Search Results (1,324)

Low-light conditions degrade remote sensing imagery by reducing contrast, distorting color, and obscuring fine terrain structures and small objects critical for Earth observation. Accurate illumination adjustment under spatially varying scene content remains challenging for existing enhancement methods, and many prior-guided approaches operate exclusively in either the spatial domain or the frequency domain. In this work, we propose a Dual-Domain Illumination Prior (DDIP), a trainable dual-domain illumination-prior module that is jointly optimized with each host backbone and exploits frequency-domain and spatial-domain illumination statistics. DDIP comprises three components: a Frequency-Domain Illumination Distribution Prior (FIDP) that performs per-color-channel amplitude calibration in Fourier space to improve global brightness; a Spatial-Domain Illumination Distribution Prior (SIDP), adapted from IDP-Net, that performs multi-scale sub-region statistical correction for local illumination adjustment; and a Selective Core Feature Fusion (SCFF) module that adaptively combines the frequency-domain output, the spatial-domain output, and the original input through an attention-based gating mechanism with dual pooling. DDIP is integrated with each host backbone while leaving its main restoration blocks unchanged. In the controlled reconstruction comparisons on iSAID-dark and the evaluated general low-light benchmarks, equipping the tested backbone networks with DDIP improves PSNR and SSIM over their corresponding baselines. Complementary LPIPS and CIELAB lightness measurements characterize perceptual similarity and lightness behavior, while a fixed-detector object-detection evaluation on the tested high-resolution iSAID-dark scenes examines the effect of the enhancement pipelines under the reported synthetic low-light conditions. The ablation studies further examine the contribution of the module components within the reported experimental settings. Full article

Geo-information texts, including geospatial data-use regulations and Earth observation metadata, are central to data governance and compliance auditing in remote sensing ecosystems. Full fine-tuning of large pre-trained language models is often computationally impractical, while standard LoRA reduces cost but assigns a fixed rank to all adapted modules, ignoring differences across layers and projection types. This paper proposes TARA, a task-adaptive rank allocation method for LoRA-based fine-tuning. TARA assigns learnable importance scores to rank dimensions and uses Gumbel–Sigmoid sampling with the Straight-Through Estimator to learn discrete rank masks under a global sparsity constraint. We further construct RSRegulation, a geospatial regulatory compliance benchmark containing 4032 English-language samples derived from 168 clauses across seven regulatory and policy sources with clause-level data isolation. Across five random seeds, TARA achieves 95.30 ± 0.10% accuracy and 95.44 ± 0.10% F1 with a maximum trainable adapter budget of 1.57 M parameters. The learned soft allocation corresponds to approximately 0.38 M effective adapter parameters and 75.8% soft rank compression. Physical hard pruning reduces the deployed adapter to 0.086 M parameters while retaining 95.12 ± 0.11% accuracy and 95.21 ± 0.10% F1. Layer-wise analysis shows that value projections retain higher ranks than query projections under the current task and backbone, revealing a task-dependent non-uniform allocation pattern. Full article

The remote sensing scene classification (RSSC) task plays a pivotal role in Earth observation missions, yet its progress remains constrained by the scarcity of high-quality labeled imagery. This article introduces a self-supervised learning (SSL) paradigm to address this challenge. First, for pseudo-label construction, a large set of long-interval satellite revisit imagery is collected and processed with pixel-level registration. The SIFT inliers retained during registration serve as saliency priors to guide asymmetric masking across views. This produces positive pairs that preserve global scene consistency while introducing controlled object-level ambiguities. Second, we propose a progressive layer-wise contrastive learning framework (MTC-Net) that couples the pseudo-label with the network’s representational hierarchy, forming a curriculum from local texture robustness to global semantic invariance. A dual-attention module with spatial–channel branches is further embedded to recalibrate intermediate features. The learning paradigm encourages the model to perform cross-view contextual reasoning rather than relying on pixel-wise correspondences. Experiments on three widely used datasets demonstrate that MTC-Net achieves competitive classification accuracy under limited-label settings, while ablation and visualization studies validate the effectiveness of establishing scene-level invariance through multi-temporal contrastive alignment. Full article

