, A685 (Manninen et al. 2001), are also diplayed (as x). B) Yläliesijoki felsic rock A659. C) Akanvaara rhyolite A1524. D) Kaunismännikkö felsic rock A157. E) Porttipahta gneiss A1671. An old TIMS analysis of roughly similar gneiss A416 (Manninen et al. 2001) is also presented. F) Sakiamaa felsic volcanic rock A1432.

20 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

studies. The main minerals in this rock are pla- NW–SE-trending intrusion has a length of ca. gioclase, clinopyroxene, orthopyroxene and olivine. 5 km and a width of approximately 0.5 km. It is dif- The obtained mineral separates were mostly clean. ferentiated, with dunites (serpentinites) occurring However, plagioclase contained dark pigment, and at the assumed bottom of the intrusion and grad- in standard Franz isodynamic separation, this mining into metapyroxenites, and these further into eral was collected from the magnetic fraction. microgabbros and metagabbros. Locally, xenoliths The five Sm–Nd analyses conducted on whole of quartzite and gneiss occur within the roof-part rock and mineral separates from the Peuratunturi microgabbros (Palmen 1997, Mutanen 2002). gabbro defined an isochron with an age of 2448 ± Five samples were collected from the lower part

30 Ma and εNd = –1.5 (MSWD = 1.8, Fig. 10, Appendix of the intrusion for dating purposes. These included 1). A similar set of minerals and whole rock ana- two from xenoliths (quartzite and gneiss), one from lysed from the Koulumaoiva intrusion yielded an the gabbroic host of the xenoliths, one from a gab-

age of 2464 ± 34 Ma (εNd = –1.0, MSWD = 1.7). broic contact variety and one from an ultramafic These data confirm that both the Koulumaoiva and cumulate. All these samples yielded by separa- Peuratunturi intrusions are members of the 2.44– tion zircon grains, which were analysed using the

2.5 Ga age group. The slightly negative initial εNd NORDSIM facility in Stockholm (Appendix 4a). The and low Sm/Nd ratio in whole-rock samples are U–Pb data on the xenolith samples give Archaean also consistent with the general characteristics of ages, and the data on their gabbroic host rock magmatic rocks of the 2.44 Ga events. and contact variety also mostly indicate Archaean xenocrystic grains. In addition, several of the stud- 3.2.4 Lehtomaa intrusion ied zircon grains have domains that yield ages of ca. 1.8 Ga, suggesting strong hydrothermal/meta- The volcanic rocks of the Salla Group are cut by morphic effects at this time. With regard to dat- a mafic–ultramafic intrusion at Lehtomaa in the ing the magmatic crystallization of the intrusion, Salla area (Manninen 1991, Fig. 2). The dyke-like, the few zircon grains obtained from the ultramafic

Tuntsa belt intrusions

0.5126 A1432 Sakiamaa felsic volcanic rock data-point error ellipses are 2s A1475cpx A1474 Peuratunturi A1474cpx 0.5122 A1474opx 0.55 A1475opx LA-MC-ICPMS Age = 2448 ± 30 Ma Concordia Age = 2411 ± 8 Ma eps = -1.5 n=17 (/19) 2600

0.5118 MSWD = 1.8 n=5 U 2400 0.45

238 143 Nd A1474 wr#2 A1475 wr 2200 0.5114 Pb/ A1474wr 144

206 2000 Nd 0.35 0.5110 TIMS Intercepts at A1475 Koulumaoiva 584 ± 58 & 2438 ± 11 Ma Age = 2464 ± 34 Ma MSWD = 2.7 (4/5) A1474plag A1475plag eps = -1.0 0.25 F) 0.5106 5 7 9 11 13 MSWD = 1.7 n=4 207Pb/235U 0.5102 0.05 0.07 0.09 0.11 0.13 0.15 0.17 0.19 0.21 147Sm/144Nd

Fig. 10. Sm–Nd isotope data for whole-rock samples and mineral separates from the Peuratunturi (A1474) and Koulumaoiva (A1475) intrusions.

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.56 n1486 Lehtomaa gabbro, zircons from pyroxene cumulate

0.52 Intercepts at 2416 ± 12 & 403 ± 490 Ma 2600 MSWD = 2.2; n=10 0.48

0.49 U 2400 238 0.44 0.47 Pb/ 0.45 206 2200 0.40 0.43

0.41 2000 0.36 8.8 9.2 9.6 10.0 10.4 dark brown, small stubby Concordia Age = 2424 ± 5 Ma; n=7 zircon (2s, decay-const. errs ignored) 0.32 5 7 9 11 13 207Pb/235U

Fig. 11. Concordia plot of U–Pb SIMS data from zircon in the Lehtomaa gabbro A1525.

cumulate sample (n1486 = A1525) provide the most comprises medium-grained hornblende metagab- likely answer. Seven spot analyses yielded concord- bro. The discordant and slightly scattered U–Pb data ant U–Pb compositions indicating an age of 2424 obtained by TIMS on multigrain zircon fractions ± 5 Ma (Fig. 11, Appendix 4a). This result is also from gabbro sample A1405 yielded a rough age esti- supported by a few discordant data points from the mate of 2383 ± 33 Ma (Manninen & Huhma 2001). contact gabbro. This age should provide a minimum The quality of the zircon grains suggests that the age for the Salla Group volcanic rocks hosting the uncertainty in the result is due to alteration rather Lehtomaa intrusion. than inheritance. This has recently been confirmed Sm–Nd analysis has been performed on sample by LA-MC-ICPMS data, which scatter along the A1525 from a gabbroic rock near the basal contact concordia between 2.0 and 2.4 Ga (Fig. 12, Appendix of the Lehtomaa intrusion. The analysed rock shows 2). There is a clear negative correlation between the 207 206 strong LREE enrichment and yielded an εNd(2424 Ma) Pb/ Pb age and U content, suggesting that the value of -3.6 (Appendix 1), suggesting together with data on high-U domains are affected by radiation the zircon data that the magma solidified in the damage and lead loss and do not register primary Lehtomaa intrusion contained abundant Archaean igneous crystallisation ages (Fig. 12b). A TIMS U–Pb crustal material. This is consistent with Nd isotope analysis of zircon subjected to a chemical abrasion data obtained in general for volcanic rocks of the treatment gave a concordant composition at 2403 2.44 Ga age group (e.g., Hanski & Huhma 2005). ± 3 Ma, which may be considered a minimum age for the intrusion of the Onkamonlehto dyke (Appendix 3.2.5 Onkamonlehto dyke 5) and also a minimum age for the Kuusamo Group volcanic rocks. The NE–SW-trending Onkamonlehto dyke, at least Gabbroic sample A1405, which was originally 12 km long and up to 150 m wide, intrudes metavol- taken for U–Pb studies, has also been analysed for canic rocks of both the Salla and Kuusamo Groups in Sm–Nd isotopes (Hanski & Huhma 2005). The anal- the Salla area (Fig. 2). No primary magmatic miner- ysis revealed that the rock has a high concentration als are preserved in the dyke, which has recrystal- of REE and a LREE-enriched chondrite-normalised lised in greenschist facies metamorphic conditions. pattern (Appendix 2). The data yield an εNd(2403 Ma) The lowermost part of the distinctly differentiated value of –0.7, which is which within the range gen- mafic body consists of a bright green, hornblend- erally observed for 2.4–2.44 Ga mafic rocks. ite-like metacumulate layer, while the middle part

Geological Survey of Finland, Bulletin 405 Sm–Nd andA1405 U–Pb isotope Onkamonlehto geochemistry of the Palaeoproterozoic mafic dyke mafic magmatism in eastern and northern Finland

data-point error ellipses are 2s 0.56A1405 Onkamonlehto mafic dyke A1405H CA-TIMS data-point error ellipses are 2s 0.560.52 Concordia Age = 2403 2 Ma A1405H CA-TIMS 2600 0.520.48 Concordia Age = 2403 2 Ma 2600 206Pb 2400 0.480.44 238 206 U 2400 Pb 2200 0.440.40 238U Concordia Age = 2395 8 Ma 2200 LA-MC-ICPMS data 0.40 2000 with U<770 ppm (n=10) 0.36 TIMS Concordia Age = 2395 8 Ma MSWD (of concordance) = 1.7 LA-MC-ICPMS data 2000 with U<770 ppm (n=10) A) 0.360.32 TIMS 5 7 MSWD9 (of concordance)11 = 1.7 13 207Pb/235U A) 0.32 5 7 9 11 13 207Pb/235U

A1405-26a A1405-5a A1405-22a A1405-2a 2400 A1405-8a A1405-9a A1405-25a A1405-23a A1405-33aA1405-12a A1405-27a A1405-21a A1405-30a A1405-26a A1405-31a A1405-36a A1405-28a A1405-22a A1405-5a A1405-15a A1405-9a A1405-25a A1405-2a 2400 A1405-8a A1405-17a A1405-19a A1405-23a A1405-33aA1405A1405-12a-32a A1405-27a A1405-21a A1405-30a A1405-10a A1405-31a A1405-A140536a -24aA1405-28a A1405-13a 2300 A1405A1405-7a-15a A1405-17a A1405-19a A1405-32a A1405-10a A1405A1405-1a -16a A1405-24a A1405-13a 2300 A1405-7a A1405-11a Pb age Pb A1405-20a A1405-4a

206 A1405-16a 2200 A1405A1405-1a -18a A1405-29a A1405-11a Pb age Pb A1405-20a A1405-4a Pb/ A1405-6a 206 2200 A1405-18a A1405-29a A1405-3a A1405-35a 207

Pb/ 2100 A1405-6a A1405-3a A1405-35aA1405-34a 207 2100 A1405-14a A1405-34a B) 2000 A1405-14a 400 600 800 1000 1200 1400 1600 B) 2000 U (ppm) 400 600 800 1000 1200 1400 1600 Fig. 12. A) Concordia diagram of the zircon analysesU (ppm) from the Onkamonlehto dyke A1405 (error ellipse – LA-MC- ICPMS, dot – TIMS (Manninen & Huhma 2001, this study). B) Pb–Pb age vs. U concentration in zircon.

23 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

3.3 The 2.22 Ga Palovaara intrusion

One of the distinct phases of the Palaeoproterozoic A previously unknown occurrence of 2.2 Ga sills mafic magmatism in Finland is represented by the was recently revealed by LA-MC-ICPMS analy- ca. 2.22 Ga layered sills, which are widely present in ses conducted on zircon grains from an old sam- eastern and northern Finland, concentrated along ple, A136 Palovaara, from an intrusion located a the basal contacts of the Karelian schist belt (Vuollo few kilometres south of the Tshokkoaivi fell in & Huhma 2005, Hanski et al. 2010). Abundant iso- Enontekiö, NW Finnish Lapland (Fig. 1), where the tope data on these rocks were published by Hanski intrusion occurs as a sill-like body within mafic et al. (2010), who reported for them an average volcanic rocks. Eight concordant analyses of the

εNd(2220 Ma) value of 0.6. best-quality zircon domains yielded an age of 2213 A136 Palovaara diabase data-point error ellipses are 2s 0.52 A136 Palovaara diabase Average Pb/Pb age data-point error ellipses are 2s 0.520.48 2213 14 Ma AverageMSWD =Pb/Pb 0.1 n=8 age 0.44 0.48 2213 14 Ma 2300 MSWD = 0.1 n=8

U 0.40 0.44 2100 2300 238

U 0.400.36 1900 2100 Pb/ 238 0.360.32 206 1700 1900 Concordia Age = 2219 13 Ma

Pb/ n=8 0.320.28 206 1500 1700 ConcordiaInterceptsAge = 2219at 13 Ma 394 230 & n=82208 15 Ma 0.280.24 MSWD = 0.16 n=10 1300 1500TIMS Intercepts at 394 230 & 2208 15 Ma A) 0.240.20 2 4 6 MSWD = 0.168 n=10 10 1300 TIMS 207 235 A) 0.20 Pb/ U 2 4 6 8 10 207Pb/235U

A136 Palovaara diabase data-point error ellipses are 2s 0.6 A136 Palovaara diabase data-point error ellipses are 2s 0.6 0.5

0.5 2200 0.4 U 2200 238 0.4 1800

U 0.3 Pb/

238 1400 1800 0.3 Baddeleyite 206 0.2 Intercepts at Pb/ 1000 1400 -72 99Baddeleyite & 2201 7 Ma 206 0.2 MSWD = 2.9 n=11 0.1 6001000 Intercepts at -72 99 & 2201 7 Ma 200 MSWD = 2.9 n=11 0.1 600 0.0 B) 0 2 4 6 8 10 200 207 235 0.0 Pb/ U B) 0 2 4 6 8 10 Fig. 13. A) Concordia plot of U–Pb zircon data from207 sample235 A136 Palovaara. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses as red dots. B)Pb/ ConcordiaU plot of U–Pb data on baddeleyite from sample A136 Palovaara.

24 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

± 14 Ma, whereas analyses of altered zircon domains Baddeleyite from sample A136 has subsequently yielded discordant data with much younger Pb–Pb been analysed using the in-house (GTK) baddeley- ages (e.g., 2a in Fig. 13a). Abundant altered zircon ite standard A974, for which a precise TIMS age of explains the discordance of the previous TIMS data 1256.2 ± 1.4 has been determined by Söderlund et al. obtained by O. Kouvo. The sample also contains (2004). These data are closer to the concordia and baddeleyite grains. Analysis of these using a zircon yield an average Pb–Pb age of 2204 ± 6 Ma (Fig. standard produced strongly reversely discordant 13b, Appendix 2). data, but with Pb–Pb ages that were nevertheless close to the 2.2 Ga indicated by the apparently well- preserved zircon domains (Appendix 2).

3.4 The 2.15 Ga intrusions

3.4.1 Rantavaara intrusion that were used for U–Pb zircon dating are included,

the result is 2233 ± 25 Ma (εNd = +3.2, MSWD = 0.95, The Rantavaara intrusion is a ca. 20-km-long (pos- Fig. 14). It should be noted that this Sm–Nd age is sibly 30 km), up to 1-km-thick, steeply dipping not consistent with the U–Pb zircon age of 2148 differentiated mafic body in the Sodankylä area in ± 11 Ma reported for the Rantavaara intrusion by central Lapland (Räsänen & Huhma 2001, Fig. 4). On Räsänen & Huhma (2001). To resolve this inconsist- the stratigraphic map of central Lapland, this intru- ency, another gabbroic sample (A1586) was taken sion is located in an area between zones occupied for U–Pb zircon studies. One U–Pb TIMS analysis by schists of the Sodankylä and Savukoski Groups performed on a fraction of clear zircon grains from (Fig. 4), but it is not clear whether the magmatic this sample plots exactly on the chord defined by body has an intrusive relationship with the black the analyses of sample A900 in Räsänen & Huhma schists of the Savukoski Group. A U–Pb age of 2148 (2001), suggesting that the published U–Pb age ± 11 Ma has earlier been reported for zircon grains of 2148 ± 11 Ma gives a better idea of the primary extracted from a coarse-grained gabbroic sample of crystallisation age of the Rantavaara intrusion. the intrusion (Räsänen & Huhma 2001). Because there is only a slight difference in Sm/ For Sm–Nd studies, a sample (A1337) of an ultra- Nd between pyroxene separates and whole-rock mafic cumulate zone was collected from the lower samples, the obtained Sm–Nd age is largely based part of the intrusion. The primary magmatic min- on plagioclase. Rejecting all data on plagioclase, an erals, i.e., pyroxene, olivine and minor plagioclase, age of 2168 ± 120 Ma (εNd = +3.2, MSWD = 0.74, are relatively well preserved in this sample. The n = 5) can be calculated. The results suggest that pyroxene occurs as large poikilitic crystals enclosing the plagioclase analysed from the peridotite sam- olivine grains. The amount of plagioclase is small ple A1337 had a lower initial 143Nd/144Nd ratio than and it generally occurs at pyroxene grain bounda- the pyroxene separates and whole-rock samples. ries. The mineral fractions used for Sm–Nd analyses This difference could be explained, for example, by were fresh and clean. minor contamination introduced into intercumu- Over the years, a total of ten Sm–Nd analyses have lus melt by externally derived fluids during the late been conducted on samples from the Rantavaara stages of crystallisation of the intrusion. intrusion (Appendix 1). Some technical problems The initial εNd of value +3.2 obtained from the were involved in the older mineral analyses, and pyroxene and whole-rock analyses is close to con- several duplicate analyses of plagioclase and pyrox- temporaneous depleted mantle values and clearly ene were therefore performed. Based on the evalu- distinct from the values obtained for most other ation of these data as a whole, the old analyses of mafic intrusions during this work. The indicated pyroxene with large errors are of questionable value high initial εNd value is supported by the initial and are rejected. In general, the reproducibility of εHf value of +9.8 reported for zircon from sample the analyses is, however, acceptable, as can be seen A900 by Patchett et al. (1981). The Sm/Nd ratios from the duplicate data on plagioclase. The seven for whole-rock samples indicate that the magma mineral analyses of sample A1337 provide an isoch- that solidified in the Rantamaa intrusion was char- ron with an age of 2236 ± 27 Ma (εNd = +3.3, MSWD acterised by a near flat chondrite-normalised REE = 1.0). If whole-rock analyses of the two samples pattern.

25 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.5134 Rantavaara intrusion Age = 2168 ± 120 Ma 0.5130 eps = +3.2 A1337 px#2 MSWD = 0.74 n=5 A1337 px#3 (plag excluded) A1586 A1337

Nd 0.5126 A900 144 Nd/ Age = 2233 ± 25 Ma 143 0.5122 eps = +3.2 MSWD = 0.95 n=9

A1337 plag 0.5118 A1337plag#2 A1337 plag#3 A1337 plag#4 Cf: U-Pb zircon age = 2148±11 Ma 0.5114 0.10 0.12 0.14 0.16 0.18 0.20 0.22 147Sm/144Nd

Fig. 14. Sm–Nd isotope data for whole-rock samples and mineral separates from the Rantavaara intrusion.

3.4.2 Tanhua intrusions gabbro, A1446, below the granophyre. The granophyre sample A1674 yielded abundant zircon, which The Kannusvaara gabbro is one of the mafic intru- occurs as long, euhedral, simple prisms with sharp sive bodies in the Tanhua area, eastern Lapland edges. Zircon grains are pale and transparent, and (Figs. 2, 4). The general geology of the Kannusvaara the population appears very homogeneous. Four intrusion has been described by Mattila (1974). multigrain TIMS U–Pb analyses were carried out Some geochemical features from a drill core from on zircon (Appendix 5), defining a chord that has the eastern (upper) part of the intrusion are pre- intercepts with the concordia curve at 2116 ± 10 sented and discussed by Mutanen (1997). The intru- and 300 ± 65 Ma (Fig. 15). Recent analyses using sion runs approximately north–south and is 1.7 km LA-MC-ICPMS suggest a slightly older age, as most (or possibly up to 3 km) long and up to 0.6 km wide. data yield a concordia age of 2148 ± 7 Ma (Fig. 15, The dip in the north appears to be steep to the west, Appendix 2). Two analyses are distinctly on the but in the south, on the best exposed Kannusvaara younger side, and the existence of such (metamor- hill, the intrusion has been twisted into an over- phic?) domains could explain the TIMS results. turned position, so that the diamond drill hole, col- Zircon from two other gabbroic samples from lared in magnetite gabbros, intersected granophyres the Tanhua area was also analysed using laser abla- deeper down (see Fig. 32 in Mutanen 1997). The tion MC-ICP-MS, one from the Kylälampi gabbro gabbros, originally plagioclase-pyroxene-magnet- and the other from the Koskenkangas gabbro. The ite cumulates, were later partly or wholly uralitised. previous TIMS results for these samples (A418 and The granophyre member consists of a lower biotite A817, Fig. 2) are discordant, suggesting ages of 2.11– granophyre unit and an upper hornblende grano- 2.14 Ga (Räsänen & Huhma 2001). All data on the phyre unit. The contact rocks (immediately east of Kylälampi gabbro sample A817 (except 5d & 5e) have the granophyre) are pyroxene-bearing metasedi- low U and give similar, slightly discordant Pb–U mentary hornfelses (Mattila 1974). results (Fig. 16, Appendix 2). The discordance could Two samples were chosen for isotopic studies. be due to a matrix effect in “albitite” zircon, which These were a granophyre, A1674, on top of the is different from the used standard. The Pb–Pb age internal stratigraphy and a magnetite-bearing of 2137 ± 5 Ma can be considered as the best age

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1674 Kannusvaara granophyre data-point error ellipses are 2s 0.48 LA-MC-ICPMS Concordia Age = 2148 7 Ma 0.44 n=18 (/20) 2300

U 2200

238 0.40 2100

Pb/ 2000 0.36 206 1900 TIMS Intercepts at 0.32 300 65 & 2116 10 Ma n=4

0.28 5 6 7 8 9 207Pb/235U

Fig. 15. Concordia plot of U–Pb zircon data from the Kannusvaara granophyres, A1674. LA-MC-ICPMS analyses shown as error ellipsoids and ID-TIMS analyses as black dots.

estimate of the magmatic zircon. Two points from although some grains are slightly yellowish. The distinctly altered domains show high U concentra- five Sm–Nd analyses performed on whole rock, tions (5d & 5e) and provide a slightly younger age pyroxene and plagioclase defined an age of 2089 of 2.10 Ga, which would explain the TIMS result. ± 33 Ma (εNd = +2.0, MSWD = 1, Fig. 18, Appendix All LA-MC-ICPMS data on zircon from the 1). As the whole-rock sample has a Sm/Nd ratio Koskenkangas gabbro sample A418 tend to plot close to that of pyroxene, the age is largely based above the concordia curve and give a Pb–Pb age on the analysis of plagioclase. It has become clear of 2110 ± 8 Ma (Appendix 2, Fig. 17). The U and Pb in this work that there are occasionally problems concentrations are extremely high, which is likely in using plagioclase for dating primary magmatic to cause problems in calibration. In fact, there is crystallisation. These problems are mostly related a slight negative correlation between the Pb con- to metamorphic alteration, which may also be the centration and Pb–Pb age (diagram in Appendix 2). case here. The concentrations in the TIMS data were much The Sm/Nd ratio for the magnetite gabbro (A1446) lower, but cannot be directly compared, since they is chondritic and the initial εNd value clearly posi- were obtained from heavier zircon. The minimum tive, suggesting derivation from depleted mantle age yielded by TIMS was 2134 ± 4 Ma (Räsänen & without major crustal contamination. In contrast, Huhma 2001). the Sm–Nd analysis from the granophyre sample

Because of its relatively well-preserved primary A1674 yielded a slightly negative εNd(2148 Ma) value magmatic mineralogy, a sample (A1446) from the of –1.0 (Appendix 1). The REE pattern for this sam- Kannusvaara intrusion was selected for Sm–Nd ple is, however, quite strange for a felsic rock, as the isotopic studies. The plagioclase separate is char- Sm/Nd ratio is higher than in chondrites. acterised by a dark pigment, but is mostly clear,

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A817 Kylälampi gabbro (albitite) data-point error ellipses are 2s 2200 0.40

Intercepts at 2100 0.38 0 0 & 2137 5 Ma MSWD = 0.82 n=14 (/16) 2000 0.36 U 1900 238 0.34

Pb/ 1800 0.32 206 1700 0.30 TIMS Intercepts at 1600 0.28 232 97 & 2114 6 Ma MSWD = 2.6 n=6 1500 0.26 3 4 5 6 7 8 207Pb/235U Fig. 16. Concordia plot of U–Pb zircon data from the Kylälampi gabbro A817. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses as black dots.

data-point error ellipses are 2s 0.49 A418 Koskenkangas gabbro

0.47 LA-MC-ICPMS Average Pb/Pb age 0.45 2110 8 Ma very high U& Pb - calibration?? 0.43 206 2280 Pb 2240 238U 0.41 2200 2160 0.39 2120 2080 TIMS_A418A TIMS A418A 2040 0.37 Pb/Pb age 2134 4 Ma

0.35 6.2 6.6 7.0 7.4 7.8 8.2 8.6 207Pb/235U Fig. 17. Concordia plot of U–Pb zircon data from gabbro A418. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses as black dots.

28 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5132 A1446 Kannusvaara A1446px Age = 2089 ± 33 Ma A1446px#2 0.5128 eps = +2.0 A1446 wr MSWD = 1.06 n=5

Nd 0.5124 144 Nd/

143 0.5120

A1446plag#2 0.5116 A1446plag

Cf: Granophyre U-Pb zircon age 2148 ± 7 Ma

0.5112 0.09 0.11 0.13 0.15 0.17 0.19 0.21 0.23 147Sm/144Nd

Fig. 18. Sm–Nd isotope data for whole rock and mineral separates from the Kannusvaara intrusion.

3.5 The 2.05 Ga intrusions

3.5.1 The Kevitsa intrusion ern margin and still 30 m thick near the eastern border. Using zircon extracted from an ultramafic The Kevitsa (also known as Keivitsa) mafic-ultra- cumulate and the U–Pb TIMS method, Mutanen & mafic intrusion is located within pelitic metasedi- Huhma (2001) were able to date the intrusion at ments and komatiitic metavolcanic rocks of the 2058 ± 4 Ma. Savukoski Group in Central Lapland, only ca. 1 A large disseminated Cu-Ni-PGE sulphide deposit km south of the large 2.44 Ga Koitelainen layered (Keivitsansarvi), which is currently under exploita- intrusion (Fig. 4). The first detailed account of the tion, is located in the middle part of the ultramafic geology of the Kevitsa intrusion was reported by zone. This part of the intrusion is also characterised Mutanen (1997), and a summary of the geology by the presence of ultramafic (including dunitic) and exploration history was recently published by and pelitic hornfelsed xenoliths. Two geochemi- Santaguida et al. (2015). On the surface, the intru- cally distinct, spatially closely associated ore types sion has a roundish shape and an area of ca. 18 km2. can be discerned (Mutanen 1997, Yang et al. 2013): The estimated maximum thickness is more than the prevalent regular ore (Cu–Ni–PGE–Au type) 1.5 km. The lower part of the layered succession with Ni and Cu tenors of 4–7 wt% and 5–12 wt%, consists of ultramafic olivine-clinopyroxene- respectively, and the less abundant sulphur- and orthopyroxene cumulates (olivine websterites and Cu-poor Ni–PGE ore with extremely high Ni tenors olivine clinopyroxenites), with a rather constant of up to ca. 40 wt%. There also exists some ore that original modal olivine content (15–25%). The over- is transitional between these two types. In addi- lying gabbroic cumulates include ferrogabbros, tion, Mutanen (1997) recognised an uneconomic graphite-bearing gabbros and magnetite gabbros, variety with equal contents of sulphur compared to which all typically contain inverted pigeonite. The the mentioned ore types, but with very poor grades upper gabbroic cumulates are topped by a layer of of Ni, Cu and PGE, referring to it as “false ore”. magnetite-bearing and sulphide-rich sodic grano- Apart from being different in their chalcophile ele- phyre; this layer is ca. 0.5 km wide along the south- ment contents, the ore types have their own litho-

29 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye phile trace element characteristics, with the most material from the intrusion, Sm–Nd analyses have remarkable of them being the high LREE contents been performed on country rock samples and two of the Ni–PGE ore type (Mutanen 1997, Hanski et dykes cutting the intrusion. Data from the dykes are al. 1997). presented separately in the following chapter, 3.5.2. An important aspect of the Kevitsa intrusion is The results of five Sm–Nd analyses from a well- that the rocks are mostly rather fresh, contain- preserved gabbro sample, A1226, provide a mineral- ing primary igneous minerals (Ca-rich pyroxene, whole rock isochron with an age of 2019 ± 26 Ma orthopyroxene, olivine, plagioclase), which allow (εNd = –3.8, MSWD = 1.6, Fig. 19). The Sm/Nd ratio the use of the Sm–Nd method for dating and genetic and concentrations in the clinopyroxene separates studies. A Sm–Nd mineral age of 2052 ± 25 Ma has are close to those of the whole-rock sample. In con- earlier been published in some conference abstracts trast, pigeonitic pyroxene from another gabbroic (Huhma et al. 1995, 1996, Hanski et al. 1997), being sample, A1316, has a very low REE level and con- within error equal to the U–Pb zircon age reported siderably higher Sm/Nd. Six analyses of this sample by Mutanen & Huhma (2001). Previously, barren yielded an age of 2067 ± 22 Ma (εNd = –3.3, MSWD and ore-bearing samples from the Kevitsa intrusion = 0.19, Fig. 19). The old analysis of poorly purified have also been studied using other radiogenic and pyroxene yielded an unusually high concentration of stable isotope systems: Hanski et al. (1997) pub- Nd and also had a large error, and has been omitted lished preliminary Os isotope data and Grinenko from the age and εNd calculations. et al. (2003) documented extensive S and C isotope The third sample, A1390, on which several Sm– data. In this chapter, we focus on Sm–Nd isotope Nd analyses have been performed, represents barren results and discuss in detail their relationship with pyroxenite from the ultramafic main suite. The ana- other isotope and geochemical data in a separate lysed fractions include whole rock, clinopyroxene, paper (Hanski et al. in prep.). orthopyroxene and apatite, and define an isochron

Sm–Nd isotope data from Kevitsa are listed in with a date of 2018 ± 56 Ma (εNd = –3.4, MSWD = Appendix 1, consisting of 33 analyses conducted on 2.6, Fig. 19). The Sm/Nd ratio in orthopyroxene is 11 mafic and ultramafic samples. In addition to the unusually low. This peculiar sample also contains

Kevitsa intrusion A1316 px #3 A1316 px#6 A1390 Pyroxenite A1316 px#5 Age = 2018 ± 56 Ma A1316 px#4 0.5132 eps = -3.4 MSWD = 2.6 n=6

A1226 cpx#1

Nd A1316 Gabbro A1316

144 0.5122 A1390 cpx #1 & #2 2067 ± 22 Ma A1390 A1226 eps = -3.3 A1390 opx#2 Nd/ MSWD = 0.19 n=6

143 A1390 opx

0.5112 A1390 ap A1226 Gabbro A1316 plag Age = 2019 ± 26 Ma A1226 plag #2 A1226 plag eps = -3.8 MSWD = 1.6 n=5 0.5102 0.04 0.08 0.12 0.16 0.20 0.24 0.28 147Sm/144Nd

Fig. 19. Sm–Nd isotope data for three whole-rock samples and related mineral separates from the Kevitsa intrusion.

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1390 Kevitsa pyroxenite data-point error ellipses are 2s 0.39 LA-MCI-CPMS average Pb/Pb age 2080 2046 8 Ma n=8 2040 0.37 2000

1960 206 Pb 0.35 1920 238U TIMS Intercepts at 476 430 & 2057 7 Ma 0.33 MSWD = 2.6 n=6

TIMS average Pb/Pb age 2052 4 Ma (data: M & H 2001) 0.31 5.4 5.6 5.8 6.0 6.2 6.4 6.6 6.8 207Pb/235U

Fig. 20. Concordia plot of U–Pb zircon data from the Kevitsa pyroxenite A1390. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses by Mutanen & Huhma (2001) as black dots.

some high-U zircon, which was studied earlier by = 0.4, εNd = –6.4), which is supported by another Mutanen & Huhma (2001). It turned out to be nearly whole-rock sample (R713/36.6-) of the Ni–PGE concordant and yielded the above-mentioned U–Pb ore type (Appendix 1). The REE concentrations in age of 2058 ± 4 Ma. Zircon from the same sam- these Ni–PGE ore samples are very high and form ple was recently re-analysed by LA-MC-ICPMS a LREE-enriched chondrite-normalised REE pat- (Appendix 2). The data tend to be slightly discord- tern. In particular, the REE concentrations in the ant and provide an average Pb–Pb age of 2046 analysed pyroxene concentrate are extremely high ± 8 Ma (Fig. 20). and yield a rather unusual chondrite-normalised Some Sm–Nd analyses were also conducted on pattern (see Fig. 3 in Hanski et al. 1997). the peridotite samples A1319, HH/18.5B-92 and In summary, the average initial εNd value of the R337/18.5–18.9 (Appendix 1), which are all related main cumulate suite, related regular ore and “false to the regular ore type. All the obtained isotope ore” is –3.4 ± 0.3, whereas the Ni–PGE ore type data from these samples, combined with the data provides an initial εNd value of –6.6. The duplicate from the three other samples discussed above, analyses of a granophyre sample yielded εNd values plot roughly on a Sm–Nd isochron with an age of of –2.7 and –3.4. In order to be able to evaluate the

2049 ± 26 Ma (MSWD = 5.7; εNd = −3.4, n = 23, Fig. potential influence of the immediate country rocks 21). The Sm–Nd analyses on a granophyre sample on the isotope composition of the Kevitsa magma (A1380) and Fe-sulphide-rich “false ore” sample and ores, 6 Sm–Nd analyses were performed on (R688/34.25) also plot on the same line (Fig. 21). samples from the surrounding pelitic schists and

In contrast, the Sm–Nd data on the samples rep- hornfelses. They provided εNd(2050 Ma) values from resenting the Ni–PGE ore type clearly lie below the –3.7 to –6.8, thus falling between the initial εNd line and thus give lower initial εNd values; isotope values measured for the Kevitsa main suite and the data on the whole-rock sample R695/67.65-67.7 Ni–PGE ore type. and related pyroxene and plagioclase separates pro- The obtained strongly negative initial εNd val- vide a Sm–Nd age estimate of 2069 ± 31 Ma (MSWD ues are consistent with the radiogenic Os isotope

31 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Kevitsa mafic intrusion 0.514 Age = 2049 ± 26 Ma Epsilon = - 3.4 MSWD = 5.7 n=23 wr, px, plag, ap from pyroxene A1316

0.513 four samples/ main rock types

Nd false ore 144 micro gabbro 0.512 A1380 granophyre Nd/ A1445 felsic dyke 143 Keivitsa country rocks R695 R695cpx Ni-PGE ore (R695) 0.511 plagioclase, Age = 2069 ± 31 Ma apatite eps = - 6.4 R695plag MSWD = 0.40 n=3 0.510 0.0 0.1 0.2 0.3 147Sm/144Nd U-Pb zircon age for Kevitsa: 2058±4 Ma

Fig. 21. Sm–Nd isotope data for whole-rock samples and mineral separates from the Kevitsa intrusion and its country rocks.

compositions reported by Hanski et al. (1997) and the extremely non-radiogenic Nd isotope composi- deviate noticeably from the composition of the con- tion of this ore type, with its εNd(2050 Ma) values temporaneous convective upper mantle, implying being mostly lower than those of the metasediments an involvement of old material with high LREE/ and approaching those of Archaean felsic gneisses HREE in the genesis or evolution of the Kevitsa (Fig. 22). magma. A major problem is whether these geochemical features are related to crustal contamina- 3.5.2 Kevitsa dykes tion or inherited from heterogeneous subcontinental lithosphere. According to Mutanen (1989, 1997), an Several types of dykes have been found to cut the important contaminant was reduced carbonaceous Kevitsa intrusion, varying in composition from material from the surrounding black schists. This ultramafic and mafic to felsic (Mutanen 1997). contamination involved reduction of the redox state Sample A1445 (from drill core) represents a of the magma, with part of the carbonaceous mate- medium-grained dioritic dyke having sharp, wind- rial ultimately crystallising as cumulus grains of ing contacts to the surrounding olivine pyroxenite. graphite (Mutanen 1989). On the other hand, in their This sample has earlier yielded a U–Pb zircon age of model of ore genesis at Kevitsa, Yang et al. (2013) 2054 ± 5 Ma, indicating that the dyke is contempo- suggested that contamination with sulphur-bearing raneous with its host intrusion (Mutanen & Huhma country rock material already happened before the 2001). The Sm–Nd isotopic results furnished by final emplacement of the Kevitsa magma in a more sample A1445 are given in Appendix 1 and plotted deep-seated magma chamber. In any case, the Nd in Figure 21 together with other Sm–Nd data from and other isotope data are consistent with a major the Kevitsa intrusion. As shown by the isochron dia- contribution of material from a source similar to gram, the isotope composition of the diorite dyke the surrounding S- and C-rich sedimentary rocks. lies very close to the line defined by the samples However, involvement of this material in the gen- from peridotite related to the regular ore type and esis of the Ni–PGE ore type does not fully explain yields an initial εNd value of –3.7.

32 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Kevitsa whole rocks

* DM

false ore granophyre diorite dyke

-5 main suite gabbros schists Ni-PGE ore & peridotites epsilon (2060)epsilon

- -10 Archean gneisses Nd

-15 0 10 20 30 40 Nd (ppm)

Fig. 22. εNd vs. Nd (ppm) diagram showing whole-rock data from the Kevitsa intrusion: main suite – red dot, false ore – green square, Ni-PGE ore – blue triangle, granophyre – red diamond, diorite dyke – diamond, and country rock schists – x. DM shows the composition of magma derived from the depleted mantle. Data on some Archaean gneisses in Finland are presented for reference (+, data from Huhma 1986, O’Brien et al. 1993, Hölttä et al. 2000, Hanski et al. 2001c, Mutanen & Huhma 2003).

0.5145 A1366 Kevitsa dyke Age = 1916 ± 67 Ma A1366px#2 0.5143 eps = +5.1 A1366px MSWD = 0.3 n=5

0.5141 Nd 144

Nd/ 0.5139

143 A1366plag#2

A1366#2 0.5137 A1366

0.5135 0.25 0.27 0.29 0.31 0.33 147Sm/144Nd Fig. 23. Sm–Nd isotope data for whole rock and mineral separates from the Kevitsa LREE-depleted dyke.

33 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Some narrow (ca. 4–5 m) dykes with olivine- strange for a mafic magmatic phase in northern rich ultramafic central parts display prominent Finland. In addition, the chemistry of the dyke flow differentiation. Mutanen (1997) called them shows similar features to those of the ca. 2050 Ma olivine gabbro diabases. These dykes have a chemi- Ti-enriched komatiites of the Savukoski Group in cal composition very distinct from their host ultra- Central Lapland (Hanski et al. 2001b). Assuming an mafic rocks, displaying strongly LREE-depleted age of ca. 2050 Ma for the dyke, individual samples chondrite-normalised REE patterns. One such would have initial εNd values of between +2.8 and ENE-trending dyke was sampled (A1366) for min- +3.9, which are consistent with the values measured eral separation and Sm–Nd isotope studies. Some for these komatiites. problems were encountered in the mineral separation, as plagioclase turned out to be slightly mag- 3.5.3 The Moskuvaara intrusion netic. The analysis of plagioclase also revealed a relatively high REE level and unusually high Sm/ The Moskuvaara intrusion is located ca. 10 km south Nd, in fact higher than in the whole-rock sample of Kevitsa, where it has intruded into black schists (Appendix 1). This composition is probably more of the Savukoski Group. Sample A1436 represents representative of the inclusions in plagioclase than the gabbroic portions of the intrusion. It yielded a the mineral itself. Nevertheless, five analyses con- small quantity of turbid small simple zircon prisms. ducted on this olivine-bearing gabbro dyke yielded The earlier multigrain TIMS data are discordant and an isochron age of 1916 ± 67 Ma (εNd = +5.1, Fig. heterogeneous, suggesting an upper intercept age

23). The highly positive initial εNd value suggests of ca. 2.0 Ga (Appendix 5, Fig. 24). The LA-MC- that the magma was derived from depleted mantle ICPMS data are also scattered and partly discord- without any contamination from the old enriched ant, with Pb–Pb ages of ca. 1.9–2.05 Ga (Appendix lithosphere. The obtained age seems to indicate 2). Rejecting compositions with elevated common that the dyke is clearly younger than the Kevitsa lead (206Pb/204Pb <3000), an age of 2039 ± 14 Ma intrusion. However, the age of ca. 1920 Ma is rather can be calculated (n = 15). The effects of alteration

A1436 Moskuvaara gabbro data-point error ellipses are 2s

0.44 A1436 Intercepts at 250 230 & 2039 14 Ma 2200 0.40 MSWD = 2.9 n=15 (/27) (data with 206Pb/204Pb>3000) 2000 0.36

1800 206Pb 0.32

238 1600 U 0.28

1400 0.24 TIMS Intercepts at 106 260 & 1996 48 Ma 0.20 MSWD = 22 n=4

0.16 2.5 3.5 4.5 5.5 6.5 7.5 207Pb/235U Fig. 24. Concordia plot of U–Pb zircon data from the Moskuvaara gabbro A1436. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses as black dots.

34 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland are obvious in the zircon Pb-U system, and the true good-quality standard zircon compared with the magmatic age is probably ~2.05 Ga, i.e., close to the turbid, high-U material of the sample (Fig. 25, age of the Kevitsa intrusion. Appendix 2). The Pb/Pb isotope ratio is less sensi- The Sm–Nd analysis of A1436 yielded an tive to this difference and, excluding one analy-

εNd(2040 Ma) value of –5.2. A clearly negative ini- sis, an average Pb–Pb age of 2035 ± 8 Ma can be tial εNd value of –4.2 has also been obtained for the calculated for zircon A2288. An analysis (13b) of a roughly coeval 2055 ± 5 Ma (Räsänen & Huhma clearly altered domain is discordant and suggests 2001) Rovasvaara gabbro further south (A820, the effects of a younger event. Appendix 1). The results from the other Puijärvi sample, A2289, are consistent with those of sample A2288, 3.5.4 The Puijärvi and Satovaara intrusions suggesting an age of 2028 ± 8 Ma. The data from these samples (and the Satovaara sample considered Recently, gabbros from the Puijärvi and Satovaara next) show a clear negative correlation between areas were selected for isotope studies by FQM the U concentration and apparent (Pb–Pb) age, FinnEx Ltd. The Puijärvi intrusion is one of the suggesting some secondary Pb loss (diagram in mafic-ultramafic bodies west of the large 2.44 Ga Appendix 2). Thus, the ages reported above may Koitelainen layered intrusion. Sedimentary and be slightly too young for magmatic crystallisation. volcanic rocks enclose the gabbro, from which two The Satovaara intrusion occurs close to the samples, A2288 and A2289, were processed for Kevitsa intrusion (Fig. 4). An extensive, NE–SW- mineral separation. Both samples yielded zircon trending shear zone separates the two intrusions that is turbid and shows domains of alteration in from each other, although the displacement along BSE images (Fig. 25). this structure is probably minor judging from The analyses by LA-MC-ICPMS revealed that the close similarity of the sedimentary and vol- zircon from A2288 has generally high U, and the canic rocks on both sides of the tectonic zone. The results tend to be slightly reversely discordant. This Satovaara intrusion is differentiated, with a perido- may well be due to difference between the clear, tite basal unit grading into an upper gabbroic unit.

A2288 Puijärvi gabbro data-point error ellipses are 2s

0.46 Average Pb/Pb age 2035 8 Ma 0.42 n=17 2200 U 2100

238 0.38 2000 Pb/ 1900 0.34 206 1800 1700 0.30

0.26 4 5 6 7 8 207Pb/235U Fig. 25. Concordia plot of U–Pb zircon data from the Puijärvi gabbro A2288.

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A2290 Satovaara gabbro data-point error ellipses are 2s

0.45

Average Pb/Pb age 0.43 2025 8 Ma n=10 2220 U 0.41 2180 238 2140 0.39 2100 Pb/ 2060 206 0.37 2020 1980 1940 0.35

0.33 5.6 6.0 6.4 6.8 7.2 7.6 8.0 207Pb/235U Fig. 26. Concordia plot of U–Pb zircon data from the Satovaara gabbro A2290.

For isotopic studies, a gabbroic sample (A2290) was centration also leaves room for speculation that the chosen from drill core SAT-007, where hydrother- real magmatic age may be slightly older (Fig. 26, mal alteration and serpentine overprinting of the Appendix 2). peridotite unit is not intensive. The initial εNd values for the Puijärvi and Satovaara Only a few zircon grains were found from the samples are close to zero, and thus distinct from sample. The U–Pb results are similar to those the Ni-ore-bearing Kevitsa intrusion (Appendix 1). obtained from the Puijärvi samples. The average The isotope results from the Puijärvi and Satovaara Pb–Pb age is 2025 ± 8 Ma, but in this case, the intrusions were earlier reported in a conference negative correlation between the date and U con- abstract (Peltonen et al. 2014).

3.6 The 2.0 Ga intrusions in Kittilä

Much of the Kittilä greenstone area represents (sheeted) (see Fig. 16 in Lehtonen et al. 1998). Mafic juvenile ca. 2.0 Ga Palaeoproterozoic oceanic crust pillow lavas in the vicinity are chemically related to with initial Nd isotope compositions close to that the mafic dykes. of coeval depleted mantle (Hanski & Huhma 2005). The mineral separation of sample A1563 yielded The age is based on U–Pb zircon data from felsic a small amount of anhedral, brown and fairly tur- porphyries and gabbroic rocks (Rastas et al. 2001) bid zircon, quite normal for gabbroic rocks. The and supported by Sm–Nd results from mafic vol- CA-TIMS U–Pb analysis of zircon gave a concordant canic rocks, including analyses of primary pyrox- composition and, together with a slightly discordant ene (Vesmajärvi Formation, Hanski & Huhma 2005). analysis of air-abraded zircon, provides an age of Further age constraints have been obtained from 2008 ± 3 Ma (Fig. 27, Appendix 5). Subsequently, the Selkäsenvuoma gabbroic dyke A1563 (Fig. 2). zircon was analysed using LA-MC-ICPMS, and This sample was collected from a large outcrop rejecting a few points with elevated common lead, area, where the gabbroic rock occurs as narrow a concordia age of 2002 ± 8 Ma (n = 21) can be screens between fine-grained, parallel mafic dykes calculated (Appendix 2). Thus, the zircon popula-

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1563 Selkäsenvuoma gabbro

data-point error ellipses are 2s

0.44 LA-MC-ICPMS Concordia Age 2002 8 Ma, n=21

0.40 2150 U

238 2050

0.36 Pb/ 1950 206 1850 TIMS A1563C 0.32 concordant 2008 3 Ma

0.28 4.8 5.2 5.6 6.0 6.4 6.8 7.2 7.6 8.0 207Pb/235U Fig. 27. Concordia plot of U–Pb zircon data from the Selkäsenvuoma gabbro A1563, Kittilä. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses as red dots.

tion appears to be homogeneous and the CA-TIMS The Tuulijoki gabbro (A1565) in the western result 2008 ± 3 Ma can be considered as the best Kittilä area (Fig. 2) represents a rock that contains age estimate for the gabbroic rocks and the whole a small amount of heavily altered zircon. Four of the mafic rock suite in the area. five U–Pb analyses by laser ablation MC-ICP-MS The new Sm–Nd analysis conducted on the yielded ages of ca. 1.79 Ga, which is considered whole-rock sample of the Selkäsenvuoma gab- to register a major metamorphic event (Fig. 31a, broic dyke (A1563) yielded an initial εNd value of +2.8 Appendix 2). One analysis plots above the concordia (Appendix 1), which is close to the isotope compo- curve, suggesting an age of ca. 2.0 Ga and provid- sition measured for most volcanic rocks from the ing an explanation for the TIMS analysis, which Kittilä Group (Hanski & Huhma 2005). resulted in a Pb–Pb age of 1.82 Ga (Appendix 5). Employing laser ablation MC-ICP-MS, we also The data are too few to determine the magmatic confirmed earlier TIMS results for zircon from age of the rock. two felsic porphyries (A581 Veikasenmaa and Subsequently, baddeleyite was analysed using a A893 Kiimarova) and a gabbroic sample (A1273 Nu Attom SC-ICP-MS. The U–Pb data are scattered Kulkujärvi) from the Kittilä Group (Figs. 28–30, and discordant and the laser beam very probably Appendix 2). All results are consistent with the TIMS hit domains that also contain zircon. The data do ages published by Rastas et al. (2001). Previously not provide any reliable age, but the rock may well unpublished discordant TIMS data on a gabbroic belong to the 2.0 Ga group, occurring within the rock sample, A721 Jolhikko, are also consistent with Savukoski Group basalts (Fig. 31b, Appendix 2). an age of ca. 2.0 Ga (Appendix 5). West of this locality in the Saarenpudas area, Kolari, In contrast to clearly positive initial epsilon Nd an age of ca. 2.03 Ga has previously been reported values in rocks of the Kittilä Group, a 2.0 Ga gabbro (sample A964, Hiltunen 1982). cutting the Savukoski Group at Pittarova (A1272,

Rastas et al. 2001) gives an εNd(2000 Ma) value of –0.4 (Appendix 1).

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A581 Veikasenmaa felsic porphyry data-point error ellipses are 2s

0.44 LA-MC-ICPMS Concordia Age 2200 0.40 2007 5 Ma 2160

U n=20 2120 2080 238 2040 2000 0.36 Pb/ 1960 1920

206 1880 1840 0.32 TIMS Intercepts at 234 140 & 2013 2 Ma MSWD = 0.17 (n=4)

0.28 5.0 5.4 5.8 6.2 6.6 7.0 7.4 7.8 207Pb/235U

Fig. 28. Concordia plot of U–Pb zircon data from the Veikasenmaa felsic porphyry A581, Kittilä. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses by Rastas et al. (2001) as red dots.

A893 Kiimarova felsic porphyry data-point error ellipses are 2s

0.42

LA-MC-ICPMS 0.40 Concordia Age 2005 6 Ma 2120 U n=8 2080

238 0.38 2040

Pb/ 2000 0.36 206 1960 1920 TIMS average Pb/Pb age 0.34 2012 2 Ma n=4

0.32 5.4 5.8 6.2 6.6 7.0 207Pb/235U Fig. 29. Concordia plot of U–Pb zircon data from the Kiimarova felsic porphyry A893, Kittilä. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses by Rastas et al. (2001) as red dots.

