Multi-temporal Images Research Articles

Coastal and landuse changes of Burullus Lake, Egypt: A comparison using Landsat and Sentinel-2 satellite images

Monitoring water bodies such as lakes is expensive, but remote sensing made it cost-effective. The extensive anthropogenic activities negatively impact the water lakes everywhere. Moreover, the Egyptian coastal lakes are not an exception. This paper investigates changes in land use/land cover and detects coastal changes of Buullus Lake, ranked to be the second-largest Egyptian northern Mediterranean coastal lake, using Remote Sensing and Geographic Information Systems (GIS) techniques. Multitemporal Landsat satellite images from 1984 to 2019 have been collected and processed where Landsat and Sentinel-2A images for 2019 have been used for comparison purposes. In addition field visits have been conducted to verify the obtained results. The maximum likelihood supervised classification approach was shown to be more accurate in monitoring changes in the Lake's water bodies. Classes of the supervised classification included eight classes; seawater, lake water, floating vegetation, sand, urban, agricultural areas, fish farms, and drainage canals. Analysis of the satellite data indicated that: between 1984 and 2019, the lake's water bodies and floating vegetation area were reduced by 16 % and 52 %, respectively. While agricultural lands increased by (648 km2). In addition, there has been a significant growth in the area of fish farms (290 km2). The resulted updated spatiotemporal maps and extracted water features assist decision-makers in taking the necessary steps to stop or minimize anthropogenic activities. These steps will help to restore or maintain the lake to host the government planned activities to achieve the Sustainable Development Goals (SDGs) on the lake level.

Multi-temporal Satellite Images and Combination of Multi-Classification Approaches for Evaluation of Changes in Forest Cover: A case study

The primary objective of this study is to develop a method to monitor the changes in forest cover using three different techniques, namely Spatial Data Mining (SDM), supervised and unsupervised classification, and Geographical Information System (GIS) overlay processing methods. The secondary objective was to monitor the land use changes using multi-temporal satellite images (Landsat 8 images of 2016 and 2020). The study area is located in Nuwara Eliya, Sri Lanka. The first and second approaches consisted of supervised and unsupervised classification approaches. The use of a combination of two classification mapping approaches provides better results compared with the use of a single classification, which is a novelty of the present study. The GIS overly processing technique combined the two maps obtained from supervised and unsupervised classifications to develop an improved map for land use changes. The improved map was reclassified and converted into ASCII (American Standard Code for Information Interchange) format, which was then pre-processed to convert the data set into the suitable format for the SDM modeling. The converted format was used to implement SDM modeling with the support of the Clustering-outline detection algorithm. The overall accuracy of the Landsat 8 images was 94.6. Results revealed that the forest cover extent was diminished by 5.28% in the study area between 2016 and 2020. The ground measurement was done with the help of the forest department to verify the results. The study revealed that the forest area decreased, and farming lands increased due to intensive agriculture. Future studies would be necessary to determine the validity and suitability of the developed model for other climatic zones in the country.

Unsupervised change detection of man-made objects using coherent and incoherent features of multi-temporal SAR images

Constrained by complex imaging mechanism and extraordinary visual appearance, change detection with synthetic aperture radar (SAR) images has been a difficult research topic, especially in urban areas. Although existing studies have extended from bi-temporal data pair to multi-temporal datasets to derive more plentiful information, there are still two problems to be solved in practical applications. First, change indicators constructed from incoherent feature only cannot characterize the change objects accurately. Second, the results of pixel-level methods are usually presented in the form of the noisy binary map, making the spatial change not intuitive and the temporal change of a single pixel meaningless. In this study, we propose an unsupervised man-made objects change detection framework using both coherent and incoherent features derived from multi-temporal SAR images. The coefficients of variation in time-series incoherent features and the man-made object index (MOI) defined with coherent features are first combined to identify the initial change pixels. Afterwards, an improved spatiotemporal clustering algorithm is developed based on density-based spatial clustering of applications with noise (DBSCAN) and dynamic time warping (DTW), which can transform the initial results into noiseless object-level patches, and take the cluster center as a representative of the man-made object to determine the change pattern of each patch. An experiment with a stack of 10 TerraSAR-X images in Stripmap mode demonstrated that this method is effective in urban scenes and has the potential applicability to wide area change detection.

