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.

The geocentric axial dipole (GAD) hypothesis states that the Earth's magnetic field may be approximated by an Earth-centric dipole aligned with the rotation axis. This hypothesis is fundamental for paleogeographic reconstructions of the tectonic plates. While global paleomagnetic directions from the last 5 million years recover a predominately GAD field structure, paleointensity estimates over the same time period do not. In this study, we reexamine the paleomagnetic field structure in the Erebus Volcanic Province, Antarctica, and recover a robust data set of directional and intensity data. We then compare the paleopole and average dipole moment against a GAD field structure and model predictions of paleosecular variation.

Twenty‐two sites, subjected to an IZZI‐modified Thellier‐Thellier experiment and strict selection criteria, recover a paleomagnetic axial dipole moment (PADM) of 62.2 ± 30.6 ZAm2 in Northern Israel over the Pleistocene (0.012–2.58 Ma). Pleistocene data from comparable studies from Antarctica, Iceland, and Hawaii, re‐analyzed using the same criteria and age range, show that the Northern Israeli data are on average slightly higher than those from Iceland (PADM = 53.8 ± 23 ZAm2, n = 51 sites) and even higher than the Antarctica average (PADM = 40.3 ± 17.3 ZAm2, n = 42 sites). Also, the data from the Hawaiian drill core, HSDP2, spanning the last half million years (PADM = 76.7 ± 21.3 ZAm2, n = 59 sites) are higher than those from Northern Israel. These results, when compared to Pleistocene results filtered from the PINT database (http://www.pintdb.org) suggest that data from the Northern hemisphere mid‐latitudes are on average higher than those from the southern hemisphere and than those from latitudes higher than 60°N. The weaker intensities found at high (northern and southern) latitudes therefore, cannot be attributed to inadequate spatiotemporal sampling of a time‐varying dipole moment or low quality data. The high fields in mid‐latitude northern hemisphere could result from long‐lived non‐axial dipole terms in the geomagnetic field with episodes of high field intensities occurring at different times in different longitudes. This hypothesis is supported by an asymmetry predicted from the Holocene, 100 kyr, and 5 million year time‐averaged geomagnetic field models.