Paper Push: 2026-07-29
Back to date list
每日论文推送：BGC-Argo、海色/海洋光学、海洋热浪与碳泵Daily Paper Push: BGC-Argo, ocean colour/ocean optics, marine heatwaves and carbon pump
本期由 GitHub Actions 自动检索生成：Nature/Science 系列优先，其次是用户指定重点期刊，再补充重点关注团队的新论文，最后纳入其他相关期刊；历史去重后保留 3 篇，不超过每日 50 篇上限。 This issue was generated automatically by GitHub Actions: Nature and Science series first, then the user-defined priority journals, then new papers from the focused team, followed by other relevant journals as topical supplements. After deduplication, 3 papers remain, below the daily limit of 50.
Download Word summary
无 mechanism sketch 图。今天的意大利语卡片： No mechanism sketch figure today. Daily Italian card:
每日一句意大利语Daily Italian
O luce etterna che sola in te sidi.
Dante, Commedia, Paradiso XXXIII, 124; Italian original from Kalliope
终章中但丁呼唤“永恒之光”。它把《天堂篇》的视觉和神学主题推到顶点。
Near the end, Dante addresses the eternal light. It brings Paradiso's visual and theological language to its height.
趋势总结Trend Summary
本期重点关注 BGC-Argo、海色遥感/海洋光学、海洋热浪、浮游植物垂向结构和碳泵过程。筛选逻辑不再只限于重点期刊；当高影响力期刊当天新增较少时，会额外检索重点关注团队作者的新论文，并用海洋、海色/光学和碳循环关键词过滤，再从其他相关期刊补充候选论文。
This issue focuses on BGC-Argo, ocean-colour remote sensing, ocean optics, marine heatwaves, vertical phytoplankton structure and carbon-pump processes. The selection is no longer limited to priority journals; when few high-impact papers are newly available, the workflow also checks focused-team authors and filters those papers with ocean, ocean-colour/optics, and carbon-cycle keywords before adding other relevant journals as supplements.
重点期刊：按影响力和相关性排序Key journals: ordered by impact and relevance
1. Improving cross-flight plastic litter segmentation with ConvNeXt V2 U-Net for UAV SWIR hyperspectral imagery
作者Authors: Lahcene Mamen; Ahmed Tibermacine; Soufyane Bouchelaghem; Marco Balsi; Monica Moroni
发表月份Publication month: 2026-07 2026-07
Frontiers in Marine Science · DOI: 10.3389/fmars.2026.1909051
关键词Tags: ocean colour ocean colour
摘要：塑料垃圾是一种持久的环境污染形式，准确的空间监测对于生态系统评估、污染缓解和循证管理至关重要。无人机短波红外高光谱成像可以检测具有高空间和光谱细节的塑料材料，但其操作使用仍然受到不同飞行日收集的采集数据之间的巨大差异的限制，其中照明、曝光、天气和表面背景的变化引入了立方体特定的辐射差异，从而降低了交叉飞行的概括性。这项研究通过在分割之前使用其自己的光谱统计数据标准化每个高光谱立方体，将塑料垃圾分割作为遥感和数据协调问题来解决，从而在不依赖训练集统计数据的情况下减少采集特定的偏差。 在此协调输入上，我们引入了一个紧凑的基于 ConvNeXt V2 的 U-Net，适用于 60 个保留的 SWIR 频段，并通过金字塔池合并多尺度上下文信息和 CBAM 解码器细化以改善背景抑制。该框架在公共多飞行无人机 SWIR 基准上进行评估，并与 PSPNet、SegFormer、DeepLabV3、U-Net、UNet++ 和我们之前的注意力门控残差 U-Net 进行比较。与我们之前的工作相比，所提出的模型将平均 Dice 分数从 64. 6% 提高到 78. 2%，平均 IoU 从 49. 0% 提高到 64. 5%；根据使两个模型的训练数据预算相等的匹配再训练协议，平均留一立方 IoU 从 0. 37 上升到 0. 69。 在困难的采集条件下，这一改进尤其重要，因为之前的模型更容易受到过度曝光、光谱对比度降低和背景混乱的影响。增量架构研究进一步表明，与具有挑战性的分布外立方体上的 ConvNeXt V2 基线相比，金字塔池化和 CBAM 细化改善了可塑性检测和背景抑制之间的平衡，提高了精度、特异性、AUC-ROC、AUC-PR 和 IoU。对开发立方体的多种子分析显示对随机初始化的敏感性较低，这表明这些运行中的变化更多地是由采集条件驱动的，而不是由初始化驱动的。总体而言，所提出的框架为现实交叉飞行条件下基于无人机的高光谱塑料垃圾监测提供了一种紧凑且可转移的方法。
