The sun is the most important energy source for the Earth’s climate system, however, influences of solar variations on climate are not fully understood. Studying past solar activity can improve our understanding of what role the sun plays in Earth’s climate. Cosmogenic radionuclides, such as 10Be in ice cores, are powerful tools to reconstruct past solar activity before the period of direct telescope observations. Because the atmospheric production rates of 10Be depend on the strengths of the solar and geomagnetic field, hence measurements of 10Be in ice can provide useful information on variabilities of these two magnetic fields back in time. However, the interpretation of 10Be data is challenged due to the influences of transport and scavenging processes on 10Be from Earth’s atmosphere to the natural archives (e.g. ice cores). Unidentified climate imprints on 10Be records could lead to errors in solar and geomagnetic reconstructions.
This project aims to improve our understanding of the climate effects on 10Be through analysis of high-resolution 10Be and 7Be data, in order to improve solar and geomagnetic reconstructions. We present a well-defined seasonally resolved 10Be record for the period 1887-2002 from a NEEM firn core (NEEM07S1) in northwestern Greenland. Through analyzing the sub-annual δ18O records at the NEEM site, we identify the seasonal signals with 30% accumulation for November-April (winter) and 70% accumulation for May-October (summer). Both summer and winter 10Be data reflect the production signal induced by solar modulation of galactic cosmic rays. Superimposed on this solar signal we find that the tropopause pressure over 30°N represents an important factor influencing NEEM 10Be concentrations. Summer 10Be concentration, on average, is higher than winter, which could be attributed to the effects of the stratospheric intrusion of 10Be. This stratospheric intrusion of beryllium is also supported by the study of weekly-resolved air 7Be records over Europe. Furthermore, by comparing the NEEM 10Be record with other Greenland 10Be records over the last 100 years, we identify that the 10Be values from the Dye3 ice core after 1958 are unusually low, which are resulting from a data quality issue instead of meteorological influences. This period of unusually low Dye3 10Be values can lead to a normalization problem when connecting radionuclide records to modern levels of solar modulation estimated from neutron monitor data over the past 70 years. We found that disagreements of different solar reconstructions based on Greenland and Antarctica 10Be records can be partly attributed to this data problem. Finally, we present a geomagnetic dipole moment reconstruction for the period from 11.7 ka BP (before present AD 1950) to 108 ka BP based on a new NEEM 10Be record and published GRIP 10Be and 36Cl records. With a first-order correction of cosmogenic radionuclides data using climate proxy data, the “climate correction” results lead to an improved agreement with independent reconstructions compared to simply using radionuclide data. Therefore, with this linear correction method, geomagnetic dipole moment reconstructions based on cosmogenic radionuclides data from ice cores can be extended back in time when there is a strong climate signal in radionuclides data.