Abstract The thermochemical evolution of oceanic lithosphere and its interaction with the underlying asthenosphere exerts a fundamental control on the dynamics of the Earth system. Since the 1960s, the range, accuracy and spatial coverage of geophysical and geochemical datasets collected in the oceanic realm have substantially increased. These additional constraints have helped to elucidate many aspects of the lithosphere-asthenosphere system, but some apparently contradictory observations have presented additional interpretational challenges. Here, we summarise the merits, limitations and ambiguities of available observational constraints on the thermomechanical evolution of oceanic upper mantle. Newly developed cooling models are generally compatible with these constraints, although there is evidence for systematic differences in behaviour between different oceanic basins. Subsidence, magnetotelluric and seismological observations from the Pacific Ocean are consistent with plate rather than half-space cooling models, whereas results from the Atlantic and Indian Oceans are more equivocal. We provide an overview of proposed mechanisms for seafloor flattening and we show that regional deviations from globally averaged trends can be attributed to asthenospheric temperature variation and to local changes in lithospheric thickness. Although the plate cooling model generally provides a good description of available observations, it is probably a crude approximation of the dynamic processes operating within the thermal boundary layer that underlies oceanic basins. By incorporating mantle density structure inferred from surface wave tomography into more sophisticated convection simulations, we show that the plate model provides a good approximation of the predicted age-dependent behaviour of bathymetric and gravity fields. While the results presented here suggest a unified understanding of the lithosphere-asthenosphere system is within reach, unambiguous evidence for small-scale convection at the base of the lithosphere remains elusive. As a result, the precise mechanism responsible for sea-floor flattening has yet to be identified. It is also unclear why models that incorporate the effect of phase changes such as the garnet-spinel transition tend to fit observations less well than simpler counterparts. These outstanding controversies suggest further research is needed to develop a complete description of the structure and dynamics of the oceanic upper mantle but, encouragingly, the tools at our disposal have never been more powerful.

中文翻译：

---

海洋岩石圈-软流圈系统的结构与动力学

摘要 海洋岩石圈的热化学演化及其与下伏软流圈的相互作用对地球系统的动力学具有根本的控制作用。自 1960 年代以来，在海洋领域收集的地球物理和地球化学数据集的范围、准确性和空间覆盖范围已大大增加。这些额外的限制有助于阐明岩石圈-软流圈系统的许多方面，但一些明显矛盾的观察结果提出了额外的解释挑战。在这里，我们总结了对海洋上地幔热机械演化可用观测约束的优点、局限性和模糊性。新开发的冷却模型通常与这些约束条件兼容，尽管有证据表明不同海洋盆地之间的行为存在系统差异。太平洋的沉降、大地电磁和地震观测与板块而非半空间冷却模型一致，而大西洋和印度洋的结果则更加模棱两可。我们概述了海底变平的拟议机制，并表明与全球平均趋势的区域偏差可归因于软流圈温度变化和岩石圈厚度的局部变化。尽管板块冷却模型通常可以很好地描述可用的观测结果，但它可能是对海洋盆地下方热边界层内运行的动态过程的粗略近似。通过将从表面波层析成像推断出的地幔密度结构结合到更复杂的对流模拟中，我们表明板块模型提供了对测深和重力场的预测年龄相关行为的良好近似。虽然这里提出的结果表明对岩石圈-软流圈系统的统一理解是触手可及的，但岩石圈底部小规模对流的明确证据仍然难以捉摸。因此，导致海底变平的确切机制尚未确定。还不清楚为什么包含相变效应（如石榴石-尖晶石转变）的模型往往不如简单的对应模型更适合观察。