Search

The generation and enhancement of satellite imagery are critical in remote sensing, requiring high-quality, detailed images for accurate analysis. This research introduces a two-stage diffusion model methodology for synthesizing high-resolution satellite images from textual prompts. The pipeline comprises a low-resolution diffusion model (LRDM) that generates initial images based on text inputs and a super-resolution diffusion model (SRDM) that refines these images into high-resolution outputs. The LRDM merges text and image embeddings within a shared latent space, capturing essential scene content and structure. The SRDM then enhances these images, focusing on spatial features and visual clarity. Experiments conducted using the Remote Sensing Image Captioning Dataset demonstrate that our method outperforms existing models, producing satellite images with accurate geographical details and improved spatial resolution. [ABSTRACT FROM AUTHOR]

Automatic classification of remote sensing images using machine learning techniques is challenging due to the complex features of the images. The images are characterized by features such as multi-resolution, heterogeneous appearance and multi-spectral channels. Deep learning methods have achieved promising results in the analysis of remote sensing satellite images in the recent past. However, deep learning methods based on convolutional neural networks (CNN) experience difficulties in the analysis of intrinsic objects from satellite images. These techniques have not achieved optimum performance in the analysis of remote sensing satellite images due to their complex features, such as coarse resolution, cloud masking, varied sizes of embedded objects and appearance. The receptive fields in convolutional operations are not able to establish long-range dependencies and lack global contextual connectivity for effective feature extraction. To address this problem, we propose an improved deep learning-based vision transformer model for the efficient analysis of remote sensing images. The proposed model incorporates a multi-head local self-attention mechanism with patch shifting procedure to provide both local and global context for effective extraction of multi-scale and multi-resolution spatial features of remote sensing images. The proposed model is also enhanced by fine-tuning the hyper-parameters by introducing dropout modules and a decay linear learning rate scheduler. This approach leverages local self-attention for learning and extraction of the complex features in satellite images. Four distinct remote sensing image datasets, namely RSSCN, EuroSat, UC Merced (UCM) and SIRI-WHU, were subjected to experiments and analysis. The results show some improvement in the proposed vision transformer on the CNN-based methods. [ABSTRACT FROM AUTHOR]

Hyperspectral remote sensing images obtained from cameras are characterized by high-dimensions and low quality, which makes them unfavorable for various analytics purposes. This is due to the presence of visible and invisible frequencies of the reflected light making it poorly reveal the spectral signatures of the image. Visual communication advancement has paved the need for Image Super-Resolution (SR) which recovers high-resolution images from low-resolution images. Several works were carried out earlier on image SR using variants of supervised and unsupervised models that still lack accuracy. In this paper, we propose an unsupervised learning model titled Gompertz Function–based Convergence War Accelerometric Optimization–GAN framework for generating of High-Resolution (HR) images. The framework comprises a pre-processing stage, where the incoming Low-Resolution (LR) image is preprocessed for noise removal by applying Shannon-Gaussian Filter (S-GF). Following is the Gradient Domain Approach based Tone-Mapping (TM). Skew correction is done to remove distortion and maintain original resolution that may change during TM stage. The next stage comprises the boundary and edge enhancement of the resulting preprocessed image generated by the method of Inverse Gradient Mapping (IGM) followed by patch extraction to extract minute low-frequency information from the resulting boundary and edge-enhanced image. The contrast of the enhanced patches is improved by removing blurriness effect. The preprocessed image patches are then fed into the Gompertz Function-based Convergence War Accelerometric Optimization – GAN for feature mapping on the trained SR Image features that are clustered using Krzanowski and Li- Kantorovich Metric-K-Means clustering Algorithm (KL-KM-KMA) for effective generation of SR image. The developed model is validated for both qualitative and quantitative measurements. Comparisons are made with several other state-of -the-art methods for accuracy of 98.05%, precision of 97.98%, inception score of 8.71, Fréchet Inception Distance of 36.4 with reduced clustering and training time proving the efficiency of the proposed model. [ABSTRACT FROM AUTHOR]

The Awash River Basin is essential in combating climate change by absorbing atmospheric carbon. The goal of this study was to determine carbon stocks using the linear correlation between field inventory and remote sensing data in the basin. The regression model was created using each plot studied from field inventory and the corresponding vegetation indices (VI) values obtained from ArcGIS software in the basin's study area. The linear correlation values for estimated carbon stock with Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) were R2 = 0.36, R = 0.60, and R2 = 0.52, R = 0.72, respectively. In contrast, the values for the Soil-Adjusted Vegetation Index (SAVI) were R2 = 0.36 and R = 0.60. The results show that the relationship between carbon stock and EVI is stronger for estimating carbon stock than other indices (NDVI and SAVI). The linear regression model or equation developed for NDVI, EVI, SAVI, and field-level carbon stock was Y = 16.25x–1.093, Y = 8.935x–1.1254, and Y = 12.988x–1.0895, respectively. The estimated carbon stock from the EVI linear regression model was 15,904,158.24 tons, with an average carbon stock value of 111.96 tons per hectare. The findings concluded that EVI has a stronger correlation with carbon stock stand estimation than NDVI and SAVI. This study could serve as a baseline for the estimation of carbon stock using vegetation indices in different and large-scale land use land cover types. High-resolution multispectral imagery and cloud-based geospatial analysis platforms could provide more accurate results. [ABSTRACT FROM AUTHOR]

Many freshwater forested wetlands along the southeastern United States coastline are rapidly transitioning from forest to marsh or open water, due to climate change-related disturbances. Recent studies have found early warning signals (EWS) of regime shifts in other ecosystems, but it is unclear if these can be detected for coastal wetlands. In this study, we examined the ability to detect EWS of regime shifts in coastal wetlands within the Albemarle Pamlico peninsula (APP), North Carolina, U.S.A. We used the Landsat record (1985–2021) to examine trends of normalized difference vegetation index (NDVI) time series for selected areas known to have undergone regime shifts. We found that while 77% of the APP was either stable or revegetating, 22% of the landscape underwent a decrease in NDVI that would indicate a transition from forest to marsh or open water. Of the areas that transitioned, about half (11%) experienced an abrupt decrease in NDVI and 10% experienced a gradual decline. Increasing standard deviation and skewness of time series could serve as EWS of abrupt transitions, but can also provide false negative and positives. Our results suggest that ecosystem transitions from a forest to a marsh or open water can occur both rapidly and slowly, and remote sensing of NDVI time series can help identify EWS for some areas, but not all. Our results allow for prioritization of conservation/restoration of coastlines which will become important in the face of climate change and sea level rise. [ABSTRACT FROM AUTHOR]

This study investigates the impacts of Land Use/Land Cover (LULC) changes and climate change on surface runoff in Gdańsk, Poland, which is crucial for local LULC planning and urban flood risk management. The analysis employs two primary methodologies: remote sensing and hydrological modeling. Remote sensing was conducted using Google Earth Engine and Land Change Modeler in IDRISI Terrset software to analyze historical (1985–2022) and future (2050–2100) LULC. Hydrological modeling was performed using the Natural Resources Conservation Service curve number method to assess the overall impact of LULC changes on Gdańsk's hydrology at the local scale. The Orunia basin, a critical area due to intensive LULC development, was selected for detailed hydrological analysis using the Hydrologic Modeling System (HEC-HMS). The analysis encompassed three scenarios: LULC changes, climate change, and combined LULC and climate change effects. The LULC analysis revealed a marked increase in urban area, a shift in forest and vegetation cover, and a reduction in agricultural land. HEC-HMS simulations showed an increase in the runoff coefficient across selected decades, which was attributed to the combined effect of LULC and climate change. The projected increases under the Representative Concentration Pathway (RCP) 4.5 and RCP 8.5 scenarios for 2050 and 2100 are projected to surpass those observed during the baseline period. The findings highlight that the synergistic effects of LULC and climate change have a more significant impact on Gdańsk's hydrology at both local and basin scales than their separate effects. These insights into LULC shifts and urban hydrological responses hold implications for sustainable urban planning and effective flood risk management in Gdańsk and similar urban settings. [ABSTRACT FROM AUTHOR]