Formic acid (HCOOH), the most abundant carboxylic acid in the troposphere, modulates rainwater acidity, aerosol water uptake, and the oxidative capacity of remote atmospheres, yet its global budget remains poorly constrained. Herein, we present the first HCOOH total-column retrieval from thermal–infrared (TIR) measurements of the Thermal And Near-infrared Sensor for carbon Observation Fourier Transform Spectrometer-2 (TANSO-FTS-2) on board GOSAT-2, providing an early-afternoon observational perspective that complements existing morning low-Earth-orbit and geostationary HCOOH products. The Optimal Estimation retrieval sequentially fits the surface state, the atmospheric background (temperature, water vapor and ozone), and the HCOOH profile in a 1104–1109 cm⁻¹ microwindow centered on the ${\nu }_6$ Q-branch, with a radiance-ratio-scaled a priori that adapts to each scene. Averaging-kernel diagnostics concentrate the sensitivity in the 500–900 hPa layer with degrees of freedom for signal of approximately 1.05 under enhanced-emission conditions. For a 2019–2020 Australian bushfire case, including HCOOH in the state vector reduces the mean spectral residual from −0.327 K to 0.033 K. Independent evaluation against 113 time-coincident Toronto NDACC FTIR overpasses gives R = 0.95 and a zero-intercept slope of 2.12 for raw FTIR versus GOSAT-2. Applying the GOSAT-2 a priori and averaging kernel to the FTIR profiles changes the slope to 0.77 and reduces the RMSE to $0.23\times {10}^{16}$ molec cm⁻²; this one-sided smoothing is treated only as a sensitivity diagnostic. Monthly global maps for December 2019 and June 2020 show cross-sensor consistency with the IASI/MetOp-B ANNI-HCOOH product at R = 0.83 and 0.76. Over East Asia during April–June 2023, GOSAT-2 correlates with FY-4B/GIIRS at R = 0.90 (April) and R = 0.65 (June), with coherent three-sensor daily variability. These satellite comparisons are treated as cross-sensor consistency assessments rather than independent validation. GOSAT-2 consistently reports lower columns, a sensitivity-limited tendency consistent with a priori dominance under weak signals, limited information content, a narrow retrieval window, and differences among retrieval frameworks. The current product is a first demonstration for cloud-free daytime land scenes; this domain defines its sampling scope and representativeness but is not interpreted as a direct cause of the lower columns. The product offers a traceable GOSAT-2 TIR observational constraint on tropospheric HCOOH for future multi-platform synergy. Full article

Effective conservation of the Near Threatened (NT) black-necked crane (Grus nigricollis) requires standardized frameworks for organizing multi-scale biodiversity data. This study proposes the Essential Biodiversity Variables (EBV) system as an optimal structure for archiving and sharing biodiversity data such as Habitat Suitability Index (HSI). Developed by the Group on Earth Observations Biodiversity Observation Network (GEO BON), the EBV framework is an emerging system offering a robust solution for standardizing data exchange. Based on 483,592 valid location records of 106 black-necked cranes using satellite telemetry, we apply species distribution models and demonstrate how the multi-dimensional EBV architecture accommodates distinct life-stage preferences: breeding sites favor mid-elevations modulated by temperature; migration staging relies on precipitation regimes; and wintering grounds are driven by moisture availability and the avoidance of human-modified landscapes. The EBV NetCDF (Network Common Data Form) format functions as a self-describing hypercube that captures spatial, temporal, and life-stage dimensions while ensuring metadata transparency. This integration facilitates critical applications, including the identification of priority conservation areas and climate vulnerability assessments, thereby bridging the gap between species-specific modeling and global biodiversity monitoring standards. Full article