Remote estimates of soil organic carbon using multi-temporal synthetic images and the probability hybrid model

Soil organic carbon (SOC) plays a key role in soil function, ecosystem services, and the global carbon cycle. Digital SOC mapping is essential for agricultural production management. Digital SOC mapping based on multi-source remote sensing data has been integrated well into prediction models and methodological approaches on different mapping scales. However, the mixed use of synthetic images and the application of the hybrid model considering the soil classification probability are rare. Here, we propose a probability hybrid model to estimate and map SOC content and distribution. Multi-temporal synthetic images were used to build the probability hybrid model. One hundred forty topsoil samples were collected in Suihua, a city in a typical black soil region in China. Cloud-free Sentinel-2 images were acquired from the bare soil periods between 2018 and 2021. Minimum, maximum, mean, and median synthetic images were calculated using single images from the same period. The random forest and support vector machine models were used for discriminating soil class and calculating soil classification probability, and then random forest regression model was applied to SOC mapping. Soil class mapping and classification probability were performed for Phaeozems, Chernozems, and Cambisols of the World Reference Base for Soil Resources (WRB) based on soil texture, climatic factors, and the normalized differential vegetation index (NDVI). Based on soil class mapping results, global models using single temporal images and multi-temporal images, the hybrid model using multi-temporal images for the three soil classes as well as a probability hybrid model using multi-temporal images were built and compared their SOC predictions. Our study showed that (1) Phaeozems, Chernozems and Cambisols could be classified accurately, and the overall validation accuracy of random forest model was 91.67%. (2) The correlations of SOC with bands and band indices improved using multi-temporal images, and the use of mixed synthetic images was better than the use of only one synthetic image. (3) The hybrid model performed far better than the global models, and the probability hybrid model led to the highest prediction accuracy, a validation R2 of 0.77 and an RMSE of 2.30 g kg−1. (4) The probability hybrid model was more accurate than the original hybrid model for digital SOC mapping, and the SOC distribution in boundary regions was smoother and more continuous. Our results suggest that the probability hybrid model has a large potential for SOC prediction and mapping.

Mapping Plant Species in a Former Industrial Site Using Airborne Hyperspectral and Time Series of Sentinel-2 Data Sets

Industrial activities induce various impacts on ecosystems that influence species richness and distribution. An effective way to assess the resulting impacts on biodiversity lies in vegetation mapping. Species classification achieved through supervised machine learning algorithms at the pixel level has shown promising results using hyperspectral images and multispectral, multitemporal images. This study aims to determine whether airborne hyperspectral images with a high spatial resolution or phenological information obtained by spaceborne multispectral time series (Sentinel-2) are suitable to discriminate species and assess biodiversity in a complex impacted context. The industrial heritage of the study site has indeed induced high spatial heterogeneity in terms of stressors and species over a reduced scale. First, vegetation indices, derivative spectra, continuum removed spectra, and components provided by three feature extraction techniques, namely, Principal Component Analysis, Minimal Noise Fraction, and Independent Component Analysis, were calculated from reflectance spectra. These features were then analyzed through Sequential Floating Feature Selection. Supervised classification was finally performed using various machine learning algorithms (Random Forest, Support Vector Machines, and Regularized Logistic Regression) considering a probability-based rejection approach. Biodiversity metrics were derived from resulted maps and analyzed considering the impacts. Average Overall Accuracy (AOA) reached up to 94% using the hyperspectral image and Regularized Logistic Regression algorithm, whereas the time series of multispectral images never exceeded 72% AOA. From all tested spectral transformations, only vegetation indices applied to the time series of multispectral images increased the performance. The results obtained with the hyperspectral image degraded to the specifications of Sentinel-2 emphasize the importance of fine spatial and spectral resolutions to achieve accurate mapping in this complex context. While no significant difference was found between impacted and reference sites through biodiversity metrics, vegetation mapping highlighted some differences in species distribution.