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1273 Kulkujärvi gabbro/ zircon data-point error ellipses are 2s 0.42 LA-MC-ICPMS Concordia Age = 1986 14 Ma 0.40 n=11 2120

U 0.38 2080 2040 238 2000 0.36

Pb/ 1960 A1565 Tuulijoki gabbro data-point error ellipses are 2s 1920 0.5 206 0.34 1880 Intercepts at 464 98 & 1786 20 Ma 1840 MSWD = 0.67 n=4 A1565-5a 0.4 Rastas et al 2001 TIMS coarse grain zircon Intercepts at 2100 0.32 U 1363 99 & 1986 10 Ma1900

238 analysis H concordant 1986 3 A1565Ma-1a 0.3 1700 A1565-2a A1565-4a

0.30 Pb/ 1500 A1565A TIMS (+4.3 turbid)

5.0 5.4 5.8 206 6.2 6.6 7.0 207 235 1300 Pb/0.2 1100 U A1565-3a 900 Fig. 30. Concordia plot of U–Pb zircon data from the Kulkujärvi gabbro A1273, Kittilä. LA-MC-ICPMS analyses presented as error ellipsoids and ID-TIMS analyses by Rastas et al. (2001) as dots. A) 0.1 1 3 5 7 207Pb/235U

A) B)

A1565 Tuulijoki gabbro data-point error ellipses are 2s 0.5 0.45

Intercepts at A1565 Tuulijoki gabbro Kittilä 464 98 & 1786 20 Ma Baddeleyite ± zircon MSWD = 0.67 n=4 A1565-5a 0.4 2100 0.35

U 1800 U

1900 238 Pb/ 238 A1565-1a 0.3 1700 A1565-2a A1565-4a 206 0.25 Pb/ 1500 A1565A TIMS (+4.3 turbid)

206 1300

0.2 1100 A1565-3a 1000 0.15 900 Intercepts at A) 432±97 & 2058±36 Ma 0.1 1 3 5 7 MSWD = 3.6 n=21 B) 207 235 0.05 Pb/ U 0 2 4 6 8 207 235 Pb/ U data-point error ellipses are 2s

Fig. 31. A) Concordia plot of U–Pb zircon data from the Tuulijoki gabbro A1565, Kittilä. LA-MC-ICPMS analyses

0.45 presented as error ellipsoids and ID-TIMS analyses as dots. B) Concordia plot of U–Pb baddeleyite data (using Attom) from the Tuulijoki gabbro A1565, Kittilä. A1565 Tuulijoki gabbro Kittilä Baddeleyite ± zircon

0.35

U 1800 238 Pb/ 206 0.25 39

1000 0.15 Intercepts at 432±97 & 2058±36 Ma MSWD = 3.6 n=21 B) 0.05 0 2 4 6 8 207 235 Pb/ U data-point error ellipses are 2s

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

3.7 The 1.8 Ga Tainio and Lotto intrusions

3.7.1 Tainio intrusion variation in the chemical composition of the suite, but all rocks are enriched in incompatible trace ele- The appinitic intrusive rocks occurring in the ments. Typical rocks are gabbroic in composition

Central Lapland granitoid area commonly give with MgO 6–9 wt.%, total FeO 7–9 wt.%, Na2O 3.5 rise to distinct magnetic anomalies. One of such wt.%, K2O 1.5 wt.%, Cr 300 ppm and Ni 100 ppm intrusions is the Tainio gabbro in the Pasmajärvi (Mutanen & Väänänen 2004, Väänänen 2004). area (map sheet 2642 12; Väänänen 2004) (Fig. 1). Primary igneous plagioclase and pyroxene, The stock- or pipe-like intrusion is distinguish- together with amphibole, are the main minerals able as a round, zoned magnetic anomaly ca. 3 km and were separated and used for Sm–Nd studies. in diameter. The outer magnetic ring is caused by The Sm–Nd reconnaissance work was carried out on small-grained contact gabbro rich in Fe-Ti oxides the same gabbroic sample (A1665) from which the and apatite. Towards the centre, the rocks are zircon age of 1796 ± 4 Ma was obtained. The three coarser in grain size, mainly consisting of plagio- analyses available provide an age of 1774 ± 54 Ma clase, pyroxenes, biotite and hornblende. The minor (εNd = –5.2, MSWD = 1.6, Fig. 32), which is consistent and accessory minerals include quartz, magnetite, within error with the zircon age. A peculiar feature ilmenite, apatite, sulphides, carbonate and zircon. of the analyses is the Sm/Nd ratio measured for the Gabbro autoliths and cavities partly filled by zeolite “pyroxene” concentrate, which is lower than the and carbonate minerals have also been observed, ratio in the whole-rock sample (Appendix 1). The with the latter feature suggesting that the magma analysis by XRD revealed that in addition to pyrox- was rich in volatiles. The U–Pb zircon age of the ene, this material also contains some amphibole. Tainio intrusion is 1796 ± 4 Ma (Väänänen 2004), Amphibole was probably formed in reaction between and the body hence represents one of the youngest pyroxene and the melt (Mutanen & Väänänen 2004), igneous rock suites in western Finnish Lapland (the and should thus be practically coeval with the other Lohiniva Suite). Crystal fractionation has produced main minerals.

0.5115 A1665 Tainio gabbro Age = 1774 ± 54 Ma eps = -5.2 A1665 0.5113 MSWD = 1.6 n=3

144 0.5111 A1665 px+amph Nd/ 143 0.5109

0.5107 A1665plag

Zircon age 1796±4 Ma 0.5105 0.03 0.05 0.07 0.09 0.11 0.13 147Sm/144Nd

Fig. 32. Sm–Nd isotope data for whole rock and mineral separates from the Tainio gabbro A1665.

40 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5124 A1916 Lotto dyke Age = 1805 ± 44 Ma eps = -5.2 A1916 px#2 MSWD = 1.00 n=3

A1916 px(+amph?) 0.5120 Nd 144

0.5116

Nd/ A1916 wr 143

0.5112 A1916 plag

0.5108 0.07 0.09 0.11 0.13 0.15 0.17 0.19 0.21 147Sm/144Nd

Fig. 33. Sm–Nd isotope data for whole rock and mineral separates from the Lotto mafic dyke A1916.

The low initial εNd value suggests major involve- erals, albeit variably altered. Sample A1916 for iso- ment of old LREE-enriched lithosphere in the gen- tope studies was taken from a drill core (R302, 41.1 esis of the Tainio gabbro and related intrusions. m) intersection in which the rock mainly consists The geochemical features are best explained by of plagioclase, clinopyroxene, orthopyroxene and interaction of mantle-derived magma with crustal olivine. Minor minerals include biotite, chromite, material at great depths, which is compatible with ilmenite, magnetite, sulphides and, as alteration the magmatic underplating model discussed above. products, amphibole, serpentine, talc, carbonate and chlorite. 3.7.2 Lotto dyke The Sm–Nd data on whole rock, plagioclase and a light-coloured, clear-looking heavy pyroxene frac- A mafic NNW–SSE-trending swarm of olivine- tion define an isochron with an age of 1805 ± 44 Ma bearing diabase dykes cuts the Lapland granulite and a low initial epsilon Nd value of -5.2 (Fig. 33, belt in the Lotto area, Inari. Individual dykes of Appendix 1). An analysis conducted on another the swarm can be traced for several kilometres on pyroxene concentrate, which has a relatively low the low-altitude aeromagnetic map as elongated density and higher REE level than the high-density positive anomalies (Mutanen 2011). The Lotto dyke fraction, plots below the isochron, probably due to sampled for this study preserves its primary min- associated metamorphic amphibole.

3.8 Intrusions with unknown age

3.8.1 Väkkärävaara intrusion the gabbroic rocks. The gabbro in the outcrops is cut by an ultramafic dyke. Sample A1715 collected Several outcrops indicate a gabbro intrusion at for isotope dating contains few grains of zircon, Väkkärävaara, south of the Sattasvaara komatiitic which are turbid and of poor quality. The U–Pb data rocks (Fig. 5). Based on an interpretation of aero- obtained by LA-MC-ICPMS mostly have high com- geophysical maps, it seems likely that schists of the mon lead and yield scattered and mostly discordant Matarakoski Formation (Savukoski Group) surround results (Fig. 34, Appendix 2). Two distinct grains

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A1715 Väkkärävaara metagabbro data-point error ellipses are 2s

0.6 Analysis A1715-4a 3000 Concordia Age = 2110 20 Ma 2600 A1715-11a

A1715-7a U 2200A1715-4a 0.4 A1715-8a 238 1800

Pb/ A1715-9a 1400 A1715-8b 206 A1715A1715-1a-12a 0.2 A1715-6a 1000 A1715-3a A1715-5a

A1715-2a A1715-10a

0.0 0 4 8 12 16 20 207Pb/235U

Fig. 34. Concordia plot of U–Pb zircon data obtained by LA-MC-ICPMS from the Väkkärävaara gabbro A1715.

give Archaean ages and are clearly xenocrystic. Most pentine, chlorite, talc and amphiboles. A narrow data plot on a chord having intercepts with the con- band of chlorite schist usually occurs between the cordia curve at 2.1 and 0.4 Ga, which is supported complex and its country rocks. by one concordant analysis at 2.1 Ga (A1714-4a). In order to constrain the age of the Värriö com- However, one spot is concordant at 2.0 Ga, but this plex, twelve samples were collected from four drill zircon grain is very high in U and the result is prob- cores representing different blocks of the intrusion ably less reliable for dating the primary magmatic complex. All samples have a low level of REE and crystallisation event. Although an age of ca. 2.1 Ga a variably LREE-enriched chondrite-normalised is possible, the available data do not allow reliable REE pattern. The Sm–Nd analyses provide a range dating of this gabbro. of Sm/Nd ratios, but the isotope data are scattered Sm–Nd analysis of A1715 indicated nearly chon- (Fig. 35, Appendix 2). Regression of all data gives dritic LREE/HREE for the sample and yielded an a date of ca. 1.8 Ga, and it is possible that meta-

εNd(2100 Ma) value of –3.0 (Appendix 1). morphic effects have disturbed the Sm–Nd system. However, the results give some constraints on the 3.8.2 Värriö intrusion initial Nd isotope composition of the magma. The

average εNd value calculated using an age of 2440 Ma The Värriöjoki intrusion complex is a large ultra- is -2.2 ± 1.6, and becomes lower if younger ages are mafic rock formation in eastern Lapland (Törmänen used (e.g., -4.6 ± 1.0 if the age is 2050 Ma). It seems et al. 2007, Fig. 2). The surrounding rocks belong evident that a major Archaean LREE-enriched com- to the Archaean Tuntsa suite, mainly consisting of ponent is present in the system. These apparent quartz-feldspar and mica gneiss with amphibo- epsilon values are similar to the 2.05 Ga Kevitsa lite and amphibole-chlorite schist intercalations intrusion, as well as the Koulumaoiva intrusion (Juopperi & Vaasjoki 2001). According to Törmänen located ca. 10 km east of Värriö, for which the Sm– et al. (2007), the rocks in the centre of the intrusion Nd mineral age is 2464 ± 34 Ma. In fact, based on the complex are mostly dunites, whereas peridotitic and geological setting, a correlation with these 2.4 Ga pyroxenitic rocks dominate in the outer part. The rocks would be conceivable (Fig. 2). primary magmatic minerals are replaced by ser-

42 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Värriö ultramafic intrusion

0.5122

0.5120

143Nd 0.5118 144Nd 0.5116

Age = 1768 ± 250 Ma eps = -6.1 0.5114 MSWD = 13 (n=12)

0.5112 0.10 0.12 0.14 0.16 0.18 147Sm/144Nd Fig. 35. Sm–Nd isotope data for whole-rock samples from the Värriö intrusion.

3.9 Volcanic rocks

Felsic volcanic rocks from Honkavaara in the Formation elsewhere are more altered and prob- Sodankylä area contain Archaean zircon with an lematic for isotope studies. Some clinopyroxene average age of ca. 2.7 Ga (Fig. 5, Rastas et al. 2001). was found in the komatiites from Mikkuurova Recent LA-MC-ICPMS analyses have revealed that (Sattasvaara, Fig. 5), but there were difficulties the zircon populations are heterogeneous, showing in separating pure mineral fractions. Analyses of a range of Archaean ages. These rocks are cut by pyroxene concentrates together with whole-rock 2.2 Ga mafic sills and have recently been assigned to samples yielded age estimates of ca. 1.8–1.9 Ga, the Central Lapland (Vuojärvi) supersuite, which is which reflects the influence of metamorphism. The considered Palaeoproterozoic in age (Lahtinen et al. komatiitic whole-rock samples from Mikkuurova 2015a, Nironen et al. 2016). The Honkavaara rocks have low REE concentrations, are depleted in LREE were one of the special targets of a large project and yield εNd(2060 Ma) values of ca. +4 (Appendix focusing on volcanic rocks in Lapland (Lehtonen 6), all features that are also typical for the Jeesiörova et al. 1998) and were also analysed using the area komatiites. Sm–Nd method (in 1991). Both felsic and mafic The three samples taken from a 5-m-thick mafic rocks are enriched in LREE, providing Archaean dyke cutting tuffitic komatiites in the northern model ages and negative initial εNd values at any slope of Sattasvaara hill provide an example of Palaeoproterozoic age (Appendix 6). secondary REE fractionation. These samples show Preservation of primary clinopyroxene in a large range in Sm/Nd (and also other chemical the komatiitic rocks at Jeesiörova (Sattasvaara parameters) and yielded an age of ca. 1.84 Ga (Fig. Formation, Savukoski Group) in the Kittilä area 36, Appendix 6). allowed Hanski et al. (2001b) to determine a direct Excluding dyke rocks, the 41 published and Sm–Nd isotope age for these rocks. Seven pyrox- unpublished Sm–Nd analyses conducted on ene–whole rock pairs yielded an average age of 2056 komatiitic rocks and pyroxene concentrates from

± 25 Ma and a range of initial εNd values from +2 the Sattasvaara Formation and the three analyses of to +4. In contrast to samples from Jeesiörova, Peuramaa area picrites are consistent with an age of most mafic and ultramafic rocks of the Sattasvaara ca. 2.06 Ga, with a clearly positive initial εNd (Fig. 37).

43 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Sattasvaara mafic-ultramafic dykes

Dyke 210/LVP 13718 (210-1/LVP-88)(1 0.514 Metamorphic fractionation Age = 1837 ± 16 Ma 7697 (096-28/LVP-85)(1 eps = +7 MSWD = 0.4 n=3 6603 (951-4/LVP-85)(1

Nd 0.513 13720 (210-3/LVP-88)(1 144 Nd/ 6598 (950-5/LVP-85)(1 143 6639 (953-12/LVP-85)(1 0.512 6614 (951-15/LVP-85)(1 13719 (210-2/LVP-88)(1 5631 (R2/84 255.6m)(1

0.511 0.04 0.08 0.12 0.16 0.20 0.24 0.28 0.32 147Sm/144Nd Fig. 36. Sm–Nd isotope data for mafic-ultramafic dykes from Sattasvaara.

Sattasvaara Fm komatiites & Peuramaa picrites 0.517

0.516 Age = 2064 ± 38 Ma epsilon =+3.1 MSWD = 21 (n=44, wr & cpx) 0.515 pyroxene Nd

144 0.514 Nd/

143 0.513

0.512 picrites

0.511 0.05 0.15 0.25 0.35 0.45 147Sm/144Nd Fig. 37. Sm–Nd isotope data for whole-rock samples and mineral separates from the Sattasvaara komatiites and Peuramaa picrites.

44 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

In addition to open systems, the scatter in the (Fig. 5). Pillow basalts are common, but primary data (MSWD = 21) is also due to slight variation in main minerals are not preserved. The rocks are the primary initial Nd isotope ratios (Hanski et al. enriched in LREE and yield an average εNd(2060 Ma) 2001b). value of +0.7 and TDM = 2.35 Ga (Appendix 6). The Linkupalo mafic volcanic rocks in Western Lapland are also assigned to the Savukoski Group

4 TAIVALKOSKI BLOCK IN THE LENTUA COMPLEX AND KUUSAMO SCHIST BELT

4.1 Geological background

The Taivalkoski basement block represents the Kuhmo block by E–W- and SW–NE-trending shear northernmost part of the more than 400-km-long zones (Hölttä et al. 2012). The dominant rock types Archaean basement complex (Lentua complex in are 2.7–2.8 Ga granodioritic and tonalitic gneisses Hölttä et al. 2012) in eastern Finland. On its south- that locally have largely anhydrous granulite-facies ern side, the Taivalkoski block is separated from the mineral assemblages (Vuollo & Huhma 2005, Lauri

^_ ^_^_ Ruukinvaara Fm/ Sodankylä Gr ¢ Kuusamo schist belt ^_ Petäjävaara Fm/ ^_Sodankylä Gr^_^_ ^_ Kun�järvi Fm/ A1868 Kuusamo Gr A0497 Laivajoki ^_ A1443 Kortejärvi ^_ ^_ Taivalkoski block Suoperä (BD) A0713^_A0610 ^_ Uolevinlehto ^_^_^_ Näränkävaara (UD) A0722Por�vaara Intrusions and ^_^_ ^_^_ volcanic rocks ^_ A1663 Kon�oluoma ^_ Taivalkoski Sm-Nd (±U-Pb) Syöte ^_ ^_ ^_ Volcanic rocks ^_ Kallioniemi ^_ Koivuvaara ^_ U-Pb ( in this paper ) ^_ A1471 ^_ Murhiniemi _ Norway ^ ^_ Hirsikangas U-Pb (age published elsewhere) Tilsanvaara Russia Sm-NdXW (age ?) 2.4 - 2.5 Ga felsic rocks ( in this paper ) Sweden A0365 Karkuvaara/Nyrhinoja ^_^_ Age ^_ ^_ <1930 Ma Ma�nvaara Fm/ A0988 ^_ 1931 - 2080 Kurkikylä Gr ^_ 2081 - 2180 Finland ^_ ^_ 2181 - 2280 A0496 Lohisärkkä ^_ 2281 - 2380 0 30 km ^_ ^_ >2380

Fig. 38. Geological map of the Taivalkoski block (northern part of Lentua complex) and Kuusamo schist belt showing sample localities. For symbols, see Figure 1. The Koillismaa layered intrusion suite refers to 2.44 Ga rocks in Syöte, Porttivaara and Näränkävaara near the Russian border.

45 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye et al. 2006). A prominent feature of the Taivalkoski 2) 2.32 Ga Fe-tholeiitic dyke swarm and block is the presence of 2.44 Ga mafic-ultramafic intrusions bodies assigned to the Koillismaa layered intrusion 3) 2.2 Ga low-Al tholeiitic layered sills suite (Alapieti 1982, Karinen 2010) (Fig. 38). It is (karjalites, gabbro-wehrlite association) divided into two intrusions, the Näränkävaara lay- 4) ~2.1 Ga Fe-tholeiitic dyke swarm ered intrusion on the Finnish–Russian border and 5) ~1.98 Ga Fe-tholeiitic–tholeiitic dyke swarm the Porttivaara layered intrusion in the western part of the Taivalkoski area. The latter intrusion is now The oldest 2.45 Ga group is further divided into 5 represented by several separate intrusion blocks, subtypes: which were formed by tectonic fragmentation of (1) NE–SW-trending boninite–noritic dykes a larger, originally cohesive, sheet-like magmatic (high MgO, SiO2, Cr, Ni, and LREE, body. low TiO2 and Zr) Other manifestations of the Palaeoproterozoic (2) NW–SE-trending gabbro-norite dykes mafic magmatism in the Taivalkoski basement block (low TiO2, Cr, and Zr) are mafic dykes that occur in several generations of (3) NW–SE-trending low-Ti tholeiitic dykes dyke swarms. A review of the age groups, areal dis- (4) NW–SE-trending Fe-tholeiitic dykes, tribution, mineralogy and geochemistry of the mafic continental type dyke swarms occurring in the Archaean basement (5) E–W-trending orthopyroxene- and plagio- and Karelian schist belts in northern and eastern clase-phyric dykes (high SiO2, LREE; Finland was published by Vuollo & Huhma (2005). calc-alkaline aff.) Before dealing with the isotope data from dyke rocks from the Taivalkoski and other basement blocks, it In contrast to the 2.45 Ga dykes, the younger dyke is worth presenting a summary of the classifica- swarms with ages between 2.32 and 1.98 Ga show tion of the dyke swarms by Vuollo & Huhma (2005). more subtle variation in chemistry, with most of Based on the emplacement ages, they distinguished them resembling continental tholeiitic basalts in the following 5 dyke swarm groups: composition. However, there may be clear differ- 1) 2.45 Ga dyke swarms ences in the orientation between the swarms of

Fig. 39. Stratigraphic column of the Kuusamo schist belt as originally established by Silvennoinen (1972) (left) and current lithostratigraphic classification as defined by GTK (right). Note that some of the formations of Silvennoinen are now lowered to the member status, being part of a single formation in the new classification.

46 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland different age groups. As shown by a map of Vuollo belt and occurs at four stratigraphic levels repre- & Huhma (2005, their Fig. 5.14), there is a dense sented by the Kuntijärvi Fm (former Greenstone network of mostly E–W-trending mafic dykes in Fm I), Petäjävaara Fm (former Greenstone Fm II), the Taivalkoski block. Only a few age determina- Ruukinvaara Fm (former Greenstone Fm III) and tions are available for these dykes. Nevertheless, Liikasenvaara Fm (former Amphibole Schist Fm). examples of the 2.45 Ga dykes have been recog- The samples studied in this work from the ca. nised, such as boninite-norites with an orientation 2.44 Ga Koillismaa layered intrusion suite represent of 60o. The U–Pb ages of 2446 ± 6 Ma and 2332 ± an anorthosite dyke from the Syöte block and previ- 18 Ma mentioned by Vuollo & Huhma (2005) were ously studied gabbros from the Porttivaara block, based on U–Pb analyses on baddeleyite carried out which hosts the Mustavaara Fe–Ti–V ore deposit in Toronto and are reported in this volume below (Karinen et al. 2015). We also document abundant (samples A1415 and A1471). Sm–Nd mineral and whole-rock data for several The Kuusamo schist belt comprises a supracrus- of the mafic dyke families distinguished in Vuollo tal sequence that was deposited on Archaean rocks & Huhma (2005). The ca. 2.44 Ga dykes that were of the Taivalkoski block (Fig. 38). The Kuusamo analysed from the Russian side of the border include belt continues to the Russian side of the border as five dykes from the Pääjärvi area, belonging to the the Kuolajärvi-Paanajärvi belt. In the west, the belt gabbro-noritic, Fe-tholeiitic and orthopyroxene- is cut by plutons of the Central Lapland granitoid phyric types, and one Fe-tholeiitic dyke from the complex, and in the north it continues as the Salla Suoperä area (Fig. 38). In an attempt to constrain belt. The formation-level stratigraphic division of the ages and initial Nd isotope compositions of the the Kuusamo schist belt, as originally established by mafic dykes occurring in a granulite-facies ter- Silvennoinen (1972), is shown in Figure 39 together rain, about 2.0 x 4.5 km2 in size, around the town of with the currently used division applying a more Taivalkoski, nine well-preserved dykes were sam- formal lithostratigraphic nomenclature. In the latter pled. One of these samples was successfully dated scheme, the stratigraphic column is divided into 11 at ca. 2.3 Ga using the U–Pb baddeleyite method. formations, which belong to the Salla, Kuusamo, Another representative of the dykes of this age Sodankylä and Savukoski Groups, i.e., the same group was studied at Karkuvaara, occurring in the groups that are distinguished in the stratigraphy of SW part of the Taivalkoski block close to the Kainuu the Central Lapland greenstone belt (see Fig. 3). With schist belt (Fig. 38). Some Nd isotope data on mafic the exception of the Salla Group, mafic extrusive volcanic rocks from three different stratigraphic magmatism is found in all groups of the Kuusamo levels of the Kuusamo schist belt are also presented.

4.2 The 2.44 Ga Koillismaa layered intrusion suite

The age of 2436 ± 5 Ma reported by Alapieti (1982) ues (analysed at VSEGEI, St. Petersburg). In order for the Koillismaa layered intrusion suite was based to confirm this result, we performed analyses on on sixteen U–Pb zircon analyses conducted on five equivalent samples from the same drill core. several gabbroic samples (Kouvo 1977). Recently, These data (Appendix 1) yielded initial εNd(2440 Ma) the Koillismaa intrusions were studied by Karinen values from -1.0 to -2.8, which are in contrast to (2010), who reported abundant Sm–Nd data on those in Karinen (2010) and typical for rocks in the whole rocks, pyroxene, and plagioclase. The min- Koillismaa layered intrusion suite (Fig. 40). eral isochrons on two samples from the Porttivaara North of the Syöte block, mafic volcanic rocks intrusion block (Fig. 38) yielded ages consistent with of the Vehnäsvaara Formation are cut by an the U–Pb zircon age, but for two other samples, the anorthositic dyke a few metres in thickness. The Sm–Nd mineral age was slightly lower, presum- anorthosite was assumed to represent the 2.44 Ga ably due to metamorphic effects (all analysed at magmatism, and a sample (A1663-Kaidansuvanto)

GTK). Nevertheless, the initial εNd(2440 Ma) values was hence collected with the aim to constrain the were found to be negative and close to -2, which age of the volcanic rocks intersected by the dyke. is characteristic for rocks of this family. Karinen The zircon population of the rock appears hetero- (2010) also published Sm–Nd data on gabbroic rocks geneous and yields scattered U–Pb data (Fig. 41, from the two PGE reefs (Rometölväs, Syöte), some Appendix 7). The eleven oldest compositions are of which yielded positive initial εNd(2440 Ma) val- concordant and yield an age of 2.82 Ga, which prob-

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Porttivaara & Syöte intrusions

0.5125

pyroxene Nd 0.5115 144 Nd/ 143 0.5105 Porttivaara plagioclase Age = 2388 75 Ma Initial eps = -2.2 MSWD = 12, n=23

0.5095 0.02 0.06 0.10 0.14 0.18 0.22 147Sm/144Nd

Fig. 40. Sm–Nd isotope data for whole-rock samples and mineral separates from the Koillismaa layered intrusion suite. Data from Karinen (2010, blue triangles), this study (Syöte, five red diamonds) and Sirniö felsic rocks from Lauri et al. (2006, x), all analysed at GTK, Espoo.

A1663 Kaidansuvanto anorthosite dyke 0.75 data-point error ellipses are 2s

0.65 Grain 13 a & b Age = 2432 21 Ma 0.55 2800

U Concordia Age = 2815 10 Ma

238 TIMS n=11 0.45 2400 Pb/ 2000 206 0.35 Alapieti 1982: Koillismaa layered intrusion 1600 Intercepts at 0.25 322 100 & 2438 7 Ma 1200 MSWD = 12 (n=17) zircon from 6 samples 0.15 0 4 8 12 16 20 207Pb/235U Fig. 41. Concordia plot of U–Pb zircon data obtained by LA-MC-ICPMS from the anorthosite dyke A1663. TIMS U–Pb data on zircon from the Koillismaa layered intrusion suite published by Alapieti (1982) are also shown for reference (x, +). The data with Archean age (+) are from the Soukeli sample A610.

48 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland ably dates zircons inherited from some felsic gneiss to plot above the concordia curve, which is prob- in the hosting Archaean basement complex. The ably due to a different matrix compared to the zir- results from six analyses range from 1.85 to 2.1 Ga, con used as the standard. However, the average and three data points yield Pb/Pb ages of ca. 2.4 Ga. 207Pb/206Pb age of 2428 ± 6 Ma can be considered as Two of the measurements are very likely to at least a reliable estimate for the age of magmatic zircon partly hit baddeleyite, yielding reversely discord- (Fig. 43), consistent with the two TIMS analyses ant U–Pb results, obviously due to the use of an reported by Alapieti (1982), as well as with old data unsuitable standard (zircon). In spite of this, the from the nearby Välivaara gabbro (sample A699) Pb/Pb isotope ratios should still be correct, as was (Fig. 43). discussed above (in section 3.3; A136). It is tempt- The third old sample re-analysed by laser abla- ing to interpret the ca. 2.4 Ga date as representing tion MC-ICP-MS from the Koillismaa layered intru- the primary igneous age and the ages of 1.85–2.1 sion suite represents a rock near the base of the Ga as the variable effects of (1.8 Ga) metamorphic/ Porttivaara intrusion. It has been called “albitite hydrothermal overprinting. or palingenic rock” (A722 Rusamo). Earlier unpub- Zircon grains from three earlier studied samples lished TIMS data (Appendix 5) suggested a con- picked from the Porttivaara layered intrusion were tribution from the Archaean basement, which was recently re-analysed using LA-MC-ICPMS. Three confirmed by one spot analysis (7a). Other analyses zircon standards were employed in calibration (GJ1, performed on selected clean domains suggest an age A1772 and A382). Only one distinct grain from the of ca. 2.45 Ga (Fig. 44). The old TIMS data from sev- Soukeli pyroxenite (A610) is close to 2.44 Ga in age, eral other samples further support this age (A577, whereas the other data give an age of ca. 2.72 Ga, A578 and A723-5, Appendix 5). suggesting contamination from Archaean country In summary, the results available on the rocks. These results are consistent with the previ- Koillismaa layered intrusion suite confirm its age ous TIMS data acquired from this sample (Fig. 42, of 2.44 Ga and also show significant communica- Alapieti 1982). tion of the magma with Archaean crustal material. New U–Pb data obtained by LA-MC-ICPMS on zircon from the Mustavaara gabbro (A713) tend

A610 Soukeli, Koillismaa intrusion pyroxenite

data-point error ellipses are 2s 0.62

0.58 2900 Concordia Age = 2453 ±12 Ma 2800 0.54 n=2, grain 4 U 2700 238

Pb/ 0.50 2600 206 2500

0.46 2400

2300 0.42

0.38 8 10 12 14 16 207Pb/235U

Fig. 42. Concordia plot of U–Pb zircon data obtained by LA-MC-ICPMS from the Soukeli pyroxenite A610. Multigrain TIMS U–Pb data on zircon by Alapieti (1982) are also presented (red triangles).

49 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A713 Mustavaara, Koillismaa intrusion gabbro

data-point error ellipses are 2s 0.60

0.56

U 2700 0.52 238

Pb/ 2600 206 0.48 2500

2400 Average Pb/Pb age 0.44 2428 ± 6 Ma MSWD = 0.22, n=6 diamond - A699-Välivaara gabbro 0.40 9 10 11 12 13 207Pb/235U

Fig. 43. Concordia plot of U–Pb zircon data obtained by LA-MC-ICPMS from the Mustavaara gabbro A713. Old multigrain TIMS U–Pb data on zircon are also presented: A713 Mustavaara gabbro – red triangle (Alapieti 1982) and A699 Välivaara gabbro – blue diamond.

A722 Rusamo, Koillismaa intrusion albitite 0.60 data-point error ellipses are 2s

0.56 2800

2700 0.52 2600

U 0.48 2500 238 2400 Pb/ 0.44 206 2300

2200 0.40

Average Pb/Pb age 0.36 2454 ± 8 Ma MSWD = 1.6 n=5

0.32 7 9 11 13 15 207Pb/235U

Fig. 44. Concordia plot of U–Pb zircon data obtained by LA-MC-ICPMS from the Rusamo rock A722. Multigrain TIMS U–Pb data on zircon are also presented (red triangle).

50 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

4.3 Dykes in the Lake Pääjärvi and Suoperä areas, Russia

Several generations of dykes are well exposed on fractions of pyroxene and plagioclase from sam- Lupzenga Island in Lake Pääjärvi, Russia, where ple A1412 are fresh, although plagioclase is slightly cross-cutting relationships seen in outcrops reli- heterogeneous in colour. The analyses conducted on ably demonstrate the relative order of emplacement the whole-rock samples are identical and, together of four dyke generations (Stepanov 1994, Figs. 1 and with the mineral separates, they give a Sm–Nd age 45). The rocks are good targets for Sm–Nd stud- of 2421 ± 32 Ma (Fig. 46, Appendix 1). The calculated ies, because the primary magmatic mineralogy is initial εNd value is -1.4 (MSWD = 1). generally well preserved. Isotope results from the Pääjärvi dykes were discussed by Vuollo & Huhma 4.3.2 “Older Fe-tholeiitic dyke” A1414 Pääjärvi (2005), but the actual data have not been published until here. Palaeomagnetic studies on the dykes The gabbro-norite dyke discussed above is cut by have been published by Mertanen et al. (1999). The a Fe-tholeiitic dyke, which was called the “Older 2.44 Ga layered intrusions of the Oulanga Complex Jatulian dyke” by Stepanov (1994). Fresh plagioclase (Kivakka, Tsipringa and Lukkulaisvaara) are located in the sample (A1414 = 38-VEN-94) from this rock some 20 km NNW of the island of Lupzenga. The unit is fairly dark due to some pigment. The three Sm–Nd analyses from these intrusions and two analyses performed on this sample plot exactly on dyke rocks have earlier yielded mineral ages con- a line (MSWD = 0.01), which gives an age of 2476 sistent with the 2.44 Ga U–Pb zircon ages (Amelin ± 30 Ma (Fig. 47). The initial εNd value is +1.7. Using

& Semenov 1996). The reported initial εNd values the age of typical mafic layered intrusions,ε Nd +1.4 range between 0 and –2. at 2.44 Ga can be calculated for the whole-rock sample. The face age value appears to contradict 4.3.1. Gabbro-norite dyke A1412 Pääjärvi the age and geological relationship of the previous sample, but considering the error limits, no Samples A1412 (35-VEN-94) and A1413 (36-VEN- difference in age can be stated. Instead, the initial 94) represent a NW–SE-trending gabbro-noritic value is very distinct from the negative value for dyke swarm, which intrudes the Archaean granitoid the gabbro-norite and layered intrusions in general. basement (Vuollo & Huhma 2005). The separated

Fig. 45. Geological map of site XD at Lake Pääjärvi after Stepanov (1994) and Mertanen et al. (1999). A1412 represents XD1 gabbro-norite dyke, A1414 – XD3 (older) Fe-tholeiitic dyke, A1492 – XD4 (younger) Fe-tholeiitic dyke.

51 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.5130 A1412 Pääjärvi gabbronorite dyke 0.5126 A1412px

0.5122 Nd

144 0.5118 Nd/ A1413wr 143 0.5114 A1412wr Age = 2421 ± 32 Ma 0.5110 A1412plag eps = -1.4 MSWD = 1.5 n=4

0.5106 0.07 0.09 0.11 0.13 0.15 0.17 0.19 0.21 147Sm/144Nd

Fig. 46. Sm–Nd isotope data for whole rock and mineral separates from the Pääjärvi gabbro-norite dyke A1412.

0.5132 A1414 Pääjärvi Fe-tholeiitic dyke A1414px

0.5128

Nd 0.5124 144

A1414wr Nd/ 0.5120 143 Age = 2476 ± 35 Ma 0.5116 eps = +1.7 A1414plag MSWD = 0.0045 n=3

0.5112 0.10 0.12 0.14 0.16 0.18 0.20 0.22 0.24 147Sm/144Nd

Fig. 47. Sm–Nd isotope data for whole-rock and mineral separates from the Pääjärvi dyke A1414.

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5140 A1492 Pääjärvi Fe-tholeiitic dyke 0.5136 Age = 2349 24 Ma eps = +1.0 A1492px 0.5132 MSWD = 1.4 n=3

0.5128 Nd

144 0.5124 Nd/ 0.5120 A1492wr 143

0.5116 A1465opx A1492plag 0.5112 A1465wr A1465plag A1465 Pääjärvi opx-phyric dyke 0.5108 0.08 0.12 0.16 0.20 0.24 0.28 147Sm/144Nd

Fig. 48. Sm–Nd isotope data for whole rock and mineral separates from the Pääjärvi Fe-tholeiitic dyke A1492 and dyke A1465.

4.3.3 “Younger Fe-tholeiitic dyke” A1492 Pääjärvi plagioclase. Sample A1465 (= 42-VEN-94) selected for isotopic studies was picked ca. 100 m west of According to field observations, the “Older Jatulian the gabbro-norites on Lupzenga Island. It contains dyke” discussed above is cut by an E–W-trending clear orthopyroxene phenocrysts (φ ca. 1 mm) in Fe-tholeiitic dyke, which has been called the a fine-grained groundmass. Fresh plagioclase is “Younger Jatulian dyke” and is considered the characterised by dark pigment and, after standard youngest dyke rock in the area (Stepanov 1994). Franz separation, it was recovered from the mag- Sample A1492 (39-VEN-94) from this rock was col- netic fraction. lected ca. 20 m NW of the location of the previous The three isotope analyses on minerals and whole sample, A1414. The magmatic mineral paragenesis rock reveal that the range in Sm/Nd is too limited is well preserved, but plagioclase i n this rock is to constrain the age of the dyke (Appendix 1). The also relatively dark. The Sm–Nd analyses on hand- REE level in the whole-rock sample and also in the picked plagioclase and pyroxene together with the analysed plagioclase fraction is high. The dykes whole-rock sample give an age of 2349 ± 24 Ma have a calc-alkaline geochemical affinity, and thus

(εNd = +1.0, MSWD = 1.4, Fig. 48). The age is in good they have been related to the 2.44 Ga boninitic dyke agreement with the field observations and results swarm. Assuming that the age is ca. 2.44 Ga, the obtained from the other rocks on Lupzenga Island. isotope data on the whole-rock sample (and plagio-

clase) suggest an initial εNd value of -2.2. However, ε 4.3.4 Orthopyroxene-phyric dyke A1465 Pääjärvi Nd(2440 Ma) for orthopyroxene is +2.2, suggesting isotopic disequilibrium with the whole rock (Fig. Orthopyroxene-phyric dykes with an east–west 48). The Sm/Nd ratio in orthopyroxene is low for a trend have only been discovered in a few places typical igneous pyroxene. The data indicate that the in Russian Karelia and near the adjacent border origin of the pyroxene is distinct from the bulk rock. in Kuusamo, Finland. The colour of the dykes is The obvious explanation would be that orthopyrox- almost black and the width varies from 1 to 3 m. ene crystallised early in the magma derived from The primary minerals include orpthopyroxene and depleted mantle, which was subsequently con-

53 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye taminated with material from the Archaean LREE- 4.3.6 Gabbro-norite dyke, A1415 Suoperä enriched lithosphere. The reason for the low Sm/Nd ratio in the orthopyroxene remains obscure. Dykes belonging to the gabbro-norite dyke swarm trend NW and are fresh on the Russian side of the 4.3.5 Fe-tholeiitic Oulanka dyke border and also in some areas on the Finnish side. These medium- to coarse-grained gabbro-norite An altered Fe-tholeiitic dyke was sampled (113- dykes, up to 50 m in thickness, consist of clino- VEN-94, 13.6 kg) near the Pääjärvi-Oulanka River pyroxene (25%), calcic plagioclase (60%), and road about 24 km north of Pääjärvi village. No con- orthopyroxene (5–10%) with minor olivine (1–2%), tact to the country rock gneiss was visible at the quartz (<5%), biotite, and Fe–Ti oxides. In the same sampling site. The trend of the dyke is 310º and its way as with boninitic-noritic dykes, they are altered minimum width is 15 m. It is totally altered and no in many places on the Finnish side. The areal dis- primary pyroxenes can be seen. A moderate quantity tribution of the gabbro-norite dykes is difficult of extremely poor-quality baddeleyite grains was to assess due to the same trend of younger dykes. discovered (in Toronto). U–Pb analysis conducted Based on geochemical studies (relatively low Cr and on the best-quality grains is 2.3% discordant, pro- Ni), these dykes are seen on both the Russian and viding a minimum age of ca. 2.3 Ga (Appendix 8). Finnish side. They can be traced over a distance of The true age of the dyke cannot be determined, but some kilometres. an age close to that of the 2.44 Ga layered intru- A fresh gabbro-noritic sample, A1415 (122-VEN- sion of the Oulanka Complex (Amelin et al. 1995) is 94, total 15.4 kg), was taken from the central part plausible (see Fig. 49 reference line 1.88–2.45 Ga). of a 50- to 60 m-thick gabbro-norite dyke in the

data-point error ellipses are 2s

Baddeleyite

2440 0.46 A1471 (AD10) Taivalkoski Fe-tholeiitic dyke Intercepts at 641±730 & 2332 ± 9 Ma 2400 A1415 (BD) MSWD = 0.49 2360

U 0.44 A1356 Boninite (VD) 238 2320 A1410 Tholeiite (UD) Pb/

206 A1415 (BD) Pääjärvi A1471 (AD10) 2280 gabbro-norite dyke 0.42 Intercepts at 837±450 & 2240 13-VEN 2447 ± 10 Ma MSWD = 0.60 (n=4)

13-VEN Oulanka Fe-tholeiitic dyke Reference line intercepts at Pb/Pb age 2292±4 Ma 1880 & 2450 Ma 0.40 7.8 8.2 8.6 9.0 9.4 9.8 10.2 207Pb/235U

Fig. 49. Concordia diagram showing the U–Pb data for baddeleyite fractions analysed at the University of Toronto. The error ellipses reflect two sigma errors.

54 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5126 A1415 Suoperä gabbronorite dyke 0.5122 A1415 px

0.5118 Nd

144 0.5114

Nd/ A1415 wr

143 0.5110 Age = 2420 ± 29 Ma eps = -2.4 A1415 plag 0.5106 MSWD = 0.54 n=3

0.5102 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 147Sm/144Nd

Fig. 50. Sm–Nd isotope data for whole rock and mineral separates from the Suoperä gabbro-norite dyke A1415.

Suoperä area (site BD) located between the 2.44 Ga a 207Pb/206Pb age of 2442 ± 5 Ma (Appendix 8). Three Näränkävaara intrusion and Lake Pääjärvi, ca. 10 additional fractions of the second best baddeleyite km east of the Russian–Finnish border. grains (2–4) are slightly discordant (1.9–2.3%), At site BD, there are no visible chilled margins. with 207Pb/206Pb ages ranging from 2434 to 2436 A large amount of relatively high-quality badde- Ma (Appendix 8). The compositions of these four leyite was recovered from the sample (in Toronto). fractions of baddeleyite define a discordia line (55% All of the baddeleyite grains that were selected for probability of fit) that has an upper-intercept age analysis were rather fresh and contained no vis- of 2447 ± 10 Ma (2σ) and a lower-intercept age of ible zircon rims. Nearly all of them were broken, ca. 840 Ma (Fig. 49). probably during the crushing process. The quan- The pyroxene fraction (obtained at GTK) from tity and grain size (5–30 µm) of these brownish sample A1415 looks good, but plagioclase is partly to pale brownish, broken baddeleyites were small. yellowish and cloudy. The three Sm–Nd analyses Four fractions of baddeleyite were analysed; the conducted on minerals and whole rock (Appendix best baddeleyite fraction without visible inclusions, 1) yielded an isochron age of 2420 ± 29 Ma with fractures or turbidity is near concordant (0.7%) with an initial εNd value of -2.4 (MSWD = 0.54, Fig. 50).

4.4 The 2.3 Ga Karkuvaara intrusion

The Karkuvaara (Nyrhinoja) intrusion is located of the Karkuvaara intrusion are mostly nearly unde- within Archaean gneisses just east of the formed and tend to contain well-preserved primary Jaurakkajärvi section of the Palaeoproterozoic igneous minerals. The intrusion dominantly com- Kainuu schist belt and about 6 km to the east of prises medium- to coarse-grained metadiabasic to the major N–S-running Proterozoic fault/shear granular gabbroic rocks, which have a fairly Fe- and zones separating the latter from the gneisses of Ti-rich subalkalic basaltic composition. the Kalhamajärvi gneiss complex (Fig. 38). Despite Both U–Pb and Sm–Nd isotope studies were plentiful late Proterozoic granite in the Kalhamajärvi performed on three samples from the intrusion. A complex and strong (1.8–1.9 Ga) mylonitic defor- coarse-grained gabbro pegmatoid sample (A988 mation associated with the faults in the west, rocks Nyrhinoja) yielded abundant zircon and some

55 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye baddeleyite, which were used for U–Pb dating. The average age; rejecting four of such analyses with extracted zircon grains were mostly euhedral crys- U >800 ppm, a concordia age of 2300 ± 10 Ma can tals with sharp edges. However, the appearance of be calculated. the population is somewhat heterogeneous due to The Sm–Nd mineral studies were carried out on variation in the colour, transparency and presence two samples, A1456b and A1456c. The major pri- of inclusions. Much effort was put into the hand- mary minerals, plagioclase and pyroxenes, appear picking of different zircon types for analyses. well preserved and the mineral fractions purified by Nine U–Pb analyses have earlier been conducted hand-picking are perfectly clean. The six analyses on zircon and baddeleyite from sample A988 under are of good quality and form an isochron that gives the supervision of O. Kouvo (Appendix 5). The data an age of 2319 ± 27 Ma (εNd= +1.8, MSWD = 1.6, Fig. are technically good and provide a regression line 52, Appendix 1). The two samples treated separately having intercepts with the concordia curve at 2306 provide ages of 2295 ± 37 Ma (A1456b) and 2345 ± ± 8 and 542 ± 150 Ma (Fig. 51). An old TIMS analysis 39 Ma (A1456c). on another sample from Karkuvaara, A365, supports Thus, as the applied U–Pb and Sm–Nd meth- this age (Appendix 5). Subsequently, zircon from ods all produce consistent results, the nine-point sample A988 was also analysed by laser ablation multigrain TIMS age of 2306 ± 8 Ma should be a MC-ICP-MS. All twelve data points are concordant reliable age estimate for the mafic intrusion at within error and yield an age of 2294 ± 8 Ma (Fig. Karkuvaara. The initial ratio (εNd = +1.8) suggests 51, Appendix 7). It is notable that the ages from that the magma was derived from depleted mantle high-U zircon tend to be on the lower side of the without major contamination from Archaean crust.

A988 Nyrhinoja gabbro data-point error ellipses are 2s 0.58

0.54 Concordia Age = 2294 8 Ma n=12 (LA-MC-ICPMS)

0.50

2500 206 Pb 0.46 2400 238 U 2300 0.42 2200

0.38 TIMS Intercepts at 542 150 & 2306 8 Ma MSWD = 4.6 n=9 0.34 7 8 9 10 11 207Pb/235U

Fig. 51. Concordia plot of U–Pb zircon data from the Nyrhinoja gabbro A988, Karkuvaara intrusion. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red triangles.

56 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5134 A1456 Karkuvaara 0.5130 A1456bpx Age = 2319 ± 27 Ma A1456cpx eps = +1.8 MSWD = 1.6 n=6

0.5126 A1456c Nd A1456b

144 0.5122 Nd/

143 0.5118

0.5114 A1456bplag A1456cplag A988 Nyrhinoja U-Pb zircon age 2306 ± 8 Ma 0.5110 0.08 0.12 0.16 0.20 0.24 147Sm/144Nd

Fig. 52. Sm–Nd isotope data for whole rock and mineral separates from the Karkuvaara gabbroic samples A1456.

4.5 Dykes in the Taivalkoski town area

Based on available geological and aeromagnetic modal composition of the dyke interior diabase anomaly maps, several dyke swarms with relatively is plagioclase (60%), clinopyroxene (30%), and well-preserved primary igneous mineralogy occur Fe–Ti-oxides (5–10%). The chemical and mineral in an area around the town of Taivalkoski, espe- composition of sample A1466 is similar to that of cially in the parts characterised by well-preserved the other well-preserved 2.3–2.0 Ga Fe-tholeiitic Archaean granulite-facies rocks. Ten samples from dykes dealt with in this work. One clear difference the dykes were selected for preliminary Sm–Nd between this dyke swarm and 2.45 Ga dyke swarms mineral analyses (Fig. 38). The mineral concen- is that plagioclase is not cloudy. trates are mostly of fairly good quality, although Standard separation including some hand- slight heterogeneity in the colour and brightness, picking produced clean fractions of pyroxene and especially in some plagioclase fractions, is appar- plagioclase. Three analyses conducted on the min- ent. No proper grain-by-grain hand-picking was eral extracts and whole-rock powder (Appendix 1) applied, but the separates were only screened yielded a Sm–Nd isochron with an age of 2407 ± under the stereomicroscope for distinctly unwanted 40 Ma (MSWD = 1.9, Fig. 53). The initial εNd value is grains, which were removed (samples A1794-1802). clearly positive (+1.6).

4.5.1 A1466 Taivalkoski 4.5.2 A1471 Taivalkoski

The Taivalkoski dyke (A1466, WD14) represents one Another E–W-trending Fe-tholeiitic dyke (A1471, of the numerous E–W-trending Fe-tholeiitic dykes AD10) was sampled ca. 3 km SE of the A1466 observed in the Taivalkoski area. The 10-m-wide, Taivalkoski site. This rock consists of plagioclase WNW–ESE-trending, subvertical dyke is exposed in (50%), clinopyroxene (15%), secondary amphibole a road cut near the town of Taivalkoski. The inte- (20%) and quartz. A large number of good qual- riors of the dyke are occupied by medium-grained ity, mostly prismatic, 5- to 70-µm-long badde- diabase, which grades to fine-grained, glassy- leyite grains were recovered from the large, 27.2 kg looking chilled margins at the dyke contacts. The sample in Toronto. Some baddeleyite grains were

57 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.5138 A1466 Taivalkoski Fe-tholeiitic dyke

A1466px 0.5134

0.5130 Nd 144 0.5126 Nd/ 143 0.5122 A1466wr Age = 2407 ± 40 Ma eps = +1.6 0.5118 A1466plag MSWD = 1.9 n=3

0.5114 0.10 0.14 0.18 0.22 0.26 147Sm/144Nd

Fig. 53. Sm–Nd isotope data for whole rock and mineral separates from the Taivalkoski Fe-tholeiitic dyke A1466.

observed with thin zircon rims. All four analysed the results for sample A1797, but the data on plagio- fractions (Appendix 8) yielded similar slightly dis- clase and pyroxene are inconsistent. The analysed cordant compositions (0.9 –1.7%). They define a plagioclase is light in colour and analogous with the discordia line giving an upper-intercept age of 2332 previous sample; the effects of metamorphism are ± 9 Ma (2σ) and a lower-intercept age of 640 Ma conceivable. The analysed pyroxene fraction may (Fig. 49). contain some amphibole. The dyke is characterised by a positive initial

4.5.3 A1797 Törninkuru εNd value, similar to that of the 2.4 Ga dyke A1466. Recently, a U–Pb age of 2339 ± 18 Ma was reported A third sample from the E–W-trending Fe-tholeiitic for baddeleyite from the same dyke (sample AD13; dykes (A1797, AD13-8) was collected from a road cut Salminen et al. 2014). The Sm–Nd and U–Pb ages ca. 10 km SW of the A1466 site. Both contacts of this from the Törninkuru dyke are thus consistent dyke are exposed and show well-preserved chilled within error. margins. Primary minerals including plagioclase There are now two samples from the E–W- and clinopyroxene are well preserved. The mineral trending Fe-tholeiitic dykes that suggest a U– separation showed that most of the plagioclase con- Pb age of ca. 2.33 Ga, whereas the Sm–Nd data sists of darkish grains, which were collected from tend to give marginally older age indications. a slightly magnetic fraction. The Sm–Nd analysis Combining the Sm–Nd data on samples A1466 conducted on this plagioclase, together with the and A1797, an age of 2349 ± 75 Ma can be cal- data on pyroxene and whole rock (two identical culated (n = 10, MSWD = 7.9). In any case, the analyses), provide an isochron and age of 2404 ± initial epsilon Nd value is positive (+1.6) and 53 Ma (MSWD = 1.5, n = 4). In contrast, the Sm–Nd clearly distinct from most 2.44 Ga mafic rocks. analysis of the light plagioclase (A1797plag#1) is distinctly off the isochron (Fig. 54), which is prob- 4.5.4 A1796 Kallioniemi ably due to metamorphic effects. Another dyke sample (AD13-9) from the same The Kallioniemi outcrop is located by the Lake outcrop was also processed. Sm–Nd analysis of the Jokijärvi, ca. 20 km southeast of the town of whole-rock powder was roughly compatible with Taivalkoski, within the farmyard of the museum

58 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5128 A1797 Törninkuru dyke Age = 2404 ± 53 Ma eps = +1.7 A1797px

0.5124 MSWD = 1.5 n=4 Nd 144 A1797 0.5120 A1797#2 Nd/ 143

0.5116 A1797plag#1 A1797plag#2

0.5112 0.10 0.12 0.14 0.16 0.18 0.20 147Sm/144Nd

Fig. 54. Sm–Nd isotope data for a whole-rock sample and mineral separates from the Törninkuru dyke A1797.