Abstract: Plastic litter represents a persistent form of environmental pollution, and accurate spatial monitoring is essential for ecosystem assessment, pollution mitigation, and evidence-based management. UAV short-wave infrared hyperspectral imaging can detect plastic materials with high spatial and spectral detail, but its operational use remains limited by strong variability between acquisitions collected on different flight days, where changes in illumination, exposure, weather, and surface background introduce cube-specific radiometric differences that reduce cross-flight generalisation. This study addresses plastic litter segmentation as both a remote sensing and data harmonisation problem by standardising each hyperspectral cube using its own spectral statistics before segmentation, reducing acquisition-specific bias without relying on training-set statistics. On this harmonised input, we introduce a compact ConvNeXt V2-based U-Net adapted to the 60 retained SWIR bands, with pyramid pooling to incorporate multi-scale contextual information and CBAM decoder refinement to improve background rejection. The framework is evaluated on a public multi-flight UAV SWIR benchmark and compared with PSPNet, SegFormer, DeepLabV3, U-Net, UNet++, and our previous attention-gated residual U-Net. Compared with our previous work, the proposed model improves the mean Dice score from 64.6% to 78.2% and the mean IoU from 49.0% to 64.5%; under a matched retraining protocol that equalises the training-data budgets of the two models, the mean leave-one-cube-out IoU rises from 0.37 to 0.69. The improvement is especially important under difficult acquisition conditions, where the previous model was more affected by overexposure, reduced spectral contrast, and background confusion. The incremental architectural study further shows that pyramid pooling and CBAM refinement improve the balance between plastic detection and background rejection, increasing precision, specificity, AUC-ROC, AUC-PR, and IoU compared with the ConvNeXt V2 baseline on a challenging out-of-distribution cube. Multi-seed analysis on the development cubes shows low sensitivity to random initialisation, indicating that variation across these runs is driven more by acquisition conditions than by initialisation. Overall, the proposed framework provides a compact and transferable approach for UAV-based hyperspectral plastic litter monitoring under realistic cross-flight conditions.
重点关注团队Focused team
2. Geomechanical Coupling of In-Situ Stress, Structural Curvature, and Permeability in Deep Coal Seams
作者Authors: Xiaozhuo Wu; Hao Xu; Shida Chen; Daojun Huang; Xiang Li; Jianling Hu; Heng Wu; Pengfei Jiang; et al.
发表月份Publication month: 2026-07 2026-07
SPE Journal · DOI: 10.2118/234704-pa
关键词Tags: vertical structure vertical structure
摘要：地应力对深部煤层渗透率和水力压裂行为具有控制作用。利用鄂尔多斯盆地的现场应力测量、结构曲率数据、测井和岩石力学数据，开发了线性多孔弹性水平应变模型来预测深煤层中的当前应力。结果揭示了应力状态的明显依赖于深度的转变：走滑（2,250 m）。水平应力差（0. 01~23 MPa）在约1 500 m处达到峰值，然后下降，而侧压力系数稳定在0. 63附近，表明深煤层中垂直应力占主导地位。随着深度的增加，储层压力梯度（0. 31–1. 42 MPa/100 m）向静水压值收敛。 渗透率演变遵循相应的三阶段模式：快速下降区、准稳定区和应力不敏感区，约77%的测量值低于0. 1 md，反映了深煤层典型的超低渗透率。应力-结构耦合控制渗透率非均质性：具有低水平应力差的构造高点表现出高渗透率并促进裂缝连通性，作为有效的注气区，最大化二氧化碳（CO2）-基质接触面积。在深层正断层区域 (>2,250 m)，最小水平应力 (Sh) 梯度稳定在约 1. 65 MPa/100 m，定义了维持盖层完整性的注入压力的物理上限。为了确保地质力学稳定性，CO2封存作业必须通过分级增压策略严格将井底压力维持在Sh以下。 该综合分析阐明了应力、结构和渗透率之间的耦合机制，为安全高效的深层碳储存提供了定量框架。