A new digital soil mapping method with temporal-spatial-spectral information derived from multi-source satellite images

Detailed soil maps are essential for effective agricultural practices and environmental protection. Yet, despite the increasing accuracy of digital soil mapping (DSM) in recent years, generating regional-scale, high-accuracy soil maps remains a challenging task. Our study objective was to propose a new DSM method based on temporal-spatial-spectral (TSS) data and to compare the DSM result with a legacy soil map of China. In this study, 13 Landsat multispectral data from 2000 to 2019 were fused by discrete wavelet transform (DWT) to obtain temporal information, a shuttle radar topography mission digital elevation model (SRTM-DEM) was used as spatial information, and Gaofen-5 satellite hyperspectral data was used as spectral information. TSS information was obtained after the DWT and spectral band segmentation methods were used to fuse the temporal and spectral information, combined it with the spatial information. Then, the TSS information and random forest model were used for DSM. The results indicated that 1) The mapping result based on TSS information was highly correlated with the legacy soil map. In different soil classes, there were minute differences in the core area and large differences in adjacent soil classes between the two maps. The overall accuracy and kappa coefficient of DSM based on TSS information were 88.11% and 0.82, respectively. 2) With the same values of the soil moisture, the overall accuracy and kappa coefficient of DSM based on hyperspectral data were 6.80% and 0.02, respectively, higher than those based on multispectral data. 3) With increasing temporal information, the DSM accuracy continuously increased, and the mapping accuracy based on multi-temporal multispectral data was higher than that based on mono-temporal hyperspectral data when the number of multi-temporal multispectral images reached 6. 4) The DSM accuracy was effectively improved when terrain factors were considered, and the terrain factors featuring a strong separability for various soil classes differed. The DSM method proposed in this study based on TSS information from multi-sources remote sensing data greatly improved the DSM accuracy and provided new insight for future DSM research.

Mapping the soil types combining multi-temporal remote sensing data with texture features

With the rapid development of remote sensing (RS) technology, remote sensing images provide an important data basis for soil type mapping. In remote sensing images, temporal factor is difficult to obtain directly, and the rich geometric features are not used adequately. Multi-temporal remote sensing data could effectively reflect the temporal variation of ground objects, while how to extract multi-temporal image features more effectively for soil type interpretation needs to be studied. Moreover, it is not clear whether multi-temporal features and texture features can be effectively integrated to improve mapping accuracies. Therefore, taking five soil types of Laoshan County, Shandong Peninsula, China as the subject investigated and six remote sensing images as data sources, this paper explored and compared two extraction methods of multi-temporal features from remote sensing images. The effects of the eight different texture features fused the multi-temporal features on digital soil mapping were also analyzed. The results showed that the principal component extraction result based on the tasseled cap transformation was better than based on the spectral band synthesis, increasing the overall accuracy by 3.83–11.41% and the kappa index by 0.03–0.13. The fusion of multi-temporal features and texture features can effectively improve accuracies of soil type mapping. After the addition of correlation texture feature parameter, the overall accuracy (86.81%) and Kappa index (0.81) were increased by 11.92% and 0.16, respectively. The research results showed that multi-temporal features in remote sensing images had great advantages in digital soil mapping, and the effective fusion with texture features provided a new idea for improving the accuracy of digital soil mapping.

Wide Area Detection and Distribution Characteristics of Landslides along Sichuan Expressways

Wide area landslide detection is a major international research hotspot in the field of geological hazards, and the integration of multi-temporal optical satellite images and spaceborne interferometric synthetic aperture radar (InSAR) appears to be an effective way to realize this. In this paper, a technical framework is presented for wide area landslide detection: (i) multi-temporal satellite optical images are used to detect landslides with distinguishable geomorphological features; (ii) Generic Atmospheric Correction Online Service (GACOS) assisted InSAR stacking is employed to generate annual surface displacement rate maps in radar line of sight using satellite SAR images from both ascending and descending tracks, which are in turn utilized to automatically detect active landslides from ground motion using hotspot analysis, and (iii) the distribution characteristics of the detected landslides are investigated by examining their relationships with topographic and hydrological factors. Three expressways in Sichuan Province, China—namely the Yakang (Ya’an-Kangding), Yaxi (Ya’an-Xichang), and Lushi (Luding-Shimian) expressways—and their surrounding regions (a total area of approximately 20,000 square kilometers) were chosen as the study area. A total of 413 landslides were detected, among which 320 were detected using multi-temporal satellite optical images, and 109 were detected using GACOS-assisted InSAR stacking. It should be noted that only 16 landslides were detected by both approaches; these landslides all exhibited not only obvious geomorphological features but also ground motion. A statistical analysis of the topographic and hydrological factors shows that of the detected landslides: 81% are distributed at elevations of 1000–2500 m, over 60% lie within the elevation range of 100~400 m, and 90% present with medium and steep slopes (20°~45°), and 80% are located within areas seeing an annual rainfall of 950~1050 mm. Nine landslides were found to pose potential safety hazards to the expressways. The research findings in this paper have directly benefitted the Sichuan expressways; equally important, it is believed that the technical framework presented in this paper will provide guidance for hazard mitigation and the prevention of transportation hazards in the future.