Uncrewed Aerial Systems (UAS) offer a versatile solution for monitoring forest ecosystems. This study aimed to develop and assess an individual tree-based methodology using multi-temporal, multispectral UAS images to track changes caused by the European spruce bark beetle (Ips typographus L.). The approach encompassed four key steps: (1) individual tree detection using structure-from-motion point clouds, (2) tree species classification, (3) health classification of spruce trees as healthy, declined, or dead, and (4) change detection, identifying fallen/removed trees and alterations in tree health status. The developed methodology was employed to quantify changes in a bark beetle outbreak area covering 215 hectares in southeastern Finland during 2019–2021. The dataset included two managed and two conserved forest areas. The uncertainty estimation demonstrated the overall accuracies ranging from 0.58 to 0.91 for individual tree detection, 0.84 for species classification, and 0.83–0.96 for health classification, and a F1-score of 0.91 for the fallen or removed tree detection. Maps and statistics were produced, containing information on the health of the spruce trees in the area and information on changes, including trees that died during monitoring and those that fell or were removed from the forest. The results demonstrated successful control of the outbreak in the managed stands, evidenced by moderate tree mortality. Conversely, in the conserved stands, the outbreak resulted in dramatic tree mortality. This method serves stakeholders by enabling large-scale outbreak impact monitoring, facilitating timely risk assessment, and validating bark beetle outbreak management strategies. [ABSTRACT FROM AUTHOR]

Remote sensing techniques have emerged as highly effective and widely encouraged tools for mineral exploration and lithological mapping. Given the association of various types of metallic minerals with collisional belts, the northern region of Pakistan presents a unique opportunity to explore metallic mineralization related to the orogenic events of the Himalaya-Karakoram-Hindukush ranges, particularly in the Chitral region. However, due to the high relief and rugged area topography, traditional detailed surveys are challenging. The Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) remote sensing data provide a better alternative for mapping the lithological units and mineralization zones. Several remote sensing methods, including False Color Composite (FCC), Decorrelation Stretch (DS), Principal Component Analysis (PCA), Supervised Classification (Train Random Trees Classification (TRTC)), and Matched Filtering (MF), were used to process the remote sensing data. The detected results revealed the presence of chlorite, limonite, sericite (muscovite), kaolinite, malachite, and calcite and effectively discriminated the lithological units. The spatial distribution of anomalies for muscovite, kaolinite, and limonite correlates with the presence of granitic plutons and carbonate rocks exposed in the region. Additionally, this study identified five new mineralization targets associated with alteration zones, particularly correlated to copper and iron mineralization. The new mineralization zones are well distributed in the valleys of Shishi, Maroi, and Drunil. These findings were confirmed through comprehensive field campaigns, petrographic investigations, X-ray Diffraction (XRD), and spectral measurements. [ABSTRACT FROM AUTHOR]

Soil salinity is one of the significant environmental issues that can reduce crop growth and productivity, ultimately leading to land degradation. Therefore, accurate monitoring and mapping of soil salinity are essential for implementing effective measures to combat increasing salinity. This study aims to estimate the spatial distribution of soil salinity using machine learning methods in Huludao City, located in northeastern China. By meticulously collecting data, soil salinity was measured in 310 soil samples. Subsequently, environmental parameters were calculated using remote sensing data. In the next step, soil salinity was modeled using machine learning methods, including random forest (RF), support vector machine (SVM), and artificial neural network (ANN). Additionally, to estimate uncertainty, the lower limit (5%) and upper limit (95%) prediction intervals were used. The results indicated that accurate maps for predicting soil salinity could be obtained using machine learning methods. By comparing the methods employed, it was determined that the RF model is the most accurate approach for estimating soil salinity (RMSE=0.03, AIC=-919, BIS=-891, and R2=0.84). Furthermore, the results from the prediction interval coverage probability (PICP) index, utilizing the uncertainty maps, demonstrated the high predictive accuracy of the methods employed in this study. Moreover, it was revealed that the environmental parameters, including NDVI, GNDVI, standh, and BI, are the main controllers of the spatial patterns of soil salinity in the study area. However, there remains a need to explore more precise methods for estimating soil salinity and identifying salinity patterns, as soil salinity has intensified with increased human activities, necessitating more detailed investigations. [ABSTRACT FROM AUTHOR]

Hyperspectral Images (HSIs) possess extensive applications in remote sensing, especially material discrimination and earth observation monitoring. However, constraints in spatial resolution increase sensitivity to spectral noise, limiting the ability to adjust Receptive Fields (RFs). Convolutional Neural Networks (CNNs) with fixed RFs are a common choice for HSI classification tasks. However, their potential in leveraging the appropriate RF remains under-exploited, thus affecting feature discriminative capabilities. This study introduces an Enhanced Adaptive Source-Selection Kernel with Attention Mechanism (EAS 2 KAM) for HSI Classification. The model incorporates a Three Dimensional Enhanced Function Mixture (3D-EFM) with a distinct RF for local low-rank contextual exploitation. Furthermore, it incorporates diverse global RF branches enriched with spectral attention and an additional spectral-spatial mixing branch to adjust RFs, enhancing multiscale feature discrimination. The 3D-EFM is integrated with a 3D Residual Network (3D ResNet) that includes a Channel-Pixel Attention Module (CPAM) in each segment, improving spectral-spatial feature utilization. Comprehensive experiments on four benchmark datasets show marked advancements, including a maximum rise of 0.67% in Overall Accuracy (OA), 0.87% in Average Accuracy (AA), and 1.33% in the Kappa Coefficient (κ ), outperforming the top two HSI classifiers from a list of eleven state-of-the-art deep learning models. A detailed ablation study evaluates model complexity and runtime, confirming the superior performance of the proposed model. [ABSTRACT FROM AUTHOR]

Hyperspectral Imaging (HSI) has revolutionized earth observation through advanced remote sensing technology, providing rich spectral and spatial information across multiple bands. However, this wealth of data introduces significant challenges for classification, including high spectral correlation, the curse of dimensionality due to limited labeled data, the need to model long-term dependencies, and the impact of sample input on deep learning performance. These challenges are further exacerbated by the costly and complex acquisition of HSI data, resulting in limited availability of labeled samples and class imbalances. To address these critical issues, our study proposes a novel approach for generating high-quality synthetic hyperspectral data cubes using an advanced Generative Adversarial Network (GAN) integrated with the Wasserstein loss and gradient penalty phenomenon (WGAN-GP). This approach aims to augment real-world data, mitigating the scarcity of labeled samples that has long been a bottleneck in hyperspectral image analysis and classification. To fully leverage both the synthetic and real data, we introduce a novel Convolutional LSTM classifier designed to process the intricate spatial and spectral correlations inherent in hyperspectral data. This classifier excels in modeling multi-dimensional relationships within the data, effectively capturing long-term dependencies and improving feature extraction and classification accuracy. The performance of our proposed model, termed 3D-ACWGAN-ConvLSTM, is rigorously validated using benchmark hyperspectral datasets, demonstrating its effectiveness in augmenting real-world data and enhancing classification performance. This research contributes to addressing the critical need for robust data augmentation techniques in hyperspectral imaging, potentially opening new avenues for applications in areas constrained by limited data availability and complex spectral-spatial relationships. [ABSTRACT FROM AUTHOR]

Purpose: Site-specific field management operations represent one of the fundamental principles of precision viticulture. The purpose of the research is to observe and analyse the evolution of a vineyard over three consecutive years to understand which factors most significantly influence the quality of the vineyard's production. Methods: The research involved technologically advanced tools for crop monitoring, such as remote and proximal sensors for vegetation surveys. In association, grape quality analyses were performed through laboratory analysis, constructing geostatistical interpolation maps and matrix correlation tables. Results: Both remote and proximal sensing instruments demonstrated their ability to effectively estimate the spatial distribution of vegetative and quality characteristics within the vineyard. Information obtained from GNDVI and CHM proved to be valuable and high-performance tools for assessing field variability. The differentiated plant management resulted in uniform production quality characteristics, a change evident through the monitoring techniques. Conclusion: The research highlights the effectiveness of using advanced technological instruments for crop monitoring and their importance in achieving uniformity in production quality characteristics through differentiated plant management. From the results obtained, it was possible to observe how differentiated plant management led to a uniformity of production quality characteristics and how the monitoring techniques can observe their evolution. This result represents a positive accomplishment in field management during the three monitoring years, responding to the principles and objectives of precision agriculture. [ABSTRACT FROM AUTHOR]