Per- and polyfluoroalkyl substances (PFASs) are persistent synthetic chemicals of global concern. While most research has focused on their occurrence and toxicity, far less attention has been paid to their impacts on the global carbon cycle. This review synthesizes current evidence on how PFASs influence carbon cycling across soils, aquatic systems, and the atmosphere. In soils, PFASs alter organic carbon inputs by affecting plant biomass and root exudates and shift microbial community composition and enzyme activities, thereby modulating organic matter decomposition. In aquatic ecosystems, PFASs biologically impair carbon sequestration by inhibiting plankton, and abiotically interact with extracellular polymeric substances to prolong the cycling of dissolved organic carbon. The atmosphere acts as a key mediator as follows: thermal treatment of PFASs generates perfluorocarbons, potent greenhouse gases that exacerbate global warming and further disturb carbon cycling. Despite clear disruptive effects, major knowledge gaps remain. Future research should use quantitative structure–property relationship modeling to assess PFAS alternatives (e.g., PFHxS), and employ advanced molecular tracking (e.g., isotopic labeling, NanoSIMS) and machine learning to unravel nonlinear PFAS–carbon dynamics. Improved detection technologies are needed to identify greenhouse gas byproducts from PFAS thermal treatment. Ultimately, deploying high-resolution flux observation networks and integrating PFAS dynamics into Earth system models and carbon-accounting frameworks are critical for predicting carbon–climate feedback and supporting global carbon neutrality goals. Full article

Reliable intensity–duration–frequency (IDF) curves are essential for the design of stormwater drainage systems and flood mitigation infrastructure; however, conventional IDF relationships assume stationarity and may underestimate future rainfall extremes under climate change. This study developed climate-informed IDF curves for Makkah, Saudi Arabia, using hourly observed rainfall records (1985–2025) and projections from five CMIP6 Global Climate Models (EC-Earth3-CC, CNRM-CM6-1, GFDL-ESM4, MPI-ESM1-2-LR, and UKESM1-0-LL) under the SSP245 and SSP585 scenarios. Spatial downscaling was first carried out using bilinear interpolation, after which the resulting data were corrected for systematic bias using the Delta Change method. Daily precipitation projections were subsequently disaggregated to an hourly timescale using an enhanced KNN-MOF approach. Annual maximum precipitation series were then derived for durations of 1, 2, 3, 6, 12, and 24 h and fitted to a range of candidate probability distributions. The goodness of fit was evaluated using the log-likelihood, Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC). Across the five CMIP6 models, two emission scenarios, and six rainfall durations, the Log-Pearson Type III distribution consistently yielded the most satisfactory fit. Historical analysis estimated 100-year rainfall depths ranging from 7.84 mm (1 h) to 38.29 mm (24 h), while future projections indicated substantially higher design rainfall intensities under several climate models. For example, under the SSP585 scenario, the 100-year 1 h rainfall intensity reached 29.73 mm h⁻¹ for EC-Earth3-CC, whereas MPI-ESM1-2-LR projected a 102% increase in the 6 h 100-year intensity relative to SSP245. Sherman equations were successfully fitted to develop continuous IDF relationships, while bootstrap resampling and Bayesian inference quantified projection uncertainty. The multi-model ensemble indicated increasing uncertainty with return period, particularly for the 100-year event, highlighting the importance of incorporating uncertainty into engineering design. The proposed framework provides robust climate-informed IDF curves for supporting resilient urban drainage design, flood-risk assessment, and water resources planning in arid environments. Full article