Damaged Building Detection with Improved Swin-Unet Model

Automatic detection of damaged buildings from satellite remote sensing data has become an urgent problem to rescue planners and military personnel. Unfortunately damaged buildings are in different dimensions and shapes with different roofs depending on the type of the material to be painted. In this study, we present an improved Swin-Unet approach that comprises three main operations. First, improved Swin-Unet as a Unet-like pure Transformer is used for multitemporal image segmentation. Second, different multitemporal features are extracted using hyperspectral image classification algorithm. Finally, a binary change map is generated, and evaluation results are obtained. This article takes AIST building change detection scene as the example, and compared with the conventional approaches tested, overall accuracy, mean intersection over union, and separated Kappa in the proposed method were improved by at least 23.36, 0.1725, and 0.202, respectively. Furthermore, different scenes, such as Gaofen-2/Jilin-1 multitemporal optical images and satellite imagery dataset (xBD), have also come to the same conclusion. Thus, it provides advantageous capabilities for monitoring damaged buildings along coastal areas.

Retrieval of Remotely Sensed Sediment Grain Size Evolution Characteristics along the Southwest Coast of Laizhou Bay Based on Support Vector Machine Learning

Grain size is the basic property of intertidal zone sediment. Grain size acts as an indicator of sedimentary processes and geomorphological evolution under human and nature interactions. The remote sensing technique provides an alternative for sediment grain-size parameter monitoring with the advantages of wide coverage and real-time surveying. This paper attempted to map the distributions of three sediment grain size contents and the mean grain size with multitemporal Landsat images along the southwestern coast of Laizhou Bay, China, from 1989 to 2015. Considering the low correlations between the measured reflectance and grain-size parameters, we used a support vector machine (SVM) to develop a nonlinear calibration model by taking several band indices as input variables. Then, the performance of the back propagation neural network (BPNN) was determined and discussed with that of the SVM. The SVM performed better than the BPNN in calibrating the four grain-size parameters based on a comparison of R2 and the root-mean-square error (RMSE). Moreover, an atmospheric correction algorithm originally proposed for case II water enabled the TM\ETM+ images to be precisely atmospherically corrected in this study. The SVM-mapped spatial-temporal grain-size variation showed a coarsening trend, which agreed with that obtained during in situ measurements in a former study. The changes in Yellow River discharge and precipitation associated with the coarsening trend were further analyzed. The yielded results showed that the coarsening trend and reduction in tidal flat area might be aggravated with overutilization. More reasonable planning would be necessary in this case.

Spatiotemporal Change Assessment of Lake Beseka, Ethiopia Using Time Series Landsat Images

Changes in lake surface area and fluctuations in water levels are common especially in the Ethiopian Rift Valley region. The aim of this study was to present the spatiotemporal changes of lake Beseka from 1985 to 2020 using multi-temporal Landsat images. Four spectral water indices, namely Normalized Differential Water Index (NDWI), Modified Normalized Difference Water Index (MNDWI) and Automated Water Extraction Index (AWEI) were used for the extraction of the lake water. The results of the study show that lake Beseka has tended to expand significantly and strongly over the past 35 years. During 1985, the lake surface area was around 30.9km2 and within the next 35 years, the lake surface area increased by 18.18 km2and in 2020, its surface area is about 49.1 km2. lake Beseka shows 0.2412km2 area change from 2015 to 2020. This indicates lake Beseka has a decreasing trend of expansion compared to its history. However, the topography of Lake Beseka is lower in elevation a small water level rise in the lake can cover a large area and lead to significant impacts on the surrounding environment. Therefore, the monitoring of lake surface changes by multi-temporal satellite images is very necessary and of decisive importance.