Yield forecasting and within-field yield variation is essential information that helps farmers develop sustainable agriculture. However, such information still needs to be included for most of them, and remote sensing is an alternative to provide it. Our objective was to assess Random Forest regression models composed of unique GLCM texture measures as an alternative to usual empirical models that use spectral response and auxiliary data, which is complex and reaches varied results. Eleven GLCM texture models based on eight texture measures of a single spectral layer were assessed to represent soybean field yield variation in two sites and seasons. Several models achieved satisfactory results, reaching R2 from 0.90 to 0.95 and RMSE from 0.06 to 0.26 t/ha. Models above 15-window size are recommended for the soybean yield prediction as window size is an essential attribute to GLCM performance. Models derived from the bands individually (red, red-edge, near-infrared, and short wavelength infrared) were more sensitive to the window size than those derived from vegetation indices (EVI, GNDVI, GRNDVI, NDMI, NDRE, NDVI, SFDVI). The data aggregated by texture measures improve the individual spectral responses, providing alternatives to predict soybean within-field yield variation using random forest models. [ABSTRACT FROM AUTHOR]

Purpose: A generalised approach to downscale areal observations of crop production data is illustrated using cotton yield and fibre quality (length and micronaire) data which is measured as a module (areal/block) average. Methods: Two features of the downscaling algorithm are; (i) to estimate spatial trends in yield and quality using regression with fine resolution predictors such as remote sensing imagery, and (ii) use area-to-point kriging (A2PK) to downscale either the observations in the absence of a useful spatial trend model or the residuals from the trend model (if useful) from areal averages. Results: Correlations with remote sensing covariates were stronger for cotton fibre yield than for cotton fibre micronaire, and much stronger compared to those for cotton fibre length. Spatial trends in cotton fibre yield and micronaire could be estimated with good model quality using regression with remote sensing covariates with or without A2PK in almost all fields. Conversely, model quality was poorer for cotton fibre length and there was only a small difference in model performance between the null and trend models. When the downscaling approach was tested using fine-resolution yield observations, model performance was poorer at a fine-resolution compared to the module-resolution, which was to be expected. Conclusion: This approach enables the creation of high-resolution raster maps of variables of interest with a much finer spatial resolution compared to the areal observations, and can be applied for any areal averaged crop production data in a range of broadacre and horticultural industries (e.g. sugarcane, apples, citrus). The finer spatial resolution may allow growers or agronomists to better understand the drivers of variability within fields, assess management implications, and create management plans at a higher resolution. [ABSTRACT FROM AUTHOR]

Purpose: This research aims to develop a meta-machine learning model to optimize soil and nitrogen management for durum wheat in Italy. It addresses the challenges of increased food production on limited land amidst rising input costs, geopolitical changes, and climate change. The goal is to aid decision-makers in achieving maximum crop yield and income margins through effective agronomic strategies. Methods: The study developed a meta-machine learning model, integrating classification and regression models, and tested it at four sites in Marche and Basilicata, Italy, over several years. The model incorporated data from remote sensing, crop phenology, soil chemical properties, weather data, soil management, and nitrogen levels. A Random Forest model was used to classify crop phenology, while a Neural Network model predicted yield. Eleven nitrogen levels were compared across these sites. Results: The Random Forest model achieved an accuracy of 0.98, kappa of 0.96, and recall of 0.98 for predicting crop phenology. The Neural Network model for yield prediction had an R squared of 0.90 and a Root Mean Square Error of 0.59 t ha-1. Key factors identified for model accuracy were temperature, precipitation, NDVI, and nitrogen input. Simulations of 30 soil management and fertilization combinations revealed that no-tillage management increased grain yield. The Marginal Fertilizer Yield Index determined optimal nitrogen application. Conclusions: The meta-machine learning model accurately predicted durum wheat yield and identified effective agronomic strategies, demonstrating the potential for broader application in field conditions. The model offers a promising approach to sustainable agriculture and climate change mitigation by utilising publicly available spatial datasets. [ABSTRACT FROM AUTHOR]

This study describes a new method for satellite-based remote sensing analysis of plant-specific biomass yield patterns for precision farming applications. The relative biomass potential (rel. BMP) serves as an indicator for multiyear stable and homogeneous yield zones. The rel. BMP is derived from satellite data corresponding to specific growth stages and the normalized difference vegetation index (NDVI) to analyze crop-specific yield patterns. The development of this methodology is based on data from arable fields of two research farms; the validation was conducted on arable fields of commercial farms in southern Germany. Close relationships (up to r > 0.9) were found between the rel. BMP of different crop types and study years, indicating stable yield patterns in arable fields. The relative BMP showed moderate correlations (up to r = 0.64) with the yields determined by the combine harvester, strong correlations with the vegetation index red edge inflection point (REIP) (up to r = 0.88, determined by a tractor-mounted sensor system) and moderate correlations with the yield determined by biomass sampling (up to r = 0.57). The study investigated the relationship between the rel. BMP and key soil parameters. There was a consistently strong correlation between multiyear rel. BMP and soil organic carbon (SOC) and total nitrogen (TN) contents (r = 0.62 to 0.73), demonstrating that the methodology effectively reflects the impact of these key soil properties on crop yield. The approach is well suited for deriving yield zones, with extensive application potential in agriculture. [ABSTRACT FROM AUTHOR]

Soil fertility is one of the most critical bases for high productivity and sustainability in crop production. Within-field heterogeneity is often problematic in both crop management practices and crop productivity. Besides, appropriate soil management practices leads to the effective carbon sequestration. Since the soil carbon content (SCC) is the most simple and effective indicator of soil fertility, accurate and high-resolution mapping of SCC is an essential basis for addressing these issues. Here, we developed a tractor-based hyperspectral sensing system for speedy and accurate mapping of SCC. A new hybrid spectral algorithm linking normalized difference spectral index (h-NDSI) and machine learning proved superior. Appropriate algorithms were implemented to generate diagnostic map and prescription map from SCC map for the variable-rate application of pellet manure. The field performance of the sensing/mapping system was tested in the farmers' fields in the Fukushima region of Japan where the within-field heterogeneity of soil fertility was disastrous due to the decontamination after the nuclear power-plant disaster. The structure and functioning of the system proved promising. Moreover, the spatial simulation by linking the SCC data and a dynamic simulation model clearly showed the significant impact of variable-rate application of pellet manure on the chronosequential change of SCC, within-field heterogeneity, and carbon stock. The systematic linkage of the sensing/mapping system with the variable-rate spreader and dynamic simulation model would be effective for improving soil fertility and soil carbon stock. Applicability of the system will be extended through an extensive validation of the predictive models. [ABSTRACT FROM AUTHOR]

Bitemporal supervised learning paradigm always dominates remote sensing change detection using numerous labeled bitemporal image pairs, especially for high spatial resolution (HSR) remote sensing imagery. However, it is very expensive and labor-intensive to label change regions in large-scale bitemporal HSR remote sensing image pairs. In this paper, we propose single-temporal supervised learning (STAR) for universal remote sensing change detection from a new perspective of exploiting changes between unpaired images as supervisory signals. STAR enables us to train a high-accuracy change detector only using unpaired labeled images and can generalize to real-world bitemporal image pairs. To demonstrate the flexibility and scalability of STAR, we design a simple yet unified change detector, termed ChangeStar2, capable of addressing binary change detection, object change detection, and semantic change detection in one architecture. ChangeStar2 achieves state-of-the-art performances on eight public remote sensing change detection datasets, covering above two supervised settings, multiple change types, multiple scenarios. [ABSTRACT FROM AUTHOR]