Extreme droughts can rapidly reshape surface-water patterns in river-connected floodplain wetlands, yet the fine-scale responses of individual dish-shaped sub-lakes remain insufficiently resolved. This study focused on the 2022 extreme drought in Poyang Lake, China’s largest freshwater lake and a globally important floodplain wetland system. The objectives were to quantify wetland landscape changes during the 2022 extreme drought event and to compare the heterogeneous responses of dish-shaped sub-lakes under contrasting surface-water linkage and management-context settings. We developed a high-resolution wetland monitoring framework on the Google Earth Engine platform by integrating Sentinel-2 multispectral imagery, Sentinel-1 synthetic-aperture radar, and the Dynamic World land-cover product. This multi-source approach was designed to reduce spectral confusion among turbid water, saturated mudflats, and exposed lakebeds during extreme low-water stages. Monthly wetland maps were generated for the Poyang Lake National Nature Reserve and compared with a five-year historical baseline from 2017 to 2021. The framework achieved an overall accuracy of 87.44%, with a Kappa coefficient of 0.80. The results revealed a rapid wet-to-dry transition in 2022. The water area contracted by 78.1% from July to September and remained 73.6–74.7% below the historical baseline from September to November. This contraction was accompanied by extensive observable surface-water fragmentation, apparent loss of visible surface-water linkage among sub-lakes, and substantial wetland habitat contraction. Sub-lakes exhibited clearly differentiated drought responses. Sub-lakes with stronger visible surface-water linkage to the main lake generally experienced more rapid water loss during recession, whereas reserve-managed or facility-present sub-lakes retained residual water to varying degrees and may have provided important refugial habitats for waterbirds and aquatic species. These findings suggest that surface-water linkage condition, local topographic setting, and 2022 management context were jointly associated with the heterogeneous drought responses of sub-lakes. The proposed framework provides a tool for monitoring wetland landscape changes associated with extreme drought events, assessing ecological vulnerability, and supporting adaptive water-level management under intensifying climate extremes. Full article

Recent advances in Earth observation and machine learning have significantly enhanced the capacity to monitor agricultural systems and terrestrial vegetation across spatial and temporal scales. This editorial synthesizes the contributions of eleven studies published in the Special Issue ‘Machine Learning for Applications in Agriculture and Vegetation Using Remote Sensing’. These contributions highlight emerging methodological trends, as well as persistent challenges in remote sensing for agriculture and vegetation measurement and monitoring, and reflect the growing dominance of deep learning in high-resolution mapping and segmentation. An increasing importance of multi-sensor data fusion, integrating multi- and hyperspectral satellites, UAV, and environmental data, is found. Deep learning is emerging as an effective approach for retrieving key biophysical parameters such as biomass, crop height, and yield. The collected studies also emphasize the critical role of sensor characteristics and scale, particularly the trade-offs between spectral, spatial, and temporal resolution in vegetation analysis. Despite notable progress, several limitations remain. Model transferability across regions and sensors is still constrained and multi-source data integration often lacks standardized frameworks. Empirical approaches still dominate the retrieval of biophysical variables, limiting robustness and physical interpretability. The contributions also reveal a persistent gap between high-resolution, small-scale analyses and their scalability to regional or global applications. Therefore, this editorial argues for a transition towards hybrid modeling approaches that combine physical knowledge with data-driven machine learning methods, the adoption of formal data assimilation frameworks for multi-source integration, and the development of scalable and uncertainty-aware workflows. The broader scientific context of the contributions is given by providing a critical perspective on the current state of the field and outlining the key research directions necessary to advance remote sensing in agriculture and ecosystem monitoring. Full article

Understanding the processes behind Earth’s CO₂ degassing is crucial for clarifying how carbon is cycled by geologic processes on our planet. Earth’s carbon degassing results from magmatic and metamorphic processes controlled by large-scale plate tectonics. However, the nature and amount of diffuse CO₂ fluxes from faults in continental rifts remain largely unconstrained. Recent research reports important CO₂ fluxes from deep faults over 17 × 10⁴ km² in the East African Rift System (EARS), highlighting the need for a more detailed inventory of diffuse soil emissions. Across such extensive regions, only satellite observations can provide the broad measurement range needed to meet the observational requirements for long-term, large-scale overlay datasets. While space-based data provide extended coverage for large-scale identification of CO₂ emission sources, to date, only the column-averaged total CO₂ atmospheric content has been measured from space. Range-resolved measurements of CO₂ concentration with the vertical accuracy needed to detect diffuse soil emissions are not available from space, preventing direct quantification of Earth degassing over large regions. In this paper, we focus on a new approach to measuring soil CO₂ fluxes by a statistical deconvolution of column measurements of global atmospheric CO₂ content. Anthropogenic increase, seasonal climatology at a small regional scale, and soil fluxes are measured without the use of ancillary measurements or model simulations. We combine petrology of fluid and melt phases in mantle rocks with a new satellite data analysis to reveal, for the first time, large CO₂ fluxes from the lithospheric mantle to the atmosphere in a much wider region than previously considered (22.4 × 10⁵ km²), including the Ethiopian and East African domes, where magmatism is no longer active or absent. We infer that diffuse soil emissions are measurable from space, and that Earth’s tectonic carbon fluxes are considerably more relevant than currently considered, with a crucial impact on the balance of the global carbon cycle. Full article