NE-trending dykes in Taivalkoski

0.5130

A1796 Kallioniemi dyke A1800px 0.5126 Age = 2352 ± 64 Ma eps = +0.5 A1798px MSWD = 0.0007 A1796px

Nd 0.5122 A1798 144 A1796 A1800

Nd/ 0.5118 A1798plag 143

A1800plag A1796plag 0.5114 A1798 Kontioluoma dyke (+) A1800 Murhiniemi dyke (x)

0.5110 0.09 0.11 0.13 0.15 0.17 0.19 0.21 147Sm/144Nd

Fig. 55. Sm–Nd isotope data for whole rock and mineral separates from the three dykes A1796, A1798 and A1800 from the Taivalkoski area.

59 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye childhood home of the renowned novelist Kalle are generally consistent with the U–Pb ages for Päätalo. What is exposed is a NE–SW trending the same samples. Fe-tholeiitic diabase dyke mainly consisting of pla- 2) Major metamorphic effects on plagioclase may gioclase and clinopyroxene that are coarse grained be questioned, since the analysed plagioclase and well preserved in the middle parts of the dyke. appears fairly fresh. The Sm–Nd analyses of clinopyroxene, plagioclase 3) The Nd contents in pyroxene and plagioclase fractions and a whole-rock aliquot from sample are low compared to the amount of Nd in the A1799 provide an isochron with an age of 2352 ± whole-rock sample. Thus, as much of the REE

62 Ma and an initial εNd value of +0.5 (Fig. 55). must be in the interstitial minor phases, they also control the whole-rock Nd isotope com- 4.5.5 A1800 Murhiniemi position. As the country rocks are Archaean gneisses, the Nd isotope composition of these Another sample from a NE–SW-trending fluids very probably had a low 143Nd/144Nd ratio Fe-tholeiitic dyke was collected at Murhiniemi compared to the mafic magma, resulting in a

(A1800, AD91-3), ca. 2 km east of the sample loca- low εNd value in the interstitial and grain bound- tion of the above-discussed Kallioniemi dyke. The ary domains. main primary minerals are again plagioclase and clinopyroxene, which in hand samples and under One may also note that minerals used for isotope the microscope appear well preserved. Sm–Nd analyses in the Taivalkoski batch (A1794-A1802) analyses of minerals and whole rock did not, how- were not properly hand-picked, and even though ever, provide any unambiguous isochron (MSWD = the fractions were generally good looking, they 5.9). We may note that the data on pyroxene and may nevertheless have contained minor impuri- whole rock plot roughly on the isochron related to ties. Similar problems were also encountered dur- the Kallioniemi dyke (Fig. 55), but as the analysis ing this work with other than Taivalkoski samples of plagioclase is clearly above that line, the age of (see below), and clearly more detailed studies would this rock remains unknown. Conceivable explana- be required to gain a better understanding of the tions for the heterogeneity include the following: factors affecting the results from these rocks. 1) Analytical errors (e.g., spike–sample disequilibrium with the plagioclase analysis, resulting 4.5.6 A1798 Kontioluoma in biased Sm/Nd); 2) Metamorphic effects restricted to plagioclase. In A third sample (A1798, AD85-3) from the NE– this case, analyses of pyroxene and whole rock SW-trending Fe-tholeiitic dykes in the Taivalkoski would give the age of crystallisation: 2295 ± area was collected at Kontioluoma ca. 20 km NE

85 Ma, εNd = +0.6; of the Kallioniemi site. The major primary miner- 3) Crustal contamination after crystallisation of als, plagioclase and clinopyroxene, are again well pyroxene and plagioclase (analyses of these preserved, but nevertheless the analyses on these minerals would then give the age of crystal- mineral concentrates and whole rock do not deter-

lisation: 2204 ± 48 Ma, εNd = +0.6); mine any isochron (MSWD = 20!). The possible 4) Pyroxene represents an early phenocryst phase, explanations for this discrepancy were discussed and the rest of the rock was contaminated (not above. Again, the analyses on pyroxene and whole applicable for this sample). rock from sample A1798 plot very close to the 2.35 Ga isochron defined by the Kallioniemi A1796 data The following comments to these alternative sce- (Fig. 55). Considering the other option, i.e., crustal narios can be made: contamination in the groundmass/whole rock, the 1) Duplicate analysis, carried out before and dur- analyses of pyroxene and plagioclase give an age of

ing the work presented in this paper, suggests 2086 ± 63 Ma (εNd +0.2). The calculated εNd(2086 Ma) that an analytical error is unlikely. Furthermore, value for the whole-rock sample is -0.7. We can many examples in this paper also suggest that only conclude that from the available Sm–Nd data, ages obtained by the Sm–Nd method from the emplacement age of the sampled dyke remains well-preserved primary minerals/whole rocks unknown.

60 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Tilsanvaara dykes 0.5132 A1795 WNW-trending dyke 0.5128 Age = 2233 ± 42 Ma A1794px eps = +1.0 A1795px MSWD = 0.71 n=4 0.5124 Nd A1795 (AD89-8) A1794 (AD89-10) 144 0.5120 Nd/

143 0.5116 A1795plag#2 A1795plag A1794plag A1794 NW-trending dyke Age = 2219 ± 37 Ma 0.5112 A1794plag#2 eps = +0.2 MSWD = 1.4 n=4 0.5108 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22 147Sm/144Nd Fig. 56. Sm–Nd isotope data for whole-rock samples and mineral separates from the Tilsanvaara dykes A1794 and A1795.

4.5.7 A1794, A1795 Tilsanvaara clinopyroxene, appear relatively well preserved, but the mineral concentrates obtained by separation of At the Tilsanvaara site, two mafic dykes with clearly sample A1802 were not entirely clean. Nevertheless, different trends are observed in the same outcrop. the Sm–Nd compositions of these concentrates and The samples picked for isotope work include A1794 the whole-rock powder define an isochron giving

(AD89-10) from a NW–SE-trending Fe-tholeiitic an age of 2058 ± 35 Ma and an εNd value of -0.1 (Fig. dyke and A1795 (AD89-8) from a WNW–ESE- 57). The amount of Nd in the analysed plagioclase trending tholeiitic dyke. The main minerals, pla- fraction is relatively high (8.7 ppm), which warrants gioclase and clinopyroxene, are well preserved in speculation concerning whether the age obtained both diabase samples. The Sm–Nd data on sample really dates the igneous crystallisation event. More A1794 give an isochron and age of 2219 ± 37 Ma, detailed studies would be required to resolve this. with an initial εNd value of +0.2 (Fig. 56). The Sm– Nd isochron age obtained for the tholeiite sample 4.5.9 A1801 Hirsikangas

A1795 is 2223 ± 45 Ma and the initial εNd value +1.0. Most of the plagioclase was found in a slightly mag- A third dyke sample, A1801 (MLJ40-1), from the netic fraction after Frantz isodynamic separation. NW–SE-trending Fe-tholeiitic swarm in the area of The analyses conducted on plagioclase from both high-grade gneisses around the town of Taivalkoski, the magnetic (plag#2) and non-magnetic fractions was collected at Hirsikangas. The main minerals plot on the isochron. in this sample are plagioclase and clinopyroxe, which appear relatively well preserved, although 4.5.8 A1802 Koivuvaara the plagioclase used for analysis was slightly yellowish. As with samples A1800 and A1798 above, the The Koivuvaara dyke ca. 6 km NW of Tilsanvaara is three Sm–Nd analyses conducted on the mineral associated with a NW–SE-trending narrow aero- separates and whole-rock powder do not yield any magnetic anomaly, which suggests that the dyke is isochron (MSWD = 47!). The age remains unknown, at least 10–15 km in length. In thin section, the main but in any case, the analysis of pyroxene suggests minerals of this Fe-tholeiitic dyke, plagioclase and a positive initial εNd.

61 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.5128 A1802 Koivuvaara dyke Age = 2058 ± 35 Ma A1802px eps = -0.1 0.5124 MSWD = 0.15

Nd 0.5120

144 A1802 (MLJ29-1) Nd/ 0.5116 143

0.5112 A1802plag

0.5108 0.07 0.09 0.11 0.13 0.15 0.17 0.19 0.21 147Sm/144Nd Fig. 57. Sm–Nd isotope data for whole rock and mineral separates from the Koivuvaara dyke A1802.

4.6 Volcanic rocks in the Kuusamo schist belt

Mafic volcanic rocks from three stratigraphic lev- tration of U in zircon is anomalously low (<30 ppm). els in the Kuusamo belt were targets for recon- Three TIMS analyses provided concordant results naissance Sm–Nd studies in the early 1990s (Fig. and gave an average age of 2428 ± 3 Ma (Fig. 58, 38). The andesites from the lowermost unit, the Appendix 5). Analyses by LA-MC-ICPMS were Kuntijärvi Formation (Greenstone I by Silvennoinen carried out during two different sessions with dif- 1991), are enriched in LREE and yield an average ferent standards. The analyses revealed that the 206 204 εNd(2400 Ma) value of -2.7 and TDM = 2.82 Ga Pb/ Pb ratios are much higher than in the TIMS (Appendix 6). In contrast, the basaltic sample from data (Appendix 7). The two sessions differed slightly the Petäjävaara Formation (Kuusamo Greenstone II in terms of the obtained Pb/U ratios and errors in by Silvennoinen 1991) has a nearly chondritic Sm/ Pb/U, which is due to differences in calibration.

Nd ratio and a clearly positive initial εNd value. The Nevertheless, the results are consistent within third unit, the Ruukinvaara Formation (Greenstone error, and the calculated 207Pb/207Pb ages are very

III Silvennoinen 1991), has yielded an initial εNd similar from both sessions. Rejecting one analysis, value close to zero. the data yield an average 207Pb/207Pb age of 2428 ± The ages of these mafic units are constrained by 4 Ma. intruding mafic sills dated at 2.22 Ga and by under- It can be concluded that the age of zircon from lying conglomerate, which contains clasts with sample A1868 is 2428 ± 3 Ma, which should also an age of ca. 2.4 Ga (Silvennoinen 1991). We have date the formation of the porphyry and constrain recently obtained a new U–Pb age for a porphyry the maximum depositional age of the Kuntijärvi clast collected from the Kuntijärvi conglomerate conglomerate and overlying Kuntijärvi Formation

(A1868). Abundant zircon extracted from sample volcanic rocks. The initial εNd(2428 Ma) value for A1868-Kuntijärvi occurs as light-coloured, fairly this sample is -2.7 (TDM = 2.81 Ga), and thus transparent, euhedral to subhedral prisms or frag- identical with the εNd of the andesites from the ments. Both TIMS and LA-MC-ICPMS were used for Kuntijärvi Formation. U–Pb dating. The U–Pb data reveal that the concen-

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1868 Kuntijärvi porphyry clast in cgl 0.49 data-point error ellipses are 2s

A1868 Kuntijärvi TIMS 2500 0.47 Concordia Age = 2428 3 Ma n=3 2460 2420 0.45 2380 U 2340 238 0.43 2300

Pb/ 2260

206 0.41 LA-MC-ICPMS Intercepts at 0.39 0 0 & 2428 4 Ma MSWD = 0.88 n=25

0.37 8.0 8.4 8.8 9.2 9.6 10.0 10.4 10.8 207Pb/235U

Fig. 58. Concordia plot of U–Pb zircon data obtained from the porphyry clast in the Kuntijärvi Formation conglomerate. LA-MC-ICPMS analyses are presented as (large black) error ellipsoids. The three ID-TIMS analyses are coeval and concordant and shown as small error ellipsoid.

5 PUDASJÄRVI COMPLEX AND THE PERÄPOHJA SCHIST BELT

5.1 Geological background

The Pudasjärvi complex consists of Archaean mig- parental magma of the layered intrusions in the matitic gneisses and amphibolites, the ca. 2.82 Ga Tornio-Näränkävaara belt (Iljina & Hanski 2005, Oijärvi greestone belt, mafic and felsic intru- Yang et al. 2016). A younger ~2.1 Ga Fe-tholeiitic sive rocks with ages mostly close to 2.7 Ga, and dyke swarm is represented by the Sipojuntti dyke paragneisses deposited after 2.74 Ga (Lauri et al. close to the coast of the Bothnian Bay, which has 2011, Huhma et al. 2012a). The oldest rocks in the been dated at 2118 ± 14 Ma (Perttunen & Vaasjoki Fennoscandian Shield, the 3.5 Ga Siurua tonalite 2001). gneisses, are located in the Pudasjärvi complex The Palaeoproterozoic Peräpohja schist belt lies (Mutanen & Huhma 2003). unconformably on the Archaean Pudasjärvi com- There is evidence for the existence of mafic plex (Fig. 59). The belt is ca. 170 km long and 80 dyke swarms of the ~2.45 Ga, ~2.1 Ga and ~1.98 Ga wide and records ca. 500 Ma of geologic history age groups in the Pudasjärvi complex, but the pub- in a stratigraphic succession attaining ca. 5 kilo- lished U–Pb and Sm–Nd isotope data are still scarce metres in thickness. The lithostratigraphy of the (Vuollo & Huhma 2005). The oldest dykes are rep- Peräpohja belt was summarised and partly revised resented by a low-Ti tholeiitic dyke swarm trend- by Kyläkoski et al. (2012) and is presented in Figure ing 330°. It has been dated, but the Sm–Nd age of 60. The supracrustal sequence is divided into two 2461 ± 150 Ma is imprecise and the U–Pb data just major lithostratigraphic units, the Kivalo and give a minimum age of 2378 Ma. One of the 2.45 Ga Paakkola Groups. The Kivalo Group starts with basal dykes is the gabbro-noritic Loljunmaa dyke, which conglomerates, which were locally deposited on is located on the southeastern side of the Penikat partly eroded ca. 2.44 Ga mafic layered intrusions, intrusion and is considered a candidate for the marking the maximum sedimentation age of the

63 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye sequence. The conglomerates are overlain by suba- minor stromatolitic dolomite interbeds, followed by erially erupted basalts of the Runkaus Formation a cyclic repetition of formations composed of mafic and thick orthoquartzite and arenite strata of the volcaniclastic rocks and sedimentary carbonate Palokivalo Formation. The exact depositional age rocks. The former are represented by mafic tuffs and of these formations is uncertain, although it is tuffites of the Tikanmaa, Hirsimaa and Lamulehto known that they are older than ca. 2.22 Ga, which Formations, and the latter by dolomites and phyl- is the age of cutting mafic-ultramafic sills (Hanski lites of the Poikkimaa Formation and stromatolitic et al. 2010). The Palokivalo Formation is overlain dolomites of the Rantamaa Formation (Fig. 60). by the Petäjäskoski Formation, which is composed The dolomitic rocks record a typical Lomagundi- of hematite-rich phlogopitic-sericitic and albitic Jatuli positive carbon isotope anomaly (Karhu 1993, schists with quartzite and dolomite interbeds. It was 2005). Important age constraints for the Peräpohja deposited earlier than 2.14 Ga, as indicated by the belt were obtained from a tuff of the Hirsimaa presence of mafic sills of this age (see below). The Formation, for which Karhu et al. (2007) determined following unit is a thick suite of subaerially erupted, a U–Pb zircon age of 2106 ± 8 Ma (Fig. 59). LREE-depleted continental flood basalts belonging The second main unit, the Paakkola Group to the Jouttiaapa Formation (Perttunen & Hanski (>1–2 km), begins with the Martimo Formation 2003). Using the whole-rock Sm–Nd method, the (Martimo suite in Nironen et al. 2016), which formation was dated at 2090 ± 70 Ma by Huhma et comprises turbidites and mica schists with graph- al. (1990) and, including a few subsequent analy- ite- and Fe sulphide-bearing black schist interlay- ses, the age has been refined to 2105 ± 50 Ma (this ers. Ranta et al. (2015) determined ages of detrital paper). zircon grains from the upper part of the Martimo The upper part of the Kivalo Group is composed of Formation, showing that these rocks were depos- orthoquartzites of the Kvartsimaa Formation having ited later than ca. 1.91 Ga. Pillowed basalts of the

^_ Kuusivaara �ou�aapa Fm ¢ ^_ ^_ R��jänkä ^_^_Koppakumpu ^_ ^_ ^_Hirsimaa Fm ^_ ^_ A0859 ^_ A0454 ^_ Tikanmaa Fm ^_ ^_ ^_ ^_^_ Runkausvaara ^_ Runkaus Fm Intrusions and volcanic rocks A0703 ^_Penikat ^_ ^_^_^_ Sm-Nd (±U-Pb) ^_^_ Loljunmaa (LD) ^_ A0662 ^_A0603 Tervonkangas Volcanic rocks ^_ ^_ ^_ Kemi U-Pb ( in this paper )

^_ Norway ^_ U-Pb (age published elsewhere) Uolevinlehto (UD) ^_ Sm-NdXW (age ?) Russia 2.4 - 2.5 Ga felsic rocks ( in this paper ) Sweden Age ^_ <1930 Ma Palomaa ^_ 1931 - 2080 ^_ ^_ 2081 - 2180 Finland ^_ ^_ 2181 - 2280 Vengasvaara ^_ 2281 - 2380

^_ >2380 0 30 km

Fig. 59. Geological map of the Pudasjärvi complex and Peräpohja schist belt showing sample localities. For symbols, see Figure 1.

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Fig. 60. Lithostratigraphy of the Peräpohja schist belt (modified after Kyläkoski et al. 2012). Labelled boxes represent samples of this study.

Väystäjä Formation are spatially associated with formations in the recent tectonostratigraphic com- mica schists, but the felsic porphyries within the pilation (Nironen et al. 2016). Väystäjä volcanic rocks have yielded a U–Pb zir- In this paper, we document U–Pb zircon and con age of 2050 ± 8 Ma (Perttunen & Vaasjoki whole-rock Sm–Nd isotope data for several sam- 2001, Lahtinen et al. 2015a), suggesting that these ples from the ca. 2.44 Ga layered intrusions and rocks are older than the upper part of the Martimo one coeval “boninitic” dyke (Loljunmaa), all close Formation and probably represent an allochthonous to the southern margin of the Peräpohja belt. Four unit. other dykes further to the southeast in the basement Among the youngest supracrustal rocks in the were also studied, with two of them (Uolevinlehto, Peräpohja belt are felsic and mafic tuffs of the Vengasvaara) being ca. 2.44 Ga in age and the other Korkiavaara Formation (Oikarila suite in Nironen two younger (Palomaa, Tervonkangas), with an age et al. 2016) and mica schists of the Pöyliövaara of 2.0–2.1 Ga. From the Peräpohja schist belt, we Formation (Oikarila suite in Nironen et al. 2016), report isotope data for two 2.13–2.14 Ga mafic dykes both of which contain zircon dated at ca. 1.98 Ga cutting the Petäjäskoski and Tikanmaa Formations (Hanski et al. 2005, Lahtinen et al. 2015a). These (Figs. 59, 60) and one 2.08 Ga mafic dyke cutting the rock units occur in the northern part of the belt and Väystäjä Formation. Mafic lava flows or tuffs were are also in tectonic contact with associated suprac- analysed for Sm–Nd isotopes from the Runkaus, rustal rocks of the belt. Thus, these rock units are Jouttiaapa, Tikanmaa and Hirsimaa Formations. redefined as suites instead of lithostratigraphic

5.2 The 2.44 Ga Kemi, Penikat, Kilvenvaara and Siikakämä intrusions

The first U–Pb zircon analyses by TIMS of sample concordant data suggesting an age of 2433 ± 4 Ma A662 from the Kemi gabbro were conducted at GTK (Perttunen & Vaasjoki 2001). We have re-analysed by O. Kouvo in 1976. These analyses yielded nearly zircon from the same sample using LA-MC-ICPMS

65 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A662 Kemi intrusion gabbro 0.58 data-point error ellipses are 2s

0.56

0.54

0.52 U 238 0.50 2600 Pb/ 2560 206 0.48 2520 2480

0.46 2440 2400 Average Pb/Pb age 2436 ± 8 Ma 0.44 MSWD = 1.9 n=10 TIMS

0.42 9.4 9.8 10.2 10.6 11.0 11.4 11.8 12.2 207Pb/235U

Fig. 61. Concordia plot of U–Pb zircon data for sample A662 from the Kemi intrusion gabbro. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red triangles.

(Fig. 61, Appendix 9). The U concentration in zir- order to complement the isotope database of the con is very low (20 ppm), and a large spot size Kemi-Penikat layered intrusions, one Sm–Nd anal- (50 µm) together with low-U standards was used in ysis was conducted on the Kemi intrusion using the analyses. The data tend to plot above the con- the U–Pb dated gabbroic sample A662 discussed cordia curve, which is probably due to an incompat- above. This sample yielded a composition that is ible standard. Nevertheless, this has only a minor now found to be exactly on the isochron as defined effect on the Pb/Pb isotope ratios, and an average by the Penikat sample with an εNd(2440 Ma) value of 207Pb/206Pb age of 2436 ± 8 Ma can be calculated, –1.6 (Fig. 65, Appendix 1). The same average initial which is highly consistent with the TIMS result value was also recently obtained by Maier et al (in obtained 40 years earlier. press) for seven samples from the Penikat intrusion. Zircon grains from two previous samples (A603, Two old unpublished discordant TIMS analy- A703) from the Penikat layered intrusion were also ses on multigrain zircon fractions are available for analysed by LA-MC-ICPMS (Fig. 62, Appendix 9). sample A859 taken from the Kilvenvaara gabbro in The bulk of the new data on both samples are con- the Portimo Complex (Narkaus layered intrusion in cordant within error and yield an average 207Pb/206Pb Nironen et al. 2016, Fig. 59). These data suggest an age of 2437 ± 6 Ma. Most data were obtained from age about 2.4 Ga (Alapieti et al. 1989). Recent U– sample A603, providing an age of 2444 ± 8 Ma Pb analyses by LA-MC-ICPMS performed on grains (n = 16, Maier et al. in press). A few data points from these same fractions provide concordant data suggest younger ages, probably due to alteration. and an age of 2430 ± 7 Ma (Fig. 63, Appendix 9). The strongly discordant TIMS data are consistent We analysed zircon from the Siikakämä gab- with this age (Fig. 62, Perttunen & Vaasjoki 2001). bro in the Narkaus layered intrusion, which is also The first ever Sm–Nd analyses on the 2.44 Ga considered a member of the chain of 2.44 Ga mafic mafic rocks in Finland were performed on samples intrusions that flank the southeastern margin from the Penikat layered intrusion, which provided of the Peräpohja schist belt (A454, Fig. 59). The an initial εNd value of –1.6 (Huhma et al. 1990). In TIMS results on multigrain zircon fractions from a

66 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

data-point error ellipses are 2s 0.6 Penikat intrusion gabbros (A603 & A703)

LA-MC-ICPMS 0.5 average 207Pb/206Pb age 2600 2437 ± 6 Ma (MSWD = 1.3, n=18/23) 2400 U

238 2200 0.4

Pb/ A603 TIMS Intercepts at 2000

206 797±230 & 2428 ± 35 Ma MSWD = 22 1800 TIMS (n=5 baddeleyite ± zircon) (Perttunen & Vaasjoki 2001) 0.3 1600 A703 TIMS Intercepts at 574±110 & 2404 ± 25 Ma MSWD = 25 n=6 (Perttunen & Vaasjoki 2001)

0.2 3 5 7 9 11 13 207Pb/235U

Fig. 62. Concordia plot of U–Pb zircon data from the Penikat intrusion gabbro samples A603 and A703. LA-MC- ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as dots (A703 as green).

data-point error ellipses are 2s

0.56 A859 Kilvenvaara gabbro

0.52

2600

U 0.48

238 2500 Pb/ 2400 206 0.44 2300

2200 0.40 Concordia Age = 2430 ± 7 Ma (n=6/7) 2100

0.36 6.5 7.5 8.5 9.5 10.5 11.5 12.5 207Pb/235U

Fig. 63. Concordia plot of U–Pb zircon data from the Kilvenvaara gabbro A859, Portimo Complex. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as triangles.

67 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A454 Siikakämä gabbro data-point error ellipses are 2s

Range of Pb/Pb ages from 2.43 Ga to 1.8 Ga 0.5 19 analyses on five grains 2500

2300 U 0.4

238 2100 A454D 4.0-4.2 >160µm A454C 4.2-4.6 Pb/ 1900 A454B +4.6 HF leached 206 A454A +4.6 clear 1700 0.3 A454E 3.8-4.0 >160µm A454F 3.8-4.0 70-160µm 1500

1300

0.2 2 4 6 8 10 12 207Pb/235U Fig. 64. Concordia plot of U–Pb zircon data from the Siikakämä gabbro A454. LA-MC-ICPMS analyses shown as error ellipsoids and ID-TIMS analyses as black dots.

gabbro pegmatoid sample A454 are heterogene- erogeneous results. However, it seems obvious ous and rather unusual, as the youngest apparent that the most pristine domains with ages close to Pb–Pb ages were obtained for the heaviest of the 2.43 Ga would give a reasonable estimate of the analysed fractions (Mertanen et al. 1989). The 19 igneous age, whereas ages from altered domains spot analyses by LA-MC-ICPMS also yielded het- range down to ca. 1.8 Ga (Fig. 64, Appendix 9).

5.3 Loljunmaa gabbro-noritic dyke

Geological and geophysical surveys undertaken by primary magmatic mineralogy is not preserved and GTK on the Archaean basement area east of the thus only a whole-rock powder has been used for Penikat layered intrusion have revealed a gabbro- Sm–Nd work. The Loljunmaa sample 4-LD-93 pro- noritic dyke at Loljunmaa, which very probably vides an εNd(2440 Ma) value of -1.2 (Appendix 1), represents a magma conduit to the system that pro- and in terms of the REE level and isotopic composi- duced the large 2.44 Ga layered intrusions (Alapieti tion, it is almost identical to the Viianki whole-rock et al. 1990, Iljina & Hanski 2005, Yang et al. 2016). sample considered below (Chapter 7.2). The compo- Consequently, chemical and isotopic composi- sition also plots very close to the isochron defined tions of the dyke have been utilised to estimate the by the Penikat intrusion samples, being consistent parental magma composition of the layered intru- with the comagmatic nature of the dyke and intrusions (Iljina & Hanski 2005, Yang et al. 2016). The sion (Fig. 65). dyke trends NW–SE and is 20–30 m in width. The

68 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5126 Penikat intrusion Age = 2422 ± 52 Ma A1012 opx eps = -1.6 0.5122 MSWD = 0.95 n=5 (Huhma etal 1990)

0.5118 A662 Elijärvi

Nd HH/13

144 LD-4-93 Loljunmaa 0.5114 A1012 wr Nd/ 143 A703 0.5110

A1012 plag 0.5106

0.5102 0.04 0.08 0.12 0.16 0.20 147Sm/144Nd

Fig. 65. Sm–Nd isotope data for whole rock and mineral separates from the Penikat layered intrusion (Huhma et al. 1990) and whole-rock samples from Kemi intrusion (A662) and Loljunmaa dyke.

5.4 Tholeiitic dykes, A1410 Uolevinlehto, Pudasjärvi

The Uolevinlehto dyke (A1410, UD) represents the but ca. 2.44 Ga is still conceivable, as the data plot low-Ti tholeiitic NW–SE-trending dyke swarm on a 1.88–2.45 Ga chord (see Fig. 49). in the Pudasjärvi complex (Fig. 59). In the out- The plagioclase in the rock is slightly cloudy. The crop, the rock is fresh and consists of plagioclase five Sm–Nd analyses available are technically good, (50%), clinopyroxene (35%), and minor amounts which is supported by the duplicated analyses con- of orthopyroxene (2%), quartz, secondary amphi- ducted on whole rock and pyroxene (Appendix 1). bole and Fe–Ti–Voxides. The dykes are more than However, the data points do not define any well- 2 km long and normally 40–70 m wide. The studied aligned isochron, but show some scatter in excess of dyke outcrop is about 20 m x 100 m is size, but no analytical error. For the five analyses, the ISOPLOT contacts to the country rocks have been observed. programme gives an age of 2447 ± 160 Ma (εNd = The rock displays a cumulus texture. Plagioclase +0.4, MSWD = 16, Fig. 66). The age calculated for is An50–60 in composition and clinopyroxene is pyroxene and plagioclase alone is about the same, Ca-rich augite with an average composition of but the initial εNd value is slightly higher (+0.7).

Wo36En43Fs20. One feature in common with other Excluding plagioclase, the four analyses of whole 2.45 Ga dyke swarms is that plagioclase is faintly rock and pyroxene yielded an age of 2537 ± 37 Ma to strongly cloudy. A large sample (35.4 kg) was (MSWD = 1.8), whereas plagioclase and whole rock collected for U–Pb dating, but only 18 baddeleyite gave an age of 2265 ± 58 Ma. Comparing with the grains were recovered (in Toronto). Some of the U–Pb data, neither of these figures can be regarded grains were of good quality, but some have inclu- as the magmatic age estimate for the dyke. The Nd sions and cracks and others have a thin (1–5 µm) concentration in the whole-rock sample is much rim of zircon. The best 12 grains were used for anal- higher than in pyroxene and plagioclase. Thus, a ysis, which yielded a 2.1% discordant U–Pb result significant amount of Nd is situated outside these and a minimum age of 2366 Ma (207Pb/206Pb age, main minerals. This, together with the obtained

Appendix 8). The exact age remains non-precise, disequilibrium in the initial εNd values between the

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.5136 A1410 Uolevinlehto tholeiitic dyke 0.5132 A1410 px A1410 px#2 0.5128

Nd 0.5124 144

Nd/ 0.5120 A1410 143 A1410#2 0.5116 Age = 2447 160 Ma 0.5112 A1410 plag eps = +0.4 MSWD = 16 n=5

0.5108 0.08 0.12 0.16 0.20 0.24 147Sm/144Nd

Fig. 66. Sm–Nd isotope data for whole rock and mineral separates from the Uolevinlehto tholeiitic dyke A1410.

whole-rock powder and separated major miner- composition of the magma may be best estimated als, suggests that some Nd in the rock was derived using the pyroxene (and plagioclase) composition, from crustal sources after the crystallisation of the which suggests a slightly positive εNd value of ca. major minerals. Consequently, the initial isotope +0.6 at 2.44 Ga.

5.5 The 2.44 Ga Vengasvaara intrusion

The Vengasvaara (Kärppäsuo) mafic intrusion an age of 2444 ± 4 Ma (Fig. 67). within the Archaean Pudasjärvi complex was dis- Sm–Nd analyses of the mineral concentrates and covered by drilling during a GTK research project whole-rock powder are technically good but do not (Fig. 59). Sample A1744 collected from a drill core yield any isochron (Fig. 68). The age calculated from is a medium-grained gabbro, with the main miner- the analyses of “pyroxene” and whole rock is 2507 als being plagioclase and clino- and orthopyroxene. ± 72 Ma, which is compatible with the U–Pb age The pyroxenes are slightly altered to amphibole. mentioned above. The date derived from plagio- Minor minerals include quartz, biotite, apatite and clase and whole rock is ca. 2.05 Ga and probably opaque minerals. Attempts were made to separate reflects metamorphic effects. The calculated ini- the main minerals for Sm–Nd analysis, but it turned tial Nd isotope ratio for the whole-rock sample at out that the quality of the separates was poor and 2.44 Ga is very unradiogenic (εNd = -5.4) and much too difficult for hand-picking. Plagioclase is partly lower than the ratios measured for other 2.4 Ga cloudy and the pyroxene concentrate is heteroge- rocks in the shield. This suggests a significant con- neous in colour. However, a few grains of brown, tribution of old LREE-enriched lithosphere in the transparent and anhedral zircon were obtained from genesis of this intrusion. It is tempting to relate the heavy (>3.6 g/cm3) fraction. this with the 3.5 Ga old Siurua gneisses, which out- Six zircon grains were analysed using SHRIMP at crop only 10 km NE of Vengasvaara (Mutanen & the VSEGEI laboratory in St. Petersburg. The U–Pb Huhma 2003). The εNd(2440 Ma) value for the Siurua data (Appendix 4b) reveal very high contents of U gneisses is still much lower (ca. -15, Mutanen & and Th in zircon. However, the compositions of all Huhma 2003, Huhma et al. 2012b). the six grains are practically concordant and yield

70 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1744 Vengasvaara gabbro 0.51 SHRIMP data-point error ellipses are 2s

Concordia Age = 2441 ± 4 Ma (n=3)

0.49

2530 U 2510

238 2490 0.47 2470

Pb/ 2450 2430 206 2410 0.45 Average Pb/Pb age 2444 ± 3 Ma MSWD = 1.5 (n=6)

0.43 9.6 10.0 10.4 10.8 11.2 207Pb/235U

Fig. 67. Concordia plot of U–Pb zircon data obtained by SIMS from the Vengasvaara gabbro A1744.

5.6 Palomaa dyke, A1743

The NW–SE-trending Palomaa dyke forms an aero- erals. The mineral concentrates from the drill core magnetic anomaly cutting the Archaean gneisses ca. sample A1743 used for Sm–Nd analyses are fairly 5 km NW of the Vengasvaara intrusion discussed fresh, although pyroxene appears slightly hetero- above. The main minerals of the 30-m-wide dyke geneous. Plagioclase is relatively dark. The Sm–Nd are plagioclase and clino- and orthopyroxene, data yield an isochron that gives an age of 2077 ± which appear slightly altered in thin section. Brown 34 Ma (Fig. 68). The calculated initial εNd value of amphibole and biotite are common and carbonate, +1.1 suggests an origin from depleted mantle for apatite and opaque are among the accessory min- the magma.

5.7 Tervonkangas dyke A1808

Drilling by GTK of an aeromagnetic anomaly at analytical results are technically good, but it is Tervonkangas, Ranua, revealed a mafic–ultra- still questionable whether a rock with such a well- mafic dyke complex within the Archaean gneisses, preserved primary texture and minerals is indeed having a length of ca. 3 km and a width of a few Archaean in age. In order for this date to represent hundred metres. Sample A1808 taken from a the real magmatic age, these two basic assumptions NE–SW-trending dyke in the complex contains of isochron dating should be valid: 1) all miner- extremely well-preserved primary igneous min- als crystallised at the same time and had the same erals: large (>5 mm), zoned augite phenocrysts, initial Nd isotope composition and 2) the Sm–Nd large biotite oikocrysts, olivine and plagioclase. system remained closed after the formation of these The Sm–Nd analyses conducted on good quality minerals. In this particular case, it is possible that mineral separates and a whole-rock powder defined large augite phenocrysts crystallised in the magma an isochron with a date of 2623 ± 34 Ma and initial already prior to its final emplacement and solidi- εNd value of +1.0 (MSWD = 1.7, Fig. 69, Appendix fication. Assuming further that significant crustal 1). Taking into account the geological setting, the contamination took place between these two stages,

71 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A1743 Palomaa dyke Age = 2077 ± 34 Ma 0.513 epsilon= +1.1 A1743 px MSWD = 0.77

Nd 0.512 A1743 wr 144 A1744 "px" Nd/ A1743 plag

143 A1744 wr 0.511 A1744 Vengasvaara wr & "px": Age = 2507 ± 72 Ma A1744 plag epsilon = -4.6

0.510 0.06 0.10 0.14 0.18 0.22 147Sm/144Nd

Fig. 68. Sm–Nd isotope data for whole-rock samples and mineral separates from the Vengasvaara gabbro A1744 and Palomaa dyke A1743.

0.5130 A1808 Tervonkangas mafic dyke

0.5126 Reference "isochron" 2100 Ma A1808px Eps = -0.7 0.5122 A1787 Heinisuo A1808

0.5118 A1687 Soidinmaa

143 Nd 0.5114 144Nd A1808: 0.5110 The analyses define likely a mixing line and the age 0.5106 has no meaning. A1808plag A1808 Tervonkangas dyke A1808plag#2 Age = 2623 ± 34 Ma 0.5102 epsilon = +1.0 MSWD = 1.7 n=4 0.5098 0.04 0.08 0.12 0.16 0.20 147Sm/144Nd

Fig. 69. Sm–Nd isotope data for whole rock and mineral separates from the Tervonkangas dyke A1808.

72 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

plagioclase and whole rock would have had lower εNd value becomes slightly negative, being -1.9 at 143Nd/144Nd ratios by the time of crystallisation. In 2.0 Ga, for example. this case, the first assumption of meaningful isoch- Attempts to obtain zircon or baddeleyite for ron dating would not be valid and the isochron dating were not successful with the Tervonkangas steepened by contamination would yield an age that sample. The same problem was also faced with is too old. We are inclined to accept this interpreta- the NW–SE-trending Heinisuo mafic dyke A1787, tion. We recall that initial isotopic disequilibrium occurring 15 km NE of the site of sample A1808. between pyroxene and whole rock was also inferred A Sm–Nd analysis conducted on whole-rock pow- in the case of the Pääjärvi/Lupzenga dyke (A1465) der yielded an initial εNd value of -1.2, if the age is treated above. Assuming a Palaeoproterozoic age assumed to be 2 Ga (Appendix 1). for the Tervonkangas dyke, the calculated initial

5.8 The 2.13–2.14 Ga dykes in the Peräpohja schist belt, A1214 Koppakumpu and A2087 Kuusivaara

According to Perttunen & Hanski (2003), mafic intercepts with the condordia curve at 75 ± 110 and tuffs of the Tikanmaa Formation in the Peräpohja 2129.5 ± 4.2 Ma (MSWD = 1.19, Fig. 70, Appendix schist belt are cut by a diabase dyke at Koppakumpu. 4b). Most of the data points are concordant despite The Tikanmaa tuffs overlie the metabasalts of the the relatively high concentrations of U. A special Jouttiaapa Formation in the Peräpohja stratigraphy feature of the zircon is its high Th/U ratio, although (Fig. 59). Sample A1214 was already collected and this is not uncommon for zircon in mafic dykes. processed for zircon in 1990, but due to the very The high levels of U and Th are consistent with the small number of recovered zircon grains, it was dark and turbid appearance of the zircon grains. not analysed until the access to in situ analytical Despite of the radiation damage, magmatic zon- techniques. The U–Pb data obtained by SHRIMP (St ing is still clearly visible in the CL images of these Petersburg) are technically of good quality, and ten zircon grains (Fig. 70). analyses on nine grains provide a chord that has

A1214 Koppakumpu diabase 0.44 SHRIMP data-point error ellipses are 2s Intercepts at 0.42 2130 ± 4 & 75 ± 110 Ma MSWD = 1.19 n=10 2200 0.40 2160

2120 U 2080 R4F2t.jpg:A1214 grains 6,4,5,7 0.38

238 2040 2000 Pb/ 0.36 1960

206 1920 0.34

0.32

0.30 5.4 5.8 6.2 6.6 7.0 7.4 7.8 207 235 Pb/ U lenght of grain 5: 120µm

Fig. 70. Concordia plot of U–Pb zircon data obtained by SIMS from the Koppakumpu diabase A1214.

73 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

The magmatic zoning preserved in Kuoppakumpu the Petäjäskoski Formation, which lies directly zircon grains backs the interpretation that the below the Jouttiaapa Formation in the Peräpohja obtained U–Pb age of 2130 ± 5 Ma can be con- lithostratigraphy (Kyläkoski et al. 2012). This age sidered the igneous crystallization age of the was based on U–Pb isotope analyses using the Koppakumpu diabase. This gives age constraints LA-MC-ICPMS technique on zircon grains from a for the supracrustal rocks. Accordingly, the basalts gabbroic sample (A2087) picked from a drill core of the Jouttiaapa Formation should be older than intersecting the dated sill at Kuusivaara (Fig. 59). 2130 ± 5 Ma, which is (within error) in agreement Both the Kuoppakumpu and Kuusivaara mafic with the Sm–Nd age of 2103 ± 50 Ma available for dykes have a nearly chondritic Sm/Nd ratio and the Jouttiaapa metabasalts (Chapter 5.10). give clearly positive initial εNd values of +3.2 and A similar U–Pb zircon age of 2140 ± 11 Ma has +3.5 (Appendix 1). recently been published for a mafic sill intruding

5.9 Rytijänkkä dyke A854

The uppermost allochthonous stratigraphic unit of most pristine-looking domains (Fig. 71, Appendix the Peräpohja belt, the Väystäjä Formation, is cut by 9). Few measurements of altered zircon yielded a metadiabase dyke at Rytijänkkä. The dyke locates ages of ca. 1.8 Ga. Based on CL images, these grains within the mafic volcanic rocks of the Väystäjä may have baddeleyite cores. Two distinct grains Formation, ca. 600 m north of its contact to the are Archaean in age, representing either inher- underlying turbiditic Martimo Formation meta- ited Archaean zircon crystals or grains obtained sediments. Sample A854 was taken from a coarse- by contamination during sample processing (the grained, light-coloured diabase and processed for latter could have happened in the 1970s, when a zircon in the 1970s. The separation yielded a small large, difficult-to-clean jaw crusher, roller mill, and amount of fairly turbid zircon, which was recently Wilfley table were used in the processing). The Sm– studied by laser ablation MC-ICP-MS. The obtained Nd analysis of sample A854 indicated a chondritic spot data are scattered, but an igneous age of 2084 Sm/Nd ratio and an initial εNd of +1.0 (Appendix 1). ± 11 Ma can be constrained based on data from the

A854 Rytijänkkä diabase data-point error ellipses are 2s

0.48 Average Pb/Pb age 2084 11 Ma 0.44 MSWD = 0.33 n=11 2300

U 0.40 2100 238

0.36

Pb/ 1900

206 0.32 1700

0.28 Concordia Age = 2089 11 Ma 1500 n=11 (/20)

0.24 2 4 6 8 10 207Pb/235U

Fig. 71. Concordia plot of U–Pb zircon data obtained by LA-MC-ICPMS from the Rytijänkkä diabase A854.

74 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

5.10 Volcanic rocks

The age of the lowermost volcanic unit of the estimate is based on whole-rock Sm–Nd isotope Peräpohja schist belt, the Runkaus Formation is analyses. Combining the analytical data published between 2.25 Ga (metamorphic titanite) and 2.44 Ga by Huhma et al. (1990) with three subsequent anal- (Penikat layered intrusion). Sm–Nd isotope data yses on strongly LREE-depleted samples, an age of on these rocks published by Huhma et al. (1990) 2105 ± 50 Ma can be calculated (Fig. 73, Appendix are presented in Figure 72, in order to emphasise 6). As was discussed above, a mafic dyke (A1214- the slight variation in the initial εNd values between Koppakumpu) cutting the Tikanmaa Formation the lower and higher lava flows in the Runkaus above the Jouttiaapa Formation in the stratigraphy Formation. For reference, estimates of the timing has yielded a U–Pb age of 2130 ± 5 Ma. Accordingly, of the regional metamorphic effects are also shown. the basalts of the Jouttiaapa Formation should be The samples from the top part of the first recognised older than 2130 ± 5 Ma, which is (within error) in lava flow (14C, 14C2) display strong LREE depletion agreement with the Sm–Nd age derived above for at ca. 1.7–1.8 Ga, and some secondary fractionation the formation. Using an age of 2130 Ma, an average may have also influenced other samples. As a whole, epsilon of +3.9 can be calculated for the analysed the isotope and chemical data suggest that the first 16 basalt samples. Highly positive initial εNd values flow had primarily a slight LREE enrichment and exceeding +3 were also obtained for the mafic dykes initial epsilon value close to zero, whereas later at Koppakumpu (A1214) and Kuusivaara (A2087, age flows show a significantly higher REE level, more 2140 ± 11 Ma) discussed above. LREE enrichment and slightly negative initial εNd A recent Sm–Nd analysis conducted on a sam- values, suggesting stronger crustal contamination ple from mafic tuffs in the Tikanmaa Formation (Huhma et al. 1990). revealed that the rock has a chondritic Sm/Nd

So far, the Jouttiaapa Formation basalts have ratio and a highly positive εNd(2130 Ma) value of not been dated precisely. The best available age +4.1 (Appendix 6). A sample from mafic tuffites

0.5132

Runkaus metabasalts 14C, C2 1st flow (triangles) 0.5128 Reference line 2.3 Ga epsilon = 0 0.5124 14C albite

Nd 0.5120 144 14A amphibole Nd/ 14A 14A albite 143 0.5116 14A amphibole Flows 2-5 (squares) Reference line 2.3 Ga

0.5112 epsilon = -1

14A epidote Metamorphic LREE depletion 0.5108 Epidote whole rock at ca. 1.7-1.8 Ga 1571±52 Ma flow top 14C, 14C2 0.5104 0.06 0.10 0.14 0.18 0.22 0.26 147Sm/144Nd

Fig. 72. Sm–Nd isotope data for whole rock and mineral separates from the Runkaus basalts. Flow top samples with secondary REE fractionation are presented as open symbols (data from Huhma et al. 1990).

75 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Jouttiaapa metabasalts 0.5155

Nd 0.5145 lower part 144 Nd/ 143 0.5135 Age = 2105 ± 50 Ma eps=+4.2 upper part MSWD = 3.9 n=16 Tikanmaa tuff CHUR 0.5125 0.16 0.20 0.24 0.28 0.32 0.36 0.40 147Sm/144Nd

Fig. 73. Sm–Nd isotope data for whole-rock samples from the Jouttiaapa basalts and Tikanmaa tuff (data from Huhma et al. 1990 and this study). CHUR = chondritic uniform reservoir (De Paolo & Wasserburg 1976)

of the Hirsimaa Formation above the Tikanmaa (Karhu et al. 2007), has a high REE abundance, low

Formation has also been analysed. This sample Sm/Nd, and gives an initial εNd(2106 Ma) value of (A1788) yielded a U–Pb zircon age of 2106 ± 8 Ma +0.9 (Appendix 6).

6 KUHMO BLOCK IN THE LENTUA COMPLEX

6.1 Geological background

The Kuhmo basement block forms the central part of chain of Archaean greenstone belts (Fig. 74), the large Archaean basement complex (Lentua com- which is collectively called the Tipasjärvi-Kuhmo- plex) in eastern Finland, being located between the Suomussalmi greenstone complex (Papunen et al. Palaeoproterozoic Kainuu schist belt and the east- 2009). The surrounding rocks are composed of ern border of Finland. As mentioned earlier, there migmatitic tonalitic gneisses, migmatised gneisses are shear zones at the contact between the Kuhmo of sedimentary origin (paragneisses), and various and Taivalkoski blocks. That these two blocks plutonic rocks, such as tonalite-trondhjemite- have undergone partly different post-Archaean/ granodiorite series and granodiorite-granite- Proterozoic geological histories is evidenced by the monzogranite series rocks, and sanukitoids. More obvious differences between the orientations of the information on the Archaean geology and geochro- Palaeoproterozoic dyke swarms occurring in these nology of the Kuhmo block is presented in Papunen regions. Moreover, observations for the youngest, et al. (2009), Huhma et al. (2012a, b), Mikkola et al. 1.98 Ga NW–SE-trending swarm, which is well- (2011) and references therein. presented in the Kuhmo block (see below), are so The Archaean basement of the Kuhmo block is far lacking from the Taivalkoski block (see Fig. 5.21 abundantly cut by Palaeoproterozoic diabase dykes in Vuollo & Huhma 2005). (Vuollo & Huhma 2005). Most of the dykes are The Kuhmo block is divided into two parts by a geochemically similar Fe-tholeiites, but in terms narrow, approximately 220-km-long N–S-trending of age and orientation, two distinct swarms can

76 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland be recognised. The older ~2.1 Ga group comprises Kuhmo block, belonging to the NE–SW-trending dykes with a trend of 280° and the younger 1.98 Ga boninite-noritic, E–W-trending orthopyroxene- dykes show a trend of 320°. Also, representatives plagioclase-phyric and NW–SE-trending low-Ti of the ca. 2.45 Ga age group can be found in the tholeiitic subtypes (Vuollo & Huhma 2005). The

Ma�nvaara Fm/ Kurkikylä Gr ^_ Intrusions and A0496 ^_ volcanic rocks Lohisärkkä ^_ Sm-Nd (±U-Pb) Honkaniemi ^_^_A0769 Viianki (VD) ^_ Kivikevä� (KD)^_ ^_ Volcanic rocks Ke�ukallio ^_^_ ^_ ^_Paha Kapustasuo ^_ Veitsivaara U-Pb ( in this paper ) A1673 ^_ ^_ Petäiskangas ^_^_A1361 ^_ _ Kapea-aho U-Pb ^ A1363 (age published elsewhere) ^_ ^_Arola ^_A1672 Sm-Nd (age ?) Varisniemi Fm ^_^_A0196 ^_^_ ^_^_^_A0729 XW Jun�lanniemi ^_ Romuvaara 2.4 - 2.5 Ga felsic rocks ( in this paper ) ^_A0261 ^_A1838 ^_^_ Age ^_ Otanmäki ^_ ^_Mus�kkarinne <1930 Ma Raatelampi ^_ 1931 - 2080 ^_ 2081 - 2180 ^_A1460 ^_ 2181 - 2280 ^_ 2281 - 2380 Ha�uselkonen ^_Koirakoski ^_ ^_ >2380 Siunaussalmi/ ^_ Tulisaari A1368 Lapinlah�^_^_ ^_Nieminen ^_Panjavaara Humppi ^_ ^_^_Syväri ^_Vuotjärvi Siilinjärvi Koli Siilinjärvi Parkkila^_^_^_ ^_ Kaavi ^_ Siilinjärvi Vuorimäki A0376 ^_ Peräaho ^_ ^_ Ilomantsi Norway ^_Kylylah� Paukkajavaara ^_ Russia

Sweden

Finland ^_A1231 Kiihtelysvaara^_ ^_ ¢ Oravaara^_ 0 30 km

Fig. 74. Geological map of eastern Finland showing the sample locations. For symbols, see Figure 1.

