P2C2: Late Holocene Changes in the SPCZ from Algal Lipid D/H in Lakes Across Oceania

热带水循环影响着全球气候。在过去的30年里，它发生了显著的变化。在通常雨水充沛的热带地区，降雨量增加了，而在干旱地区，降雨量减少了。与此同时，两个半球的热带地区向极地扩张，导致大陆上沙漠地区的扩张。虽然热带降雨在许多气候模式中表现不太好，但简单的物理论据和广泛的模式表明，在全球变暖的情况下，这些趋势可能会继续下去。目前尚不清楚当前热带降雨的变化是否超出了工业化前气候的自然变化范围。这是因为主要来自卫星的广大热带海洋的降雨数据仅限于过去35年。如此短暂的天气记录不足以确定最近的降雨变化是否异常。通过将热带南太平洋地区的降雨记录从目前的35年延长到2000年，该项目将产生必要的数据，以评估过去35年的降雨变化是否异常，如果是，是否可以归因于大气中温室气体的积累。它们还将提供必要的数据，以确定哪些气候模式能够准确地再现热带降雨模式，从而确定哪些模式可能提供最可靠的预报。为了将热带南太平洋最近的降雨变化置于更长期的背景下，该项目将对整个南太平洋辐合带（SPCZ）过去2000年的降雨量进行定量重建。（3β,4α,5α,22E）-4,23-二甲基麦角甾醇-22-烯-3-醇，dinosterol的氢同位素比值（2H/1H，或D/H）将在现代SPCZ地区8个岛屿的沉积物岩心中进行测量。甲藻甾醇几乎完全由某些属的甲藻微藻合成。由于降雨的D/H比与热带地区的降雨率呈反比关系，即所谓的“数量效应”，我们发现整个SPCZ地区21个地表沉积物中恐龙醇（δDdino）的D/H比与降雨率高度相关，关系为&amp；#948;Ddino=-16.6（±2.45）*P-180（±15.3）[R2=0.71， N=21]，其中P为毫米/天的降雨率。蒙特卡罗误差分析表明，使用这种技术可以估计出0.3毫米/天的不确定性的降雨率。来自所罗门群岛、瓦努阿图、瓦利斯富图纳和斐济的沉积物岩心，代表了5个研究区域中的4个，已经被收集起来。作为该项目的一部分，将从萨摩亚和美属萨摩亚的湖泊中收集新的岩心。初步结果表明，在1000年前的中世纪暖期到600年前的小冰期过渡期间，SPCZ西侧的所罗门群岛的降雨量增加了2.6 mm/d，而SPCZ东侧的沃利斯和富图纳岛的降雨量减少了1.1 mm/d。

Non-Technical AbstractThe tropical water cycle influences the global climate. Over the last 30 years it has undergone remarkable changes. Rainfall has increased in the regions of the tropics that normally receive lots of rain, and decreased in the drier regions. This was accompanied by a poleward expansion of the tropics in both hemispheres that has led to an expansion of desert regions over continents. Although tropical rainfall is poorly represented in many climate models, simple physical arguments and a wide range of models suggest these trends are likely to continue under global warming scenarios. It is unknown at this point whether the current changes in tropical rainfall are outside the range of natural variations of the pre-industrial climate. This is because rainfall data over the vast tropical oceans, which is derived primarily from satellites, is limited to the last 35 years. Such a brief weather record is insufficient to establish whether the recent changes in rainfall are anomalous. By extending the record of rainfall in the tropical South Pacific from the current 35 years to 2000 years, this project will produce the data necessary to assess whether rainfall changes over the last 35 years are unusual, and if so, whether they can be attributed to the buildup of greenhouse gases in the atmosphere. They will also provide the data necessary to determine which climate models are able to accurately reproduce tropical rainfall patterns, and therefore which models are likely to provide the most reliable forecasts.Technical AbstractIn order to place recent changes in rainfall over the tropical South Pacific Ocean into a longer-term context, this project will produce quantitative reconstructions of rainfall throughout the South Pacific Convergence Zone (SPCZ) that span the last 2,000 years. The hydrogen isotope ratio (2H/1H, or D/H) of (3β,4α,5α,22E)-4,23- dimethylergost-22-en-3-ol, dinosterol, will be measured in sediment cores from 8 islands across the modern SPCZ region. Dinosterol is synthesized almost exclusively by certain genera of dinoflagellate microalgae. Owing to the inverse relationship between D/H ratios in rain and rainfall rates in the tropics, the so-called "amount effect," we find that D/H ratios of dinosterol (δDdino) in 21 surface sediments throughout the SPCZ region are highly correlated with rainfall rates according to the relationship δDdino=-16.6(±2.45)*P-180(±15.3) [R2=0.71, N=21], where P is the rainfall rate in mm/day. A Monte Carlo analysis of errors indicates that rainfall rate estimates with 0.3 mm/day uncertainty are possible using this technique. Sediment cores from the Solomon Islands, Vanuatu, Wallis & Futuna, and Fiji, representing 4 of the 5 study regions have already been collected. New cores from lakes in Samoa and American Samoa will be collected as part of this project. Initial results indicate a 2.6 mm/day increase in rainfall in the Solomon Islands, on the western side of the SPCZ, and a 1.1 mm/day decrease in Wallis and Futuna, on the eastern side of the SPCZ, during the transition from the Medieval Warm Period 1,000 years ago to the Little Ice Age 600 years ago.