Irrigation water management in arid systems requires spatially distributed estimates of crop evapotranspiration (ET) that fixed crop coefficients cannot provide. The actual ET of flood-irrigated rice (Oryza sativa L.) on the arid northern coast of Peru was mapped with the METRIC surface energy balance model (Mapping EvapoTranspiration at high Resolution with Internalized Calibration) on Google Earth Engine (GEE). Ten cloud-free Landsat 8/9 scenes (January–July 2022) were processed over 113 ha at Ferreñafe (Lambayeque) on the 30 m product grid, onto which the 100 m native thermal observation was resampled, with internal calibration based on automatic anchor-pixel selection and hourly ERA5-Land data. Daily field-mean ET ranged from 4.2 to 8.1 mm d⁻¹, peaking during flooding and establishment and declining towards harvest. Because the same reference $E{T}_o$ underlies the METRIC internal calibration and the FAO-56 estimate, this is a comparison between two modelling approaches rather than an independent validation. Against the FAO-56 reference ET, METRIC showed a positive bias of $+0.65$ mm d⁻¹ (percent bias (PBIAS) $=+13\%$; root mean square error (RMSE) $=1.23$ mm d⁻¹; $r^2=0.57$; $n=9$, after excluding one date with anomalous reanalysis forcing), concentrated during flooding and after harvest, whereas at full canopy cover the two estimates converged. Two global ET products that share neither the METRIC formulation nor the ERA5-Land forcing reproduce the same seasonal decline once the canopy closes ($r=0.63$ and $0.91$) but stay far below in magnitude, as expected from their 500 m pixel. ET did not differ between sowing methods and varied only slightly among cultivars (∼0.3 mm d⁻¹), against marked intra-field variability. The METRIC–GEE workflow offers a low-cost, high-resolution tool for monitoring water use in data-scarce arid rice systems. Full article

As China’s first commercial low Earth orbit meteorological constellation, Tianmu-1 (TM) radio occultation (RO) provides high-density observations, with approximately 30,000 profiles per day globally, offering new opportunities for research in observationally sparse regions. TM RO bending angle and refractivity retrievals from June to December 2023 are evaluated using ERA5 reanalysis, global radiosonde observations, and COSMIC-2 RO as reference datasets. TM observation errors are further estimated using the three-cornered hat (3CH) method. The results show that, in the upper troposphere and lower stratosphere, TM bending angle biases are within ±0.19%, ±0.40%, and ±0.37% relative to ERA5 (8–30 km), radiosonde (8–25 km), and COSMIC-2 (8–30 km), respectively. The corresponding refractivity biases are within ±0.09%, ±0.21%, and ±0.14%, respectively. Both bending angle and refractivity from TM RO exhibit high accuracy and precision. The observation errors estimated using the 3CH method show a clear latitudinal dependence. The observation errors in the tropics (±30° latitude) are larger than those in the middle and high latitudes. This study quantitatively characterizes the accuracy, precision, and error structure of TM RO bending angle and refractivity, providing a quantitative basis for future quality control and observation error specification in numerical weather prediction data assimilation. Full article

