Search

The use of bi-temporal remote sensing images for detecting changes in land cover is an important means of obtaining surface change information, thus contributing to urban governance and ecological environment monitoring. In this article, we propose a deep learning model named Siam-UNet3+ for high-resolution remote sensing image change detection. This model integrates the full-scale skip connections and full-scale deep supervision of the network UNet3+, which can achieve the multi-scale feature fusion of remote sensing images, effectively avoiding the locality disadvantage of convolution operations. Different from UNet3+, Siam-UNet3+ has made major improvements, including the following: (1) incorporating a Siamese network in the encoder, which can process bi-temporal remote sensing images in parallel; (2) leveraging the residual module as the backbone, which can avoid gradient vanishing (or exploding) and model degradation problems; (3) adding a Triplet Attention module to the decoder, which can avoid information redundancy that may occur in full-scale skip connections and increase the ability to focus on changing patterns; and (4) designing a hybrid loss function consisting of focal loss and dice loss, which is more suitable for remote sensing image change detection tasks. In this study, we conducted change detection experiments using the publicly available LEVIR-CD dataset, as well as two local datasets in Beijing. Through comparative experiments with five other models and ablation experiments, the proposed model Siam-UNet3+ in this article demonstrated significant advantages and improvements in four evaluation metrics, namely, precision, recall, F1-score, and overall accuracy (OA), proving to have great potential in the application to highresolution remote sensing image change detection tasks. [ABSTRACT FROM AUTHOR]

A study of land use and land cover (LULC) mapping using satellite data was conducted in the Congo Basin rainforest in the Sangha area of the northern Congo Republic, which is crucial for sustainable land management. However, the heterogeneous nature of LULC features makes data classification difficult in this area because persistent cloud cover is challenging for observing the land surface and tropical humid conditions characterizing the study area. This study compares and evaluates the potential of two of the most widely used and popular satellite image sources, the recently launched Landsat 9 and Sentinel-2, to generate land-use land cover (LULC) maps in the Sangha area using pixel-based supervised machine learning algorithms, Random Forest (RF), for classification. All procedures were performed using ArcGIS Pro. The potential to improve Landsat 9 performance was tested using the nearest resampling technique, and different bands were resampled from 30 × 30 to 10 × 10 m. Owing to the heterogeneous characteristics of the Sangha study area, six classes were defined: dense forest, open forest, wetland forest, water bodies, urban areas, and bare soils. To classify and compare Sentinel-2 and Landsat 9, the shortwave infrared (SWIR), Near Infrared (NIR), and Red, Green, Blue (RGB) bands were used. These two satellite images were compared to test the quality of their results, particularly for assessing the accuracy of the LULC classifications and identifying which dataset had the highest accuracy. Results show that overall accuracy was 93.80% for Sentinel-2 while it was 91.60% for Landsat 9, similarly, the Kappa coefficient was calculated 0.89 and 0.85 for Sentinel-2 and Landsat 9, respectively. Therefore, Sentinel-2 exhibits significant classification ability compared to the nearest resample technique Landsat 9, despite the improved resolution of the latter, which offers a scientific basis for choosing the appropriate satellite imagery to create accurate LULC maps. [ABSTRACT FROM AUTHOR]

Land use/land cover (LULC) change information is crucial for monitoring purposes, formulating strategies, socioeconomic progress, and decision-making. The main objective of this study was to analyze and quantify the changes in land use as well as land cover patterns within the Maamora forest in Morocco, and to identify the key factors that influenced its trend from 1989 to 2022. In this study, multispectral remote sensing (RS) data were employed to detect land cover changes in the Maamora forest using Landsat images for the years 1989, 1999, 2009, 2019 and 2022. The maximum likelihood classification (MLC) method was applied to classify the Landsat images using ArcMap 10.4 software to analyze the current state of the study area. Seven LULC classes (cork oak, eucalyptus, pine, acacia, bare land, daya, and others) were successfully classified, achieving overall accuracies surpassing 86% and Kappa coefficients greater than 0.85 for all selected dates. The results of the land use/land cover change detection indicate a decrease in the cork oak area from 60.71% to 44.42%, along with an increase in the eucalyptus area from 18.11% to 39.31%. Moreover, the pine, acacia, bare land, daya, and other classes went from 17.22, 2.80, 0.95, 0.05, and 0.12% to 4.58, 0.02, 10.84, 0.34, and 0.48% respectively. Indeed, from 1989 to 2022, around 50.84% of the study area’s surface remained unchanged, whereas 49.16% underwent changes, transitioning to other land cover classes or endured degradation. This research underscored the anthropogenic transformation of the Maamora woodland, which has led to the degradation of its natural resources. Broadly, these findings can serve as foundational data for future research endeavors and offer valuable insights to concentrate on the key factors driving forest degradation in order to inform the development of interventions aimed at preserving the sustainability of natural species and the overall ecosystem. [ABSTRACT FROM AUTHOR]

The Greater Amanzule Peatlands (GAP) in Ghana is an important biodiversity hotspot facing increasing pressure from anthropogenic land-use activities driven by rapid agricultural plantation expansion, urbanisation, and the burgeoning oil and gas industry. Accurate measurement of how these pressures alter land cover over time, along with the projection of future changes, is crucial for sustainable management. This study aims to analyse these changes from 2010 to 2020 and predict future scenarios up to 2040 using multi-source remote sensing and machine learning techniques. Optical, radar, and topographical remote sensing data from Landsat-7, Landsat-8, ALOS/PALSAR, and Shuttle Radar Topography Mission derived digital elevation models (DEMs) were integrated to perform land cover change analysis using Random Forest (RF), while Cellular Automata Artificial Neural Networks (CA-ANNs) were employed for predictive modelling. The classification model achieved overall accuracies of 93% in 2010 and 94% in both 2015 and 2020, with weighted F1 scores of 80.0%, 75.8%, and 75.7%, respectively. Validation of the predictive model yielded a Kappa value of 0.70, with an overall accuracy rate of 80%, ensuring reliable spatial predictions of future land cover dynamics. Findings reveal a 12% expansion in peatland cover, equivalent to approximately 6570 ± 308.59 hectares, despite declines in specific peatland types. Concurrently, anthropogenic land uses have increased, evidenced by an 85% rise in rubber plantations (from 30,530 ± 110.96 hectares to 56,617 ± 220.90 hectares) and a 6% reduction in natural forest cover (5965 ± 353.72 hectares). Sparse vegetation, including smallholder farms, decreased by 35% from 45,064 ± 163.79 hectares to 29,424 ± 114.81 hectares. Projections for 2030 and 2040 indicate minimal changes based on current trends; however, they do not consider potential impacts from climate change, large-scale development projects, and demographic shifts, necessitating cautious interpretation. The results highlight areas of stability and vulnerability within the understudied GAP region, offering critical insights for developing targeted conservation strategies. Additionally, the methodological framework, which combines optical, radar, and topographical data with machine learning, provides a robust approach for accurate and detailed landscape-scale monitoring of tropical peatlands that is applicable to other regions facing similar environmental challenges. [ABSTRACT FROM AUTHOR]

Accurate burn severity mapping is essential for understanding the impacts of wildfires on vegetation dynamics in arid savannas. The frequent wildfires in these biomes often cause topkill, where the vegetation experiences above-ground combustion but the below-ground root structures survive, allowing for subsequent regrowth post-burn. Investigating post-fire regrowth is crucial for maintaining ecological balance, elucidating fire regimes, and enhancing the knowledge base of land managers regarding vegetation response. This study examined the relationship between bush burn severity and woody vegetation post-burn coppicing/regeneration events in the Kalahari Desert of Botswana. Utilizing UAV-derived RGB imagery combined with a Random Forest (RF) classification algorithm, we aimed to enhance the precision of burn severity mapping at a fine spatial resolution. Our research focused on a 1 km2 plot within the Modisa Wildlife Reserve, extensively burnt by the Kgalagadi Transfrontier Fire of 2021. The UAV imagery, captured at various intervals post-burn, provided detailed orthomosaics and canopy height models, facilitating precise land cover classification and burn severity assessment. The RF model achieved an overall accuracy of 79.71% and effectively identified key burn severity indicators, including green vegetation, charred grass, and ash deposits. Our analysis revealed a >50% probability of woody vegetation regrowth in high-severity burn areas six months post-burn, highlighting the resilience of these ecosystems. This study demonstrates the efficacy of low-cost UAV photogrammetry for fine-scale burn severity assessment and provides valuable insights into post-fire vegetation recovery, thereby aiding land management and conservation efforts in savannas. [ABSTRACT FROM AUTHOR]

