Atmospheric and Oceanic Sciences

Pakistan has experienced several meteorological drought disasters in recent decades; their frequency and duration were more pronounced in the southern region. In the modulation of drought events, the significant contribution of regional and global climatic factors and large-scale ocean-atmospheric circulations was evident in this study.

Deep neural networks have substantially improved the performance of hyperspectral image classification, yet they remain vulnerable to adversarial attacks. Existing attack methods usually manipulate pixel spectra directly, ignoring the physical mixing mechanism of remote sensing imaging and potentially generating adversarial samples with limited physical consistency and interpretability. Moreover, balancing attack effectiveness and perturbation imperceptibility remains a challenging multi-objective optimization problem. To address these issues, this paper proposes an evolutionary multi-task multi-objective adversarial attack framework based on inter-task knowledge transfer. Instead of perturbing raw pixel spectra, the proposed method introduces perturbations into abundance maps obtained through spectral unmixing, thereby improving the physical plausibility of the generated adversarial samples. The generation of class-specific universal perturbations is formulated as a collaborative multi-task optimization problem. To solve this problem, we develop a Self-Adaptive Multi-Objective Multi-Factorial Evolutionary Algorithm for Adversarial Attacks (SAMO-MFEA-AA). By modeling the attack generation processes for different land-cover classes as distinct yet correlated optimization tasks, SAMO-MFEA-AA dynamically captures synergistic relationships among tasks. An asymmetric adaptive cooperation matrix is employed to regulate the intensity of knowledge transfer, allowing beneficial perturbation patterns to be shared across related classes while reducing the risk of negative transfer. Extensive experiments on the Indian Pines and Salinas datasets demonstrate that the proposed framework achieves competitive hypervolume performance and favorable solution diversity compared with existing multi-objective optimization algorithms. In adversarial attack scenarios, the proposed method achieves effective attack success rates against representative classification networks while maintaining the physical plausibility of abundance-space perturbations.

Abstract The Hangai Dome in central Mongolia represents one of the most prominent intracontinental uplifts, yet the relative contributions of crustal and mantle dynamics in sustaining its high topography remain unresolved. Here we jointly invert Rayleigh wave dispersion and receiver function data to constrain crustal and uppermost mantle structures, including Moho and lithosphere‐asthenosphere boundary (LAB) architecture, crustal Vp/Vs ratios and Vs. Our results reveal the significantly thickened crust (50–53 km) and thinned lithosphere (65–70 km) beneath the Hangai Dome, with pronounced lateral variations indicating presence of fluids and volatiles in the lower crust beneath the eastern Hangai Dome that generates low‐velocity zone and promotes rapid ascent of mantle‐derived melts. Furthermore, surface elevation exhibits strong linear correlations with crustal thickness and Bouguer gravity anomalies, consistent with near‐isostatic equilibrium. Our results indicate a leading role of inherited thickened crust in controlling topographic uplift of the Hangai Dome, whereas asthenospheric upwelling likely exerts a secondary influence.

Abstract Forested watersheds with limited human intervention are a vulnerable and diminishing feature in modern temperate latitudes landscapes. They are essential in terms of conserving biodiversity, as reference systems for forest management, restoration, stream ecosystem function, and associated ecosystem services. In Patagonia, some of the largest remaining intact watersheds are juxtaposed among areas heavily impacted from human‐caused fires and clearing, a pattern extending across several forested biomes. We measured in‐stream large wood storage and riparian forest structure across paired reference and impacted headwater watersheds (1–3 km 2 ) representing deciduous/dry and evergreen/humid biomes to evaluate baseline conditions and stream geomorphic response to wood storage. Deciduous forested watersheds had significantly higher LW volumes at impacted sites, relative to reference sites, together with higher variability across transects. Meanwhile, evergreen forested streams had significantly higher volumes of LW compared to deciduous sites, which were also consistently higher for impacted streams, although the relative divergence of stream LW volume between reference and impacted decreased in these evergreen sites. Compared to other regions, overall high LW volume may be a function of recent inputs, large amounts of legacy LW following an intense history of recent (50–70 years) wildfire, limited LW removal, low decay rates, and larger diameter species. The riparian forest structural variables that best explained the volumes of wood were live tree density and basal area; however, there was high uncertainty of our GLMM model given the limited number of replicates and high variability within streams. There was also a significant relationship between LW density and stream geomorphic features such as frequency of debris jams, step‐pools and bankfull width for reference condition sites, yet this relationship was not evident in the land‐use impacted sites. Despite clear evidence of land‐use impacts across stream pairs, our observations suggest a less intuitive relation between riparian forest structure, LW and stream geomorphology.

