Timely crop mapping is crucial for field management, policy formulation, phenological monitoring, and yield forecasting. However, acquiring sufficient labeled samples in the current year presents a formidable challenge for in-season mapping. Previously proposed solutions mainly include classifier transfer and sample transfer strategies. The classifier transfer strategy trains classifiers with historical samples associated with historical-year features and then transfers the trained historical sample classifiers (HSC) to classify remote sensing data in the current year; the sample transfer strategy generates trusted samples associated with current-year remote sensing features by predicting labels of the current-year sample based on some prior knowledge (e.g., crop rotation pattern) and then trains trusted sample classifiers (TSC) for current-year classification. However, the performance of the classifier-transfer strategy may degrade when there is large interannual feature variation, while the performance of the sample-transfer strategy depends on the reliability of the generated trusted samples. This study proposes a novel approach that integrates the above two strategies for in-season mapping through a sample weighting technique. Firstly, two sample sets, trusted samples and classified samples associated with current-year features, are generated by crop rotation prediction and HSC, respectively. Subsequently, based on an independent assumption between the rotational prediction errors and the current-year remote sensing features, the optimal weights of these two sample sets are derived based on the Bayesian principle. Finally, an optimal weighted sample classifier (OWSC) is trained using the weighted samples for in-season classification. To illustrate the robustness of the proposed OWSC, we compared it with different methods combined with various classification models across four regions with different interannual feature variation and crop rotation stability. Results demonstrated that OWSC maintained its advantages across various regions and different available lengths of historical crop-type sequences. Owing to its independence from specific classifiers, the proposed sample weighting method can be seamlessly applied to any classification model and thus continues to benefit from advances in classification algorithms. Additionally, sensitivity experiments regarding the uncertainty in trusted samples and historical crop-type sequences showed that OWSC performed stably across different scenarios. Therefore, OWSC provides a promising solution for in-season crop mapping without current-year samples.

Vegetation indices (VIs) are widely used as proxies for vegetation greenness and vigor. However, soil background effects—especially those caused by variations in soil moisture and type—create complex spectral mixtures of vegetation and soil, substantially impacting the accuracy and reliability of these indices. This study assessed the sensitivity of normalized difference vegetation index (NDVI) and 31 soil-resistant indices to soil effects, focusing on the independent effects of soil moisture and soil type, and overall soil effect. The soil-resistant indices are categorized into 6 groups according to the design mechanisms: soil-line adjusted, photosynthesis oriented, shape separation, shortwave infrared (SWIR) adjusted, RedEdge adjusted, and green triangular indices. Using both 3-dimensional (3D) radiative transfer model (RTM) simulations and ground-based experiments, the analysis shows that (a) in sparse canopy conditions where soil background effects are most pronounced, 22 (simulation experiments) and 26 (ground-based experiments) of the 31 soil-resistant indices exhibited lower overall sensitivity compared to NDVI, with soil-line adjusted and SWIR adjusted indices outperforming the other VIs. (b) Fifteen indices demonstrated superior resistance to both soil type and soil moisture compared to NDVI, and 8 exhibited enhanced resistance specifically to soil type, primarily belonging to the SWIR adjusted group, while 3 indices, all RedEdge adjusted, exhibited improved resistance to soil moisture. (c) Sensitivity to soil type predominantly determines overall soil effects. These findings provide important insights into understanding how soil-resistant indices can reduce soil-related uncertainties, and provide a sound basis for selecting appropriate VIs in remote sensing applications related to agricultural and ecological studies.