The emergence of crowdsourced geographic information (CGI) has markedly accelerated the evolution of land-use and land-cover (LULC) mapping. This approach taps into the collective power of the public to share spatial information, providing a relevant data source for producing LULC maps. Through the analysis of 262 papers published from 2012 to 2023, this work provides a comprehensive overview of the field, including prominent researchers, key areas of study, major CGI data sources, mapping methods, and the scope of LULC research. Additionally, it evaluates the pros and cons of various data sources and mapping methods. The findings reveal that while applying CGI with LULC labels is a common way by using spatial analysis, it is limited by incomplete CGI coverage and other data quality issues. In contrast, extracting semantic features from CGI for LULC interpretation often requires integrating multiple CGI datasets and remote sensing imagery, alongside advanced methods such as ensemble and deep learning. The paper also delves into the challenges posed by the quality of CGI data in LULC mapping and explores the promising potential of introducing large language models to overcome these hurdles. [ABSTRACT FROM AUTHOR]

Effective groundwater management is crucial under the current climatic conditions, addressing both qualitative and quantitative aspects. An important step in delineating groundwater potential zones involves remote sensing (RS) data and geographic information systems (GISs), facilitating resource assessment, and the implementation of suitable field data management. This study introduces the delineation of potential groundwater zones using seven layers and the Multi-Criteria Decision Analysis (MCDA) method. Satty's Analytic Hierarchy Process (AHP) was employed to rank the seven selected parameters, contributing to the advancement of groundwater research and resource assessment. All seven thematic layers (Rainfall, Geology, Land Use/Land Cover, Drainage Density, Elevation, Slope, and Soil) were prepared and analyzed to delineate groundwater potential zones. The resulting groundwater potential zone map was categorized into four classes, Very Good, Good, Moderate, and Poor, covering areas of 81.53 km2 (45.1%), 56.36 km2 (31.2%), 19.54 km2 (10.8%), and 23.17 km2 (12.8%) of the total area, respectively. The accuracy of the output was validated by comparing it with information on groundwater prospects in the area, and the overall accuracy of the method was approximately 72%. High-yield boreholes were drilled and concentrated in the Very Good groundwater potential zones, while low-yield ones were developed in the Poor areas. [ABSTRACT FROM AUTHOR]

Investigating the effects of spatial scales on the uncertainty and sensitivity analysis of the social vulnerability index (SoVI) model output is critical, especially for spatial scales finer than the census block group or census block. This study applied the intelligent dasymetric mapping approach to spatially disaggregate the census tract scale SoVI model into a 300‐m grids resolution SoVI map in Davidson County, Nashville. Then, uncertainty analysis and variance‐based global sensitivity analysis were conducted on two scales of SoVI models: (a) census tract scale; (b) 300‐m grids scale. Uncertainty analysis results indicate that the SoVI model has better confidence in identifying places with a higher socially vulnerable status, no matter the spatial scales in which the SoVI is constructed. However, the spatial scale of SoVI does affect the sensitivity analysis results. The sensitivity analysis suggests that for census tract scale SoVI, the indicator transformation and weighting scheme are the two major uncertainty contributors in the SoVI index modeling stages. While for finer spatial scales like the 300‐m grid's resolution, the weighting scheme becomes the uttermost dominant uncertainty contributor, absorbing uncertainty contributions from indicator transformation. [ABSTRACT FROM AUTHOR]

The urban functional zone, serving as a bridge to understanding the complex interactions between human spatial activities and surface thermal environmental changes, explores the driving force information of its internal temperature changes, which is crucial for improving the urban thermal environment. However, the impacts of the current urban functional zones on the thermal environment, based on the delineation of human activities, have yet to be sufficiently investigated. To address the issue, we constructed a two-factor weighted dominant function vector model of "population heat--land use scale" to identify urban functional zones. This model is based on multisource data and considers the perspective of urban functional supply and demand matching. We then analyzed the spatial differentiation and driving factors of the relationship between urban functional zones and the surface thermal environment using the random forest algorithm, bivariate spatial autocorrelation, geographical detectors, and geographically weighted regression models. The results showed that there are significant differences in the Land Surface Temperature among different urban functional zones in the central urban area of Lanzhou. Among these, the life service zone has the greatest impact on the surface thermal environment, followed by the industrial zone and catering service zone, while the green space zone has the least impact. The surface thermal environment exhibits high-high clusters in localized spatial clustering patterns with life service, industrial, catering service, and residential zones. In contrast, it tends to exhibit low-high clusters with green spaces. Significant spatial clustering and dependence exist between various functional zones and the surface thermal environment. The land cover types characterized by the Normalized Difference Bare Land and Building Index, the vegetation coverage represented by the Fraction of Vegetation Cover, and the density of industrial activities indicated by the Industrial POI Kernel Density Index are the main drivers of the surface thermal environment in the various functional zones of the central urban area of Lanzhou, and all exhibit significant spatial heterogeneity. [ABSTRACT FROM AUTHOR]

Improvements in high-resolution satellite remote sensing and computational advancements have sped up the development of global datasets that delineate urban land, crucial for understanding climate risks in our increasingly urbanizing world. Here, we analyze urban land cover patterns across spatiotemporal scales from several such current-generation products. While all the datasets show a rapidly urbanizing world, with global urban land nearly tripling between 1985 and 2015, there are substantial discrepancies in urban land area estimates among the products influenced by scale, differing urban definitions, and methodologies. We discuss the implications of these discrepancies for several use cases, including for monitoring urban climate hazards and for modeling urbanization-induced impacts on weather and climate from regional to global scales. Our results demonstrate the importance of choosing fit-for-purpose datasets for examining specific aspects of historical, present, and future urbanization with implications for sustainable development, resource allocation, and quantification of climate impacts. There has been a surge in global datasets of urban land recently. This paper shows large discrepancies in urban area across scales among multiple such datasets, which can influence the magnitude and direction of estimated urban climate impacts. [ABSTRACT FROM AUTHOR]

Wetlands play a vital role in regulating the global carbon cycle, providing biodiversity, and reducing flood risks. These functions maintain ecological balance and ensure human well-being. Timely, accurate monitoring of wetlands is essential, not only for conservation efforts, but also for achieving Sustainable Development Goals (SDGs). In this study, we combined Sentinel-1/2 images, terrain data, and field observation data collected in 2020 to better understand wetland distribution. A total of 22 feature variables were extracted from multi-source data, including spectral bands, spectral indices (especially red edge indices), terrain features, and radar features. To avoid high correlations between variables and reduce data redundancy, we selected a subset of features based on recursive feature elimination (RFE) and Pearson correlation analysis methods. We adopted the random forest (RF) method to construct six wetland delineation schemes and incorporated multiple types of characteristic variables. These variables were based on remote sensing image pixels and objects. Combining red-edge features, terrain data, and radar data significantly improved the accuracy of land cover information extracted in low-mountain and hilly areas. Moreover, the accuracy of object-oriented schemes surpassed that of pixel-level methods when applied to wetland classification. Among the three pixel-based schemes, the addition of terrain and radar data increased the overall classification accuracy by 7.26%. In the object-based schemes, the inclusion of radar and terrain data improved classification accuracy by 4.34%. The object-based classification method achieved the best results for swamps, water bodies, and built-up land, with relative accuracies of 96.00%, 90.91%, and 96.67%, respectively. Even higher accuracies were observed in the pixel-based schemes for marshes, forests, and bare land, with relative accuracies of 98.67%, 97.53%, and 80.00%, respectively. This study's methodology can provide valuable reference information for wetland data extraction research and can be applied to a wide range of future research studies. [ABSTRACT FROM AUTHOR]

The application of remote sensing technology in water body detection has become increasingly widespread, offering significant value for environmental monitoring, hydrological research, and disaster early warning. However, the existing methods face challenges in multi-scene and multi-temporal water body detection, including the diverse variations in water body shapes and sizes that complicate detection; the complexity of land cover types, which easily leads to false positives and missed detections; the high cost of acquiring high-resolution images, limiting long-term applications; and the lack of effective handling of multi-temporal data, making it difficult to capture the dynamic changes in water bodies. To address these challenges, this study proposes a novel network for multi-scene and multi-temporal water body detection based on spatiotemporal feature extraction, named TSAE-UNet. TSAE-UNet integrates convolutional neural networks (CNN), depthwise separable convolutions, ConvLSTM, and attention mechanisms, significantly improving the accuracy and robustness of water body detection by capturing multi-scale features and establishing long-term dependencies. The Otsu method was employed to quickly process Sentinel-1A and Sentinel-2 images, generating a high-quality training dataset. In the first experiment, five rectangular areas of approximately 37.5 km2 each were selected to validate the water body detection performance of the TSAE-UNet model across different scenes. The second experiment focused on Jining City, Shandong Province, China, analyzing the monthly water body changes from 2020 to 2022 and the quarterly changes in 2022. The experimental results demonstrate that TSAE-UNet excels in multi-scene and long-term water body detection, achieving a precision of 0.989, a recall of 0.983, an F1 score of 0.986, and an IoU of 0.974, significantly outperforming FCN, PSPNet, DeepLabV3+, ADCNN, and MECNet. [ABSTRACT FROM AUTHOR]

