40Ar/39Ar chronology and paleomagnetism of Quaternary basaltic lavas from the Perşani Mountains (East Carpathians)

Accepted Manuscript

40Ar/39Ar chronology and paleomagnetism of Quaternary basaltic lavas from the

Perşani Mountains (East Carpathians)

C.G. Panaiotu, B.R. Jicha, B.S. Singer, A. Ţugui, I. Seghedi, A.G. Panaiotu, C.

Necula

PII: S0031-9201(13)00084-8

DOI: http://dx.doi.org/10.1016/j.pepi.2013.06.007

Reference: PEPI 5629

To appear in: Physics of the Earth and Planetary Interiors

Received Date: 3 February 2013

Revised Date: 13 June 2013

Accepted Date: 19 June 2013

Please cite this article as: Panaiotu, C.G., Jicha, B.R., Singer, B.S., Ţugui, A., Seghedi, I., Panaiotu, A.G., Necula,

C., 40Ar/39Ar chronology and paleomagnetism of Quaternary basaltic lavas from the Perşani Mountains (East

Carpathians), Physics of the Earth and Planetary Interiors (2013), doi: http://dx.doi.org/10.1016/j.pepi.2013.06.007

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers

we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and

review of the resulting proof before it is published in its final form. Please note that during the production process

errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

40Ar/

39Ar chronology and paleomagnetism of Quaternary basaltic lavas from the Perşani 1

Mountains (East Carpathians) 2

3

Panaiotu, C. G.1*

, Jicha, B.R.2, Singer, B.S.

2, Ţugui, A.

3, Seghedi, I.

4, Panaiotu, A.G.

1, Necula, C.

1 4

1University of Bucharest, Paleomagnetic Laboratory, Bălcescu 1, 010041 Bucharest, Romania 5

([email protected]; [email protected]; [email protected]) 6 2Department of Geoscience, University of Wisconsin-Madison, 1215 W. Dayton St., Madison, WI, 53706, USA 7

([email protected]; [email protected]). 8 3 National Institute for Earth Physics, National Institute for Earth Physics, Călugăreni 12, 077125 Măgurele, Romania 9

([email protected]) 10 4Institute of Geodynamics, Romanian Academy, Jean-Luis Calderon 19-21, 020032, Bucharest, Romania 11

([email protected]). 12

13

Abstract 14

Quaternary volcanism in the Perşani Mountains forms an Na-alkali basaltic province inside the bend 15

area of the Carpathians in the southeastern part of Europe. Previous K-Ar ages and paleomagnetic 16

data reveal several transitional virtual geomagnetic poles, which were tentatively associated with the 17

Cobb Mountain subchron and a Brunhes chron excursion. We report a new paleomagnetic and rock-18

magnetic study coupled with 40

Ar/39

Ar geochronology to better constrain the age of geomagnetic 19

reversals or excursions that might be recorded and the timing of volcanism. Of the paleomagnetic 20

directions obtained from sampled lava flows 4 are reversed polarity, 19 are normal polarity and 16 21

have transitional polarity. 40

Ar/39

Ar plateau ages determined from incremental heating experiments 22

on groundmass indicate that two of the reversely magnetized lavas erupted at 1142 ± 41 and 800 ± 25 23

ka, four of the normally magnetized lavas erupted at 1060 ± 10, 1062 ± 24, 684 ± 21, and 683 ± 28 24

ka, and two transitionally magnetized lavas formed at 1221 ± 11 and 799 ± 21 ka. Both the new 25

40Ar/

39Ar ages and the paleomagnetic data suggest at least five episodes of volcanic activity with the 26

most active periods during the Jaramillo and Brunhes chrons. This results shows that the last phases 27

of alkalic and calc-alkaline magmatism in the South-East Carpathians were contemporaneous. The 28

age of the older transitionally magnetized lava flow is within error of recent unspiked K-Ar and 29

astrochronologic ages for the reversal that defines the onset of the Cobb Mountain normal polarity 30

2

subchron. The age of the younger transitional lava is similar to that of an excursion that preceded the 31

Matuyama-Brunhes polarity reversal and which has come to be known as the Matuyama-Brunhes 32

precursor. Omitting the excursion data, the dispersion of the virtual geomagnetic poles (around 19°) 33

is larger than the expected value around 45°N from the global compilation, but closer to the value 34

obtained only from the Time Averaged geomagnetic Field Initiative studies. 35

36

Keywords: Perşani Mountains, East Carpathians, Quaternary, basalts, 40

Ar/39

Ar dating, 37

paleomagnetism. 38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

* corresponding author: tel.: +040740060659; e-mail: [email protected] 54

55

3

1. Introduction 56

Na-alkalic basaltic volcanism in the Perşani Mountains, although of modest extent (176 km2), 57

represents an important Quaternary alkali basaltic province inside the Carpathians and south-eastern 58

Europe (Seghedi et al., 2004b). Its evolution is intimately connected with the last phase of tectonic 59

deformation in the bending area of the East Carpathians (Seghedi et al., 2011), which is the locus of 60

most active intermediate-depth seismicity (e.g., Popa et al., 2012). 61

Previous K-Ar ages and paleomagnetic data (Panaiotu et al., 2004) constrain the timing of 62

volcanism between 1.5 Ma and 0.5 Ma. This study suggested that the volcanism took placed in two 63

phases: one between 1.5 – 1.2 Ma, but mainly around 1.2 Ma, and a second around 0.6 Ma. 64

Paleomagnetic data reveals several transitional virtual geomagnetic poles (VGP) that were associated 65

with the Cobb Mountain subchron and an excursion within the Brunhes chron based on K-Ar ages. 66

Although, the uncertainties on the K-Ar ages are ± 80-200 ka. 67

Important refinements of the geomagnetic timescale for the Matuyama and Brunhes chrons have 68

been made in the last decade (Singer et al., 2004, 2008; 2013; Channell and Guyodo, 2004). Despite 69

recent advances in both the 40

Ar/39

Ar dating of lava flows that record transitional magnetization 70

directions and astronomical dating of marine sediment, several geomagnetic excursions and polarity 71

reversals have only been dated in one or two localities. For example, only two localities are known to 72

record transitional magnetization directions associated with the Cobb Mountain normal polarity 73

subchron: the Alder Creek rhyolite at Cobb Mountain, California (Mankinnen et al., 1978; Turrin et 74

al., 1994; Nomade et al., 2005) and two mafic flows on São Jorge Island, Azores (Silva et al, 2012). 75

The highest resolution record of the Cobb Mountain subchron is from sediment cored at ODP sites 76

983 and 984 where the continuous magnetostratigraphy and an O isotope-based astrochronologic age 77

model have been obtained by Channell et al. (2002). The reverse-normal and normal-reverse polarity 78

transitions that bound the Cobb Mountain subchron in these cores show that virtual geomagnetic 79

4

poles cross the equator at 1215 and 1190 ka coincident with a sharp drop and rise in paleointensity 80

together indicating a duration of >25 kyr (Channell et al., 2002). 81

To constrain the age of geomagnetic reversals or excursions that might be recorded we have 82

measured for the first time 40

Ar/39

Ar ages from several lava flows in the Perşani Mountains volcanic 83

field. New palaeomagnetic measurements were also conducted to enhance the number of sites and 84

the quality of the directional measurements. 85

86

2. Geological setting 87

In the eastern Carpathian–Pannonian region during the last 15 Ma, westward-dipping subduction 88

in a land-locked basin caused collision of a lithospheric block from the west with the southeastern 89

border of the European plate (e.g. Seghedi et al. 2004a; van Hinsbergen et al. 2008; Seghedi and 90

Downes, 2011). After the main collisional events at 11 Ma (Maţenco et al., 2007), calc-alkaline post-91

collisional volcanism took place in the East Carpathians forming the Călimani – Gurghiu – Harghita 92

(CGH) volcanic chain (e.g. Szakacs and Seghedi, 1995; Mason et al., 1998; Seghedi et al., 1998). 93

This volcanic chain is around 160 km long and the volcanic activity gradually migrated to the south 94

between the Miocene (~11 Ma) and the Quaternary (~0.23 Ma) (e.g., Pécskay et al. 1995). The last 95

phase of volcanic eruptions during Pliocene–Quaternary (Pécskay et al., 2006) occurred in the South 96

Harghita Mountains, which form the southern end of the CGH volcanic chain, and the Perşani 97

Mountains (Fig. 1). This volcanism was coeval with the last peak of crustal deformation in the 98

Carpathian bending zone (Merten et al. 2010) and the formation of the Braşov Basin (Gîrbacea et al., 99

1998). The deformation was coeval with uplift in the orogen and subsidence in the foreland (Moesian 100

platform) with similar amplitudes in the order of 2–4 km (Maţenco et al. 2010). The bending area of 101

the East Carpathians is characterized by significant intermediate and crustal seismic activity in the 102

Vrancea area (Fig. 1), which is accompanied by crustal seismicity in the Braşov Basin below the 103

Quaternary volcanic areas (Popa et al., 2012). 104

5

Na-alkalic mafic volcanism from the Perşani Mountains along the western margin of the Braşov 105

Basin (Fig. 1) occurs at several Quaternary vents that cross the south-eastern part of the 106

Transylvanian basin in the interior of the bend area of the East Carpathians (Seghedi and Szakacs, 107

1994; Downes et al., 1995; Panaiotu et al., 2004). It is situated 40 km west of the NW-SE-oriented 108

Pliocene-Quaternary calc-alkaline volcanism in the South Harghita Mountains (Szakács and Seghedi, 109