Fractional vegetation cover (FVC) as a key indicator for assessing ecological degradation and recovery in arid desert regions. However, the selection of suitable vegetation indices for desert areas remains controversial because of limitations imposed by soil background interference and regional heterogeneity. This study employs a pixel-level binary classification model, integrating three commonly used vegetation indices—EVI, NDVI, and MSAVI—to systematically evaluate the spatiotemporal dynamics of FVC in Xinjiang’s desert regions from 2016 to 2025. A comparative analysis of their inversion accuracies is also conducted. The results are as follows: (1) Leveraging its multiband atmospheric and soil correction mechanisms, the EVI significantly outperformed the NDVI and MSAVI in terms of vegetation–soil discrimination capability (endpoint difference of 0.185) and inversion accuracy (R 2 = 0.79), establishing itself as the optimal index for estimating FVC in Xinjiang’s desert regions. In contrast, the NDVI and MSAVI systematically overestimated FVC in areas with low vegetation density because of their sensitivity to the soil background. (2) The Junggar Basin exhibited high FVC (>50%) with high temporal variability (CV) and a predominant decreasing trend, whereas the Tarim Basin showed low FVC with low variability and a increasing trend. This study not only validates the superiority of the EVI in monitoring vegetation in arid desert regions but also maps the spatiotemporal evolution trends of vegetation coverage in Xinjiang’s desert areas, providing scientific evidence and theoretical support for regional ecological conservation and restoration practices.

Background In the era of the digital economy, the rapid expansion of data centers has precipitated severe environmental challenges. While the operational efficiency of Artificial Intelligence (AI) is well-recognized, its impact on the sustainability of the construction phase remains ambiguous. This study aims to empirically examine the direct and moderating roles of AI technology in the sustainable development of data center construction. Methods Adopting a Resource-Based View (RBV), this research constructed a Structural Equation Model (SEM). Empirical data were collected through a questionnaire survey of 851 professionals involved in data center design and engineering in China. Results The findings demonstrate that the level of AI application has a significant direct positive impact on sustainable development (p < 0.001). Furthermore, AI acts as a critical moderator, significantly amplifying the positive effects of green policy support, computing market demand, technology capital investment, and top management support on sustainability outcomes. Conclusion These results suggest that AI functions not merely as a technical tool but as a catalyst that bridges the gap between strategic drivers and green execution. This study offers theoretical insights and practical recommendations for policymakers and industrial stakeholders to leverage AI for green infrastructure.

Abstract Although magnetic data are widely inverted to estimate subsurface susceptibility or remanent magnetization, most current methods do not explicitly account for the complicating factors of both remanent magnetization and self-demagnetization. This limits their applicability to recover accurate physical property distributions when highly susceptible and remanently magnetized materials are present. Here, we introduce a general finite-volume framework for forward modeling and inversion of magnetic data using the differential form of Maxwell's equations for magnetics. The aims of this framework are twofold. First, the approach models directly in terms of intrinsic physical properties, specifically magnetic susceptibility and remanent magnetization, which is advantageous for incorporating a priori information. Second, the differential approach enables magnetic vector inversion (MVI) to be performed on large scales with lower memory requirements than standard integral approaches. We validate the formulation against analytic and integral solutions and show that it accurately models induced magnetization in highly susceptible materials while accounting for remanent magnetization. Using a synthetic inversion example, we show that jointly inverting for susceptibility and remanent magnetization increases the non-uniqueness of the problem, but that incorporating appropriate a priori information through a parametric formulation leads to improved recovery of both geometry and physical properties, as compared to a parametric MVI inversion. We then apply the differential approach to magnetic vector inversion and show that it scales favorably to very large-scale problems. We invert data from the Mt. Isa Inlier in Australia for a model with 47.3 million model parameters in under two hours.

Abstract Accurate formation density measurement during logging while drilling (LWD) is often compromised by dynamic tool standoff variations, especially in eccentric or enlarged boreholes. Traditional methods struggle with errors caused by asymmetric standoff and rotational tool movement. To overcome these limitations, an optimized azimuthal gamma density method based on the weighted fusion of sixteen-sector measurement data is presented. By leveraging the tool's spatial nuclear response characteristics, a directional quality factor is assigned to each sector. A first-order Fourier expansion is then used to identify the optimal azimuthal direction that best reflects the true formation signal. Within a ±90° window centered on this optimal direction, high-quality sector data are selectively fused to compute a more accurate formation density. This approach effectively suppresses azimuthal noise and enhances measurement reliability under complex wellbore conditions. Validation through numerical simulations and field data demonstrates significant improvements: in a vertical well section (X500-X600 m), the proposed method achieved a root mean square error (RMSE) of 0.038 g/cm3 and a mean absolute error (MAE) of 0.029 g/cm3; in a deviated section (Y650-Y750 m), the RMSE and MAE were further reduced to 0.036 g/cm3 and 0.028 g/cm3, respectively—representing over 30% improvement compared to conventional methods. These results demonstrate the method's effectiveness and adaptability for real-time formation evaluation and geosteering applications.