Ratoon rice plays a vital role in boosting land productivity and contributing to stable food supplies under the context of global climate change. This system offers these advantages by demonstrating the capacity to produce an additional grain yield of 5-6 t ha-1 from the ratoon season, while simultaneously reducing the growth duration by 44-48 days compared to conventional double-season rice cultivation. However, its precise spatial distribution remains unclear under varying climatic and surface conditions, and remote sensing-based monitoring of ratoon rice has received limited attention. Existing methods often struggle to accurately distinguish ratoon rice from other morphologically similar types, such as double-season rice, and are further hampered by frequent cloud cover and rainfall, compromising optical sensing effectiveness. This study proposes a two-step ratoon rice (TSRR) mapping method using multi-source remote sensing data at the parcel scale. The TSRR method integrates Sentinel-1A synthetic aperture radar (SAR) and Sentinel-2 optical imagery, utilizing the distinct growth characteristics of main-season and ratoon rice. It employs object-based segmentation and two novel indices-the SAR-based paddy rice index (SPRI) and the SAR-based ratoon rice index (SRRI)-without relying on detailed phenological information. Results indicate that the TSRR method effectively distinguishes ratoon rice from other paddy rice types, achieving an average overall accuracy (OA) of 0.87, with particularly high performance in separating ratoon rice from double-season rice. The TSRR method demonstrates strong robustness and transferability across different regions, which can provide a reliable solution for large-scale paddy rice mapping, especially in cloud-prone areas with limited optical data availability, and offers valuable support for crop monitoring, yield estimation, and national agricultural inventory initiatives.

Dengue is a widespread arboviral disease and poses a persistent global health threat, particularly in Southeast Asia, a region characterized by frequent dengue and extreme weather. Urbanization, infrastructure, and mosquito thermal suitability are pivotal factors in shaping dengue dynamics. However, their roles in modulating the causal effects of extreme weather on dengue transmission remain unclear. Here, we compile monthly dengue data during 2000-2019 covering Singapore and 162 first–level administrative units in six Southeast Asian countries. A spatiotemporal Bayesian framework is developed to explicitly capture the interactions of extreme weather with urbanization and with infrastructure. We also apply convergent cross mapping (CCM) to identify causal associations between extreme weather and dengue, incorporating vector thermal suitability, urban development, and access to improved water and sanitation. Among extreme weather, heatwaves pose the highest relative risk (RR) [max RR 5.5, 95% CI: 3.27–9.25], followed by drought [max RR 5.21, 95% CI: 3.31–8.21], wet conditions [max RR 2.76, 95% CI: 1.79–4.26], and compound drought–heatwave events [max RR 2.55, 95% CI: 1.3–5.01]. Causal analysis indicates divergent pathways: in urban settings with improved infrastructure and high mosquito thermal suitability, drought and compound drought–heatwave events emerge as key drivers; in rural settings with poor infrastructure and low mosquito thermal suitability, their effects are not significant. This study reveals that extreme weather–driven dengue is modulated by social vulnerability and mosquito thermal suitability, and provides practical insights for integrating extreme weather into adaptation strategies for dengue control.

Timely and accurate estimation of crop area is fundamental for agricultural policy formulation, production forecasting, and economic assessment. Probability-sampling-based area estimation provides statistically rigorous inference with quantifiable uncertainty, but its operational implementation remains challenging. Two barriers are particularly important for probability-sampling-based in-season crop area estimation. First, collecting spatially dispersed probability samples in field is highly costly, especially in regions under complex traffic transportation conditions. Second, although progressively updated in-season crop maps provide valuable auxiliary information for improving estimation precision, standard post-stratified estimation is often impractical when samples were originally selected using an earlier stratification map (the updated post-strata may strongly overlap with the original sampling strata, leaving some substrata with insufficient samples). To address these two challenges, we present a practical framework for progressive in-season crop area estimation by introducing two complementary techniques. First, a vehicle–unmanned aerial vehicle (UAV) cooperative response design assigns sample units to multiple park-and-launch sites and minimizes the overall time cost of the vehicle tour and UAV sorties using ant colony optimization, thereby improving the efficiency of the probability samples collection. Second, a poststratified combined ratio estimator, a design-consistent estimation method that does not require sufficient samples within each substratum, is applied to progressively refine crop area estimates by leveraging the updated in-season crop maps in later season. In a field experiment conducted for rapeseed area estimation in Yanting County, 263 probability sample units selected under the initial stratification (the first in-season crop map) were collected in 31 hours using the proposed UAV cooperative sampling method. The poststratified combined ratio estimator was then introduced to update the crop area estimate using later-season crop maps, which reduced the standard error of the area estimate from 1.34% to 1.06% (≈21% relative reduction). Overall, the proposed framework provides a statistically robust rigorous and operationally feasible solution for progressive in-season crop area estimation.