1995; Seghedi et al, 2011). The Na-alkalic volcanic centers are arranged parallel to a ~ NE-SW 110

normal fault system with the same orientation as the main normal faults of the Braşov basin (Ciulavu 111

et al., 2000; Gîrbacea et al., 1998). According to previous studies, Perşani Mts. volcanism took place 112

between 1.2-1.5 Ma and ca. 0.6 Ma as small monogenetic volcanoes or fields with maars, scoria 113

cones and lava flows in an area of ca. 22 x 8 km (Seghedi and Szakács, 1994; Panaiotu et al., 2004). 114

In the first stage, each volcano started with phreatic/phreatomagmatic explosive activity followed by 115

a less energetic explosive strombolian and lastly as effusive. A number of volcanoes were generated 116

(Sărata, Racoş, Mateiaş, Măguricea monogenetic volcanoes and Bârc volcanic area); Racoş volcano, 117

one of the best examples, developed in a short time span at ~ 1.2 Ma. It started with 118

phreatic/phreatomagmatic explosive activity followed by strombolian and effusive lavas that extruded 119

at the base of the strombolian cone, cutting parts of the previous phreatic/phreatomagmatic ring-120

deposits (Seghedi and Szakács, 1994; Lexa et al, 2010). Volcanism in the Bârc volcanic area and was 121

less voluminous, and was dominantly strombolian and effusive (Panaiotu et al., 2004). Thin lake 122

sediments with occasional tephra intercalations or palesoil separate the above mentioned sequences in 123

this area. Due to the vegetation cover and the lack of regional reference levels it is not possible to 124

establish a relative stratigraphy of the volcanic units in the Perşani Mountains. Na-alkalic basaltic rocks 125

from Perşani Mts. are similar to other continental intraplate alkali basalts, but have distinctively 126

higher La/Nb and lower Ce/Pb ratios than similar Na-alkaline basalts in the Pannonian Basin, 127

suggesting the involvement of a subduction-related component in their source region (Embey-Isztin 128

and Dobosi, 1995; Downes et al., 1995; Seghedi et al., 2004a, b). 129

6

130

3. Methods 131

3.1 Paleomagnetic methods 132

Oriented cores were collected at 42 sites (~6-10 cores per site) (Table 1, Fig. 2). The samples were 133

oriented using a Brunton magnetic compass and a sun compass where possible. Several localities 134

(Racoş Quarry, Bârc Quarry and Bogata Quarry) reported in the previous study (Panaiotu et al., 2004) 135

as recording transitional paleomagnetic directions were sampled in more detail to better facilitate 136

detection of potentially rapid changes of the geomagnetic field. Based on magnetic polarity 8 sites 137

were also sampled for 40

Ar/39

Ar dating. The location of sampling sites is presented in supplementary 138

data as a kml file. 139

Laboratory analyses were carried out in the Palaeomagnetic Laboratory at the University of 140

Bucharest. Standard paleomagnetic specimens (11 cm3) were cut from each core. Remanent 141

magnetizations were measured using a JR5 spinner magnetometer (AGICO). Alternating-field (AF) 142

demagnetization was done using a LDA-3A - AF demagnetizer (AGICO). AF demagnetization was 143

performed in steps from 0 to maximum 100 mT, with 7 to 10 steps per specimen. Thermal 144

demagnetization was performed with a Thermal demagnetizer TD 700 (Magnon International). The 145

susceptibility variation upon thermal treatment was measured on MFK1-A kappabridge (AGICO). 146

Thermal demagnetization was performed in 50°C or 25°C steps (depending on magnetic mineralogy) 147

between room temperature and the maximum unblocking temperature. Demagnetization data were 148

plotted on orthogonal demagnetization diagrams (Zijderveld, 1967), and magnetization components 149

were isolated by principal component analysis (Kirschvink, 1980) using the Remasoft 3.0 software 150

(Chadima and Hrouda, 2006). The line fits were based on the following constraints: 1. a minimum of 151

4 demagnetization steps; 2. the line fit was anchored to the origin; 3. the maximum angular deviation 152

was <5°. 153

7

Several rockmagnetic properties were measured for minimum a sample from each site to 154

determine the magnetic mineralogy. The hysteresis properties were measured at room temperature 155

using a VSM model 3900 (Princeton Measurements) with a maximum applied field of 1 T. The 156

saturation magnetization (Ms), saturation remanent magnetization (Mrs) and coercive force (Bc) 157

values were calculated after correction for the paramagnetic contribution. The coercivity of 158

remanence (Bcr) and the ratio between isothermal remanent magnetization at 300 mT and Mrs (S 159

ratio) were determined by applying a progressively increasing backfield after saturation. First-order 160

reversal curve (FORC) diagrams were measured using the same instrument and processed with the 161

FORCinel 1.18 software (Harrison and Feinberg 2008). 162

To further understand the magnetic properties the field dependence of the magnetic 163

susceptibility between 2 A/m and 700 A/m was determined using the MFK1A kappabridge. These 164

curves were characterized using the V parameter (Hrouda et al., 2006). This parameter is define as 165

V=100(k700-k50)/k50, where k50 and k700 are the susceptibilities measured at 50 and 700 A/m, 166

respectively. The temperature dependence of magnetic susceptibility was measured with a CS-L 167

apparatus from liquid nitrogen temperature to room temperature and with CS3 apparatus from room 168

temperature to 700°C. Both instruments were coupled with the MFK1A kappabridge. The heating-169

cooling cycle above room temperature was performed in argon atmosphere. 170

3.2 40

Ar/39

Ar Methods 171

40Ar/

39Ar experiments were undertaken on eight basaltic lavas from the Perşani Mountains. 172

Sample preparation and irradiation procedures follow Jicha et al. (2011). At the University of 173

Wisconsin-Madison Rare Gas Geochronology Laboratory, ~ 200 mg groundmass packets were 174

incrementally heated in a double-vacuum resistance furnace attached to a 300 cm3 gas clean-up line. 175

Prior to sample introduction, furnace blanks were measured at 100 °C increments throughout the 176

temperature range spanned by the incremental heating experiment and interpolated. Following blank 177

analyses, samples were degassed at 550 °C for 60 minutes to remove atmospheric argon. Fully 178

8

automated experiments consisted of 8-10 steps from 725-1280 °C; each step included a two-minute 179

increase to the desired temperature that was maintained for 15 minutes, followed by an additional 15 180

minutes for gas cleanup. The gas was cleaned during and after the heating period with three SAES 181

C50 getters, two of which were operated at ~450 °C and the other at room temperature. Argon 182

isotope analyses were done using a MAP 215-50 mass spectrometer using a single Balzers SEM-217 183

electron multiplier, and the isotopic data were reduced using ArArCalc software version 2.4 184

(http://earthref.org/ArArCALC/). Atmospheric argon was measured 6–10 times prior to, and 185

following each furnace incremental heating experiment. Measured 40

Ar/36

Ar ratios of atmospheric 186

argon were normalized to 40

Ar/36

Ar = 295.5 (Steiger and Jäger, 1977). 187

188

4. Results 189

4.1 Rock magnetics 190

Low-field variation of magnetic susceptibility of investigated samples is presented in Fig. 3A. 191

The shapes of the magnetic susceptibility curves suggest a magnetic mineralogy from magnetite with 192

no field dependence to titanomagnetite with increasing Ti content and field dependence (e.g. Hrouda 193

et al., 2006). The values of V parameter range from virtually zero to around 90% (Fig. 3B). Most of 194

samples have S ratio larger than 0.9 in agreement with mineralogy dominated of titanomagnetite (Fig. 195

3B). This magnetic mineralogy is also reflected in the variation of magnetic susceptibility with 196

temperature, which indicates a large range of Curie temperatures from around 150°C to 580°C (Fig. 197

4A and 4B). Samples with low V values (< 10%) have Curie temperatures above 500°C compatible 198

with presence of Ti-poor titanomagnetites (e.g. Lattard et al., 2006, Zhao et al., 2008). Samples with 199

higher V values show magnetic susceptibility - temperature curves suggesting both the presence of 200

Ti-rich titanomagnetites with Curie temperatures below 250°C and Ti-intermediate and poor 201

titanomagnetites with Curie temperatures above 400°C or only Ti-rich titanomagnetites. Low 202

temperature variation of magnetic susceptibility is also compatible with this magnetic mineralogy 203

9

with samples dominated by Ti-poor titanomagnetite showing the Verwey transition (Lattard et al., 204

2006). 205

FORC diagrams evolve with V values. Samples with low V values dominated by Ti-poor 206

titanomagnetite (Fig. 4C) display distinct closed-contour peaks around 20 mT, which represent the 207

signature of SD grains, and divergent contours that intersect the HU axis suggesting a contribution of 208

the MD and/or PSD grains. (Roberts et al., 2000, Muxworthy and Dunlop, 2002). If Ti-rich 209

titanomagnetite grains are dominant (Fig. 4D) FORC diagrams have no central peak and show only 210

divergent contours compatible with MD and/or PSD grains. Because these contours do not tend to 211

spread broadly parallel to the Hu axis these types of FORC diagrams suggest that PSD grains are 212

dominant (Muxworthy and Dunlop, 2002). Magnetic granulometry observed in FORC diagrams is 213

also reflected (Fig. 5) in the Day plot (Day et al. 1977). Majority of samples are distributed in the 214

PSD region. This distribution is delimited by theoretical SD-MD mixing curves for magnetite and 215

SD-MD mixing curves for TM60 (Dunlop, 2002) reflecting also the variable magnetic mineralogy 216

observed in our samples. 217

4.2 Paleomagnetic results 218

Natural remanent magnetization and demagnetization behavior were measured on a total of 219

247 independent samples. At least two pilot specimens from each site were subjected to AF and 220

thermal demagnetization. For the majority of the lava flows thermal and AF demagnetization were 221

equally effective for determining the characteristic remanent magnetization of the samples (Fig. 6). 222