Automatically detecting Ulva prolifera (U. prolifera) in rainy and cloudy weather using remote sensing imagery has been a long-standing problem. Here, we address this challenge by combining high-resolution Synthetic Aperture Radar (SAR) imagery with the machine learning, and detect the U. prolifera of the South Yellow Sea of China (SYS) in 2021. The findings indicate that the Random Forest model can accurately and robustly detect U. prolifera, even in the presence of complex ocean backgrounds and speckle noise. Visual inspection confirmed that the method successfully identified the majority of pixels containing U. prolifera without misidentifying noise pixels or seawater pixels as U. prolifera. Additionally, the method demonstrated consistent performance across different images, with an average Area Under Curve (AUC) of 0.930 (±0.028). The analysis yielded an overall accuracy of over 96%, with an average Kappa coefficient of 0.941 (±0.038). Compared to the traditional thresholding method, Random Forest model has a lower estimation error of 14.81%. Practical application indicates that this method can be used in the detection of unprecedented U. prolifera in 2021 to derive continuous spatiotemporal changes. This study provides a potential new method to detect U. prolifera and enhances our understanding of macroalgal outbreaks in the marine environment. [ABSTRACT FROM AUTHOR]

The occurrences of large rock slides often result in catastrophic debris flow within high mountain environments. Discontinuity intersected blocks meeting kinematic conditions stemming from deglaciation-related damage can be triggered by external factors, leading to massive rock slides with a significant downstream hazard. This study presents a comprehensive analysis underlining the mechanism and evolution of the failure during the 2017 Santa Lucía landslide, Patagonian Andes, Chile, utilizing remote sensing and numerical modelling. Due to the remote location, aerial photogrammetry was used to unravel the structural and geomorphological configuration, and four discontinuity sets were identified. Based on colour-shaded relief and slope kinematic analysis, it was found that the failure is governed by combinations of three different discontinuity sets. The failure in the crown portion is complex due to resulting planar and wedge surfaces, whereas in the toe region, the failure is governed by the wedge formation between bedding and other joint set. To further examine its mechanism and evolution, rigid block numerical models were developed in 3DEC to reproduce the failure with real topography and joint parameters. The maximum displacement was observed in the same topographical region where the actual failure occurred, thus conforming to the role of discontinuities in the evolution of the catastrophic failure. Acting on a reduced strength due to rock damage, the modelled slope boosts the instability leading to higher displacements along bonding surfaces with similar attributes as observed in the field. A detailed methodology is discussed regarding coupling remote sensing and 3D numerical modelling for detailed insights into the failure mechanism of the landslides. Overall, our results demonstrate that the Santa Lucía rock slide is a structurally controlled failure where joints provided kinematic freedom, favoured by long-term rock slope damage due to deglaciation. [ABSTRACT FROM AUTHOR]

For the last seven years, PlanetScope satellites have started near-daily imaging of parts of the Earth's surface, making high-density multitemporal, multispectral, 3-m pixel imagery accessible to researchers. Multitemporal satellite data enables landscape archaeologists to examine changes in environmental conditions at time scales ranging from daily to decadal. This kind of temporal resolution can accentuate landscape features on the ground by de-emphasizing non-permanent signatures caused by seasonal or even daily changes in vegetation. We argue that the availability of high spatial and temporal resolution multispectral imagery from Planet Inc. will enable new approaches to studying archaeological visibility in landscapes. While palimpsests are discrete overlapping layers of material accumulation, multitemporal composites capture cyclical and seasonal time and can be used to interpret past landscape histories at multiple scales. To illustrate this perspective, we present three case studies using PlanetScope imagery in tropical environments on the Indian Ocean islands of Madagascar, Mauritius, and Zanzibar. [ABSTRACT FROM AUTHOR]

Coastal green infrastructure provides numerous ecosystem services, including flood protection, erosion control, carbon sequestration, and habitat for marine life. Mapping and monitoring these critical coastal habitats are essential for sustainable management and conservation efforts. This study employed the Google Earth Engine (GEE) cloud platform with high-resolution multispectral satellite imagery (Sentinel-2 data with 10 m spatial resolution), Airborne Laser Terrain Mapper (ALTM) elevation data with 5 m resolution, and advanced machine learning (ML) algorithms used to map the distribution and extent of coastal green infrastructure along the Pondicherry coastline in southern India. A random forest (RF) classifier was trained on a diverse set of reference data (70% for training and 30% for validation) collected through extensive field surveys and visual interpretation of very high-resolution aerial imagery. The model integrated spectral information from multiple satellite sensors along with derived biophysical indices to accurately delineate different coastal vegetation types. The resulting maps revealed detailed spatial patterns of mangroves, sand dunes, and coastal plantations with an overall accuracy exceeding 90% verified with the field data. Analyses quantified their spatial coverage and fragmentation along the study area. This high-resolution, accurate baseline data can inform coastal management strategies, including targeted conservation efforts, ecological restoration projects, climate change adaptation planning, and sustainable development practices that preserve vital green infrastructure. The workflow demonstrated the robust capabilities of ML methods coupled with multi-source remote sensing data for effectively mapping complex and dynamic coastal ecosystems at a regional scale. The techniques can be adapted for other coastal regions to understand green infrastructure dynamics better and support evidence-based policies promoting ecological and community resilience. [ABSTRACT FROM AUTHOR]

The Madurai local planning authority, encompassing a land area of 726.34 sq. km, has encountered challenges of droughts and flash floods during the north–east monsoon season. These issues have arisen as a result of notable alterations in land use and the swift pace of urbanization. This comprehensive study aims to assess the effects of land use changes on direct runoff within the study area over a span of 40 years, from 1980 to 2020. This study attempted to understand the evolution of land use and land cover change in the Madurai LPA region over the past 2 decades and its corresponding impact on runoff generation. The study also predicted the trend of LULC in 2040. LANDSAT images from different years were acquired to create land use and land cover maps using ERDAS IMAGINE 9.1 and Arc GIS Version 10.1. Four hydrological soil groups were determined using data from the Madurai Atlas, and the surface runoff was calculated using the soil conservation service–curve number. The accuracy of the land use and land cover maps was evaluated using an error matrix and kappa index. LULC predictions for 2040 were made using the cellular automata and artificial neural network model. The analysis showed that agricultural land increased by 5.9% between 1980 and 2020, while forest cover decreased by 0.2% and urban settlements grew by 7.4% in the D hydrological soil group. The predicted land use for 2040 indicates that agricultural land will account for 54.1%, followed by 1% forest cover and 15.8% urban areas. The accuracy of the predicted land use map was validated using the 2020 map, with a 91% accuracy and a kappa coefficient of 0.8. The Madurai region has experienced a notable surge in urbanization, highlighting the urgency for effective flood management and the implementation of urban development strategies that prioritize the creation of green spaces and efficient storm water drainage systems. [ABSTRACT FROM AUTHOR]

In 2022 and 2023, the Defense POW/MIA Accounting Agency (DPAA) partnered with Marine Imaging Technologies to conduct remote sensing survey operations in Guam to search for and possibly confirm the locations of four separate B-29 Superfortress wreck sites. The reported locations for these sites are in challenging environments; high seas, fast currents, and strong winds pose hazardous conditions for survey teams and equipment. Additionally, the bathymetry of the areas consists of dynamic changes in depths with sudden valleys, rises, and seamounts, mirroring the rocky coastline which poses threats to successful, strategic surveying and quality results. Marine Imaging Technologies worked with Eastern Search and Survey to develop an innovative survey system that allowed flexibility and pliability for small boat, deeper water, and limited crew operations. This paper discusses the goals, challenges, and results of this project and looks at ways forward on similar missions partnering with DPAA to account for MIA service members. [ABSTRACT FROM AUTHOR]

Mapping floods is crucial for effective disaster management. This study focuses on flood assessment in northern Morocco, specifically Tangier, Tetouan, and Larache. Due to the lack of a comprehensive flood inventory map, we used unsupervised learning techniques, such as K-means clustering and fuzzy logic algorithms, to predict flood-prone areas. We identified nine conditioning factors influencing flood risk: elevation, slope, aspect, plan curvature, profile curvature, land use, soil type, normalized difference vegetation index (NDVI), and topographic position index (TPI). Using Landsat-8 imagery and a Digital Elevation Model (DEM) within a Geographic Information System (GIS), we analyzed topographic and geo-environmental variables. K-means clustering achieved silhouette scores of 0.66 in Tangier and 0.70 in Tetouan, while the fuzzy logic method in Larache produced a Davies-Bouldin Index (DBI) score of 0.35. The maps classified flood risk levels into low, moderate, and high categories. This research demonstrates the integration of machine learning and remote sensing for predicting flood-prone areas without existing flood inventory maps. Our findings highlight the main factors contributing to flash floods and assess their impact, enhancing the understanding of flood dynamics and improving flood management strategies in vulnerable regions. [ABSTRACT FROM AUTHOR]