Planted fields, which are defined by areas containing different crop types, are essential for agricultural planning, crop mapping, and yield estimation. Accurate and timely information on planted fields is crucial. The Segment Anything Model 2 (SAM2) offers advanced image segmentation but is not directly suited for planted field segmentation. We propose ASAMPS, an Adaptive SAM2 model tailored for segmenting planted fields from single-date or time series remote sensing images. Unlike other SAM-based adaptations that require fine-tuning the model, ASAMPS incorporates three external modules to automatically generate and optimize prompt points, aligning SAM2 with the specific requirements of planted field segmentation without any retraining. Testing across six agricultural regions in China and the United States, ASAMPS demonstrated robust performance, achieving accuracy comparable to or surpassing state-of-the-art supervised methods. It demonstrated strong performance in complex landscapes and was effective across multiple satellite platforms, including Planet, GF-2, Sentinel-2, and Landsat 8 OLI. Additionally, ASAMPS efficiently extracted minimal planted fields from multi-temporal imagery. ASAMPS leverages SAM2's capabilities without requiring retraining and offers enhanced flexibility for optimizing prompts, making it suitable for diverse agricultural monitoring applications.

The Earth is experiencing continuous anthropogenic and natural changes. Very high resolution (VHR) remote sensing imagery-based change detection provides an effective means to monitor these dynamics at fine spatial scales. Although deep learning has significantly advanced supervised change detection (CD), it heavily relies on large amounts of human-labeled samples. In real-world CD application scenarios, acquiring sufficient change samples is challenging due to the labor-intensive nature of pixel-level labeling. This challenge has motivated the development of unsupervised change detection (UCD). However, existing UCD methods still struggle in complex scenes with bi-temporal domain shifts caused by different imaging conditions. This is largely due to the absence of high-quality samples needed to guide CD-oriented optimization. To address this challenge, we propose DreamCD, a change-label-free framework that synthesizes change samples for UCD. DreamCD consists of: (1) a weakly conditional semantic diffusion model trained with pseudo-semantic masks, (2) a Content-Semantic-Style synthesis strategy that synthesizes realistic pre- and post-event image pairs of the application domain, and (3) an arbitrary contemporal deep change detector trained solely on synthetic samples. We further introduce LsSCD-Ex, a large-scale semantic change detection (SCD) dataset consistent with OpenEarthMap semantics, enabling evaluation of synthetic-sample-based SCD. Experiments on the SECOND and LsSCD-Ex datasets demonstrate that DreamCD achieves state-of-the-art (SOTA) UCD performance, improving the average F1 score by 14.01% over existing methods for binary CD and outperforming the SOTA unsupervised SCD model, Changen2, by 2.15% in F1 and 3.63% in separated kappa coefficient (SCD metric). These results suggest that DreamCD provides a promising and extensible solution for CD in real-world remote sensing applications. Code and LsSCD-Ex dataset are available at https://github.com/tangkai-RS/DreamCD.