When magnetic mineralogy of samples was dominated by Ti-rich titanomagnetite grains AF 223

demagnetization was more effective. Thus, AF demagnetization was the preferred technique to 224

analyze the rest of the collection. A primary characteristic component was unambiguously identified 225

in all analyzed samples. In some samples soft components, considered viscous overprints, were 226

observed, but they were usually removed at demagnetization steps below 20 mT. The mean site 227

directions were obtained by averaging the AF and thermal results using Fisher statistics (Fisher 1953). 228

10

Summary of site mean directions from the Perşani Mountains is presented in Table 1 and plotted in 229

Fig. 7A. All sites passed site quality selection criteria of Johnson et al. (2008): n (number of samples) 230

is at least 5 and k (precision parameter from Fisher statistics) > 50. Most of sites (34) pass also the 231

more stringent criteria of Tauxe et al. (2003), which demand k > 100. All sites with k > 50 have also 232

α95 < 10°being also in agreement with cut-off value for paleosecular variation studies proposed by 233

Opdyke et al. (2010). When multiple sites, sampled in the same lava flow, gave similar directions we 234

computed a mean lava flow direction (Table 1), which was used for further analyses. In case of 235

resampled sites there is a very good agreement between our new site mean directions and those 236

presented by Panaiotu et al. (2004). 237

The VGPs are presented in Fig. 7B. Transitional data are often defined as measurements with 238

absolute value of VGP latitude typically ranging from 45° to 55° (e.g. Tauxe et al., 2003) or can be 239

selected using the Vandamme (1994) algorithm for VGP latitude cut-off. All the VGPs from the 240

Racoş Quarry can be define as transitional VGPs because their latitudes are less than 45°. The rest of 241

VGPs have absolute latitudes above 55°, with one exception the mean VGP from the Bogata Quarry 242

(Fig. 3B). Using the Vandamme (1994) criterion resulted in a cut-off value of 42.7°, which include 243

slightly the VGP from Bogata Quarry. However if several low-VGP-latitude sites are present the 244

Vandamme algorithm will converge without removing these sites (Johnson et al., 2008). We consider 245

that the VGP from the Bogata Quarry, with latitude around 50°S, can be classified as transitional 246

(e.g., Brown et al., 2004; Leonhardt and Fabian, 2007; Leonhardt et al., 2009). The majority of sites 247

(19) are within the secular variation limits and have normal polarity; only 4 sites have reversed 248

polarity (Fig. 7B). 249

4.3 40

Ar/39

Ar Results 250

Results from the eight incremental heating experiments are summarized in Table 2 and Figure 251

8; age uncertainties reflect 2analytical contributions. All samples define statistically acceptable 252

plateaus and have isochrons with trapped 40

Ar/36

Ar ratios that are indistinguishable from the 253

11

atmospheric 40

Ar/36

Ar ratio; thus, the plateau ages are preferred. Plateau ages for the Perşani 254

Mountain lavas range from 1221 ± 11 ka to 683 ± 28 ka. These ages, as well as others that are 255

discussed in subsequent sections, are calculated using an age of 28.201 Ma for the FCs standard 256

(Kuiper et al., 2008). 257

258

5. Discussion 259

5.1 Timing of volcanic activity 260

Previous K-Ar ages and paleomagnetic data were interpreted as an indication of two short 261

phases of volcanic eruption in the Perşani Mountains: a first between 1.5 and 1.2 Ma and second 262

between 0.67 and 0.52 Ma (Panaiotu et al., 2004). The new 40

Ar/39

Ar ages (Table 2) show that the 263

volcanic activity in the Perşani Mountains took place in at least five episodes between 1.2 and 0.6 264

Ma. The oldest identified structure is the Racoş volcano with an age of 1221 ± 11 ka. All 265

paleomagnetic directions from this structure are transitional (Table 1; Fig. 7), suggesting that this 266

complex was constructed rapidly. The second volcanic event occurred at 1142 ± 41 ka and it is 267

recorded by a lava flow in the southern part of the Perşani Mountains volcanic field (Fig. 2). Both 268

this lava flow from the Comana Quarry (PN35) and the neighboring lava (site CO) have a reversed 269

polarity and similar paleomagnetic directions, which suggest that they erupted during the same 270

volcanic event. Two lava flows with normal polarity, dated at 1062 ± 24 ka (PN 47) and 1060 ± 10 271

ka (PN 23), indicate that the third volcanic event took place during the beginning of the Jaramillo 272

normal subchron (e.g. Chanell et al., 2010). The next volcanic event took place around 800 ka; it is 273

recorded by a lava flows with reversed polarity from the Bârc Quarry and a lava flow with a 274

transitional direction from the Bogata Quarry. The last volcanic event at ~ 683 ka is recorded by lava 275

flows with normal polarity from the Mateiaş Quarry and Pietrele Valley, consistent with eruption 276

during the Brunhes normal chron. 277

12

The geographical distribution of magnetic polarity (Fig. 2) supports this volcanic episodicity 278

and gives also an idea about the most active volcanic periods. The presence of only a few lava flows 279

with reversed polarity shows that during the Matuyama chron volcanic eruptions were probably 280

episodic. Most lava flows have normal polarity and these cover large areas in the Perşani Mountains 281

volcanic field. The 40

Ar/39

Ar ages suggest that the main eruption periods took place during the 282

Jaramillo normal subchron and the lower part of the Brunhes normal chron. The time lapses between 283

each of the last three volcanic periods, which overlap in the central part of the Perşani Mountain 284

volcanic field, are in agreement both with the presence of paleosols between some lava flows from 285

this area and the magnetic polarity data. 286

The last three phases of mafic Na-alkalic volcanic activity in the Perşani Mountain partly 287

overlap adakite-like calc-alkaline volcanism from southern South Harghita Mountain. According to 288

K-Ar ages (Pécskay et al., 1995, 2006) and paleomagnetic data (Panaiotu et al., 2012), the last phase 289

of volcanic activity in the South Harghita Mountains began during the Jaramillo normal subchron 290

(around 1 Ma) with two isolated bodies and developed during the Brunhes chron (until around 0.3 291

Ma) with the main Ciomadu volcanic structure (Fig. 1). Our new data prove the Perşani volcanism as 292

contemporaneous with earlier products of Ciomadu volcano and do not support the hypothesis of 293

Seghedi et al. (2011) that the most important Na-alkalic magmatic activity took place during a short 294

interruption in adakite-like calc-alkaline magmatism in the South Harghita Mountains. Adakite-like 295

calc-alkaline magmatism in the South Harghita area account for a corner-flow around the steepening 296

and melting Vrancea slab below Moesia, while the Na-alkalic magmatism at the westernmost edge of 297

the Braşov basin system was associated with asthenosphere uprise during slab steepening and its 298

decompression melting (Seghedi et al., 2011). Even the magma sources are not the same, the ascent 299

of magmas was probably facilitated by the same regional tectonic event. This tectonic event is 300

probably the last episode of the generalized tectonic inversion that began during the latest Pliocene 301

13

and it remains active in the bending area of the Southeastern Carpathian Mountains (Seghedi et al., 302

2011, Popa et al., 2012). 303

5.2 Cobb Mountain normal subchron 304

Mankinen et al. (1978) documented 3 sites with transitional VGPs in the Alder Creek rhyolite 305

at Cobb Mountain and obtained a K-Ar age of 1200 ± 20 ka on sanidine. Sanidine from the Alder 306

Creek rhyolite yields a 40

Ar/39

Ar age of 1201 ± 1 ka relative to the 28.201 Ma calibration of the Fish 307

Canyon sanidine standard (Turrin et al., 1994; Nomade et al., 2005). The Alder Creek Rhyolite lava 308

is considered the type occurrence of the Cobb Mountain subchron, because other transitionally 309

magnetized lavas from the Cobb Mountain area and Coso Range have been shown to record a later 310

geomagnetic excursion at 1122 ± 10 ka named Punaruu excursion (Singer et al., 1999). A recent 311

study by Silva et al. (2012) reported two transitionally magnetized lavas from São Jorge Island, 312

Azores, one of which gives an unspiked K-Ar age of 1207 ± 17 ka that likely represent the onset of 313

the Cobb Mountain subchron. Unequivocal documentation of a normal polarity zone of similar age is 314

provided by deep-sea sediment cores recovered by the Ocean Drilling Program and its predecessor, 315

the Deep Sea Drilling Project (e.g. Clement, 2000; Yang et al., 2001; Channell et al., 2002; Laj and 316

Channell, 2007). From these widely distributed records, the highest resolution astrochronologic age 317

model from ODP sites 983 and 984 suggests that Cobb Mountain subchron is bounded by full polarity 318

reversals at 1215 and 1190 ka (Channell et al., 2002). Moreover, the PISO-1500 stack of Channell et 319

al (2009) that is based on 14 high resolution relative paleointensity and δ18

O records, including those 320

from ODP sites 983 and 984, reveals that the Cobb Mountain subchron is associated with a ~30 kyr 321

period of exceptionally low intensity. Channell et al. (2009) propose that drops in field intensity 322

below a threshold virtual axial dipole moment (VADM) value of 2.5×1022

Am2 may trigger 323

excursions and reversals. For the reversals bounding the Cobb Mountain normal subchron the 324

intensity drops below this threshold at 1210 ka and remains below it until 1183 ka (Channell et al., 325

2009). 326

14

Clement (2000) has analyzed directional changes in several sediment records during the Cobb 327

Mountain subchron. He found these records to exhibit VGP paths that are remarkably similar. A 328

cartoon illustrating segments of the VGP path identified by Clement (2000) during the reverse – 329

normal (R-N) transition is presented in Fig. 9. Detailed sedimentary records of the Cobb Mountain 330

subchron (Clement, 2000; Yang et al., 2001) show that the beginning R-N transition is characterized 331

by a movement of the VGPs away from the south geographic pole to a position or loop centered over 332

