Improving late Holocene radiocarbon-based chronologies by matching paleomagnetic secular variations to geomagnetic field models - Examples from Nam Co (Tibet) and Lake Kalimpaa (Sulawesi)

Precise age control is a prerequisite for reliable paleoenvironmental reconstruction. Almost all Holocene chronologies of lacustrine sequences are based on radiocarbon dating. Most reliable ages are obtained from fragile terrestrial vegetation remnants (e.g., leaves) as these neither yield a reservoir or hard water effect nor are reworked as they would be destroyed during this process. However, in many records no terrestrial plant remains or macro remains in general are found in the sediments. In this case the only option to establish a radiocarbon based chronology is to date bulk sediment. This sediment sometimes contains reworked material or, in hard water lakes, aquatic organic remains which contain old carbon - both resulting in older ages. Here we present two records dated by bulk material showing these inconveniences. Nam Co is a hard water lake on the Tibetan Plateau (30.5°N, 91°E). Previous studies from various sites in this lake revealed different hard water effects. A radiocarbon date of bulk sediment from the top of a gravity core yielded an age of 1420 ±40 BP for the sediment/water interface. In order to test if this value can be constantly extrapolated back in time we established a reservoir corrected chronology with this value. Subsequently, we compared inclination and declination data to the CALS3k.3 and the CALS7k geomagnetic field models which currently provide the best representation of the late Holocene geomagnetic field. This showed excellent accordance for the period covered by the gravity core, i.e., ~4000 cal BP. Lake Kalimpaa (1.3°S, 120.3°E) is located on the island of Sulawesi, Indonesia. Radiocarbon-dated bulk samples also showed inconsistencies. Most likely not only the autochthonous carbon fraction was dated but also some allochthonous organic matter was incorporated into the samples. If this is the case, all ages are tentatively too old. Following a conservative approach, in order to minimize this error, only the youngest dates were chosen for age-depth modeling and a linear interpolation was applied. Subsequently, inclination, declination and intensity were compared to the CALS3k.3 model also showing an excellent match from ~1350 cal BP to the present. Although, the age-depth models of both lakes are conservative, comparisons of paleomagnetic data with geomagnetic spherical harmonic models support this approach. This leads to the conclusion that the presented chronologies are suited for further paleoenvironmental investigations. This is important as both areas lack well-dated records. On the other hand, our data also support the validity of the CALS-models for the past ~4000 and ~1350 cal BP on the Tibetan Plateau and Indonesia, where paleomagnetic data are very scarce.