Whole-mantle body-wave tomography under the hotspot of La Réunion

留尼旺岛是世界上最活跃、寿命最长的火山之一。它的时间渐进，5500公里长的热点轨道出现在印度的德干陷阱，地球上最大的洪水玄武岩省之一，当它们爆发时，对全球气候和生命产生了巨大的影响。我们想建立热源和“管道系统”，它为这个强大的热点提供了6500万年的燃料。通过主要由DFG和ANR资助的德国-法国合作项目RHUM-RUM，自2011年以来，我们在西印度洋部署了90多个宽带海底和沙地地震仪。RHUM-RUM的目标是对从地表到核-幔边界的地幔柱（或缺乏地幔柱）进行地震成像，并从物质和热流的角度来理解结果。这包括羽流管道中的垂直上升流，以及扩散到西印度洋上地幔的任何横向流。我们在这里申请一个博士学位。学生将获得的波形数据转换为地幔的三维全地幔P波和S波模型。我们将使用多频层析成像，一种最先进的波形反演方法。技术上的新颖性或接近新颖性包括（i）在体波的整个可用频率范围内对海底和岛屿地震图进行完全有限频率处理和反演。(ii)将RHUM-RUM数据嵌入到半球尺度的全地幔模型中，通过与半球的所有其他宽带数据进行联合反演。(iii)包括核心衍射波，由于完全数值波形建模。其目标是显著改善最低地幔的结构，这是深地幔柱的假定基础区域。

The island of La Reunion is one of the most active and long-Iived volcanoes in the world. Its time- progressive, 5500-km long hotspot track emerges from the Deccan Traps of India, one of the largest flood basalt provinces on earth, which had a massive impact on global climate and life when they erupted. We want to establish the heat source and "plumbing system" that has been fueling this powerful hotspot for 65 million years. Through the RHUM-RUM project, a German- French effort funded mainly by the DFG and the ANR, we have deployed more than 90 broadband, ocean-bottom and Sand seismometers across the western Indian Ocean since 2011.The goal of RHUM-RUM is to seismically image a mantle plume - or lack of plume - from the surface down to the core-mantle boundary, and to understand the results in terms of material and heat flow. This comprises vertically ascending flow in the plume conduit, as well as any lateral flow spreading into the upper mantle of the western Indian Ocean. Here we are asking for funding for a Ph.D. Student to work the acquired waveform data into three-dimensional, whole-mantle P- and S-wave models of the mantle. We will be using multi-frequency tomography, a state-of-the art waveform Inversion method. Technical novelties or near-novelties include(i) the fully finite-frequency processing and Inversion of ocean-bottom and island seismograms across the entire usable frequency range of body waves.(ii) embedding of the RHUM-RUM data into a hemisphere-scale, whole-mantle model by joint Inversion with all other broadband data available for the hemisphere.(iii) the inclusion of core-diffracted waves, thanks to fully numerical waveform modeling. The goal is a significant improvement in structure of the lowermost mantle, the presumed footing region of deep mantle plumes.