South Africa (segment B, Fig. 9) followed by a return to the south geographic pole. A second loop to 333

positions off of southwest Australia (segment C in Fig. 9) occurs next, followed by a return to the 334

pole. Then the VGPs proceed on a curved track through the central north Pacific (segment D and E in 335

Fig. 9) to positions close to the North Pole. According to Clement (2000), the VGPs path during N – 336

R transition are characterized by an initial loop over Asia followed by a quick southward movement 337

over the Atlantic Ocean (segment I in Fig. 9) to the south geographic pole. The VGPs then loop twice 338

over southern Africa and finally over southern South America. 339

The transitional field directions of lavas in the Racoş volcano yield VGPs over the Indian 340

Ocean, west of Australia (Fig. 9). The large directional swings may occur rapidly, but given the 341

limited geological field evidence, the sequence of eruptions and the duration of time recorded by sites 342

within the Racoş volcano remain uncertain. This makes difficult a detailed interpretation of this 343

elongated VGPs distribution. The 40

Ar/39

Ar age of 1221 ± 11 ka of transitionally magnetized site 344

PN11 from the Racoş basalt overlaps with the astronomical age for the onset of Cobb Mountain 345

subchron, around 1215 to 1210 ka (Channell et al., 2002; 2009). The VGP positions from the Racoş 346

basalts spread between the two initial loops of the VGPs path during R – N transition (Fig. 9). If we 347

look at individual records the VGPs from the Racoş volcano overlap with several VGPs from DSDP 348

Site 609 in the North Atlantic (Clement, 2000; Yang et al., 2001) during the initial loop over the 349

Indian Ocean. The 40

Ar/39

Ar age, together with the VGP positions of these flows suggest that the 350

transitional directions from the Racoş volcano record the beginning of the Cobb Mountain subchron. 351

15

The VGPs from São Jorge Island located near Hawaii along segment D (Fig. 9) suggest that they 352

record a subsequent part of the R-N transition (Silva et al., 2012). 353

Transitional VGPs from the Alder Creek rhyolite (Mankinen et al., 1978) cluster above South 354

America (Fig. 9), and the 40

Ar/39

Ar age of 1201 ± 1 ka for the ACr, is younger than the age of the 355

Racoş basalts. Taken together, this suggests that the Alder Creek rhyolite was magnetized during the 356

N – R transition that terminated the Cobb Mountain normal subchron (segment I). Based on 357

40Ar/

39Ar ages from the Racoş basalts and Alder Creek rhyolite, a minimum duration of the Cobb 358

Mountain subchron can be estimated at around 20 ± 11 ka. This value is similar to the 359

astrochronologic estimate of about 30 kyr (Channell et al., 2002; 2009). 360

The clustering of Racoş lavas VGPs southwest of Australia is similar to other records of 361

geomagnetic excursions and reversals during the Matuyama chron, which show clusters or 362

movements of VGPs in the beginning of these geomagnetic events over the same region (Hoffman 363

and Singer, 2004, 2008; Leonhardt et al., 2009). Thus, our results support the hypothesis that the 364

physical processes operating during the onset of each of these geomagnetic instabilities may be 365

similar. This may reflect the long-term influence of lower mantle heterogeneities on the weak, non-366

axial dipole fields that emerge during excursions and reversals (e.g., Hoffman and Singer, 2004; 367

2008; Leonhardt et al., 2009). 368

5.2 Matuyama–Brunhes precursor 369

Precise 40

Ar/39

Ar dating of transitionally magnetized lava flow sequences on the ocean islands 370

of Tahiti, La Palma, Maui and at Tatara-San Pedro volcano, Chile led Singer et al. (2005) to conclude 371

that all but the lavas on Maui record an excursion that preceded the Matuyama-Brunhes reversal by 372

about 18 kyr. The ages of these three lava flow sequences are indistinguishable from one another and 373

yield a weighted mean of 798 ± 3 ka (recalculated relative to 28.201 for Fish Canyon sanidine 374

standard), although, this weighted mean age may be ~1% or 8 ka too old (Singer et al., 2012). The 375

directional recordings in these lava flows have been correlated with a period of low paleointensity 376

16

that precedes the M–B reversal in many marine sediment cores and is astronomically dated at about 377

794 ka (Channell et al., 2010 and references therein). 378

Two lava flows with mid-latitude VGPs from the Perşani Mountains yield 40

Ar/39

Ar plateau 379

ages of 800 ± 25 and 799 ± 21 ka (Table 2) that overlap with the age of the M-B precursor. Note that 380

these two lavas were only analyzed once each and thus the plateau ages are relatively imprecise. 381

Additional experiments as part of a focused study of transitionally magnetized lavas in this region 382

could be undertaken to more precisely constrain the timing of the excursions and reversals. The 383

VGPs from these ~ 800 ka lava flows are located near the southern tip of South America. The mean 384

VGP from the Bogata Quarry has latitude of 50.6°S (Table 1) and can be categorized as transitional 385

VGP. The second VGP from the Bârc Quarry may be transitional, but its latitude of 57.6 ± 4.6°S only 386

slightly exceeds the paleosecular variation limit (55°S). We propose that these two lavas provide spot 387

recordings of the M-B precursor excursion in eastern Europe. The transitional VGP from the Bogata 388

Quarry basalt is positioned not far from transitional VGPs recorded by lava flows from La Palma 389

(Singer et al., 2002). Our results give additional support for presence of transitional VGPs over 390

southern South America during the M-B precursor. Transitional VGPs concentrated near Australia, 391

or the South America-South Atlantic region are characteristic of several reversals and excursions 392

during Matuyama and Brunhes chrons (e.g. Hoffman and Singer, 2004; 2008; Hoffman et al., 2008; 393

Leonhardt et al., 2009). 394

5.3 Paleosecular Variation 395

In a recent paper Linder and Gilder (2012), using numerical simulations with synthetic data, 396

has shown that the latitude dependency of the root mean square angular deviation of VGPs (SB) owes 397

its origin to a geometrical aberration from the mathematical conversion of directions to virtual 398

geomagnetic poles and is not related to geomagnetic phenomena. To avoid this mathematical artifact 399

they have proposed to quantify the paleosecular variation (PSV) using k, the precision parameter for 400

directions (Fisher, 1953), and not the root mean square angular deviation of VGPs. Their proposal is 401

17

based on the hypothesis that directions are Fisherian distributed and not the VGPs. This question 402

whether poles or directions are more Fisherian distributed is still under debated (e.g. Lund, 2007; 403

Tauxe, 2013). For this reason we have analyzed the paleosecular variation using the classical method 404

based on the dispersion of VGPs (e.g. Johnson et al., 2008, Cromwell et al., 2013). 405

The 23 VGPs from the Perşani Mountains within the secular variation limits were combined 406

with three VGPs from contemporaneous volcanic rocks from the neighboring South Harghita 407

Mountains (Panaiotu et al., 2012) to estimate paleosecular variation (PSV) between 1.2 Ma and 0.6 408

Ma. The PSV was described using SB, the root mean square angular deviation of VGPs about the 409

geographic axis (e.g. Johnson et al., 2008). SB and its confidence limits were calculated using the 410

PmagPy-2.73 software packge (Tauxe, 2010). To compute SB the reverse polarity data has been 411

flipped to its equivalent normal polarity. The result is presented in Table 3 together with dispersion of 412

VGPs recorded in the South Harghita Mountains between 2 Ma and 4.3 Ma (Panaiotu et al., 2012) 413

and dispersion of combined VGPs from the Perşani Mountains and the South Harghita Mountains. 414

Fig. 10A shows our results (Table 3) and regional compilations for the Time Averaged 415

geomagnetic Field Initiative (TAFI) studies (Johnson et al., 2008, their Table 6) and global 416

compilation (Johnson et al, 2008, their Table 8) in the latitudinal band 42°N-55°N. In this latitudinal 417

band there are the nearest PSV data relevant for the sampling area latitude. For clarity in figure 10 we 418

moved slightly the data with respect to their original latitude (around 0.1°). In the same figure we 419

plotted the expected dispersion from PSV model G (McElhinny and McFadden, 1997) and for the 420

GAD version of the TK03 statistical model (Tauxe and Kent, 2004). 421

The VGP dispersion in our data is higher than that for the global compilation at 44.8°N 422

latitude, but it is in better agreement, at 95% confidence level, with dispersions obtained from the 423

TAFI studies at the latitude of 43°N (Johnson et al., 2008). This value of SB between 1.2 Ma and 0. 6 424

Ma is closer to the expected dispersion from model TK03 (Tauxe and Kent, 2004) than that from 425

model G (McElhinny and McFadden, 1997). 426

18

When compared with VGP dispersion between 2.0 and 4.3 Ma the SB value between 0.6 – 1.2 427

Ma is slightly lower, but at 95% confidence the two values can be also considered identical. The 428

Matuyama data set presented by Johnson et al. (2008) shows several estimates of SB around 53°N 429

latitude that are higher than during the Brunhes (Fig. 10A). The results from Perşani and Harghita 430

Mountains might imply an increase of VGP dispersion with time starting around 45°N, but more data 431

are needed at a global level to determine if this reflects the behavior of the geomagnetic field, or an 432

incomplete database. 433

The time-average field (TAF) is examined using the inclination anomaly (Table 3). 434

Inclination anomaly is defined as the difference between the observed inclination and the expected 435

inclination from a GAD at the same latitude (e.g. Johnson et al., 2008). The inclination anomaly (ΔI) 436

is 1.6° ± 3.6° for the whole data set (0.6 – 4.3 Ma), 2.6° ± 5.5° between 0.6 Ma and 1.2 Ma and 0.7° 437