Due to their coastal location, mangrove ecosystems are especially susceptible to various human-caused disturbances. These disturbances can result in a loss of original mangrove cover, a decline in biodiversity, and a degradation of their ecological functions. This study examines a restored mangrove area within the Barão de Mauá Natural Municipal Park (PNMBM) in Guanabara Bay, Brazil, two decades after the initiation of a restoration project. We used a multidisciplinary approach, combining remote sensing analyses, assessment of the vegetation structure, and genetic diversity analysis to evaluate this restored mangrove forest. Remote sensing data demonstrated high Normalized Difference Vegetation Index values following the restoration project's completion, indicating the revegetation's success. Exploring the structure of the new forest, we found Laguncularia racemosa to be the most abundant species, with a higher density and dominance for adults and seedlings. Conversely, Avicennia schaueriana was the least abundant in all areas, indicating a need for species enrichment. Planting age affected the forest structure, suggesting the capacity of this mangrove to achieve a natural maturity state. There was a genetic erosion in L. racemosa individuals, both adults and seedlings and low diversity was observed in A. schaueriana individuals. The PNMBM mangrove forest has shown remarkable resilience in maintaining its vegetation cover after restoration despite previous deforestation cycles. Moreover, as a young mangrove forest, it will continue to go through successional stages naturally. The loss of genetic diversity could be a concern for long-term survival, highlighting the need for genetic management. Integrating remote sensing, phytosociology, and genetic diversity analyses provided a comprehensive and detailed view of the park's restoration outcome. These results should be used as guidelines for future interventions outlined in this work and can help ensure the resilience and sustainability of any restored mangrove. [ABSTRACT FROM AUTHOR]

Coastal risk assessment requires accurate coastal bathymetry, which is largely lacking in many regions of the world's coastal oceans. This is particularly true in Guinea-Bissau, where the tidal range is the highest of any country in West Africa, affecting several socio-economic sectors such as agriculture and urban development. Here, we present a selection of coastal bathymetry datasets available for generally poorly sampled coastal regions. These include the General Bathymetric Chart of the Oceans (GEBCO), digital nautical charts (CMAP), and depth retrieval from satellite wave detection (S2Shores). While GEBCO provides the right regional envelope of bathymetry, the coastal zone is plagued with random bumpiness at scales below 10 km ("orange peel" texture). S2Shores can be used to correct these errors and instead reveal the presence of small-scale channels, reminiscent of very high-resolution satellite color images. These structures are also confirmed by nautical charts, which show the channels to be well correlated but deeper than that indicated by the satellite retrieval (with a 40% higher variance, and 40 m instead of 20 m for Orango, the main and deepest channel). Simulations with a coastal ocean hydro-sedimentary model (CROCO) further show that submerged rivers can help drive a smooth flow of tidal energy into the main estuary of Guinea-Bissau. Tidal energy flux is better channeled in the CMAP and S2shores cases, while GEBCO produces an order of magnitude greater form drag, resulting in a stronger response from the sediment bed. Tides therefore have the effect of smoothing bathymetry along fine-scale channels, minimizing friction on the way to equilibrium. This result highlights the potential of combining different approaches to estimate updated coastal bathymetry and its effect on coastal dynamics in the most challenging areas. [ABSTRACT FROM AUTHOR]

Remote sensing data prove to be an effective resource for constructing a data-driven predictive model of mineral prospectivity. Nonetheless, existing deep learning models predominantly rely on neural networks that necessitate a substantial number of samples, posing a challenge during the early stages of exploration. In order to predict mineral prospectivity using remotely sensed data, this study introduced deep forest (DF), a non-neural network deep learning model. Mainly based on ASTER multispectral imagery supplemented by Sentinel-2 and geological data, gold ore in Hamissana area, NE Sudan was used to test the DF predictive model capability. In addition to four geological-based evidential layers, 20 remote sensing-based evidential layers were generated using remote sensing enhancing techniques, forming the predictor variables of the proposed model. The applicability of the DF was thoroughly examined including its accuracy for delineating prospective areas, sensitivity to amount of training samples, and adjustment of hyperparameters. The results demonstrate that DF model outperformed conventional machine learning models (i.e., support vector machine, artificial neural network, and random forest) with AUC of 0.964 and classification accuracy of 93.3%. Moreover, the sensitivity analysis demonstrated that the DF model can be trained with a limited number (i.e., < 15) of mineral occurrences. Therefore, the DF algorithm has great potential and proves to be a viable solution for data-driven prospectivity mapping, particularly in scenarios with data availability constraints. [ABSTRACT FROM AUTHOR]

High-quality precipitation data input and the selection of reasonable and applicable hydrological models are the main ways to improve the accuracy of runoff simulation, and are crucial for flood control, drought resistance and comprehensive water resource management in the basin. This study takes the Jingle Basin as the research area, establishing a transformer model that integrates rainfall data from multiple sources considering environmental factors. It combines six types of remote sensing data with rainfall data, which are then used as inputs for the XAJ model, LSTM model, and Prophet model, respectively. The output results are further separately using the ensemble mean method and the Bayesian mean method for ensemble forecasting. The results show that: Compared with a single precipitation product, the fusion model considering environmental factors significantly enhances the correlation between the predicted rainfall and the observed rainfall, with the CC value reaching 0.72; Compared with the other two models, the LSTM model has the NSE value of 0.89, showing a better runoff prediction effect; Compared with the LSTM model with the NSE value of 0.85 and the ensemble average method with the NSE value of 0.76, the Bayesian model averaging method demonstrates the best runoff prediction and simulation effect, with the NSE value of 0.88. [ABSTRACT FROM AUTHOR]

Aiming at the low tracking accuracy and longtime tracking problems of the traditional automatic mobile human body tracking methods, this research proposes an automatic tracking method of the moving human body based on remote sensing spatial information. The proposed automatic tracking method for the moving human body first utilizes remote sensing technology to obtain the spatial information of the moving human body and then constructs a human body target motion model. By segmenting and fusing human body motion images, automatic matching and tracking of human body motion is finally realized. The experimental results showed that the auto-tracking time of the moving human body was only 0.3 s, while the auto-tracking accuracy rate was as high as 98.29%. In summary, the method used in this research has a high human motion tracking recognition effect. In addition, the study still has shortcomings, such as how to maintain high accuracy tracking in low light and bad weather conditions, still need to be studied. [ABSTRACT FROM AUTHOR]

The Amazon Basin is experiencing large-scale land use conversion from primary forest to pasture. While several land cover datasets map cleared areas in the Amazon, the percent cover of woody vegetation (trees and shrubs) in cleared areas and its association with clearing age, soil type, and geology is poorly understood, despite its importance for carbon emissions, biodiversity, land–atmosphere interactions, and monitoring of pasture condition. We used temporal mixture analysis on Sentinel-2 imagery from 2019 to map woody vegetation cover on cleared areas in Rondônia, Brazil. Binary woody vegetation masks were generated at 10-m resolution using a threshold of the evergreen endmember, with an overall accuracy of 84%. The age of clearing for each pixel was calculated from MapBiomas (Sousa and Davis, Remote Sens Environ 247, 2020) with a 2019 base year. We find little evidence of large-scale abandonment of pasture: most (53%) of the cleared area in 2019 was “clean pasture” (< 10% woody vegetation cover), 34% was “dirty pasture” (10–90% woody vegetation cover), 10% was forest (90–100% woody vegetation cover), and 3% was early stage clearing (> 10% woody vegetation cover, cleared 1–5 years ago). Recently cleared areas (1–2 years) had high (60%) woody vegetation cover, woody vegetation cover decreased with pasture age, and older pastures (20–34 years) had consistently low woody vegetation cover (25% on average). The commonly observed decrease in greenness with increasing clearing age, which is sometimes interpreted as decreasing grass health, was due in part to decreasing woody vegetation cover as pastures were gradually cleared over a decade. These results suggest modifications to existing conceptual models that describe clearing as a rapid process with high rates of secondary growth. We found a gradual and semi-permanent clearing of woody vegetation and proposed a revised conceptual model of deforestation dynamics. [ABSTRACT FROM AUTHOR]