The digital number of a given pixel in a satellite image does not only measure the radiance originating from the area of the Earth’s surface represented by this pixel, but also measures radiance originating from surrounding areas represented by adjacent pixels. This adjacency effect, also known as image blurring, is a fundamental limitation of satellite images, and introduces errors in all image processing techniques that quantify the properties of the Earth’s surface on a per-pixel basis. Image deblurring via deconvolution, is a traditional technique to reduce the adjacency effect, which has proven effective in a variety of remote sensing applications. For example, it has provided an almost fourfold reduction of the adjacency error, when using support vector machine classifiers to estimate land cover proportion at a subpixel level. The remote sensing literature usually assumes that satellite images should be deconvolved with a kernel defined by the sensor’s Point Spread Function PSF(x, y), overlooking the results of three seminal empirical studies which systematically demonstrated that the optimum deconvolution kernel is a shrunk PSF of the form PSF(x/β, y/β), where β < 1 is the optimum shrink factor. The current empirical procedure to find the optimum shrink factor is laborious and restrictive since it requires suitable satellite images of much higher resolution, such that their Ground Sampling Distance (GSD) is equal to the GSD of the sensor of interest divided by an integer number much greater than one. A new theoretical procedure that uses synthetic edge images to find the optimum shrink factor is proposed and applied to the same cases considered by the current empirical procedure, showing that the same results are obtained. The new procedure is much simpler to apply and more accurate and versatile, opening a variety of research paths on satellite image deblurring. [ABSTRACT FROM AUTHOR]

Considering the characteristics of remote sensing land cover images with a wide variety of types and complex object edges as well as the limited receptive field of local convolutions in existing segmentation networks resulting in inadequate utilization of contextual information, leading to issues such as blurred object edges and low segmentation accuracy, this paper proposes a remote sensing land cover segmentation algorithm based on the UNet3+ network architecture. Firstly, during the decoding process, a similarity-aware point affiliation operator is introduced as an upsampling method. This operator aggregates multiple proposals from the feature pyramid to enhance the segmentation capability for object boundary details. Secondly, in the encoding process, a selective kernel module is introduced to optimize the downsampling approach. This module enables neurons to achieve an adaptive receptive field size, facilitating the acquisition of multi-scale information from target features and precise capture of valuable detailed semantic information. Finally, in the skip-connection phase, a dual multi-scale attention module is added to perform weighted fusion of features from different scales, enabling the model to better focus on both local details and global contextual information. Experimental results on the WHDLD and ISPRS Potsdam datasets demonstrate that the proposed algorithm achieves mean intersection over union (MIoU) improvements of 64.4% and 75.4% respectively, compared to baseline models, the improvement is about 2.6 percentage points and 3.2 percentage points respectively. This also validates the effectiveness of the proposed algorithm in addressing the issue of blurry segmentation edges. [ABSTRACT FROM AUTHOR]

Large-scale, high-resolution land cover classification is a prerequisite for constructing Earth system models and addressing ecological and resource issues. Advancements in satellite sensor technology have led to improvements in spatial resolution and wider coverage areas. Nevertheless, the lack of high-resolution labelled data is still a challenge, hindering the large-scale application of land cover classification methods. In this study, a Transformer-based weakly supervised method for cross-resolution land cover classification using outdated data is proposed. First, to capture long-range dependencies without overlooking the fine-grained details of objects, a U-Net-like Transformer based on a reverse difference mechanism (RDM) using dynamic sparse attention is designed. Second, an anti-noise loss calculation module based on optimal transport (OT) is proposed. The anti-noise loss calculation identifies confident areas and vague areas based on the OT matrix, which relieves the effect of noises on outdated land cover products. By introducing a weakly supervised loss with weights and using an unsupervised loss, the RDM-based U-Net-like Transformer was trained. Remote sensing images with 1 m resolutions and the corresponding ground truths of six states in the United States were used to validate the performance of the proposed method. The experiments used outdated land cover products with 30 m resolutions from 2013 as training labels and produced land cover maps with 1 m resolutions from 2017. The results showed the superiority of the proposed method over state-of-the-art methods. The code is available at https://github.com/yu-ni1989/ANLC-Former. [ABSTRACT FROM AUTHOR]

Objectives: In the land cover classification study from multi-source remote sensing images, domain adaptation method can align images or extracted image features from source and target images, thus improves the generalization ability of deep learning models and plays an important role in intelligent remote sensing image interpretation. Methods: A self-training domain adaptation method based on information entropy uncertainty estimation for pseudo label correction is proposed, its core is an entropy uncertainty loss function for land cover classification between cross source remote sensing images. First, a land cover classification model is pretrained on the source domain training set with ground truth, and is applied on the target domain images without ground truth labels to generate pseudo labels. Then, the pseudo labels are used to further train the model, the information entropy of the prediction result is calculated and used as the uncertainty estimation of the pseudo labels to further correct the pseudo labels with self-training, so as to obtain weights of the classification model more suitable for the target domain dataset. Finally, a cross domain classification experiment was conducted on three data sets, namely, the WHU building change detection data set, the ISPRS 2D semantic annotation competition data set, and the Wuhan land cover classification data set. Results: Experimental results show that:(1) The proposed method improved the mean intersection over union(mIoU) and overall accuracy(OA) of semantic segmentation network by 0.3%-3.1% and 1.2%-4.5%, respectively. (2) Compared with the traditional self-training method, the proposed method can improve the mIoU and OA by 0.1%-1.5%. (3) Compared with the most recent uncertainty estimation method based on Kullback-Leibler divergence, the proposed method can improve the mIoU and OA about 0.6% in average. Conclusions: The proposed method can further improve the performance of a trained segmentation model for the target domain images without the requirement of target labels. There is also no need to introduce additional modules or parameters on the existing segmentation model. [ABSTRACT FROM AUTHOR]

Land cover change research is salient in providing information for land use planning and management in urban areas. This study therefore analysed land cover change in four local government areas in Ogun State, Nigeria. Using remote sensing and geographic information system, Landsat images were analysed between 2001, 2013 and 2020 with both unsupervised and supervised image classifications. Results showed that between 2002 and 2013, there was a 41.14% decrease in forest area which further reduced by 41.92% by 2020. Agricultural land increased by 66.47% between 2002 and 2013 but reduced by 43.24% by 2020. Urban areas increased by 108.66% between 2002 and 2013 while by 2020 they had gained 42.19% additional land. It is concluded that the 2005–2025 Ogun State regional planning policy has so far been less than effective due to massive ecological change caused by urban expansion in the region. [ABSTRACT FROM AUTHOR]

The post‐globalization of the Indian economy marks the beginning of a new era for various industries, including manufacturing and the service sector. As a result, people begin to migrate from sub‐urban/rural area to nearby major cities in search of employment and access to modern conveniences. Land Use Land Cover (LULC) pattern change is caused by human and natural factors which include geology, elevation, and slope. LULC change assessment is best ways to manage and understand landscape transformation. Bilari block in Moradabad is India's brass city and one of Uttar Pradesh's major cities, its population growth has pushed the rapid urbanization of city in last two decades. In the present study Landsat 7 Enhanced Thematic Mapper (ETM) and Landsat 8 Operational Land Manager (OLI) data for years 2000 and 2023 have been employed to assess. ArcGIS Pro 2.9.0 software is used to perform layer stacking and on the raw Landsat data for pre‐processing. To find LULC changes in the study area between 2000 and 2023, supervised classification is done by adopting Maximum Likelihood Classification (MLC). From the results, it is observed that the area from pervious to impervious changes very quickly in Bilari. This research will aid in regional and urban planning, as well as agricultural management, in the years ahead. [ABSTRACT FROM AUTHOR]