Ground crop yield records are typically published only at regional scales to protect farmers’ privacy. To predict regional crop yields, remote sensing and meteorological variables are commonly aggregated (e.g., averaged) to match the spatial scale of ground yield records. However, spatial aggregation of remote sensing data—such as the simple averaging of pixel-wise vegetation index (VI) time series—ignores intra-regional phenological variations. Several vegetation phenology studies unveiled a phenological bias caused by the simple average, which distort the representation of crop-specific phenological characteristics and subsequently compromising the accuracy of crop yield modeling. To address this critical limitation, we therefore proposed a novel yield prediction framework that explicitly incorporates intra-regional phenological heterogeneity. Instead of relying on aggregated VI time series, our method first applies adaptive clustering to automatically group pixels according to their growth dynamics. The resulting subregions and their respective area proportions are then integrated into a customized loss function within a convolutional neural network (CNN)-based modeling framework, termed Group-CNN. We evaluate the performance of Group-CNN in predicting soybean yields across 13 states in the U.S. Corn Belt, comparing it against multiple benchmark models. Results demonstrate that Group-CNN consistently outperforms the benchmarks, achieving a lower root mean square error (RMSE) (5.16 vs. 5.94–6.76 bu/ac), a higher coefficient of determination (R2) (0.83 vs. 0.72–0.76), and a lower mean absolute percentage error (MAPE) (9.78% vs. 11.72–12.81%). Ablation experiments confirm a notable decline in model performance when intra-regional phenological variations are excluded. This study highlights the importance of accurately representing intra-regional crop phenological characteristics—an often-overlooked factor in current yield prediction models.

Urban nighttime vitality significantly contributes to a city's economic, social, and cultural attractiveness. This study develops a dual-perspective framework that integrates annual mean nighttime light (NTL) intensity indicating stable infrastructural capacity, with a novel Nighttime Light Fluctuation Index (NLFI) capturing the dynamic pulse of human activity. Rather than measuring vitality directly, the framework extends intensity-based vitality evaluations by incorporating temporal dynamics, enabling the categorization of urban zones by distinct vitality-related patterns. Applying the framework to 25 global megacities, with detailed analyses in Los Angeles and Beijing, we identified three distinct patterns: "Bright-Fluctuating", "Bright-Steady", and "Low-Lit" zones. A consistent cross-city association was found between these patterns and urban functions: residential areas consistently exhibit low fluctuation, while commercial, industrial, and transportation hubs are hotspots of high fluctuation. Spatially, fluctuation hotspots in Los Angeles are concentrated in its central commercial and activity hubs, whereas in Beijing, they are also prominent in newly developing areas in the urban periphery, unveiling different urban dynamics and development. An XGBoost model combined with SHapley Additive exPlanations (SHAP) analysis revealed the key built-environment drivers behind these patterns. In Los Angeles, functional diversity and density (e.g., non-residential land use) are the primary drivers. In contrast, Beijing's patterns are predominantly shaped by infrastructure, with distance to the metro network being the most critical factor. These findings demonstrate the value of capturing NTL's temporal dynamics in understanding the interplay between built environment and nighttime urban dynamics, offering a powerful tool for urban planning and policy.

Autumn leaf fall is a critical phenological event in temperate deciduous forests, with important ecological and socioeconomic implications. Traditional estimates based on daytime vegetation indices primarily capture foliage color changes rather than the actual timing of leaf fall, and are often affected by mixed-pixel effects that reduce the accuracy. This study proposes a novel workflow for detecting the autumn leaf fall date (LFD) using nighttime light (NTL) data in urban areas, validated through in-situ observations from phenological cameras and city-scale assessments. Results showed that NTL-derived LFDs closely matched in-situ observations across three cities (New York City, Boston, and Beijing), with an RMSE of around 5 days and a bias of 0.77 days. In Beijing, both interannual (2012-2024) and spatial variations (2024) in LFD were delayed by higher preseason temperature and precipitation but advanced by greater strong-wind frequency, consistent with known autumn phenological controls and supporting the reliability of the NTL-based approach. These results demonstrate that NTL data can provide more accurate and interpretable LFD estimates than traditional daytime remote sensing, enabling detailed city-scale mapping. The use of NTL-derived LFD dynamics facilitates a more comprehensive understanding of their spatiotemporal patterns in urban environments and their linkages to climate change and human activities.

