1. Aerosol and ocean color retrievals from FastMAPOL/component algorithm using NASA’s PACE polarimetric measurements
关键词Tags: ocean colour ocean colour
摘要：Crossref 未提供该 DOI 的摘要。
Abstract: Crossref did not provide an abstract for this DOI.
Paper Push: 2026-07-22
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
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
In la sua volontade è nostra pace.
这句说“在他的意志中有我们的平安”。它是《天堂篇》中最常被引用的神学句之一。
In his will is our peace. This is one of Paradiso's most quoted theological lines.
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
重点关注团队Focused team
2. Interpretable Machine Learning for Predicting Blast-Induced Particulate Matter Emissions in Surface Mines
关键词Tags: ocean biogeochemistry ocean biogeochemistry
摘要：爆破是露天采矿的一项重要作业，但它会在短时间内产生高浓度的颗粒物。准确预测爆炸引起的粉尘浓度对于空气质量管理、工人保护和粉尘控制规划很有用。在这项研究中，开发了一个可解释的机器学习框架来预测大型露天煤矿爆破引起的空气动力学直径小于 10 μm 的颗粒物 (PM10) 和总悬浮颗粒物 (TSP) 浓度。该数据集来自已发布的现场监测记录，包括爆破设计参数、监测距离、材料相关变量和测量的粉尘浓度。共有148个有效样本用于模型开发和评估。 建立并比较了六种基于树的集成模型，包括额外树回归、随机森林、梯度提升回归树、极限梯度提升（XGBoost）、轻梯度提升机（LightGBM）和分类提升（CatBoost）。 使用贝叶斯优化进行超参数调整，并使用确定系数 (R²)、均方根误差 (RMSE)、平均绝对误差 (MAE) 和平均绝对百分比误差 (MAPE) 评估模型性能。 在采用的 80:20 保留验证方案下，CatBoost 在 PM10 预测方面取得了最佳测试性能，测试 R2 为 0。 9555，RMSE 为 444。 87 μg/m3，MAE 为 343。 48微克/立方米。 额外树回归对于 TSP 预测表现最好，测试 R2 为 0。 8803，均方根误差为 3899。 04 μg/m3，MAE 为 3036。 27微克/立方米。 残差分析进一步表明最优模型不存在明显的系统偏差。由模型内特征重要性排名支持的 Shapley 附加解释 (SHAP) 分析表明，爆炸物数量、爆破孔数量和监测距离是影响 PM10 和 TSP 预测的主要变量。爆炸量和炮孔数量主要增加了预测粉尘浓度，而监测距离则普遍降低了预测粉尘浓度。所提出的框架可以为露天矿的爆破参数优化、监测点布置和粉尘控制决策提供有用的支持，但仍需要使用更大的多站点数据集进行进一步验证。
Abstract: Blasting is an essential operation in surface mining, but it can generate high particulate matter concentrations within a short period. Accurate prediction of blast-induced dust concentration is useful for air quality management, worker protection, and dust control planning. In this study, an interpretable machine learning framework was developed to predict particulate matter with an aerodynamic diameter less than 10 μm (PM10) and total suspended particulate matter (TSP) concentrations induced by blasting in a large surface coal mine. The dataset was derived from published field monitoring records and included blasting design parameters, monitoring distance, material-related variables, and measured dust concentrations. A total of 148 valid samples were used for model development and evaluation. Six tree-based ensemble models, including Extra Trees Regression, Random Forest, Gradient Boosting Regression Trees, extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost), were established and compared. Bayesian optimization was used for hyperparameter tuning, and model performance was evaluated using the coefficient of determination (R²), root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). Under the adopted 80:20 hold-out validation scheme, CatBoost achieved the best test performance for PM10 prediction, with a test R2 of 0.9555, RMSE of 444.87 μg/m3, and MAE of 343.48 μg/m3. Extra Trees Regression performed best for TSP prediction, with a test R2 of 0.8803, RMSE of 3899.04 μg/m3, and MAE of 3036.27 μg/m3. Residual analysis further indicated that the optimal models had no obvious systematic bias. Shapley additive explanations (SHAP) analysis, supported by within-model feature importance rankings, showed that explosive quantity, number of blastholes, and monitoring distance were the dominant variables affecting PM10 and TSP predictions. The explosive quantity and number of blastholes mainly increased the predicted dust concentration, whereas the monitoring distance generally reduced it. The proposed framework may provide useful support for blast parameter optimization, monitoring point arrangement, and dust control decision-making in surface mines, but further validation using larger multi-site datasets is still needed.
