Data on the geology and paleomagnetism of the Miocene on the southern margin of the Asahi Mountains in Northeast Japan

Table S1. Major and trace element analyses. [Footnote] * Total Fe as Fe₂O₃. BAM: Lavas of the Bentozawabashi Andesite Member, INT: intrusive rocks, KF: Welded tuff of the Kitaoguni Formation, MTM: Welded tuff of the Myozawagawa Tuff Member. 

Table S2. Petrographic summary of the analyzed rocks, including phenocrysts and groundmass minerals. [Footnote] BAM: Bentozawabashi Andesite Member, INT: intrusive rocks, KF: Kitaoguni Formation, MTM: Myozawagawa Tuff Member, Af: alkali feldspar, Ap: apatite, Bi: biotite, Cpx: clinopyroxene, Gl: glass, Hb: hornblende, Kf: K-feldspar, Ol: olivine, Opx: orthopyroxene, Ore: ore mineral, Pl: plagioclase, Qz: quartz, Zir: zircon. 

Table S3. K–Ar ages of a groundmass separate from an andesite intrusion at Site IN3.  [Footnote] * 10^－8 cc STP/g.

Table S4. Site-mean remanent magnetization directions. [Footnote] Lat, Long: site latitude and longitude; Str1/dip1: strike/dip of bedding plane measured at outcrop (righthand rule for strike, corrected for current local geomagnetic declination); N: number of specimens used to calculate site-mean direction; D_a/I_a: in situ declination/ inclination; D_b/I_b: declination/inclination after correcting for the plunge of the Myozawa Syncline; Str₂/dip₂: strike/ dip of bedding plane after correcting for the plunge of the syncline; D_c/I_c: declination/inclination after untilting the plunge-corrected limbs of the syncline, except for the Kitaoguni Formation, which was corrected using Str₁/dip₁; a₉₅: 95% confidence bounds; k: precision parameter; V_lat, V_long: latitude and longitude of virtual geomagnetic pole (poles for the andesite intrusions were determined using D_b/I_b, and those for other units using D_c/I_c); ND: not determined. *For B3, B5, and B6a, a great circle (GC) method was used for site mean estimation (B3: one direction + five GCs, B5: six GCs, B6a: three directions + two GCs).

Table S5. Isothermal remanent magnetization (IRM) coercivity components. [Footnote] SIRM: saturation isothermal remanent magnetization, Contr.: contribution to total SIRM, H0.5: mean coercive force (at which half of the SIRM value is reached), DP: dispersion parameter (one standard deviation of logarithmic distribution). Analyses were conducted using IRM-CLG version 1.0 (Kruiver et al., 2001) and IRMUNMIX version 2.2 (Heslop et al., 2002).

Fig. S1. Correlation between lithostratigraphic divisions of the Miocene strata in the study area.

Fig. S2. Outcrop photos. (a) Rhyolitic welded tuff of the Kitaoguni Formation at Site K7. (b) Unsorted breccia of the main facies of the Meganebashi Formation at 38.0477°N, 139.8499°E. (c) Platy-jointed andesitic lava of the Bentozawabashi Andesite Member at Site B2b. (d) Rhyolitic welded tuff of the Myozawagawa Tuff Member at Site M2. (e) Heavily bioturbated silty sandstone of the Myozawabashi Formation at 38.0477°N, 139.8699°E. (f) Wellbedded dark gray shale of the Numazawa Formation at 38.0392°N, 139.8796°E. (g) Unconformity between the Myozawabashi and Numazawa formations at 38.0393°N, 139.8773°E. (h) Sketch of the unconformity shown in (g).

Fig. S3. (a) Stratigraphic columns for the eastern part of the Oguni area. See (b) for locations. Stratigraphic horizons of the paleomagnetic samples (sites K1 and K2 in the Kitaoguni Formation and M1–M4 in the Myozawagawa Tuff Member) and the horizon with Arcid-Potamid fauna and mangrove pollen fossils (Sato et al., 1989; Abe et al., 2006; Yamanoi et al., 2008) are also shown. (b) Topographic map showing the locations of the columns shown in (a) and the Arcid-Potamid fauna locality. (c) Stratigraphic columns for the Bentozawabashi Andesite Member mapped along three sections (3B, 4B, and 5B; see (a) for locations). Stratigraphic horizons of the paleomagnetic samples from the Bentozawabashi Andesite Member (17 sites) are shown.

Fig. S4. Orthogonal plots of alternating-field (AF) and thermal (Th) demagnetization results. Solid and open circles represent vectors on the horizontal and vertical N–S planes. Values in parentheses are natural remanent magnetization (NRM) intensities (×10^－3 A/m). BAM: Bentozawabashi Andesite Member, KF: Kitaoguni Formation, MTM: Myozawagawa Tuff Member, RWT: rhyolitic welded tuff.

Fig. S5. Isothermal remanent magnetization (IRM) component analyses for four rock samples. See Table S5 for IRM component values. Panels on the left are linear acquisition plots (LAP), and panels on the right are gradients of acquisition plots (GAP). Open squares are data points, thin solid lines are coercivity component 1, thin dashed lines are coercivity component 2, and thick red lines are the sum of the individual components. Analyses were conducted using IRM-CLG version 1.0 (Kruiver et al., 2001) and IRMUNMIX version 2.2 (Heslop et al., 2002).