International Journal of Remote Sensing

Wen, Yanan; Wang, Tianyi; Liu, Meiling; Zhang, Hui; Zhao, Yuxin; Yang, Ben; Yan, Xutao; Wu, Ling

doi: 10.1080/01431161.2026.2680265pmid: N/A

Crops often experience heavy metal stress alongside other agricultural stresses. However, distinguishing heavy metal stress from other types of stress is difficult because of the similar canopy reflectance responses. The study area is located in the Zhuzhou section of the Xiangjiang River Basin, Hunan Province, China. A total of 39 GaoFen-6 Wide Field View (WFV) images with less than 20% cloud cover were acquired during the rice-growing period (1 April to 31 October) from 2019 to 2022. The GF-6 WFV sensor provides multispectral imagery with 16 m spatial resolution and red-edge spectral bands, which are suitable for constructing the red-edge chlorophyll index. To capture the persistent spectral response of rice to heavy metal stress, this study constructed a CIre time series during the rice growth period and decomposed it using complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) to isolate the relatively stable low-frequency fluctuation component associated with heavy metal stress. The extracted component was further analysed using emerging hot spot analysis (EHSA) to identify spatiotemporal stress patterns, from which a heavy metal stress index (HMSIEC-n) was developed to quantify rice stress intensity. Field-sampled and laboratory-measured soil Cd concentration was used to validate whether the detected rice stress pattern was consistent with the actual contamination level. The HMSIEC-n showed a significantly and positively correlated with measured soil Cd concentration (r = 0.844), and severe heavy metal stress was mainly concentrated along the Xiangjiang River and in the nearby northwestern and central parts of the study area. These findings indicate that the combination of signal decomposition and EHSA can effectively identify and characterize heavy metal stress in rice by capturing the spatiotemporal characteristics of the CIre response. This study will use denser field sampling, more complete time series, and optimized spatiotemporal analysis to improve model generalizability.

Kenzhebay, Meruyert; Khan, Mehak; Hanan, Abdul; Arghandeh, Reza

doi: 10.1080/01431161.2026.2665974pmid: N/A

Change detection using remote-sensing data has shown remarkable progress through deep learning approaches, particularly with convolutional neural networks (CNNs) and transformers, yet both architectures face limitations either in terms of computational efficiency or accuracy. In this paper, we introduce MCD-Mamba (Multimodal Change Detection Mamba), a novel framework that leverages state space models for efficient and scalable multimodal change detection using optical and Synthetic Aperture Radar (SAR) data. The proposed architecture is built around modality-specific Visual Mamba encoders with strategic weight sharing, enabling robust joint representation learning across heterogeneous data sources. To effectively capture cross-modal and temporal interactions, we introduce an efficient multi-scale late fusion strategy and a difference-based spatio-temporal (DST) block, which models bi-temporal changes while preserving the linear complexity of the Mamba architecture. Extensive experiments on the Multimodal ONERA Satellite Change Detection (OSCD) dataset validate the effectiveness of the proposed design, demonstrating competitive performance, while maintaining favourable computational efficiency. These results highlight the potential of state space modelling as a principled and efficient alternative for multimodal remote sensing change detection.

Shi, Run; Liu, Yankai; Liu, Song; Peng, Xiangbin; Song, Bowen; Gao, Jianwei; Bian, Fang; Zhong, Teng; Cui, Yuanzheng