± 4.3° between 2 Ma and 4.3 Ma. Taking into account the 95% confidence limits these values are 438

compatible with data for 0-5 Ma data set (Johnson et al., 2008) both for the latitude of 44.8°N and 439

52.8°N (Fig 10B). Our data with a small positive inclination anomaly can be interpreted as a support 440

for the trend observed in the global data by Johnson et al. (2008), which suggest a change from a 441

negative inclination anomaly to a small positive inclination anomaly around 50°N. Both this very 442

small inclination anomaly and the mean pole position of the 0.6-4.3 Ma combined data set, which is 443

indistinguishable from the spin axis, are consistent with field geometry with insignificant nonzero 444

nonaxial dipole contributions (Tauxe et al. 2004). 445

446

6. Conclusions 447

Of the paleomagnetic directions obtained from lava flows in the Perşani Mountains, 4 are 448

reversed polarity, 19 are normal polarity and 16 have transitional polarity. 40

Ar/39

Ar isochron ages 449

determined from incremental heating experiments on groundmass indicate that two of the reversely 450

magnetized lavas erupted at 1142 ± 41 and 800 ± 25 ka, four of the normally magnetized lavas 451

19

erupted at 1060 ± 10, 1062 ± 24, 684 ± 21, and 683 ± 28 ka, and two transitionally magnetized lavas 452

formed at 1221 ± 11 and 799 ± 21 ka. Both the new 40

Ar/39

Ar ages and the paleomagnetic data 453

suggest at least five episodes of volcanic activity with the most active periods during the Jaramillo 454

and Brunhes chrons. The age of the older transitionally magnetized lava flow agrees with independent 455

estimates from astrochronology for the age of the reversal that defines the onset of the Cobb 456

Mountain normal polarity subchron. Given the uncertainty, the age of the younger transitional lava 457

is similar to that of an excursion that preceded the Matuyama-Brunhes polarity reversal and which has 458

come to be known as the Matuyama-Brunhes precursor. The VGP of this transitional lava, dated at 459

799 ± 21 ka, falls over southern South America and is remarkably similar to VGPs obtained from 460

lavas on La Palma, Canary Islands that also record the Matuyama-Brunhes precursor. The 461

transitional lava in Romania erupted about 4200 km northeast of La Palma, thus the common VGPs 462

from these two sites in Europe may reflect a regional, and perhaps deep-seated, control on the non 463

axial dipole field that emerged during the Matuyama-Brunhes precursor (e.g., Singer et al., 2005; 464

Hoffman and Singer, 2008). Omitting the excursion data, PSV, around19°, is larger than the expected 465

value around 45°N from the global compilation of Johnson et al. (2008), but closer to the value 466

obtained only from the TAFI studies. 467

468

Acknowledgements 469

C.G.Panaiotu and A.Ţugui were supported by grant PNII-IDEI 974/2007. A.G. Panaiotu was 470

partially supported by University of Bucharest, grant 19601.20/2011. C. Necula was supported by the 471

strategic grant POSDRU/89/1.5/S/58852. I. Seghedi thanks Institute of Geodyanamics for support. 472

We thank Bernard Henry and an anonymous reviewer for their helpful comments. 473

474

475

476

20

Appendix. Supplementary data 477

Sampling sites location: persani.kml 478

Complete 40

Ar/39

Ar results 479

480

References 481

Brown, L. L., Singer, B. S., Gorring, M. L., 2004. Paleomagnetism and 40

Ar/39

Ar Chronology of Lavas from Meseta del 482

Lago Buenos Aires, Patagonia. Geochem. Geophys. Geosyst. 5, Q01H04, doi:10.1029/2003GC000526. 483

Ciulavu, D., Dinu, C., Szakács, A., Dordea, D., 2000. Neogene kinematics of the Transylvanian Basin (Romania). AAPG 484

Bulletin 84 (10), 1589-1615. 485

Chadima, M. & Hrouda, F., 2006. Remasoft 3.0 — a user-friendly paleomagnetic data browser and analyzer. Travaux 486

Géophysiques XXVII, 20-21. 487

Channell, J.E.T., A. Mazaud, P. Sullivan, S. Turner, M.E. Raymo., 2002. Geomagnetic excursions and paleointensities in 488

the 0.9-2.15 Ma interval of the Matuyama Chron at ODP Site 983 and 984 (Iceland Basin). .J. Geophys. Res. 107 489

(B6), 10.1029/2001JB000491. 490

Channell, J.E.T, Guyodo, Y., 2004. The Matuyama Chronozone at ODP Site 982 (Rockall Bank): Evidence for decimeter-491

scale magnetization lock-in depths., In: "Timescales of the Internal Geomagnetic Field". J.E.T. Channell, D.V. 492

Kent, W. Lowrie and J. Meert (editors), AGU Geophysical Monograph 145, 205-219. 493

Channell, J.E.T., Xuan, C., Hodell, D.A., 2009. Stacking paleointensity and oxygen isotope data for the last 1.5 Myr. 494

Earth and Planetary Science Letters 283, 14–23. 495

Channell, J. E. T., Hodell, D. A., Singer, B. S., Xuan, C., 2010. Reconciling astrochronological and 40Ar/39

Ar ages for the 496

Matuyama‐Brunhes boundary and late Matuyama Chron. Geochem. Geophys. Geosyst. 11, Q0AA12, 497

doi:10.1029/2010GC003203. 498

Clement, B.M., 2000. Comment on the Lau Basin Cobb Mountain records by Abrahamsen and Sager. Physics of the Earth 499

and Planetary Interiors 119, 73–184. 500

Cromwell, G., Tauxe, L., Staudigel, H., Constable, C.G., Koppers, A.A.P., Pedersen, R-B., 2013. In search of long term 501

hemispheric asymmetry in the geomagnetic field: results from high northern latitudes. Geochemistry, 502

Geophysics, Geosystems, DOI 10.1002/ggge.20174. 503

Day, R., Fuller, M.D., Schmidt, V., 1977. Hysteresis properties of titanomagnetites: Grain-size and compositional 504

dependence. Phys. Earth Planet. Int. 13, 260–267. 505

Downes, H., Seghedi, I., Szakács, A., Dobosi, G., James, D.E., Vaselli, O., Rigby, I.J., Ingram, G.A., Rex, D., Pécskay, 506

Z., 1995. Petrology and geochemistry of the late Tertiary/Quaternary mafic alkaline volcanism in Romania. 507

Lithos 35, 65-81. 508

Dunlop, D.J., 2002.Theory and application of the Day plot (Mrs/Ms versus Hcr/Hc) 1. Theoretical curves and tests using 509

titanomagnetite data. J. Geophys. Res, 107 (B3), 2056, doi:10.1029/2001JB000486. 510

Embey-Isztin, A., Dobosi, G., 1995. Mantle source characteristics for Miocene-Pleistocene alkali basalts, Carpathian-511

Pannonian Region: a review of trace elements and isotopic composition. Acta Vulcanologica 7, 155-166. 512

Fielitz, W., Seghedi, I., 2005. Late Miocene - Quaternary volcanism, tectonics and drainage system evolution in the East 513

Carpathians, Romania. Tectonophysics 410, 111-136. 514

21

Fisher, R. A., 1953. Dispersion on a sphere. Proc. R. Soc. London, Ser. A, 217, 295–305. 515

Gîrbacea, R., Frisch, W., Linzer, H.-G., 1998. Post-orogenic uplift induced extension: a kinematic model for the Pliocene 516

to recent tectonic evolution of the Eastern Carpathians (Romania). Geologica Carpathica 49, 315-327. 517

Harrison, R. J., Feinberg, J. M., 2008. FORCinel: An improved algorithm for calculating first-order reversal curve 518

distributions using locally weighted regression smoothing, Geochem. Geophys. Geosyst., 9(5), Q05016, 519

doi:10.1029/2008GC001987. 520

Hoffman, K. A., Singer, B., 2004. Regionally recurrent paleomagnetic transitional fields and mantle processes. In: 521

Channell, J. E. T., Kent, D. V., Lowrie, W., Meert, J (Eds.),Timescales of the Paleomagnetic Field. Am. 522

Geophys. Un. Geophys. Monogr. 145, pp. 223–243 523

Hoffman, K.A., Singer, B.S., 2008. Magnetic source separation in Earth’s outer core. Science 321, 1800. 524

Hoffman, K.A., Singer, B.S., Camps, P., Hansen, L.N., Johnson, K., Clipperton, S., Carvallo, C., 2008. Stability of mantle 525

control over dynamo flux since the mid-Cenozoic. Physics of the Earth and Planetary Interiors 169, 20-27. 526

Hrouda, F., Chlupáčová, M., Mrázová, S., 2006. Low-field variation of magnetic susceptibility as a tool for magnetic 527

mineralogy of rocks. Physics of the Earth and Planetary Interiors 154, 323-336. 528

Jicha, B.R., Kristjánsson, L., Brown, M.C., Singer, B.S., Beard, B.L., Johnson, C.M., 2011. New age for the Skálamælifell 529

excursion and identification of a global geomagnetic event in the late Brunhes chron. Earth and Planetary 530

Science Letters 310, 509-517. 531

Johnson, C. L., Constable, C. G., Tauxe, L., Barendregt, R., Brown, L. L., Coe, R. S., Layer, P., Mejia, V., Opdyke, N. D., 532

Singer, B. S., Staudigel, H., Stone, D. B., 2008. Recent investigations of the 0–5 Ma geomagnetic field recorded 533

by lava flows. Geochem. Geophys. Geosyst. 9, Q04032, doi:10.1029/2007GC001696. 534