The Ganges–Brahmaputra–Meghna delta is one of the most populous deltas in the world, threatened by diverse environmental stressors, including salinity intrusion, causing adverse consequences on livelihood adaptation. Shrimp farming has been recognized as one of the most practiced livelihoods among various livelihood opportunities. Growing global demand and higher economic returns encouraged many farmers to convert their agricultural land to shrimp farms by intruding saline water into the farmlands to produce shrimps. In this study, we investigated whether the growing trend of shrimp farming influences the migration aspirations of the vulnerable coastal communities in Bangladesh. The study employed qualitative interviews with individuals in four villages in southwest coastal Bangladesh. Satellite images were analyzed to detect land-use changes and changes in soil salinity in the last 28 years, confirming a rapid increase in shrimp farms. Results from the qualitative interviews indicate that poor people with limited resources cannot produce paddy on their land because of increased soil salinity caused by shrimp farming. Such a situation leads them to aspire to migrate. Thus, our analysis contributes to the discourse on “aspiration and capability” in environmental migration by adding the impacts of land-use changes on livelihoods. [ABSTRACT FROM AUTHOR]

The tropical Andes, the highest global biodiversity hotspot, is highly threatened by habitat loss and intense land use. Forest loss rates have remained persistent throughout the region during this century, resulting in a spread fragmented landscape composed of small forest fragments. However, the spatial distribution and magnitude of the proximate causes of associated land-cover change across the tropical Andes remain unclear at regional scales (e.g., provinces). Here, we combined climate, socioeconomic, and remote sensing data with machine-learning methods to model the proximate causes of forest loss across Antioquia, Colombia, where mountains dominate the landscape. We found that temperature is one of the main proximate causes of deforestation because it defines the distribution of crops and cattle (i.e., livestock) along an elevational gradient. We also detected that vegetation features and preceding deforestation predict current deforestation increasing our ability to identify proximate causes of deforestation. Additionally, there is a high influence of local-scale socioeconomic factors driving deforestation trends over the last decade, providing evidence that implementing control measures and economic alternatives tailored to each region can reduce deforestation rates. Our model would be used as a valuable tool to inform decisions aimed at reducing pressure on forests. [ABSTRACT FROM AUTHOR]

Satellite remote sensing (SRS) is a technique that can provide effective method on surface water quality assessment at large spatial scale studie. The analysis research involves: (1) analysis of changes in surface water quality in the Vam Co Tay River, Long An province, Vietnam in the period 2015–2020, (2) selection a model to estimate water quality assessment index from remote sensing data based on Bayesian Model Averaging (BMA); and (3) quantitative assessment of surface water pollution risks in the study area. The results show that the predictive coefficients of determination (R2) for water quality (BOD5, COD, and TSS) are higher than 0.70 for all three parameters. In particular, the upstream of Vam Co Tay river with "very high risk" in 2015 tended to decrease to "high risk" in 2020. Besides, the results also show an increasing risk in downstream from "low risk" in 2015 to "moderate risk" in 2020. The study demonstrated the potential of SRS for providing an overall assessment of the spatial distribution of risks associated with surface water pollution and forecasting the concentration changing trends in the future, and supporting to overcome data shortage in water monitoring. [ABSTRACT FROM AUTHOR]

In recent years, drought has become a global problem. Undoubtedly, drought monitoring is an important step for combating and reducing the resultant damage. Soil moisture, namely its spatial and temporal variability, is one of the most important environmental variables. Due to the difficulty cost, and timeliness of field measurements, this parameter has not been used widely in drought indexes. The recent development of global databases based on satellite imagery as well as rapid progress in hardware and software for modeling complex processes governing the water balance at the land surface employ these new tools to reduce the limitations in this field. The purpose of this research is to provide a comprehensive drought monitoring approach by integrating remote sensing data and the variable infiltration capacity (VIC) model with the Palmer Index (PDSI) in central Iran. In this study, the components of water and energy balance in the Central Iran region were simulated using the VIC land surface model. The output components of this model, especially soil moisture after evaluation, were used as inputs in the drought index based on Palmer's water balance. The integrated index of the VIC-PDSI in comparison with conventional Palmer indices and the SPI index at the 3, 6, 12, 24, and 48-month intervals was fitted with increments in moisture data and variations in the storage of water extracted from GRACE satellite data. Results showed that the combination of VIC-PDSI had the highest correlation coefficient of 0.87 with groundwater level change compared with other drought indices. [ABSTRACT FROM AUTHOR]

Grasslands and savannas are experiencing transformation and degradation due to bush encroachment (BE). BE has been monitored using restrictive traditional techniques that include field surveys and manual long-term observations. Owing to the limitations of traditional techniques, remote sensing (RS) is an attractive alternative to assess BE because of its generally high precision and return interval, cost-effectiveness, and availability of historical data archives. Furthermore, RS has an added advantage in its ability of acquiring global coherent data in near-real time compared to the snapshot acquisition mode with traditional surveying techniques. Despite its extensive application and vast possibilities, a critical synthesis for RS successes, shortcomings, and best practices in mapping BE in savannas and grasslands is lacking. Thus, broadly, the direction, which this type of investigation has taken over the years is largely unknown. This study sought to connect and measure the progress RS has made in mapping BE in grassland and savanna ecosystems through bibliometric analysis. One hundred and twenty-three peer-reviewed English written documents from the Web of Science and Scopus databases were evaluated. The study revealed 13.05% average annual publication growth, indicating that RS and BE mapping research in grasslands and savannas has been increasing over the survey period. Most published studies came from the USA, while the rest came from South Africa, China, and Australia. The results indicate that BE has been extensively mapped in grasslands and savannas using coarse to medium resolution data. As a result, there is a weak relationship (r² = 0.324) between the dependent variable (aerial images) and the independent variable (percentage of woody cover). This connotes the need to improve BE assessments in grasslands and savannas by integrating recent high-resolution data, machine learning algorithms and artificial intelligence. [ABSTRACT FROM AUTHOR]

Jordan is severely affected by climate change, it suffers from significance fluctuation and decrease in the amounts of the annual precipitation basically during the last decade which had dire consequences for farmers and the provision of fresh water. In this study, the impact of climate change on the Al-Wala basin was analyzed during the period 2013 to 2024 using Geomatics techniques, Google Earth Engine (GEE) and machine learning codes. Soil and Water Assessment Tool (SWAT) model was used to simulate the hydrological process up to year 2064. Moreover, the Meteorological Research Institute Earth System Model (MRI-ESM2-0) was used to predict the change of water surface area of the Al-Wala dam lake in the future. Annual satellite images: Lanadsat and sentinel, covering the period of the study area were downloaded and enhanced. They permit to provide the necessary information to carry out this study. As result, an important fluctuation of the amount of annual rainfall quantity was observed as well as, the amounts of annual rainfall expected to increase and decrease wobbly for several years in the future. Overall the average annual runoff will increase by 10% compared to the baseline scenario. The minimum temperature is expected to be higher than their rates throughout the year by 0.09°- 0.11o C, this will increase the evaporation rates with about 0.03%. The analysis of the sensitivity using the SWAT model was identified by 6 parameters out of 17. The regression coefficient (R2), Nash and Sutcliffe efficiency (NSE), on monthly basis, were above 0.60 for both of them which indicates satisfactory model results. [ABSTRACT FROM AUTHOR]