doi: 10.1080/01431161.2026.2675719pmid: N/A

Commercial vitality represents the intensity and spatial presence of consumer-facing services that support everyday urban life. Although this dimension of urban development is economically important, it is difficult to observe at fine-granularity for long periods of time. At present, night-time light (NTL) data are widely used to examine urban vitality. However, the raw radiance reflects multiple sources of illumination including residence, infrastructure, and industry. We develop a remote-sensing framework that integrates annual VIIRS day/night band composites (2013–2024) with approximately 5.7 million commercial point-of-interest (POI) records to derive a POI-constrained commercial vitality index for 334 first-level subnational units in nine Southeast Asian countries. Commercial radiance is aggregated within a 5 km POI buffer and validated against service-sector GDP, yielding coefficients of determination of 0.80 in 2020. To characterize temporal development pathways, country-specific Markov transition matrices are estimated from within-country commercial vitality tiers. The results indicate strong persistence in commercial hierarchies: diagonal transition probabilities generally exceed 0.80, while expected upgrading from middle to top tiers typically requires 11–33 years. Transition indices vary markedly across countries (0.08–0.38), with Cambodia and Indonesia showing higher mobility and Singapore exhibiting near-static hierarchy. A regression model with country-fixed effects, with R2 = 0.849, further shows that volatility is negatively associated with mean mid-tier upgrading time (β =  −0.070, p = 0.002) and positively associated with absolute annual growth (β = 4.864, p = 0.002). The results show that combining VIIRS NTL with commercial POI constraints improves the functional specificity of night-time-light monitoring and provides an approach for monitoring service-related urban dynamics in data-scarce regions.

Sunil, Sneha; Padmakumari, B.; Madan Mohan Rao, K

doi: 10.1080/01431161.2026.2679163pmid: N/A

The study of cloudiness is fundamental to improving meteorological understanding, enhancing climate resilience, and predicting weather and climate extremes. This study investigates the spatial, temporal, and diurnal variability of cloud fraction (CF) over the Indian region using INdian SATellite (INSAT-3D) data for 2017–2018 and examines its relationship with aerosol optical depth (AOD). The INSAT-3D cloud mask is validated using ground-based infrared sky imagery and depicts a high hit rate (HR ≈ 86%), with improved performance during night-time and monsoon seasons, demonstrating reliable detection of clear and cloudy conditions when high-confidence cloud flags are used. Regional analysis reveals a predominantly monsoon-driven CF cycle with clear continental–marine contrasts. During JJAS, high clouds dominate the oceanic and southern regions, whereas low and mid-level clouds are more prevalent over continental interiors. AOD exhibits distinct seasonal patterns, with elevated values over inland and eastern regions during the pre-monsoon and monsoon periods, driven by dust transport, anthropogenic emissions, and wet scavenging. Spectral analysis of CF and AOD indicates dominant periodicities of 5–20 days, with enhanced power during the monsoon, suggesting the influence of monsoon intraseasonal oscillations, while a weaker 6–8-day signal suggests a possible weekly anthropogenic cycle. The diurnal cycle shows an afternoon CF peak over land driven by thermodynamic convection, while marine regions exhibit morning peaks associated with stratiform cloud formation. The CF–AOD relationship displays significant spatial heterogeneity, with both positive (Twomey) and negative (anti-Twomey) correlations. Overall, the results highlight strong regional variability in aerosol–cloud interactions and provide new observational evidence linking CF variability with AOD and monsoon intraseasonal dynamics, emphasizing the need for sustained multi-platform observations to improve aerosol–cloud representation over the Indian subcontinent.

Tsunoda, Yuuki; Takeuchi, Fumio

doi: 10.1080/01431161.2026.2676899pmid: N/A

Estimating understory light environments using simple yet spatially extensive approaches is essential for advancing forest ecosystem research. Moreover, predicting shifts in light regime following tree loss due to thinning or windthrow is critical for elucidating forest dynamics and developing effective management strategies. Consequently, a methodological framework that is (i) simple, (ii) capable of broad and spatially explicit estimation, and (iii) able to quantitatively predict the effects of canopy opening is required; however, previous studies have not achieved these goals simultaneously. Here, we introduce the neighbourhood light index (NLI), a novel metric derived from 2-m grid laser penetration indices (LPI) obtained via UAV-LiDAR, which integrates the effects of distance, direction, and solar altitude. Understory light estimation based on cumulative NLI values accurately reproduced relative light intensity (RMSE = 6.19%, MAE = 4.57%) and effectively quantified the influence of LPI spatial configuration and solar altitude on understory light environments. Additionally, the proposed method has the potential to support the virtual exploration of canopy-opening scenarios, thereby providing a basis for anticipating how changes in stand structure may influence understory light conditions. Although the present study does not include explicit simulations of post-treatment or disturbance scenarios, the framework offers a simple, spatially comprehensive approach for estimating understory light environments. It may serve as a foundation for future predictive applications. This approach provides information that can support sustainable, adaptive, and multipurpose forest management.