Assessment of land cover changes on Subi Reef in Truong Sa Islands, Vietnam using multi-temporal Landsat images

In this article, the author proposes that Landsat images should be pre-processed using the DOS method for atmospheric correction and the Gram-Schmidt method for image enhancement. According to the results, the overall classification accuracy of multi-temporal images achieved over 85%, and Landsat images are suitable for classifying land cover on small islands in Truong Sa Islands, such as Subi Reef.

Monitoring deforestation in Jordan using deep semantic segmentation with satellite imagery

Jordan is witnessing major transformations in its environmental and topographic features, where the desert dominates a large part of the territory, with very limited forest areas. Over the last three decades, Jordan has lost about a third of its natural forests at an annual rate of 1.6%. In this study, we develop a deep learning model to automatically monitor deforestation in Jordan based on the semantic segmentation of multitemporal Landsat-8 satellite images. Very few studies and datasets are devoted to monitoring forest cover changes using semantic image segmentation with deep neural networks. Therefore, we have collected a new benchmarking dataset of Jordanian forests between 2010 and 2020. The proposed model includes an efficient encoder-decoder architecture, with which we can extract a set of discriminating features that semantically assign every key image pixel into either forest or non-forest classes. The deep architecture is first initialised by a CNN-based pre-trained model and then it is fine-tuned on the forest images through an effective transfer learning procedure to improve its generalisation ability. Then, a set of key features are extracted by encoding each forest image into low-dimensional semantic maps to formulate the generic descriptors used in image segmentation. Finally, the output of the segmentation process is used to detect any dissimilarity in the forest area or boundaries using an absolute pixel-pixel similarity check over many years. The experimental results proved the effectiveness of the proposed model in segmenting the forest images and predicting any loss (deforestation) or gain (reforestation), and the model achieved an accuracy of 94.8% and MIoU of 82.1%. Moreover, the deep semantic features are discriminative enough to efficiently identify and estimate the amount of deforestation change in terms of accuracy and computational resources.

Extracting Coastal Water Depths from Multi-Temporal Sentinel-2 Images Using Convolutional Neural Networks

Satellite-Derived Bathymetry (SDB) can be calculated using analytical or empirical approaches. Analytical approaches require several water properties and assumptions, which might not be known. Empirical approaches rely on the linear relationship between reflectances and in-situ depths, but the relationship may not be entirely linear due to bottom type variation, water column effect, and noise. Machine learning approaches have been used to address nonlinearity, but those treat pixels independently, while adjacent pixels are spatially correlated in depth. Convolutional Neural Networks (CNN) can detect this characteristic of the local connectivity. Therefore, this paper conducts a study of SDB using CNN and compares the accuracies between different areas and different amounts of training data, i.e., single and multi-temporal images. Furthermore, this paper discusses the accuracies of SDB when a pre-trained CNN model from one or a combination of multiple locations is applied to a new location. The results show that the accuracy of SDB using the CNN method outperforms existing works with other methods. Multi-temporal images enhance the variety in the training data and improve the CNN accuracy. SDB computation using the pre-trained model shows several limitations at particular depths or when water conditions differ.

Multi-temporal satellite image correction based on CNN features and adaptive contrast enhancement

In recent years, identifying changes in multi-temporal images in terms of land use and land cover has been significant in a variety of applications, including urban planning. Due to weather and environmental effects, optical remote sensing has limitations in obtaining images where the image quality may be degraded. It’s because the images being registered are taken at various times, viewpoints, and types of sensors. In this article, the pre-processing methods, which include radiometric correction and geometric correction, are introduced to enhance the quality of satellite images and identify correct spatial alignment. For radiometric correction, adaptive contrast enhancement is done by combining histogram- and non-linear transfer function-based approaches in CIELAB color space. A comparison study is done to see how the new method compared to other methods. For geometric correction, the features from two images are extracted using Convolutional Neural Network to match and align them. The introduced approach for radiometric correction gave the best average rank of BRISQUE scores and RMSE of contrast scores, and the geometric correction can align two images together with an average accuracy of improvement of 91.78 percent. The findings of this research will provide the preliminary step for any change detection activities.