From land degradation and desertification to cyclones and tropical storms, and so on, the repercussions of global change have become increasingly severe in recent years. Such environmental impacts require continuous assessment and monitoring. Thus, to study and analyse these impacts, a variety of time-series forecasting approaches have been developed, including statistical ones [i.e. Moving Average (MA), Auto-Regressive Integrated Moving Average (ARIMA), etc.], and machine-learning approaches such as Recurrent Neural Networks (RNN), Long-Short-Term Memory Network (LSTM) and Convolutional Neural Networks (CNN). In this study, accuracy of the most used forecasting approaches (i.e. MA, LSTM and Conv-LSTM) has been quantified, and three impacts (i.e. climate regions, land cover and satellite sensors) have been brought to light. Firstly, eight Mediterranean regions were selected (i.e. two hot arid regions, two cold arid regions, two regions with temperate hot summers and two regions with temperate warm summers) based on Köppen climate classification. Secondly, 654 hyperspectral images retrieved from three different satellites (i.e. Sentinel-2, Landsat-8 and MODIS) from 2016 to 2022 were computed in order to predict 29 vegetation biophysical indices (i.e. NDVI, GNDVI, EVI, CVI, etc.). Accordingly, more than 18,000 images were computed, resulting in 696 time series forecasted using the aforementioned approaches. Finally, 2088 forecasted time series have been determined, and their accuracy has been compared. As a result, Landsat-8 and Sentinel-2 images had the highest forecasting accuracy in the three approaches, reaching 86% of the computed indices being over 50% accurately predicted in LSTM model, while MODIS data had the highest forecasting accuracy only in MA model, with a percentage of 72% of the computed indices being over 50% accurately predicted. Furthermore, we have identified that region climate impacts vegetation forecasting accuracy. For instance, arid regions showed low accuracies across all models, while temperate regions showed higher accuracies in the Mediterranean region. [ABSTRACT FROM AUTHOR]

The semantic segmentation of satellite and UAV remote sensing imagery is pivotal for address exploration, change detection, quantitative analysis and urban planning. Recent advancements have seen an influx of segmentation networks utilizing convolutional neural networks and transformers. However, the intricate geographical features and varied land cover boundary interferences in remote sensing imagery still challenge conventional segmentation networks' spatial representation and long-range dependency capabilities. This paper introduces a novel U-Net-like network for UAV image segmentation. We developed a link aggregation Mamba at the critical skip connection stage of UNetFormer. This approach maps and aggregates multi-scale features from different stages into a unified linear dimension through four Mamba branches containing state-space models (SSMs), ultimately decoupling and fusing these features to restore the contextual relationships in the mask. Moreover, the Mix-Mamba module is incorporated, leveraging a parallel self-attention mechanism with SSMs to merge the advantages of a global receptive field and reduce modeling complexity. This module facilitates nonlinear modeling across different channels and spaces through multipath activation, catering to international and local long-range dependencies. Evaluations on public remote sensing datasets like LovaDA, UAVid and Vaihingen underscore the state-of-the-art performance of our approach. [ABSTRACT FROM AUTHOR]

In remote sensing applications, image classification algorithms normally require parameter optimization strategies to adapt to the complexities of the data and determine the model's parameters for optimal performance. However, a unique set of hyper-parameters is generally chosen, without acknowledging the presence of subsets of classes that exhibit different levels of spectral similarity. A model used to distinguish between well-separable classes might not be optimal for delineating other classes. This study explores a classification method that extends the standard One-Against-One (OAO) and One-Against-All (OAA) strategies for Support Vector Machines (SVMs), based on agglomerative hierarchical clustering of spectrally-similar classes. The main objective is to provide additional classification scenarios where more detailed per-class analyses can be performed, rather than simple comparisons of global accuracies, particularly for hard-to-distinguish classes. Clusters of classes closely located in the feature space are initially classified using an OAA strategy, while each cluster undergoes further subdivision using an OAO approach. This strategy aims to identify the most effective sequence for incorporating subgroups of classes into the tuning process. The experimental results indicate that the proposed method can be efficiently applied to classify remote sensing data, producing varying levels of enhancement in per-class accuracies at each stage of the agglomerative process. [ABSTRACT FROM AUTHOR]

Due to Sharkia population's rapid growth, the crises of atmospheric pollutants are causing significant concern. These circumstances have positive and negative impacts on both environment and climate. This research aims to investigate and evaluate changes in land and air characteristics of the governorate using remotely sensed data from several satellites (Landsat, Aqua, Terra, and Sentinel-5P) in conjunction with geographic information system (GIS) techniques. Three multitemporal Landsat images were acquired in July 2002, 2012, and 2022, to monitor land use/land cover (LULC) changes during the past 20 years. In addition, aerosol optical depth (AOD) data were obtained daily over 2022, and Particulate Matter (PM2.5) was calculated and then averaged seasonally. Sentinel-5P was utilized for CO, NO2, and SO2 retrievals. Results showed that the governorate was continuously subjected to spatiotemporal changes in LULC during the whole study period. It was mainly occupied by vegetation with a total area of 3711.64, 3851.05, and 3794.59 km2 in 2002, 2012, and 2022, respectively. Seasonal means of PM2.5 followed this order: spring (77.59 µg/m3) > summer (50.76 µg/m3) > winter (49.29 µg/m3) > autumn (42.71 µg/m3). However, the winter season recorded the highest mean values for all gases, CO (946.58 µg/m3), NO2 (9.70 µg/m3), and SO2 (13.07 µg/m3). It was concluded that the southern region of the governorate is more vulnerable to environmental stresses, which can reduce biodiversity and ultimately affect the study area's climate. In order to manage fragile ecosystems sustainably, this study recommends the creation of efficient land and air management information systems and regulations. [ABSTRACT FROM AUTHOR]

Responding to the shortcomings of China's civil remote sensing data in land cover classification, such as the difficulty of data acquisition and the low utilization rate, we used Landsat-8, China Orbita Zhuhai-1 hyperspectral remote sensing (OHS) data, and Landsat-8 + OHS data combined with band (red, green, and blue) and vegetation index features to classify land cover using maximum likelihood (ML), Mahalanobis distance (MD), and support vector machine (SVM). The results show that Landsat-8 + OHS data have the highest classification accuracy in SVM, with an overall accuracy of 83.52% and a kappa coefficient of 0.71, and this result is higher than that of Landsat-8 images and OHS images separately. In addition, the classification accuracy of OHS images was higher than that of Landsat-8 images. The results of the study provide a reference for the use of civil satellite remote sensing data in China. [ABSTRACT FROM AUTHOR]

The study of changes in land use and land cover (LULC) is helpful in the understanding of change and management of environmental sustainability. As a result, the central Telangana districts are the focus of this study since they are under stress from both natural and human-caused problems. The examination of LULC variations and predictions for the region between 2007–2015 and 2021–2030 employed Landsat OLI datasets, TerrSet, and geographic information system (GIS) tools. The LULC image is produced using a Landsat dataset and classified using a support vector machine (SVM). Then, consecutively to project future LULC change, LULC maps were constructed using the CA Markov model. The four stages included were change analysis, transition possibility, change forecasting, and model validation. It is found that the vegetation and the arid landscape are stressed and accumulating. The total accuracy was above 87 percent, and the kappa statistic measurement was above 85 percent with a three-year target. The study has found using the Markov chain land change modeler that Medchal–Malkajgiri district urban settlements will grow by 46, 37, and 26% from 2021–2030, 2030–2050, and 2050–2100, respectively. On other hand, the Warangal (Hanamkonda) observed 39, 45, and 30% between 2021–2030, 2030–2050, and 2050–2100, respectively, and Rangareddy districts observed 60, 24, and 12% between 2021–2030, 2030–2050, and 2050–2100, respectively. Given that urban areas are especially susceptible to flash flooding, this research will offer policymakers advice and a framework on behalf of planning city growth and managing the available resources judiciously with utmost planning. [ABSTRACT FROM AUTHOR]

The study aimed to determine how changes in land cover and surface water are being made using stratified objectoriented analysis based on the interpretation of remote sensing images. It is the first step toward managing the region’s annual land-use inventories projects. The study used Sentinel-2 images from 2019 through 2021 to delineate the changing urban land cover in the Ninh Kieu District, Can Tho City, Viet Nam. The study used QGIS software to interpret the images and eCognition software to classify the objects based on the NDBI, NDVI, and NDWI indices. The interpretation results were checked for the accuracy, and the land cover was changed over the years. The results show that urban land cover changes with the increase of urban land and the decrease of vegetation land used for urban land, while water surface area inwards decreased from 2019 to 2020 but increased in 2021. Maps of the current state of the urban land covers in the study area were delineated. The interpretation results contribute to the preliminary method by using satellite images for the annual land use inventory project in the region, even though some difficulties still exist and need to be modified. [ABSTRACT FROM AUTHOR]