Sun, Lei; Wang, Hui; Lu, Zhizhong; Wen, Baotian; Lin, Lepeng

doi: 10.1080/01431161.2026.2666913pmid: N/A

To accurately obtain rainfall intensity information during ship navigation and maritime operations, an unsupervised self-organizing map (SOM) method for retrieving rainfall intensity levels from X-band marine radar images is proposed in this paper. Since the spatial distribution of rainfall can be represented by the distribution of wavenumbers in the frequency domain, a new method for extracting image texture features is presented. The distribution characteristics of wavenumbers in the wave-number energy spectrum are extracted and combined with local histogram feature techniques to obtain multi-dimensional feature vectors. These vectors are fed into an unsupervised 20 × 20 neuron SOM to complete the classification of rainfall intensity levels. The method is validated using data collected from the ocean observation station of Haitan Island in Pingtan County. The results show that the retrieval accuracy of the proposed method is 3.3% higher than that of the Genetic algorithm-back propagation neural network (GA-BPNN) method and 16.7% higher than that of the ratio of zero intensity to echo (RZE) method. Meanwhile, the experiments show that the proposed method has better reliability and more stable detection performance.

Yao, ZiLin; Li, Mei; Xiong, Biao; Tan, XiaoKang; Bi, JingWen; Shi, LingChen

doi: 10.1080/01431161.2026.2676258pmid: N/A

Studying the spatiotemporal evolution patterns of carbon storage and their driving factors based on high-precision predictions of land-use changes is of significant strategic importance for maintaining ecological security in the Three Gorges Reservoir Area and promoting the region’s green and low-carbon transition. This paper utilizes a new method combining the Random Forest (RF) algorithm, the Future Land Use Simulation (FLUS) model, and the Integrated Valuation of Ecosystem Services and Trade-offs (InVEST) model (RF-FLUS-InVEST) to simulate and predict future land-use changes and their carbon storage effects. It also uses the optimal parameter geographical detector to further study the driving mechanisms of carbon storage changes. The results show: (1) The total carbon storage in areas above 800 m is more than 2.2 times that in low-elevation regions; (2) The Kappa consistency coefficient of the FLUS model improved by 1.06% and the overall classification accuracy increased by 0.19% after the RF algorithm improvement compared to the original FLUS model; (3) the interaction between Digital Elevation Model (DEM) and Normalized Difference Vegetation Index (NDVI) had the most significant impact (0.5978) on carbon storage in the average values of 2012 and 2022, while the interaction between slope aspect and distance to water bodies had the weakest impact. Both were greater than single-factor effects, indicating that carbon storage changes in the Three Gorges Reservoir Area are driven by multiple socio-economic and natural factors; (4) The conversion of cropland to forest from 2012 to 2022 increased carbon storage by 16.5250 Tg, highlighting the success of ecological policies like afforestation and soil erosion control. These results provide a scientific basis for ecosystem carbon management in the Three Gorges Reservoir Area.

Wang, Xuqiao; Du, Lan; Zhou, Peiyuan; Liu, Zejun; Zhang, Zhongkai; Deng, Ken; Bao, Lin