其他相关期刊：按主题相关性补充Other relevant journals: topical supplements
3. Diurnal sea surface temperature amplification enhances surface extremes during marine heatwaves
关键词Tags: marine heatwaves marine heatwaves
摘要：海洋热浪通常使用“基础”海面温度进行监测，以确保稳定的气候参考，但这种做法忽略了海洋上层约 10 m 内发生的短暂昼夜极端事件，那里近地表生物暴露，海气交换过程受到调节。利用卫星观测、大气再分析和海洋模型模拟，我们表明，在海洋热浪条件下，昼夜周期被系统地放大，白天变暖增强，昼夜周期延迟。在全球范围内，白天皮下峰值增强了 20-30%，使事件平均最大强度提高了 0. 5-1。 0 °C 并在累积暴露量上增加数十度日。 这种广泛的放大源于热力学反馈，其中海洋热浪期间的微弱风和增强的短波辐射使混合层浅滩并稳定，从而使近地表热量积聚并持续到晚上。在低纬度海洋中，高达 40% 的海洋热浪日数超过了基金会定义的皮下层事件峰值，使生态系统面临无法解释的热应力。我们建议，需要将解析良好的昼夜基于皮下的诊断集成到未来的海洋热浪监测和预测框架中。
Abstract: Marine heatwaves are typically monitored using “foundation” sea surface temperature to ensure a stable climate reference, but this practice neglects the short-lived diurnal extremes occurring within the upper ~10 m of the ocean, where near-surface organisms are exposed and air-sea exchange processes are modulated. Using satellite observations, atmospheric reanalysis, and ocean model simulations, we show that during marine heatwave conditions, the diurnal cycle is systematically amplified, with enhanced daytime warming and delayed diurnal phases. Globally, daytime subskin peak is intensified by 20–30%, elevating event-mean maximum intensity by 0.5–1.0 °C and adding tens of degree-days to cumulative exposure. This widespread amplification arises from a thermodynamic feedback in which weak winds and enhanced shortwave radiation during marine heatwaves shoal and stabilize the mixed layer, allowing near-surface heat to accumulate and persist into evening hours. In low-latitude oceans, up to 40% of marine heatwave days exceed the foundation-defined event peak at the subskin level, exposing ecosystems to unaccounted thermal stress. We suggest well-resolved diurnal, subskin-based diagnostics need to be integrated into future marine heatwave monitoring and prediction frameworks.
4. Vertical profiles of aerosol chemical species concentration retrieved through synergy of spaceborne lidar and polarimeter observations
关键词Tags: ocean colour; backscattering; bio-optics; vertical structure ocean colour; backscattering; bio-optics; vertical structure
摘要：我们提出了一种新颖的方法，AEROCHEMPro/GRASP（AEROsol CHEMical PROfiling），通过协同组合多波长激光雷达和多角度旋光计的同位测量来检索气溶胶化学物质的垂直浓度剖面。 AEROCHEMPro 代表了第一个垂直分辨气溶胶化学成分遥感检索框架。它基于 GRASP（气溶胶和表面特性的通用检索）化学成分框架的改进版本。该方法是在大气观测系统（AOS）国际倡议的背景下开发的，该倡议提出了一个星载观测系统，以增进我们对气溶胶、云、对流和降水的了解。 此外，检索策略仍然广泛适用于未来的卫星任务和涉及激光雷达和多角度旋光计气溶胶遥感组合的观测系统。基于统计优化和物理约束反演，AEROCHEMPro/GRASP 提供三种不同的气溶胶垂直剖面：(i) 由黑碳、棕碳、无机盐和相关水吸收组成的精细模式； (ii) 代表矿物粉尘的疏水性粗模式，分解为氧化铁和石英物质； (iii)由海盐颗粒及其相关水含量组成的亲水粗模式。该方法明确检索这三种气溶胶剖面，以及提到的六种气溶胶化学物质的丰度及其含水量。 这种气溶胶化学成分垂直剖面的检索首次提供了气溶胶光学测量与其在高度上解析的形态形成之间的直接观测联系。我们通过创新的检索实验证明了该技术的可行性和性能，其中使用 MOCAGE 化学传输模型和综合辐射传输模拟器（GRASP 正向模型）生成合成激光雷达和旋光计观测结果。这些伪观测包括多波长衰减后向散射和去偏振比，以及跨多个视角和光谱带的偏振辐射率。 跨越海洋、城市、灰尘和复杂混合物主导区域的全球尺度横断面的结果表明，该反演以良好的保真度捕获了垂直气溶胶成分的主要特征及其整体光学特性。精细模式物质可以很好地恢复，特别是在边界层中，甚至对于具有多种气溶胶物质的复杂混合物的大气也是如此。灰尘被准确地回收，并在垂直范围、负载和成分方面得到了很好的定义，并且海盐浓度也得到了很好的再现。指出了导出高层大气层中水含量的一些限制。 AEROCHEMPro/GRASP 可靠地导出气溶胶光学深度 (AOD)、单散射反照率 (SSA) 和激光雷达比 (LR) 等光学特性，证明了该方法的稳健性。 由于此初始验证依赖于空间均匀表面基线上的自洽模拟框架，因此这些结果代表了乐观、理想化的理论性能极限。这种新的检索方法代表了星载气溶胶遥感的重大进步，因为它提供了与化学传输模型结果直接相关的垂直解析化学形态。它为提高我们对气溶胶过程及其对气候和空气质量影响的理解提供了新的机会。