Decadal Changes in Land use and Land Cover of Noyyal River Basin using Geo-spatial Techniques

Aims: River Noyyal was the life line of the people of Coimbatore, Tirupur and Karur districts of Tamil Nadu and has nurtured a rich civilization. The river is mentioned in many ancient travelogues by European travelers which suggest the importance of the river. But over the years, the condition of the river, both in terms of quantity and quality has deteriorated owing to the expanding population size and its related land use changes.
 Place and Duration of Study: The study was conducted to investigate the decadal land use and land cover changes of the Noyyal basin for the year of 2000 to 2020 in the year 2021- 2022.
 Methodology: The study was undertaken to produce the land use/land cover map and to explore the change detection analysis of the Noyyal river basin for 20 years. Based on RS and GIS for monitoring the temporal variations of land use land cover, multi-temporal Landsat satellite 30m spatial resolution images of Landsat 4/5 MSS and TM 2000, 2010, and Landsat 8 (OLI) 2020 were obtained from the google earth engine. At the first stage NDVI calculation was done by using ArcGIS software and the second stage supervised classification maximum likelihood classification was done for 3 years 2000,2010 and 2020.
 Results: The analysis suggests that Normalized Difference Vegetation Index NDVI of without any vegetation (Class1), medium density (Class3), and high density (Class4) increased by 8.37%, 1.29%, 0.42% respectively. Low density (Class2) decreased by 10.1%. The urban area and agriculture land increased by 13.82% and 18.46%.The forest cover, waste land and barren land decreased by 12.24% 11.99% and 7.90% over the 2 decades and water bodies increased in the year of 2010 and then decreased.
 Conclusion: The study has revealed a decline in area under forest and wasteland and an increase in area under built up activities and agriculture land.

Automatic binary and ternary change detection in SAR images based on evolutionary multiobjective optimization

In most of the previous works, changed and unchanged regions are detected by analyzing the changes of backscattering coefficients for SAR images, which is termed as binary change detection. In fact, due to the increase and decrease of backscattering coefficients, the changed regions can be further analyzed as two kinds of changes, which is termed as ternary change detection. In this paper, a change detection method based on evolutionary multiobjective optimization is proposed to automatically perform binary and ternary change detection of multitemporal SAR images. First, the log-likelihood function of the Gaussian mixture model and the Bhattacharyya distance are designed as two objectives, respectively. In particular, a novel measurement method based on Bhattacharyya distance is designed for the ternary change detection task. Not only the separability between each two classes is maximized, but also the Bhattacharyya distance between two changed classes and unchanged class is kept closer to obtain a more balanced classification performance. Then a multiobjective optimization method based on non-dominated sorting is used to optimize these two objectives simultaneously. In the proposed approach, chromosome ranking and perturbation probability selection operators are designed to make high-quality solutions with a high probability of being exploited and improve the performance of the algorithm. In addition, a one-step local search strategy based on the expectation–maximization method is integrated into the proposed algorithm to accelerate the convergence. Experimental results on simulated and real-world datasets demonstrate the effectiveness and robustness of the proposed algorithm.

Multitemporal seagrass carbon assimilation and aboveground carbon stock mapping using Sentinel-2 in Labuan Bajo 2019–2020

Seagrass is considered one of the most effective and efficient natural carbon sinks. In order to fully understand how it can help to adapt and mitigate climate change, information on the variability of the extent of its area, carbon stock, assimilation, sequestration, and potential emission is critical. However, details on spatially and temporally explicit seagrass biomass carbon stock, assimilation and sequestration are lacking in many parts of the world, including Indonesia as the seagrass biodiversity hot spot. To address this issue, this research aims to develop a remote sensing model to map aboveground seagrass carbon stock (AGC) and seagrass carbon assimilation rates (CA) based on multitemporal Sentinel-2 image analysis that can be used to assess the variability of seagrass over times and assess the potential carbon loss and emission reduction by seagrass. Sentinel-2 images with ideal conditions (cloud, haze, sunglint-free) and less ideal ones (cloud and haze-free, but with sunglint) were both used as the basis to develop the model. The research used two regression models, the random forest (RF) regression and stepwise regression (SW), and four input types, i.e., the reflectance bands, deglint bands, Principal Component (PC) bands, and co-occurrence texture bands. Field AGC and CA data estimated from percent cover were used to train and assess the accuracy of each model. Cross-validation was also performed to determine the model consistency when applied to images with different conditions. The results showed that the SW regression model using reflectance and deglint bands produced the most accurate and consistent AGC and CA maps. Applied to a 24-month Sentinel-2 image, from 2019 until 2020, the model showed a consistent pattern where the AGC and CA were lower from September to February and higher in March to August for both 2019 and 2020. Based on the two-year seagrass AGC and CA maps, the monthly average AGCs for 2019 and 2020 were estimated at 28.4 ± 2.7 and 28.4 ± 1.3 g C/m2, respectively, while the monthly average CAs were 6.7 ± 1.4 and 6.6 ± 0.7 Mg C/ha/year. The total AGCs in 2019 and 2020 were 113.8 ± 10.9 and 113.9 ± 5.2 Mg C, and the total CAs were 2648.3 ± 623.4 and 2636.9 ± 281.4 Mg C/year. All these values were from 400.71 ha of seagrass meadows. Finally, the availability of such remote sensing models helps to solve the time and cost-effectiveness challenges in acquiring spatially and temporally extensive seagrass AGC and CA information. Moreover, the research was able to produce the first map of multitemporal seagrass AGC and CA in specific seagrass meadow environments in Indonesia, which can be further improved to map these data rapidly at the national level and will be very useful for future work that implements Tier 3 methodological approach on carbon inventory.