Ground-level ozone (O3) pollution threatens global agroecological security. In China, the world's largest tea producer, widespread O3 exposure has reduced tea quality, largely due to stomatal uptake of O3 at phytotoxic concentrations. However, available approaches inadequately capture stomatal regulation dynamics and spatial heterogeneity, limiting precision risk management in tea cultivation. We developed a novel semi-mechanistic exploratory index (Rtea) at 1 km resolution, integrating environmental exposure, physiological response and spatial weighting. Using a stomatal conductance model, we quantified the O3 stomatal-modulated exposure index and identified spatiotemporal risk patterns using decomposition methods. Ground-level O3 hotspots (characterized by multiyear mean daily maximum 8-h average (MDA8) O3 concentrations of up to 56 ppb) were concentrated in the northeastern part of the study area and coastal regions, with a widespread upward trend observed across most tea-producing areas from 2000 to 2023. Provincial mean Rtea exhibited a fluctuating upward trend, peaking at 17.10 ppb·days of O3 exposure in Zhejiang by 2020, with 2017 as a critical inflection point. Spatial-temporal logarithmic mean Divisia index (ST-LMDI) analysis revealed a geographic dichotomy where environmental exposure intensity drives risk in industrialized regions, whereas planting density acts as the primary driver of increased O3 exposure in regions with high tea plantation density. For precision governance, a bivariate risk matrix was constructed, categorizing regions by static risk (grades I-IV) and dynamic annual growth rates (grades A-D). Grade I-A (core control) zones, representing the highest risk base and experiencing the most rapid deterioration, were predominantly located in parts of Sichuan, Fujian, and Henan. In this novel study, a comparative assessment is conducted and spatial maps of relative O3 exposure risks within tea production systems are constructed, supporting targeted mitigation strategies, sustainable tea production and ecological risk assessment in global leaf-based economic crop systems, with applicability across regions.

Alpine grasslands are vital for biodiversity and ecosystem service delivery, yet their responses to climate change remain a focus of intense scientific interest. While studies have examined separate ecological processes such as biodiversity changes or treeline shifts, broader ecosystem changes are not as well understood. Here, leveraging time series of high-resolution images from the Landsat satellite, we developed an automated method to track the shifts in the upper limits of alpine grasslands on the high and expansive Tibetan Plateau, a region of high ecological and climatic significance. Our analysis revealed modest upward boundary shifts of -0.55 to 0.99 m year-1 (2nd-98th percentile; mean = 0.12 m year-1) across the plateau over nearly four decades, from 1986 to 2023, with faster rates in wetter areas, yet drastically slower than the rapid climate-driven isotherm shifts of 3.35 to 12.04 m year-1. This substantial lag, further confirmed by ≤ 3 m resolution satellite images, is potentially attributed to water scarcity, poor soil quality, and a lack of stable substrates beyond the current boundaries and geomorphological features. Consequently, the upward expansion of alpine grasslands on the Tibetan Plateau-attributed to the shift of the upper grassland boundary-was limited to approximately 6100 km2. Notably, two high-spatial-resolution CMIP6 models simulated more rapid upward expansion of alpine grasslands and greater carbon sequestration than observed. These findings underscore the need to integrate local environmental nuances into future predictions. Our study elucidates the resilient yet vulnerable nature of alpine ecosystems, sparking new conversations regarding effective strategies to safeguard these extraordinary landscapes under changing climate.