Abstract: We present a novel methodology, AEROCHEMPro/GRASP (AEROsol CHEMical PROfiling), for retrieving vertical concentration profiles of aerosol chemical species by synergistically combining co-located measurements from a multiwavelength lidar and a multi-angular polarimeter. AEROCHEMPro represents the first retrieval framework for remote sensing of the vertically-resolved aerosol chemical composition. It is based on an improved version of the GRASP (Generalized Retrieval of Aerosol and Surface Properties) chemical component framework. The methodology is developed within the context of the Atmosphere Observing System (AOS) international initiative, which proposes a spaceborne observing system to advance our knowledge of aerosols, clouds, convection, and precipitation. Moreover, the retrieval strategy remains broadly applicable to future satellite missions and observing systems involving combined lidar and multi-angular polarimeter aerosol remote sensing. Based on a statistically optimized and physically-constrained inversion, AEROCHEMPro/GRASP delivers three distinct aerosol vertical profiles: (i) a fine mode composed of black carbon, brown carbon, inorganic salts, and associated water uptake; (ii) a hydrophobic coarse mode representing mineral dust, decomposed into iron oxide and quartz species; and (iii) a hydrophilic coarse mode consisting of sea salt particles and their associated water content. The approach explicitly retrieves these three aerosol profiles, along with the fractional abundance of each of the six mentioned aerosol chemical species and their water content. This retrieval of aerosol chemical composition vertical profiles offers, for the first time, a direct observational link between aerosol optical measurements and their speciation resolved in altitude. We demonstrate the feasibility and performance of this technique through an innovative retrieval experiment, where synthetic lidar and polarimeter observations are generated using the MOCAGE chemical transport model and a comprehensive radiative transfer simulator (GRASP forward model). These pseudo-observations include multi-wavelength attenuated backscatter and depolarization ratios, along with polarized radiances across multiple viewing angles and spectral bands. Results from global-scale transects spanning marine, urban, dust, and complex mixture-dominated regions show that the retrieval captures with good fidelity the main features of vertical aerosol composition and their bulk optical properties. Fine-mode species are well retrieved, particularly in the boundary layer, even for atmospheres with complex mixtures of multiple aerosol species. Dust is accurately retrieved and well defined in terms of vertical extent, load, and composition, and sea salt concentrations are very well reproduced. Some limitations for deriving water content in the upper atmospheric layers are remarked. AEROCHEMPro/GRASP reliably derives optical properties such as aerosol optical depth (AOD), single scattering albedo (SSA), and lidar ratio (LR), demonstrating the approach's robustness. As this initial validation relies on a self-consistent simulation framework over a spatially uniform surface baseline, these results represent an optimistic, idealized theoretical performance limit. This new retrieval approach represents a significant advancement in spaceborne aerosol remote sensing, as it provides vertically resolved chemical speciation that is directly related to chemical transport model results. It offers new opportunities to improve our understanding of aerosol processes and their effects on climate and air quality.