The rate of population growth in Sukabumi Regency continues to grow, along with the increasing need for food. This population growth, combined with the constant changes in land cover can reduce the productivity of environment in providing natural capital for food availability. Therefore, this study aimed to examine the condition of ecosystem service provision for a decade in Sukabumi Regency due to changes in land cover. In general, the efficient use of remote sensing method and geographic information systems to monitor ecosystem services had received widespread recognition. Following this scenario, the current study used geospatial analysis with dasymetric method which was integrated with supply and demand formulas for ecosystem services provision, food availability, and remote sensing. Geographic information system was also used for land cover interpretation data. The results showed that three districts in Sukabumi Regency, namely Cicurug, Cibadak, and Cicantayan, had "exceeded" condition when the environmental condition already passed the threshold or were unable to support population's needs. Meanwhile, the other districts have "not exceeded" condition, when the environmental conditions were still able to fulfill the needs of population. Finally, the changes in agricultural land cover had a significant influence on the condition of ecosystem services. [ABSTRACT FROM AUTHOR]

Information on Land Use and Land Cover Map (LULCM) is essential for environment and socioeconomic applications. Such maps are generally derived from Multispectral Remote Sensing Images (MRSI) via classification. The classification process can be described as information flow from images to maps through a trained classifier. Characterizing the information flow is essential for understanding the classification mechanism, providing solutions that address such theoretical issues as "what is the maximum number of classes that can be classified from a given MRSI?" and "how much information gain can be obtained?" Consequently, two interesting questions naturally arise, i.e. (i) How can we characterize the information flow? and (ii) What is the mathematical form of the information flow? To answer these two questions, this study first hypothesizes that thermodynamic entropy is the appropriate measure of information for both MRSI and LULCM. This hypothesis is then supported by kinetic-theory-based experiments. Thereafter, upon such an entropy, a generalized Jarzynski equation is formulated to mathematically model the information flow, which contains such parameters as thermodynamic entropy of MRSI, thermodynamic entropy of LULCM, weighted F1-score (classification accuracy), and total number of classes. This generalized Jarzynski equation has been successfully validated by hypothesis-driven experiments where 694 Sentinel-2 images are classified into 10 classes by four classical classifiers. This study provides a way for linking thermodynamic laws and concepts to the characterization and understanding of information flow in land cover classification, opening a new door for constructing domain knowledge. [ABSTRACT FROM AUTHOR]

The research presented in this study introduces a novel methodology for delineating flood-prone regions within the Kosi River Basin, utilizing a multi-criteria decision-making approach. This method integrates multi-criteria analysis, Geographical Information Systems (GIS), and Remote Sensing (RS). The specific process involves the creation of flood susceptibility maps based on five crucial factors: rainfall, land use/cover, slope, drainage density, and distance from the river. Expert judgments were incorporated and translated into weighted values to ascertain the relative significance of each factor in determining flood susceptibility. Weight calculations were performed using the Fuzzy Analytic Hierarchy Process (FAHP). The findings of this study indicate that across all 10 districts in the region, a varying degree of land area is classified as high-risk, with Madhubani displaying the highest percentage of land area categorized as of very high-risk. Key challenges include data accuracy and model generalization, with potential applications in other flood-prone areas. This approach not only improves the precision of flood susceptibility mapping, but also offers valuable insights for disaster management and planning in areas with limited data availability. [ABSTRACT FROM AUTHOR]

Uncontrolled human activity and nature are causing the deterioration of Saint Martin Island, Bangladesh's only tropical island, necessitating sustainable land use strategies and ecological practices. Therefore, the present study measures the land use/cover transition from 1974 to 2021, predicts 2032 and 2042, and constructs the spatiotemporal features of the Landscape Ecological Risk Index based on land use changes. The study utilized Maximum Likelihood Classification (MLC) on Landsat images from 1974, 1988, 2001, 2013, and Sentinel 2B in 2021, achieving ≥ 80% accuracy. The MLP-MC approach was also used to predict 2032 and 2042 LULC change patterns. The eco-risk index was developed using landscape disturbance and vulnerability indices, Bayesian Kriging interpolation, and spatial autocorrelations to indicate spatial clustering. The research found that settlements increased from 2.06 to 28.62 ha between 1974 and 2021 and would cover 41.22 ha in 2042, causing considerable losses in agricultural areas, waterbodies, sand, coral reefs, and vegetation. The area under study showed a more uniform and homogenous environment as Shannon's diversity and evenness scores decreased. The ecological risk of Saint Martin Island increased from 4.31 to 31.05 ha between 1974 and 2042 due to natural and human factors like erosion, tidal bores, population growth, coral mining, habitat destruction, and intensive agricultural practices and tourism, primarily in Nazrul Para, Galachipa, and Western Dakhin Para. The findings will benefit St. Martin Island stakeholders and policymakers by providing insights into current and potential landscape changes and land eco-management. [ABSTRACT FROM AUTHOR]

Groundwater depletion and water scarcity are pressing issues in water-limited regions worldwide, including Pakistan, where it ranks as the third-largest user of groundwater. Lahore, Pakistan, grapples with severe groundwater depletion due to factors like population growth and increased agricultural land use. This study aims to address the lack of comprehensive groundwater availability data in Lahore's semi-arid region by employing GIS techniques and remote sensing data. Various parameters, including Land Use and Land Cover (LULC), Rainfall, Drainage Density (DD), Water Depth, Soil Type, Slope, Population Density, Road Density, Normalized Difference Vegetation Index (NDVI), Normalized Difference Built-Up Index (NDBI), Moisture Stress Index (MSI), Water Vegetation Water Index (WVWI), and Land Surface Temperature (LST), are considered. Thematic layers of these parameters are assigned different weights based on previous literature, reclassified, and superimposed in weighted overlay tool to develop a groundwater potential zones index map for Lahore. The groundwater recharge potential zones are categorized into five classes: Extremely Bad, Bad, Mediocre, Good, and Extremely Good. The groundwater potential zone index (GWPZI) map of Lahore reveals that the majority falls within the Bad to Mediocre recharge potential zones, covering 33% and 28% of the total land area in Lahore, respectively. Additionally, 14% of the total area falls under the category of Extremely Bad recharge potential zones, while Good to Extremely Good areas cover 19% and 6%, respectively. By providing policymakers and water supply authorities with valuable insights, this study underscores the significance of GIS techniques in groundwater management. Implementing the findings can aid in addressing Lahore's groundwater challenges and formulating sustainable water management strategies for the city's future. [ABSTRACT FROM AUTHOR]

Monitoring the land use/land cover (LU/LC) changes that have occurred with rapid population growth and urbanization since the Industrial Revolution is important for the optimal configuration of landscape patterns and ensuring the sustainability of ecological functions. Spatiotemporal dynamic pattern of LU/LC change using high-resolution land use data is an indicator to evaluate the landscape ecological risk through landscape pattern index analysis. In this study, the landscape ecological risk index (LERi) based on LU/LC change was calculated using remote sensing images of Landsat TM (Thematic Mapper) and OLI (Operational Land Imager) Rdata of a Gediz Mainstream Sub-basin in Turkiye between 1992 and 2022, and the spatial distribution regularity of LERi values was determined with spatial statistical analysis. According to the results, it was determined that the LERi values of the study area changed by 45% in 30 years. The highest change is in the very high-risk class, with an increase of 10.96%, and the least change occurred in the very low-risk class, with a decrease of 1.29%. According to the obtained statistical analysis results, it was determined that the global spatial autocorrelation values analyzed at different grain levels showed positive autocorrelation for both years and that the LERi values tended to have strong spatial clustering. As a result, it is emphasized that strict control measures should be taken for areas showing High-High (HH) autocorrelation type located in the southeast and north-southwest line of the study area at the local level, and ecological restoration applications should be given priority in these areas. [ABSTRACT FROM AUTHOR]