Timely and accurate monitoring of abrupt natural disasters, such as floods, landslides, and wildfires, is critical for protecting lives and property. On-orbit satellite image change detection offers near-real-time disaster insights, yet existing methods, designed for ground-based tasks, overlook the resource constraints of on-orbit computing environments. To fill this study gap, we propose a novel change detection method, Single-temporal HIgh-spatial rEsoLution image unsupervised change Detection (Shield), tailored to the constraints of on-orbit computing environments. Shield integrates the capabilities of change detection models that can identify various disaster events, with the efficiency of anomaly detection methods, which demand fewer computational resources and only require a single post-disaster image along with prior knowledge as input. First, Shield generates lightweight prior knowledge from pre-disaster imagery to establish connections between the changes and anomalies. Then, Shield employs a 2-step localization strategy to progressively identify change patches and pixels in post-disaster imagery to provide disaster-affected areas of interest. We validated Shield’s performance in 4 typical disaster scenarios: landslides, floods, wildfires, and deforestation. When evaluated against 8 classical and state-of-the-art bi-temporal change detection methods, as well as 2 single-temporal anomaly detection methods, Shield exhibited superior performance, achieving an average F1 score improvement of 24.37%. Furthermore, Shield’s application in 2 large-scale disaster scenarios in 2023 highlighted its robust performance and efficiency. It achieved a 5- to 239-fold reduction in data storage requirements and up to a 136-fold increase in on-orbit real-time detection speed compared to alternative methods, reinforcing its substantial potential for on-orbit operations. Code available at https://github.com/tangkai-RS/Shield.

Monitoring grain harvest events is crucial for crop management, yield estimation, and food security. Timely detection supports machinery allocation and mitigates risks from extreme weather. However, most existing studies rely on dense time series or labeled data, limiting real-time applications. This study proposes an unsupervised, index-based thresholding framework using Sentinel-2 imagery for major grain crops (paddy rice, maize, and wheat). A new Reaped Index (RI) was developed to distinguish harvested from unharvested fields based on their spectral differences. An adaptive thresholding algorithm addresses class imbalance across the early and late harvest stages. The framework was validated in Wuchang and Huaxian (China) and central Iowa (US), achieving F1 scores of 85–98 %, outperforming existing unsupervised approaches and vegetation indices. Therefore, this is a novel method for near-real-time harvest event detection, enabling label-free, rapid, and large-scale monitoring to support agricultural management under increasing weather risks.

The escalating impacts of global climate change and extreme weather have intensified flood risks worldwide, including in arid and semi-arid regions traditionally considered low-risk. This study examines the spatiotemporal dynamics of flood events across Kazakhstan from 2000 to 2024 by integrating remote sensing (RS) with machine learning (ML). Using Google Earth Engine (GEE), we address data gaps and cloud interference through spatiotemporal fusion (STARFM), denoising, smoothing, and sample transferring techniques. In addition, this study incorporates the TimeDisaggregated Water Frequency (TWF) method, which enables the identification of water bodies with temporal variability, eliminates permanent water bodies, and distinguishes flood from non-flood conditions in seasonal water bodies, thereby enhancing the accuracy of flood reconstruction and enabling precise delineation of flood inundation areas. Landsat and MODIS imagery are combined to produce high-resolution flood distribution maps, while spectral similarity indicators guide the transfer of samples from the Global Flood Database. A range of spectral, texture, environmental, and socioeconomic features is extracted, with flood classification performed using random forest (RF) and attribution analysis conducted via XGBoost and SHAP. Results highlight a high flood risk in northern, southwestern, and western Kazakhstan, primarily driven by changes in precipitation (PRE), temperature (TEM), soil moisture (SM), and land use. Floods occur most frequently in spring - especially in March and April - due to snowmelt and extreme precipitation. The ML models achieve over 80 % classification accuracy, demonstrating their reliability. This work improves flood monitoring and provides essential insights for climate adaptation and targeted flood risk management in Kazakhstan.