5. Pleistocene benthic foraminifera bioevents in the Central Arctic Ocean: stratigraphic and paleoceanographic implications
关键词Tags: carbon pump carbon pump
摘要：底栖有孔虫在北冰洋中部 (CAO) 显示出独特的时空分布模式，表明它们具有提供强大的年龄限制和解决更新世古海洋变化的潜力。之前已经报道过更新世的几个底栖有孔虫生物事件，这里通过研究门捷列夫和罗蒙诺索夫海脊的三个沉积岩芯并分析已发表的数据集对这些生物事件进行了严格的评估。基于这些数据，生物事件是通过使用>63微米粒度部分中物种的绝对丰度来定义的，而相对丰度被认为是不可靠的，因为诸如分解和/或溶解之类的埋藏过程会覆盖原始的组合成分。 生物事件与岩性层位相关，并与基于可用独立地层数据的海洋同位素阶段（MIS）相关。中更新世可以定义三个钙质生物事件：（1）最常见的Bolivina arctica（~MIS 9），（2）最低常见的Oridorsalis umbonatus（~MIS 7），（3）在水深小于~2000米的Bulimina aculeata（~MIS 7）达到顶峰。 Oridorsalis umbonatus 最低常见程度与浅处 Bulimina aculeata 的顶端基部同时出现。由于辐射年龄和生物地层年龄的数量有限，因此所提出的生物事件与海洋同位素阶段的相关性应被视为临时的。 进一步的底栖有孔虫生物事件可能有助于区域到超区域尺度的地层相关性，但需要评估先前的分类学鉴定和额外的沉积岩芯研究。已灭绝的凝集物种 Haplophragmoides obscurus 在中更新世的罗蒙诺索夫海岭消失，但复杂的分类学和北极沉积物岩心中出现的数据很少，目前禁止将其用作生物地层标记。从凝集的底栖有孔虫到钙质底栖有孔虫的组合转换发生在罗蒙诺索夫海岭和莫里斯杰苏普隆起的第一次正常到反向磁极性的下岩心变化附近，并且可能是北冰洋东部早于 MIS 11 的中更新世沉积物中的同步事件。 然而，这种组合组成的根本性变化在整个北冰洋上具有时间海侵性，因为它发生在早更新世的美亚盆地。 CAO 中的生物事件是由各种生物过程的复杂相互作用引起的。除了可能在北冰洋进化的 B. arctica 和 H. obscurus 之外，B. aculeata 和 O. umbonatus 一定是从副极地纬度入侵北冰洋的。由于只有通过弗拉姆海峡和巴伦支海才能促进中层到深水物质与亚极地纬度的无限制交换，因此这些中层到深水物种必须在有利于其繁殖的时间段内作为幼体标本（繁殖体）由大西洋水运到CAO地点。 北冰洋内定居的最大可到达位置取决于物种、当地环境条件和大西洋水平流的强度。环境条件，特别是食物的可获得性，对于在特定地点成功定殖发挥了重要作用，不仅对于入侵物种，而且对于 CAO 特有物种（H. obscurus、B. arctica）也是如此。这些地点一定面临着比今天（B. aculeata）高（H. obscurus、B. arctica、O. umbonatus）或明显更高的向海底输出的颗粒有机碳。这种环境条件必须在整个盆地范围内发生，才能引发组合成分的同步和同时变化。此外，外部强迫可能引发了环境变化。 富含碎屑白云石的冰筏碎片的大量排放可能导致了以玻利维纳北极为主的动物群的突然崩溃以及Haplophragmoides obscurus的即将消失。环境中最显着的变化表现为从以粘着性为主的底栖有孔虫到以钙质底栖有孔虫为主的转变，这要么是由于食物供应及其质量的根本变化，要么是由于底层水的腐蚀性。一般来说，由于薄壳表层动物类群的选择性溶解，组合中富含健壮的表层动物和/或动物内钙质物种，或者可能仅由凝集的地表动物群组成。
Abstract: Benthic foraminifera show distinct temporal and spatial distribution patterns in the Central Arctic Ocean (CAO) demonstrating their potential to provide robust age constraints and to address paleoceanographic change in the Pleistocene. Several benthic foraminifera bioevents have been previously reported from the Pleistocene that are critically evaluated here by studying three sediment cores from the Mendeleev and Lomonosov ridges and analysing published data sets. Based on these data bioevents are defined by using absolute abundances of species in the >63 µm grain-size fraction, whereas relative abundances are considered not reliable because taphonomic processes such as disintegration and/or dissolution overprint the original assemblage composition. Bioevents are correlated to lithological horizons, and linked to marine isotope stages (MIS) based on available independent stratigraphic data. Three calcareous bioevents can be defined in the Middle Pleistocene: (1) the highest common occurrence of Bolivina arctica (∼MIS 9), (2) the lowest common occurrence of Oridorsalis umbonatus (∼MIS 7), and (3) the acme of Bulimina aculeata (∼MIS 7) in water depths of less than ∼2000 m. The lowest common occurrence of Oridorsalis umbonatus is coeval with the base of the acme of Bulimina aculeata at shallow sites. Since the number of radiometric and biostratigraphic ages is limited, the proposed correlation of bioevents to marine isotope stages