77 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye boninite-norites include the Viianki dyke, which In this paper, we present isotope data for dykes potentially gives important geochemical evidence of the age groups mentioned above. In addition, we for the parental magma composition of the 2.44 provide new isotopic evidence for the presence of ca. Ga layered intrusions in Finland (Vogel et al. 1998, 2.3 Ga dykes in a WNW–ESE-trending dyke swarm Yang et al. 2016). Hanski et al. (2010) documented close to the Suomussalmi greenstone belt and in a the occurrences of 2.22 Ga sills and dykes (GWA SE–NW-trending dyke swarm in an area between association) within the Archaean Kuhmo green- the Kuhmo greenstone belt and Kainuu schist belt. stone belt, but no signs of dykes of this age group Furthermore, we have confirmed an age of ca. 2.2 have been discovered from the Archaean granitoid Ga for a dyke that cuts the Saari-Kiekki belt, a small basement in the Kuhmo block, except at the eastern Palaeoproterozoic schist belt crossing the Finnish– margin of the Kainuu schist belt. In general, the Russian border (Luukkonen 1989), and an age of mafic dykes are better preserved, containing pri- 2.15 Ga for a dyke representing a NW–SE-trending mary plagioclase and clinopyroxene, in the eastern swarm that occurs just east of the Kainuu schist part of the Kuhmo block, close to the Russian border belt. (see Fig. 5.23 in Vuollo & Huhma 2005).

6.2 The 2.4 Ga boninite-norite Viianki dyke, A1356

The Viianki dyke (A1356, VD) occurs on the Finnish– NE, are 30–60 m thick, and some can be traced Russian border (Fig. 74) and represents the group along strike for more than 40 km. They consist of noritic dykes with boninitic geochemical affin- typically of coarse plagioclase (35%), orthopyrox- ity (Vuollo & Huhma 2005). Dykes of this type ene (30%), clinopyroxene (20%), with minor oli- are encountered in many places in the Kuhmo vine (5%), chromite and Fe–Ti oxides. These dykes block, near the 2.44 Ga layered intrusions from are found well preserved only near the Finnish– Näränkävaara through Koillismaa to Peräpohja, Russian border and further east on the Russian side. and also in Russian Karelia (Stepanov 1994, Vuollo Elsewhere, pyroxenes are altered to amphiboles and & Fedotov 2005). In the Kuhmo block, they trend olivine to serpentines. The dykes tend to lack well-

0.5128 A1356 Viianki A1356 px#3 0.5124 boninite-norite dyke Age = 2409 46 Ma A1356 px eps = -1.3 A1356 px#2 0.5120 MSWD = 1.6 n=5 Nd

144 0.5116 A1356 wr Nd/

143 0.5112

A1356 plag 0.5108

0.5104 0.06 0.08 0.10 0.12 0.14 0.16 0.18 0.20 147Sm/144Nd

Fig. 75. Sm–Nd isotope data for whole rock and mineral separates from the Viianki dyke A1356.

78 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland developed chilled margins. The boninite-norite isotope composition of these grains is 1.7% discor- dykes are characterised by high contents of MgO, dant (Fig. 49, Appendix 8) and gives a minimum 207 206 SiO2, Cr and Ni, whereas TiO2 and Zr are low. age of 2385 Ma ( Pb/ Pb age). As many of the Chondrite-normalised REE patterns are fairly steep, baddeleyite grains have a thin rim of polycrystalline with enrichment in LREE. metamorphic zircon (cf. Heaman & LeCheminant The width of the Viianki dyke is 30 m and it can 1993), the primary age of the magmatic baddeleyite be traced in the NNE direction for about 200 m in should be older than that obtained from the bulk the field (Kilpelä 1991). The major minerals are material, conceivably close to 2.44 Ga (see Fig. 49). plagioclase (An 52%), clinopyroxene, and olivine, Five Sm–Nd analyses were performed on a which are well preserved in places. A large sample whole-rock powder and plagioclase and pyroxene (29 kg), labelled A1356, was collected from the dyke separates from Viianki (Appendix 1). These data plot for U–Pb dating, but only 11 grains of baddeleyite along a chord with an age estimate of 2409 ± 46 Ma were obtained (in Toronto). Some of the grains were and an initial εNd of -1.3 (MSWD = 1.6, Fig. 75), thus rather good, but some had inclusions, cracks, and being consistent with the age obtained by U–Pb. a thin (1–5 µm) rim of zircon. The measured U–Pb

6.3 The 2.3 Ga dykes, Lohisärkkä A1914, Kovavaara A1361, Karhuvaara A1672

The WNW–ESE-trending, 200-m-wide Lohisärkkä 206Pb/207Pb ages on zircon are younger, down to dyke swarm extends more than 30 km, cutting 1.8 Ga, which well explains the age from the bulk Archaean rocks in the Suomussalmi area (Bernelius TIMS data. We may conclude that the Lohisärkkä 2009, Fig. 74). The dykes are usually only slightly dyke swarm belongs to the 2.3 Ga dyke group and strained, coarsely plagioclase-phyric metadiabases records 1.8 Ga metamorphic effects, which are of Fe-tholeiitic composition. The main minerals are common throughout Finland. Sm–Nd whole-rock typically labradoritic plagioclase, tschermakitic analysis yielded an εNd(2321 Ma) value of +1.5, which hornblende–tschermakite and epidote. A small is similar to those from the 2.3 Ga dykes in the amount of fine-grained baddeleyite was recov- Iisalmi complex (Chapter 8, Appendix 1). ered from the sample (A1914) collected for isotopic Samples from several Fe-tholeiitic dykes occur- studies. The two multigrain TIMS analyses revealed ring in the Archaean migmatite gneiss-dominated fairly high common lead and provided discordant area between the Kainuu schist and Kuhmo schist data with 207Pb/206Pb ages of ca. 2.2 Ga (Fig. 76, belts were collected for isotope dating in 1993 in Appendix 5). A closer look at the grains mounted connection with the project “The age and character on epoxy revealed that many baddeleyite grains of the Proterozoic mafic magmatism” (Department are mantled by zircon. For LA-MC-ICPMS analy- of Geology, University of Oulu). These originally sis, we have used an in-house baddeleyite standard, diabasic rocks have been variably sheared/foli- A974, which has a TIMS U–Pb age of 1256 ± 2 Ma ated and metamorphosed under amphibolite facies (Söderlund et al. 2004). Forty spots were analysed conditions to more or less schistose metadiabases using laser ablation (Fig. 76, Appendix 10); most usually containing amphibole and oligoclase as of these were targeted at baddeleyite, but some their main minerals. Because of the metamorphic measurements may also have hit zircon deeper in effects, many of the samples produced discordant the sample. Many analyses, especially those of zir- multigrain zircon TIMS U–Pb data, often difficult con, showed a high amount of common lead. Zircon to interpret reliably and, in the best cases, yielding domains were analysed using the baddelyite stand- imprecise ages. ard calibration, and the reported Pb/U ratios are not Sample A1361-Kovavaara was picked from the therefore correct, but Pb–Pb ages should be rel- coarse-grained inner part of a SE–NW-trending evant. This conclusion is based on the finding that dyke with a thickness of 60–80 m (Fig. 74). In the measurements on our in-house zircon standard contrast to many mafic dykes occurring in the area A382 during the same session closely produced the between the Kuhmo greenstone belt and Kainuu expected age of 1877 ± 2 Ma (Appendix 10). schist belt, this rock is only mildly deformed and Rejecting data with high common lead preserves its primary diabasic texture fairly well, (206Pb/204Pb<600), the analyses on baddeley- although without any traces of primary main min- ite give an age of 2321 ± 21 Ma (n = 27). The erals. A small amount of translucent, brownish

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A1914 Lohisärkkä diabase data-point error ellipses are 2s

0.6 Baddeleyite Intercepts at 504 150 & 2321 21 Ma MSWD = 2.8 n=27 0.5 2600 (data with 206/204Pb>600)

2200 206Pb 0.4 238U 1800 0.3 1400 black diamonds - TIMS

0.2 1000 red -zircon

0.1 1 3 5 7 9 11 13 207Pb/235U

Fig. 76. Concordia plot of U–Pb zircon data obtained from the Lohisärkkä dyke A1914. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as dots.

"2.3 Dykes in Kainuu" data-point error ellipses are 2s A1672 Karhuvaara Intercepts at 0.7 76 ± 75 & 2314 ± 6 Ma MSWD = 1.1 n=9

0.6 3000

0.5 2600 206Pb 238 U 2200 0.4

A1361 Kovavaara 1800 reference line intercepts 0.3 700 & 2280 Ma 1400 (3 TIMS analyses)

0.2 2 6 10 14 18 22 207Pb/235U

Fig. 77. Concordia plot of U–Pb zircon data obtained from the Kovavaara (A1361) and Karhuvaara (A1672) dykes. LA-MC-ICPMS analyses for A1672 are shown as error ellipsoids and ID-TIMS analyses for A1361 as red dots.

8080 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland zircon was obtained from the sample. Three multi- thickness. The zircon population extracted from grain TIMS analyses yielded discordant results and sample A1672 looks heterogeneous, and many grains do not define an unambiguous straight-line chord. appear turbid and hence altered. The 18 analyses by However, the average age of the three fractions, ca. LA-MC-ICPMS conducted on 15 grains were tech- 2.28 Ga, could well be close to the magmatic age of nically good and the data from the best-preserved the dyke (Fig. 77, Appendix 5), which is supported domains yield Pb–Pb ages close to 2.3 Ga (Fig. 77, by the fact that the zircon has an unusually high Appendix 10). The analyses of altered domains Th/U ratio, similarly to the zircon grains in the 2.3 tended to give significantly younger Pb–Pb ages, Ga Karkuvaara dyke occurring in the Taivalkoski and two Archaean ages were obviously measured block (Chapter 4.4). from xenocrystic grains. The obtained isotope data Sample A1672-Karhuvaara was collected 16 km suggest a magmatic age of 2314 ± 6 Ma for this south of the Kovavaara dyke, from a similar though Fe-tholeiitic dyke, being similar to the age that more strongly strained/foliated, SE–NW-trending was determined for the Kovavaara dyke (A1361) Fe-tholeiitic metadiabase dyke 60–80 metres in discussed above.

6.4 The 2.2 Ga Rasiaho dyke A261

The Rasiaho dyke is a major metadiabase dyke in the In situ LA-MC-ICPMS analyses revealed different Kuhmo area close to the Russian border, where it ages for zircon and baddeleyite, with the badde- intrudes the supracrustal rocks of the Saari-Kiekki leyite in the zircon cores giving ages of ca. 2.2 Ga schist belt (A261, Fig. 74). The Rasiaho trunk dyke, and the zircon rims recording ages of ca. 1.8 Ga. which has several sizeable apophyses (pajonettes), It is obvious that the multigrain TIMS data rep- is up to 240 m wide and can be traced in the NW– resent mixtures between these two components. SE direction for more than 14 km. The dyke rocks Unfortunately, the analytical session that included are mostly only very weakly foliated to not foliated the baddelyite measurements was short and with at all, preserving their primary diabasic textures, only a few measurements of the standard, resulting although they mostly contain pervasively hydrated in large errors in the analyses (Fig. 78a, Appendix mineral assemblages of lower amphibolite facies, 10). Subsequently, the age of ca. 2.2 Ga has been including calcic amphibole, sodic plagioclase and confirmed for baddeleyite, using SC-ICP-MS Attom epidote as their main minerals. (Fig. 78b, Appendix 10). A sample from a coarse-grained segregation from It is to be noted that the 2.2 Ga age and weakly Rasiaho has earlier been studied by Luukkonen to non-deformed nature of the Rasiaho dyke (1989), who published multigrain TIMS U–Pb data raise doubts over the presently assumed Sumi- on zircon (with baddeleyite inclusions) and titanite. Sariolan (2.5–2.3 Ga) age of the Saari-Kiekki belt The studied zircon grains extracted from sample (Luukkonen 1989). Namely, if the belt was indeed A261 were mostly turbid, altered-looking and con- younger than 2.5 Ga, then the 2.2 Ga age of the tained some baddeleyite cores. The obtained TIMS Rasiaho dyke would constrain the timing of the data were discordant with Pb–Pb ages mostly at complex lower amphibolite facies deformation 1.9–1.97 Ga. Interestingly, an analysis of low-den- of the Saari-Kiekki belt (Luukkonen 1989) to a sity material (3.8–4.0) gave the highest Pb–Pb age pre-2.2 Ga, i.e., Sumi-Sariolan or Jatulian event. A of 2.14 Ga. The emplacement age of the rock was problem with this is that no such dynamothermal estimated from two analyses of titanite at 2.2 Ga. event has been reported from other parts of the We observed that many zircon grains have a badde- Karelian craton. lyite core, which is clearly visible in the BSE images.

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A261 Rasiaho diabase data-point error ellipses are 2s A261 Rasiaho diabase data-point error ellipses are 2s

0.48 0.48 BaddeleyitesBaddeleyites (n=5) (n=5) ConcordiaConcordia Age = 2197Age = 382197 Ma 38 Ma 0.44 0.44 2300 2300

0.40 0.40 2100 2100 206 206 Pb Zircon rims Pb 0.36Zircon rims 2380.36 238 U 1900 1900 U TitanitesTitanites TIMS ca. TIMS 2.2 Ga ca. 2.2 Ga (Luukkonen 1989) 0.32 (Luukkonen 1989) 0.32 1700 1700 Zircon cores (with 0.28 Zircon cores (with 0.28 baddeleyitebaddeleyite mixture) mixture) A) TIMS zircon (+badd) A)0.24 TIMS zircon (+badd) 0.24 3.5 4.5 5.5 6.5 7.5 8.5 9.5 3.5 4.5 5.5 6.5 7.5 8.5 9.5 207 235 207Pb/235Pb/U U

data-point error ellipses are 2s data-point error ellipses are 2s

0.50 0.50 A261 RasiahoA261 Rasiaho diabase diabase baddeleyitebaddeleyite (+-zircon) (+-zircon) 0.46 Average Pb/Pb age 0.46 Average Pb/Pb age 2400 2400 2207±312207 Ma±31 Ma MSWDMSWD = 4.1 = 4.1 U 2300

U 2300

238 0.42 238 0.42 2200 Pb/ 2200 Pb/ 206 206 2100 2100 0.38 0.38 2000 2000

1900 19000.34 0.34

B) B)0.30 0.30 5 6 7 8 9 10 5 6 7207 235 8 9 10 207Pb/235UPb/ U

Fig. 78. A) Concordia plot of U–Pb data obtained from the Rasiaho diabase A261. LA-MC-ICPMS analyses conducted on zircon and baddeleyite are presented as error ellipsoids and ID-TIMS analyses of zircon (±baddeleyite) as black dots. B) Concordia plot of U–Pb data obtained by Attom on baddeleyite from the Rasiaho diabase A261. In the back-scattered electron image, baddeleyite forms distinct cores mantled by zircon.

82 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

6.5 The 2.15 Ga Petronjärvi dyke A1363

The Petronjärvi dyke (A1363) is one of the many Pb–Pb ages of 1.93–1.97 Ga. Subsequent to the com- Fe-tholeiitic metadiabase dykes that cut the pletion of the U–Pb zircon dating, zircon grains Archaean gneissic terrain immediately to the east from the original mineral separate were analysed of the Kainuu schist belt as NW–SE-trending, 10- by laser ablation MC-ICP-MS. Five analyses of the to 50-m-wide magmatic bodies (Fig. 74). Most of best-preserved domains yielded a concordia age of these dykes are metamorphosed to amphibolite 2148 ± 8 Ma, which may be considered as the igne- facies mineral assemblages and deformed to show ous age of the rock. Several altered zircon domains at least some obvious foliation and lineation. Zircon gave ages of ca. 1.87 Ga (Fig. 79, Appendix 10). We grains obtained from sample A1363 are turbid and point out that the appearance and chemical com- appear to be extensively altered. Five multigrain position of this dyke are very similar to those of the fractions were initially analysed using TIMS. The 2.3 Ga dykes discussed above. U–Pb data are discordant and plot in a cluster with

6.6 The 2.1 Ga Kapea-aho dyke A1212

The Kapea-aho dyke represents a local swarm of crops and as a positive aeromagnetic anomaly. At E–W-trending diabase dykes, which are common the Kapea-aho sampling site, where the dyke has a in the area east of the northern Kuhmo green- width of approximately 50 metres, the central part stone belt and where they tend to be nonfoliated of the dyke is typically hydrated to metadiabase and often comprise only slightly altered primary containing mainly actinolitic hornblende, horn- mineral assemblages (Fig. 74). The Kapea-aho dyke blende, andesine and muscovite. However, closer has been studied by Kilpelä (1991), who states in to the margins, zones of unaltered rock are found, his Master’s thesis that the dyke is 50–60 m wide where the major minerals are labradorite (An 51%) and can be traced for 6–7 km along strike in out- and clinopyroxene. The chemical characteristics of

A1363 Petronjärvi metadiabase (Fe-tholeiitic) 0.48 data-point error ellipses are 2s

0.44 2300

0.40

U 2100 238 0.36 Concordia Age = 1900 Pb/ 2148 ± 8 Ma n=5 206 0.32 1700 TIMS

0.28 Average Pb/Pb age 1868 ± 31 Ma 4 youngest 0.24 3.5 4.5 5.5 6.5 7.5 8.5 207Pb/235U Fig. 79. Concordia plot of U–Pb zircon data obtained from the Petronjärvi dyke A1363. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red triangles.

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.5132 A1212 Kapea-aho dyke Age = 2133 29 Ma A1212px 0.5128 eps = +0.7 A1212px#2 MSWD = 0.20 n=5 Nd 0.5124 144 Nd/ A1212wr #1, #2

143 0.5120

0.5116 A1212plag

0.5112 0.09 0.11 0.13 0.15 0.17 0.19 0.21 0.23 147Sm/144Nd

Fig. 80. Sm–Nd isotope data for whole rock and mineral separates from the Kapea-aho dyke A1212.

the Kapea-aho dyke indicate an evolved, Cr- and and clean-looking fractions of both minerals. The Ni-poor composition with Fe-tholeiitic affinity performed five analyses, including two for the (Kilpelä 1991, Vuollo & Huhma 2005). whole-rock powder, yielded an isochron with an

The main minerals, plagioclase and pyroxene, age of 2133 ± 29 Ma (εNd = +0.7, MSWD = 0.2, Fig. were separated from sample A1212 for Sm–Nd iso- 80, Appendix 1), which we interpret to represent tope measurements. Hand-picking was used as the the most probable emplacement age of the Kapea- final step of the extraction process, yielding fresh aho dyke.

6.7 The 2.0–1.95 Ga dykes, Kivikevätti A1409, Puuropuro A1673, Peräaho A1519, Kivimäki A1460

The Kivikevätti dyke is one of the numerous NW– 1991). At the A1409 sampling site, both margins of SE-trending dykes that occur in the Kuhmo block, the dyke are exposed and show chilling against the in this case in the area east of the Kuhmo green- country gneiss. The primary magmatic minerals in stone belt (Fig. 74). The Fe-tholeiitic dyke swarm the sample are well preserved. represented by the A1409 Kivikevätti sample is in Hundreds of relatively high-quality baddeleyite its area of occurrence the youngest known and least grains were recovered from the sample (in Toronto, deformed mafic swarm. In well-preserved samples, sample wt 16.8 kg). Most baddeleyites were brown- the mostly coarse- to medium-grained dykes con- ish needles without inclusions. The four U–Pb anal- sist of subhedral coarse plagioclase (40%), set in a yses conducted on baddeleyite by TIMS were slightly matrix of fine-grained, anhedral-subhedral clino- discordant (1.4–2.3%) and yielded 207Pb/206Pb ages pyroxene (25%), Fe–Ti oxide (10%) with quartz of 1977–1979 Ma (Appendix 8). A regression line (3%), biotite (4%) and uralitic amphibole (10%). through the data gives intercepts of 1980 ± 4 Ma Plagioclase laths are characterised by tea-coloured (2σ) and 329 Ma (Fig. 81). clouding just as plagioclase, e.g., in well-preserved Well-preserved primary minerals, plagioclase 2.45 Ga dykes. The observed maximum width of the and clinopyroxene, were used for Sm–Nd dating. Kivikevätti dyke is ca. 60 m, and along strike, it can The analysis of plagioclase (#1) involved problems be traced in the field for more than 10 km (Kilpelä related to the dissolution and subsequent spike-

84 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

data-point error ellipses are 2s 0.368

0.364 1990

0.360 1980 1970 U

238 0.356 1960

Pb/ 1950

206 0.352 1940

0.348 Intercepts at 239±280 & 1980 ± 4 Ma MSWD = 0.5 0.344

0.340 5.7 5.8 5.9 6.0 6.1 207Pb/235U Fig. 81. Concordia diagram showing U–Pb TIMS data for baddeleyite fractions analysed in Toronto from the Kivikevätti dyke. The error ellipses reflect two sigma errors.

0.5128 A1409 Kivikevätti dyke A1409cpx

0.5124 Nd 144 0.5120

Nd/ A1409wr 143

A1409plag#2 Age = 2014 33 Ma 0.5116 eps = +0.3 MSWD = 0.86 n=3

0.5112 0.10 0.12 0.14 0.16 0.18 0.20 0.22 147Sm/144Nd

Fig. 82. Sm–Nd isotope data for whole rock and mineral separates from the Kivikevätti dyke A1409.

85 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye sample homogenisation. These problems were The Koidanvaara dyke is a large tholeiitic diabase avoided in the duplicated analysis (plag#2), and due dyke close to the Russian border in the Ilomantsi to uncertainty in Sm/Nd, the analysis plag#1 should area (Fig. 74; Kärenlampi 2015). The dyke persists be omitted. The data on the whole-rock sample and along strike for more than ten kilometres and has plagioclase (#2) and pyroxene separates give an age a width of 0.5–1 kilometres. Signs of deformation estimate of 2014 ± 33 Ma (εNd = +0.3, MSWD = 0.9, are mostly slight and, despite amphibolite facies Fig. 82, Appendix 1), which is consistent with the metamoprphism, primary minerals including pla- baddeleyite U–Pb age within error. gioclase, clinopyroxene, Fe-Ti oxides, quartz and The NW–SE-trending Fe-tholeiitic Puuropuro potassic feldspar with minor subsolidus ferrohorn- dyke (A1673) is located about 8 km to the east of blende and fayalite are widely preserved, but usu- the Kainuu schist belt, about 5 km west of the 2.3 ally with at least some secondary amphibole and Ga Kovavaara dyke (Fig. 74). The width of the dyke plagioclase. is 60–80 metres and along strike, it can be fol- The sample for zircon dating (A1519), which was lowed in outcrops for at least 4 km. The rock in the taken at the Peräaho locality from a pegmatoid seg- dyke varies from nearly non-foliated to conspicu- regation in the diabase, produced abundant light- ously foliated metadiabase that contains actinolitic coloured, although mostly fairly turbid, acicular hornblende, plagioclase, and in places also garnet zircon grains. Several conventional U–Pb TIMS as the main minerals. Compared with many other analyses yielded discordant and slightly scattered metadiabase dykes of this study, the zircon grains data with apparent 207Pb/206Pb ages of 1.91–1.96 Ga extracted from sample A1673 appear exceptionally (Fig. 84, Appendix 5). The common-lead content in fresh and good in their preservation quality. The these data is also fairly high. Instead, a more recent LA-MC-ICPMS data (n = 14) are concordant and analysis using the chemical abrasion technique gave suggest an age of 1979 ± 6 Ma, which is probably a concordant result at 1956 ± 3 Ma. Nevertheless, the igneous age for this dyke. A multigrain TIMS due to the slightly heterogenous population, it was analysis is consistent with this result (Fig. 83). still questionable whether this could be considered

A1673 Puuropuro metadiabase (Fe-tholeiitic) data-point error ellipses are 2s 0.46

Concordia Age = 1979 ± 6 Ma 0.42 n=14 2200

U 2100

238 0.38 2000 Pb/ 1900

206 0.34 1800

1700 TIMS 0.30

0.26 4 5 6 7 8 207Pb/235U

Fig. 83. Concordia plot of U–Pb zircon data obtained from the Puuropuro dyke A1673. LA-MC-ICPMS analyses are presented as error ellipsoids and an ID-TIMS analysis as a red dot.

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1519 Peräaho gabbro data-point error ellipses are 2s

Concordia Age = 1973 13 Ma 0.42 n=7 (Pb206/204>4000)

CA-TIMS A1519J 2100 0.38 Concordia Age = 1956 3 Ma 2000

1900 206Pb 0.34 1800 238 U 1700 TIMS 0.30 1600 1500 0.26 Intercepts at 537 220 & 1962 12 Ma MSWD = 0.83 n=23 0.22 3 4 5 6 7 207Pb/235U

Fig. 84. Concordia plot of U–Pb zircon data obtained from the Peräaho dyke A1519. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

as an unambiguous igneous age. Therefore, zir- bly also foliation and rotation of the dyke. One of the con from sample A1519 was also hit by laser and studied grains appears more pristine. Two analyses analysed by MC-ICP-MS. Rejecting a few analyses of this grain yielded an age of 1954 ± 19 Ma (grain with high common lead, an age of 1962 ± 12 Ma 10, Fig. 85, Appendix 10). It is tempting to consider (n = 23) can be calculated. Using only the spots (n = this as an igneous age, but the data are too scarse 7) on domains of apparently best preservation, the for a solid conclusion. The three multigrain TIMS age becomes 1973 ± 13 Ma (Fig. 84, Appendix 10). analyses conducted before the laser work are dis- This may be considered as the igneous age of this cordant and consistent with the laser data (Fig. 85, dyke. The Sm–Nd analysis of whole rock yielded an Appendix 5). However, a closer inspection reveals an

εNdvalue of -0.7 (Appendix 1). unusual pattern, with the analyses of heavier zircon The Kivimäki dyke (A1460) is one of the several giving slightly younger Pb–Pb ages and lower Th/U N–S-trending, mostly conspicuously foliated meta- compared to the analysis of the 4.0–4.2 g/cm3 den- diabase dykes that cut Archaean gneisses south of sity fraction, and also to the main ICP-MS data (the the Archaean Tipasjärvi greenstone belt. Towards mount was very likely made on heavy zircon). This the south and southeast from Kivimäki, the dykes suggests that a lot of material must be older than become less foliated and gradually turn to the 1.83 Ga, the age that was obtained for the dominant NW-SE direction. The rock in the sampled dyke is grain type by LA-ICP-MS. The low Th/U ratio of strongly foliated, and one reason for the dating was the heaviest zircon fraction is a characteristic of to constrain the age of the deformation, including metamorphic zircon (e.g., Rubatto 2002). the map-scale northwards (clockwise) rotation of Recently, another diabase dyke of this age the Kivimäki dykes. Zircon grains from the sample group was recognised by Mikkola et al. (2013) at are mostly turbid, appearing poor in their preserva- Hattuselkonen, in the northern Lieksa area (Fig. 74). tion quality. Accordingly, most LA-MC-ICPMS data They reported a U–Pb age of 1989 ± 9 Ma and an give ages of ca. 1.83 Ga, which may be considered an initial εNd of -0.7 for this diabase (A2071, Fig. 86). approximate age of the metamorphism and proba-

87 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A1460 Kivimäki diabase data-point error ellipses are 2s

0.4 2100

1900

U 1700 Concordia Age 0.3

238 1954 ± 19 Ma 1500 (10a,b) Pb/ 1300 TIMS 206

0.2 1100 Intercepts at 198 ± 92 & 1832 ± 13 Ma MSWD = 1.07 n=14

0.1 1.5 2.5 3.5 4.5 5.5 6.5 7.5 207Pb/235U Fig. 85. Concordia plot of U–Pb zircon data obtained from the Kivimäki dyke A1460. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

A2071 Hattuselkonen diabase data-point error ellipses are 2s 0.42

Concordia Age = 1989 9 Ma (2s, decay-const. errs ignored) 0.40 n=10 (all)

2100 U 0.38 2060 238 2020 Pb/ 0.36 1980 206 1940

0.34

0.32 5.5 5.7 5.9 6.1 6.3 6.5 6.7 6.9 207Pb/235U

Fig. 86. Concordia plot of U–Pb zircon data obtained by LA-MC-ICPMS from the Hattuselkonen dyke A2071 (Mikkola et al. 2013).

88 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

6.8 Dykes in the Veitsivaara area, A1489b & A1489c

Studies in the Veitsivaara area, a former candidate age grouping with the dyke orientation. Namely, area for a nuclear waste repository, have given valu- two E–W-trending (280o) dykes have yielded Sm– able information on the trends and age relation- Nd mineral ages of 2133 ± 29 Ma (A1212 Kapea- ships of mafic dykes in the Kuhmo block (material aho, which is located 20 km east from Veitsivaara) from Posiva Ltd). Two main swarms have been rec- and 2054 ± 40 Ma (A1489c Veitsivaara), whereas ognised. The E–W-trending (280o) swarm is con- three NW–SE-trending dykes have given a U–Pb sidered older than the other principal dyke swarm, age of 1981 ± 4 Ma (Kivikevätti) and Sm–Nd ages which has a trend to NW (320o). In the Veitsivaara of 2014 ± 33 Ma (A1409 Kivikevätti) and 2005 ± area, both diabase dyke swarms contain rock varie- 40 Ma (A1489b Veitsivaara). A slight difference is ties which still, to a variable extent, preserve their also apparent in the chemical compositions, as the primary magmatic textures and major minerals NW–SE-trending dykes seem to have higher REE plagioclase and clinopyroxene. Fresh samples for concentrations (Appendix 1). isotopic studies were taken from drill cores, with Provided that the two E–W-trending dykes (A1212 sample A1489b representing the NW–SE-trending and A1489c) are coeval with an identical intial Nd dykes and sample A1489c the E–W-trending dykes. isotope ratio, one can calculate a regression line The four analyses conducted on whole rock using eight analyses. This gives an age of 2106 ± 40 and primary minerals from sample A1489b gave Ma (εNd = +0.6, MSWD = 1.9, Fig. 88). Similarly, the an age of 2005 ± 40 Ma (εNd = +0.3, MSWD = 0.5). seven analyses available from the NW–SE-trending

The three Sm–Nd results from the other sample dykes yield an age of 2010 ± 26 Ma (εNd = +0.3,

A1489c yielded an age of 2054 ± 40 Ma (εNd = +0.3, MSWD = 0.6). The small MSWD is well consistent MSWD =1, Fig. 87, Appendix 1). Considering the with a common origin for these geographically dis- error limits, the Sm–Nd ages derived for the two tant rocks. Instead, if all 15 analyses from the two dyke swarms at Veitsivaara overlap. Neither can the dyke groups discussed here are regressed together, swarms be distinguished based on their initial Nd the MSWD rises to 5, which probably means that the isotope compositions. samples are unrelated. In a broader view, the isotope data available from the Kuhmo block, however, show some consistent

0.5128 Veitsivaara

0.5126 E-trending dyke A1489cpx A1489c A1489bpx 0.5124 Age = 2054 40 Ma eps = +0.3 Nd MSWD = 1.0 n=3

144 0.5122 A1489c Nd/ 0.5120 143 A1489b NW-trending dyke 0.5118 A1489b Age = 2005 40 Ma A1489cplag 0.5116 eps = +0.3 A1489bplag#2 MSWD = 0.5 n=4 A1489bplag 0.5114 0.10 0.12 0.14 0.16 0.18 0.20 0.22 147Sm/144Nd Fig. 87. Sm–Nd isotope data for whole rock and mineral separates from two Veitsivaara dykes.

89 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

6.9 Dykes in the Romuvaara area

In the Romuvaara area, both boninitic and samples instead gave a rather strange pattern, sug- Fe-tholeiitic dykes have been recognised. The gesting an age of ca. 1.0 Ga (Fig. 88). Two of the Sm–Nd data on four whole-rock samples from the samples have a Sm/Nd ratio fairly typical for 2.44 Fe-tholeiitic dykes are scattered and do not provide Ga rocks, whereas the two other samples have a any ages. However, the data are roughly consistent much higher Sm/Nd ratio, suggesting hydrother- with the data acquired for the 2.1 Ga dykes, yielding mal alteration and associated depletion of LREE at

εNd(2100 Ma) values from -0.5 to +0.8. The eight ca. 1.0 Ga. analyses conducted on four boninitic whole-rock

Mafic dykes in Kuhmo

NW-trending dykes 0.5132 Age = 2010 ± 26 Ma eps = +0.3 MSWD = 0.58 n=7 A1212px A1212px#2 0.5128

A1409px E-trending dykes A1489c_px A1489bpx

Nd Age = 2106 ± 40 Ma 0.5124 eps = +0.6 189-SSP 144 MSWD = 1.9 n=8 KR-25 A1489c Nd/ RO-KR3 A1212 & #2 0.5120 A1409 RO-KR-5 4-ROM & #2 143 A1489b 2-ROM & #2 & #3 A1409plag#2 A1489cplag 1.1-ROM Romuvaara boninitic dykes 0.5116 A1212plag 3-ROM & #2 A1489bplag#2 Age = 914 ± 200 Ma A1489bplag eps = -12 2.44 Ga rocks MSWD = 5.3 0.5112 0.09 0.11 0.13 0.15 0.17 0.19 0.21 0.23 147Sm/144Nd Fig. 88. Sm–Nd isotope data for whole-rock samples from Romuvaara compared to results from other dykes in the Kuhmo area. Green triangle – Fe-tholeiitic dyke, red dot – boninitic dyke.

7 KAINUU SCHIST BELT

7.1 Geological background

The Palaeoproterozoic Kainuu schist belt (KSB) The KSB is dominantly metasedimentary in forms a N–S-trending, folded and faulted inlier composition. The lowermost sedimentary rocks between the Archaean Iisalmi–Pudasjärvi and along with the contacts with the Lentua and Lentua complexes (Fig. 89, Laajoki 2005). Tapering Iisalmi complexes lie unconformably on the mig- to the north and south, the belt is about 220 km matitic-plutonic gneisses of the named blocks. The long and has a maximum width of 40 km in its Archaean-Palaeoproterozoic contact, where not middle part. It was deformed and metamorphosed faulted, is generally defined by Jatulian (2.30–2.10 during the Svecofennian orogenic tectonism, with Ga) quartz-pebbly arenites sitting on the Archaean the conditions reaching upper greenschist facies in gneisses. Remnants of older Sumi-Sariolan (2.50– the north and amphibolite facies in the south. 2.30 Ga) volcanic and metasedimentary rocks are

90 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland locally met, but only in any significant volume in basement. Extensive detrital zircon dating for the the north. These volcanic rocks of the Kurkikylä Central Puolanka Group in the Oikarila area/struc- Group are chemically similar to and probably cor- ture has shown that the metasediments, including relate with the Kuntijärvi Formation metabasalts in assumed felsic tuffites of the CPG, exclusively con- the Kuusamo schist belt (Fig. 39). The quartz-pebble tain Archaean detrital zircon grains, mostly aged ca. conglomerates and quartzites of the immediately 2.72 Ga, but for a small part up to 3.7 Ga (Kontinen overlying Korvuanjoki Group have been correlated et al. 2014). with the Vesivaara and Koli Formations in the North At Varisniemi, a basaltic metalava (Pitkälika Karelia schist belt (Kohonen & Marmo 1992). metabasalt), probably correlating with the Central At several locations, the Jatulian sequence attains Puolanka Group, is intruded by the 2.44 Ga a thickness of more than 2 km (up to 3.5 km). As (Chapter 7.2 of this work) granophyre of the Juntti- in the prototypical Eastern Puolanka Group in the lanniemi layered gabbro intrusion (Kontinen et northern part of the belt (Laajoki 1991), it comprises al. 2014). This basalt (Pitkälika metabasalt) is mainly shallow-marine platformal feldspathic and overlain by a 50-m-thick conglomerate-wacke quartz arenites, with the latter commonly having layer with lithic clasts and zircon grains from a supermature, extremely quartz-rich character the granophyre, which in turn is topped by sev- (Kontinen 1986; Laajoki 1991). The upper part in the eral hundreds of metres of metabasalt (Varisniemi sequence has been proposed (Nironen et al. 2016) metabasalt). The volcanic rocks are physically to correlate, for instance, with the Rukatunturi (dominantly richly amygdalous) and chemically Formation (Silvennoinen 1972) and Puso Formation similar to Sumi-Sariolan metabasalts in north- (Kohonen & Marmo 1992) in the Kuusamo and ern Kainuu (Kurkikylä Group), Posio (Karkuvaara North Karelia schist belts, respectively. The Jatuli Formation) and Kuusamo (Kuntijärvi Formation). sequence is topped by dolomite intercalated with These relationships suggest that the Oikarila mafic volcanic rocks. These rocks probably corre- structure contains both <2.44 Ga and >2.44 Ga late with ca. 2.10 Ga upper Jatulian dolomite–vol- supracrustal units. Assuming a correlation of canite sequences elsewhere, such as the Tikanmaa the Pitkälika basalt with the CPG, Kontinen et al. Formation (upper part of the Kivalo Group) in the (2014) proposed an age of >2.44 Ga for the Central Peräpohja area (Fig. 60) (Perttunen & Vaasjoki 2001, Puolanka Group. Kyläkoski et al. 2012). According Laajoki (2005), in the Puolanka area, For a length of about 100 km, the western mar- the Central Puolanka Group schists grade at the gin of the KSB against the Pudasjärvi complex is western margin of the KSB to gneisses of the Kalpio defined by a mostly narrow (<1–5 km), folded– Complex, with the latter making up the SE corner of folded stripe of the dominantly metasedimen- the Pudasjärvi complex. However, in the Kivesvaara tary Central Puolanka Group (CPG). The Central area, some 40 km to the south, there is clearly a fault Puolanka Group is a tripartite sequence of inter- between these units, as there is an abrupt change bedded feldspathic wackes–aluminous pelites in the metamorphic grade from upper greenschist (Puolankajärvi Formation), fluvial-shallow marine facies (garnet-biotite) in the CPG schists to middle feldspathic arenites (Akanvaara Formation), and amphibolite facies (garnet-staurolite-sillimanite) shallow marine interbedded quartzite–pelite with in the Kalpio Complex gneisses just at their con- mafic and intermediate-felsic tuff–tuffite interca- tact. Also, there are abundant late Palaeoproterozoic lations (Pärekangas Formation). The age of deposi- (1.86–1.80 Ga, Vaasjoki et al. 2001) granites in the tion of the Central Puolanka Group has so far only Kalpio Complex gneisses immediately on the west- tentatively been defined as Archaean at the latest ern side of the contact fault, but none in the Central or Palaeoproterozoic at the earliest (Laajoki 2005, Puolanka Group schists on its eastern side. This Kontinen et al. 2014). An age of >2.20 Ga is obvious, stark contrast introduced by post-1800 Ma tec- as over its whole extent the sequence is intruded by tonism/faulting characterises the entire length of differentiated sills of the 2.22 Ga gabbro-wehrlite the Pudasjärvi complex (Kalpio-Kalhamajärvi)–KSB (karjalite) event. The presence of 2.22 Ga sills within contact. the CPG means that this unit was buried at that time The middle parts of the KSB are occupied by less than 4–5 km below the surface. The 2.22 Ga Kalevian (2100–1900 Ma) deep-water gravity sills are, all over the KSB, also abundant in the Jatuli flow metasediments. A two-part succession has strata and their immediately underlying Archaean been distinguished (Kontinen 1987, Laajoki 2005,

91 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A1456 Karkuvaara Upper Kaleva A0988 Nyrhinoja Kurkikylä Lower Kaleva A0613 Hukkavaara Marine Jatuli A0496 Jokijyrkkä Jatuli PUDASJÄRVI 7220000 COMPLEX Sumi-Sariola

KUHMO 3570000 Central Puolanka Group COMPLEX Kalpio & Kalhamajärvi Complexes A0651 Honkaniemi Archaean complexes

A0759 Late Svecofennian Kettukallio A0769 Niskansuo granitoids (1.86-1.80 Ga) A1373 Paha Kapustasuo Kaarakkala gabbro- A0437 Latolan- diorite (1.86 Ga) vaara Jormua Ophiolite Complex (magmatic units 1.95 Ga) Otanmäki suite, A-type gneissic granitoids (2.05 Ga) A1362 Petäiskangas A1673 Otanmäki gabbros (2.06 Ga) Puuropuro A1457 Kapustakangas suite, Liminpuro A1363 ultramafic-mafic intrusions Petronjärvi A1595 (2.3 Ga?) Parvialankangas

Junttilanniemi intrusion A1596 (2.44 Ga) Varisniemi 2.43 Ga gneissic A-type MANAMANSALO granitoids COMPLEX Jormua

Otanmäki

A1838 A0199 Jäkäläkangas Raatelampi A0198 Mustikkarinne

IISALMI COMPLEX

3500000 10 km 7070000

Fig. 89. Geological map of the Kainuu schist belt with the sample locations.

92 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Kontinen & Hanski 2015). The Lower Kaleva records granites of the Otanmäki suite (Talvitie & Paarma a lithologically heterogeneous package with geo- 1980, Kontinen et al. 2013), restricted in the area graphical variation in its internal stratigraphy. It south of Lake Oulunjärvi in a 1- to 6-km-wide, consists of mass-flow conglomerates, interbedded 60-km-long, E–W-trending fault-bound sliver metagreywackes, metepelites, and quartz-rich (Fig. 89). These granites locally contain Jatuli-type metawackes with iron formations and sulphide- rocks (orthoquartzite-dolomite) as inclusions. In graphite-rich metashales in its upper part. The the Ristijärvi area, two roundish (<10 km) grano- provenance of the detrital material was in the diorite–granite plutons (Fig. 89) dated at ca. 1860 Archaean basement and its pre-Kalevian sedimen- Ma (GTK, unpublished data) are emplaced in Upper tary cover (Kontinen & Hanski 2015). Kaleva metasediments. The Upper Kaleva, in turn, is a relatively monoto- In Finland, the ca. 2.44 Ga mafic-ultramafic nous unit mainly consisting of turbidite-type inter- layered intrusions are concentrated in the Tornio- bedded metagreywacke and pelite with infrequent Näränkävaara belt supplemented with a few occur- black metashale interbeds. Additionally, ophi- rences further to the north in Lapland (Alapieti et al. olitic fragments are locally found as fault-bound 1990). Recently, petrological evidence for the pres- lenses, of which the largest, in the middle of the ence of intrusive magmatism of that age has been KSB, belong to the Jormua ophiolite (Kontinen found in the Kainuu schist belt at Junttilanniemi, 1987, Peltonen 2005a). In marked contrast to the near Paltamo (Halkoaho & Niskanen 2013), repre- lower Kaleva and older Palaeoproterozoic units, senting the southernmost discovery of these intru- which mostly only contain Archaean detrital zircon sions in Finland. In this work, the granophyre of the grains, Upper Kaleva metasediments contain a large Junttilanniemi layered gabbro was dated to confirm Palaeoproterozoic detrital component, as revealed by the age of the intrusion, but also to place constraints zircon grain populations having ages in the range of on the age of the closely associated pre-Jatulian 2.00–1.92 Ga (Lahtinen et al. 2010). It is noteworthy metavolcanic rocks. We have made an attempt to that the 1.95 Ga magmatic age of the gabbros and constrain the depositional age of Central Puolanka plagiogranites in the ophiolite fragments (Kontinen Group by dating a sill of the Kapustakangas intru- 1987, Peltonen et al. 1998) is by a significant margin sive suite intruding the Puolankajärvi Formation. older than the ca. 1920 Ma maximum age indicated The dated mafic sills also include those cutting the by detrital zircon data for the Upper Kaleva depo- Archaean basement and its quartzitic cover at the sition. The Upper Kaleva, with its exotic ophiolite eastern contact of the KSB and those intruding the fragments, is interpreted as an allochthonous unit Kalpio Complex on the western side of the KSB. A thrust on the Archaean-Karelian basement-cover couple of samples from the Jormua ophiolite pro- structure at ca. 1900 Ma (Peltonen et al. 2008). viding updated results are also included. Other important rock units within the KSB include the ca. 2060–2050 Ma gabbros and A-type

7.2 The 2.44 Ga Junttilanniemi plutonic-volcanic complex (A1595-6)

The Junttilanniemi layered intrusion is found at Soidinsuu conglomerate contains pebbles and gran- the western margin of the Kainuu schist belt in a ules of granophyre, obviously from the underlying stratigraphically and structurally complex setting Junttilanniemi granophyre. including both Archaean and Palaeoproterozoic Two geochemically similar felsic granophyre rocks (Fig. 74, Fig. 89). According to Kontinen et samples (A1595 Parvialankangas granophyre, A1596 al. (2014), the Junttilanniemi gabbro intruded the Varisniemi “rhyodacite”) were used for age deter- Pitkälika metavolcanic rocks, which possibly cor- mination. A small amount of mostly light-coloured, relate with the basalts of the Pärekangas Formation, turbid zircon was obtained from both samples. The the uppermost unit of the Central Puolanka Group mineral separate from sample A1596 also contains (cf. Laajoki 1991). The Pitkälika metavolcanic a few brown, translucent crystals. The results of rocks and the apparent roof granophyre of the multigrain TIMS analyses are discordant, with Junttilanniemi gabbro are unconformably over- 207Pb/206Pb ages of ca. 2.35–2.38 Ga (sample A1595) lain by a thin veneer of conglomerate succeeded and ca. 2.08 Ga (A1596, Fig. 90, Appendix 5). by metabasalts of the Varisniemi Formation. The

93 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Junttilanniemi granophyre data-point error ellipses are 2s

0.50 A1596 Varisniemi rhyodacite 2600 SHRIMP Concordia Age = 2444 4 Ma, n=6 0.46 2400

U 0.42 2200

238 A1595B Parvialankangas granophyre 0.38 A1595A

Pb/ 2000 206 0.34 1800 A1596A

0.30 A1373A & B Paha Kapustasuo 1600 A1373-n652 Paha Kapustasuo NORDSIM (green error ell) 0.26 2 4 6 8 10 12 207Pb/235U

Fig. 90. Concordia plot of U–Pb zircon data obtained from the Junttilanniemi layered intrusion (A1595, A1596) and Kapustakangas suite gabbro A1373. SIMS analyses presented as error ellipsoids and ID-TIMS analyses as dots.

In search for more concordant data, zircon grains andesibasaltic Varisniemi Formation. In thin sec- from sample A1596 were analysed by SHRIMP at tion, the sample resembles arkosic wacke up to VSEGEI in St. Petersburg. Despite the high U con- ca. 1.2 mm in its grain size. Some of the larger tent in zircon of sample A1596, the SIMS results for clasts are clearly granophyre, similar to that in the brown, translucent crystals (image in Appendix 4b) nearby 2.44 Ga Junttilanniemi differentiated intru- are concordant, good in quality and provide an age sion. There are a few, up to 60-μm-sized zircon of 2444 ± 4 Ma (Fig. 90, Appendix 4b). Instead, an grains visible in the thin section, most of them dull, analysis of a turbid, dark grain (A1596.7.1, marked brownish prisms. in CL-light grain without label) revealed a high Fifty U–Pb analyses on zircon performed abundance of common lead and gave an imprecise using LA-MC-ICPMS provided a range of mostly date of ca. 1.7 Ga. The discordant TIMS data are Archaean ages (Appendix 11, Fig. 91A). However, a consistent with the SHRIMP results. The Sm–Nd few grains constrain the depositional time as early analyses conducted on these two samples give simi- Palaeoproterozoic. Sm–Nd analysis on whole rock lar negative initial εNd(2444 Ma) values of -1.8 and gave a T-DM model age of 2.92 Ga (Appendix 1).

-1.9, which are within the range of εNd values typical Sample A2105 represents the trondhjemitic xeno- for the 2.44 Ga intrusions (Appendix 1). liths that are found in abundance in the Pitkälika Sample A2050, picked ca. 400 metres west of the Formation meta-dacite, which is cut by the grano- granophyre sample A1596, represents the massive- phyre of the Junttilanniemi intrusion. All zircon schistose, pink-grey weathering feldspathic wacke grains analysed from this sample were found to be that dominates the Soidensuu conglomerate-wacke Archaean, mostly between 2.6 and 2.8 Ga in age member at the base of the dominantly basaltic to (Appendix 11, Fig. 91B).

94 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A2050 Soidensuu arkosic wacke 14

12 Relative probability

6 Number

A) 0 2300 2400 2500 2600 2700 2800 2900 3000 3100 3200 Pb/Pb age

A2105 (165-AVL-89) Pitkälika trondhjemite xenolith

6 Relative probability

3 Number

B) 0 2550 2650 2750 2850 2950 3050 Pb/Pb Age

Fig. 91. Distribution of 207Pb/206Pb ages of concordant U–Pb data for A) Soidensuu arkosic wacke (A2050) and B) Pitkälika trondhjemite xenoliths (A2105).