Existing methods for dehazing remote sensing (RS) images using deep learning have typically relied on convolutional frameworks. However, the limitations inherent in convolution, such as local receptive fields and independent input elements, hinder the network's ability to grasp long-range dependencies and nonuniform distributions. Consequently, the network is constrained in its capacity to learn these aspects. In response to this challenge, a proficient architecture for enhancing the clarity of remote sensing images through transformation, labeled RSDformer, has been designed. The architecture is structured to tackle the non-regular formations and varied spreads of hazing commonly found inside pictures of Remote Sensing. Emphasizing the importance of acquiring features from both nearby and distant areas, the design incorporates a novel detail-compensated transposed attention (DCTA) mechanism. This mechanism aims to get both localized and globalized dependency throughout channel. Furthermore, for enhancing model's capability towards learning from aspects that have undergone degradation and direct processes of restoring effectively, the DFBA or dualized frequencies blocks (adaptive) with filters of dynamic type has been developed. In the end DGBF or blocks of fusion that are dynamic has been devised to facilitate the effective fusion along with exchanging of aspects over differing levels. Through these innovations, these frameworks demonstrate robustness in their capability in capturing dependency in local regions and regions that are globalized, thereby enhancing restoration of visual information within the image. Wide-ranging experimental evaluations confirm superiorities of the proposed methodology over other competitive approaches. [ABSTRACT FROM AUTHOR]

Remote sensing scene classification is a critical task in the processing and analysis of remote sensing images. Traditional methods typically use standard convolutional kernels to extract feature information. Although these methods have seen improvements, they still struggle to fully capture unique local details, thus affecting classification accuracy. Each category within remote sensing scenes has its unique local details, such as the rectangular features of buildings in schools or industrial areas, as well as bridges and roads in parks or squares. The most important features are often these rectangular structures and their spatial positions, which standard convolutional kernels find challenging to capture effectively.To address this issue, we propose a remote sensing scene classification method based on a Rectangle Convolution Self-Attention Fusion Network (RCSFN) architecture. In the RCSFN network, the Rectangle Convolution Maximum Fusion (RCMF) module operates in parallel with the first 4 × 4 convolutional layer of VanillaNet-5. The RCMF module uses two different rectangular convolutional kernels to extract different receptive fields, enhancing the extraction of shallow local features through addition and fusion. This process, combined with the concatenation of the original input features, results in richer local detail information.Additionally, we introduce an Area Selection (AS) module that focuses on selecting feature information within local regions. The Sequential Polarisation Self-Attention (SPS) mechanism, integrated with the Mini Region Convolution (MRC) module through feature multiplication, enhances important features and improves spatial positional relationships, thereby increasing the accuracy of recognising categories with rectangular or elongated features. Experiments were carried out on AID and NWPU-RESISC45 data sets, and the overall classification accuracy was 96.56% and 92.46%, respectively. This shows that the RCSFN network model proposed in this paper is feasible and effective for class classification problems with unique local detail features. [ABSTRACT FROM AUTHOR]

Analysing land transformation and its driving factors is of utmost importance for the optimal land use planning for development at grassroots level. The south-western zone of Punjab, India is vulnerable in terms of adverse climatic conditions, problem of sand dunes, water logging and brackish ground water for irrigation. Therefore, it was felt imperative to understand the long term land use land cover dynamics in the intensively cultivated region of south-western Punjab having more than 80% area under agriculture. The objective was to study and analyse the conversion of agricultural land to urbanization, industrialization, and other classes, which is of significant concern to food production and security. Change detection for last two decades (2000, 2010 and 2020) was carried out using Landsat archive data to observe distinct land transformation. An increase has been observed in the agricultural land from 86.60 to 88.96% during 2000–2010 whereas it decreased to 87.38% during 2020. In contrast, the area under settlement shows a gradual increase i.e. 4.43, 6.22, 7.57% for 2000, 2010 to 2020, respectively indicating major conversion of agricultural and wastelands to this category. Further, reduction of wastelands from 6.82 to 2.03 to 1.4% can be attributed to ameliorative measures taken up by the government and farmers. The waterlogged areas showed the extent as 1.24%, 1.09%, and 2.0% for 2000, 2010 and 2020, respectively. The results of this study can be effectively utilised to alter the land use practices and policies for south-western region of Punjab, as the study area represents the similar ecological conditions of the entire region. [ABSTRACT FROM AUTHOR]

Soil degradation in Guinea has become a subject of national concern in general, and in Lower Guinea in particular. We set ourselves the objective of understanding the source of this phenomenon, evaluating its extent and proposing perspectives for its mitigation. To do this, we conducted about 30 surveys in the Boké region (Lower Guinea), combined with a very thorough geoscientific analysis (Teledetection, GIS, Cartography and Geology). Our results show that between 1982 and 1992, fresh vegetation cover was 67% and about 11% for each of the other classes: dry vegetation, barren soil and sand mineral soil. Between 1992 and 2002, the trend was 70% for fresh vegetation cover; 23% for dry vegetation; 2% for barren soil and 5% for sand mineral soil. Between 2002 and 2012, the same entities each and respectively increased to: 61%; 16%; 7% and 16% respectively. The period between 2012 and 2021 was marked by: 73% fresh vegetation cover; 22% dry vegetation; 2% barren soils and 3% sand mineral soil combination. Domestic needs, deforestation, meteorological variations and mining activity (bauxite mining affects the health of the local and surrounding population in several ways: by polluting the environment with charged air masses that move over several kilometres or by destroying its physical state; by the uncontrolled use of chemical explosives that impact human and animal wastewater in rivers and lakes) play a key role in this temporal and geographical fluctuation. In addition, to providing a more stable land management policy, we demonstrate here the natures of the negative impacts of all these factors and their magnitudes. The locations of the main pollution units in the region have been displayed. Degradation over time and in space was tracked, and then position the main rivers in relation to the high-risk areas. [ABSTRACT FROM AUTHOR]

Hyperspectral-multispectral image fusion (HSI-MSI Fusion) for enhancing resolution of hyperspectral images is a hot topic in remote sensing. An important category of approaches for HSI-MSI Fusion is based on deep learning. The main challenges in deep learning based fusion methods include the lack of training data, poor generalization to various datasets, and high computational costs. This paper suggests a new approach to tackle these difficulties by introducing an innovative technique for HSI-MSI fusion. The proposed method involves training a tiny deep neural network that can reconstruct high-resolution hyperspectral images through spectral super-resolution of high-resolution multispectral images. This method does not require high resolution training data and they are artificially generated based on the spatial degradation model of the input observation images. Therefore, the problems of data scarcity and poor generalization are addressed, and also the computational burden is significantly reduced. After conducting thorough experiments, it was found that the proposed method provides promising results. The source code of this method is available at https://github.com/rajaei-arash/SSSR-HSI-MSI-Fusion. [ABSTRACT FROM AUTHOR]

Greater Dhaka area is home to large industrial clusters that are driving economic growth and the poverty reduction efforts of Bangladesh. These clusters are located around peripheral rivers- Turag, Buriganga, Dhaleswari, Balu, Shitalakhya, Bangshi, and Tongi-Khal, which are important for water transport, environment, and eco-systems where flooding of floodplains in monsoon is an integral part. The urban and industrial growth stressing natural resources has led to severe degradation of the rivers and floodplains, affecting the livelihoods, health, and well-being of the people. Monsoon-time exposure to polluted water is yet to be studied and addressed scientifically. This study looked into the water quality and flooding situation of Greater Dhaka for two successive monsoons through extensive river sampling coupled with the estimation of flooded area and exposed population using remote sensing tools. Sentinel 1’s Synthetic Aperture Radar (SAR) images are used for flood mapping considering cost-effectiveness and its advantages for data-scarce regions. The estimated population exposed to flooding was over 668,000 in 2019, and this number increased by 1.53 times in 2020, totaling over one million. During the monsoon and post-monsoon periods of 2019 and 2020, Buriganga, Tongi Khal, and Balu were consistently in poor condition. The lowest water quality index (WQI) was observed in Balu during the monsoon of 2019 (32.28), post-monsoon of 2019 (35.71), and post-monsoon of 2020 (29.58). The lowest WQI during the monsoon of 2020 was recorded in Tongi Khal (35.75). Among the four districts, Dhaka and Gazipur were the most affected by floods in terms of inundation area and exposed population. Our study indicates that most rivers remain in poor condition during the monsoon when exposure is also high. This highlights the need for policymakers to take monsoon exposure seriously and design appropriate interventions. [ABSTRACT FROM AUTHOR]