should be considered provisional. Further benthic foraminifera bioevents may be useful for stratigraphic correlation on a regional to supra-regional scale but require evaluation of previous taxonomic identifications and additional sediment core studies. The extinct agglutinated species Haplophragmoides obscurus disappeared at the Lomonosov Ridge in the Middle Pleistocene but the complex taxonomy and the few data on the occurrence in Arctic sediment cores currently prohibits the application as biostratigraphic marker. The assemblage turnover from agglutinated to calcareous benthic foraminifera occurred close to the first downcore change of normal to reverse magnetic polarity at the Lomonosov Ridge and Morris Jesup Rise and might be a synchronous event in the eastern Arctic Ocean in middle Pleistocene sediments older than MIS 11. However, this fundamental change in assemblage composition is time-transgressive across the Arctic Ocean because it occurred in the Amerasian Basin in the Early Pleistocene. The bioevents in the CAO are caused by a complex interplay of various biological processes. Apart from B. arctica and H. obscurus that likely evolved in the Arctic Ocean, B. aculeata and O. umbonatus must have invaded the Arctic Ocean from subpolar latitudes. Since an unrestricted exchange of intermediate to deep-water masses with subpolar latitudes is only facilitated through the Fram Strait and the Barents Sea, these intermediate to deep-water species had to be transported as juvenile specimens (propagules) by Atlantic Water to CAO sites during time periods favourable for their propagation. The maximum reachable location for settlement within the Arctic Ocean depended on the species, the local environmental conditions, and the strength of Atlantic Water advection. Environmental conditions, in particular the availability of food, played then a major role for the successful colonization at a particular site, not only for the invading species but also for the species endemic to the CAO (H. obscurus, B. arctica). These sites must have faced a high (H. obscurus, B. arctica, O. umbonatus) or significantly higher particulate organic carbon export to the sea floor than today (B. aculeata). Such environmental conditions must have occurred basin-wide to trigger the synchronous and coincident changes in assemblage compositions. Moreover, external forcing may have triggered environmental change. A massive discharge of detrital dolomite-rich ice-rafted debris might have induced the abrupt collapse of a Bolivina arctica dominated fauna and imminent disappearance of Haplophragmoides obscurus. The most conspicuous change in the environment is expressed in the turnover from a predominance of agglutinated to calcareous benthic foraminifera which was either caused by a fundamental change in food supply and its quality or by corrosive bottom waters. In general, due to selective dissolution of thin-shelled epifaunal taxa, assemblages are enriched in robust epifaunal and/or infaunal calcareous species, or may consist only of an agglutinated taphocoenosis.