Abstract The hot springs of Yellowstone National Park (YNP) provide a natural laboratory for investigating the influence of chemical speciation and energy availability on the distribution of microbes. Among 133 inorganic chemotrophic energy supplies, total dissolved ammonia oxidation was consistently one of the highest across 63 YNP hot spring samples. Despite this, the distribution of ammonia‐oxidizing archaea (AOA) does not reflect the energy supply calculated using total chemical abundances, and AOA were detected in only ∼25% of the hot springs sampled and constituted <5% of any microbial community. To investigate the role of substrate speciation preference, total dissolved ammonia (NH 3 (aq) + NH 4 + ) measurements were made in coordination with the collection of biological samples. DNA was extracted from these samples for 16S rRNA gene sequencing. By speciating the total dissolved ammonia concentrations into NH 3 (aq) and NH 4 + , a minimum concentration for substrate availability was identified at 0.07 μmolal NH 3 (aq). A maximum NH 4 + concentration of 20.5 μmolal, indicates that NH 4 + may be inhibitory at higher concentrations. Energy supply calculations performed using speciated chemical abundances reveal a minimum of 0.017 J (kg of water) −1 required for NH 3 (aq) oxidation. Both the NH 3 (aq) concentration and NH 3 (aq) oxidation energy supply minimums, as well as the NH 4 + maximum concentration, lead to significant differences in AOA distribution ( p < 0.0001; Mann‐Whitney U test). Thus, microbial distribution is better defined by speciated substrate concentrations, such as NH 3 (aq) and NH 4 + , and the energy supply from NH 3 (aq) oxidation, than by total chemical abundances or energy supplies calculated from total chemical abundances.

Introduction This study examines the effects of de facto and de jure financial globalization, environmental policy stringency, and economic growth on green energy consumption in OECD countries. Methods The study employs the Driscoll–Kraay robust estimation method and Method of Moments Quantile Regression (MMQR) using data for OECD countries covering the period 1995–2020. Results The findings reveal that environmental policy stringency makes a considerable contribution to green energy consumption. When environmental policy stringency prevails and green energy investments are low, the de facto effect of financial globalization substantially improves the green energy transition. Discussion As green energy investments increase, the de facto effect weakens while the de jure effect strengthens, indicating that different forms of financial globalization contribute differently across the green energy distribution.

Marine ranching has emerged as an important governance instrument for marine ecosystem restoration and the sustainable enhancement of fisheries within the contemporary global ocean governance paradigm. Against this background, this study examines the evolution, institutional structure, and governance challenges of China’s marine ranching legal system. Methodologically, the study employs normative legal analysis, policy text analysis, and comparative institutional analysis. It systematically reviews national legislation, administrative regulations, departmental rules, technical standards, and local regulations concerning marine ranching, and further compares China’s institutional framework with relevant governance experiences in the United States, Japan, and Norway. The findings indicate that China has gradually developed a policy-driven governance framework for marine ranching, supported by national marine ranching demonstration zones, technical standards, and local legislative initiatives. However, the existing legal system continues to face several structural constraints, including a relatively low legislative hierarchy, fragmented regulatory authority, incomplete life-cycle supervision mechanisms, and insufficient institutionalization of ecological responsibilities. Comparative experience suggests that the sustainability of marine ranching depends not only on technological development but also on stable legal frameworks and well-coordinated regulatory mechanisms. Accordingly, this study argues that China should strengthen national-level legislation, improve cross-departmental coordination, and incorporate ecological accountability into legally binding governance structures, thereby enhancing both the ecological sustainability and institutional governance capacity of marine ranching.

Aurelia coerulea and Rhizostoma pulmo are two bloom-forming scyphomedusae with a complex life cycle characterized by benthic scyphistoma and pelagic medusa stages. Understanding how environmental variability regulates developmental transitions and asexual reproductive pathways in medusozoans is central to evolutionary developmental biology, particularly under ongoing climate change. The Mediterranean Sea is a climate-change hotspot, where rising temperatures and increasingly frequent marine heatwaves provide a natural context to examine how seasonal and extreme thermal conditions shape life-cycle regulation. Here, we conducted a nine-months experiment rearing polyps of both species under controlled conditions that reproduced weekly in situ temperature variations from the Bages-Sigean lagoon, combined with two experimentally imposed heatwaves (+5 °C for 6 days in December and +3.5 °C for 5 days in April). A. coerulea polyps exhibited strong developmental plasticity, proliferating through budding at low temperatures and rapidly colonizing the substrate (0.12 cm² day -1 ), before a marked decline in asexual propagation when temperatures persistently exceeded 15 °C. In contrast, R. pulmo polyps followed a highly conservative developmental pathway, almost exclusively producing podocysts throughout the experiment, with minimal substrate colonization (<1.5% of a 100 cm² surface), and summer temperatures acting as a trigger for podocysts excystment. Strobilation timing and magnitude further differed between species: A. coerulea strobilated mainly from mid-November to February (>200 ephyrae per 100 cm²), whereas R. pulmo showed an extended strobilation period from January to July, peaking in late March (>30 ephyrae per 100 cm²). Temperature significantly regulated the production of buds, podocysts, and strobilae in both species, but with contrasting thermal optima: cold conditions favored A. coerulea strobilation, whereas warming enhanced R. pulmo strobilation. Heatwave exposure further modified these responses, indicating context-dependent developmental sensitivity to extreme events. Together, these results demonstrate that closely co-occurring medusozoans can express fundamentally different life-cycle regulation strategies under identical thermal regimes, highlighting how developmental pathways may mediate species-specific responses to climate warming and shape future jellyfish bloom dynamics.