Jointly building the Economic Corridor of the Eurasia Continent (ECEC), which is one of the most important parts of the Silk Road Economic Belt, is a pivotal initiative for fostering regional development. Forests, which serve as a green foundation of economic resilience, underpin this effort. However, there is an imbalance in ecological status due to differences in natural resources and the social economy along the economic corridor. This imbalance has led to alterations in landscapes, yet the specific changes and their underlying relationships are still much less understood. Here, we quantitatively detected changes in the forest landscape and its ecological efforts over the ECEC via widespread, satellite-based and long-term land cover maps released by the European Space Agency (ESA) Climate Change Initiative (CCI). Specifically, the coupling between changes in forest coverage and landscape patterns, e.g., diversity, was further examined. The results revealed that forest coverage fluctuated and declined over the ECEC from 1992 to 2018, with an overall reduction of approximately 9784.8 km2 (i.e., 0.25%). Conversions between forests and other land cover types were widely observed. The main displacements occurred between forests and grasslands/croplands (approximately 48%/21%). Moreover, the landscape diversity in the study area increased, as measured by the effective diversity index (EDI), during the study period, despite obvious spatial heterogeneity. Notably, this pattern of landscape diversity was strongly associated with forest displacement and local urban development through coupling analysis, consequently indicating increasing fragmentation rather than biological diversity. This study highlights the coupled relationship between quantitative and qualitative changes in landscapes, facilitating our understanding of environmental protection and policy management. [ABSTRACT FROM AUTHOR]

The present work was carried out on River Yamuna passing through Delhi-National Capital Region (Delhi-NCR) of India for a stretch of about 125 km to assess the fluvial characteristics in the highly human-dominated region for a period of 200 years with the help of historical maps, topographical maps and satellite data with integration of geographic information system (GIS) environment. Erosion, deposition and channel stability data were analyzed for 1977–1986, 1986–1996, 1996–2006, 2006–2016, and 2016–2022 period. Digital Shoreline Analysis System (DSAS) application was utilized to quantify river channel movement with average channel migration of 22.8 m/year. Westward migration is 4149.2 m maximum in length and eastward migration is about 4083.8 m. The river has migrated a total of 32.26 km2 of area during 1955-2022. The findings indicate that various human activities, such as engineering structures, sand mining, embankments, urbanization, land use/land cover changes, and canal networks, have significantly impacted the river. The DSAS application was also used to predict the position of river channel centerline in future for 2032 and 2042 with channel length of 132.5 and 141.6 km respectively. [ABSTRACT FROM AUTHOR]

Introduction: Groundwater demand has been considerably heightened due to rapid urban growth, specifically in arid areas that rely primarily on groundwater. This study aims to utilize remote sensing and aeromagnetic data, combined with the Analytic Hierarchy Process (AHP) based GIS, to evaluate potential groundwater zones in the Sohag area, Egypt. Methods: Nine thematic layers, including soil moisture, rainfall, lithology, normalized difference vegetation index (NDVI), drainage density, lineament density, slope, and land use/land cover, were developed using various remote sensing datasets. Besides the remote sensing-derived thematic layers, a geophysics-derived thematic layer represented by the RTP aeromagnetic map was included. The aeromagnetic data were analyzed and interpreted to outline the subsurface structure affecting groundwater storage and flow. Also, the aeromagnetic data analysis helps estimate the basement depth that constitutes the Nubian Aquifer's base and identifies regions with considerable thick sedimentary deposits and significant water reserves. Results and discussion: The groundwater potentiality map was consistent with production wells in the area, and sites for drilling new wells were predicted, especially in the Nile Valley around the Tahta, El-Hamimia, and west Sohag cities. The most promising sites are clustered along the Nile Valley, and the study area's northwestern and northeastern parts. The results indicate that the predominant magnetic structural trends are NW-SE, NE-SW, N-S, and E-W, which contribute to the formation of a series of subsurface horsts (H) and grabens (G). Three main basins (A, B, and C) were identified as the most profound areas. These basins represent the most promising areas for groundwater accumulation, making them attractive for future hydrogeological exploration. This integrated approach strongly offers a powerful and effective tool to assist in developing an appropriate plan to manage groundwater in arid regions. [ABSTRACT FROM AUTHOR]

Appropriate environment management requires an understanding of how mining activity alters environmental characteristics and how those changes affect an area. Therefore, to reduce the adverse effects of mining activity on the land, it becomes crucial to have relevant information about responses to environmental degradation. This study aims to assess the impact of semi-mechanised and artisanal mining activities on the land cover using remote sensing data and photogrammetric analysis, in the Mbale locality, Northern Cameroon. For this purpose, the maximum likelihood classification algorithm of the supervised classification method combined with field surveys was used to map environmental changes, based on Sentinel-2 images of 2019, 2021, and 2023. Normalized Difference Vegetation Index (NDVI), Normalised Difference Water Index (NDWI), Brithness index (BI), and Soil crust index (SCI), were calculated to assess changes in vegetation, bare soil, water body, and exploited area. The orthophoto obtained from photogrammetric processing was performed to outline river network change through visual interpretation techniques and to calculate the volume of pits created by mining. The result of classified images indicated that vegetation cover decreased by 11.74% over the studied years. However, bare soil and exploited areas increased by 9.2% and 5.4% respectively. The calculated spectral indices show that between 2019 and 2023 the locality of Mbale considerably lost its vegetation cover, in favor of bare soil. The color of the soil and the granulometric size of the topsoil have also changed. The photogrammetry analysis highlighted the deviation of the main river and estimated the volume of pits created by mining activity to 22188.7 m3. The mining activities caused a loss of the vegetation cover, generated big pits, and multiple deviations of the Lom River from its natural course, which have a substantial negative influence on the ecosystem. Such data can be used for long-term environmental management, reclamation and rehabilitation monitoring, and mining area restoration. [ABSTRACT FROM AUTHOR]

The deep pools are considered vital habitats for both aquatic and terrestrial biodiversity in arid and semi-arid rivers. These 'refugia' habitats sustain the aquatic biodiversity of local and regional importance when water flow ceases. Banas is an ecologically unique and non-perennial river in the Ganga Basin originating from the Aravalli Range and flowing through the semi-arid region of Rajasthan, India. This study maps and characterises the deep pools in the water stressed river using Sentinel-2 satellite data (2019–2022). Mapping and analysis were done using geospatial tools and field data. The composite map reported 2.18 km2 (0.6% of the total area) and 72.42 km2 (19.0% of the total area) of permanent water spread in the floodplain and reservoirs of Banas River, respectively with seasonal variations. A total of 558 contiguous habitats with varying sizes (50 to 314,422 m2) were delineated and most of them were located in the downstream of Bisalpur Dam especially along the river meandering. The composition of the area under different land use land cover classes in the riparian zone varied across the deep pools with medium land use intensity. The high proportion of vegetation and cropland near and far from the riparian buffer indicated existence of the natural and agrarian landscapes, respectively. The indications of various ecosystem services by the deep pools necessitate spatial quantification. Additionally, impact of the various anthropogenic threats on aquatic habitats recommends measures for habitat restoration and conservation planning of Banas River. [ABSTRACT FROM AUTHOR]

Accurate and timely global evapotranspiration (ET) data are crucial for agriculture, water resource management, and drought forecasting. Although numerous satellite-based ET products are available, few offer near-real-time data. For instance, products like NASA's ECOsystem Spaceborne Thermal Radiometer Experiment mounted on the International Space Station (ECOSTRESS) and MOD16 face challenges such as uneven coverage and delays exceeding 1 week in data availability. In this study, we refined the Variation of the Standard Evapotranspiration Algorithm (VISEA) by fully integrating satellite-based data, e.g., European Centre for Medium-Range Weather Forecasts ERA5-Land shortwave radiation (which includes satellite remote sensing data within its assimilation system) and MODIS land surface data (which include surface reflectance, temperature and/or emissivity, land cover, vegetation indices, and albedo as inputs). This enables VISEA to provide near-real-time global daily ET estimates with a maximum delay of 1 week at a resolution of 0.05°. Its accuracy was assessed globally using observation data from 149 flux towers across 12 land cover types and comparing them with five other satellite-based ET products and Global Precipitation Climatology Centre (GPCC) data. The results indicate that VISEA provides accurate ET estimates that are comparable to existing products, achieving a mean correlation coefficient (R) of about 0.6 and an RMSE of 1.4 mmd-1. Furthermore, we demonstrated VISEA's utility in drought monitoring during a drought event in the Yangtze River basin in 2022 in which ET changes correlated with precipitation. The near-real-time capability of VISEA is, thus, especially valuable in meteorological and hydrological applications for coordinating drought relief efforts. The VISEA ET dataset is available at 10.11888/Terre.tpdc.300782 (Huang, 2023a). [ABSTRACT FROM AUTHOR]