95 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

7.3 The Kapustakangas intrusive suite (A1373)

207 206 The Kapustakangas suite refers to mafic-ultra- yielded Pb/ Pb ages of ca. 2.33–2.36 Ga, whereas mafic intrusions within the Central Puolanka the data obtained from turbid zircon plot on a loose- Group flanking the western margin of the Kainuu fitting chord with an upper intercept age of ca. schist belt (Laajoki 1991). In an attempt to date 1.8 Ga. The Th/U ratio in the older grains is rela- these intrusions, sample A1373 was collected from tively high, as is often typical for igneous zircon of a coarse-grained pegmatoid segregation in the mafic rocks. In contrast, in the 1.8 Ga zircons, the upper gabbroic part of a gabbro-peridotite sill at Th/U ratio is low, as it is in many cases for meta- Paha Kapustasuo (Fig. 89). The lower peridotite morphic zircon grains. part of the intrusion is now serpentinite, and an Recently, zircon from sample A1373 was also intrusive contact of the upper metagabbro against analysed using laser ablation MC-ICP-MS. The staurolite schist of the Central Puolanka Group can fourteen data points analysed were all from turbid be seen at the sampling site. The effects of pervasive zircon grains. Seven of them were concordant at amphibolite facies metamorphism and subsequent 1794 ± 15 Ma. Six analyses yielded Pb–Pb ages of retrograde alteration of the Paha Kapustasuo rocks ca. 1.86–1.90 Ga and one spot on a distinctly more are also reflected in the zircon grains, which are pristine domain had a Pb–Pb age of ca. 2.36 Ga reddish and turbid. The two multigrain analyses by (Fig. 92, Appendix 11). The U–Pb results suggest TIMS yielded discordant results, with Pb/Pb ages of that zircon in A1373 was badly altered by high- ca. 1.95 Ga (Fig. 92, Appendix 5). temperature fluid activity at ca. 1.8 Ga, which was Subsequent to the TIMS analyses, zircon grains probably related to the extensive late granite vein- from sample A1373 were analysed using NORDSIM ing west of the Kainuu schist belt approximately at in Stockholm (n652_1999, Fig. 92, Appendix 4a). the same time. The three analyses on the apparently The SIMS U–Pb data on nine zircon grains are scat- best-preserved zircon domains suggested that the tered and discordant. Two analyses on the appar- emplacement age of the Kapustakangas intrusions ently most pristine, slightly translucent domains was at least 2.36 Ga. Of particular note is that no

A1373 Paha Kapustasuo metagabbro 0.55 data-point error ellipses are 2s

2500

0.45 Concordia Age = 2300

1794 ±15 Ma

U n=7 youngest 2100 Concordia Age = 238 0.35 1900 2358 ± 24 Ma (grain A1373-12a) Pb/ 1700 206 1500 TIMS 0.25 1300 SIMS

0.15 2 4 6 8 10 12 207Pb/235U Fig. 92. Concordia plot of U–Pb zircon data obtained from the Kapustakangas suite gabbro A1373. SIMS analyses are presented as blue triangles (also shown in Figure 90 as green error ellipsoids), ID-TIMS analyses as red dots and LA-MC-ICPMS data as error ellipsoids.

96 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland ages close to 2.2 Ga, which would suggest a linkage A1373 is strongly enriched in LREE and yielded an to the 2.22 Ga gabbro-wehrlite (karjalite) sills, were εNd value of -0.6 at 2.4 Ga. obtained. Sm–Nd analysis indicated that sample

7.4 The 2.22 Ga intrusions in the Kainuu schist belt

In the past years, many samples have been taken dyke forms a ca. 10-km-long, mosty 200-m-wide from mafic sills and dykes in the Kainuu schist belt, intrusion within the Archaean rocks and seems to but magmatic ages have often remained unresolved turn into a sill within the Jatulian quartzites when because of problems with metamorphic overprint- entering the eastern boundary of the Kainuu schist ing and typically a high degree of discordance of belt (Fig. 89). Sample A198 is from a medium- zircon. Recent analyses by LA-MC-ICPMS have grained, only slightly deformed metadiabase, with confirmed that several of these dyke samples actinolitic hornblende and sodic plagioclase as its come from differentiated intrusions of the ca. 2.22 main constituents. Ga gabbro-wehrlite association (GWA), which are Most zircon grains in sample A198 are turbid and characteristic of Karelian schist belts throughout appear to be strongly altered. The recently acquired Finland (Hanski et al. 2010). LA-MC-ICPMS data on better preserved domains Two of these 2.22 Ga intrusions, Mustikkarinne give concordant U–Pb results, which suggest an (A198) and Raatelampi (A199), are located in the age of 2210 ± 10 Ma (Fig. 93, Appendix 11). Some southern part of the Kainuu schist belt. The sam- of the analysed zircon spots have high U (>1000 ples were collected by Matti Havola in the 1980s. ppm) and tend to give younger Pb/Pb ages. The Sample A198 Mustikkarinne is from a major mafic three multigrain TIMS analyses performed before dyke east of the central part of the Kainuu schist the laser experiment yielded discordant results and belt in the Sotkamo area. Owing to its high mag- were consistent with the interpretation based on the netite content, the Mustikkarinne dyke is traceable LA-MC-ICPMS data (Fig. 93, Appendix 5). on geophysical maps as a distinct aeromagnetic Sample A199 Raatelampi is from a leucogabbroic anomaly. Based on outcrop and magnetic maps, the part of a mafic sill intruding the interface of narrow

A198 Mustikkarinne metadiabase 0.50 data-point error ellipses are 2s

Concordia Age = 2210 10 Ma 0.46 MSWD=2 (n=12/15) 2400

2300 0.42

U 2200

238 2100 0.38 2000 Pb/

206 1900 0.34 1800

0.30 Intercepts at 992 250 & 2192 19 Ma TIMS MSWD = 0.43 n=15 (all) 0.26 4.5 5.5 6.5 7.5 8.5 9.5 207Pb/235U

Fig. 93. Concordia plot of U–Pb zircon data obtained from the Mustikkarinne diabase A198. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A199 Raatelampi metadiabase

data-point error ellipses are 2s 0.48 Concordia Age = 2222 7 Ma MSWD= 0.3 (n=23/25) 2400 0.44

2200

U 0.40 238 2000 0.36 Pb/

206 1800 0.32

1600 0.28 TIMS

0.24 3 5 7 9 207Pb/235U

Fig. 94. Concordia plot of U–Pb zircon data obtained from the Raatelampi mafic sill (sample A199). LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

stripes of Archaean gneisses and Jatulian quartzites ervation and the U–Pb data obtained by LA-MC- in a fault-bound (thrust) sliver within the Kainuu ICPMS provide concordant results with an age of schist belt, ca. 5 km to the north of the Talvivaara 2214 ± 7 Ma (Fig. 95, Appendix 11). Three earlier mine (Fig. 89). In contrast to the grains in sample discordant multigrain TIMS analyses have yielded A198, most zircon grains in sample A199 are fairly an upper intercept age of 2189 ± 8 Ma. pristine. The LA-MC-ICPMS data on good zircon Sample A759 Kettukallio is from a mafic dyke domains yield a concordant age of 2222 ± 7 Ma in the Kettukallio quartzites of the Kalpio complex (Fig. 94, Appendix 11). Analyses on altered domains (Laajoki 1991), collected close to the sediment con- yielded much younger apparent ages and explain tact. Most zircon grains in this sample are turbid the previous discordant multigrain TIMS results. and appear altered, but there are also more pristine-

The initial εNd(2220 Ma) values for these dykes are looking grains. The LA-MC-ICPMS data on these close to zero (A198 +0.7, A199 -1.0, Appendix 1), grains are concordant and give an age of 2202 ± which is typical for rocks of this family (Hanski et 5 Ma (n = 26, Fig. 96, Appendix 11). The results of al. 2010). analyses on altered domains (10b) are discordant Three of the samples with an age of ca. 2.2 with younger Pb/Pb ages, explaining well the scat- Ga, A769 Niskansuo, A 759 Kettukallio and A651 ter in the previously obtained TIMS data. A TIMS Honkaniemi, are from mafic intrusions in the analysis on monazite yielded a concordant age of mostly metasedimentary schists and gneisses of the 1792 ± 3 Ma, recording the timing of metamor- Prejatulian Central Puolanka Group and the Kalpio phism, implying growth or blocking of the monazite Complex, its presumed lithodemic correlative. during the 1800 Ma “Kajaani event” that produced According to Laajoki (1991), the Niskansuo metadi- voluminous Himalaya-type leucogranite-pegmatite abase (A769) has intruded into the contact between granite west of the KSB. The initial εNd value of the the Pärekangas and Mäntykangas Formations, i.e. whole-rock powder of this sample is -0.4 (Appendix between metasediments of the Central Puolanka 1), typical for the 2.22 Ga GWA intrusions (Hanski Group and Jatulian Vihajärvi Group (Fig. 89). Zircon et al. 2010). grains obtained from sample A769 show good pres-

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A769 Niskansuo metadiabase data-point error ellipses are 2s

0.48

Concordia Age = 2214 7 Ma n=11/13 0.44

U 2300 238 2200

Pb/ 0.40 2100 206

2000 0.36 TIMS Intercepts at 204 25 & 2189 8 Ma n=3 (2 outside figure) 0.32 5.5 6.5 7.5 8.5 9.5 207Pb/235U

Fig. 95. Concordia plot of U–Pb zircon data obtained from the Niskansuo diabase A769. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

Kettukallio data-point error ellipses are 2s

0.50 A759 Kettukallio metadiabase LA- MC-ICPMS Concordia Age 0.46 2400 2202 5 Ma (n=26)

0.42

U 2200 TIMS A759 Monazite 1792 3 Ma A760 zr 238 0.38 TIMS A760 Titanite 1784 5 Ma2000 Pb/ 0.34

206 1800

0.30 1600

0.26

TIMS A759 0.22 3 5 7 9 207Pb/235U Fig. 96. Concordia plot of U–Pb data on zircon, monazite and titanite obtained from the Kettukallio diabase A759 and amphibolite A760. LA-MC-ICPMS analyses on zircon are presented as error ellipsoids and ID-TIMS analyses as dots.

99 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

A sample (A760) from a banded amphibolite poor in their preservation. Three TIMS analyses interlayer within the Kettukallio quartzite was on multigrain zircon fractions were performed, also collected for isotope studies. This sample has resulting in scattered data with Pb–Pb ages of only been studied for multigrain fractions by TIMS. ca. 2.1–2.2 Ga. The data subsequently obtained by Both of the performed two TIMS analyses yielded LA-MC-ICPMS are also strongly scattered (Fig. 97, discordant results with Pb–Pb ages of ca. 2.5 Ga Appendix 11). Analyses of the domains that appear (Laajoki 1991, Fig. 96, Appendix 5). Titanite from best preserved have tended to give ages approach- the amphibolite gave a concordant TIMS U–Pb age ing 2.2 Ga. Some technically good data give clearly of 1784 ± 5 Ma, providing further support for the younger ages, obviously due to metamorphic effects age of metamorphism as discussed above. Ages of (15a, 21a), which can be expected based on the con- ca. 1.8 Ga have previously been obtained for mona- cordant TIMS age of ca. 1.84 obtained for titanite. zites from the Kalpio complex and also from the Part of the analyses produced discordant data with correlative Kalhamajärvi complex further north Pb–Pb ages exceeding 2.2 Ga. For these analyses, (Vaasjoki et al. 2001). Sm–Nd analysis indicated the relatively low 206Pb/204Pb ratios imply possible that the Kettukallio amphibolite is strongly enriched problems with common-lead correction. According in LREE and provided a TDM model age of 2.85 Ga to Laajoki (1991), the sampled intrusion is part of (Appendix 1), indicating an Archaean age or a large the Kapustakangas igneous suite, as is the Paha component of inherited Archaean material. Kapustasuo sample A1373 discussed above, for Sample A651-Honkaniemi (Vihajärventie) is which an age exceeding 2.36 Ga could be specu- from a layered mafic–ultramafic intrusion found lated. The data available for sample A651 do not within staurolite mica schists of the Puolankajärvi allow strict conclusions on the magmatic age of the Formation, the lowermost main unit of the Central Honkaniemi intrusion, although an age of ca. 2.2 Ga Puolanka Group (Laajoki 1991, Fig. 89). It was col- could be suggested as the most probable one. Sm– lected from the upper gabbroic part of the intrusion Nd analysis on sample A651 gave an εNd(2200 Ma) ca. 3 km west of the Niskansuo (A769) site discussed value of -2.7, which is much lower than those above. The extracted zircon grains appear mostly measured for typical GWA intrusions.

A651 Vihajärventie gabbro data-point error ellipses are 2s

0.52 LA-MCICPMS data with 206/204Pb > 4000 (n=21/33)

0.48 "7 best preserved domains" Pb/Pb ages ca. 2.16-2.19 Ga 2400 0.44

2200 206Pb 0.40 238 2000 U 0.36 206Pb/204Pb<4000

1800 0.32 TIMS 1600 0.28 TIMS A651D titanite Concordia Age = 1838 13 Ma 0.24 3 5 7 9 11 207Pb/235U Fig. 97. Concordia plot of U–Pb zircon and titanite data obtained from the Honkaniemi gabbro A651. LA-MC- ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as dots.

100 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Gabbroic rocks within Jatulian quartzites

A496 Jokijyrkkä (red circle) 2200 0.4A437 Latolanvaara (green square) A613 Hukkavaara (blue triangle) 2000

0.3 1600 U 238 1200

Pb/ 0.2 206

800 A496 Jokijyrkkä Intercepts at 0.1 269 150 & 2192 37 Ma 400 MSWD = 47 n=6(/8)

0.0 0 2 4 6 8 207Pb/235U

Fig. 98. Concordia plot of TIMS U–Pb zircon data obtained from four mafic samples from the Kainuu schist belt.

Over the past decades, several other samples the dominant rock unit in the eastern part of the from the Puolanka area have been collected for dat- Kainuu schist belt (Laajoki 1991). Only discordant ing purposes. Three of these, A496 Jokijyrkkä, A613 multigrain TIMS data on zircon (and baddeleyite) Hukkavaara and A437 Latolanvaara, are from three are available, but emplacement ages close to 2.2 Ga separate mafic intrusions, all intruding Jatulian seem most probable for the three sampled intru- quartzites of the East Puolanka Group, which is sions (Fig. 98, Appendix 5).

7.5 The 1.95 Ga Jormua ophiolite

Kontinen (1987) published an age of ca. 1.95–1.96 Ga for a Jormua gabbro pegmatoid (A1402; Peltonen for the rocks of the gabbro-Fe-gabbro-plagiogran- et al. 1998). ite suite located in the upper part of the Jormua Peltonen et al. (2003) published U–Pb SIMS anal- ophiolite. This age estimate was based on the U–Pb yses conducted on zircon from two clinopyroxene- zircon data produced by O. Kouvo on trondhjemite rich dyke samples, A1528 and JCX-23B, from the (A196) and gabbro samples (A729). Given the some- central (Lehmivaara) and western (Hannusranta) what discordant and heterogeneous data on which mantle blocks of the Jormua ophiolite, respectively. this age estimate was based, we re-analysed zircon They regarded sample A1528 as belonging to their from these two samples using the CA-TIMS method “OIB dykes” and interpreted sample JCX-23B and by Mattinson (2005). The chemically abraded zir- other similar dykes in the Hannusranta block as con fractions yielded concordant data with ages of coeval-cogenetic with the cumulate dykes. In addi- 1950 ± 3 Ma for the trondhjemite sample A196 and tion to zircon grains with ages of ca. 2 Ga, several 1952 ± 2 Ma for the gabbro sample A729 (Fig. 99, xenocrystic Archaean zircon grains were discovered Appendix 5). These results are identical to the pre- in both samples. We note here that a previously cise TIMS age of 1953 ± 3 Ma previously obtained unpublished multigrain TIMS result from sample

101

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

data-point error ellipses are 2s 0.37 A196 Jormua trondhjemite CA-TIMS analysis A196G 1990 0.36 Concordia Age = 1950 3 Ma 1970

1950 A196G 4.3-4.5 +200 CA 0.35 1930 1910 206 1890 Pb 0.34 1870 A196B 4.45-4.55 238 U 1850 A196C 4.3-4.55 A196D 4.2-4.3 0.33 A196A +4.5 Intercepts at A196F 4.0-4.2 CRU A196E 4.0-4.2 HF -33 230 & 1948.9 6.9 Ma 0.32 MSWD = 5.1 n=7

0.31 A) 5.1 5.3 5.5 5.7 5.9 6.1 207 235 Pb/ U Kontinen 1987: 1954 11 Ma

data-point error ellipses are 2s 0.365 A729 Jormua gabbro 1990 CA-TIMS analysis E 1970 0.355 Concordia Age = 1952 2 Ma 1950 A729E 4.2-4.4 +125 CA 1930

U 0.345 1910

238 1890 A729C +4.3 1870 A729A GABBRO +4.2 Jormua Pb/ A729D 4.2-4.3 0.335 206

A729B 4.0-4.2 0.325 Intercepts at 220 180 & 1952 4 Ma MSWD = 0.5, n=5 B) 0.315 5.2 5.4 5.6 5.8 6.0 207 235 Pb/ U Kontinen 1987: 1960 12 Ma

Fig. 99. A) Concordia plot of U–Pb TIMS data on zircon obtained from the Jormua trondhjemite A196 (data from Kontinen 1987 and this study). B) Concordia plot of U–Pb TIMS data on zircon obtained from the Jormua gabbro A729 (data from Kontinen 1987 and this study).

102 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

A1528 is nearly concordant and shows a Pb–Pb age Zr- and P-poor serpentinite (peridotite) host are of 2.01 Ga, which implies that zircon in this sample rich in zircon and apatite, which favours a magmatic must be dominantly Palaeoproterozoic in age (Fig. rather than metasomatic-hydrothermal origin. Two 100, Appendix 5). U–Pb analyses by SIMS on dark zircon cores yielded Zircon from two other samples of the Jormua dates of ca. 2.08 Ga, but the other data on fairly clear central and western block dykes has also been ana- homogeneous zircon are concordant at ca. 1.95 Ga. lysed (Appendix 4a). Sample A1529 picked from the These clear zircon grains have very low concentra- central block represents a heavily altered (chlori- tions of U and Pb, and consequently larger than tised), apatite-bearing OIB dyke, which yielded a typical analytical errors (Fig. 101, Appendix 4a). In small quantity of zircon grains. Five concordant contrast to Peltonen et al. (2003), we are inclined to analyses by SIMS on these grains indicated ages accept these low-U zircon grains as primary igneous from 1.96 to 2.02 Ga (Fig. 100, Appendix 4a). A TIMS phases from the carbonatitic magma and interpret analysis on a multigrain fraction was slightly dis- the older grains as xenocrystic. cordant, with a Pb–Pb age of 1974 Ma, which is The earlier Sm–Nd results from the Jormua consistent with the results of in situ analyses (Fig. ophiolite suggested that the E-MORB-type 100, Appendix 5). basaltic, gabbroic and plagiogranitic rocks have

The other sample (60L-ATK) represents carbon- εNd(1950 Ma) values of ca. +2, whereas the at atitic veins occurring at the southern margin of the least somewhat older OIB-like dykes tend to give

Hannusranta block of the Jormua ophiolite. It is to εNd(1950 Ma) values close to zero (Huhma 1986, be noted that the sample was collected from a boul- Peltonen et al. 1996, 1998). The dykes discussed der. However, there are several serpentinite boul- here also yield εNd values at 1.95 Ga close to zero, ders with carbonatitic veins in the sampled boulder except the carbonatitic (boulder) sample 125-JCB, field, attesting for a local source in the Hannusranta which gives εNd(1950 Ma) of +1.9 (Appendix 1). In sliver. Also, a drill core at the site has intersected addition to whole-rock powder, we have also ana- thin carbonatitic veins in serpentinite. Several of lysed apatite from this sample, which together these carbonate-rich veins occurring in a very with the whole-rock powder gave a Sm–Nd age

A1529 Jormua OIB dyke 0.41 data-point error ellipses are 2s

0.39 2100

U 2060 238 0.37 2020 Pb/ 1980 206 A1528 TIMS 1940 0.35 A1529 OIB dyke Concordia Age = 1970 ± 8 Ma A1529A TIMS n=4

0.33 5.5 5.7 5.9 6.1 6.3 6.5 6.7 6.9 207Pb/235U Fig. 100. Concordia plot of U–Pb zircon data obtained from the Jormua OIB dyke A1529. SIMS analyses are presented as error ellipsoids and an ID-TIMS analysis as a red dot. An ID-TIMS analysis on a clinopyroxenite dyke A1528 is also shown.

103 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

60L-ATK, Jormua carbonatitic vein 0.55 data-point error ellipses are 2s

60L-ATK, carbonatitic vein 0.45 Concordia Age = 1948 ±30 Ma n=6 2200

206 2000 Pb 0.35 238U 1800 1600 Concordia Age = 0.25 1400 2084 ±10 Ma, n=2

0.15 2.5 3.5 4.5 5.5 6.5 7.5 8.5 207Pb/235U

Fig. 101. Concordia plot of U–Pb zircon SIMS data obtained from the Jormua carbonatitic vein.

Jormua 0.517 97-JCX zirconolite Nd 97-JCX zirconolite-whole rock

144 Age = 1782±16 Ma Nd/

143 0.515

125-JCB carbonatite boulder apatite-whole rock 0.513 Age = 1765±74 Ma 97-JCX 125-JCB apatite carbonatitic vein 60L-ATK 125-JCB clinopyroxenite dyke

0.511 OIB dykes 0.05 0.15 0.25 0.35 0.45 0.55 0.65 147Sm/144Nd

Fig. 102. Sm–Nd isotope data for whole rock and mineral separates from two Jormua samples.

104 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland estimate of 1765 ± 74 Ma (Fig. 102). We have also ably records the time of formation of zirconolite. analysed a zirconolite (CaZrTi2O7) concentrate from This is consistent with the discordant results of a a clinopyroxenite-ilmenite-magnetite cumulate poor U–Pb analysis, giving a Pb–Pb age of 1751 ± (97-JCX). This analysis indicated that the zircono- 69 Ma (unpublished). Secondary REE mobility was lite has a high concentration of REE and a strongly also involved in the Jormua metabasalts, which subchondritic LREE/HREE ratio. Sm–Nd analysis have yielded a Sm–Nd isochron age of 1.72 ± 0.12 Ga of zirconolite combined with whole-rock isotope (Peltonen et al. 1996). data suggests an age of 1782 ± 16 Ma, which prob-

7.6 Volcanic rocks

In addition to the Jormua metabasalts, Sm–Nd is underlain by a conglomerate, which contains isotope data are available from two other mafic granophyre clasts from the 2.44 Ga Junttilanniemi volcanic formations in the Kainuu schist belt, intrusion. The volcanic rocks in both formations are the Matinvaara and Varisniemi Formations. The enriched in LREE and have yielded clearly nega-

Matinvaara Formation represents the lowermost tive εNd(2400 Ma) values and Archaean TDM ages Palaeoproterozoic volcanic unit on the Archaean (Appendix 6). These results compare well with the basement in the northern part of the Kainuu data obtained from the Kuntijärvi Formation, the schist belt in Kurkikylä (Fig. 89). The Varisniemi lowest unit in the Kuusamo schist belt (Chapter Formation in the central part of the Kainuu belt 4.6).

8 IISALMI COMPLEX

8.1 Geological background

The Archaean Iisalmi complex is located between the Iisalmi complex and have yielded ages of ca. 2.3 and Kainuu schist belt in the NE and the Svecofennian 2.1 Ga (Paavola 1988, Toivola et al. 1991, Hölttä et domain in the WSW. It is separated from the Lentua al. 2000). These rocks have been utilised to study complex on its eastern side by a Palaeoproterozoic the effects of the Palaeoproterozoic deformation SE–NW-trending shear zone. The rocks in the east- and metamorphism on Archaean rocks in the area. ern and western parts of the Iisalmi complex dif- In the areas retaining granulite-facies mineralogy, fer significantly in their lithology and ages. In the the dykes are mostly undeformed and unmetamor- east (Rautavaara complex of Hölttä et al. 2012), the phosed, whereas in the areas retrogressed during rocks are dominated by TTG gneisses with vari- the ca. 1.9 Ga Svecofennian metamorphism, they are able amounts of amphibolite and biotite-plagioclase strongly deformed and display mineral assemblages paragneiss enclaves with strong signs of chemical of hydrated metadiabases (Paavola 1988, Toivola alteration, and the obtained ages fall in the range of et al. 1991). The western areas of the Iisalmi com- 2.66–2.75 Ga. In the west (Iisalmi complex of Hölttä plex also contain, in addition to felsic plutons, a et al. 2012), the oldest rocks are Mesoarchaean 3.2– few mafic intrusions related to the Svecofennian 3.1 Ga migmatitic gneisses (Mänttäri & Hölttä 2002), magmatism (Paavola 1988, Peltonen 2005b), and which were intruded by 2.70 Ga quartz diorites. At the southern areas, such as Kuopio and Siilinjärvi, ca. 2.68–2.62 Ga, the rocks were metamorphosed contain small belts of Palaeoproterozoic supracrus- up to medium-pressure granulite facies conditions tal rocks (Lukkarinen 2008). producing garnetiferous two-pyroxene mafic and In the southern part of the Iisalmi complex, intermediate granulites. Archaean granulites occur Palaeoproterozoic mafic volcanic rocks erupted on in an area of approximately 20 by 70 km in extent, Archaean basement are found in small supracrus- having been studied in detail in the Varpaisjärvi area tal belts, e.g., in the Siilinjärvi and Kuopio areas. In (Hölttä et al. 2000). the Siilinjärvi belt, they form bulk of the Koivusaari A large number of NW–SE-trending Palaeo Formation, occurring as pillowed and massive lavas proterozoic diabase dykes cut Archaean rocks in the with minor pyroclastic interbeds (Lukkarinen 2008).

105 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

An intervening felsic member has been dated at 2062 2.32 Ga, 2.13 Ga and 2.0 Ga, two gabbro–anorthosite ± 2 Ma using the U–Pb zircon method (Lukkarinen plutons, the 2.06 Ga Otanmäki Fe–Ti–V ore-bear- 2008). Mafic intrusive magmatism of approxi- ing intrusion, and the 1.89 Ga Lapinlahti intrusion, mately the same age is represented by the Otanmäki with the last-mentioned intrusion representing the Fe–Ti–V oxide-bearing gabbroic intrusion, which Svecofennian synorogenic magmatism. In addition, is located in the NE part of the Iisalmi complex, we document Nd isotope characteristics of Jatulian close to the Kainuu schist belt (Lindholm & Anttonen 2.06 Ga volcanic rocks that occur in the Siilinjärvi 1980, Talvitie & Paarma 1980, Nykänen 1995). area (Fig. 74). In this paper, we report U–Pb and Sm–Nd isotope data on several mafic dykes with ages of ca.

8.2 The 2.3 Ga dykes, Humppi A135, Siunaussalmi A1369, Petäiskangas A1362

The NW–SE-trending Humppi dyke (A135) cuts ple A135. Under the microscope, the plagioclase in NW–SE-trending Archaean gneisses at Lapinlahti this coarse-grained rock looks fairly fresh, with in the Iisalmi complex (Fig. 74). Paavola (1988) only some patchy clouding, whereas the pyroxene reported an age estimate of 2331 ± 33 Ma for the is partly altered to amphibole. The three analyses dyke. This was based on slightly heterogeneous yielded an age estimate of 2270 ± 40 Ma (εNd = +1.3, U–Pb data on zircon. Recently, zircon grains from MSWD = 0.8, Fig. 105). However, the overall REE this sample were re-analysed using laser ablation level in the analysed pyroxene is relatively high MC-ICP-MS. Nine data points gave a concordia (Appendix 1) and some metamorphic effects on the age of 2323 ± 13 Ma (Fig. 103, Appendix 11), being Sm–Nd system are likely, as also witnessed by the consistent within error with the date obtained by amphibole growth on pyroxene grains. Paavola (1988). Another example of the ca. 2.3 Ga mafic dykes is We analysed Sm–Nd isotope ratios for the provided by the Tulisaari dyke, which is one of the pyroxene and plagioclase concentrates from sam- relatively wide and long, E–W-trending diabases

A135 Humppi diabase data-point error ellipses are 2s 0.60

0.56 A135 Concordia Age = 0.52 2323 ±13 Ma n=9 2600 0.48 206 Pb 2400 238U 0.44 2200 0.40

2000 TIMS A135 A1333 Tulisaari Intercepts at 0.36 801±320 & 2307±21 Ma MSWD = 6.5 (n=8, x) 0.32 5 7 9 11 13 207Pb/235U Fig. 103. Concordia plot of U–Pb zircon data obtained from the Humppi diabase A135. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red triangles. TIMS data on Tulisaari dyke A1333 are shown for reference (x).

106 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland in the Varpaisjärvi area (Fig. 74). It has a width The age and initial 143Nd/144Nd ratio of the Tulisaari of ca. 200 m and a length of at least several kilo- dyke (A1333 & A1369) are thus roughly similar to metres. Based on discordant multigrain TIMS data, those of the Humppi dyke (A135). The samples from Hölttä et al. (2000) reported a U–Pb age of 2295 these dykes have similar trace element characteris- ± 5 Ma for zircon from a coarse-grained gabbroic tics; however, the samples are rather course grained sample A1333. The data are slightly heterogeneous, and thus may not precisely represent the composi- and including all eight data points, a somewhat tion of the injected Fe-tholeiitic magma. Provided less precise upper intercept age of 2307 ± 21 Ma is that the dykes are coeval, the best estimate for the obtained (Fig. 103). intrusion age is 2323 ± 13 Ma. With this age, the

A new sample, A1369 Siunaussalmi, was taken initial εNd values for the two samples from the dyke from the Tulisaari dyke for Sm–Nd isotope analy- are +1.3 and +1.6 (Appendix 1). Using all Sm–Nd data sis. The plagioclase in this sample appears fresh in gathered from these two samples in the regression thin section, but the margins of pyroxene grains yields an age of 2322 ± 67 Ma (εNd = +1.6, MSWD = are slightly altered to amphibole. However, after 3.5, n = 6). separation, including hand-picking as its final The Petäiskangas mafic dyke (A1362) intrudes step, the pyroxene fraction was clean looking. Due Archaean rocks in the Manamansalo complex west to slight technical problems, the Sm/Nd error in the of the Kainuu schist belt (Fig. 89). The Archaean plagioclase analysis is estimated larger (1%) than gneisses in the region are variably affected by the usual. Three analyses on the mineral separates and 1.8–1.9 Ga Svecofennian tectonic and metamorphic one whole-rock sample gave an age of 2350 ± 40 overprinting, and cross-cutting Palaeoproterozoic

Ma (εNd = +1.8, MSWD = 0.2, Fig. 104). It should diabase dykes are often foliated and even show fold- be noted that the Sm–Nd age estimates based on ing. In spite of this, in several locations, the dykes so few analyses are sensitive to slight variation in have surprisingly well-preserved parts with little the 143Nd/144Nd ratio, e.g., the age here would be 50 strain and partially preserved primary minerals. The 143 144 Ma younger if the Nd/ Nd ratio in pyroxene was ophitic texture of the dyke at the Petäiskangas sam- 0.005% smaller than the measured value. pling site is well preserved and the rock contains

0.5138

0.5134 A1369 Siunaussalmi A1369px Age = 2350 40 Ma eps = +1.8 0.5130 MSWD = 0.19 n=3

Nd 0.5126 A135px

144 A135 wr A135 Humppi

Nd/ 0.5122 A1369wr Age = 2270 38 Ma 143 eps = +1.3 0.5118 MSWD = 0.66 n=3

A1369plag A135 plag 0.5114 A135 Humppi zircon U-Pb age 2323 ± 13 Ma 0.5110 0.08 0.12 0.16 0.20 0.24 0.28 147Sm/144Nd

Fig. 104. Sm–Nd isotope data for whole rock and mineral separates from the Humppi (A135) and Siunaussalmi (A1369) dykes.

107 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye mainly clinopyroxene and plagioclase. However, an age of ca. 1.96 Ga, a lot of the zircon in A1362 garnet replacing plagioclase surrounding ilmeno- must be older than 1.84 Ga. All matters considered, magnetite grains and partial replacement of clino- it apprears conceivable that 2322 ± 28 Ma could be pyroxene grains by secondary amphiboles are the igneous age of this dyke, although more data evidence that the mineral assemblage is not purely are needed for a definite conclusion. primary magmatic in origin. In the sampled out- The mineral fractions from sample A1362 that crop, the diabase is crossed by many decimetre- to were used for Sm–Nd studies appear fairly clean, metre-wide shear zones with pervasive foliation although plagioclase is slightly turbid. The three and purely amphibolite facies secondary mineral Sm–Nd analyses yielded an isochron with an age composition. of 1932 ± 42 Ma (εNd = -1.4, MSWD = 0.8, Fig. 106), A small number of light turbid zircon grains were which may well register metamorphic effects (see obtained from sample A1362. The four multigrain below). TIMS analyses were all discordant and yielded a In order to date the garnet-producing metamor- chord with an upper intercept with the concordia phic event, another NW–SE-trending dyke was curve at 1964 ± 32 Ma (Fig. 105, Appendix 5). As the sampled at Liminpuro, ca. 1 km south of the A1362 multigrain data were quite scattered, this sample locality. Sample A1457 Liminpuro contains abun- was also recently studied using the laser ablation dant garnet replacing ilmeno-magnetite and plagi- MC-ICP-MS. Seventeen of the total of 22 analyses oclase. The stepwise dissolution method by DeWolf gave an age of 1842 ± 10 Ma. Two of the analysed et al. (1996), involving powdering in a boron carbide spots suggest ages of 2.0–2.1 Ga, but must be con- mortar as the first step and leaching in 6N HCl as sidered suspect because of the considerable amount the second step, was used for Sm–Nd analysis of of common lead, whereas two analyses on a single garnet (Fig. 107). Combining the garnet data with grain (18 in Fig. 105) were concordant at 2322 ± the analytical results from two whole-rock pow- 28 Ma (Appendix 11, Fig. 105). The grain (18) is dis- ders, 1794 ± 15 Ma can be calculated for the clo- tinct in its BSE image, appearing better preserved sure timing of garnet. This is close to the age of the than the main population. As the TIMS data suggest dominant zircon phase in the previously considered

A1362 Petäiskangas Vaala metadiabase 0.52 data-point error ellipses are 2s

Primary igneous ? 0.48 Concordia Age = 2322 28 Ma 2500 n=2 (only grain 18) A1362-18a, 18b 0.44 2300

U Strong metamorphic stage 0.40

238 Concordia Age = 2100 1842 10 Ma (n=17)

Pb/ 0.36 LA-MC-ICPMS Intercepts at 1900 206 1800 65 & 2337 160 Ma 0.32 MSWD = 0.76 (n=22) 1700 TIMS Intercepts at 0.28 492 330 & 1964 32 Ma 1500 MSWD = 9.1 (n=4) 0.24 2 4 6 8 10 207Pb/235U grain length ca. 100µm

Fig. 105. Concordia plot of U–Pb zircon data obtained from the Petäiskangas dyke A1362. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

108 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland sample A1362, as well as the age of the latest major igneous age could well be close to 2.3 Ga. The initial metamorphic-magmatic event in the area (e.g., εNd(2320 Ma) values for the whole-rock samples are Vaasjoki et al. 2001). It seems conceivable that the +1.1 (A1362) and +1.5 (A1457), and are thus similar to Sm–Nd age of ca. 1.93 Ga for the Petäiskangas dyke the results obtained for the 2.3 Ga dykes discussed also reflects major metamorphic effects, and the above.

0.5126 A1362 Petäiskangas dyke 0.5124 Age = 1930 ± 42 Ma epsilon = -1.3 A1362px 0.5122 MSWD = 0.77

0.5120

Nd A1362wr

144 0.5118 Nd/ 0.5116 143

0.5114 A1362plag 0.5112

0.5110 0.07 0.09 0.11 0.13 0.15 0.17 0.19 147Sm/144Nd Fig. 106. Sm–Nd isotope data for whole rock and mineral separates from the Petäiskangas dyke A1362.

0.520 A1457 Liminpuro dyke (cf. A1362)

A1457 grt 0.518

Nd 0.516 144 Garnet-whole rock Age Nd/ 0.514 1794 ± 15 Ma 143

A1457 wr 0.512

0.510 0.0 0.2 0.4 0.6 0.8 147Sm/144Nd Fig. 107. Sm–Nd isotope data for whole-rock sample and garnet from the Liminpuro dyke A1457.

109 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

8.3 The 2.13 Ga Nieminen dyke, A1223 and A1368

The E–W-trending, 25- to 30-m-wide Nieminen heterogeneous (Fig. 108, Appendix 5). Especially dyke is one of the well-preserved diabase dykes analysis C on a few brownish long crystals with intruding Archaean granulites in the Iisalmi com- a low density was very distinct from the other plex (Fig. 74). The dyke is nearly undeformed, data, providing a slightly discordant result with a shows a homogeneous diabasic texture, and is well 207Pb/206Pb age of 1866 Ma. The other seven analyses exposed in a dimension stone quarry, where the conducted on heavy zircon were variably discord- sampling for isotope studies was carried out. The ant, but all with 207Pb/206Pb ages of ca. 2.1 Ga (Fig. main silicate constituents are plagioclase and clino- 108). Rejecting one technically poor analysis (E), the pyroxene accompanied by minor orthopyroxene. six data points provide a chord with intercepts at Secondary minerals, such as light-green amphibole, 2100 ± 3 Ma and -256 ± 150 Ma (MSWD = 0.8). The biotite, and epidote, occur only sporadically along negative lower intercept age is obviously impossi- grain boundaries. Geochemically, the Nieminen ble, questioning the value of the data in determining dyke is tholeiitic basalt with 1.2 wt% TiO2 and the magmatic age of the Nieminen dyke. 6.2 wt% MgO and is in this respect similar in com- Subsequently, zircon was also re-analysed using position to many other diabase dykes in the area laser ablation and MC-ICP-MS. The obtained data (Toivola 1988). are all concordant within analytical error and yield The diabase in the Nieminen quarry contains a Concordia age of 2121 ± 7 Ma (Fig. 108, Appendix patches where the rock is more coarse-grained 11). As a whole, the TIMS data are roughly consistent and plagioclase-rich. Sample A1368 was taken from with the in situ analyses. such a pocket, which after crushing and separation The Nieminen dyke was also sampled for Sm–Nd yielded a small amount of dominantly clear, col- isotopic studies. The diabase sample A1223 consists ourless, long and acicular zircon grains. The eight of fresh and clean plagioclase, clinopyroxene and multigrain U–Pb TIMS analyses performed on vari- minor orthopyroxene as the main minerals. Early ous fractions of the zircon separate were, however, dating results on this sample were published by

A1368 Nieminen diabase data-point error ellipses are 2s

0.44

Concordia Age = 2121 ± 7 Ma 0.42 2200 0.40 206Pb 2100 0.38 238U 2000 0.36

1900 0.34 TIMS minimum age 2111 ± 5 Ma TIMS 0.32 5.0 5.4 5.8 6.2 6.6 7.0 7.4 7.8 207Pb/235U Fig. 108. Concordia plot of U–Pb zircon data obtained from the Nieminen diabase A1368. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

110

Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

0.5145 A1223 Nieminen dyke Age = 2127 42 Ma 0.5135 eps = +2.5 A1223px#2, #3, MSWD = 2.7 n=12 #4, #5, #6 A1223 px#1 Nd A1223wr#1, #2, #3 144 0.5125 Nd/ 143 A1223plag#5 A1223plag#4 0.5115 A1223plag

U-Pb zircon age 2121 ± 7 Ma 0.5105 0.06 0.10 0.14 0.18 0.22 0.26 0.30 147Sm/144Nd

Fig. 109. Sm–Nd isotope data for whole rock and mineral separates from the Nieminen dyke A1223.

Toivola et al. (1991), who reported an age estimate than 1 ppm in plagioclase), the error in the old of 2085 ± 95 Ma (εNd = +2.1), which was based on analyses is relatively large. There is, however, no Sm–Nd mineral data analysed by a non-commer- systematic error in the old data, and using all the cial, in-house-built mass spectrometer at GTK. 12 analyses available, the obtained isochron gives

These old analyses also involved aliquoting the HCl an age of 2127 ± 42 Ma (εNd = +2.5, MSWD = 2.7, Fig. solution prior to the addition of the Sm–Nd spike. 109, Appendix 1), which is consistent with the U–Pb

Subsequently, five more analyses have been con- zircon dating results. The positive initial εNd sug- ducted using the currently employed techniques. gests that the magma was derived from depleted Due to the low concentrations of Sm and Nd (less mantle without significant crustal contamination.

8.4 The 2.06 Ga Otanmäki intrusion, A1381

The Fe-Ti oxide-bearing Otanmäki intrusion is The reported eleven analyses, which are discord- located SE of Lake Oulunjärvi, west of the Kainuu ant with a relatively limited spread in Pb/U, yield schist belt (Fig. 89). Together with another mafic 207Pb/206Pb ages from 2020 to 2045 Ma. Using all intrusion, Vuorokas, it was mined for iron, tita- these data and the current regression algorithm nium and vanadinium from 1953 to 1985 (30 Mt). (Ludwig 2003), an upper intercept age of 2058 The southern contacts of the intrusions show that ± 13 Ma can be calculated (lower intercept at 809 ± they were emplaced into Archaean amphibolite- 220 Ma, MSWD = 2). banded TTG migmatites. At their northern mar- The Otanmäki and Vuorokas intrusions are for gins, the bodies have sharp, fault-defined contacts large parts strongly deformed and pervasively met- to ca. 2050 Ma syenites and peralkaline granites amorphosed. Nevertheless, especially the southern (Kontinen et al. 2013). parts of the Vuorokas intrusion also contain gabbro In earlier studies on the Otanmäki intrusion, zir- varieties in which primary igneous phases, pyrox- con occurring in mafic pegmatoids was used for U– enes and plagioclase, are surprisingly well preserved Pb dating. Based on the data provided by O. Kouvo (Nykänen 1995). Three samples of such well-pre- on two samples (A671, A672), an age of 2065 ± 4 Ma served gabbros from the middle and lower parts of has been published by Talvitie and Paarma (1980). the Vuorokas block were collected by V. Nykänen

111

Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

0.5132 Otanmäki intrusion 0.5128 A1381 Age = 2043 19 Ma px px Epsilon = -0.8 wr 0.5124 px MSWD = 1.5 n=9 wr Nd 0.5120 wr 144

Nd/ 0.5116 143 0.5112 plag plag plag 0.5108 A671& A672 U-Pb zircon age 2058 ± 15 Ma 0.5104 0.04 0.08 0.12 0.16 0.20 0.24 147Sm/144Nd

Fig. 110. Sm–Nd isotope data for whole-rock samples and mineral separates from the Otanmäki mafic intrusion (A1381). for mineral separation. They are the clinopyrox- cate that there is no significant variation in the ini- ene-orthopyroxene-olivine-plagioclase cumulate tial ratios between the different samples. The REE A1381a, clinopyroxene-orthopyroxene-plagioclase- level is relatively low and the most mafic cumulates olivine cumulate A1381b, and plagioclase-clinopy- have nearly chondritic relative REE concentrations. roxene-orthopyroxene cumulate A1381c. The initial 143Nd/144Nd ratio obtained from the isoch- As usual, the Sm–Nd analyses were conducted ron is not far from the chondritic ratio, either, and on high-purity hand-picked mineral fractions. The shows that the source for the magma was not in the nine analyses available define an isochron, which convective/depleted mantle, which is considered to gives an age of 2043 ± 19 Ma (εNd = -0.8, MSWD = have been characterised by positive time-integrated

1.5, Fig. 110, Appendix 1), corresponding to the U–Pb εNd values throughout the Earth’s history. age given by the pegmatoid zircons. The data indi-

8.5 The ca. 2.0 Ga dykes Koirakoski A1875, Jäkäläkangas A1838

Traditionally, it was commonly assumed that the Sonkajärvi area (Fig. 74). The dyke is ca. 30 m wide dominantly Fe-tholeiitic NW–SE-trending diabase and contains primary igneous pyroxene and pla- dykes occurring abundantly in the Archaean terrains gioclase in places. However, in the sampled outc- in eastern Finland are mostly ca. 2.1 Ga in age, as rop, the NE margin of the dyke is metamorphosed with the Nieminen dyke discussed above (e.g., Vuollo to hornblende-bearing metadiabase with zones of et al. 1992). Subsequently, it has become apparent schistose and folded amphibolite. Sample A1875 that this is not true, but this group of dykes com- collected for Sm–Nd studies represents the well- prises both significantly older and younger dykes. preserved SW part of the dyke. A thin section pre- The Koirakoski and Jäkäläkangas below are exam- pared from the separated mineral fractions revealed ples of the latter. that both the plagioclase and pyroxene fractions The Koirakoski dyke, which is located ca. 26 km are fresh; however, a few grains of amphibole were north of the Nieminen dyke, is one of the abun- observed in the pyroxene fraction. Sm–Nd analyses dant NW–SE-trending dykes cutting Archaean conducted on minerals and whole rock are techni- high-grade gneisses in the Varpaisjärvi-Iisalmi- cally of good quality and provide an isochron that

112 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

gives an age of 1968 ± 38 Ma (MSWD = 1.5, εNd = distinct within error, but it is possible that met- +0.5, Fig. 111, Appendix 1). Using analyses of the amorphic amphibole may have some effect on the whole-rock powder and plagioclase only, the result analytical results of the pyroxene fraction. is 2000 ± 63 Ma, while whole rock and pyroxene The Jäkäläkangas dyke is an at least 60-m-wide, give a date of 1913 ± 95 Ma. These dates are not SW–NE-trending mafic dyke cutting Archaean

0.5130 A1875 Koirakoski dyke Age = 1968 ± 38 Ma 0.5126 epsilon = +0.5 A1875px MSWD = 1.5

Nd 0.5122 A1875 144

Nd/ 0.5118 143

0.5114 A1875plag

0.5110 0.08 0.10 0.12 0.14 0.16 0.18 0.20 0.22 147Sm/144Nd Fig. 111. Sm–Nd isotope data for whole rock and mineral separates from the Koirakoski dyke A1875.

data-point error ellipses are 2s

0.354 A1838 Jäkäläkangas metadiabase Intercepts at 1940 0.350 289 ± 150 & 1946 ± 4 Ma A1838B . 1920

0.346 U 1900

238 0.342

Pb/ 1880 0.338 206

0.334 A1838A

0.330

0.326 5.3 5.4 5.5 5.6 5.7 5.8 207 235 Pb/ U zircon grains euhedral long turbid

Fig. 112. Concordia plot of U–Pb zircon TIMS data obtained from the Jäkäläkangas diabase A1838.

113 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye migmatitic gneisses west of the central Kainuu Only two multigrain TIMS U–Pb analyses have schist belt (Fig. 74). The dyke is only weakly been performed, but the one on a zircon fraction deformed to non-deformed and, in the sampled purified by mechanical abrasion overnight was outcrop, contains many irregular, metres-across nearly concordant (Fig. 112, Appendix 5). The two patches, which preserve primary minerals, the fore- fractions yielded an upper intercept age of 1946 ± most being clinopyroxene and plagioclase, although 4 Ma, which is also proposed to be the magmatic the rock is mostly hydrated to amphibolite-facies age of the Jäkäläkangas dyke. The previous discor- metadiabase. Sample A1838 Jäkäläkangas was dant U–Pb zircon analyses of some diabase samples collected from a decimetre-wide, coarse-grained (A256-8, A991; collected by M. Havola) from the band in a relatively well-preserved part of the Kainuu schist belt, east of the site of sample A1838, dyke. Mineral separation yielded a small amount are roughly compatible with the results from the of light-coloured, slightly turbid, needle-shaped Jäkäläkangas dyke. (>100 µm) zircon grains.

8.6 The 1.89 Ga Lapinlahti intrusion

The Lapinlahti gabbro refers to a concentrically and several tens of metres in length that contain zoned, circular (6–7 km) gabbro-anorthosite some zircon. Sample A1688 was taken from such intrusion and is among the rare examples of the a segregation in a gabbro exposed in a macadam synorogenic Svecofennian mafic intrusions that quarry at the Taskilanmäki locality. A small amount were emplaced into the Archaean crust close to the of dominantly transparent, pale zircon was recov- margin of the Karelian craton (Peltonen 2005b). ered and analysed using LA-MC-ICPMS. All data Paavola (1988) published an age of ca. 1.89 Ga are concordant and yield an age of 1888 ± 5 Ma, for the Lapinlahti gabbro. The main gabbro and which is identical to the age obtained in an earlier anorthosite phases of the Lapinlahti intrusion are multigrain TIMS analysis (Fig. 113, Appendix 11, 5). low in Zr and do not produce zircon grains in sepa- The Sm–Nd data on the two dated pegmatoid ration. However, locally, the gabbros have pegma- gabbro samples (A708, A1688) give distinctly nega- toid segregations up to several metres in thickness tive initial εNd values of -6.5 and -5.3, suggesting a

Fig. 113. Concordia plot of U–Pb zircon data obtained from the Lapinlahti gabbro (A1688). LA-MC—ICP-MS analyses are presented as error ellipsoids and an ID-TIMS analysis as a red dot.

114 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland significant contribution of Archaean LREE-enriched gabbros from the Svecofennian domain, having lithosphere (crust, Appendix 1). In this respect, a εNd values from +3 to 0 (Huhma 1986, Makkonen sharp contrast is obvious to the roughly coeval & Huhma 2007).

8.7 Volcanic rocks in the Siilinjärvi area

The Koivusaari Formation refers to a 20-km- obtained for 8 multigrain fractions are partly con- long and 3- to 4-km-wide, fault-bounded block cordant, providing an age of 2062 ± 6 Ma (Fig. 114, of Palaeoproterozoic metavolcanic rocks lying on Appendix 5). Archaean basement gneisses in the Siilinjärvi area, Whole-rock Sm–Nd isotope analyses conducted 20 km north of Kuopio (Lukkarinen 2008). The for- in the late 1980s on samples of mafic and felsic vol- mation consists of three mafic units, the Parkkila, canic rocks from the Koivusaari Formation yielded Vehkasuo and Vuorimäki Members, and one felsic somewhat scattered results. Although the analytical unit, the Kirjoniemi Member. The mafic amphib- errors are slightly larger than in recent analyses, the olite-facies metavolcanic rocks comprise massive scatter is mostly due to variation in the initial Nd and pillowed lavas with intercalated tuffs. The isotope composition of the samples (Appendix 6). rocks in the Parkkila and Vuorimäki Members clas- The tholeiitic pillow lavas of the Parkkila Member sify mainly as tholeiitic and those in the Vehkasuo (the lowest stratigraphic unit) and the Kirjoniemi Member alkaline basalts. The Kirjoniemi Member Member felsic rocks show clearly negative initial felsic rocks are amygdaloidal, quartz-phyric A-type εNd values, suggesting a major involvement of older rhyolites. LREE-enriched lithosphere in their genesis. The The age of the Koivusaari Formation is based on other mafic rocks, including the Vehkasuo Member old U–Pb TIMS analyses conducted on zircon grains alkali basalts, have yielded a positive εNd(2062 Ma) from two samples, A242 and A481, both represent- value of about +2. Using the whole dataset, the ing metarhyolites of the Kirjoniemi Member. The range in Sm/Nd allows an age of 2090 ± 62 Ma to U content of zircon is relatively low and the data be calculated (Fig. 115). Combining these isotope

Siilinjärvi felsic volcanics TIMS data-point error ellipses are 2s

0.39 2110 A481C +4.6/abr 5 h A481D +4.6/abr 3 h 2090

0.38 2070 A481F +4.6/>70/abr 5 h A481E +4.6/<70/abr 3 h 2050 206Pb 2030 A242A +4.6/HF 238U 0.37 A481A +4.6/HF/crushed 2010 A481B +4.6/HF/uncr.