Kirschvink, J.L., 1980. The least-squares line and plane and the analysis of palaeomagnetic data. Geophysical Journal of 535

the Royal Astronomical Society 62, 699-718. 536

Kuiper K.F., Deino A., Hilgen F.J., Krijgsman W., Renne P.R., Wijbrans J.R., 2008. Synchronizing rock clocks of Earth 537

history. Science 320, 500-504. doi: 10.1126/science/1154339 538

Laj, C., Channell, J. E. T., 2007. Geomagnetic Excursions. In: Kono, M. (ed), Treatise in Geophysics Volume 5 539

Geomagnetism. Elsevier B.V., pp. 373-416. 540

Lattard, D., Engelmann, R., Kontny, A., Sauerzapf, U., 2006. Curie temperatures of synthetic titanomagnetites in the Fe-541

Ti-O system: Effects of composition, crystal chemistry, and thermomagnetic methods. J. Geophys. Res. 111, 542

B12S28, doi:10.1029/2006JB004591. 543

Leonhardt, R., Fabian, K., 2007. Paleomagnetic reconstruction of the global geomagnetic field evolution during the 544

Matuyama/Brunhes transition: Iterative Bayesian inversion and independent verification. Earth and Planetary 545

Science Letters 253, 172–195. 546

Leonhardt, R., McWilliams, M., Heider, F., Soffel, H. C., 2009. The Gilsá excursion and the Matuyama/Brunhes 547

transition recorded in 40Ar/39Ar dated lavas from Lanai and Maui, Hawaiian Islands. Geophys. J. Int. 179, 43–548

58. 549

Lexa, J., Seghedi, I., Németh, K., Szakács, A., Konečný, V., Pécskay, Z., Fülöp, A., Kovacs, M., 2010. Neogene-550

Quaternary volcanic forms in the Carpathian-Pannonian Region: a review. Cent. Eur. J. Geosci. 2(3), 207-270, 551

DOI: 10.2478/v10085-010-0024-5. 552

Linder, J., Gilder, S.A., 2012. Latitude dependency of the geomagnetic secular variation S parameter: A mathematical 553

artifact. GEOPHYSICAL RESEARCH LETTERS 39, L02308, doi:10.1029/2011GL050330. 554

22

Lund, S.P., 2007. Paleomagnetic Secular Variation. In: Gubbins, D., Herrero-Bervera, E., (eds), Encyclopedia of 555

Geomagnetism and Paleomagnetism. Springer, pp. 766-776. 556

Mankinen, E.A., Donnelly, J.M., Grommé, C.S., 1978. Geomagnetic polarity event recorded at 1.1 m.y.b.p. on Cobb 557

Mountain, Clear Lake volcanic field, California. Geology 6, 653–656. 558

Mason, P.R.D., Seghedi, I., Szakács, A., Downes, H., 1998. Magmatic constraints on geodynamic models of subduction in 559

the Eastern Carpathians, Romania. Tectonophysics 297, 157-176. 560

Maţenco L, Bertotti G, Leever K, Cloetingh S., Schmid S.M., Tărăpoancă M. and Dinu C., 2007. Large-scale deformation 561

in a locked collisional boundary: Interplay between subsidence and uplift, intraplate stress, and inherited 562

lithospheric structure in the late stage of the SE Carpathians evolution. Tectonics 26, TC4011, doi 563

10.1029/2006TC001951. 564

Maţenco, L., Krezsek, C., Merten, S., Schmid, S., Cloetingh, S., Andriessen, P.A.M., 2010. Characteristics of collisional 565

orogens with low topographic build-up: an example from the Carpathians. Terra Nova 22(3), 155-165. 566

McElhinny, M. W., P. L. McFadden, 1997. Palaeosecular variation over the past 5 Myr based on a new generalized 567

database. Geophys. J. Int. 131, 240–252. 568

Merten, S., Maţenco, L., Foeken, J.P.T., Stuart, F.M., Andriessen, P.A.M., 2010. From nappe-stacking to out-of-sequence 569

post-collisional deformations: Cretaceous to Quaternary exhumation history of the SE Carpathians assessed by 570

low-temperature thermochronology. Tectonics 29(3), TC3013, doi: 10.1029/2009tc002550. 571

Min K., Mundil R., Renne P. R. and Ludwig K. R., 2000. A test for systematic errors in 40

Ar/39

Ar geochronology through 572

comparison with U–Pb analysis of a 1.1 Ga rhyolite. Geochim. Cosmochim. Acta 64, 73–98. 573

Muxworthy, A. R., Dunlop, D. J., 2002. First-order reversal curve (FORC) diagrams for pseudo-single-domain magnetites 574

at high temperature. Earth and Planetary Science Letters 203, 369-382. 575

Nomade, S., Renne, P. R., Vogel, N., Deino, A. L., Sharp, W. D., Becker, T. A., Jaouni, A. R., Mundil R., 2005. Alder 576

Creek sanidine (ACs-2): A Quaternary 40

Ar/39

Ar dating standard tied to the Cobb Mountain geomagnetic event. 577

Chem. Geol. 218, 315–338. 578

Opdyke, N. D., Kent, D. V., Huang, K., Foster, D. A., Patel, J. P., 2010. Equatorial paleomagnetic time-averaged field 579

results from 0–5 Ma lavas from Kenya and the latitudinal variation of angular dispersion. Geochem. Geophys. 580

Geosyst. 11, Q05005, doi:10.1029/2009GC002863. 581

Panaiotu, C. G., Pécskay, Z., Hambach, U. Seghedi, I. Panaiotu, C.C., Itaya, C. E. T., Orleanu, M., Szakács, A., 2004. 582

Short-lived Quaternary volcanism in the Perşani Mountains (Romania) revealed by combined K-Ar and 583

paleomagnetic data. Geologica Carpathica 55, 333-339. 584

Panaiotu, C.G., Vişan, M., Ţugui, A., Seghedi, I., Panaiotu, A.G., 2012. Palaeomagnetism of the South Harghita volcanic 585

rocks of the East Carpathians: implications for tectonic rotations and palaeosecular variation in the past 5 Ma. 586

Geophysical Journal International, 189, 369-382. 587

Pécskay, Z., Edelstein, O., Seghedi, I., Szakács, A., Kovacs, M., Crihan, M., Bernad, A., 1995. K-Ar datings of the 588

Neogene-Quaternary calc-alkaline volcanic rocks in Romania. Acta Vulcanologica 7, 53-63. 589

Pécskay, Z. Lexa, J. Szakács, A. Seghedi, I. Balogh, K. Konečny, V. Zelenka, T. Kovacs, M. Poka, T., Fülop, A., Marton, 590

E., Panaiotu, C., Cvetković, V., 2006. Geochronology of Neogene magmatism in the Carpathian arc and intra-591

Carpathian area. Geologica Carpathica 57, 511-530. 592

23

Popa, M., Radulian, M., Szakács, A. Seghedi, I., Zaharia, B., 2012. New Seismic and Tomography Data in the Southern 593

Part of the Harghita Mountains (Romania, Southeastern Carpathians): Connection with Recent Volcanic Activity. 594

Pure Appl. Geophys., 169, 9, 1557-1573. 595

Roberts, A. P., Pike, C. R., Verosub, K. L., 2000. First-order reversal curve diagrams: A new tool for characterizing the 596

magnetic properties of natural samples. J. Geophys. Res. 105, 28,461–28,475. 597

Seghedi, I., Szakács, A., 1994. The Upper Pliocene-Pleistocene effusive and explosive basaltic volcanism from the 598

Perşani Mountains. Rom. J. Petrology 76, 101-107. 599

Seghedi, I., Balintoni I., Szakács A., 1998. Interplay of tectonics and Neogene post-collisional magmatism in the 600

Intracarpathian area. Lithos 45, 483-499. 601

Seghedi, I., Downes, H., Szakács, A., Mason, P.R.D., Thirlwall, M.F., Roşu, E., Pécskay, Z., Marton, E., Panaiotu, C., 602

2004a. Neogene-Quaternary magmatism and geodynamics in the Carpathian-Pannonian region: a synthesis. 603

Lithos 72, 117-146. 604

Seghedi I., Downes H., Vaselli O., Szakács A., Balogh K., Pécskay Z., 2004b. Post-collisional Tertiary–Quaternary mafic 605

alkalic magmatism in the Carpathian–Pannonian region: a review. Tectonophysics 393, 43-62. 606

Seghedi, I., Downes, H., 2011. Geochemistry and tectonic development of Cenozoic magmatism in the Carpathian–607

Pannonian region. Gondwana Research 20 (4), 655-672. 608

Seghedi I., Maţenco L., Downes H., Mason P. R. D., Szakács A., Pécskay Z., 2011. Tectonic significance of changes in 609

post-subduction Pliocene-Quaternary magmatism in the south east part of the Carpathian-Pannonian Region. 610

Tectonophysics 502, 146-157. 611

Silva P. F., Henry B., Marques F. O., Hildenbrand A., Madureira P., Mériaux C. A., Kratinova Z., 2012. Palaeomagnetic 612

study of a subaerial volcanic ridge (São Jorge Island, Azores) for the past 1.3 Myr: evidence for the Cobb 613

Mountain Subchron, volcano flank instability and tectonomagmatic implications. Geophys. J. Int. , 188, 959–614

978. 615

Singer, B. S., K. A. Hoffman, K. A., A. Chauvin, A., Coe, R. S., Pringle, M. S., 1999. Dating transitionally magnetized 616

lavas of the late Matuyama Chron: Toward a new 40Ar/39Ar timescale of reversals and events. J. Geophys. Res. 617

104, B1, 679-693. 618

Singer, B. S., Relle, M. K., Hoffman, K. A., Battle, A., Laj, C., Guillou, H., Carracedo, J. C., 2002. Ar/Ar ages of 619

transitionally magnetized lavas on La Palma, Canary Islands, and the Geomagnetic Instability Timescale. J. 620