Recent advancements in deep learning have significantly improved the performance of remote sensing scene classification, a critical task in remote sensing applications. This study presents a new aerial scene classification model, the Multi-Scale Swin–CNN Aerial Classifier (MSCAC), which employs the Swin Transformer, an advanced architecture that has demonstrated exceptional performance in a range of computer vision applications. The Swin Transformer leverages shifted window mechanisms to efficiently model long-range dependencies and local features in images, making it particularly suitable for the complex and varied textures in aerial imagery. The model is designed to capture intricate spatial hierarchies and diverse scene characteristics at multiple scales. A framework is developed that integrates the Swin Transformer with a multi-scale strategy, enabling the extraction of robust features from aerial images of different resolutions and contexts. This approach allows the model to effectively learn from both global structures and fine-grained details, which is crucial for accurate scene classification. The model's performance is evaluated on several benchmark datasets, including UC-Merced, WHU-RS19, RSSCN7, and AID, where it demonstrates a superior or comparable accuracy to state-of-the-art models. The MSCAC model's adaptability to varying amounts of training data and its ability to improve with increased data make it a promising tool for real-world remote sensing applications. This study highlights the potential of integrating advanced deep-learning architectures like the Swin Transformer into aerial scene classification, paving the way for more sophisticated and accurate remote sensing systems. The findings suggest that the proposed model has significant potential for various remote sensing applications, including land cover mapping, urban planning, and environmental monitoring. [ABSTRACT FROM AUTHOR]

The quality and level of groundwater tables have rapidly declined because of intensive pumping in Punjab (Pakistan). For sustainable groundwater supplies, there is a need for better management practices. So, the identification of potential groundwater recharge zones is crucial for developing effective management systems. The current research is based on integrating seven contributing factors, including geology, soil map, land cover/land use, lineament density, drainage density, slope, and rainfall to categorize the area into various groundwater recharge potential zones using remote sensing, geographic information system (GIS), and analytical hierarchical process (AHP) for Punjab, Pakistan. The weights (for various thematic layers) and rating values (for sub-classes) in the overlay analysis were assigned for thematic layers and then modified and normalized using the AHP. The result indicates that about 17.88% of the area falls under the category of very high groundwater potential zones (GWPZs). It was found that only 12.27% of the area falls under the category of very low GWPZs. The results showed that spatial technologies like remote sensing and geographic information system (GIS), when combined with AHP technique, provide a robust platform for studying GWPZs. This will help the public and government sectors to understand the potential zone for sustainable groundwater management. [ABSTRACT FROM AUTHOR]

Land-use and land-cover change (LULCC) is a critical environmental issue that has significant effects on biodiversity, ecosystem services, and climate change. This study examines the land-use and land-cover (LULC) spatiotemporal dynamics across a three-decade period (1998–2023) in a district area. In order to forecast the LULCC patterns, this study suggests a hybrid strategy that combines the random forest method with multi-layer perceptron (MLP) and Markov chain analysis. To predict the dynamics of LULC changes for the year 2035, a hybrid technique based on multi-layer perceptron and Markov chain model analysis (MLP-MCA) was employed. The area of developed land has increased significantly, while the amount of bare land, vegetation, and forest cover have all decreased. This is because the principal land types have changed due to population growth and economic expansion. This study also discovered that between 1998 and 2023, the built-up area increased by 468 km2 as a result of the replacement of natural resources. It is estimated that 25.04% of the study area's urbanization will increase by 2035. The performance of the model was confirmed with an overall accuracy of 90% and a kappa coefficient of around 0.89. It is important to use advanced predictive models to guide sustainable urban development strategies. The model provides valuable insights for policymakers, land managers, and researchers to support sustainable land-use planning, conservation efforts, and climate change mitigation strategies. [ABSTRACT FROM AUTHOR]

Long-term monitoring studies on the transition of different land cover units to barren areas are crucial to gain a better understanding of the potential challenges and threats that land surface ecosystems face. This study utilized the Moderate Resolution Imaging Spectroradiometer (MODIS) land cover products (MCD12C1) to conduct geospatial analysis based on the maximum extent (MaxE) concept, to assess the spatiotemporal changes in barren areas from 2001 to 2022, at global and continental scales. The MaxE area includes all the pixels across the entire period of observations where the barren land cover class was at least once present. The relative expansion or reduction of the barren areas can be directly assessed with MaxE, as any annual change observed in the barren distribution is comparable over the entire dataset. The global barren areas without any land change (UA) during this period were equivalent to 12.8% (18,875,284 km2) of the global land surface area. Interannual land cover changes to barren areas occurred in an additional area of 3,438,959 km2 (2.3% of the global area). Globally, barren areas show a gradual reduction from 2001 (91.1% of MaxE) to 2012 (86.8%), followed by annual fluctuations until 2022 (88.1%). These areas were mainly interchanging between open shrublands and grasslands. A relatively high transition between barren areas and permanent snow and ice is found in Europe and North America. The results show a 3.7% decrease in global barren areas from 2001 to 2022. Areas that are predominantly not barren account for 30.6% of the transitional areas (TAs), meaning that these areas experienced short-term or very recent transitions from other land cover classes to barren. Emerging barren areas hotspots were mainly found in the Mangystau region (Kazakhstan), Tibetan plateau, northern Greenland, and the Atlas Mountains (Morocco, Tunisia). [ABSTRACT FROM AUTHOR]

An accurate assessment of the spatial–temporal variations in regional net ecosystem productivity (NEP), water use efficiency (WUE), and carbon use efficiency (CUE) are vital for understanding the water–carbon cycle. We analyzed the spatial–temporal patterns of the NEP, WUE, and CUE in the middle reaches of the Yellow River (MRYR) from 2001 to 2022, and the factors that influenced them using remote sensing data, NEP estimation models, and various statistical methods. The results indicate that the recovery of the ecosystem in the MRYR is a result of the combined effects of climate change and human activities. Climate change in the MRYR led to warming and humidification from 2001 to 2022. The NEP, WUE, and CUE were characterized by increasing trends, with average growth rates of 7.75 gC m−2a−1, 0.012 gC m−2 mm−1a−1, and 0.009a−1, respectively. For four vegetation types, the interannual rates of change were, in descending order, grassland, cropland, shrubs, and forest. Spatially, the NEP, WUE, and CUE showed significant regional heterogeneity, increasing from the northwest to the southeast. Based on an analysis of the interannual anomalies, precipitation accumulation contributed to carbon sink accumulation. The correlation of the NEP, WUE, and CUE with the drought severity index (DSI) was high, and their correlation with precipitation showed latitudinal zonality, which suggests that precipitation (PRE) is the main climatic factor influencing the water–carbon cycle in the MRYR rather than temperature (TEM). There were 67,671.27 km2 of land that changed use during 2001–2022, and 15.07 Tg of NEP was added to these areas. [ABSTRACT FROM AUTHOR]

Remote sensing allows us to conduct large-scale scientific studies that require extensive mapping and the amalgamation of numerous images. However, owing to variations in radiation, atmospheric conditions, sensor perspectives, and land cover, significant color discrepancies often arise between different images, necessitating color consistency adjustments for effective image mosaicking and applications. Existing methods for color consistency adjustment in remote sensing images struggle with complex one-to-many nonlinear color-mapping relationships, often resulting in texture distortions. To address these challenges, this study proposes a convolutional neural network-based color consistency method for remote sensing cartography that considers both global and local color mapping and texture mapping constrained by the source domain. This method effectively handles complex color-mapping relationships while minimizing texture distortions in the target image. Comparative experiments on remote sensing images from different times, sensors, and resolutions demonstrated that our method achieved superior color consistency, preserved fine texture details, and provided visually appealing outcomes, assisting in generating large-area data products. [ABSTRACT FROM AUTHOR]

The integration of multi-source remote sensing data, bolstered by advancements in deep learning, has emerged as a pivotal strategy for enhancing land use and land cover (LULC) classification accuracy. However, current methods often fail to consider the numerous prior knowledge of remote sensing images and the characteristics of heterogeneous remote sensing data, resulting in data loss between different modalities and the loss of a significant amount of useful information, thus affecting classification accuracy. To tackle these challenges, this paper proposes a LULC classification method based on remote sensing data that combines a Transformer and cross-pseudo-siamese learning deep neural network (TCPSNet). It first conducts shallow feature extraction in a dynamic multi-scale manner, fully leveraging the prior information of remote sensing data. Then, it further models deep features through the multimodal cross-attention module (MCAM) and cross-pseudo-siamese learning module (CPSLM). Finally, it achieves comprehensive fusion of local and global features through feature-level fusion and decision-level fusion combinations. Extensive experiments on datasets such as Trento, Houston 2013, Augsburg, MUUFL and Berlin demonstrate the superior performance of the proposed TCPSNet. The overall accuracy (OA) of the network on the Trento, Houston 2013 and Augsburg datasets is of 99.76%, 99.92%, 97.41%, 87.97% and 97.96%, respectively. [ABSTRACT FROM AUTHOR]