1990 A242B 4.2-4.6/HF 0.36 Intercepts at 2062 ± 6 & 770 ± 260 Ma MSWD = 1.7 n=8

0.35 6.0 6.2 6.4 6.6 6.8 7.0 207Pb/235U Fig. 114. Concordia plot of U–Pb zircon data obtained from the Siilinjärvi felsic volcanic rocks (analyses by Kouvo/ GTK).

115115 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Siilinjärvi volcanics 0.5129

0.5127 Age = 2090 ± 62 Ma 0.5125 Nd-epsilon = +2.2 MSWD = 1.1 n=5

0.5123

143 Nd 0.5121 144 Nd Parkkila pillow lava 0.5119 Kirjoniemi felsic tuff Parkkila pillow lava 0.5117

Kirjoniemi felsic tuff 0.5115

0.5113 0.10 0.12 0.14 0.16 0.18 0.20 147Sm/144Nd Fig. 115. Sm–Nd isotope data for whole-rock samples of volcanic rocks from the Siilinjärvi area.

data with geochemical data, Lahtinen et al. (2015b) nental lithospheric mantle (SCLM) magma sources recently concluded that the Siilinjärvi volcanic rocks and were also contaminated by Archaean crust. were derived from both plume (OIB) and sub-conti-

9 THE OUTOKUMPU AREA

The Outokumpu allochthon in the North Karelia OKU794B/490.80–482.70 and chlorititised alka- schist belt comprises numerous fault-bound ophi- line dyke sample R393 (M52/4224/R393/422.40– olite fragments dominated by serpentinised metape- 427.40), both from the southern part of the Kylylahti ridotites, but many also with metagabbro and basalt body. Sample OKU794B/490.80–482.70 represents as small stocks and dykes (Peltonen et al. 2008). a pervasively carbonate-altered and subsequently Based on TIMS U–Pb data on zircon, Huhma (1986) metamorphosed (low amphibolite facies) mafic determined an age of 1972 ± 18 Ma for a coarse- rock, probably originally gabbro. Zircon grains grained metagabbro occurring in such a fragment at extracted from the sample are mostly euhedral, Horsmanaho. The relevance of this result was con- small and low in U. The U–Pb data obtained by TIMS firmed by Peltonen et al. (2008) by TIMS analyses and SIMS yield results that are consistent with the conducted on another gabbro (A1029 Huutokoski) earlier data from much less altered metagabbros, from another ophiolite fragment, which yielded an also suggesting an age of ca. 1.96 Ga. age of 1959 ± 5 Ma. They also reported a SIMS age Sample R393 comes from a several-metre-wide, of 1971 ± 15 Ma for a trondhjemite gneiss (A1754) pervasively chloritised and subsequently metamor- that is associated with metagabbroic rocks in the phosed, probably basaltic dyke occurring in a talc- Kylylahti ophiolite fragment. In addition, SIMS carbonate rock-serpentinite environment. High (SHRIMP) and TIMS data are available for two drill TiO2 and P2O5 suggest an alkaline affinity for the core samples from the Kylylahti ophiolitic ultramafic basaltic protolith. Most data on the dyke give simi- body. These are the skarnoid metabasite sample lar ages to the above-mentioned gabbros, but there

116 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland are also a few significantly older grains suggesting even higher value (~ +2.5) has been obtained from a xenocrystic origin (Fig. 116, Appendix 4b, 5). The the Horsmanaho gabbro (Huhma 1986). However, three earlier-performed TIMS analyses resulted in the Huutokoski gabbro sample A1029 has given an discordant Pb–Pb dates of ca. 2.1, probably repre- εNd(1960 Ma) value of ca. +0.5 (Appendix 1). The senting analyses of mixed zircon populations. large Sm–Nd database available from the ophiolitic Sm–Nd analysis conducted on sample R393 gabbroic-basaltic rocks of the Outokumpu region indicated a strong LREE enrichment and gave an do not suggest a significant quantity of rocks with ε (1950 Ma) value of -1.3 (Appendix 1). Peltonen et depleted mantle-like Nd isotope signatures, as Nd al. (2008) published an εNd(1960Skarn, Ma) Outokumpuvalue of +1.7 ophiolitethe initial εNd values are mostly close to zero (A. for the Kylylahti plagiogranite sample A1754, and an Kontinendata & -pointH. Huhma, error ellipses areunpublished). 2s 0.46 OKU794B AverageSkarn, 207Pb/206Pb Outokumpu age ophiolite 2300 0.42 1963 21 Ma MSWD = 0.46 data-point error ellipses are 2s 0.46 n=7 (nearly OKU794Bconcordant data) 2100 0.38 Average 207Pb/206Pb age

U 2300 1963 21 Ma 0.42 1900 238 0.34 MSWD = 0.46 n=7 (nearly concordant data) 2100 0.38

Pb/ 1700 U 0.30 1900 206 238 0.34 1500 0.26 OKU794B

Pb/ 1700 0.301300 nearly concordant data Intercepts at 0.22 206 300 300 (forced) & 1955 26 Ma 1500 0.26 MSWD = 0.29 n=7 OKU794B A) 0.18 2 1300 4 nearly6 concordant data Intercepts8 at 0.22 300 300 (forced) & 1955 26 Ma 207 235 triangle - TIMS d ata Pb/ UMSWD = 0.29 n=7 A) 0.18 2 4 6 8 triangle - TIMS d ata 207Pb/235U

Alkaline dyke, Outokumpu ophiolite 0.7 data-point error ellipses are 2s R393 Kylylahti Intercepts at 0.6 Alkaline417 240 dyke, & 1959 Outokumpu 14 Ma ophiolite 0.7 MSWD = 1.20 n=10 data-point error ellipses are 2s R393 Kylylahti 0.5 Intercepts at 2600 0.6 417 240 & 1959 14 Ma 206 Pb MSWD = 1.20 n=102200 0.4 238U 0.5 2600 1800 206 0.3 Pb 2200 0.4 1400 238U 0.2 1800 1000 Concordia Age = 0.3 1962 20 Ma 1400 n=6 B) 0.1 0.20 4 8 12 16 1000 Concordia Age = 207Pb/1962 235 20 UMa triangle - TIMS d ata n=6 B) 0.1 Fig. 116. A) Concordia plot of U–Pb0 zircon data4 obtained for 8the metabasaltic12 skarn rock sample16 OKU794B from Kylylahti. SIMS analyses are presented as error ellipsoids and ID-TIMS analyses as red triangles. B) Concordia 207 235 triangle - TIMS d ata plot of U–Pb zircon data obtained for the alkaline dyke R393Pb/ from Kylylahti.U SIMS analyses are presented as error ellipsoids and ID-TIMS analyses as red triangles.

117 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

10 TOHMAJÄRVI VOLCANIC COMPLEX AND A BASEMENT DYKE

10.1 Oravaara gabbro A398

The Tohmajärvi volcanic complex is a 20-km-long and are thought to be genetically related to a WNW- and 2-km-wide sliver of mafic sills and metavol- trending swarm of mafic diabase dykes cutting canic rocks located in the SE corner of the North the Archaean basement. Pekkarinen & Lukkarinen Karelia schist belt close to the Finnish–Russian bor- (1991) published a joint U–Pb zircon age of 2113 ± der (Nykänen 1971) (Fig. 74). Occurring in an anti- 4 Ma for two samples from such a dyke close to clinal structure together with Jatulian quartzites, the contact to the Palaeoproterozoic cover sequence conglomerates and dolomites, the volcanic complex (Kiihtelysvaara in Fig. 74). Although the Tohmajärvi is surrounded by younger Kalevian metawackes. It volcanic rocks differ in chemistry from the Koljola is composed of pillow lavas, pyroclastic deposits Formation rocks, they appear to be more or less and subvolcanic intrusions varying in composi- coeval. Namely, Huhma (1986) obtained an age of tion from magnesian basalts to Fe-rich andesites 2105 ± 15 Ma for the Tohmajärvi mafic magmatism with non-fractionated or slightly LREE-enriched based on slightly discordant TIMS analyses con- chondrite-normalised REE patterns (Nykänen et al. ducted on zircon separated from a gabbroic rock 1994). Pekkarinen & Lukkarinen (1991) correlated (sample A398), probably a sill, occurring in meta- the metavolcanics of the Tohmajärvi area with the volcanic rocks at Oravaara. Koljola Formation in the Kiintelysvaara area, which The age from the Oravaara gabbroic sample was is found 20–30 km north of Tohmajärvi in a well- recently tested by utilising the original zircon extract studied Jatulian supracrustal succession lying on from sample A398 and subjecting it to laser ablation Archaean basement gneisses. The Koljola Formation MC-ICP-MS analysis. Sixteen isotope measurements mafic lava flows and pyroclastic rocks are sand- were carried out on grain domains with the best wiched between two Jatulian quartzite formations apparent preservation quality. All the compositions

A398 Oravaara gabbro Tohmajärvi complex data-point error ellipses are 2s 0.43 Concordia Age = 2103 8 Ma n=16 (LA-MC-ICPMS) 2220 0.41 2180 2140 0.39 206Pb 2100 238U 2060 0.37 2020 1980

TIMS 0.35

0.33 5.8 6.2 6.6 7.0 7.4 7.8 207Pb/235U Fig. 117. Concordia plot of U–Pb zircon data obtained from the Oravaara gabbro A398. LA-MC-ICPMS analyses are presented as black error ellipsoids and ID-TIMS analyses by Huhma (1986) as red dots. The blue error ellipse denotes the final age calculated by the ISOPLOT program (Ludwig 2003) from the LA-MC-ICPMS data.

118 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland overlap with the concordia curve and provide an trometer. The new analyses yielded a slightly lower, age of 2103 ± 8 Ma, thus confirming the previously although within errors similar, εNd value of +1.6 ± obtained TIMS age (Fig. 117, Appendix 10). 0.5 (Appendix 1), which should better characterise Huhma (1986) analysed a whole-rock powder the sample because of the newer improved tech- of sample A398 for Sm–Nd isotopes, obtaining an nology involved. Compared to the Jatulian magma- initial εNd value of +2.6 ± 1.0. We re-analysed the tism at Kiihtelysvaara, as represented by the Hyypiä same sample in order to assess the reliability of sill, which has yielded an εNd(2100Ma) value of -1.7 the earlier Sm–Nd results, which were obtained (Huhma 1987), the Tohmajärvi rocks are clearly using aliquoting and a home-made mass spec- more depleted in their Nd isotope signature.

10.2 Purola dyke A1231

The southernmost part of the Archaean basement discordant, with a Pb–Pb age of 1.97 Ga (Appendix complex in eastern Finland, called the Ilomantsi 5). The new results obtained using laser ablation complex by Hölttä et al. (2012), is cut by tholeiitic MC-ICP-MS are concordant within error for the mafic dykes with their thickness falling commonly most pristine zircon domains and provide an age of in the range of 10–30 m, sometimes exceeding 100 2106 ± 11 Ma, whereas the data on altered domains m, and their orientation varying mostly between give Pb–Pb ages of ca. 1.87 Ga (Fig. 118). The older 275o and 325o (Pekkarinen & Lukkarinen 1991). age may be considered as the igneous age of the The Purola metadiabase dyke (A1231) represents a rock. Consequently, the Purola dyke appears coeval NW–SE-trending dyke swarm. The sampling site is with the Oravaara gabbro in the Tohmajärvi area and located 20 km northeast of Kiihtelysvaara (Fig. 74). the basement dyke dated from the Kiintelysvaara An old TIMS U–Pb analysis on turbid zircon was area in Pekkarinen & Lukkarinen (1991).

A1231 Purola metadiabase data-point error ellipses are 2s 0.50

0.46

2300

0.42 U

238 2100 0.38 Concordia Age = Pb/ 2106 ±11 Ma

206 1900 n=7 0.34

1700 0.30 TIMS Concordia Age = 1870 ±17 Ma n=7 0.26 3.5 4.5 5.5 6.5 7.5 8.5 9.5 207Pb/235U

Fig. 118. Concordia plot of U–Pb zircon data obtained from the Purola dyke A1231. LA-MC-ICPMS analyses are presented as error ellipsoids and an ID-TIMS analysis as a red triangle.

119 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

11 CARBONATITES AND LAMPROPHYRES

11.1 Geological background

There are around ten known carbonatite occur- age data from the Siilinjärvi, Laivajoki and Kortejärvi rences in Finland. The most prominent of them are carbonatites, obtained using the ID-TIMS, LA-MC- the Archaean ca. 2.6 Ga Siilinjärvi carbonatite in the ICPMS and SIMS methods. The same samples were Iisalmi complex (O’Brien et al. 2015), the host rock also used to measure the Nd isotope composition of of an active apatite mine, and the Sokli carbonatite these carbonatites. in eastern Lapland (O’Brien and Hyvönen 2015), Like carbonatites, lamprophyres also provide which is part of the Devonian Kola alkaline province useful probes of mantle sources, as due to their (Fig. 119) and contains large regolithic phosphorite usually high concentrations of REE, they are rela- resources. Smaller Palaeoproterozoic occurrences tively insensitive to crustal contamination. In this include the ~2.0 Ga Laivajoki and Kortejärvi carbon- paper, we deal with Palaeoproterozoic lamprophyres atites (O’Brien 2015), which are located in a shear occurring at Niinivaara, North Karelia, and Kuotko zone between the Archaean Lentua and Pudasjärvi and Palovaara in the Central Lapland greenstone complexes. In this paper, we provide refined zircon belt (Fig. 119).

Fig. 119. Location of the studied alkaline rocks. The dashed line shows the boundaries of the Devonian Kola alkaline province after O’Brien (2015).

120 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

11.2 The 2.61 Ga Siilinjärvi carbonatite

The first U–Pb analyses of zircon from the Siilinjärvi using a 50 µm laser spot and utilising three carbonatite were carried out in the late 1970s by low-U zircon standards having a range of ages up Olavi Kouvo. The results suggested an age close to Archaean (GJ1, A1912, A2024). These analyses to 2.6 Ga. These data on large (several cm) single yielded a concordia age of 2608 ± 6 Ma (Fig. 120, zircon crystals indicated an unusually low abun- Appendix 11). dance of uranium (5 ppm). Later, in the 1980s, U–Pb Sm–Nd analyses on two of the U–Pb-dated sam- analyses were conducted on three other carbonatite ples gave initial εNd values close to zero and model samples at GTK. The analytical results are listed in ages TDM of ca. 2.75 Ga (Appendix 1, analysed in Appendix 5 and plotted in the concordia diagram in 2001). These values are supported by the Nd isotope Figure 120. These data suggest an age of 2610 ± 4 composition of the apatite concentrate collected Ma. This result is supported by the 207Pb/206Pb age of from the production line of the Siilinjärvi mine (Nd subsequent CA-TIMS analysis on a zircon fraction = 700 ppm, εNd(2610 Ma) = -0.3). Based on a mineral from one (A187) of the samples (No Pb/U available, isochron with an age of 2615 ± 57 Ma, a similar εNd Appendix 5). value (+0.4 ± 0.2) has been reported by Zozulya

Recently, zircon grains from the Siilinjärvi et al. (2007). carbonatite were also analysed by LA-MC-ICPMS

11.3 The Sokli carbonatite

The Sokli carbonatite complex is one of the 22 alka- on several methods, a Devonian age was already line complexes that constitute the Devonian Kola established in the 1970s (Vartiainen & Woolley alkaline province (O’Brien & Hyvönen 2015). Based 1974), and a Rb–Sr age of 365 ± 3 Ma was later

Siilinjärvi carbonatite 0.57 data-point error ellipses are 2s

0.55 TIMS Intercepts at 686±170 & 2610 ± 4 Ma

0.53 MSWD = 1.6

U (n=6, A187, A300, A376) 2720

238 2680 Pb/ 0.51 2640 206 2600

0.49 2560 LA-MC-ICPMS Concordia Age = 2608 ± 6 Ma 0.47 (2s, decay-const. errs ignored) MSWD (of concordance) = 0.68, Probability (of concordance) = 0.41

0.45 11.0 11.4 11.8 12.2 12.6 13.0 13.4 13.8 207Pb/235U

Fig. 120. Concordia plot of U–Pb zircon data obtained by ID-TIMS (red error ellipse) and LA-MC-ICPMS from the Siilinjärvi carbonatite.

121 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye reported by Kramm et al. (1993). Our contribution initial εHf value of +8.2 suggesting a primary origin here is to present an unpublished U–Pb analysis from a depleted mantle-type reservoir. We have also on zircon carried out by O. Kouvo (in 1980), which included a U–Pb analysis of pyrochlore (conducted gave a concordant age of 366 ± 3 Ma (Appendix in 1969 by Kouvo), which suggested a U–Pb age of 5). Zircon from Sokli was also used in the Lu-Hf ca. 350 Ma (Appendix 5). studies by Patchett et al. (1981), who reported an

11.4 The 2.0 Ga Laivajoki and Kortejärvi carbonatites

The Laivajoki and Kortejärvi carbonatites are located provides an explanation for the slightly younger in the Koillismaa area between the Archaean Lentua multigrain TIMS result. A closer inspection of the and Pudasjärvi complexes (Nykänen et al. 1997, Fig. ICP-MS data reveals large variation in the U (and 38). The first U–Pb analysis conducted on Laivajoki Pb) concentrations. Three analyses of BSE-darker zircon that was already obtained in 1973 suggested domains yielded less than 8 ppm U (analysis 6a, an age of ca. 2.0 Ga, which was confirmed by sub- Fig. 121) whereas the rest had up to 400 ppm U, but sequent U–Pb TIMS data on the same sample (A497 the ages from both domains are indistinguishable. Laivajoki). The data are slightly heterogeneous and Two other samples from these carbonatites were discordant, providing an average 207Pb/206Pb age collected for dating during the research project on of 1980 Ma (Fig. 121, Appendix 5). Recently, more alkaline magmatism in the 1990s. Zircon from the U–Pb analyses were performed using laser abla- Laivajoki sample A1443 mainly consists of large tion MC-ICP-MS. Excluding one analysis, the data crystals. In BSE images, the mega-grains often yielded a concordant age of 2001 ± 7 Ma, which is show lighter inner domains surrounded by wide considered the best estimate for the igneous age darker outer domains. The U–Pb analyses revealed of this rock (Fig. 121, Appendix 7). One analysis low U in these darker domains, as was also obtained suggested a younger age, which was probably due in sample A497 above. TIMS and SIMS methods to a crack hit by the laser beam. The crack effect were applied, with both producing slightly scattered

A497 Laivajoki carbonatite data-point error ellipses are 2s

0.44

Concordia Age = 2001 7 Ma n=14 (LA-MC-ICPMS) 0.40 U 2100 238 2000 0.36 Pb/ 1900 206

1800 0.32

TIMS average Pb/Pb age 1980 5 Ma 0.28 4.6 5.0 5.4 5.8 6.2 6.6 7.0 7.4 207Pb/235U

Fig. 121. Concordia plot of U–Pb zircon data obtained from the Laivajoki carbonatite A497. LA-MC-ICPMS analyses are presented as error ellipsoids and ID-TIMS analyses as red dots.

122 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Laivajoki (A1443) and Kortejärvi (A1444) carbonatites data-point error ellipses are 2s

0.44

A1443 NORDSIM 2200 0.40 Average Pb/Pb age 1990 ± 10 Ma 2100 (all data) 2000 206 0.36 Pb 1900 238 U 1800 0.32 TIMS data 1700 diamond - A1443 zircon triangle - A1444 zircon 0.28 square - A1444 baddeleyite A1443G monazite TIMS Pb/Pb age 1687 ± 20 Ma 0.24 4 5 6 7 8 207Pb/235U Fig. 122. Concordia plot of U–Pb data obtained from the Laivajoki and Kortejärvi carbonatites. SIMS analyses conducted on zircon A1443 are presented as error ellipsoids and ID-TIMS analyses as dots.

data that were roughly consistent with an age of consistent with the zircon U–Pb data indicating an 2.0 Ga (Fig. 122, Appendix 4a & 5). A small amount age of ca. 2.0 Ga (Fig. 122, Appendix 5). Zircon was of monazite was also found from this sample. A also analysed by SIMS, but due to extremely low multigrain TIMS analysis dated the monazite at ca. concentrations of U and Pb, the errors are very large 1.7 Ga. This supports the assumption that meta- and the data useless (Appendix 4a). morphic effects are the reason for the scatter in the The C and O isotope compositions reported by zircon U–Pb data. Based on six concordant SIMS Nykänen et al. (1997) for the Laivajoki and Kortejärvi analyses on inner zircon domains, an age of 1999 carbonatites are within the range obtained for car- ± 9 Ma can be calculated, which may be considered bonatite melts in equilibrium with mantle miner- as the best estimate for the age of this rock. als (δ13C ~ -4.2, δ18O ~ +7). We have used the same Sample A1444 collected from the Kortejärvi car- samples for Sm–Nd isotope analyses. The REE level bonatite contains zircon, which is bright and clear in these rocks is high (Nd ~350 x chondrites) and up to a gem quality. The U–Pb analyses using TIMS the obtained initial εNd(2000 Ma) values are system- demonstrated that the U concentrations are low, but atically positive, providing an average value of +2.4 the data still tend to be discordant. A small amount (Appendix 1). This suggests that convective depleted of baddeleyite found from this sample was also ana- mantle was the ultimate source for these rocks. lysed by TIMS. These data are also discordant, but

11.5 Lamprophyres

A Palaeoproterozoic U–Pb age of ca. 1.8 Ga was 4 Ma (Woodard et al. 2014). The initial εNd(1784 Ma) obtained for the Kaavi lamprophyres in the early value and TDM for this sample are +0.4 and 1.98 1980s (Huhma 1981). Recently, zircon from the Ga, respectively (Woodard & Huhma 2015, Appendix same sample (A159 Niinivaara, Fig. 74) was re- 1). Near chondritic initial isotope ratios are also analysed by SIMS (NORDSIM facility), yielding evident in other high-REE dykes in the north- good concordant U–Pb data and an age of 1784 ± ern Savo region, such as the Syväri lamprophyres

123 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

(average εNd(1784 Ma) = -0.5, Woodard & Huhma ated mantle material. It may be of interest to note 2015, Appendix 1) and the Panjavaara carbonatite that the Pb isotope ratios obtained for galena from

(average εNd(1800 Ma) = -0.5, unpublished, Torppa the Panjavaara and related Petäiskoski carbon- & Karhu 2007). The roughly coeval dykes of this rock atites plot exactly on the terrestrial evolution line of family in the Svecofennian and NW Ladoga regions Stacey & Kramers (1975) at 1.8 Ga (Tyni et al. 2003). also share this feature. These include the Naantali Two lamprophyres from Central Lapland have and Halpanen carbonatites (average εNd(1800 Ma) = also been analysed by the Sm–Nd method (Appendix -0.1 and -0.3, respectively) and NW Ladoga lam- 1). The Kuotko lamprophyre (A1168), which occurs prophyres (average εNd(1800 Ma) = -0.3, Woodard within the 2.0 Ga Kittilä Group volcanic rocks about & Huhma 2015, Appendix 1). This rock suite is also 45 km NE of Kittilä, just west of the Ruoppapalo characterised by distinctively low δ13C values from granodiorite intrusion (Fig. 2), gave an εNd(2000 Ma)

-12 to -16 (Tyni et al. 2003, Torppa & Karhu 2007, value of -3.1 (TDM = 2.43 Ga). A dyke at Palovaara Woodard & Huhma 2015). The lamprophyres and (A1439), 15 km east of Kittilä, was found to have related 1.8 Ga rocks have been interpreted to origi- a εNd(2000 Ma) value of +2.3 (TDM = 2.07 Ga). The nate from a metasomatically enriched lithospheric Palovaara sample was obtained from a drill core mantle (Eklund et al. 1998, Andersson et al. 2006, penetrating sedimentary rocks of the Sodankylä Woodard et al. 2014). Furthermore, Torppa & Karhu Group close to the contact of the Savukoski Group (2007) considered that the depletion in 13C relative metavolcanics. Another lamprophyre sample to the average mantle value could be related to sub- (A1441) from the same locality has yielded a U–Pb duction of organic-rich crustal material. monazite age of 1771 ± 8 Ma, which was thought to

This might suggest that the chondritic initial εNd record the latest major stage of alteration in Central values are a result of mixing of various components Lapland (Rastas et al. 2001). rather than a signature of the primary unfraction-

12 DISCUSSION

12.1 Episodic rifting stages of the Archaean lithosphere

The U–Pb and Sm–Nd ages that are available on and introduced a totally new discovery of 2.5 Ga mafic rocks in the Fennoscandian Shield manifest rocks. Obviously, the new data are largely based several stages of rifting of the Archaean lithosphere on U–Pb spot analyses on zircon, but the Sm–Nd before its major breakup ca. 2 Ga ago, which is the mineral ages determined for well-preserved rocks age of ophiolites (Kontinen 1987). The results from are generally consistent with the U–Pb zircon ages. the Finnish Karelia province are summarised in There are a few exceptions, such as the Rantavaara Figure 123. Most of the analytical results provide intrusion case, in which the apparent old Sm–Nd geochronological information on the emplacement mineral age is explained by slight crustal contami- of gabbroic rocks in intrusions and dykes, but also nation in the intercumulus plagioclase occurring as constrain the timing of mafic volcanism and the a minor phase in an ultramafic rock. Problems in overall geological evolution. From northern Finland, dating often relate to secondary open systems, but many of these ages were already reported in a spe- can also be due to isotopic disequilibrium between cial volume containing U–Pb isotope data (Vaasjoki phenocrysts and their host rock, as was observed, 2001), and together with new results suggest a for example, in the Suisarian picrites in the Onega range of ages, mostly at ca. 2.44 Ga, 2.22 Ga, 2.15 region, Russian Karelia (Puchtel et al. 1998, Huhma Ga, 2.12 Ga, 2.05 Ga and 2.0 Ga (Hanski et al. 2001a), et al. unpublished), or in some dykes in this work providing cornerstones for chronostratigraphy over (A1465, A1808). the Karelian craton (Fig. 3). An overview of the ages within each area is pre- The isotopic results reported from northern sented in Figures 124a–f, which also give individual and eastern Finland in this work confirmed new sample numbers. localities for ca. 2.3 Ga and 2.44 Ga mafic rocks

124 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Age of mafic rocks in the Karelia province/ Finland Blue - Lapland (n=76) Red - Pudasjärvi & Taivalkoski (n= 56) Green - Kuhmo & Iisalmi (n= 57)

Relative probability Relative

1700 1800 1900 2000 2100 2200 2300 2400 2500 2600 Age (Ma)

Fig. 123. U–Pb and Sm–Nd age determinations for the Palaeoproterozoic mafic rocks in the Karelia province of Finland.

125 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

data-point error symbols are 2s data-point error symbols are 2s

2600 Lapland mafic rocks 2100

2500

PGE A0412

A0818

A1316

A0816

A0863 A0450

A1390

A0820 A0861 A1445 A1226 A0469

A1475 A1317 A0401 A0655 A1318

2000

2400 A1436

A1474 A2288

A0742 Jeesiörova A0741 A2289

A0964 A1204 A2290 A0580

A0666 A1312 A1311 A1563 A0944 Keivitsa NI Keivitsa A0946 A1405

A1272 A1273 A0360 Tshokkoaivii 132

2300 - L

Age (Ma) Vesmajärvi Age (Ma) A1310 Akanvaara A0604 Koitelainen A0604

A1525 Lehtomaa 1900

A1366

2200

A1337 A1431 Tokka A0142 A0281 A0892 A0470 A0825 A0136 A0379 A0916 A1448

A0915

A1700

1800

A1408 A1430

2100

A0900 A0414

A0962 A1916 A0472 A0817 A0841 A1665 A0066 A0077

A0078 A1441 A) A1446 B) 2000 1700 A0959 A1674 Kannusvaara

data-point error symbols are 2s data-point error symbols are 2s 2600 2700 Taivalkoski block, mafic rocks (n=25) Pudasjärvi complex & Peräpohja belt, mafic rock ages (n=31)

2500

A1414

2400

01 A0722 - A0709 A0700 A0919 A0698 A1663 A0713 A1868 A1410 A1412 A1415

A1744 A0603

TTK

A0662 A1466 A0859 A1797 A0703 - A1012

2300

2300 22B A1796 A1492

A1456

A0988

Age (Ma) A0999 A0476

2200 A0408

Age (Ma) A0407

A0305

A0592 A0477 A0304 A0409

A0474 A0706

A0865 A0475

2100

A1795

A1794 A0847 A2087 A1214 A0986 A0496 A0480 A0410 A1788 2100 A1009 A0854 A1743

1900

2000

A1802 A0755 A0907 A0497 A1443 D) 1900 C) 1700

data-point error symbols are 2s data-point error symbols are 2s

2600 2600 Kuhmo block mafic rock ages (n=30) Iisalmi complex mafic rock ages (n=27)

2400 2400 A1596

A1356

A1369

A0135

A1914

A1672

2200

2200 A1333

A1361

A0586

A0199

A0769 A1182 A0198 A0261

A0759 A1220 A1096

A0977 Age (Ma) Age (Ma)

A1363

A1212 A1368 A1223 A0465 A0427 A1231

2000 A0398 2000

A0481 A0671

A0672

A1489 A0242

A1381

A1409 A2071

794B

A1673 R393 A1519 A1529 A0149 A1754 - A1875 A1144 A1029 A1402 A1460 A0729 A0196 A1838

1800 OKU 1800 A1688 A0918

05 05 - - JA A1457 JD A0760 A0159 Kalto 2A Kalto F) 1600 E) 1600

Fig. 124. Ages of mafic rocks in different areas, with two-sigma error bars.

126 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

12.2 Range of εNd – evidence for heterogeneous mantle and crustal contamination

In order to constrain the origin and processes the 2.6 Ga Siilinjärvi carbonatite, 2.0 Ga Kortejärvi involved in the formation of crust, the Sm–Nd and Laivajoki carbonatites (Nykänen et al. 1997), method has been used at GTK since 1981. The avail- 1.97 Ga Jormua OIB dykes and 1.8 Ga Kaavi lampro- able Sm–Nd data comprise ca. 2600 analyses, with phyres (Huhma 1981, Woodard & Huhma 2015), for ca. 1100 of them being from mafic rocks in Finland which Sm–Nd analyses are presented in Appendix (ca. 800 from the Karelia province). The main Sm– 1. They suggest that mantle reservoirs with a nearly Nd results of the mafic-ultramafic rocks and some chondritic Nd isotope composition were the dom- related felsic rocks within the Karelia province are inant sources for the 2.6 Ga carbonatite, Jormua summarised in Table 1 and shown in the εNd vs. age OIB, and 1.8 Ga lamprophyres. Similar results have diagrams (Fig. 125a–e). Most of the initial εNd val- been obtained for 1.8 Ga shoshonitic magmas in the ues for mafic rocks are based on Sm–Nd mineral Svecofennian domain, as well as Russian Karelia isochrons and should thus give reliable estimates (Patchett & Kouvo 1986, Eklund et al. 2000). In con- for the initial isotope composition of the rocks in trast, the 2.0 Ga Kortejärvi and Laivajoki carbon- question. It should be noted, however, that some atites provide clearly positive initial εNd(T) values results are based on only a few analyses of whole- (+2.5) and suggest an origin from a mantle source rock samples. Some points in Figure 125 have a with time-integrated depletion in LREE. These and relatively large error in age, which is shown by the other available data thus indicate that the mantle evolution line following typical compositions of the had long-term heterogeneity. rock units in question. Nevertheless, a wide range in Some features of the chemical composition of the initial Nd isotope compositions is evident. Some the mafic rocks are given in the 147Sm/144Nd vs. Nd rocks were clearly derived from depleted mantle diagrams (Figs. 126, 127), which mirror the level of sources, whereas others show a major contribution LREE enrichment and the quantity of incompat- from old enriched lithosphere. It can be questioned ible elements. In this diagram, one “typical” sam- whether this is due to crustal contamination in the ple from the data set was selected to represent the final magma chamber, at deep crustal levels, or an whole mafic unit. This may obviously result in bias, inherited feature from the heterogeneous subcon- as a range of compositions is present due to crystal tinental lithospheric mantle. fractionation in magma chambers. Nevertheless, High-REE mantle-derived rocks should provide most of the age groups and magma types appear useful information on the isotope composition of to form distinct systematic trends or groupings in the source, since they are less sensitive to crus- the diagram. tal contamination. Such rocks in Finland include

127 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Mafic rocks & Felsic porphyries/ Lapland Mafic rocks/ Taivalkoski block Kapsajoki 5 Keivitsa dyke Veikasenmaa 5 Yräjärvi Kiimaselkä Vesmajärvi 3 Selkäsenvuoma Rantavaara 3 Kortejärvi Laivajoki Karkuvaara Tanhua Kannusvaara Törninkuru 1 Ahvenvaara 1 Puijärvi Haaskalehto CHUR CHUR Silmäsvaara Onkamonlehto Pittarova Satovaara -1 Kannusvaara Tanhua -1 epsilon epsilon Nyssäkoski - - Tshokkoaivi Syöte Koitelainen Akanvaara Porttivaara Nd

Nd Latvajärvi Akanvaara Suoperä (BD) -3 Mäntyvaara -3 Keivitsa Lehtomaa Salla Matarakoski Möykkelmä Rovasvaara Tainio Koutoiva -5 Lotto Moskuvaara -5

A) Keivitsa NI-PGE Sakiamaa B) -7 -7 1800 2000 2200 2400 2600 1800 2000 2200 2400 2600 Age (Ma) Age (Ma)

Mafic rocks/ Pudasjärvi & Peräpohja Mafic rocks/ Kuhmo block

5 5 Tikanmaa Kuusivaara 3 Koppakumpu 3 Horsmanaho

Paukkajavaara Tohmäjärvi Rytijänkä Lohisärkkä 1 1 Koli Hirsimaa Kivikevätti Mustikkarinne CHUR Niinivaara Arola CHUR Runkausvaara Huutokoski -1 epsilon -1 epsilon Peräaho Hattuselkonen - Kemi - Penikat Nd Nd -3 -3

-5 -5 Vengasvaara C) D) -7 -7 1800 2000 2200 2400 2600 1780 1980 2180 2380 2580 Age (Ma) Age (Ma)

Mafic rocks/ Iisalmi complex

3 Siilinjärvi Vuorimäki Nieminen Jormua main suite Siunaussalmi/Tulisaari Humppi 1 CHUR Jormua OIB suite Vuotjärvi Kettukallio Siilinjärvi -1 Syväri Otanmäki epsilon Raatelampi - Junttilanniemi Nd -3 Siilinjärvi Parkkila

-5

E) Lapinlahti -7 1780 1980 2180 2380 2580 Age (Ma) Fig. 125. Epsilon-Nd vs. age diagram for 1.8–2.6 Ga mafic rocks (green – volcanic rocks) and some related felsic lithologies (data sources shown in Table 1 and Appendix 6). Evolution lines are presented if the error in age is in excess of 20 Ma. The trend of the line follows the typical composition of the rocks in question and the length of the line approximates the error in age. Depleted mantle evolution is according to DePaolo (1981). CHUR is the Bulk Earth evolution (De Paolo & Wasserburg 1976). The evolution of typical Neoarchaean granitoid is also shown (A1611, Mutanen & Huhma 2003).

128 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland 3540160 3498750 3296659 YKJ-East 3408843 3507700 3433816 3603128 3617874 3553500 3425956 3587537 3600569 3588191 3508700 3440300 3466640 3502700 3524900 3476945 3498820 3499170 3496250 3501900 3491350 3500925 3387300 3447180 3388066 7597510 7512300 7624845 YKJ-North 7452883 7524300 7550572 7490727 7507142 7462900 7511986 7489163 7382926 7399823 7429990 7499000 7487530 7498900 7494900 7499262 7512500 7511460 7503250 7471250 7522500 7509260 7528761 7534560 7525559 383302 371412 183401 Map 264212 374101 274404 471306 471411 373304 273406 4711 461402 461212 364103 371207 371303 373201 373207 371404 371412 371412 371411 373102 274104 372108 U-Pb age on A360 comments within Tuntsa suite within Tuntsa suite negative eps at any prot age crosscuts Salla Group crosscuts Salla & Kuusamo Groups U-Pb age, cf. A281 (Ref 8) U-Pb age, cf. A281 (Ref 8) plag excluded plag disequilibrium U-Pb age on granophyre A1674 age? crosscuts Savukoski Group age minimum age minimum within Kittilä Group within Kittilä Group crosscuts Savukoski Group A1916 A1366 A1317, A1318, A360 A1318, A1317, samples A1665 A1204, etc A1168 (R308) A1475 A1474 A1312 A1439 4711/06/R25 etc. A1525 A1405 A1431 A1430 A1408 etc. A1337, A900, A1586 A1337, A1446, A1674 A1715 R695/67.65-67.70 A1226, A1316, A1319, A1390, etc. A1436 A820 A2288-9 A2290 A1563 R4/9.40 A1272 x x x x x x x x x x x x x x x x x x x x 7 7 7 15 15 17 4, x Ref* 17, x Sm-Nd 4 5 6 3 1 4 5 6 1 1 4 4 8 1 9 5 1 3 1 1 2 1 2 2 1 n* 22 12 23 ± 44 67 81 54 49 34 30 26 27 35 44 25 33 31 26 1804 1916 2458 1774 2438 2464 2448 2432 2231 2185 2187 2233 2089 2069 2049 Sm-Nd Age (Ma) x x x x x x x x x 1 1 8 8 7 8 8 24 1, x Ref* 12, x U-Pb ± ? 4 3 6 5 3 6 7 5 5 8 8 3 8 11 11 14 (4 U-Pb 2499 1796 2439 2436 2424 2403 2211 2148 2148 2100 2058 2039 2055 2035 2025 2008 1996 c. 2220 c. 2220 Age (Ma) ? 0.7 3.1 0.8 0.0 2.0 0.5 2.8 2.3 5.1 eps -1.8 -5.0 -5.2 -2.0 -5.8 -2.2 -0.6 -3.4 -1.0 -1.5 -0.7 -3.0 -3.5 -5.2 -6.4 -4.2 -0.5 -0.4 -3…-5 ± ? ? ? ? 4 3 6 5 3 6 7 5 5 5 8 8 3 8 11 44 34 30 11 27 35 14 67 ? 2499 1796 1804 2439 1800 2436 1800 2424 2464 2448 2403 2211 2148 2231 2185 2148 2100 2058 2039 2058 2055 2035 2025 2008 2000 1996 1916 Age (Ma) (1 L L L L L L L L L L L L L L L L L L L L L L L L L L L L D results on mafic-ultramafic rocks in the Karelia province Finland. (dyke trend) mafic intrusion mafic intrusion (appinite) mafic dyke Rock type mafic intrusion lamprophyre dyke mafic intrusion lamprophyre dyke mafic intrusion mafic intrusion mafic intrusion mafic-ultramafic intrusion boninite-norite dyke (NE) “Karjalite/ GWA” mafic intrusion “Karjalite/ GWA” “Karjalite/ GWA” mafic intrusion mafic intrusion mafic intrusion mafic intrusion mineralization mafic intrusion mafic intrusion mafic intrusion mafic “sheeted” dyke mafic dyke (calc-alkaline) mafic intrusion olivine gabbro dyke Lapland: Tshohkkoaivi Tainio Lotto, Inari Table 1. Summary of the Sm-Nd Name/ Location Koitelainen Kuotko Akanvaara Palovaara Kittilä Lehtomaa Salla Koulumaoiva Peuratunturi Värriö Onkamonlehto Haaskalehto Rantavaara Ahvenvaara Silmäsvaara Tanhua Kannusvaara Väkkärävaara Keivitsa Moskuvaara Keivitsa NI-PGE Rovasvaara Puijärvi Satovaara Selkäsenvuoma Nuttio Pittarova Keivitsa dyke

129 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye YKJ-East 3548780 3679957 3679970 3527510 3679953 3679852 3558600 3650896 3573079 3548690 3550074 3550107 3534040 3585084 3574957 3544886 3559318 3535150 3529500 YKJ-North 7299240 7346042 7346124 7278908 7346133 7346039 7276750 7304580 7268263 7270390 7263323 7263247 7240360 7285883 7268724 7267231 7263963 7320400 7310700 Map 4633 4633 353209 4633 4633 3534 4541 451205 353403 353405 353405 353112 452107 451205 353402 353405 354211 354207 comments opx phenocr. disequilibrium analysis of light plag excluded heterogeneous? heterogeneous? heterogeneous? 59-W-73 etc. samples A1412 (35-VEN-94) A1414 (38-VEN-94) B7-11.56 etc. A1492 (39-VEN-94) A1465 (42-VEN-94) A1466 (WD 14) A1415 (122-VEN-94) A1415 A1796 (2-JIV-03) A1797 (AD13-8) A1795 (AD89-8) A1794 (AD89-10) A1456, A988 A1798 (AD85-3) A1800 (AD91-3) A1802 (MLJ29-1) A1801 (MLJ40-1) A497, A1443,… A1444, … x x x x x x x x x x x x x x x x x x 26 Ref* Sm-Nd 4 3 5 3 2 3 3 3 4 4 4 6 3 3 3 3 4 4 n* 23 ± 75 32 35 24 35 29 62 53 42 37 27 35 2388 2421 2476 2349 2407 2420 2352 2404 2233 2219 2319 2058 Sm-Nd Age (Ma) x x x x 25 25 28 Ref* U-Pb ± 5 5 8 7 9 10 18 (4 U-Pb 2436 2436 2447 2339 2306 2001 1999 Age (Ma) 1.7 1.0 1.2 1.6 1.8 0.5 1.0 0.2 2.4 2.6 eps -2.1 -1.7 -2.2 -1.4 -2.4 -0.1 ± 5 5 6 8 7 9 ?? 32 35 24 18 35 62 42 37 35 ? ? ? 2436 2436 2440 2421 2476 2446 2349 2339 2407 2306 2352 2233 2219 2001 2058 1999 Age (Ma) (1 T T T T T T T T T T T T T T T T T D (dyke trend) mafic intrusion Rock type mafic intrusion opx-phyric dyke gabbronorite dyke (NNW) Fe-tholeiitic dyke (NW) gabbronorite dyke (NNW) Fe-tholeiitic dyke (W) Fe-tholeiitic dyke (W) Fe-tholeiitic dyke (W) mafic intrusion Fe-tholeiitic dyke (NE) Fe-tholeiitic dyke (NE) tholeiitic dyke (WNW) Fe-tholeiitic dyke (NW) Fe-tholeiitic dyke (NE) Fe-tholeiitic dyke (NW) carbonatite Fe-tholeiitic dyke (NW) carbonatite Taivalkoski block (northern part of the Lentua complex): Porttivaara Name/ Location Syöte Pääjärvi Pääjärvi (XD1) Pääjärvi (XD3) Suoperä (BD) Pääjärvi (XD4) Törninkuru Taivalkoski Karkuvaara/ Nyrhinoja Kallioniemi Kontioluoma Kuusamo Tilsanvaara 2 Tilsanvaara Murhiniemi Hirsikangas Laivajoki Koivuvaara Kortejärvi Table 1. Cont.

130 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland 3417703 3545060 3428578 3472600 3502900 3418244 3395397 YKJ-East 3544630 3412956 3521875 3466147 3468985 3406326 3362620 3554020 3537250 3566160 3527140 3530010 3521560 3514100 3544300 3554500 3531800 3525100 3540380 3537990 3508490 7318622 7137220 7329965 7259000 7288380 7314041 7301076 YKJ-North 7139070 7366297 7030850 7303000 7263183 7353659 7354715 7022440 7000140 7008800 7140340 7179750 7025690 7158040 7045360 7103000 7194050 7184580 7030820 7056220 7111630 254404 343401 2544 351404 354110 2544 254111 Map 3434 263303 333209 352303 351404 263302 261311 333112 333208 343206 334301 333403 334111 343102 at least 17 analyses plag disequilibrium SHRIMP Nd disequlibrium feeder dyke for Penikat? comments at least 7 analyses within Petäjäskoski Fm “mixing line age” ca. 2.6 Ga? within Tikanmaa Fm data from felsic rocks A1012, A603, A703 A1744 A1410 LD-4-93 A662 samples A2087 A1688, A708 A1808 A1743 A1214 A854 JD-05 A300, A71, A187, A376 JA-05 A1595-6 A1373 A135 A1362 A1369 A199 A651 A759 A1223, A1368 A1875 A1381 x x x x x x x x x x x x x x x x x x x x x 16 16 21 21 Ref* 13, x 13, x 14, x Sm-Nd 5 5 3 5 1 1 7 1 2 4 3 1 1 6 3 1 2 1 3 3 3 1 1 1 3 9 n* 17 12 ± 52 72 58 34 38 42 40 42 38 19 43 190 160 2422 2196 2507 2447 1968 2077 2270 1930 2350 2127 1968 2043 1936 Sm-Nd Age (Ma) x x x x x x x x x x x x x 2 3 2 6 9 13 23 23 13 5,x 5,x Ref* U-Pb ± 4 8 4 5 5 5 5 4 4 4 5 7 7 5 11 20 11 13 15 28? (4 U-Pb 2433 2444 2140 2210 2444 2378 1953 2130 1888 2084 1784 2610 1790 2444 2323 2295 2222 2322 2121 2058 1953 (min) Age (Ma) 3.5 0.2 0.4 3.2 1.0 0.0 1.1 1.3 1.8 1.0 2.5 2.0 0.5 eps -1.6 -1.6 -5.4 -1.2 -0.1 -6.5 -0.6 -0.3 -1.8 -0.6 -1.0 -2.7 -0.4 -0.8 ± ? ? 4 8 4 5 5 5 5 4 4 4 5 7 5 7 5 ?? ?? ?? 11 20 11 34 13 13 38 ? 2433 2444 2140 2210 2444 2433 2433 1953 2130 1888 2084 1784 2610 2077 1790 2444 2440 2323 2295 2222 2300 2200 2202 2121 2058 1953 1968 Age (Ma) (1 I I I I I I I I I I I I I I I P P P P P P P P P P P D (dyke trend) mafic intrusion mafic intrusion Rock type OIB-suite dykes mafic dyke “Karjalite/ GWA” mafic intrusion tholeiitc dyke (NW) boninite-norite dyke (WNW) mafic intrusion mafic dyke lamprophyre carbonatite mafic dyke (NE) mafic intrusion gabbro dyke (NW) lamprophyre mafic intrusion mafic intrusion mafic dyke (NW) mafic dyke (W) mafic intrusion mafic dyke mafic intrusion mafic intrusion tholeiitic dyke (W) mafic intrusion basalts (main suite) mafic dyke (NW) Kemi Pudasjärvi complex: Penikat Name/ Location Jormua OIB suite Kuusivaara/ Peräpohja belt Runkausvaara Vengasvaara Uolevinlehto (UD) Loljunmaa (LD) Lapinlahti Koppakumpu/ Peräpohja belt Syväri Nilsiä Iisalmi complex: Siilinjärvi Tervonkangas Ranua Rytijänkä Palomaa Vuotjärvi Nilsiä Junttilanniemi Paha Kapustasuo Humppi Siunaussalmi/ Tulisaari Raatelampi Petäiskangas Honkaniemi Kettukallio Nieminen Otanmäki Jormua main suite Koirakoski Table 1. Cont.

131 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye 3611880 3646329 3632425 3655111 3611623 3632160 3610031 YKJ-East 3655282 3643657 3673371 3650177 3598495 3564650 3730177 3593000 3594966 3614763 3622792 7167213 7189752 7187531 7058711 7168402 7164598 7218527 YKJ-North 6984529 7126840 6904818 6995436 7150240 7104520 6989912 7026000 6989664 6970714 6981251 4423 4441 4423 434102 4423 442307 Map 433101 441312 431311 4412 531101 4312 431107 NW-trend. comb. Sm-Nd: 2010±26 (MSWD=0.6) 2106±40 (MSWD=1.9) E-trend. comb., Sm-Nd: comments open systems? plag excluded A1489B A1356 A1409 A2071 A1489C A1212 A1914 samples A1144 ROM A398 A1220, A1221 A198 A1519 OKU-1-J2-2, … A159 OKU-602,… A1029 x x x x x x x x x x 7 7 22 10 30 21 Ref* 11, x 11, x Sm-Nd 4 5 3 1 3 5 1 4 3 5 1 1 7 1 4 2 n* 12 10 ± 40 46 33 40 29 41 58 75 2005 2409 2014 2054 2133 1967 2201 2125 Sm-Nd Age (Ma) x x x x x 7 7 22 10 23 11 19 Ref* 11, x U-Pb ± 9 4 8 8 4 5 21 10 13 20 20 18 (4 U-Pb 1989 2321 2395 1980 2210 1973 1963 2103 2210 2210 1784 1971 1959 (min) Age (Ma) 1.5 0.3 0.3 0.7 1.4 1.6 0.3 0.7 1.0 0.6 0.4 2.5 0.5 eps -0.7 -1.2 -0.6 -0.3 ± ? ? 9 4 8 8 4 5 21 40 10 13 40 29 20 20 18 ? 1989 2321 2440 1980 2005 2210 1973 1963 2103 2054 2133 2210 2210 1784 1971 1800 1959 Age (Ma) (1 K K K K K K D K K K K K K K K K K K (dyke trend) mafic dyke mafic dyke (WNW) boninite-norite dyke (NE) Rock type boninitic & Fe-tholeiitic dykes Fe-thol dyke (NW) Fe-thol dyke (NW) mafic intrusion mafic dyke mafic dyke tholeiitic dyke (NW) Fe-thol dyke (E tr) Fe-thol dyke (E) “Karjalite/ GWA” “Karjalite/ GWA” lamprophyre mafic intrusion carbonatite mafic intrusion Hattuselkonen Lohisärkkä Kuhmo block (in the Lentua complex): Viianki (VD) Name/ Location Romuvaara Kivikevätti (KD) Veitsivaara Oravaara Tohmäjärvi Mustikkarinne Peräaho Ilomantsi Paukkajavaara Veitsivaara Kapea-aho Koli Arola Niinivaara Kaavi Outokumpu/ Horsmanaho Panjavaara Outokumpu/ Huutokoski Table 1. Cont.