Abstract: In-situ stress exerts a controlling effect on permeability and hydraulic fracturing behaviors in deep coal seams. Using in-situ stress measurements, structural curvature data, well logging, and rock mechanical data from the Ordos Basin, a linear poroelastic horizontal strain model was developed to predict the present-day stress in deep seams. The results reveal a clear depth-dependent transition of stress regimes: strike-slip ( 2,250 m). The horizontal stress difference (0.01–23 MPa) peaks at approximately 1,500 m before declining, while the lateral pressure coefficient stabilizes near 0.63, indicating vertical stress dominance in deep seams. Reservoir pressure gradients (0.31–1.42 MPa/100 m) converge toward the hydrostatic value as depth increases. Permeability evolution follows a correspondingly three-stage pattern: rapid decline, quasistable, and stress-insensitive zones with approximately 77% of measured values below 0.1 md, reflecting the typical ultralow permeability of deep seams. Stress-structure coupling governs permeability heterogeneity: Structural highs with low horizontal stress difference exhibit high permeability and promote fracture connectivity, serving as efficient gas injection zones that maximize the carbon dioxide (CO2)-matrix contact area. In the deep normal-faulting regime (>2,250 m), the minimum horizontal stress (Sh) gradient stabilizes at approximately 1.65 MPa/100 m, defining the physical upper limit for injection pressure to maintain caprock integrity. To ensure geomechanical stability, CO2 storage operations must strictly maintain bottomhole pressures below Sh through staged pressurization strategies. This integrated analysis elucidates the coupling mechanism among stress, structure, and permeability, providing a quantitative framework for safe and efficient deep carbon storage.
其他相关期刊：按主题相关性补充Other relevant journals: topical supplements
3. Seasonal chlorophyll-a prediction in the tropical Pacific with a global climate model incorporating marine biogeochemistry (SINTEX-F2bio)
作者Authors: Takeshi Doi; Yuya Baba; Hakase Hayashida
发表月份Publication month: 2026-07 2026-07
Journal of Climate · DOI: 10.1175/jcli-d-25-0533.1
关键词Tags: phytoplankton phytoplankton
摘要：由于全球变暖和厄尔尼诺-南方涛动（ENSO）的综合影响，太平洋岛国正面临极端天气和气候事件的严重威胁。他们还依赖渔业。低营养级海洋生态系统的季节性预测有可能支持海洋政策制定者和利益相关者的决策，但这种预测服务尚不具备可操作性。为了解决这个问题，我们通过将海洋生物地球化学模型纳入可操作的季节性气候预测系统中，开发了一种新的季节性海洋生物地球化学预测系统（SINTEX-F2bio）。 SINTEX-F2bio 巧妙地预测了未来 5-7 个月由卫星产生的季节性叶绿素 a 围绕太平洋暖池东缘的波动。 该地区表面硝酸盐和叶绿素a的减少/增加与厄尔尼诺/拉尼娜期间暖池向东/向西的运动有关。这些发现为季节性尺度的全球海洋生物地球化学的业务预测提供了前景。
Abstract: Pacific island countries are facing serious threats from extreme weather and climate events due to the combined effects of global warming and El Niño–Southern Oscillation (ENSO). They also depend on fisheries. Seasonal prediction of lower-trophic marine ecosystems has the potential to support the decision-making of marine policymakers and stakeholders, but such a forecasting service does not yet exist on an operational basis. To address this issue, we developed a new seasonal marine biogeochemical prediction system (SINTEX-F2bio) by incorporating a marine biogeochemical model into an operational seasonal climate prediction system. SINTEX-F2bio skillfully predicts satellite-derived seasonal chlorophyll-a fluctuations around the eastern edge of the Pacific warm pool 5-7 months ahead. Reduction/increase of surface nitrate and thus chlorophyll-a in the region is associated with eastward/westward movement of the warm pool during El Niño/La Niña. These findings provide prospects for operational forecasting of global ocean biogeochemistry at seasonal scale.
