Beryllium-10 concentration in the EPICA Dome C core between 2384.36 and 2626.25 m deep (269-355 ka on the EDC3 age scale)

Ice cores are exceptional archives which allow us to reconstruct a wealth of climatic parameters as well as past atmospheric composition over the last 800 kyr in Antarctica. Inferring the variations in past accumulation rate in polar regions is essential both for documenting past climate and for ice core chronology. On the East Antarctic Plateau, the accumulation rate is so small that annual layers cannot be identified and accumulation rate is mainly deduced from the water isotopic composition assuming constant temporal relationships between temperature, water isotopic composition and accumulation rate. Such an assumption leads to large uncertainties on the reconstructed past accumulation rate. Here, we use high-resolution beryllium-10 (10Be) as an alternative tool for inferring past accumulation rate for the EPICA Dome C ice core, in East Antarctica. We present a high-resolution 10Be record covering a full climatic cycle over the period 269 to 355 ka from Marine Isotope Stage (MIS) 9 to 10, including a period warmer than pre-industrial (MIS 9.3 optimum). After correcting 10Be for the estimated effect of the palaeomagnetic field, we deduce that the 10Be reconstruction is in reasonably good agreement with EDC3 values for the full cycle except for the period warmer than present. For the latter, the accumulation is up to 13% larger (4.46 cm ie per yr instead of 3.95). This result is in agreement with the studies suggesting an underestimation of the deuterium-based accumulation for the optimum of the Holocene (Parrenin et al., 2007, doi:10.5194/cp-3-243-2007). Using the relationship between accumulation rate and surface temperature from the saturation vapour relationship, the 10Be-based accumulation rate reconstruction suggests that the temperature increase between the MIS 9.3 optimum and present day may be 2.4 K warmer than estimated by the water isotopes reconstruction. We compare these reconstructions to the available model results from CMIP5-PMIP3 for a glacial and an interglacial state, i.e. for the Last Glacial Maximum and pre-industrial climates. While 3 out of 7 models show relatively good agreement with the reconstructions of the accumulation-temperature relationships based on 10Be and water isotopes, the other models either underestimate or overestimate it, resulting in a range of model results much larger than the range of the reconstructions. […]

冰芯（Ice cores）是极为卓越的古气候档案，可用于重建南极地区过去80万年（kyr，kiloyear）内的各类气候参数与古大气组分。推断极地地区古降雪累积速率的变化，无论是对于记录古气候还是冰芯年代学研究而言，均至关重要。在东南极高原，降雪累积速率极低，无法识别年层，因此累积速率主要通过水同位素组成推算，其前提假设是温度、水同位素组成与累积速率之间存在恒定的时间关系。这类假设会导致重建的古累积速率存在较大不确定性。本文以高分辨率铍-10（beryllium-10, ¹⁰Be）作为替代手段，对东南极的EPICA Dome C冰芯开展古累积速率重建。我们构建了一条覆盖完整气候周期的高分辨率¹⁰Be记录，时间跨度为269至355千年，对应海洋同位素阶段（Marine Isotope Stage, MIS）9至10，其中包含一段较工业革命前更为温暖的时段（MIS 9.3暖期最佳期）。在对¹⁰Be记录进行古地磁场效应校正后，我们发现，除了较现代更为温暖的时段外，整条气候周期内的¹⁰Be重建结果与EDC3的数值均较为吻合。针对该暖期，累积速率较估算值高出最多13%（即年累积厚度从3.95厘米冰当量升至4.46厘米冰当量）。这一结果与相关研究的结论一致，即基于氘（deuterium）同位素推算的全新世暖期累积速率存在低估（Parrenin et al., 2007, doi:10.5194/cp-3-243-2007）。结合饱和水汽关系得到的累积速率与地表温度之间的关系，基于¹⁰Be的累积速率重建结果表明，MIS 9.3暖期与现代之间的温度增幅，可能比通过水同位素重建得到的估算值高出2.4开尔文。我们将这些重建结果与耦合模式比较计划第五阶段-古气候模式比较计划第三阶段（CMIP5-PMIP3）针对冰期与间冰期状态（即末次冰盛期（Last Glacial Maximum）与工业革命前气候）的现有模式模拟结果进行了对比。7个模式中有3个与基于¹⁰Be和水同位素的累积-温度关系重建结果吻合度相对较好，其余模式要么低估要么高估了该关系，导致模式结果的分布范围远大于重建结果的分布范围。……