Geophys. Res. 107 (B11), 2307, doi:10.1029/2001JB001613. 621

Singer, B.S., Brown, L.L., Rabassa, J.O., Guillou, H., 2004. 40Ar/39Ar chronology of Late Pliocene and early Pleistocene 622

geomagnetic and glacial events in southern Argentina. In: Channell, J.E.T., Kent, D.V., Lowrie, W., Meert, J. 623

(Eds.), Timescales of the Paleomagnetic Field: American Geophysical Union, Geophysical Monograph, 145, pp. 624

175–190. 625

Singer, B.S., Hoffman, K.A., Coe, R.S., Brown, L.L., Jicha, B.R., Pringle M.S., Chauvin, A., 2005. Structural and 626

temporal requirements for geomagnetic field reversal deduced from lava flows. Nature, 434, p. 633-636. 627

Singer, B.S., Hoffman, K.A., Schnepp, E., Guillou, H., 2008. Multiple Brunhes chron excursions in the West Eifel 628

Volcanic Field: support for long-held mantle control on the non-axial dipole field. Phys. Earth Planet. Inter. 169, 629

28–40. 630

Singer, B.S., Jicha, B.R., Coe R.S., Mochizuki, N., 2012. An EARTHTIME chronology for the Matuyama-Brunhes 631

reversal. EOS, Transactions of the American Geophysical Union abstract, V21E-06. 632

24

Singer, B.S., Guillou, H. Jicha, B.R., Zanella, E., Camps, P., 2013. Lava flow recordings of the Blake and post-Blake 633

excursions: refining the Quaternary Geomagnetic Instability Time Scale (GITS), Quaternary Geochronology, in 634

press, http://dx.doi.org/10.1016/j.quageo.2012.12.005. 635

Steiger, R.H., Jäger, E., 1977. Subcommission on geochronology: convention on the use of decay constants in geo- and 636

cosmochronology. Earth Planet. Sci. Lett. 36, 359– 362. 637

Szakács, A., Seghedi, I., 1995. The Călimani - Gurghiu - Harghita volcanic chain, East Carpathians, Romania: 638

volcanological features. Acta Volcanologica 7, 145-153. 639

Tauxe, L., 2010. PmagPy , software package, Available at: http://magician.ucsd.edu/Software/PmagPy/index.html (last 640

time accessed: 2012 February 15). 641

Tauxe, L., 2013. Essentials of Paleomagnetism: Second Web Edition, Available at: http://magician.ucsd.edu/Essentials_2/ 642

((last time accessed: 2013 June 6) 643

Tauxe, L., Constable, C., Johnson, C. L., Koppers, A. A. P., Miller, W. R., Staudigel, H., 2003. Paleomagnetism of the 644

southwestern U.S.A. recorded by 0–5 Ma igneous rocks. Geochem. Geophys. Geosyst. 4(4), 8802, 645

doi:10.1029/2002GC000343. 646

Tauxe, L., Kent D., 2004. A simplified statistical model for the geomagnetic field and the detection of shallow bias in 647

paleomagnetic inclinations: Was the ancient magnetic field dipolar?. In: Channell, J.E.T., Kent, D.V., Lowrie, 648

W., Meert, J.G. (Eds),Timescales of the Internal Geomagnetic Field. Geophys. Monogr. Ser., AGU, Washington, 649

D. C., 145, pp. 101–115 650

Tauxe, L., Luskin, C., Selkin, P., Gans, P., Calvert, A., 2004. Paleomagnetic results from the Snake River Plain: 651

Contribution to the time-averaged field global database. Geochem. Geophys. Geosyst. 5, Q08H13, 652

doi:10.1029/2003GC000661. 653

Turrin, B.D., Donnely-Nolan, J.M., Hearn Jr., B.C., 1994. 40Ar/39Ar ages from the rhyolite of Alder Creek, California: 654

age of the Cobb Mountain normal-polarity subchron revisited. Geology 22, 251– 254. 655

Vandamme, D. ,1994. A new method to determine paleosecular variation, Phys. Earth Planet. Inter., 85, 131–142. 656

van Hinsbergen, D.J.J., Dupont-Nivet, D., Nakov, R., Oud, K., Panaiotu, C., 2008. No significant post-Eocene rotation of 657

the Moesian Platform and Rhodope (Bulgaria): Implications for the kinematic evolution of the Carpathian and 658

Aegean arcs. Earth and Planetary Science Letters 273, 345-358. 659

Yang, Z., Clement, B. M., Acton, G. D., Lund, S. P., Okada, M., Willams. T., 2001. Records of the Cobb Mountain 660

Subchron from the Bermuda Rise (ODP LEG 172). Earth and Planetary Science Letters 193, 303-313. 661

Zijderveld, J.D.A., 1967. A.C. demagnetization of rocks: analysis of results. In: Collinson, D.W., Creer, K.M., Runcorn, 662

S.K., (Eds.), Methods of Palaeomagnetism. Elsevier, Amsterdam, pp. 254—286. 663

Zhao, X., Riisager, P., Antretter, M., Carlut, J., Lippert, P., Liu, Q., Galbrun, B., Hall, S., Delius, H., Kanamatsu, T., 2006. 664

Unraveling the magnetic carriers of igneous cores from the Atlantic, Pacific, and the southern Indian oceans with 665

rock magnetic characterization. Physics of the Earth and Planetary Interiors 156, 294–328. 666

667

668

669

670

671

672

673

25

Figure captions 674

675

Fig. 1. Geological sketch map of the bending area of the East Carpathians and surrounding areas 676

(modified after Fielitz and Seghedi, 2005). Symbols: 1 = Quaternary volcanism younger than 1.2 Ma: 677

Perşani Mountains(PM) and Ciomadu volcanic structure (CV); 2 = Miocene to Quaternary volcanic 678

rocks older than 2 Ma: Călimani-Gurghiu (CG), North Harghita (NT) and South Harghita (SH) 679

Mountains; 3 = Moldavide nappes ; 4 = Dacides and other inner Carpathians nappes; 5 = Miocene-680

Quaternary intramontane basins : Gheorgheni Basin (GB), Ciuc Basin (CB) and Braşov Basin (BB); 6 681

= present day seismic active faults in the Moesian Platform. 682

Fig. 2. Location of sampling sites in the Perşani Mountains (circles). Legend: 1. Normal polarity; 2. 683

Reversed polarity; 3. Intermediate polarity; 4. 40

Ar/39

Ar age; 5. Holocene alluvia; 6. Lava flows; 7. 684

Scoria cones; 8. Pyroclastics (phreatomagmatic deposits); 9. Prevolcanic basement (Mesozoic and 685

Cenozoic). Map is modified after Panaiotu et al. (2004). 686

Fig. 3. A. Low-field variation of magnetic susceptibility normalized to initial value; B. S ratio vs. V 687

parameter. 688

Fig. 4. Typical examples of: A. field dependence (H) of magnetic susceptibility normalized to initial 689

value for a Ti-rich titanomagnetite sample (site PN1) and Ti-poor titanomagnetite sample (PN9); B. 690

temperature (T) dependence of magnetic susceptibility normalized to room temperature value for a 691

Ti-rich titanomagnetite sample ( site PN1) and Ti-poor titanomagnetite sample (site PN9); C. FORC 692

diagrams for Ti-poor titanomagnetite sample (site PN9); D. FORC diagram for Ti-rich 693

titanomagnetite sample (site PN1). 694

Fig. 5. Day plot of site-representative samples. The boundaries between SD, PSD and MD regions 695

and SD-MD mixing curves for magnetite (black lines) and TM60 (grey lines) are after Dunlop 696

(2002). 697

Fig. 6. Orthogonal projections showing typical demagnetization behavior. Full dots = horizontal 698

projection; Open dots = vertical projection. 699

26

Fig. 7 A. Site-mean directions; B. VGPs distribution and PSV limit (dotted line). 700

Fig. 8. 40

Ar/39

Ar age spectra and isochrons for eight analyzed lavas. 701

Fig. 9. Cartoon shown segments of the VGP path for the R–N transition (segments A to F) and N-R 702

(segment I) from the available sediment records of the Cobb Mountain Subchron (Clement, 2000). 703

Transitional VGPs from lavas are plotted with full stars (Racoş, PerşaniMountains, this study), circles 704

(São Jorge Island, Azores, Silva et al., 2012) and squares (Cobb Mountain, California, Mankinnen et 705

al., 1978). 706

Fig. 10. VGP dispersion from lavas (A) and inclination anomaly (B) and their 95% confidence 707

intervals. Normal and reversed combined data from Perşani and South Harghita Mountains are 708

represented with: full square = data younger than 1.2 Ma; open square = data between 2 Ma and 4.3 709

Ma; grey square = all data between 0.6 Ma and 4.3 Ma; Data from Johnson et al. (2008) are 710

represented with full symbols for Brunhes-age normal polarity, open symbols for Matuyama-age 711

reversed polarity data and grey symbols for 0-5 Ma combined data set: inverted triangles represent 712

data only from TAFI studies, circles are latitudinally binned global data. Predicted dispersions for 713

Model G and TK03: horizontal bars. 714

715 716

717

718

719

720

721

722

723

Figure1

Figure2

Figure3

Figure4

Figure5

Figure6

Figure7

Figure8

Figure9

Figure10

1

Table 1 Paleomagnetic results from the Perşani Mountains Site Locality

Age (ka)

D (°) I (°) N k α95

(°)

P PLon (°) Plat (°)