6. Toward Sustainable Marine Governance in the Egyptian Red Sea: A Multi-Decadal Integration of Climate-Driven and Anthropogenic Pressures on Coral Reef Resilience
关键词Tags: phytoplankton phytoplankton
摘要：埃及红海的珊瑚礁生态系统面临着气候变暖和沿海开发加剧带来的不断升级的压力，在根据第 4419/2025 号总理令指定大边缘礁海洋保护区之前，这些双重压力因素的空间表达和治理影响仍未得到充分表征。这项研究开发了一个综合的多十年框架，结合了 32 年的海面温度和表面海流速度记录、二十年的卫星叶绿素 a 观测，以及四十年的卫星衍生的五个监测沿海地点的珊瑚礁范围测绘。全流域海温变暖约 1。 15°C 存在空间异质性，北部沿海地区升温幅度比南部地区高出近 25%。 拉斯巴纳斯的地表环流在结构上出现分歧：北部的加速与人为输入的平流稀释增强一致，而南部的一致减速（平均-0. 03 m/s）与干河沉积物和养分排放的生态足迹增加有关。明显局部的叶绿素-a 异常，在赫尔格达达到 0. 97 mg/m3，并且在空间上与记录的发育年代一致，这与人为破坏寡营养水质是导致北部地区珊瑚礁退化的重要因素是一致的。 来自 2026 年重复调查的现场证据，与记录在案的 2024 年影响埃及南部珊瑚礁的严重区域白化现象一起解释，表明气候驱动的热应力构成了平行的主动压力，特别是在南部地区，在珊瑚礁收缩最严重的时期，热应力和人为应力源似乎已经聚合。陆地卫星造成的损失为 24. 1–50.自 1985 年以来，所有监测地点的收缩率均下降了 2%，其中仅 2022 年至 2025 年间，Halaib-Shalateen 就收缩了 33. 3%，反映了这些相互作用压力的累积结果。这些发现支持 2025 年海洋保护区的生态必要性，同时确定空间差异性、双重压力源治理优先事项，这对于将其政策承诺转化为可衡量的保护成果至关重要。
Abstract: Coral reef ecosystems of the Egyptian Red Sea face escalating pressure from compounding climate-driven warming and intensifying coastal development, dual stressors whose spatial expression and governance implications remained insufficiently characterized prior to the designation of the Great Fringing Reef Marine Protected Area under Prime Ministerial Decree No. 4419/2025. This study develops an integrated multi-decadal framework combining 32 years of sea surface temperature and surface current velocity records, two decades of satellite-derived chlorophyll-a observations, and four decades of satellite-derived coral reef extent mapping across five monitored coastal sites. Basin-wide SST warming of ≈1.15 °C is spatially heterogeneous, with the northern coastal sector warming nearly 25% above the southern sector rate. Surface circulation diverges structurally at Ras Banas: acceleration in the north is consistent with enhanced advective dilution of anthropogenic inputs, whereas consistent deceleration in the south (mean −0.03 m/s) is associated with an increased ecological footprint of wadi-derived sediment and nutrient discharge. Sharply localized chlorophyll-a anomalies, reaching 0.97 mg/m3 at Hurghada and spatially congruent with documented development chronologies, are consistent with anthropogenic disruption of oligotrophic water quality as an important contributor to reef degradation at northern sites. Field evidence from 2026 repeat surveys, interpreted alongside documented severe regional bleaching affecting southern Egyptian reefs during 2024, suggests that climate-driven thermal stress constitutes a parallel and active pressure, particularly in the southern sector where thermal and anthropogenic stressors appear to have converged during the period of most acute reef contraction. Landsat-derived losses of 24.1–50.2% across all monitored sites since 1985, including a 33.3% contraction at Halaib–Shalateen between 2022 and 2025 alone, reflect the cumulative outcome of these interacting pressures. These findings support the ecological necessity of the 2025 MPA while identifying spatially differentiated, dual-stressor governance priorities essential to translating its policy commitment into measurable conservation outcomes.