Optical remote sensing imagery can serve as a crucial data source in numerous fields. However, the susceptibility to weather conditions (like clouds and rain) can render the cloud contamination, leading to the significant influence on optical images. The image quality can be required to restore the spectral information in cloud-contaminated regions. In this study, the spectral information was reconstructed in the cloud-contaminated regions using an object-oriented approach. Synthetic aperture radar (SAR) data was integrated with the radar signals and land cover information, in order to serve as the constraints for the reconstruction. The distance matrix was calculated on the SAR images, and then transplant onto the optical images for the replacement. The pixel-level matching accuracy was enhanced to maximize the retention of spectral information from the original images. A solid foundation was provided for the accurate extraction of subsequent land surface information. The performance was also evaluated to verify the improved model. The comparative experiments were conducted with the traditional pixel replacement, weighted linear regression, and curvature-driven. Visual inspection was carried out over various scenes. The results revealed that the superior quality was achieved in the imagery reconstruction in the thick cloud areas and cloud shadows. Particularly, the consistency was maintained among various land cover pixels in the reconstructed areas by the object-oriented method. The boundaries and areas were accurately delineated in the various types of fine-grained land cover. Quantitative analysis was conducted to validate using some metrics, such as correlation coefficient, root mean square error (RMSE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM). The spectral characteristic curves of the pixels were reconstructed to closely match those of the reference images, with the correlation coefficients exceeding 0.99. Notably, significant improvement was observed in the spectral curve recovery of vegetation pixels with the enhanced relative to the rest. The radar vegetation indices were optimized to employ an object-oriented matching mechanism during pixel reconstruction. The traditional pixel replacement model was optimized effectively. Furthermore, the band reflectance was closely resembled that of the reference images. Among the five bands of cloud-contaminated images after reconstruction, the highest correlation coefficients were observed with Band 2 and Band 3 reaching 0.925 and 0.922, respectively, and RMSE values of 0.009 and 0.007, respectively. The reconstructed images shared the minimal quality loss and high similarity to the reference images, with the highest PSNR and SSIM values across all bands. Specifically, the structural similarity values of the reconstructed Band 2 and Band 3 images were 0.902 and 0.910, respectively, with the PSNR values reaching 31.978 and 33.173, respectively. The superior image quality was obtained similar to reference images, which was consistence with spectral characteristics. The experimental results were used to the qualitatively and quantitatively validated the effectiveness of the reconstruction. The object-oriented approach can be expected to reconstruct the spectral information in the cloud-contaminated regions. The remotely sensed images were consistently produced with the higher spectral reconstruction accuracy in the different types of land cover and spectral bands. However, the replacing pixel spectral information was relied primarily on from single-image data. Some limitations was matched the effectiveness between cloudy and cloud-free regions, when reconstructing the large-scale, thick cloud areas in single remote sensing images. Additionally, the global searching can fail to the optimal reconstruction for the locally contaminated cloud pixels. Therefore, further investigation is required to determine the optimal local search range for the pixel matching using different extents of cloud contamination. [ABSTRACT FROM AUTHOR]

Pansharpening is an important technology for obtaining high-resolution multispectral (HRMS) images by fusing low-resolution multispectral (LRMS) images and high-resolution panchromatic (PAN) images. Although many pansharpening models have emerged by taking advantage of deep learning (DL) technology, there remains a pressing need to further assess pansharpening accuracy and stability when LRMS images with complex land-cover types. What's more, these models often overlook the exploitation of PAN images' inherent high-frequency information. To address these issues, we propose a pansharpening model combining multi-level and multi-scale network architectures. The multi-level network architecture is used to build spatial-spectral dependence on LRMS-PAN pairs, and strengthen the network's feature capture capability by keeping the multi-level texture details. The multi-scale architecture is subsequently used to extract the spatial structure and deep texture of the PAN images at different scales. Downsampled experiments and real experiments in four standard datasets show that the proposed model achieves a state-of-the-art performance. [ABSTRACT FROM AUTHOR]

One of the most prominent factors contributing to global landscape cover change is land use and land cover (LULC). Such changes are the outcome of several interrelated factors. This study assessed trends in land use and land cover within Hirpora from 1992 to 2006 and 2021 utilizing remote sensing data and satellite pictures from Landsat 5-TM in 1992, Landsat 7-enhanced TM in 2006, and Landsat 8-OLI in 2021. The images were analyzed using ArcGIS v10.1 and ERDAS Imagine v14. From 1992 to 2021, there was a substantial shift in land use trends in a few targeted classes, with snow cover losing the most (33.76%), followed by dense forests (6.80%), and economically grasslands (21.46%), barren/rocky (11.43%), and scrub (5.41%) gaining the most. During the observation period, moderately dense forests and build-up increased in area by 4.50 and 0.35%, respectively. Due to large-scale human intrusion and habitat fragmentation, the Hirpora Wildlife Sanctuary has swiftly transitioned from a natural to a cultural landscape. As part of the LULC change process, natural, demographic, and economic factors have impacted land and had environmental implications. [ABSTRACT FROM AUTHOR]

• Improving Coastline Edge Detail: EDS module enhances edge fitting. • Bridging Semantic Gap: CSDS and AFM collaboration enhances semantic information. • Outperforms with mIoU: CSAFNet achieves remarkable 96.72% mIoU value. The extraction of coastlines from remote sensing images is vital for promoting sustainable development in coastal areas, conserving marine environments, strengthening disaster response capabilities, and supporting scientific research. However, current coastline detection approaches using remote sensing face challenges related to resolution, terrain, boundary, and data, requiring accurate solutions for reliability. Here, we introduce the Collaborative Supervision and Attention Fusion (CSAFNet) model for pixel-level sea-land segmentation, with a primary goal of improving the accuracy of coastline extraction. The model integrates the Edge Deep Supervision (EDS) module to enhance coastline edge detail fitting. Additionally, the Collaborative Semantic Deep Supervision (CSDS) module and Attention Fusion Module (AFM) collaborate to bridge the semantic gap between different hierarchical features, resulting in a more precise and detailed delineation of coastlines. Experimental validation on the publicly available SLSD 1 1 The dataset was initially labeled as "sea-land segmentation data" by the authors. For ease of discussion, we will adopt the practice of using the initial letters of each word as an abbreviation. dataset has demonstrated superiority over various advanced methods, with an impressive mIoU value of 96.72%. Through simple optimization, detailed and rich coastlines can be extracted, validating the feasibility of coastline extraction. [ABSTRACT FROM AUTHOR]

• Landsat bands alone may not give adequate quality of LULC mapping in a heterogeneous landscape. • Remote sensing-derived indices and auxiliary data improve the accuracy of LULC mapping. • LULC maps classified under different classification algorithms have shown a significant discrepancy. Land use/land cover (LULC) information is regarded as one of the most important variables in global change studies. Several LULC classification algorithms are available with different levels of accuracy in LULC mapping under different geographical setups. Furthermore, different remote sensing (RS) based indices and auxiliary data have been reported to have contributed to the accuracy of LULC mapping. Hence, the main aim of this study was to compare the different ML algorithms for the LULC mapping and explicitly assess the potential contribution of different RS-derived indices and auxiliary data on classification accuracy in the Upper Tekeze-Atbarah (UTA) river basin in Ethiopia. The overall classifiers evaluation result revealed that random forest (RF) performed best and followed by K-Nearest Neighbour (KNN), Support Vector Machine (SVM), Artificial Neural Network (ANN), Maximum Likelihood Classification (MLC), and Classification and Regression Tree (CART). Compared to RF, the MLC method showed less accuracy by 16.2% (19.7%) in OA (kappa coefficient) respectively, which showed that the currently emerging ML classifiers contributed to an improvement in LULC mapping. The use of RS-derived indices and DEM as predictors together with the Landsat bands improved the overall accuracy and kappa coefficient by 5.5% and 7% respectively. Furthermore, the discrepancy in the classified LULC map from RF and other classifiers was compared at pixel-level and the result showed a considerable disagreement between the LULC maps. For instance, when the LULC map from RF and KNN are compared, only 80.6% of the area has the same LULC condition. This result reflects the potential discrepancy that could exist in LULC maps from different classifiers. The hydrological simulation under each of the classified map also resulted in a considerable difference in the major water balance components. In a conclusion, given the considerable difference in classifier performance and the potential propagation of errors in LULC maps to further applications such as water resources assessment, it is highly recommended to explicitly evaluate the available classifiers. Users should also harness the power of RS indices and auxiliary datasets for improved LULC mapping. [ABSTRACT FROM AUTHOR]