The elevated D14C levels during the last ice age show a good correlation with the strong climatic changes during the last deglaciation. However, carbon cycle models fail to explain the major part of the increased 14C concentrations in the atmosphere during the last ice age (e.g. Köhler et al., 2006).The geomagnetic modulation of the 14C production rate is partly responsible for the elevated 14C level during the last 50 kyr. However, the errors of the geomagnetic field reconstructions are large. In addition, geomagnetic field records that can well explain the elevated D14C during the last ice age don't yield a good explanation for D14C during the climatically stable Holocene period. Alternative estimates of the 14C production rate based on 10Be from polar ice cores can explain D14C during the Holocene very well. However, the differences to D14C during MIS2 and MIS3 seem too large to be explained by carbon cycle changes alone.We will discuss potential causes for the disagreement between measured and modeled D14C during the last 50 kyr. In the view of the new dating of the Greenland ice cores (Svensson et al., QSR2006) we will reassess the reliability of the 10Be record as a proxy for the global changes in the 10Be and 14C production rates and discuss the implications for the Radiocarbon-based reconstructions of past ocean circulation.References:Köhler, P., Muscheler, R., Fischer, H. (2006). A model-based interpretation of low frequency changes in the carbon cycle during the last 120,000 years and its implications for the reconstruction of atmospheric D14C, Geochemistry Geophysics Geosystems, doi:10.1029/2005GC001228, in press.Svensson et al. (2006). The Greenland Ice Core Chronology 2005, 15-42 ka.Part 2: Comparison to other records, Quaternary Science Reviews 2006. in press.