7. Embodied Encounters with Plankton: Ground Truth Methods as Sites for Human-Ocean Entanglement
关键词Tags: ocean colour ocean colour
摘要：本文通过 2025 年 2 月在圣卢西亚悬崖附近进行的为期十天的研究巡航的直接参与，探讨了当代海洋学中浮游生物研究方法的转变。作为一名女权主义视觉科学研究研究员，我反思了晕船、感官沉浸和具体参与如何塑造了我的浮游生物采样体验，并揭示了海洋学知识与其背景的纠缠。虽然卫星、数字孪生和远程操作车辆等遥感技术日益塑造海洋学并产生强化的海洋量化模式，但我认为触觉物理采样方法在浮游生物研究中仍然不可或缺。 这些方法并没有被取代，而是转变为地面实况方法：观察和定位浮游生物和科学家之间的接触实例，用于校准、验证和丰富遥感数据。通过将物理收集方法与遥感抽象并置，我展示了不同的技术如何产生不同的浮游生物观察方式。借鉴女权主义、视觉和残疾科学理论，我强调海洋学研究，即使使用遥感技术，本质上也是通过与海洋本身的不断谈判行为来体现、定位和共同产生的。我认为，地面实况方法是将遥感海洋数据与直接、具体的海洋接触联系起来的重要枢纽。
Abstract: This article examines shifting methods of plankton research within contemporary oceanography through direct participatory engagement on a ten-day research cruise off the Santa Lucia Escarpment in February 2025. As a feminist visual science studies researcher, I reflect on how seasickness, sensory immersion, and embodied participation shaped my experience of plankton sampling and reveal the entanglement of oceanographic knowledge with its contexts. While remote sensing technologies – such as satellites, digital twins, and remotely operated vehicles – are increasingly shaping oceanography and producing intensified modes of ocean quantification, I argue that tactile, physical sampling methods remain indispensable in plankton research. Rather than being supplanted, these transform into ground truth methods: observed and situated instances of contact between plankton and scientist used to calibrate, validate, and enrich remote sensed data. By juxtaposing physical collection methods with abstractions of remote sensing, I demonstrate how different technologies generate distinct ways of seeing plankton. Drawing on feminist, visual, and disability theories of science, I emphasize that oceanographic research, even when using remote sensing, is inherently embodied, situated, and coproduced through constant acts of negotiation with the ocean itself. Ground truth methods, I contend, are essential hinges that tether remotely sensed ocean data to direct, embodied ocean encounters.
8. AI in Ocean Monitoring: Deep Learning for Marine Pollution, Fisheries, and Underwater Robotics
关键词Tags: ocean colour ocean colour
摘要：海洋覆盖了地球表面百分之七十以上，但由于人类直接大规模观测（尤其是深度观测）的成本、难度和危险，相对于陆地环境而言，海洋采样仍然相对不足。人工智能已成为缩小这一观察差距的日益重要的工具，支持水下图像和声学记录的自动分析、自主水下航行器导航、基于卫星的海洋遥感以及根据船只跟踪数据检测非法捕鱼活动。本文调查了深度学习在海洋监测中的应用，将方法组织成涵盖水下视觉和声学分析、自主水下机器人、卫星海洋遥感、海洋污染检测和渔业资源评估的分类法。 我们回顾了海洋塑料碎片监测、非法、未报告和无管制捕捞检测、珊瑚礁健康评估、海洋哺乳动物保护和海洋气候监测等应用，总结了该领域使用的基准和数据集，并提出了一个案例研究，说明用于非法捕捞检测的综合卫星和船只跟踪管道。我们讨论了挑战，包括相对于陆地数据集标记的水下图像的稀缺性、在功率和带宽受限的水下平台上运行机器学习系统的难度以及相对于可用监测基础设施的海洋的广阔性，并概述了将海洋人工智能与自主海洋观测网络和国际渔业治理连接起来的未来方向。
Abstract: The ocean covers more than seventy percent of the Earth's surface, yet remains comparatively undersampled relative to terrestrial environments due to the cost, difficulty, and hazard of direct human observation at scale, particularly at depth. Artificial intelligence has become an increasingly important tool for closing this observation gap, supporting automated analysis of underwater imagery and acoustic recordings, autonomous underwater vehicle navigation, satellite-based ocean remote sensing, and detection of illegal fishing activity from vessel tracking data. This paper surveys the application of deep learning to ocean monitoring, organizing methods into a taxonomy spanning underwater visual and acoustic analysis, autonomous underwater robotics, satellite ocean remote sensing, marine pollution detection, and fisheries stock assessment. We review applications including marine plastic debris monitoring, illegal, unreported, and unregulated fishing detection, coral reef health assessment, marine mammal conservation, and ocean climate monitoring, summarize benchmarks and datasets used in this domain, and present a case study illustrating an integrated satellite and vessel-tracking pipeline for illegal fishing detection. We discuss challenges including the scarcity of labeled underwater imagery relative to terrestrial datasets, the difficulty of operating machine learning systems on power- and bandwidth-constrained underwater platforms, and the vastness of the ocean relative to available monitoring infrastructure, and outline future directions connecting ocean AI with autonomous ocean observation networks and international fisheries governance.