Collaborative Research: Understanding Contemporary Variability in Antarctic Sea Ice: Ensemble Reconstruction of Sea Ice Extent and Concentration for the 20th Century

与北极相反，自1979年以来，南极大部分地区的海冰覆盖面积有所增加。在过去6年中，有4年的海冰面积达到了创纪录的最大值，但2015-16年夏季的海冰覆盖面积却创下了历史新低。虽然令人印象深刻，但由于连续的卫星记录只能追溯到1978年，因此很难在长期变率的背景下评估这些最近记录的意义。连续海冰记录的有限长度是确定观测到的洋流变化是仅仅由于自然变率还是代表一种强迫的人为反应的一个重要混淆因素。因此，对南极海冰趋势的科学认识仍然很差，对未来南极海冰趋势预测的信心也很差。为了应对这一挑战，该项目试图利用卫星观测到的海冰、海平面压力、热带海面温度、ENSO指数、一些代理数据（冰芯等）和原位南大洋温度数据之间的关系，重建海冰范围和海冰浓度。该研究的目的是尽可能准确地收集和组合这些辅助记录，同时保留与现有现场数据中固有不确定性相关的可变性。将使用一系列统计方法来模拟卫星时代海冰数据之间的关系，包括灵活回归、贝叶斯和多元动态时空（MDST）方法。这项研究将为加州大学洛杉矶分校和俄亥俄大学的本科生和一名研究生提供跨学科培训。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

In contrast to the Arctic, sea ice cover in most Antarctic regions has increased since 1979. The area-integrated total sea ice extent grew to record maximum values in four of the last six years, yet the 2015-16 summer was marked by record low ice cover. While impressive, it is difficult to assess the significance of these very recent records in the context of longer term variability, since the continuous satellite record only dates back to 1978. The limited length of the continuous sea ice record, is a significant confounding factor in ascertaining whether the observed current changes are due to natural variability alone, or represent a forced anthropogenic response. As a result, the scientific understanding of the Antarctic sea ice trends remains poor, as does confidence in projections of future Antarctic sea ice trends. To address this challenge, this project seeks to reconstruct sea ice extent and sea ice concentration, using the relationships between satellite-observed sea ice, sea level pressure, tropical sea surface temperature, ENSO indices, some proxy data (ice cores, etc.), and in situ Southern Ocean temperature data. The aim of the study is to collect and combine these ancillary records as accurately as possible while retaining the variability associated with the intrinsic uncertainty in the available field data. A range of statistical methods for modelling the relationship between satellite era sea-ice data using flexible regression, Bayesian and multivariate dynamic spatial temporal (MDST) methods will be used.The study will entrain cross-disciplinary training of undergraduate and a graduate student at UCLA and Ohio University.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.