132 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland 3473150 YKJ-East 3586268 3607695 3538540 3526517 3456600 3395021 3417837 3605396 3434872 3418695 3428578 3428578 3605716 3434600 3473360 3405496 3386699 3439534 3443041 3670161 3528520 3526520 7510500 YKJ-North 7427294 7340634 7229140 7140447 7491900 7354702 7343145 7341805 7512831 7346342 7329965 7329965 7364349 7511017 7502820 7528198 7519277 7520904 7507727 6930826 6996280 6997430 371406 Map 462112 452406 353110 371210 263108 461304 273409 263304 254408 254408 461306 273409 371405 274110 273206 372107 371208 4241 333108 333108 comments younger than A1868 (2428±4 Ma) older than A1214 (2130+-5 Ma) age inferred from A2087 (“Greenstone Fm II”, Kuusamo) range of epsilon (“Greenstone Fm III”, Kuusamo) range of epsilon open systems age, cf. Jeesiörova cf. felsic porphyry age inferred from Vesmaj. age inferred from A481 age inferred from A481 A1435 samples R1/79/ 4.4m A654 A969 A1009 etc. 10J-HAH-80 A1788 A985 etc. 11B-HAH-80 6621 (85/952-4) 800.56-LVP86 A972 2B x x x x x x x x x x 17 17 18 16 16 18 18 17 17 27 20 20 Ref* 16, x Sm-Nd 1 8 5 3 4 6 2 2 1 4 2 6 1 6 2 n* 16 12 14 12 11 20 12 15 ± 50 25 41 61 2105 2056 1982 2089 Sm-Nd Age (Ma) x x 29 12 Ref* U-Pb ± 7 5 (4 U-Pb 2106 2017 Age (Ma) 4.1 0.9 2.8 3.9 0.0 0.6 2.1 3.9 3.7 1.1 3.8 2.2 eps -3.7 3±1 -2.4 -2.7 -2.8 -2.1 -1.9 -1.0 -1.2 -1.7 -2.6 ± ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? 7 5 5 5 25 25 2400 2400 2400 2400 2400 2250 2130 2106 2250 2140 2300 2300 2250 2056 2056 2060 2060 2060 2017 2017 2100 2062 2062 Age (Ma) (1 I I I L L L L L L L L L L T T T T P P P P P K D (dyke trend) andesites Rock type komatiites andesites andesites basalts basalts basalts tuffites tuffites basalts komatiites komatiites basalts basalts (other flows) basalts (1. flow) komatiites picrites basalts, variolitic lava basalts basalts basalts basalts basalts Volcanic rocks Möykkelmä/ Kuusamo Gr Name/ Location Mäntyvaara Fm/ Kuusamo Gr Kuntijärvi Fm/ Kuusamo Gr Matinvaara Fm/ Kurkikylä Gr Varisniemi Fm Honkavaara Fm/ Sodankylä Gr Petäjävaara Fm/ Sodankylä Gr Tikanmaa Fm/ Kivalo Gr Hirsimaa Fm/ Kivalo Gr Ruukinvaara Fm/ Sodankylä Gr Jeesiörova Sattasvaara Fm Jeesiörova Sattasvaara Fm Jouttiaapa Fm/ Kivalo Gr Runkaus Fm/ Kivalo Gr Runkaus Fm/ Kivalo Gr Sattasvaara Fm/ Savukoski Gr Peuramaa/ Savukoski Gr Linkupalo Fm/ Savukoski Gr Vesmajärvi/ Kittilä Suite Kautoselkä/ Kittilä Suite Siilinjärvi Parkkila Kiihtelysvaara Hyypiä Siilinjärvi Vuorimäki Table 1. Cont.

133 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye 17) Hanski & Huhma (2005) 18) Hanski et al. (2001b) 19) Peltonen et al. 2008 20) Lahtinen et al. 2015b 21) Woodard & Huhma 2015 22) Mikkola et al. 2013 23) Woodard et al. (2014) 24) Väänänen (2004) 25) Alapieti (1982) 26) Karinen (2010) 27) Huhma (1987) 28) Salminen et al. 2014 29) Karhu et al. 2007 30) Torppa & Karhu 2007 3) Kyläkoski et al. 2012 4) Manninen & Huhma (2001) 5) Paavola (1988) 6) Hölttä et al. (2000) 7) Hanski et al. (2010) 8) Räsänen & Huhma (2001) 9) Talvitie & Paarma (1980) 10) Vuollo et al. (1992) 11) Huhma (1986) 12) Rastas et al. (2001) 13) Peltonen et al. (1996, 1998) 14) Toivola et al. (1991) 15) Hanski et al. (2001c) 16) Huhma et al. (1990) : U-Pb age in Ma, (min)= minimum (4 : Domain, L= Lapland, P=Pudasjärvi (1 Table 1. Cont. D age deduced from discordant analyses n*: number of Sm-Nd analyses Ref*: Reference for isotope data: x) This volume 1) Mutanen & Huhma (2001) 2) Perttunen & Vaasjoki (2001) complex, T=Taivalkoski block in Lentua complex (& Pääjärvi), K= Kuhmo block in Lentua complex, I=Iisalmi complex Fm=Formation, Gr= Group Age (Ma) eps Age (Ma)

134 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Mafic intrusions

0.22 NMORB Oravaara 2.4-2.5 Ga Ahvenvaara 2.3-2.4 Ga Mantle Silmäsvaara Rytijänkä 0.20 primitive Tanhua Kannusvaara 2.22 Ga Koppakumpu Tervola Koirakoski Paukkajavaara Jormua Eastern block Kontioluoma 2.1-2.15 Ga Kuusivaara Nieminen Veitsivaara Rantavaara Kapea-aho 1.96-2.06 Ga 0.18 Satovaara

Nd Keivitsa Karkuvaara Otanmäki Koli Selkäsenvuoma Haaskalehto Humppi db 144 Arola Peräaho Tilsanvaara Siunaussalmi Hattuselkonen 0.16 Törninkuru Pääjärvi 38-VEN Tilsanvaara Raatelampi Taivalkoski Crust, Lower EMORB Hirsikangas Moskuvaara KemiLohisärkkä Kallioniemi Pääjärvi 39-VEN

Sm/ Murhiniemi Petäiskangas Koivuvaara Värriö Uolevinlehto Mustikkarinne Kivikevätti Runkausvaara Palomaa Veitsivaara 147 0.14 Keivitsa Ni PGE Pittarova Kittilä Koulumaoiva Loljunmaa Rovasvaara Onkamonlehto Syöte Viianki Puijärvi Pääjärvi 35-VEN OIB Penikat Peuratunturi Kettukallio Porttivaara Akanvaara Lehtomaa Salla 0.12 Vengasvaara Crust, bulk Koitelainen Suoperä Pääjärvi 42-VEN Tsohkoaivi Crust, Middle Paha Kapustasuo A) Crust, Upper Nuttio 0.10 0 10 20 30 40 Nd (ppm)

Mafic intrusions

NMORB 0.22 1.6 2.4-2.5 Ga 0.8 2.3-2.4 Ga Mantle 0.0 1.0 0.20 primitive 1.7 ca. 2.22 Ga 3.2 1.2 2.1 3.5 2.1-2.15 Ga 3.2 2.5 0.18 1.96-2.06 Ga -0.5 1.7 Nd -0.9 -3.5 0.9 144 2.8 0.4 EMORB 0.9 -0.6 1.2 0.4 0.9 1.5 -0.7 0.16 1.0 -1.0 1.4 Sm/ Crust, Lower 0.5 0.3 0.8 1.6 1.5 0.4- 0.7 -5.1 147 -2.5 -1.6 1.0 0.4 0.6 -0.1 0.7 -0.3 -0.1 1.0 0.2 0.2 0.14 0.5 -6.3 -1.2 -4.3 -0.4 -1.2 0.3 -1.0 -1.2 OIB -2.1 -1.3 -0.4 -0.7 -1.3 -1.6 -2.4 -3.4 0.12 -5.3 -1.8 Crust, bulk -1.8 -2.4 Crust, Middle -0.6 -2.2 B) Crust, Upper 2.3 0.10 0 10 20 30 40 Nd (ppm)

126. 147Sm/144Nd vs. Nd (ppm) diagram for mafic intrusive rocks divided into different age groups, with Figure B giving the initial εNd values for intrusions named in Figure A. One representative whole-rock analysis was selected from each intrusive unit. The age grouping for few samples is not based on absolute dating but is inferred (Värriö, Pääjärvi 42-VEN, Kontioluoma, Murhiniemi, Hirsikangas). Global reference compositions are from Rudnick & Gao (2003) and Klein (2003).

135 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Mafic volcanic rocks

Jouttiaapa Fm, lower part 0.30 Text color - initial epsilon: red >+1.5, blue <-1.5 0.28

0.26 Jeesiörova komatiite cumulates ca. 2.4 Ga

2.2-2.3 Ga? 0.24 Jeesiörova komatiites

Nd 2.1-2.15 Ga 0.22 NMORB Jouttiaapa Fm, upper part 144 1.98-2.06 Ga Island arc tholeiite Vesmajärvi 0.20 Petäjävaara Fm Tikanmaa Fm Mantle primitive Pechenga tholeiite Sm/ 0.18 Siilinjärvi, Vuorimäki

147 Runkaus Fm, 1. flow Ruukinvaara Fm 0.16 Crust, Lower EMORB Peuramaa Siilinjärvi, lower Parkkila Honkavaara Fm Linkupalo Fm 0.14 Runkaus Fm BCR-1 Köngäs, Vesmajärvi Island arc calc-alk basalt Hyypiä Pechenga Fe-picrite Hirsimaa Fm Mäntyvaara Fm OIB 0.12 Matinvaara Fm Crust, bulk Kautoselkä Fm. Möykkelmä Crust, Middle Siilinjärvi, Vehkasuo Kuntijärvi Fm Crust, Upper 0.10 0 10 20 30 40 50 60 70 80 Nd (ppm) Fig. 127. 147Sm/144Nd vs. Nd (ppm) diagram for mafic-ultramafic volcanic rocks. One representative whole-rock analysis is selected from each rock formation. Global reference compositions are from Rudnick & Gao (2003) and Klein (2003).

12.3 The 2.44–2.50 Ga intrusions and dykes

Since the pioneering U–Pb dating studies by Kouvo the younger one at ca. 2.44–2.45 Ga. The younger (1977), the large 2.44 Ga mafic layered intrusions in one is widespread, being found as intrusions and northern Fennoscandia have been targets for sev- lavas in different parts of the Archaean basement eral isotopic studies (e.g., Alapieti 1982, Mutanen & of the shield. In contrast, the older event is only Huhma 2001, Hanski et al. 2001c). These intrusions represented by three mafic–ultramafic layered manifest the first major Palaeoproterozoic mafic intrusions (Monchegorsk, Fedorovo-Pansky, Mt. magmatism and have generally been considered Generalskaya) dated by Amelin et al. (1995), occur- a sign of break-up of an Archaean (super)conti- ring in the Kola Peninsula, i.e. in the Kola craton, nent (Buchan et al. 2000, Mertanen et al. 1999). with no examples found in the Karelian craton. The Our studies have brought new occurrences to this isotopic data published by Amelin et al. (1995) were family, i.e., the mafic intrusions of Tshokkoaivi in generated in a highly respected geochronological NW Lapland (ca. 2.5 Ga), Lehtomaa, Peuratunturi laboratory (the Royal Ontario Museum, University and Koulumaoiva in eastern Lapland, Vengasvaara of Toronto), and there appear to be no reasons to in the Pudasjärvi block and Junttilanniemi in the doubt their validity. However, there is a problem Kainuu schist belt, and several dykes cutting the with the geochronological data obtained from the Archaean crust. three mentioned intrusions, because other studies In their review of the published geochrono- have produced variable dates from a single intru- logical data on the 2.44–2.50 Ga magmatism in sion, ranging down to ca. 2.44 Ga (see Fig. 3.10 in the Fennoscandian Shield, Hanski and Melezhik Hanski & Melezhik 2012). Regarding the dating (2012) critically evaluated the notion of two major results of Amelin et al. (1995) as trustworthy ages apparently distinct magmatic phases, with the for the three Kola craton intrusions, the establish- older one having taken place at ca. 2.50 Ga and ment of the presence of two age populations has

136 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland implications for the potential correlation of early intrusions in Russia (e.g., Amelin & Semenov 1996, Palaeoproterozoic magmatic and related ore-for- Balashov et al. 1993), and many felsic lithologies mation events between different cratons (Karelia, related to the layered intrusions also share this Kola, Superior, Hearne, Wyoming). Within the feature (e.g., Lauri et al. 2006). Moreover, roughly Fennoscandian Shield, the Kola and Karelian cratons coeval mafic and ultramafic metavolcanic rocks, are separated by the Belomorian mobile belt and the for example, the Mäntyvaara Formation (Kuusamo Lapland granulite belt, and they are thought to have Group) in the Salla area, eastern Lapland, provide collided together during the Kola-Lapland orog- initial εNd values at 2.44 Ga close to –2 (Hanski eny at ca. 1.91–1.93 Ga (Daly et al. 2001, Lahtinen & Huhma 2005; Figs. 125, 127). The age of the et al. 2005). Thus, the cratons potentially had dif- Mäntyvaara Formation is constrained by the cutting ferent histories before that tectonic event, and the Onkamonlehto dyke (A1405), providing a minimum magmatism in the early Palaeoproterozoic might age of 2.4 Ga for the volcanic rocks. These LREE- have been different, consistent with the appar- enriched komatiites resemble the komatiites from ent lack of ca. 2.50 Ga intrusions in the Karelian the Vetreny belt (Puchtel et al. 1997). All these mafic craton. However, the situation is not as simple, as rocks of the “2.44 Ga” group with negative initial this work has shown that the Karelian craton was εNd values are characterised by significantly LREE- magmatically not entirely dormant at ca. 2.50 Ga. enriched chondrite-normalised REE patterns (Figs. The new isotopic data for the Tshokkoaivi layered 126, 127). intrusion in NW Lapland indicate that its emplace- Based on their field relationships and geochemi- ment took place at 2499 ± 11 Ma. Also, some felsic cal and isotopic characteristics, the 2.4–2.5 Ga dyke rocks (Rookkiaapa Formation) in Central Lapland swarms have been divided into five groups by Vuollo have an age close to 2.50 Ga. & Huhma (2005): (1) NE–SW-trending boninite- Rocks of the 2.44–2.50 Ga group are gener- noritic dykes, (2) NW–SE-trending gabbro-norite ally characterised by negative initial εNd values of dykes, (3) NW–SE-trending tholeiitic dykes, (4) NW– –1 to –2 (Figs. 125, 126b). Similar negative initial SE- and E–W-trending Fe-tholeiitic dykes, and (5)

εNd values have been obtained for 2.4–2.5 Ga mafic E–W-trending orthopyroxene-plagioclase-phyric

0.514

"2.22 Ga intrusions" Age = 2223 ± 28 Ma

0.513 eps = +0.6 MSWD = 3.0 n=21

144 0.512 Keivitsa intrusion Age = 2053 ± 26 Ma Nd/ eps = -3.3 143 MSWD = 7.2 n=27

0.511 "2.44 Ga intrusions" Age = 2446 ± 31 Ma eps = -1.9 MSWD = 7.7 n=50 0.510 0.04 0.08 0.12 0.16 0.20 0.24 0.28 147Sm/144Nd

Fig. 128. Sm–Nd isochron diagram illustrating the limited variation in the initial ratio within each rock association. Data sources are presented in Table 1.

137 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye dykes. Groups 1 and 2 represent a (boninitic) magma been observed in the world’s largest layered intru- type with negative initial εNd(T) values, consistent sion, the Bushveld Complex, where Sm–Nd data with the results obtained for most layered intru- suggest that crustal contamination in the upper sions. These dykes may be coeval with the 2.44 Ga part of the intrusion was slightly higher than in intrusions and relate to boninite-like Cr-rich (group the lower part (Maier et al. 2000). 1) and Cr-poor (group 2) parental magmas. The According to Mutanen (1997), several lines of pet- data on groups 3 (A1410 Uolevinlehto) and 4 (A1414 rographic, geochemical and mineralogical evidence Pääjärvi, A1466 Taivalkoski) show slightly posi- suggest that crustal contamination at the final site tive initial εNd(T) values and are related to tholei- of emplacement has been common in large layered itic parental magmas. The chemical distinction is intrusions. Provided that during emplacement the also evident in the 147Sm/144Nd vs. Nd diagram (Fig. magma was already contaminated by crustal mate- 126), as the Fe-tholeiites tend to have higher Nd rial at depth, it is conceivable that the further in situ and Sm/Nd ratios. These Fe-tholeiitic dykes appear contamination processes may be minor in terms of to share isotopic signatures with slightly younger isotope diagnostics. 2.3 Ga mafic rocks. Based on the combined Nd and Os isotopic sys-

The low εNd values suggest that the REE were tematics of the Koitelainen and Akanvaara intru- largely derived from lithospheric sources that were sions, Hanski et al. (2001c) suggested that extensive enriched in LREE in late Archaean times. Debate crustal contamination took place deep in the crust, has surrounded whether this is a feature inherited while in situ contamination after the final emplace- from the subcontinental lithosphere or is due to ment of the magma was relatively insignificant. The crustal contamination deep in the crust or in the fact that the Os isotope compositions remained final magma chamber. In spite of a few exceptions, almost immune to crustal contamination, whereas the available Sm–Nd data on the layered intrusions the Nd isotope compositions were strongly affected, suggest that the variation in the initial Nd isotope led Hanski et al. (2001c) to conclude that the pri- composition is very limited over the shield and mary magma producing the layered intrusions was within intrusions (Huhma et al. 1990, Tolstikhin et a high-Os, low-Nd magma, potentially a komati- al. 1992, Balashov et al. 1993, Amelin and Semenov ite or komatiitic basalt. They suggested that this 1996, Hanski et al. 2001c, Hanski 2012, this study). magma evolved through ACF processes in a deep- This is shown in the isochron diagram (Fig. 128), in crustal magma chamber into a contaminated, low- which 50 Sm–Nd analyses conducted on all intru- Ti basaltic composition, and it was later emplaced sions in Finland give an “age” of 2446 ± 31 Ma into upper crustal Archaean gneisses and overly- and an εNd value of –1.9. The MSWD of 7.7 sug- ing, broadly coeval volcanic rocks. This hypothesis gests scatter in excess of analytical error, which is is supported by the Nd–Os isotopic systematics mainly due to metamorphic effects experienced by of the contemporaneous, crustally contaminated the Koitelainen and Akanvaara intrusions (Hanski komatiitic basalts in the Onkamo Group (Hanski et al. 2001c), not due to variation between intru- & Huhma 2005) and in the Vetreny belt, Russian sions. A possible exception is provided by the large Karelia (Puchtel et al. 2001). It is generally thought

Burakovka Complex in Russia, where εNd(T) in the that the formation of 2.45 Ga rocks was related to lower zone appears to be marginally higher (from upwelling of a mantle plume, which was related to –1 to 0) than in the upper zone (from -2 to -1, the fracturing and rifting of a supercontinent. Amelin & Semenov 1996). A similar trend has also

12.4 The 2.3 Ga mafic rocks

The reduction in magmatic activity on Earth begin- ing ages between 2295 ± 5 Ma and 2339 ± 18 Ma. In ning about 2.45 Ga and lasting for 200–250 Ma has addition to these, Sm–Nd mineral data from two been emphasised by Condie et al. (2009) and is also other sites suggest emplacement ages within this evident in the database obtained from Finland (e.g., range. All samples in this age group are located in Huhma et al. 2011). However, several new occur- eastern Finland, in the Lentua and Iisalmi com- rences of ca. 2.3 Ga mafic intrusions and dykes have plexes. NW–SE-trending ca. 2.31 Ga Fe tholeiitic been recognised during recent studies. Currently, dykes have also been identified in Russian Karelia U–Pb results are available from six localities yield- (Stepanova et al. 2014a). The Sm–Nd results for

138 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland all these rocks provide clearly positive initial Nd cal composition (Vuollo & Fedotov 2005). The dated epsilon values, suggesting that the mafic magma 2.3 Ga mafic rocks are mainly dykes and do not had an origin from a depleted mantle source with- form sills in the Jatulian metasediments, suggesting out major communication with Archaean LREE- that the Jatulian was deposited after a strong post- enriched material (Figs. 125, 126). These rocks are 2.3 Ga erosional stage of the Karelian craton. The bulk thus very distinct from the majority of the 2.44 Ga of the Sm–Nd data available on the 2.22–2.44 Ga magmatism. The dykes and intrusions of the 2.3– mafic volcanic rocks do not yield clearly positive 2.4 Ga group have relatively high levels of REE with initial Nd epsilon values (Table 1). This suggests a moderate enrichment in LREE (Fig. 126), and that these volcanic rocks are not related to the resemble continental tholeiitic basalts in chemi- 2.3 Ga high-epsilon dykes.

12.5 The 2.22 Ga intrusions

The differentiated mafic intrusions of the ca. 2.22 Ga Central Finnish Lapland to northern Norway as the gabbro-wehrlite association (GWA) are widespread Kautokeino and Karasjok belts. in eastern and northern Finland, and often located A special feature of the 2.22 Ga magmatism is near the Archaean-Palaeoproterozoic unconform- that it occurs overwhelmingly as sill-like bodies ity, intruding conformably into Jatulian quartzites. near the Archaean-Palaeoproterozoic unconformity; In Central Lapland, they occur as sills in quartz- there are no known volcanic equivalents, and steeply ites of the Sodankylä Group (Fig. 3). Because of the dipping dykes cutting the Archaean basement are extensive occurrence of these sills in the Karelian very rare. One of the confirmed dykes belonging to formations and the ease of discovering zircon in this magmatic event is found within the Archaean their gabbroic differentiates, this igneous phase has Kuhmo greenstone belt. It occurs as a mostly ultra- become the most frequently dated episode of mafic mafic, strongly flow-differentiated dyke, penetrated magmatism in Finland, as summarised by Hanski by drilling at Petäjäniemi, with its strike being et al. (2001a). Zircon in these rocks often displays approximately north–south, i.e. the same as that a turbid colour and porous texture due to a high of the greenstone belt (Hanski 1984). Differentiated degree of metamictisation, resulting in a strong U– sills, similar to those of the 2.22 Ga magmatism Pb discordance, but spot analyses on pristine zir- elsewhere, have also been recognised within the con domains have still yielded concordant ages of Kuhmo belt (Hanski et al. 2010). For example, in the 2210–2220 Ma (Hanski et al. 2010). We have used Ensilä area, they trend approximately perpendicu- laser ablation MC-ICP-MS to analyse some previ- larly to the narrow greenstone, but are not found ously studied samples and obtained similar results. in the granitoids outside the belt (Hanski 1984, Despite common albitisation and alteration, the Tulenheimo 1999). This suggests that the green- Sm–Nd data from rocks of the GWA association form stone belt has faulted contacts to the surrounding a coherent group yielding an initial εNd(2220 Ma) granitoids in this area and the Palaeoproterozoic value of ca. +0.6 (Fig. 128, Fig. 126; Hanski et al. intrusions were preserved in a graben-like structure 2010). This composition was also obtained for a within the belt due to its post-2.22 Ga downward granophyre in the Koli intrusion. As the roof rocks displacement. The same structure may have also at Koli consist of Archaean gneisses, in situ con- preserved some potentially Jatulian-age quartzites tamination was not important in the genesis of the within the belt (at Hietaperä), also known to be spa- granophyre or the whole-rock association. tially associated with 2.22 Ga sills. In summary, as Given the widespread regional distribution of opposed to the several mafic dyke swarms of the the 2.22 Ga sills in Finland, Hanski et al. (2001a) basement, the emplacement of the 2.22 Ga intru- regarded it as surprising that the existence of simi- sions appears to be unrelated to fracturing of the lar intrusive bodies has not been reported from the Archaean crust due to shield-wide tensional forces. Russian part of the Fennoscandian Shield. Recently, Instead, they were emplaced as horizontal sills with Bingen et al. (2015) published geochronological evi- an extensive lateral flow of magma and strongin situ dence for the presence of 2.22 Ga mafic rocks in gravitative differentiation. The sills were fed from the Palaeoproterozoic Kautokeino greenstone belt, sparse, rarely identified magma conduits, which northern Norway, which is plausible given the were channelled through earlier zones of weakness, fact that the Karelian formations continue from such as the narrow Kuhmo greenstone belt.

139 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

12.6 The 2.1–2.15 Ga mafic rocks

Ages of 2.1–2.15 Ga have been obtained for mafic also provide a minimum age of 2130 ± 5 Ma for the rocks from ca. 20 localities, mostly from north- Jouttiaapa Formation. ern Finland (Räsänen & Huhma 2001, Perttunen & The E–W-trending Fe-tholeiitic dykes in the Vaasjoki 2001, Manninen et al. 2001, Pekkarinen Kuhmo block appear to belong in this age group (e.g. & Lukkarinen 1991, Huhma 1986, this paper). A1212 Kapea-aho), whereas the NW–SE-trending The intrusions analysed for Sm–Nd isotopes are dykes (e.g. A1409 Kivikevätti) are younger, being characterised by positive εNd(T) values (Fig. 126), ca. 2 Ga. Dykes of this family have also been rec- some approaching the composition of the (model) ognised from Russian Karelia by Stepanova et al. depleted mantle (Fig. 125). The most positive εNd(T) (2014b), who report ages of 2140 ± 3 Ma and an value for the 2.1 Ga mafic rocks (+4.2) has been initial εNd of +3. obtained for the Jouttiaapa Formation metabasalts in The positive εNd(T) may indicate that major the Peräpohja belt (Huhma et al. 1990). In this for- attenuation of the lithosphere took place at ca. mation, the first lava flows are extremely depleted 2.1 Ga and eventually allowed material from the con- in LREE, and the Sm–Nd data on 16 whole-rock vective mantle to escape to the surface. This is also samples yield an age of 2103 ± 50 Ma for the major well exemplified by the Jeesiörova (and Sattasvaara) Sm–Nd fractionation, which can be also considered komatiites (Savukoski Group) in Central Lapland, as an estimate for the age of eruption (recalculated for which a slightly younger age of 2056 ± 25 Ma using three additional analyses). Recent analyses was obtained using several Sm–Nd analyses on of the roughly coeval Tikanmaa Formation as well pyroxene separates and whole-rock samples as two dykes (Koppakumpu, Kuusivaara) yielded (Hanski et al. 2001b). similar positive εNd values (Fig. 125c). These dykes

12.7 The 1.95–2.06 Ga mafic rocks

Mafic intrusions with an age of ca. 2050 Ma exist that the Kittilä Group contains products of bimodal in many places in northern Finland, particularly magmatism generated within an oceanic domain within the Savukoski Group (Rastas et al. 2001, (Hanski & Huhma 2005). Räsänen & Huhma 2001). These include the Ni ore- The other two sites of voluminous ca. 2.0 Ga bearing Kevitsa mafic-ultramafic body (Mutanen mafic magmatism in the Fennoscandian Shield are 1997, Mutanen & Huhma 2001) and recently dated the Onega region and Pechenga belt in Russia. In

Moskuvaara, Puijärvi and Satovaara intrusions. In the Onega region, basalts with an initial εNd(T) of central Finland, this age group is represented by the +3 suggest an origin from depleted mantle, but the

Otanmäki intrusion (Talvitie & Paarma 1980). The lower initial εNd values in some volcanic rocks have available Sm–Nd results provide initial Nd isotope been interpreted to result from significant crus- compositions very distinct from the coeval depleted tal contamination, and in this sense, these lavas mantle and suggest the involvement of old enriched resemble lavas from other continental flood basalt lithosphere in their genesis (Fig. 125). In particu- provinces (Puchtel et al. 1998, Huhma et al. unpub- lar, the Ni–PGE-bearing ore type from Kevitsa with lished). The Pechenga ferropicrites have provided a

εNd(T) of –6.4 has Nd isotope characteristics typi- Sm–Nd age of 1977 ± 52 Ma, which was supported cal for late Archaean crust, suggesting significant by U–Pb zircon analysis on a felsic rock (Hanski crustal contamination (Huhma et al. 1995, Hanski et al. 1990, Hanski 1992, Hanski et al. 2014). The et al. 1997, Hanski & Huhma 2005). initial εNd of +1.4 suggests that the source had slight

Rifting of the Archaean continent finally led to its time-integrated LREE depletion. A similar εNd(T) breakup and the formation of seafloor at ca. 2.0 Ga. value was also obtained from the nearby Nyasyukka The ~2015 Ma felsic porphyries and the associated dyke, which has yielded well-constrained ages of (Vesmajärvi Formation) mafic metavolcanic rocks 1941 ± 3 Ma (U–Pb on baddeleyite) and 1956 ± 19 Ma have yielded initial εNd compositions close to the (Sm–Nd ) (Smolkin et al. 2015). The analyses con- depleted mantle (Fig. 125). This, together with the ducted on associated tholeiitic basalts, instead, overall characteristics, has led to the conclusion have an initial εNd value of +3.5 and suggest an

140 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland origin from a depleted mantle, which also appears to connected to the depleted mantle, which is rep- be the case with slightly older Kolosjoki Formation resented, for example, by the high εNd(T) basalts tholeiites (Hanski et al. 2014). from the Vesmajärvi Formation or the Kevitsa dyke Several Sm–Nd analyses are also available from (Fig. 125a). the Jormua ophiolite complex, which has been Other alkaline rocks associated with the 2.0 Ga dated at 1953 ± 5 Ma by U–Pb on zircon (Peltonen rifting occur ca. 180 km north of Jormua, where the et al. 1996) and refined here to 1952 ± 2 Ma using Kortejärvi and Laivajoki carbonatites are found in concordant CA-TIMS analyses. The Jormua ophi- a N–S-trending shear zone between two Archaean olite represents a 1.95 Ga seafloor from an ocean blocks (Nykänen et al. 1997). In contrast to the to continent transition zone that mainly consisted Jormua OIB, these carbonatites provide a clearly of Archaean subcontinental lithospheric mantle positive initial εNd(T) of ca. +2.5 (Fig. 125b). As these (Peltonen et al. 2003). The E-MORB-type basalt rocks have high concentrations of REE, the obtained suite yielded a Sm–Nd age of 1936 ± 43 Ma and initial Nd isotope composition should well repre- an initial εNd value of +2, whereas the OIB-type sent the average mantle source that contributed dykes in the mantle peridotites gave a Sm–Nd age to the formation of carbonatitic magma at 2.0 Ga. of 1968 ± 58 Ma with a nearly chondritic initial Evidently, the mantle below the Archaean craton

εNd(T) (Peltonen et al. 1996, 1998). The source for was heterogeneous. the main E-MORB-type suite is thus not directly

13 CLOSING REMARKS

Nearly 400 Sm–Nd analyses of whole-rock samples reservoirs deep in the crust may explain many fea- and mineral separates from ca. 70 mafic intrusions tures observed in mafic-ultramafic rocks, such as and ten mafic volcanic formations in the Finnish the near constant εNd of -2 in the 2.44 Ga intrusions, Karelia province are reported in this volume. These but the isotopic results also show that various man- data, together with abundant U–Pb age results and tle sources with distinct isotope compositions have the previously published Sm–Nd analyses on ca. existed during the Palaeoproterozoic. Examples 30 mafic rock units, provide tools for constrain- are provided by high-REE mantle-derived rocks ing igneous petrogenesis and the Palaeoproterozoic showing a range of initial εNd values from nearly evolution in the Fennoscandian Shield. Over a time chondritic (e.g., 2.6 Ga Siilinjärvi carbonatite, 2.0 period of 700 Ma, mafic magmas were emplaced in Ga Jormua OIB, 1.8 Ga lamprophyres) to highly several episodes occurring at ca. 2.50 Ga, 2.44 Ga, positive (e.g., the 2.0 Ga Laivajoki and Kortejärvi 2.3 Ga, 2.22 Ga, 2.15 Ga, 2.12 Ga, 2.05 Ga, 2.0 Ga, 1.95 Ga, carbonatites). 1.88 Ga and 1.78 Ga. Many of the rock associations Contamination with country rock material in formed during these events may be regarded as the final site of intrusion has also been considered examples of ancient large igneous provinces, in this important in modifying the chemical composition case related to long-lasting episodic rifting of the of some rocks, for example in the 2.06 Ga Kevitsa

Archaean lithosphere. mafic intrusion, with anε Nd value of –3.4, and par- The emphasis of the Sm–Nd studies was on most ticularly the Kevitsa Ni-PGE mineralisation with an pristine mafic rocks available, which generally pro- εNd value of –6.4. duced Sm–Nd mineral ages consistent with the The age and initial Nd isotope composition, available U–Pb zircon ages. As many of the initial together with other relevant information, should

εNd values were based on Sm–Nd mineral isochrons, be used to correlate dykes, intrusions and mafic they should give reliable estimates of the initial iso- extrusive units in the Fennoscandian Shield. tope composition of the studied rocks. The initial εNd Although in situ contamination may occasionally values range from very positive to strongly nega- cause problems, the Nd isotope studies suggest tive and suggest that some rocks were derived from that the chemical characterisation of various for- a depleted mantle source, whereas others have a mations should provide useful tools for correlation. large contribution from old enriched lithosphere. The results could also be used in correlating events Contamination of ultramafic magma in well-mixed in different cratons, particularly across the Atlantic

141 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye to the Canadian Shield, from where compilations (2004), 2.3 Ga mafic rocks appear to be absent in of magmatic events (LIPs) have been published by Canada, but the 2.2 Ga event is well represented by Ernst & Buchan (2004) and Ernst & Bleeker (2010). the Ungava dykes and Nipissing sills. In Finland, Some of these events are analogous on both sides 2.2 Ga mafic rocks form layered intrusions and of the Atlantic, but differences also seem to exist. sills close to the Archaean craton margins, but in The global occurrence of 2.44 Ga mafic magmatism Canada, they also occur as dyke swarms. The 2.1 Ga is well known (e.g., Alapieti 1982, Heaman 1997). event is found in the Superior, for example, as the It has been termed the Matachewan event in the Marathon linear dyke swarm. Superior Province. According to Ernst & Buchan

ACKNOWLEDGEMENTS

We want to express our gratitude to Olavi Kouvo for Museum of Natural History. This is NORDSIM con- the farsighted vision he showed when initiating the tribution number #547. isotope research in Finland. His contribution to the Tom Andersen is acknowledged for providing the data presented in this paper is also enormous. The program used for MC-ICPMS data reduction. personnel of the isotope laboratory are also greatly Excellent sample material was provided by seve- acknowledged, particularly Tuula Hokkanen, Arto ral geologists: Tapani Mutanen, Vesa Perttunen, Pulkkinen, Leena Järvinen, Matti Karhunen, Marita Jorma Räsänen, Petri Peltonen, Perttu Mikkola, Niemelä, Mirja Saarinen and Hugh O’Brien. Vesa Nykänen, Jorma Palmen, Tuomo Karinen, We acknowledge the laboratories in Stockholm Juha Karhu, Jorma Paavola, Erkki Luukkonen and (NORDSIM/ Martin Whitehouse) and St. Petersburg Heikki Juopperi. We appreciate discussion with (VSEGEI, SHRIMP/ Sergei Sergeev) for SIMS analy- these geologists, as well as discussions with Pentti ses and Sandra Kamo in Toronto for U–Pb bad- Hölttä, Jarmo Kohonen and Raimo Lahtinen, and deleyite data. comments on the manuscript by Jouni Luukas. The The NORDSIM facility has been supported by the review comments by Stefan Claesson are greatly Research Councils of Denmark, Norway and Sweden, acknowledged. E.H. acknowledges support from the Geological Survey of Finland and the Swedish Academy of Finland grant 281859.

REFERENCES

Alapieti, T. 1982. The Koillismaa layered igneous complex, Amelin, Yu. V., Heaman, L. & Semenov, V. 1995. UPb Finland: its structure, mineralogy and geochemistry, geochronometry of layered mafic intrusions in the with emphasis on the distribution of chromium. Geo- eastern Baltic Shield: implications for the timing and logical Survey of Finland, Bulletin 319. 116 p. Available duration of Palaeoproterozoic rifting. Precambrian at: http://tupa.gtk.fi/julkaisu/bulletin/bt_319.pdf Research 75, 31–46. Alapieti, T. T., Filen, B. A., Lahtinen, J. J., Lavrov, M. M., Andersen, T., Griffin W. L., Jackson S. E., Knudsen T.-L. Smolkin, V. F. & Voitsekhovsky, S. N. 1990. Early Pro- & Pearson N. J. 2004. Mid-Proterozoic magmatic arc terozoic layered intrusions in the northeastern part of evolution at the southwest margin of the Baltic Shield. the Fennoscandian Shield. Mineralogy and Petrology Lithos 73, 289–318. 42, 1–22. Andersson, U. B., Eklund, O., Fröjdö, S. & Konopelko, D. Alapieti, T., Lahtinen, J., Huhma, H., Hänninen, E., Pi- 2006. 1.8 Ga magmatism in the Fennoscandian Shield; irainen, T. & Sivonen, S. 1989. Platinum-group ele- lateral variations in subcontinental mantle enrichment-bearing Cu-Ni sulphide mineralization in the ment. Lithos 86, 110–136. marginal series of the early Proterozoic Suhanko- Balashov, Y. A., Bayanova, T. B. & Mitrofanov, F. P. 1993. Konttijärvi layered intrusion, northern Finland. In: Isotope data on the age and genesis of layered basic- Prendergast, M. D. & Jones, M. J. (eds) Magmatic sul- ultrabasic intrusions in the Kola Peninsula and north- phides - the Zimbabwe volume. London: The Institu- ern Karelia, northeastern Baltic Shield. Precambrian tion of Mining and Metallurgy, 177−187. Research, 64 197–205. Amelin, Yu. V. & Semenov, V. S. 1996. Nd and Sr isotope Bayanova, T., Smolkin, V., Leckovich, N. & Ryungenen, geochemistry of mafic layered intrusions in the east- G. 1999. U-Pb age of rocks of the Mt. Generals- ern Baltic Shield: Implications for the sources and kaya layered intrusion, Kola Peninsula. Geochemical contamination of Palaeoproterozoic continental mafic International 37 (1), 1–10. magmas. Contributions to Mineralogy and Petrology Belousova, E. A., Griffin W. L. & O’Reilly S. Y. 2006. Zir- 124, 255–272. con crystal morphology, trace element signatures

142 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

and Hf isotope composition as a tool for petrogenetic Grinenko, L. N., Hanski, E. & Grinenko, V. A. 2003. For- modeling: examples from Eastern Australian grani- mation conditions of the Keivitsa Cu-Ni deposit, toids. Journal of Petrology 47, 329–353. northern Finland: evidence from S and C isotopes. Bergman, S., Kubler, L. & Martinsson, O. 2001. Descrip- Geochemistry International 41 (2), 154–167. tion of regional geological and geophysical maps of Halkoaho, T. & Niskanen, M. 2013. New PGE-Cu-Ni ob- northern Norrbotten County (east of the Caledonian servations from the early Palaeoproterozoic Juntti- orogen). Sveriges Geologiska Undersökning Ba 56. lanniemi layered intrusion, Paltamo, eastern Finland. 110 p. Geological Survey of Finland, Report of Investigation Bernelius, S. O. 2009. Suomussalmen varhaisproterot- 198, 29–33. Available at: http://tupa.gtk.fi/julkaisu/ sooisten diabaasien mineralogia tektonisessa tulkin- tutkimusraportti/tr_198.pdf nassa. Unpublished M.Sc. Thesis, University of Hel- Hanski, E. 1984. Geology of the gabbro-wehrlite associa- sinki. (in Finnish) tion in the eastern part of the Baltic Shield. Research Bingen, B., Solli, A., Viola, G., Torgersen, E., Sandstad, Project of Archean Areas, Report No. 20, University of J. V., Whitehouse, M. J., Røhr, T. S., Ganerød, M. & Oulu. 78 p. Nasuti, A. 2015. Geochronology of the Palaeoprotero- Hanski, E. 1992. Petrology of the Pechenga ferropicrites zoic Kautokeino Greenstone Belt, Finnmark, Norway: and cogenetic, Ni-bearing gabbro-wehrlite intru- Tectonic implications in a Fennoscandia context. Nor- sions, Kola Peninsula, Russia. Geological Survey of wegian Journal of Geology 95, 365–396. Finland Bulletin 367. 192 p. Available at: http://tupa. Buchan, K. L., Mertanen, S., Park, R. G., Pesonen, L .J., gtk.fi/julkaisu/bulletin/bt_367.pdf Elming, S.-A., Abrahamsen, N. & Bylund, G. 2000. Hanski, E. 1997. The Nuttio serpentinite belt, Central Comparing the drift of Laurentica and Baltica in the Lapland: An example of Paleoproterozoic ophiolitic Proterozoic: the importance of key palaeomagnetic mantle rocks in Finland. Ofioliti 22, 35–46. poles. Tectonophysics 319, 167–198. Hanski, E. 2012. Evolution of the Palaeoproterozoic Condie, K. C., O’Neill, C. & Aster, R. C. 2009. Evidence and (2.50–1.95 Ga) non-orogenic magmatism in the east- implications for a widespread magmatic shutdown for ern part of the Fennoscandian Shield. In: Melezhik, V. 250 My on Earth. Earth and Planetary Science Letters A., Prave, A. R., Hanski, E. J., Fallick, A. E., Lepland, A., 282, 294–298. Kump, L. R. & Strauss, H. (eds) Reading the Archive of Daly, J. S., Balagansky, V. V., Timmerman, M. J., White- Earth’s Oxygenation, Volume 1: The Palaeoproterozo- house, M. J., de Jong, K., Guise, P., Bogdanova, S., ic of Fennoscandia as Context for the Fennoscandian Gorbatschev, R. & Bridgwater, D. 2001. Ion micro- Arctic Russia - Drilling Earth Project. Berlin, Heidel- probe U–Pb zircon geochronology and isotopic evi- berg: Springer-Verlag, 179–245. dence for a trans-crustal suture in the Lapland–Kola Hanski, E. & Huhma, H. 2005. Central Lapland green- Orogen, northern Fennoscandian Shield. Precambrian stone belt. In: Lehtinen, M., Nurmi, P. A. & Rämö, O. Research 105, 289–314. T. (eds) Precambrian Geology of Finland. Key to the DePaolo, D. J. 1981. Neodymium isotopes in the Colorado Evolution of the Fennoscandian Shield. Amsterdam: Front Range and crust-mantle evolution in the Prote- Elsevier Science B.V., 139–193. rozoic. Nature 291, 684–687. Hanski, E. J. & Melezhik, V. A. 2012. Litho- and chronos- DePaolo, D. J. & Wasserburg, G. J. 1976. Nd isotopic vari- tratigraphy of the Karelian formations. In: Melezhik, ations and petrogenetic models. Geophysical Research V. A., Prave, A. R., Hanski, E. J., Fallick, A. E., Lepland, Letters 3, 249−252. A., Kump, L. R. & Strauss, H. (eds) Reading the Archive De Wolf, C. P., Zeissler, C. J., Halliday, A. N., Mezger, K. of Earth’s Oxygenation. Volume 1: The Palaeoprotero- & Essene, E. J. 1996. The role of inclusions in U‑Pb zoic of Fennoscandia as Context for the Fennoscandi- and Sm‑Nd garnet geochronology: Stepwise dissolu- an Arctic Russia – Drilling Early Earth Project. Berlin, tion experiments and trace uranium mapping by fis- Heidelberg: Springer-Verlag, 39–110. sion track analysis. Geochimica Cosmochimica Acta Hanski, E. J., Grinenko, L. N. & Mutanen, T. 1996. Sulfur 60, 121−134. isotopes in the Keivitsa Cu-Ni-bearing intrusion and Eilu, P. 1994. Hydrothermal alteration in volcano-sedi- and its country rocks, northern Finland: evidence for mentary rocks in the Central Lapland greenstone belt, crustal contamination. IGCP 336 Project, Field Confer- Finland. Geological Survey of Finland, Bulletin 374. ence on Layered Mafic Complexes and Related Ore De- 145 p. Available at: http://tupa.gtk.fi/julkaisu/bulle- posits of Northern Fennoscandia, August 21–23, 1996, tin/bt_374.pdf Abstracts, p. 15–16. Eklund, O., Konopelko, D., Andersson, U. B., Korja, A. Hanski, E., Huhma, H. & Melezhik, V. 2014. New isotopic & Shebanov, A. D. 2000. 1.8 Ga Svecofennian post- and geochemical data from the Palaeoproterozoic Pe- collisional magmatism in the Fennoscandian Shield: chenga Greenstone Belt, NW Russia: implication for Evidence of mantle enrichment and detachment of the basin development and duration of the volcanism. lithospheric mantle. 24. Nordiske Geologiske Vinter- Precambrian Research 245, 51–65. möte, Trondheim 6.–9.1.2000. Geonytt 1, p. 61. Hanski, E., Huhma, H., Rastas, P. & Kamenetsky, V. S. Eklund, O., Konopelko, D., Rutanen, H., Fröjdö, S. & She- 2001b. The Palaeoproterozoic komatiite-picrite as- banov, A. D. 1998. 1.8 Ga Svecofennian post-colli- sociation of Finnish Lapland. Journal of Petrology 42, sional shoshonitic magmatism in the Fennoscandian 855–876. Shield. Lithos 45, 87–108. Hanski, E., Huhma, H., Smolkin, V. F. & Vaasjoki, M. Ernst, R. & Bleeker, W. 2010. Large Igneous provinces 1990. The age of the ferropicritic volcanics and co- (LIPs), giant dyke swarms, and mantle plumes: sig- magmatic Ni-bearing intrusions at Pechenga, Kola nificance for breakup events within Canada and adja- Peninsula, U.S.S.R. Bulletin of the Geological Society of cent regions from 2.5 Ga to the Present. Can. J. Earth Finland 62, 123–133. Available at: http://www.geolog- Sci. 47, 695–739. inenseura.fi/bulletin/Volume62/sgs_bt_062_2_ Ernst, R. E. & Buchan, K. L. 2004. Igneous rock associa- pages_123_133.pdf tions in Canada 3. Large Igneous Provinces (LIPs) in Hanski, E. J., Huhma, H., Suominen, I. M. & Walker, R. J. Canada and adjacent Regions: 3 Ga to Present. Geos- 1997. Geochemical and isotopic (Os, Nd) study of the cience Canada, 31, 103–126. early Proterozoic Keivitsa intrusion and its CuNi de-

143143 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

posit, northern Finland. In: Papunen, H. (ed.) Mineral Huhma, H., Kontinen, A., Mikkola, P., Halkoaho, T., Deposits: Research and Exploration Where Do They Hokkanen, T., Hölttä, P., Juopperi, H., Konnunaho, J., Meet? Proceedings of the Fourth Biennial SGA Meet- Luukkonen, E., Mutanen, T., Peltonen, P., Pietikäi ing, Turku/Finland/11–13 August 1997. Rotterdam: A. nen, K. & Pulkkinen, A. 2012b. Nd isotopic evidence A. Balkema, 435–438. for Archaean crustal growth in Finland. In: Hölttä P. Hanski, E., Huhma, H. & Vaasjoki, M. 2001a. Geochronol- (ed.) The Archaean of the Karelia Province in Fin- ogy of northern Finland: a summary and discussion. land. Geological Survey of Finland, Special Paper 54, In: Vaasjoki, M. (ed.) Radiometric age determinations 176−213. Available at: http://tupa.gtk.fi/julkaisu/spe- from Finnish Lapland and their bearing on the timing cialpaper/sp_054.pdf of Precambrian volcano-sedimentary sequences. Geo- Huhma, H., Mänttäri, I., Peltonen, P., Kontinen, A., logical Survey of Finland Special Paper 33, 255–279. Halkoaho, T., Hanski, E., Hokkanen, T., Hölttä, P., Available at: http://tupa.gtk.fi/julkaisu/specialpaper/ Juopperi, H., Konnunaho, J., Lahaye, Y., Luukkonen. sp_033.pdf E., Pietikäinen. K., Pulkkinen, A., Sorjonen-Ward, Hanski, E., Huhma, H. & Vuollo, J. 2010. SIMS zircon ages P., Vaasjoki, M. & Whitehouse, M. 2012a. The age of and Nd isotope systematics of the 2.2 Ga mafic intru- the Archaean greenstone belts in Finland. In: Hölttä sions in northern and eastern Finland. Bulletin of the P. (ed.) The Archaean of the Karelia Province in Fin- Geological Society of Finland 82, 31–62. Available at: land. Geological Survey of Finland, Special Paper 54, http://www.geologinenseura.fi/bulletin/Volume82/ 74−175. Available at: http://tupa.gtk.fi/julkaisu/spe- Hanski2010.pdf cialpaper/sp_054.pdf Hanski, E., Walker, R. J., Huhma, H. & Suominen I. Huhma, H., Mutanen, T., Hanski, E., Räsänen, J., Man- 2001c. The Os and Nd isotopic systematics of c. 2.44 ninen, T., Lehtonen, M., Rastas, P. & Juopperi, H. Ga Akanvaara and Koitelainen mafic layered intru- 1996. Sm-Nd isotopic evidence for contrasting sources sions in northern Finland. Precambrian Research 109, of the prolonged Palaeoproterozoic mafic–ultramafic 73–102. magmatism in Central Finnish Lapland. IGCP Project Heaman, L. M. 1997. Global mafic magmatism at 2.45 Ga: 336 Symposium, Rovaniemi, Finland, August 21–23, Remnants of an ancient large igneous province? Geol- 1996, Program and Abstracts, Publications of the De- ogy 25, 299–302. partment of Geology and Mineralogy of the University Heaman, L. M. & LeCheminant, A. N. 1993. Paragenesis of Turku 38, p. 17. and U–Pb systematics of baddeleyite (ZrO2). Chemical Huhma, H., Mutanen, T., Hanski, E. & Walker, R. J. 1995. Geology 110 (1–3), 95–126. Sm-Nd, U-Pb and Re-Os isotopic study of the Kei- Hiltunen, A. 1982. The Precambrian geology and skarn vitsa Cu-Ni-bearing intrusion, northern Finland. 22nd iron ores of the Rautuvaara area, northern Fin- Nordic Geological Winter Meeting, Turku, 1996, Ab- land. Bull. Geological Survey of Finland, Bulletin 318, stracts, p. 72. 1–133. Available at: http://tupa.gtk.fi/julkaisu/bulle- Huhma, H., O’Brien, H., Lahaye, Y. & Mänttäri, I. 2011. tin/bt_318.pdf Isotope geology and Fennoscandian lithosphere evo- Hölttä, P., Heilimo, E., Huhma, H., Juopperi, H., Konti- lution. Geological Survey of Finland, Special Paper 49, nen, A., Konnunaho, J., Lauri, L., Mikkola, P., Paavo- 35−48. Available at: http://tupa.gtk.fi/julkaisu/spe- la, J. & Sorjonen-Ward, P. 2012. Archaean complex- cialpaper/sp_049.pdf es of the Karelia Province in Finland. In: Hölttä, P. Iljina, M. & Hanski, E. 2005. Layered mafic intrusions (ed.) The Archaean of the Karelia Province in Finland. of the Tornio-Näränkävaara belt. In: Lehtinen, M., Geological Survey of Finland, Special Paper 54, 9−20. Nurmi, P. & Rämö, O. T. (eds) Precambrian Bedrock of Available at: http://tupa.gtk.fi/julkaisu/specialpaper/ Finland – Key to the Evolution of the Fennoscandian sp_054.pdf Shield. Amsterdam: Elsevier, 103–138. Hölttä, P., Huhma, H., Mänttäri, I. & Paavola, J. 2000. Jackson S. E., Pearson N. J., Griffin W. L. & Belousova E. PTt development of Archaean granulites in Varpais- A. 2004. The application of laser ablation-inductively järvi, Central Finland, II: Dating of highgrade meta- coupled plasma-mass spectrometry to in-situ U-Pb morphism with the UPb and SmNd methods. Lithos zircon geochronology. Chemical Geology 211, 47–69. 50, 121–136. Juopperi, H. & Vaasjoki, M. 2001. U-Pb mineral age de- Horstwood, M. S. A., Foster, G. L., Parrish, R. R., Noble, S. terminations from Archean rocks in eastern Lapland. R. & Nowell, G. M. 2003. Common-Pb corrected in situ In: Vaasjoki, M. (ed.) Radiometric age determinations U-Pb accessory mineral geochronology by LA-MC- from Finnish Lapland and their bearing on the timing ICPMS, J. Anal. At. Spectrom, 18, 837–846. of Precambrian volcano-sedimentary sequences. Geo- Huhma, A. 1981. Youngest Precambrian dyke rocks in logical Survey of Finland, Special Paper 33, 209–227. north Karelia, east Finland. Bulletin of the Geologi- Available at: http://tupa.gtk.fi/julkaisu/specialpaper/ cal Society of Finland 53, 67–82. Available at: http:// sp_033.pdf www.geologinenseura.fi/bulletin/Volume53/sgs_ Kantti, J.-P. 2002. Kaamajoen gabronoriitti-intruusio bt_053_2_pages_067_082.pdf Enontekiöllä. Unpublished M.Sc. Thesis, Departure of Huhma, H. 1986. Sm-Nd, U-Pb and Pb-Pb isotopic ev- Geology and Mineralogy, University of Turku. 115 p. idence for the origin of the Early Proterozoic Sve- (in Finnish) cokarelian crust in Finland. Geological Survey of Fin- Kärenlampi, K. 2015. The geology and industrial potential land, Bulletin 337. 48 p. Available at: http://tupa.gtk. of the Koidanvaara mafic intrusion in Ilomantsi (Ilo- fi/julkaisu/bulletin/bt_337.pdf mantsin Koidanvaaran mafisen intruusion geologia ja Huhma, H. 1987. Provenance of early Proterozoic and Ar- teollinen potentiaali) Unpublished M.Sc. Thesis, Oulu chean metasediments in Finland: a Sm-Nd isotopic Mining School, University of Oulu. 161 p. (in Finnish) study. Precambrian Research 35, 127–143. Karhu, J. A. 1993. Paleoproterozoic evolution of the car- Huhma, H., Cliff, R. A., Perttunen, V. & Sakko, M. 1990. bon isotope ratios of sedimentary carbonates in the Sm-Nd and Pb isotopic study of mafic rocks associ- Fennoscandian Shield. Geological Survey of Finland, ated with early Proterozoic continental rifting: the Bulletin 371. 87 p. Available at: http://tupa.gtk.fi/ju- Peräpohja schist belt in northern Finland. Contribu- lkaisu/bulletin/bt_371.pdf tions to Mineralogy and Petrology 104, 367–379.

