Paleomagnetism and Paleosecular Variations From the Plio‐Pleistocene Golan Heights Volcanic Plateau, Israel

Abstract

Statistical analysis of geomagnetic paleosecular variation (PSV) and time-averaged field has been largely based on global compilations of paleomagnetic data from lava flows. These show different trends in the averaged inclination anomaly (ΔI) between the two hemispheres, with small positive (<2°) anomalies in midsouthern latitudes and large negative (> −5°) anomalies in midnorthern latitudes. To inspect the large ΔI between 20°N and 40°N we augment the global data with a new paleomagnetic data set from the Golan-Heights (GH), a Plio-Pleistocene volcanic plateau in northeast Israel, located at 32-33°N. The GH data set consists of 91 lava flows sites: 40 sites obtained in the 1990s and 51 obtained in this study. The chronology of the flows is constrained by 57 40 Ar/ 39 Ar ages: 39 from previous studies and 18 from this study, which together cover most of the GH plateau. We show that the 1990s data set might be affected by block rotations and does not fully sample PSV. The Plio-Pleistocene pole (86.3°N, 120.8°E, N = 44, k = 25, α 95 = 4.4°), calculated after applying selection criteria with Fisher precision parameter (k) ≥ 100 and number of specimens per site (n) ≥ 5 is consistent with a geocentric axial dipole field and shows smaller inclination anomaly (ΔI = −0.4°) than predicted by global compilations and PSV models. Reexamination of the inclination anomaly in the global compilation using different calculation methods and selection criteria suggests that inclination anomaly values are affected by (1) inclusion of poor quality data, (2) averaging data by latitude bins, and (3) the way the inclination anomaly is calculated. Plain Language SummarySince the early days of paleomagnetic research it has been inferred that the time-averaged structure of the field is a geocentric axial dipole (GAD)-equivalent to a field generated by a dipole at the center of the Earth aligned with its rotation axis. This so-called GAD hypothesis is fundamental in paleomagnetism and is the basis for plate tectonic reconstructions. However, recent studies have suggested persistent departures from a GAD structure, manifested as an anomaly in the inclination angle of the field in northern hemispheric low to middle latitudes. To address this problem, we analyzed 91 basaltic lava flows from the Golan Heights volcanic plateau in Israel (32-33°N), spanning the past 5 Myr. As each basaltic flow captured the direction of the ambient magnetic field when it cooled, these flows provide us information about the averaged direction of the ancient geomagnetic field. Our results show that the averaged field in the Golan Heights is in agreement with a GAD structure. We also show that if we reanalyze the global paleomagnetic data from lava flows spanning the past 10 Myr using stricter selection criteria and a different inclination anomaly calculation method, the data do not support a global non-GAD field.

Impact of the Dead Sea Transform Kinematics on Adjacent Volcanic Activity

“…While K‐Ar ages of the last 1 Ma indicated continuous activity east of the DST (e.g., Giannérini et al, ; Mor, ; Trifonov et al, ; Weinstein et al, ), Ar‐Ar studies from two areas in the northern Golan Heights (Shaanan et al, ; Weinstein et al, ; Shtober‐Zisu et al 2017) suggested that activity was probably intermittent, with cessation periods during 0.6–0.2 Ma and post 0.1 Ma. In this study, we report Ar‐Ar data from different areas in the Golan, which supports the notion that volcanism on the eastern shoulder of the DST (15–20 km from the transform) was periodic and that activity occurred mostly during 0.8–0.6 and 0.2–0.1 Ma (see also Behar et al, ).…”

Covariant Giant Gaussian Process Models With Improved Reproduction of Palaeosecular Variation

“…Untangling the contributions to misfit from biases in the PSV10 data set and issues inherent in GGP models is nontrivial; however, it is clear that the PSV10 data set may contain some biases which affect model construction (cf. ΔI in PSV10 vs. Behar et al, 2019). The approach we have taken prevents the biasing of S by individual studies which may be affected by unrecognized tectonic effects (Opdyke et al, 2015).…”

“…However, significant deviations from either zero inclination anomaly or the antisymmetric anomaly arising from unit vector treatment may be due to persistent non-GAD contributions to the time-averaged field. Alternative methods to calculate ΔI and additional data since PSV10, presented in Behar et al (2019), suggest that the inclination anomaly estimates of PSV10 may be biased due to data selection and inclination anomaly calculation methods. If this is the case, then the persistent non-GAD contribution to the time-averaged field is likely to be negligible, and the BB18 model is optimal.…”

Uncertainty Propagation in Hierarchical Paleomagnetic Reconstructions

“…As an example of our proposed approach, we have reanalyzed two site mean directions from the Golan Heights volcanic plateau, Israel, presented by Behar et al (2019). We do not consider the experimental protocol of Behar et al 2019; rather, we work simply with their PCA-quantified ChRMs downloaded from Journal of Geophysical Research: Solid Earth 10.1029/2020JB019488 Note.…”

“…As an example of our proposed approach, we have reanalyzed two site mean directions from the Golan Heights volcanic plateau, Israel, presented by Behar et al (2019). We do not consider the experimental protocol of Behar et al (2019); rather, we work simply with their PCA‐quantified ChRMs downloaded from the Magnetics Information Consortium (MagIC) database (https://doi.org/10.7288/V4/MAGIC/16676). The mean direction of site GH01 reported by Behar et al (2019) was based on seven specimens with low MADs (Table 2).…”