In low-Earth-orbit (LEO) satellite constellation remote sensing for surface maritime target detection, the inherent characteristics of discontinuous detection epochs, non-uniform temporal intervals, and clutter contamination invariably cause conventional data association algorithms to suffer from validation gate degradation, covariance divergence, and combinatorial explosion. To circumvent these limitations, this paper proposes a multi-target, time-adaptive fast association method tailored for discontinuous sparse observations. Within the joint probabilistic data association (JPDA) framework, the proposed method analyzes the mismatch between the Kalman filter prediction covariance and the actual error under discontinuous observations. A time-interval adaptive gating mechanism maintains the gate detection probability across arbitrary revisit intervals. Secondly, to resolve the massive connected cluster problem triggered by the densification of the validation matrix, a progressive clustering strategy inspired by simulated annealing is designed, which recursively decomposes the global, exponentially scaling association graph into independent subgraphs of manageable sizes. Building upon this, a depth-first search (DFS) heap pruning technique is integrated with the Hungarian hard assignment algorithm as a safety-degradation mechanism to safeguard numerical robustness in extreme scenarios. Comparative experiments demonstrate that the proposed method significantly enhances both tracking accuracy and track completeness across various constellation coverage characteristics and maritime clutter intensities. Furthermore, its execution efficiency satisfies real-time simulation requirements, effectively supporting engineering application for surface maritime target detection. Full article

Regional development disparities remain a persistent global challenge, yet existing assessment approaches often face a trade-off between spatial detail and temporal coverage. Conventional socioeconomic statistics provide relatively reliable information but are typically limited by coarse spatial representation and low update frequency, whereas high-resolution remote sensing-based studies often focus on individual time points, making it difficult to capture the temporal evolution of regional development. Remote sensing observations provide valuable opportunities for regional development assessment by offering extensive spatial coverage and repeated observations over time. To address this gap, this study proposes a multisource remote sensing framework for characterizing the spatiotemporal dynamics of regional development in Chongqing Municipality across four temporal nodes (2014, 2016, 2018, and 2020). We first construct a county-level Development Intensity Index (DII) using socioeconomic indicators derived from statistical data. Subsequently, we integrate nighttime light, DEM, NDVI, and POI data to generate a 500 m gridded Comprehensive Spatial Development Index (CSDI), which captures spatial heterogeneity at a fine spatial scale. The CSDI exhibits strong correspondence with the DII, and its spatial validity is further corroborated through visual interpretation of Google Earth imagery. Results indicate that areas with higher development levels are predominantly concentrated in Chongqing’s central urban core, while less-developed counties are concentrated in the northeastern and southeastern peripheries. Although a general upward trend in development is observed across the study period, notable spatial disparities persist. Overall, the proposed CSDI-based framework offers an effective and replicable approach for gridded regional development assessment, with implications for targeted regional planning and differentiated policy design. Full article

Russia’s invasion of Ukraine has posed serious risks to global food security, by causing substantial crop yield losses since 2022. Accurate yield estimation helps policymakers to plan compensation, yet modelling yields in conflict regions remains challenging due to significant non-meteorological disruptions. This study proposes a novel framework to quantify war-related crop yield losses by comparing estimations derived from meteorological data, representing weather-driven yield variability, with those based on Earth observation (EO) data, reflecting actual crop conditions influenced by both weather and conflict. Thus, meteorologically based yield estimates are expected to exceed those derived from EO data, with the difference indicating war-related losses. Both, meteorological- and EO-based models, are developed using transfer learning (TL) to estimate yields of maize, winter wheat, and spring barley. Models are initially trained on EU country data and subsequently finetuned with Ukrainian data. Their performance is compared to two non-TL approaches: Extreme Gradient Boosting (XGB) and Artificial Neural Network (ANN) to test their reliability. Results show crop yield losses for maize; however, since we do not detect losses in the other crops, we conclude that simply comparing meteorological- and EO-based models proves insufficient to fully isolate conflict effects due to strong interactions of EO and meteorological data. Nevertheless, TL substantially enhances prediction accuracy (R² around 0.7), exceeding alternative models by 0.05–0.2 across crops. These findings demonstrate the value of TL for yield modelling in data-scarce environments and underscore the need for improved methodologies to quantify conflict-induced agricultural losses. Full article