doi: 10.1080/01431161.2026.2672065pmid: N/A

Global Navigation Satellite System Reflectometry (GNSS-R) has been successfully applied in fields including ground target detection and imaging. However, the low Power Flux Density (PFD) of signals reflected from the Earth’s surface has long been a critical bottleneck restricting the performance improvement of target detection. The rapid development and large-scale deployment of Low Earth Orbit (LEO) navigation satellites have furnished external illuminators with superior operational performance. The optimization of the quantity and spatial distribution of these transmitters can effectively enhance the accuracy of target state estimation. This paper first analyzes the Signal-to-Noise Ratio (SNR) improvement effect of the LEO-R system from three key dimensions: the orbital altitude of LEO navigation satellites, detection range, coherent time and incoherent time. Furthermore, a moving target state estimation algorithm specifically designed for LEO-R scenarios is proposed. This algorithm extracts the bistatic range and target radial velocity of a single satellite via the Radon-Fourier Transform (RFT), establishes a path difference-based observation model, and adopts the Levenberg-Marquardt (L-M) algorithm to solve the nonlinear least squares problem for precise target position estimation. Subsequently, target velocity is resolved by integrating the position estimation results with a direct physical model. Both simulation and field experimental results validate that the comprehensive performance of the proposed algorithm is significantly superior to that of the conventional Time Difference of Arrival (TDOA) algorithm.

Teo, Tee-Ann; Lin, Yu-Liang

doi: 10.1080/01431161.2026.2667023pmid: N/A

Cloud cover presents a significant challenge for optical image classification, particularly in the analysis of multi-temporal satellite imagery, as it obscures surface features and affects data continuity. This study investigated cloud-filling methods and U-Net-based architectures for time-series Sentinel-2 satellite images in land use and land cover (LULC) classification. Four cloud-filling techniques were evaluated – image mosaicking, null value retention, multiple imputation by chained equations, and principal component analysis (PCA)-enhanced imputation. Among these, PCA-enhanced imputation exhibited the best performance, achieving an overall accuracy of 84% and a macro F1-score of 74.87% by effectively preserving the temporal features and mitigating cloud impact. Four U-Net-based architectures with temporal features—3D U-Net, U-Net with a temporal attention encoder, U-Net convolutional long short-term memory, and U-Net bidirectional convolutional long short-term memory – were compared using the PCA-enhanced imputed dataset. The 3D U-Net outperformed the others due to its ability to capture spatial and temporal continuity. Other models showed limitations in effectively leveraging temporal dependencies. The results demonstrate the effectiveness of multi-temporal imagery and the benefits of cloud-filling methods in enhancing classification accuracy, especially for complex LULC categories such as changed areas. These findings highlight the value of robust preprocessing techniques and temporal modelling for detailed and accurate LULC mapping, offering practical solutions for remote sensing applications.

Jia, Dunxin; Chen, Zhihong; Fei, Teng; Xie, Hui; Guo, Ziyue; Bian, Meng

doi: 10.1080/01431161.2026.2672063pmid: N/A

High-quality acoustic management constitutes a core element in promoting the long-term well-being and growth of urban settings. Evaluating the sound environment is a critical step in assessing urban liveability. Existing evaluation approaches primarily rely on objective measurements such as sound pressure level (SPL) and the creation of urban noise maps. However, SPL-based metrics do not fully capture residents’ subjective responses to acoustic environments. To address these limitations, this study proposes a novel framework that integrates subjective perception data with remote sensing for urban sound environment. Specifically, residents’ perceived annoyance towards environmental noise is defined as ‘sound environment annoyance (SEA)’ following established soundscape frameworks, and used as a proxy for subjective sound perception. SEA is then estimated using machine learning and multi-scale, multispectral remote sensing imagery. A questionnaire survey was conducted in the urban core to collect residents’ SEA (n = 1190) using a five-point Likert scale from very comfortable (1) to very uncomfortable (5). Trained investigators guided respondents to ensure standardized completion. The survey and Sentinel-2 imagery acquisition were both conducted in 2024 to ensure temporal consistency between perceived SEA and environmental observations. Based on Sentinel-2 imagery, we extracted remote-sensing features capturing multiple spatial scales, spectral bands, and neighbourhood contexts. An XGBoost model was trained to link these features with SEA, and model predictions were evaluated using standard performance metrics. The model achieved an MAE of 0.32, an RMSE of 0.48, and an R2 of 0.65 on the test set, demonstrating strong predictive performance. Furthermore, we applied the SHAP method to interpret the model and identify key remote sensing features that contribute to the predictions. The proposed framework enables rapid, low-cost, and spatially continuous estimation of urban sound environment perception. This work contributes to cross-modal research in environmental science and offers a novel pathway for understanding urban sound environment.