Unsupervised change detection in SAR images based on generalized likelihood ratio test and a two-stage morphological filter

ABSTRACT Generalized likelihood ratio test (GLRT) is an efficient method to generate difference image (DI) for change detection (CD) using synthetic aperture radar (SAR) images. GLRT is usually applied with a fixed-size moving window to the neighbourhood regions in multitemporal SAR images. The fixed window may be however not optimal for all the pixels under test. To solve disadvantages of the GLRT method, an adaptive circular window is proposed in this work. While the adaptive square window and the best fixed square window achieve an average Kappa coefficient of 83.55% and 82.12%, respectively, the adaptive circular window improves the average Kappa coefficient by 83.65% in six datasets. As another difficulty, speckle noise reduces quality of DI in SAR change detection. Three steps are considered to minimize effects of the speckle noise: (1) a two-stage morphological filter is suggested to reduce the speckle noise; (2) to generate DI, the adaptive circular window for generalized likelihood ratio test (ACWGLRT) is proposed that reduces influence of the speckle noise while preserves the edge details of multitemporal images; and (3) spatial fuzzy c-means (SFCM) is used to reduce effects of the residual speckle-noise during DI classification. The experimental results show superior performance of the proposed change detection method with respect to several competitors. The proposed method has the best Kappa coefficient and percentage correct classification (PCC) in four datasets of Ottawa, San Francisco, Farmland C, and Inland water.

Estimating Suspended Sediment Fluxes from the Largest Glacial Lake in Svalbard to Fjord System Using Sentinel-2 Data: Trebrevatnet Case Study

Glacier-fed hydrological systems in high latitude regions experience high seasonal variation in meltwater runoff. The peak in runoff usually coincides with the highest air temperatures which drive meltwater production. This process is often accompanied by the release of sediments from within the glacier system that are transported and suspended in high concentrations as they reach the proglacial realm. Sediment-laden meltwater is later transported to the marine environment and is expressed on the surface of fjords and coastal waters as sediment plumes. Direct monitoring of these processes requires complex and time-intensive fieldwork, meaning studies of these processes are rare. This paper demonstrates the seasonal dynamics of the Trebrevatnet lake complex and evolution of suspended sediment in the lake and sediment plumes in the adjacent Ekmanfjorden. We use the Normalized Difference Suspended Sediment Index (NDSSI) derived from multi-temporal Sentinel-2 images for the period between 2016–2021. We propose a new SSL index combining the areal extent of the sediment plume with the NDSSI for quantification of the sediment influx to the marine environment. The largest observed sediment plume was recorded on 30 July 2018 and extended to more than 40 km2 and a SSL index of 10.4. We identified the greatest sediment concentrations in the lake in the beginning of August, whereas the highest activity of the sediment plumes is concentrated at the end of July. The temporal pattern of these processes stays relatively stable throughout all ablation seasons studied. Sediment plumes observed with the use of optical satellite remote sensing data may be used as a proxy for meltwater runoff from the glacier-fed Trebrevatnet system. We have shown that remote-sensing-derived suspended sediment indexes can (after proper in situ calibration) serve for large scale quantification of sediment flux to fjord and coastal environments.