Land use and land cover (LULC) changes are recognized as drivers of environmental changes at the local, regional, and global levels. Detecting these changes is essential in developing land management plans and strategies. More particularly, the LULC changes constitute one of the sub-indicators used for setting the baseline for land degradation neutrality (LDN) planning and assessing the progress toward the sustainable development goal 15.3. This study aims to determine the implication of LULC dynamics on land cover degradation in Eastern DR Congo. This region has been subjected to uncontrolled LULC changes over the last three decades but the impact of these changes on land degradation has not been assessed yet at the territorial level. To fill this gap, the territory of Kalehe has been used as a case study to monitor the LULC changes during the 1987–2020 period based on the classification of Landsat images, to forecast the future LULC for the 2030–2070 period through the hybrid Markov-Support Vector Machine modeling, to determine the extent of land cover degradation associated with these changes and to assess the impact of biophysical and socio-economic factors on the occurrence of land cover degradation using the binary logistic regression model. The results reveal that the overall trends of LULC changes during the 1987–2020 period are the increasing of built-up area, shrubland, and cropland at the expense of forestland, wetland, and grassland. This situation is expected to continue in the future and contribute to the degradation of land cover within the study area. Under the current situation of LULC changes, 34.17% of the land has been subjected to potential degradation. Furthermore, under the business-as-usual scenario 28.28%, 27.28%, and 33.65% of the land will be degraded by 2030, 2050, and 2070 respectively. This land cover degradation is more likely to occur in the proximity of roads, localities, highly populated areas, mining concessions, and high-altitudinal zones. Since the current land use system is unsustainable, there is a necessity to implement sustainable land management strategies that take into account the biophysical and socio-economic specificities of this region to reverse the problem of land cover degradation. [ABSTRACT FROM AUTHOR]

Crop classification using remote sensing technology is highly important for monitoring agricultural resources and managing water usage, especially in water-scarce regions like the Hei River. Crop classification requires a substantial number of labeled samples, but the collection of labeled samples demands significant resources and sample data may not be available for some years. To classify crops in sample-free years in the middle reaches of the Hei River, we generated multisource spectral data (MSSD) based on a spectral library and sample data. We pre-trained a model using labeled samples, followed by fine-tuning the model with MSSD to complete the crop classification for the years without samples. We conduct experiments using three CNN-based deep learning models and a machine learning model (RF). The experimental results indicate that in the model transfer experiments, using a fine-tuned model yields accurate classification results, with overall accuracy exceeding 90%. When the amount of labeled sample data is limited, fine-tuning the model based on MSSD can enhance the accuracy of crop classification. Overall, fine-tuning models based on MSSD can significantly enhance the accuracy of model transfer and reduce the reliance of deep learning models on large-scale sample datasets. The method to classify crops in the middle reaches of the Hei River can provide data support for local resource utilization and policy formulation. [ABSTRACT FROM AUTHOR]

Monitoring within-season dynamics of aboveground biomass (AGB) is critical for adaptive decision-making in grasslands to respond to changing conditions caused by frequent disturbances (e.g., grazing activities, fires, climate variations). This study trained the random forest model based on intensive observations from harmonious remote sensing data (Landsat, Sentinel-2, Sentinel-1) and an accurate date grassland sample database. Then we used monthly average remote sensing data to estimate monthly AGB at 10 m across three distinct grassland types in China: temperate meadow steppe, temperate typical steppe, and temperate desert steppe. Considering all the grassland types, the mean predicted AGB values were 83.02 g/m2 in June, 103.02 g/m2 in July, 108.49 g/m2 in August, and 106.89 g/m2 in September. The overall prediction accuracy was evaluated as R2 = 0.68, RMSE = 47.87 g/m². Additionally, the comparison of our monthly AGB maps with 3 m PlanetScope images and the published 30 m yearly AGB map shows the significant advantages of monthly 10 m products to capture the spatial and within-season dynamics of grassland AGB. [ABSTRACT FROM AUTHOR]

In the Antarctic, Arctic, and Tibetan Plateau—recognized as the Earth's three poles characterized by extremely harsh environments—lichens prevail in the ecosystem and play crucial roles as pioneer species. Despite their importance, studies investigating the spatial distribution patterns of lichen attributes are scarce due to a lack of appropriate datasets. To bridge this gap and enhance our understanding of the growth preferences of lichens in these areas, here we present a geospatial dataset encompassing key attributes of lichens, such as color type and growth form, for over 2800 lichen species and 170,000 in-situ lichen records. The dataset facilitates the creation of the first spatial distribution map illustrating the variation of lichen attributes across different latitudes and longitudes. This can serve as a foundational resource for studies on the relationship between lichen types and their growing environment, which is a vital scientific question in the ecology domain. Additionally, it can contribute to the development of specialized remote sensing technique tailored for lichen monitoring, which is currently lacking. [ABSTRACT FROM AUTHOR]

Urbanization and changes in land use are crucial for national development. Managing and monitoring these changes have become easier and more accurate due to advancements in imaging systems and change detection methods. In recent years, convolutional neural networks have gained significant attention and advancement in remote sensing change detection research. However, designing a model that reduces computational complexity while improving detection accuracy is still under discussion. In this regard, an efficient model based on pseudo-siamese convolutional networks had been developed in this research, which by combining multi-scale features at different levels, increased the accuracy of the final change map, especially in images with both different spatial resolutions and scales. Our model is inspired by Scale Invariant Feature Transform (SIFT), which uses multiple scale-spaces to identify scale-independent keypoints in images. We designed our model based on this concept, in which After passing the bi-temporal images through the encoder network and making the initial feature maps, the feature learning process takes place at several levels of different scales and finally, the change map is calculated. In addition, the proposed architecture leverages point-wise, depth-wise, and neighboring features across various scale levels to harness the optimal features present in high-resolution remote sensing images. Simplicity in design, effectiveness in calculations, and accuracy make our model surpass popular methods such as BIT, DSIFN, etc. The evaluation results on 4 different datasets and comparison with 12 state-of-the-art models demonstrate that the proposed method outperforms other popular models in change detection. Available at: https://github.com/farhadinima75/OctaveNet. [ABSTRACT FROM AUTHOR]

High Frequency Radar (HFR) has gained world-wide use as a land based remote sensing technology capable of measuring ocean surface currents and ocean waves at ranges up to 200 km or more. Regional HFR networks support a variety of services including support for search and rescue, marine spill response, and resource management operations. HFR data is useful for validation and calibration of hydrodynamic models that are used to forecast storm-surge, combined flooding and surface currents. Long-term time series data are also needed to assess changes in ocean/estuary currents in response to changes in climate, hydrology, and major coastal infrastructure. The main objective of this paper is to document the site selection process applied for two HFR networks, commissioned in Galveston Bay and Sabine Lake in Southeast Texas. The general process involved several steps including: identify sites meeting technical requirements and constraints; site access negotiation, permitting, station design, station commissioning, network operation and maintenance. To some degree, these processes could be considered sequential; however, in practice intermediate steps were essentially iterative in nature. [ABSTRACT FROM AUTHOR]

Here we describe the data obtained by a successful proof-of-concept initiative to launch the first ocean color imager on board a CubeSat satellite and collect research-grade imagery at severalfold higher spatial resolution than any other ocean color satellite mission. The 3U CubeSat, named SeaHawk, flew at a nominal altitude of 585 km. Its ocean color sensor, HawkEye, collected 7,471 research-grade push-broom images of 230 × 780 km2 at best-in-class 130 × 130 m2 per pixel. The sensor is built with comparatively low-cost commercial off-the-shelf optoelectronics and was designed to match NASA SeaWiFS ocean color specifications, including wavelengths, bandwidths, and signal-to-noise ratios. HawkEye's design for ocean color remote sensing combined with its high spatial resolution make the imagery especially well-suited for coastal, estuarine, and limnological applications. Ultimately, the successful mission provided open access to a rich global dataset of calibrated and quality-controlled imagery for use in aquatic ecology and environmental change studies. [ABSTRACT FROM AUTHOR]