144 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Karhu, J. A. 2005. Paleoproterozoic carbon isotope excur- Group keeps fighting us. In: Lauri, L. S., Heilimo, E., sion. In: Lehtinen, M., Nurmi, P. A. & Rämö, O. T. (eds) Leväniemi, H., Tuusjärvi, M., Lahtinen, R. & Hölttä, Precambrian Geology of Finland – Key to the Evolu- P. (eds) Current Research: 2nd GTK Mineral Poten- tion of the Fennoscandian Shield. Amsterdam: Else- tial Workshop, Kuopio, Finland, May 2014. Geological vier B.V., 669–680. Survey of Finland, Report of Investigation 207, 68–71. Karhu, J., Kortelainen, N. M., Huhma, H., Perttunen, V. Available at: http://tupa.gtk.fi/julkaisu/tutkimus- & Sergeev, S. 2007. New time constraints for the end raportti/tr_207.pdf of the Paleoproterozoic carbon isotope excursion. 7th Korsman, K., Koistinen, T., Kohonen, J., Wennerström, International Symposium on Applied Isotope Geoche- M., Ekdahl, E., Honkamo, M., Idman, H. & Pekkala, Y. mistry, 10th–14th September, 2007, Stellenbosch, (eds) 1997. Bedrock map of Finland 1:1 000 000. Es- South Africa, 76–77. poo: Geological Survey of Finland. Karinen, T. 2010. The Koillismaa Intrusion, northeastern Kouvo, O. 1977. The use of mafic pegmatoids in geochro- Finland – evidence for PGE reef forming processes in nometry. In: ECOG V. Fifth European Colloquium of the layered series. Geological Survey of Finland, Bul- Geochronology, Cosmochronology and Isotope Geolo- letin 404. 176 p. Available at: http://tupa.gtk.fi/julkai- gy, Pisa, September 5–10, 1977. 1 p. su/bulletin/bt_404.pdf Kramm, U., Kogarko, L. N., Kononova, V. A. & Vartiai- Karinen, T., Hanski, E. & Taipale, A. 2015. The Musta- nen, H. 1993. The Kola Alkaline Province of the CIS vaara Fe-Ti-V oxide deposit. In: Maier, W., O’Brien, and Finland: precise Rb–Sr ages define 380–360 Ma H. & Lahtinen, R. (eds) Mineral Deposits of Finland. age range for all magmatism. Lithos 30, 33–44. Amsterdam: Elsevier, 179–192. Krogh, T. 1973. A low-contamination method for hydro- Karvinen, A., Kojonen, K. & Johanson, B. 1988. Geology thermal decomposition of zircon and extraction of U and mineralogy of the Karhujupukka Ti-Fe deposit in and Pb for isotopic age determinations. Geochimica Kolari, northern Finland. Geological Survey of Fin- et Cosmochimica Acta 37, 485–494. land, Special Paper 10, 95–99. Available at: http:// Krogh, T. E. 1982. Improved accuracy of U-Pb zircon ages tupa.gtk.fi/julkaisu/specialpaper/sp_010.pdf by the creation of more concordant systems using an Kilpelä, M. 1991. Early Proterozoic diabases in the East- air abrasion technique. Geochimica et Cosmochimica ern Kainuu area. (Itä-Kainuun varhaisproterotsooiset Acta 46, 637–649. diabaasit) Unpublished M.Sc. Thesis, University of Krogh, T. E., Corfu, F., Davis, D. W., Dunning, G. R., Turku. 113 p. (in Finnish) Heaman, L. M., Kamo, S. L., Machado, N., Greenough, Klein, E. M. 2003. Geochemistry of the igneous oceanic J. D. & Nakamura, E. 1987. Precise U–Pb isotopic ages crust. In: Rudnick, R. L. (ed.) The Crust. Treatise on of diabase dykes and mafic to ultramafic rocks using Geochemistry, Volume 3. Oxford: Elsevier, 433–463. trace amounts of baddeleyite and zircon. In: Halls, H. Kohonen, J. & Marmo, J. 1992. Proterozoic lithostratig- C. & Fahrig, W. F. (eds) Mafic Dyke Swarms. Geological raphy and sedimentation of Sariola and Jatuli-type Association of Canada, Special Paper 34, 147–152. rocks in the Nunnanlahti–Koli–Kaltimo area, east- Kyläkoski, M., Hanski, E. & Huhma, H. 2012. The Petä- ern Finland; implications for regional basin evolution jäskoski Formation, a new lithostratigraphic unit in models. Geological Survey of Finland, Bulletin 364. the Paleoproterozoic Peräpohja Belt, northern Fin- 67 p. Available at: http://tupa.gtk.fi/julkaisu/bulletin/ land. Bulletin of the Geological Society of Finland 84, bt_364.pdf 85–120. Available at: http://www.geologinenseura.fi/ Koistinen, T., Stephens, M. B., Bogatchev, V., Nordgulen, bulletin/Volume84/Kylakoski.pdf Ø., Wennerström, M. & Korhonen, J. (comp.) 2001. Laajoki, K. 1991. Stratigraphy of the northern end of the Geological Map of the Fennoscandian Shield, scale early Proterozoic (Karelian) Kainuu Schist Belt and 1:2 000 000. Geological Survey of Finland (Espoo), associated gneiss complexes, Finland. Geological Sur- Geological Survey of Norway (Trondheim), Geologi- vey of Finland, Bulletin 358. 105 p. Available at: http:// cal Survey of Sweden (Uppsala), Ministry of Natural tupa.gtk.fi/julkaisu/bulletin/bt_358.pdf Resources of Russia (Moscow). Laajoki, K. 2005. Karelian supracrustal rocks. In: Leht- Kontinen, A. 1986. Early Proterozoic stratigraphy and inen, M., Nurmi, P. A. & Rämö, O. T. (eds) Precam- sedimentation in the Hyrynsalmi area, eastern Fin- brian Geology of Finland – Key to the Evolution of land. In: Sokolov, V. A. & Heiskanen, K. I. (eds) Early the Fennoscandian Shield. Amsterdam: Elsevier B.V., Proterozoic of the Baltic Shield, Proceedings of the 279−342. Finnish – Soviet Symposium held in Petrozavodsk Lahtinen, R., Huhma, H., Kontinen, A., Kohonen, J. & 19th–27th August, 1985. Petrozavodsk: Karelian Sci- Sorjonen-Ward, P. 2010. New constraints for the ence Centre of the USSR Academy of Sciences, 75–103. source characteristics, deposition and age of the 2.1– Kontinen, A. 1987. An early Proterozoic ophiolite - the 1.9 Ga metasedimentary cover at the western margin Jormua mafic-ultramafic complex, northeastern Fin- of the Karelian Province. Precambrian Research 176, land. Precambrian Research 35, 313–342. 77–93. Kontinen, A. & Hanski, E. 2015. The Talvivaara black Lahtinen, R., Huhma H, Lahaye Y., Jonsson E., Manninen shale-hosted Ni-Zn-Cu-Co deposit in eastern Fin- T., Lauri L., Bergman S., Hellström F., Niiranen T. & land. In: Maier, W. D., Lahtinen, R. & O’Brien, H. (eds) Nironen M. 2015a. New geochronological and Sm-Nd Minerals Deposits of Finland. Amsterdam: Elsevier, constraints across the Pajala shear zone of northern 557−612. Fennoscandia: reactivation of a Paleoproterozoic su- Kontinen, A., Huhma, H., Lahaye, Y. & O’Brien, H. 2013. ture. Precambrian Research 256, 10–119. New U-Pb zircon age, Sm-Nd isotope and geochemical Lahtinen, R., Huhma, H, Lahaye, Y., Kousa, J. & Luukas, data for Otanmäki suite granites in the Kainuu area, J. 2015b. Archean-Proterozoic collision boundary in Central Finland. In Hölttä P. (ed.) Current Research: central Fennoscandia: Revisited. Precambrian Re- GTK Mineral Potential Workshop, Kuopio, May 2012. search 261, 127−165. Geological Survey of Finland, Report of Investigation Lahtinen, R., Korja, A. & Nironen, M. 2005. Paleoprotero- 198, 65–69. Available at: http://tupa.gtk.fi/julkaisu/ zoic tectonic evolution. In: Lehtinen, M., Nurmi, P. A. tutkimusraportti/tr_198.pdf & Rämö, O. T. (eds) Precambrian Geology of Finland Kontinen, A., Huhma H., Lahaye Y. & O’Brien H. 2014. – Key to the Evolution of the Fennoscandian Shield. The problem with the age of the Central Puolanka Amsterdam: Elsevier B.V., 481–532.

145 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Larionov, A. N., Andreichev, V. A. & Gee, D. G. 2004. The area, northeastern Finland. Geological Survey of Fin- Vendian alkaline igneous suite of northern Timan: ion land, Report of Investigation 104, 97 p. Available at: microprobe U–Pb zircon ages of gabbros and syenite. http://tupa.gtk.fi/julkaisu/tutkimusraportti/tr_104. In: Gee, D. G. & Pease, V. L. (eds) The Neoproterozoic pdf Timanide Orogen of Eastern Baltica. London: Geolog- Manninen, T. & Huhma, H. 2001. A new U-Pb zircon con- ical Society, Memoirs 30, 69–74. straint from the Salla schist belt, northern Finland. Lauri, L. S., Andersen, T., Hölttä, P., Huhma, H. & Gra- In: Vaasjoki, M. (ed.) Radiometric age determinations ham, S. 2011. Evolution of the Archaean Karelian prov- from Finnish Lapland and their bearing on the timing ince in the Fennoscandian Shield in the light of U–Pb of Precambrian volcano-sedimentary sequences. Geo- zircon ages and Sm–Nd and Lu–Hf isotope systemat- logical Survey of Finland, Special Paper 33, 201–208. ics. Journal of the Geological Society 168, 201–218. Available at: http://tupa.gtk.fi/julkaisu/specialpaper/ Lauri, L. S., Rämö, O. T., Huhma, H., Mänttäri, I. & sp_033.pdf Räsänen, J. 2006. Petrogenesis of silicic magmatism Manninen, T., Pihlaja, P. & Huhma, H. 2001. U-Pb related to the ~2.44 Ga rifting of Archean crust in geochronology of the Peurasuvanto area, northern Koillismaa, eastern Finland. Lithos 86, 137–166. Finland. In: Vaasjoki, M. (ed.) Radiometric age deter- Lehtonen, M., Airo, M.-L., Eilu, P., Hanski, E., Kortelainen, minations from Finnish Lapland and their bearing on V., Lanne, E., Manninen, T., Rastas, P., Räsänen, J. & the timing of Precambrian volcano-sedimentary se- Virransalo, P. 1998. Kittilän vihreäkivialueen geolo- quences. Geological Survey of Finland, Special Paper gia. Summary: The stratigraphy, petrology and geo- 33, 189–200. Available at: http://tupa.gtk.fi/julkaisu/ chemistry of the Kittilä greenstone area, Northern specialpaper/sp_033.pdf Finland. Geological Survey of Finland, Report of In- Mänttäri, I. & Hölttä, P. 2002. U-Pb dating of zircons and vestigation 140. 144 p. Available at: http://tupa.gtk.fi/ monazites from Archean granulites in Varpaisjarvi, julkaisu/tutkimusraportti/tr_140.pdf Central Finland: Evidence for multiple metamorphism Lindholm, O. & Anttonen, R. 1980. Geology of the Otan- and Neoarchean terrane accretion. Precambrian Rese- mäki Mine. In: Häkli, T. A. (ed.) Precambrian Ores of arch 118, 101−131. Finland: Guide to Excursions 078 A + C, Part 2 (Fin- Mattila, H. 1974. Karelidit Savukosken Tanhuan alueella, land). Proceedings of the 26th International Geolog- Keski-Lapissa. Unpublished M.Sc. Thesis, Departure ical Congress. Espoo: Geological Survey of Finland, of Geology, University of Oulu. 89 p. (in Finnish) 25–33. Available at: http://tupa.gtk.fi/julkaisu/eriko- Mattinson, J. M. 2005. Zircon U-Pb chemical abra- isjulkaisu/ej_003.pdf sion (“CA-TIMS”) method: Combined annealing and Ludwig, K. R. 1991. PbDat 1.21 for MS-dos: A computer multi-step partial dissolution analysis for improved program for IBM-PC Compatibles for processing raw precision and accuracy of zircon ages. Chemical Geol- Pb-U-Th isotope data. Version 1.07. U.S. Geological ogy 220, 47–66. Survey. Open File Report, 88−542, 35 p. Mertanen, S. 1995. Multicomponent remanent mag- Ludwig, K. R. 2003. Isoplot/Ex 3. A Geochronological netizations reflecting the geological evolution of the Toolkit for Microsoft Excel. Berkeley Geochronologicy Fennoscandian Shield – a palaeomagnetic study with Center, Special Publication No. 4. emphasis on the Svecofennian orogeny. Espoo: Geo- Lukkarinen, H. 2008. Siilinjärven ja Kuopion kartta- logical Survey of Finland. 187 p. Available at: http:// alueiden kallioperä. Summary: Pre-Quaternary Rocks tupa.gtk.fi/julkaisu/erikoisjulkaisu/ej_012_synopsis. of the Siilinjärvi and Kuopio Map-Sheet areas. Geo- pdf logical Map of Finland 1:100 000, Explanation to the Mertanen, S., Halls, H. C., Vuollo, J., Pesonen, L. J. & Ste- Maps of Pre-Quaternary Rocks, Sheets 3331 and 3242. panov, V. S. 1999. Palaeomagnetism of 2.44 Ga ma- Geological Survey of Finland. 228 p. + 2 app. maps. fic dykes in Russian Karelia, eastern Fennoscandian Electronic publication. Available at: http://tupa.gtk.fi/ Shield—implications for continental reconstructions. kartta/kallioperakartta100/kps_3331_3242.pdf Precambrian Research 98, 197–221. Luukkonen, E. J. 1989. The Early Proterozoic Saari-Kiekki Mertanen, S., Pesonen, L. J., Huhma, H. & Leino, M. A. greenstone belt: a representative of the Sariola group H. 1989. Palaeomagnetism of the Early Proterozoic at Kuhmo, eastern Finland. Bulletin of the Geologi- layered intrusions, northern Finland. Geological Sur- cal Society of Finland 61, 161–187.Available at: http:// vey of Finland, Bulletin 347. 40 p. Available at: http:// www.geologinenseura.fi/bulletin/Volume61/sgs_ tupa.gtk.fi/julkaisu/bulletin/bt_347.pdf bt_061_2_pages_161_187.pdf Mikkola, P., Heilimo E., Paavola J., Halkoaho T., Äikäs O. Maier, W. D., Arndt, N. T. & Curl, E. A. 2000. Progres- & Huhma H. 2013. Bedrock of the southern part of the sive crustal contamination of the Bushveld Complex: Lentua complex. Geological Survey of Finland, Report evidence from Nd isotopic analyses of the cumulate of Investigation 202. 96 p. Available at: http://tupa. rocks. Contributions to Mineralogy and Petrology 140, gtk.fi/julkaisu/tutkimusraportti/tr_202.pdf 316–327. Mikkola, P., Huhma, H., Heilimo, E. & Whitehouse, M. Maier, W. D., Halkoaho, T., Huhma, H., Hanski, E. & 2011. Archean crustal evolution of the Suomussalmi Barnes, S.-J. (in press) The Penikat intrusion, Fin- district as part of the Kianta Complex, Karelia: Con- land: Geochemistry, geochronology, and origin of straints from geochemistry and isotopes of granitoids. Pt-Pd reefs through tectonically induced injection of Lithos 125, 287–307. volatile-rich crystal mushes. Journal of Petrology. Müller, W., Shelley, M., Miller, P. & Broude, S. 2009. “In- Makkonen, H. & Huhma, H. 2007. Sm-Nd data for maf- itial performance metrics of a new custom-designed ic-ultramafic intrusions in the Svecofennian (1.88 ArF excimer LA-ICPMS system coupled to a two-vol- Ga) Kotalahti Nickel Belt, Finland – implications for ume laser-ablation cell.” Journal of Analytical Atomic crustal contamination at the Archaean/Proterozoic Spectrometry 24, 209−214 boundary. Bulletin of the Geological Society of Finland Mutanen, T. 1989. Koitelainen intrusion and Keivitsa- 79, 175–201. Available at: http://www.geologinenseu- Satovaara complex. 5th International Platinum Sym- ra.fi/bulletin/Volume79/Makkonen_Huhma_2007. posium. Geological Survey of Finland, Guide 28. 49 p. pdf Available at: http://tupa.gtk.fi/julkaisu/opas/op_028. Manninen, T. 1991. Sallan alueen vulkaniitit. Lapin vul- pdf kaniittiprojektin raportti. Volcanic rocks in the Salla

146 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

Mutanen, T. 1997. Geology and ore petrology of the Nykänen, J., Laajoki, K. & Karhu, J. 1997. Geology and Akanvaara and Koitelainen mafic layered intrusions geochemistry of the Early Proterozoic Kortejärvi and and the KeivitsaSatovaara layered complex, northern Laivajoki carbonatites, Central Fennoscandian Shield, Finland. Geological Survey of Finland, Bulletin 395. Finland. Bulletin of the Geological Society of Finland 233 p. Available at: http://tupa.gtk.fi/julkaisu/bulle- 69, 5–30. Available at: http://www.geologinenseura. tin/bt_395.pdf fi/bulletin/Volume69/sgs_bt_069_pages_005_030. Mutanen, T. 2002. Tutkimustyöselostus Sallan kunnassa pdf valtausalueilla Ahmalampi 1 (kaivosrekisterinume- Nykänen, O. 1971. On the Karelides in the Tohmajärvi area, ro 5573/1), Kieskisjärvi 1-2 (kaivosrekisterinumerot eastern Finland. Bulletin of the Geological Society of 5574/1-2) ja Kapitvaara 1-2 (kaivosrekisterinumerot Finland 43, 93–108. Available at: http://www.geolog- 5572/1-2) tehdyistä malmitutkimuksista. Geologian inenseura.fi/bulletin/Volume43/sgs_bt_043_1_pag- tutkimuskeskus, archive report M06/4614/2002/1/10, es_093_108.pdf Salla, Lehtomaa. Available at: http://tupa.gtk.fi/ra- Nykänen, V. 1995. Otanmäen intruusiokompleksin portti/valtaus/m06_4614_2002_1_10.pdf Vuorokkaan lohkon ala- ja keskiosan kiteytyminen ja Mutanen, T. 2003. Magmatismi ja malminmuodostus. kemiallinen evoluutio. Lisensiaatintyö, Oulun yliopis- Loppuraportti toiminnasta 1998–2001. Geological to, Geotieteiden ja tähtitieteen laitos. 135 p. Survey of Finland, archive report M10.4/2003/1. 41 p. Nykänen, V. M., Vuollo, J. I., Liipo, J. P. & Piirainen, T. (in Finnish). Available at: http://tupa.gtk.fi/raportti/ A. 1994. Transitional (2.1 Ga). Fe-tholeiitic-tholeiitic arkisto/m10_4_2003_1.pdf magmatism in the Fennoscandian Shield signifying Mutanen, T. 2011. Alkalikiviä ja appiniitteja. Raportti lithospheric thinning during Palaeoproterozoic ex- hankkeen Magmatismi ja malminmuodostus II to- tensional tectonics. Precambrian Research 70, 45–65. iminnasta 2002 – 2005. 627 p. Geological Survey of O’Brien, H. 2015. Introduction to carbonatite deposits Finland, archive report. 627 p. (in Finnish with Eng- of Finland. In: Maier, W., O’Brien, H. & Lahtinen, R. lish summary). Available at: http://tupa.gtk.fi/raport- (eds) Mineral Deposits of Finland. Amsterdam: Else- ti/arkisto/9_2011.pdf vier, 291–304. Mutanen, T. & Huhma, H. 2001. U-Pb geochronology of O’Brien, H. & Hyvönen, E. 2015. The Sokli carbonatite the Koitelainen, Akanvaara and Keivitsa layered in- complex. In: Maier, W., O’Brien, H. & Lahtinen, R. trusions and related rocks. In: Vaasjoki, M. (ed.) Radi- (eds) Mineral Deposits of Finland. Amsterdam: Else- ometric age determinations from Finnish Lapland and vier, 305–326. their bearing on the timing of Precambrian volcano- O’Brien, H., Huhma, H. & Sorjonen-Ward, P. 1993. Petro- sedimentary sequences. Geological Survey of Finland, genesis of the late Archean Hattu schist belt, Ilomant- Special Paper 33, 229–246. Available at: http://tupa. si, eastern Finland: Geochemistry and Sr, Nd isotopic gtk.fi/julkaisu/specialpaper/sp_033.pdf composition. In: Nurmi, P. & Sorjonen-Ward, P. (eds) Mutanen, T. & Huhma, H. 2003. The 3.5 Ga Siurua trond- Geological development, gold mineralization and ex- hjemite gneiss in the Archaean Pudasjärvi Granulite ploration methods in the late Archean Hattu schist Belt, northern Finland. Bulletin of the Geological So- belt, Ilomantsi, eastern Finland. Geological Survey ciety of Finland 75, 51–68. Available at: http://www. of Finland, Special Paper 17, 145–184. Available at: geologinenseura.fi/bulletin/Volume75/MutanenHuh- http://tupa.gtk.fi/julkaisu/specialpaper/sp_017.pdf ma.pdf O’Brien, H. E., Peltonen, P. & Vartiainen, H. 2005. Kim- Mutanen, T. & Väänänen, J. 2004. PGE-Au-Cu-Ni berlites, carbonatites, and alkaline rocks. In: Lehtin- potential of postkinematic appinitic (1.79 Ga) en, M., Nurmi, P. A. & Rämö, O. T. (eds) Precambrian intrusions in Finland. In: McPhie, J. & McGoldrick, P. Geology of Finland – Key to the Evolution of the Fen- (eds) Dynamic Earth: Past, Present and Future, 17th noscandian Shield. Amsterdam: Elsevier Science B.V., Australian Geological Convention, February 8–13, 605–644. 2004, Tasmania, Australia, Geological Society of Aus- O’Brien, O., Heilimo, E. & Heino, P. 2015. The Archean tralia, Abstracts 73, p. 104. Siilinjärvi carbonatite complex. In: Maier, W., O’Brien, Neilson, J. C., Kokelaar, B. P. & Crowley, Q. G. 2009. Tim- H. & Lahtinen, R. (eds) Mineral Deposits of Finland. ing, relations and cause of plutonic and volcanic ac- Amsterdam: Elsevier, 327–343. tivity of the Siluro-Devonian post-collision magmatic Öhlander, B., Skiöld, T., Elming, S.-Å., BABEL Working episode in the Grampian Terrane, Scotland. Journal of Group, Claesson, S. & Nisca, D. H. 1993. Delineation the Geological Society 166, 545−561. and character of the Archaean-Proterozoic boundary Nironen, M. 2005. Proterozoic orogenic granitoid rocks. in northern Sweden. Precambrian Research 64, 67–84. In: Lehtinen, M., Nurmi, P. A. & Rämö, O. T. (eds) The Paavola, J. 1988. Lapinlahden kartta-alueen kallioperä. Precambrian Geology of Finland – Key to the Evolu- Summary: Pre-Quaternary Rocks of the Lapinlahti tion of the Fennoscandian Shield. Developments in Map-Sheet area. Geological Map of Finland 1:100 000, Precambrian Geology 14. Amsterdam: Elsevier Science Explanation to the Maps of Pre-Quaternary Rocks, B.V., 443–479. Sheet 3332. Geological Survey of Finland. 60 p. Avail- Nironen, M. (ed.) 2017. Bedrock of Finland at the scale 1:1 able at: http://tupa.gtk.fi/kartta/kallioperakartta100/ 000 000 - Major stratigraphic units, metamorphism kps_3332.pdf and tectonic evolution. Geological Survey of Finland, Palmén, J. 1997. Sallan emäksisen intruusioryhmän geo- Special Paper 60. 126 p. Available at: http://tupa.gtk. logiasta. Pro gradu – tutkielma. Helsingin yliopisto, fi/julkaisu/specialpaper/sp_060.pdf geologian ja mineralogian osasto. 68 p. Nironen, M. & Mänttäri, I. 2003. Structural evolution of Papunen, H., Halkoaho, T. & Luukkonen, E. 2009. Ar- the Vuotso area, Finnish Lapland. Geological Society chean evolution of the Tipasjärvi-Kuhmo-Suomus- of Finland, Bulletin 75, 93−101. Available at: http:// salmi greenstone complex, Finland. Geological Survey www.geologinenseura.fi/bulletin/Volume75/Nironen- of Finland, Bulletin 403. 68 p. Available at: http:// Manttari.pdf tupa.gtk.fi/julkaisu/bulletin/bt_403.pdf Nironen, M., Kousa, J., Luukas, J. & Lahtinen, R. (eds) Patchett, J. & Kouvo, O. 1986. Origin of continental crust 2016. Geological Map of Finland 1:1 000 000. Geologi- of 1.9–1.7 Ga age: Nd isotopes and U-Pb zircon ages in cal survey of Finland. Available at: http://tupa.gtk.fi/ the Svecokarelian terrain of south Finland. Contribu- kartta/erikoiskartta/ek_098_300dpi.pdf tions to Mineralogy and Petrology 92, 1–12.

147 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Patchett, P. J., Kouvo, O., Hedge, C. E. & Tatsumoto, M. Puchtel, I. S., Haase, K. M., Hofmann, A. W., Chauvel, C., 1981. Evolution of Continental Crust and Mantle Het- Kulikov, V. S., Garbe-Schönberg, C.-D. & Nemchin, erogeneity: Evidence from Hf Isotopes. Contributions A. A. 1997. Petrology and geochemistry of crustal- to Mineralogy and Petrology 78, 279–297. ly contaminated komatiitic basalts from the Vetreny Pekkarinen, L. J. & Lukkarinen, H. 1991. Paleoproterozoic Belt, southeastern Baltic Shield: Evidence for an ear- volcanism in the Kiihtelysvaara - Tohmajärvi district, ly Proterozoic mantle plume beneath rifted Archean eastern Finland. Geological Survey of Finland, Bulletin continental lithosphere. Geochimica et Cosmochimica 357. 30 p. Available at: http://tupa.gtk.fi/julkaisu/bul- Acta 61, 1205–1222. letin/bt_357.pdf Ranta, J.-P., Lauri, L., Hanski, E., Huhma, H., Lahaye, Peltonen, P. 2005a. Ophiolites. In: Lehtinen, M., Nur- Y. & Vanhanen, E. 2015. U–Pb and Sm–Nd isotopic mi, P. A. & Rämö, O. T. (eds) Precambrian Geology of constraints on the evolution of the Paleoproterozoic Finland – Key to the Evolution of the Fennoscandian Peräpohja Belt, northern Finland. Precambrian Re- Shield. Amsterdam: Elsevier B.V., 237−273. search 266, 246–259. Peltonen, P. 2005b. Svecofennian mafic–ultramafic in- Räsänen, J. & Huhma, H. 2001. U-Pb datings in the So- trusions. In: Lehtinen, M., Nurmi, P. A. & Rämö, O. dankylä schist area, central Finnish Lapland. In: T. (eds) Precambrian Geology of Finland – Key to the Vaasjoki, M. (ed.) Radiometric age determinations Evolution of the Fennoscandian Shield. Amsterdam: from Finnish Lapland and their bearing on the timing Elsevier B.V., 407–442. of Precambrian volcano-sedimentary sequences. Geo- Peltonen, P., Huhma, H., Santaguida, F. & Beresford, S. logical Survey of Finland, Special Paper 33, 153–188. 2014. U-Pb zircon and Sm-Nd isotopic constraints for Available at: http://tupa.gtk.fi/julkaisu/specialpaper/ the timing and origin of magmatic Ni-Cu-PGE de- sp_033.pdf posits in Northern Fennoscandia. SEG 2014: Building Rastas, P., Huhma, H., Hanski, E., Lehtonen, M. I., Här- Exploration Capability for the 21st Century, Keystone, könen, I., Kortelainen, V., Mänttäri, I. & Paakkola, J. Colorado, USA. Electronic Abstract. 1 p. 2001. U-Pb isotopic studies on the Kittilä greenstone Peltonen, P., Kontinen, A. & Huhma, H. 1996. Petrolo- area, central Lapland, Finland. In: Vaasjoki, M. (ed.) gy and geochemistry of metabasalts from the 1.95 Ga Radiometric age determinations from Finnish Lap- Jormua Ophiolite, northeastern Finland. Journal of land and their bearing on the timing of Precambrian Petrology 37, 1359–1383. volcano-sedimentary sequences. Geological Survey of Peltonen, P., Kontinen, A. & Huhma, H. 1998. Petrogen- Finland, Special Paper 33, 95–141. Available at: http:// esis of the mantle sequence of the Jormua ophiolite tupa.gtk.fi/julkaisu/specialpaper/sp_033.pdf (Finland): melt migration in the upper mantle during Richard, P., Shimizu, N. & Allégre, C. J. 1976.134Nd/146Nd, a Palaeoproterozoic continental break-up. Journal of natural tracer: an application to oceanic basalts. Earth Petrology 39, 297–329. Planetary Science Letters, 31, 269–278. Peltonen, P., Kontinen, A., Huhma, H. & Kuronen, U. Rosa, D. R. N., Finch A. A., Andersen T. & Inverno C. M. 2008. Outokumpu revisited: New mineral deposit C. 2009. U-Pb geochronology and Hf isotope ratios of model for the mantle peridotite-associated Cu–Co– magmatic zircons from the Iberian pyrite belt. Min- Zn–Ni–Ag–Au sulphide deposits. Ore Geology Re- erology and Petrology 95, 47–69. views 33, 559–617. Rubatto, D. 2002. Zircon trace element geochemistry: Peltonen, P., Mänttäri, I., Huhma, H. & Kontinen, A. partitioning with garnet and the link between U–Pb 2003. Archean zircons from the mantle: The Jormua ages and metamorphism. Chemical Geology, 184, ophiolite revisited. Geology 31, 645–648. 123–138. Perttunen, V. & Hanski, E. 2003. Törmäsjärven ja Koivun Rudnick, R. L. & Gao, S. 2003. Composition of the conti- kartta-alueiden kallioperä. Summary: Pre-Quaterna- nental crust. In: Rudnick, R. L. (ed.) The Crust. Trea- ry Rocks of the Koivu and Törmäsjärvi Map-Sheet ar- tise on Geochemistry, Volume 3. Oxford: Elsevier, eas. Geological Map of Finland 1:100 000, Explanation 1–64. to the Maps of Pre-Quaternary rocks, Sheets 3631 and Salminen, J., Halls H. C., Mertanen S., Pesonen L. J., 2633. Geological Survey of Finland. 88 p. (in Finnish Vuollo J. & Söderlund U. 2014. Paleomagnetic and with English summary). Available at: http://tupa.gtk. geochronological studies on Paleoproterozoic dia- fi/kartta/kallioperakartta100/kps_2631_2633.pdf base dykes of Karelia, East Finland—Key for testing Perttunen, V. & Vaasjoki, M. 2001. U-Pb geochronology the Superia supercraton. Precambrian Research 244, of the Perääpohja Schist Belt, northwestern Finland. 87–99. In: Vaasjoki, M. (ed.) Radiometric age determinations Santaguida, F., Luolavirta, K., Lappalainen, M., Ylinen, from Finnish Lapland and their bearing on the tim- J., Voipio, T. & Jones, S. 2015. The Kevitsa Ni-Cu-PGE ing of Precambrian volcano-sedimentary sequences. Deposit in the Central Lapland Greenstone Belt in Fin- Geological Survey of Finland, Special Paper 33, 45–84. land. In: Maier, W., O’Brien, H. & Lahtinen, R. (eds) Available at: http://tupa.gtk.fi/julkaisu/specialpaper/ Mineral Deposits of Finland. Amsterdam: Elsevier, sp_033.pdf 195–210. Puchtel, I. S., Arndt, N. T., Hofmann, A. W., Haase, K. Schoene, B., Crowley J. L., Condon D. J., Schmitz M. D. M., Kröner, A., Kulikov, V. S., Kulikova, V. V., Garbe- & Bowring, S. A. 2006. Reassessing the uranium de- Schönberg, C.-D. & Nemchin, A. A. 1998. Petrology of cay constants for geochronology using ID-TIMS U–Pb mafic lavas within the Onega plateau, central Kare- data. Geochimica et Cosmochimica Acta 70, 426–445. lia: Evidence for the 2.0 Ga plume-related continental Silvennoinen, A. 1972. On the stratigraphic and struc- crustal growth in the Baltic Shield. Contributions to tural geology of the Rukatunturi area, northeastern Mineralogy Petrology 130, 134–153. Finland. Geological Survey of Finland, Bulletin 257. Puchtel, I. S., Brügmann, G. E., Hofmann, A. W., Kulikov, 48 p. Available at: http://tupa.gtk.fi/julkaisu/bulletin/ V. S. & Kulikova, V. V. 2001. Os-isotope systematics of bt_257.pdf komatiitic basalts from the Vetreny belt, Baltic Shield: Silvennoinen, A. 1991. Kuusamon ja Rukatunturin kartta- evidence for a chondritic source of the 2.45 Ga plume. alueiden kallioperä. Summary: Pre-Quaternary Rocks Contributions to Mineralogy and Petrology 140, 588– of the Kuusamo and Rukatunturi Map-Sheet areas. 599. Geological Map of Finland 1: 100 000, Explanation to

148 Geological Survey of Finland, Bulletin 405 Sm–Nd and U–Pb isotope geochemistry of the Palaeoproterozoic mafic magmatism in eastern and northern Finland

the Maps of Pre-Quaternary Rocks, Sheets 4524+4542 dian carbonatites. Abstracts of the 17th Annual V.M. and 4613. Geological Survey of Finland. 62 p. Avail- Goldschmidt Conference, Cologne, Germany, August able at: http://tupa.gtk.fi/kartta/kallioperakartta100/ 2007. Geochimica et Cosmochimica Acta 71, p. A1032. kps_4524_4542_4613.pdf Tulenheimo, T. 1999. Kuhmon Kellojärven kerrokselli Skiöld, T., Öhlander, B., Vocke, R. D. & Hamilton, P. J. nen ultramafinen muodostuma. Pro gradu -tutkielma, 1988. Chemistry of proterozoic orogenic processes at Turun yliopisto. 199 p. a continental margin in Northern Sweden. Chemical Tyni, M., Puustinen, K., Karhu, J. & Vaasjoki, M. 2003. Geology, 69, 193–207. The Petäiskoski carbonate veins at Juuka, eastern Fin- Smolkin, V. F., Hanski, E., Huhma, H. & Fedotov, Zh. A. land. In: Autio, S. (ed.) Geological Survey of Finland, 2015. Sm-Nd and U-Pb isotopic study of the Nyasyuk- Current Research 2001–2002. Geological Survey of ka dike complex, Kola Peninsula, Russia. Transactions Finland, Special Paper 36, 13–16. Available at: http:// of Karelian Research Centre of Russian Academy of tupa.gtk.fi/julkaisu/specialpaper/sp_036.pdf Science, No. 7, Precambrian Geology Series, 74−84. Tyrväinen, A. 1983. Sodankylän ja Sattasen kartta-aluei- Söderlund, U., Patchett, P. J., Vervoort, J. & Isachsen, den kallioperä. PreQuaternary Rocks of the Sodankylä C. E. 2004. The 176Lu decay constant determined by and Sattanen MapSheet areas. Geological Map of Fin- Lu–Hf and U–Pb isotope systematics of Precambrian land 1:100 000, Explanation to the Maps of PreQuater- mafic intrusions. Earth and Planetary Science Letters nary Rocks, Sheets 3713 and 3714. Geological Survey of 219, 311–324. Finland. 59 p. (in Finnish, English summary). Avail- Stacey, J. S. & Kramers, J. D. 1975. Approximation of ter- able at: http://tupa.gtk.fi/kartta/kallioperakartta100/ restial lead isotope evolution by a two-stage model. kps_3713_3714.pdf Earth and Planetary Science Letters 26, 207–221. Väänänen, J. 2004. Sieppijärven ja Pasmajärven kartta- Stepanov, V. S. 1994. Magmatiti Pjaozerskovo blocka alueiden kallioperä. Summary: Pre-Quaternary Rocks (petrohimisekie osobennosti I postedovatelnost obra- of the Sieppijärvi and Pasmajärvi Map-Sheet areas. zovania kompleks). (Magmatic rocks of the Pjaozerskij Geological Map of Finland 1:100 000, Explanation block (petrochemical characteristics and formation of to the Maps of Pre-Quaternary Rocks, Sheets 2624 the complex)). In: Stepanov, V. S. (ed.) Dokembrii and 2642. Geological Survey of Finland. 55 p. Avail- Severnoi Karelii (Precambrian of the Northern Kare- able at: http://tupa.gtk.fi/kartta/kallioperakartta100/ lia). Petrozavodsk: Russian Academy of Sciences, 118– kps_2624_2642.pdf 170. (in Russian) Vaasjoki, M. 1971. Tshohkoaivin alueen geologiasta. Stepanova, A. V., Sal’nikova, E. B., Samsonov, A. V., La- Unpublished M.Sc. Thesis, Department of Geology, rionov Yu. O. & Stepanov, S. 2014a. 2.3 Ga intraplate University of Helsinki. 43 p. (in Finnish). magmatism on the Karelian Craton: Implications for Vaasjoki, M. (ed.) 2001. Radiometric age determinations the problem of “endogenic shutdown” in the Paleo- from Finnish Lapland and their bearing on the tim- proterozoic. Doklady Earth Sciences 457, 965–970. ing of Precambrian volcano-sedimentary sequences. Stepanova, A. V., Samsonov, A. V., Salnikova, E. B., Puch- Geological Survey of Finland, Special Paper 33. 279 p. tel, I. S., Larionova, Yo. O., Larionov, A. N., Stepanov, Available at: http://tupa.gtk.fi/julkaisu/specialpaper/ V. S., Shapovalov, Y. B. & Egorova, S. V. 2014b. Palae- sp_033.pdf oproterozoic continental MORB-type tholeiites in the Vaasjoki, M., Kärki, A. & Laajoki, K. 2001. Timing of Pa- Karelian Craton: Petrology, geochronology, and tec- leaeoproterozoic crustal shearing in the central Fen- tonic Setting. Journal of Petrology 55, 1719–1751. noscandian Shield according to U-Pb data from asso- Talvitie, J. & Paarma, H. 1980. Precambrian basic mag- ciated granitoids, Finland. Bulletin of the Geological matism and the Ti-Fe ore formation in central and Society of Finland 73, 87–101. Available at: http:// northern Finland. In: Siivola, J. (ed.) Metallogeny of www.geologinenseura.fi/bulletin/Volume73/Vaasjo- the Baltic Shield. Proceedings of the symposium held kietal.pdf in Helsinki, Finland, June 12–21, 1978. Geological Vartiainen, H. & Woolley, A. R. 1974. The age of the Sokli Survey of Finland, Bulletin 307, 98–107. Available at: carbonatite, Finland, and some elationships of the http://tupa.gtk.fi/julkaisu/bulletin/bt_307.pdf North Atlantic alkaline igneous province. Bulletin of Toivola, V. 1988. Sonkajärven-Varpaisjärven alueen the Geological Society of Finland 46 (1), 81−91. Avai- diabaasit. Pro-gradu tutkielma, Helsingin yliopisto. lable at: http://www.geologinenseura.fi/bulletin/Vo- 73 p. lume46/sgs_bt_046_1_pages_081_091.pdf Toivola, V., Huhma, H. & Paavola, J. 1991. The diabase Vogel, D. C., Vuollo, J. I., Alapieti, T. T. & James, R. S. dykes in the Sonkajärvi - Varpaisjärvi area, central 1998. Tectonic, stratigraphic, and geochemical com- Finland. In: Autio, S. (ed.) Geological Survey of Fin- parisons between ca. 2500–2440 Ma mafic igneous land, Current Research 1989–1990. Geological Sur- events in the Canadian and Fennoscandian Shields. vey of Finland, Special Paper 12, 59–61. Available at: Precambrian Research 92, 89–116. http://tupa.gtk.fi/julkaisu/specialpaper/sp_012.pdf Vuollo J. & Fedotov, Zh. A. 2005. Fifth International Dyke Tolstikhin, I. N., Dokichaeva, V. S., Kamensky, I. L. & Conference “Dyke swarms - time markers of crustal Amelin Yu. V. 1992. Juvenile helium in ancient rocks. evolution”, Pohtimolampi Wilderness Hotel, Polar II. U-He, K-Ar, Sm-Nd and Rb-Sr systematics in Circle, Rovaniemi, Finland 31 June - 3 August, 2005: the Monche pluton: 3He/4He ratios frozen in urani- Post-conference field trip B guidebook : Eastern and um-free ultramafic rocks. Geochimica et Cosmochi- Northern Finland - Belomorian Mobile Belt and Kola mica Acta 56, 987–999. Peninsula, 3 August - 9 August, 2005. 84 p. Available Törmänen, T., Iljina, M., Heikura, P. & Salmirinne, H. at: http://idc5.gsf.fi/IDC5_post_excursion.pdf 2007. Tutkimustyöselostus Savukosken kunnassa Vuollo, J. & Huhma, H. 2005. Palaeoproterozoic mafic valtausalueilla Värriö 1–6 (Kaivosrekisterinumerot dykes in NE Finland. In: Lehtinen, M., Nurmi, P. A. 7910/1-7910/6) suoritetuista tutkimuksista. 13 s., 12 & Rämö, O. T. (eds) Precambrian Geology of Finland liites. Geological Survey of Finland, archive report – Key to the Evolution of the Fennoscandian Shield. M06/4712/2007/10/75. Available at: http:// tupa.gtk. Amsterdam: Elsevier Science B.V., 193–235. fi/raportti/valtaus/m06_4712_2007_10_75.pdf Vuollo, J., Piirainen, T. & Huhma, H. 1992. Two Early Torppa, A. & Karhu, J. 2007. Ancient subduction recorded Proterozoic tholeiitic diabase dyke swarms in the Ko- in the isotope characteristics of ~1.8 Ga Fennoscan- li-Kaltamo area, eastern Finland. Geological Survey of

149 Geological Survey of Finland, Bulletin 405 Hannu Huhma, Eero Hanski, Asko Kontinen, Jouni Vuollo, Irmeli Mänttäri and Yann Lahaye

Finland, Bulletin 363. 32 p. Available at: http://tupa. Woodard, J., Kietäväinen, R. & Eklund, O. 2014. Svecofen- gtk.fi/julkaisu/bulletin/bt_363.pdf nian post-collisional shoshonitic lamprophyres at the Walker, R. J. & Nisbet, E. 2002. 187Os isotopic constraints margin of the Karelia Craton: implications for mantle on Archean mantle dynamics. Geochimica et Cosmo- metasomatism. Lithos 205, 379–393. chimica Acta 66, 3317–3325. Yang, S. H., Hanski, E., Li, C., Maier, W. D., Huhma, H., Wasserburg, G. J., Jacobsen, S. B., DePaolo, D. J., Mc- Mokrushin, A. V., Latypov, R., Lahaye, Y., O’Brien, H. Culloch, M. T. & Wen, T. 1981. Precise determination & Qu, W. J. 2016. Mantle source of the 2.44–2.50-Ga on Sm/Nd ratios, Sm and Nd isotopic abundances in mantle plume-related magmatism in the Fennoscan- standard solutions. Geochimica et Cosmochimica Acta dian Shield: evidence from Os, Nd and Sr isotope com- 45, 2311–2324. positions of the Monchepluton and Kemi intrusions. Watson, E. B. 1982. Basalt contamination by continental Mineralium Deposita 51, 1055–1073. crust: some experiments and models. Contributions to Yang, S.-H., Maier, W.D., Hanski, E., Lappalainen, M., Mineralogy and Petrology 56, 119–139. Santaguida, F. & Määttä, S. 2013. Origin of ultra- Whitehouse, M. J., Claesson, S., Sunde, T. & Vestin, J. nickeliferous olivine in the Kevitsa Ni-Cu-PGE-min- 1997. Ion-microprobe U–Pb zircon geochronology eralised intrusion, Lapland, Finland. Contributions to and correlation of Archaean gneisses from the Lew- Mineralogy and Petrology 166, 81–95. isian Complex of Gruinard Bay, north-west Scotland. Zozulya, D. R., Bayanova, T. B. & Serov, P. N. 2007. Age Geochimica et Cosmochimica Acta 61, 4429–4438. and isotopic geochemical characteristics of Archaean Wiedenbeck, M., Alle, P., Corfu, F., Griffin, W. L., Meier, carbonatites and alkaline rocks of the Baltic Shield. M. & Oberli, F. et al. 1995. Three natural zircon stand- Doklady Earth Sciences 415A (6), 874–879. ards for U–Th–Pb, Lu–Hf, trace element and REE analysis. Geostandards Newsletter 19, 1–23. Woodard, J. & Huhma, H. 2015. Paleoproterozoic mantle enrichment beneath the Fennoscandian Shield: Iso- topic insight from carbonatites and lamprophyres. Lithos 236–237, 311–323.

150 All GTK’s publications online at hakku.gtk.fi

Mafic igneous rocks have been a major focus of isotope research since the establishment of the isotope laboratory at the Geological Survey of Finland in the 1960’s. This volume presents abundant isotope data on Palaeoproterozoic mafic rocks of the Karelia Province in eastern and northern Finland. The previously unpublished data consist of ca. 400 Sm-Nd analyses on 80 mafic rock formations and ca. 1700 U-Pb analyses by TIMS, SIMS and LA-MC-ICP-MS on 80 samples. These results together with published material provide tools for constraining the age and petrogenesis of mafic magmas and the overall geological evolution in the Fennoscandian Shield.

ISBN 978-952-217-394-2 (PDF) ISSN 0367-522X (print) gtk.fi ISSN 2489-639X (online)