Tasseled Cap Transformation (TCT) is a widely applied remote sensing technique for dimensionality reduction and physical feature enhancement, valued for its interpretability and efficiency. While TCT coefficients have been developed for numerous sensors, no dedicated coefficient set has been proposed to date for Landsat 9 Operational Land Imager-2 (OLI-2) sensor. This study addresses this gap by deriving the first TCT coefficients for Landsat 9 OLI-2 based on surface reflectance data. To ensure cross-sensor consistency, we simultaneously recalculated Landsat 8 Operational Land Imager (OLI) coefficients using strictly matched underfly image pairs and identical sample selection protocols. This harmonized derivation strategy minimizes methodological and sampling-induced discrepancies, enhancing compatibility between the two Landsat sensors. More importantly, the proposed coefficients offer improved performance in capturing wetness-related information across diverse ecological settings. This makes them especially suitable for applications involving soil moisture monitoring, vegetation stress detection, and hydrological modeling in spatially and temporally heterogeneous landscapes. The validation results demonstrate that the newly proposed coefficients effectively enhance spectral differences among surface features with varying brightness, greenness, and moisture content. Moreover, the TCT components from Landsat 8 OLI and Landsat 9 OLI-2 exhibit strong agreement across all three components (R-2 > 0.96, approaching 1), underscoring the high consistency of the derived coefficients. Sensitivity analyses further reveal that the wetness and greenness components remain highly stable under varying sample selection conditions, while the brightness component, though slightly more sensitive, still maintains angular differences within 5 degrees. The results highlight the improved physical consistency and cross-sensor compatibility of the proposed coefficients, facilitating more robust long-term environmental monitoring and multi-source data integration in Earth observation studies.

Low-light image enhancement (LLIE) requires a delicate balance between modeling local details and cap turing global context, a challenge that current architectures struggle to address effectively. Convolutional neural networks are limited by local receptive fields, while Transformers incur prohibitive computational costs. Although State-Space Models (SSMs) offer improved efficiency, they often exhibit weak local feature extrac tion and multi-path scanning-induced feature redundancy. Moreover, standard MLP-based units lack the ability to model the complex nonlinearities inherent in low-light degradation. To address these challenges, we pro pose RetinexKANMamba, a novel framework anchored by the KANMamba module. This module incorporates three strategic components: (1) an Illumination-Guided Block (IGB) that constructs an illumination-aware fea ture pyramid for adaptive enhancement across varying exposures; (2) an Enhanced Multi-Scale Mamba (EMM), which combines a Local Spatial Enhancement Block (LSE-Block) for detail aggregation and a Multi-Scale Spatial Mamba (MSP-Mamba) to mitigate feature redundancy; and (3) a Kolmogorov-Arnold Network (KAN) that re places standard feed-forward networks with learnable basis functions, effectively capturing intricate nonlinear imaging characteristics. Extensive experiments on LOLv1, LOLv2-real, LOLv2-syn, and MIT-Adobe FiveK demon strate that RetinexKANMamba consistently outperforms existing methods in terms of brightness enhancement, noise suppression, and color fidelity.

Paddy rice cultivation can be implemented as single rice, double rice, and upland rice. Information on rice cropping patterns (CP) is crucial for enriching our understanding of sustainable intensification for a larger rice bowl. However, existing rice mapping efforts mainly focus on estimating planting area. It is challenging to produce annual rice CP datasets due to the difficulties of locating inundation signals from rice transplanting in complex-cropping regions. There is a lack of annual rice CP datasets in China in recent years, and there is an obvious omission error in existing rice algorithms in multiple-cropping regions. This study proposes a hierarchical Framework for the Automatic Mapping of complex Rice cropping patterns through the coupling of management practices and canopy structure. The challenge of the non-uniqueness of the "V" shaped feature in the Synthetic Aperture Radar (SAR) time series dataset was addressed by a phenology-assisted "V" shaped identification method using the Sentinel-1/2 datasets. The omission errors among multiple cropping regions were delivered using the hierarchy framework. This framework illustrated good performances among different rice cropping systems in China. This study developed the first 10-m annual rice Cropping Patterns (ChinaCP-Rice10m) maps over conterminous China from 2019 to 2024. An overall accuracy of over 90% was obtained when evaluated by ground-truth reference datasets. The upland rice cultivated areas have constantly increased, despite an overall declining trend of rice plantation areas since 2019. This study opens a new direction towards automatic complex rice cropping patterns by coupling management practices and canopy structure.