Very low archeointensities found for Suriname from 670 to 960 AD

Geomagnetic field models predict dipole moments higher than today for the past ~3000 years (e.g. Korte et al., 2011). These models are based on data that is geographically biased towards certain regions on the Northern hemisphere; the Southern hemisphere and equatorial regions are currently underrepresented. Here we present archeointensity data obtained from potsherds from the Wageningen-1 mound on the Northwest coast of Suriname (5.8°N 56.7°W); this settlement of the Hertenrits culture was excavated in the 1970s. We obtained potsherds from six levels between 0.75 - 3 meter below the surface, corresponding to 960 and 670 AD respectively (Versteeg, 2003, p. 269-270) based on four radiocarbon datings. The remanent magnetization is carried by small pseudo-single domain grains, is stable, and small overprints are typically removed below 150 °C. Archeointensities were obtained using the IZZI-Thellier protocol; laboratory alteration was monitored by pTRM checks and pTRM Tail checks. All 23 technically acceptable archeointensities (out of 52 samples measured) are very low: between 8.0 and 18.5 μT. A trend is suggested from an average of 10-12 μT at ~700 AD, to ~18 μT at ~950 AD. This corresponds to VADMs from ~28 at ~700 AD to ~47 ZAm2 at ~950 AD. This dramatically contradicts field model predictions that range from 38-39 μT (PFM9K.1b) to 38-46 μT (SHA.DIF.14k) for this location and the same ages. Our results predate the occurrence of the South Atlantic Anomaly (SAA) by 500-1000 years; without a longer record it is difficult to decide whether our reported lows are a different discrete geomagnetic feature, or whether they are a precursor of the South Atlantic Anomaly.