PN09 Racoş volcano (scoria cone) 127.1 12.0 5 116 7.1 T 85.5 -21.3

PN10 Racoş volcano (scoria cone) 119.2 8.9 6 391 3.4 T 93.8 -12.7

PN46 Racoş volcano (scoria cone) 129.0 15.9 6 340 3.6 T 78.6 -19.8

PN06 Racoş volcano (dyke) 133.7 9.5 5 100 7.7 T 74.8 -24.5

PN48 Racoş volcano (dyke) 144.2 4.4 9 109 4.9 T 62.3 -34.0

PN08 Racoş volcano (dyke) 139.4 6.1 6 174 5.9 T 79.7 -27.0

PN07 Racoş volcano (lava flow) 149.0 0.0 7 106 5.9 T 70.5 -30.2

PN05 Racoş volcano (lava flow) 147.2 15.4 5 66 9.5 T 57.9 -31.5

PN41 Racoş volcano (lava flow) 160.3 7.9 5 173 5.8 T 49.1 -37.6

PN04 Racoş volcano (lava flow) 127.5 4.1 5 141 6.5 T 85.9 -20.0

PN42 Racoş volcano (lava flow) 143.3 1.3 8 187 4.1 T 76.3 -34.3

PN01 Racoş volcano (lava flow) 151.8 1.3 6 980 2.1 T 62.2 -37.9

PN11 Racoş volcano (lava flow) 134.9 2.7 5 83 8.4 T 84.1 -23.4

PN13 Racoş volcano (lava flow) 125.1 -4.7 5 390 3.9 T 92.3 -23.1

PN12 Racoş volcano (lava flow) 120.4 -5.3 5 71 9.1 T 94.5 -22.0

PN35 Comana Quarry (lava flow) 168.1 -69.0 5 254 4.8 R 144.0 -66.2

CO* Comana Valley (lava flow) 152.5 -61.7 6 295 4.4 R 116.1 -70.2

TZ* Turzun hill (lava flow) 168.1 -57.1 6 140 4.3 R 76.2 -77.9

PN14 Sărata volcano (lava flow) 357.2 49.2 5 132 6.7 N 237.3 73.0

PN17 Măguricea Quarry (scoria cone) 23.8 61.0 5 115 7.2 N 115 75.5

PN21 Măguricea hill (lava flow) 10.7 60.2 6 54 9.1 N 180.1 70.5

PN18 Hoghiz (lava flow) 17.0 64.1 6 91 7.0 N 87.1 73.7

PN19 Hoghiz (lava flow) 31.9 64.5 6 177 5.0 N 93.3 61.6

PN20 Hoghiz (lava flow) 15.5 62.5 6 94 6.9 N 114.8 75.3

PN22 Bârc valley (lava flow) 10.9 51.7 5 63 9.6 N 237.3 79.5

PN23 Bârc valley (lava flow) 19.4 62.7 7 234 4.0 N 95.1 79.9

PN24 Bârc valley (lava flow) 7.3 72.1 6 286 4.0 N 358.6 79.2

PN28 Bârc valley (lava flow) 6.9 75.7 6 401 3.3 N 4.6 74.7

PN27 Bârc Old Quarry (lava flow) 350.6 74.5 5 92 8.0 N 53.1 60.5

PN45 Stânii Quarry (lava flow) 335.2 64.2 5 153 6.2 N 302.5 72.2

PN33 Bogata valley (lava flow) 36.2 63.0 5 310 4.3 N 115.1 68.2

PN32 Bogata valley (lava flow) 15.6 77.6 7 153 4.9 N 39.9 55.5

PN47 Bogata valley (lava flow) 340.9 61.3 6 201 4.7 N 333.1 69.7

BV1* Bogata valley (lava flow) 20.7 57.2 5 76 8.8 N 136.1 72.6

PN36 Trestia valley (lava flow) 341.1 69.8 8 173 4.2 N 22.1 76.4

PN34 Mateiaş volcano (lava flow) 346.4 64.5 6 571 2.8 N 298.1 78.3

PN30 Pietrele valley (lava flow) 9.9 71.0 8 109 5.8 N 50.5 65.7

PN31 Pietrele valley (lava flow) 10.2 72.4 7 487 2.7 N 55.2 76.9

PVA* Pietrele valley (lava flow) 13.6 74.2 6 1585 1.7 N 49.1 73.4

PVmean Pietrele valley (mean value) 11.1 72.5 3 2237 2.6 N 49.3 73.4

PN25 Barc Quarry (lava flow) 215.1 -53.2 6 524 2.9 R 310.7 -63.2

PN37 Barc Quarry (lava flow) 231.5 -51.4 5 1327 2.1 R 300.3 -52.1

PN38 Barc Quarry (lava flow) 222.9 -53.6 6 732 2.5 R 309 -58.6

PN39 Barc Quarry (lava flow) 223.1 -55.6 6 221 4.5 R 306.8 -62.5

PN26 Barc Quarry (lava flow) 212.2 -53.1 5 1042 2.4 R 306.8 -62.1

BAQ* Barc Quarry (lava flow) 216.6 -55.9 15 504 1.7 R 301.4 -61.2

BAQmean Barc Quarry (mean value) 220.3 -54.0 6 320 3.7 R 301.5 -57.6

PN43 Bogata Quarry (lava flow) 232.6 -49.7 5 319 4.3 T 291 -42.1

PN29 Bogata Quarry (lava flow) 221.8 -55.6 8 520 2.4 T 303.6 -62.6

BQA* Bogata Quarry (lava flow) 232.8 -50.6 19 260 2.1 T 296.2 -47.1

2

BQB* Bogata Quarry (lava flow) 225.4 -47.5 6 154 6.2 T 305.4 -50.6

BQmean Bogata Quarry (mean value) 228.3 -50.9 4 283 5.5 T 305.3 -50.6

Notes: D, I: mean-site direction (declination, inclination). N: number of samples (sites for mean). k, α95: Fisher’s

precision parameters and semi-angle of 95 per cent confidence. P: polarity (N=normal, R=reversed, T=transitional).

Plat, Plon: coresponding VGP latitude and longitude. For lava flow means (italics) N represent number of sites:

PVmean = mean value for sites PN30, PN31 and PVA (Pietrele Valley lava flow); BAQ mean = mean values for

sites PN25, PN26, PN37, PN 38 and PN39 (Bârc Quarry lava flow); BQmean = mean value for sites PN29, PN44,

BQA and BQB (Bogata Quarry lava flow). * Data from Panaiotu et al. (2004).

Table 2 Summary of 40Ar/39Ar experiments

Site/Sample N K/Ca Total fusion Plateau Isochron K-Ar age**

total Age (ka) ± 2 39

Ar% MSWD 40

Ar/36

Ari± 2 Age (ka) ±2σ‡ ±2σ

† Age (ka) ±2σ

‡ ±2σ

† (ka) ±2

PVA/2555 9of 9 0.39 686 ± 37 100 0.40 296.6 ± 4.1 683 ± 28 ± 28 674 ± 43 ± 43 668 ± 80

PN34/PN34-1 8of 8 0.63 680 ± 25 100 0.34 294.0 ± 3.0 684 ± 21 ± 21 692 ± 26 ± 26

PN29/PN29-1 8of 8 0.31 794 ± 26 100 0.96 292.0 ± 4.1 799 ± 21 ± 21 823 ± 35 ± 35 1270 ± 200

PN26/PN26 8of 8 0.30 788 ± 32 100 0.37 293.7 ± 3.4 800 ± 25 ± 26 821 ± 46 ± 46 631 ± 80

PN23/PN23-1 8of 8 0.47 1055 ± 13 100 0.56 293.3 ± 4.9 1060 ± 10 ± 10 1066 ± 16 ± 17 1360 ± 140

PN47/2548 9of 9 0.35 1067 ± 31 100 0.07 297.8 ± 10.3 1062 ± 24 ± 25 1056 ± 35 ± 35 1440 ± 130

PN35/2546 10 of 10 0.18 1140 ± 45 100 0.42 287.3 ± 16.0 1142 ± 41 ± 41 1177 ± 78 ± 79 1530 ± 230

PN11/2534 9of 9 0.14 1226 ± 12 100 0.34 293.7 ± 3.4 1221 ± 11 ± 11 1229 ± 17 ± 18 1210 ± 120

Ages relative to 28.201 Ma for Fish Canyon sanidine (Kuiper et al. 2008) using Min et al. (2000) decay constant and 40Ar/36Aratmos = 295.5

‡ 2σ analytical uncertainty

† Fully propagated uncertainty (includes decay constant and analytical uncertainties)

** K-Ar ages from Panaiotu et al. (2004)

Table 3 VGP dispersion and inclination anomaly for the Perşani and South Harghita Mountains Latitude (°N) Age (Ma) Polarity N Plong

(°)

Plat

(°)

Α95

(°)

SB (°) SBlo

(°) SBhi

(°) ΔI

(°)

45.89 - 46.11 0.6-1.2 All combined 26 76.0 84.2 6.8 19.6 16.7 22.3 2.6±5.5

46.10 - 46.36 2-4.3* All combined 53 310.5 86.3 5.7 23.1 20.4 25.9 0.7±4.3

45.89 – 46.36 0.6-4.3 All combined 79 359.9 87.9 4.5 21.9 19.8 24.1 1.6±3.6

Notes: N: number of sites.; Plong: mean VGP longitude; Plat: mean VGP latitude; A95: .semi-angle of 95 per

cent confidence of mean VGP; SB: the between-site VGP dispersion, along with 95% confidence limits (SBlo

, SBhi

);

ΔI : inclination anomaly. * Data from Panaiotu et al. (2012).

40Ar/39Ar ages and paleomagnetism of Quaternary basalts from the East Carpathians.

The onset of the Cobb Mountain normal polarity subchron is dated at 1221 ± 11 ka.

A geomagnetic excursion associated with the Matuyama-Brunhes precursor is identified.