Vegetation Red Edge Research Articles

Investigation of teak defoliator, Hyblaea puera (Cramer, 1777) outbreak on mangrove ecosystem in Mumbai metropolitan region using multi-spectral satellite data

The mangrove ecosystem supports biodiversity and ecology and plays a vital role in providing economic security to coastal fishers. A severe outbreak of the moth Hyblaea puera occurred along the Thane creek in Mumbai (Maharashtra, India) between September and October 2019. The ecological interaction of pest species in a mangrove ecosystem is complex and very few detailed studies have been undertaken along the Indian coast. There is evidence that infestation can have a negative impact on the already threatened mangrove ecosystem. This study investigates the mangrove infestation, the influence of the infestation on the Leaf Area Index (LAI), rainfall’s effect on such outbreaks’ occurrence, and spectral reflectance in infested and non-infested mangroves. Sentinel 2 MSI data was processed using the QGIS platform. There were two dominant mangrove species, Avicennia marina and Sonneratia apetala, of which the most affected species was A. marina. The study revealed that 67.90% (50.61 km2) of the mangrove area defoliated due to this outbreak, out of which 30.20% (22.50 km2) was highly defoliated, 37.70% (28.11 km2) was less defoliated, and 32.10% (23.95 km2) was unaffected. LAI is the health indicator of a forest affected by an infestation; it indicates the reduction in canopy cover due to the infestation. The t-test was applied to determine the significance of LAI before and after the infestation along the upper stretch of the creek,t-value was 3.886 (p = 0.0011, p ≤ 0.05 significant). The spectral reflectance analysis showed a reduction in the infested mangrove’s Vegetation Red Edge and Near Infra-Red reflectance. The ANOVA result shows that the three groups, Mangrove Highly Infested, Mangrove Less Infested, and Non-Infested Mangrove, are likely spectrally separable (p = 0.020, p ≤ 0.05). The analysis revealed that rainfall intensity during an Infested year significantly differs from that in a Non-Infested year (p = 0.0067, p ≤ 0.05).

Mapping seagrasses on the basis of Sentinel-2 images under tidal change

Tidal variations make the water bodies in satellite remote sensing images on different shooting dates have different inundation ranges and depths. Although the underwater substrates do not change, the spectral properties differ due to attenuation effects. These differences have an impact on the results when multi-temporal remote sensing images are used to analyze seagrasses. This paper proposes a remote sensing mapping method for seagrasses taking the tidal influence, using the seagrasses growth area in Xincun Bay, Hainan Province, China as a case study. a) The seagrasses growth area was determined from remote sensing images. The seagrasses were divided into two types: the seagrasses exposed to water surface or tidal flats (non-submerged seagrasses) and the seagrasses submerged in water (submerged seagrasses). b) The spectral features of seagrasses in Sentienl-2 image were analyzed. We found that the spectral characteristics of non-submerged seagrasses were similar to terrestrial vegetation and these seagrasses could be extracted by using NDVI. The submerged seagrasses spectral was different, forming a reflection peak at the first vegetation red edge band (i.e.705 nm) in Sentinel-2 images. This reflection peak was used to design the Submerged Seagrasses Identification Index (SSII) for extracting underwater seagrass. c) The extraction results of non-submerged seagrasses and submerged seagrasses were merged to map the seagrasses in the study area. The experimental results show that the mapping method proposed in this study can fully consider the influence of tidal changes in remote sensing images on seagrasses identification. The SSII constructed based on Sentinel-2 images extracted submerged seagrasses effectively. This study will provide references to remote sensing mapping of seagrasses and integrated ecological management in coastal zones.

Photosynthetic Fluorescence from Earthlike Planets around Sunlike and Cool Stars

Remote sensing of the Earth has demonstrated that photosynthesis is traceable as the vegetation red edge (VRE), which is a steep rise in the reflection spectrum of vegetation, and as solar-induced fluorescence. This study examines the detectability of biological fluorescence from two types of photosynthetic pigments, chlorophylls (Chls) and bacteriochlorophylls (BChls), on Earthlike planets with oxygen-rich/poor and anoxic atmospheres around the Sun and M dwarfs. Atmospheric absorption, such as H2O, CH4, O2, and O3, and the VRE obscure the fluorescence emissions from Chls and BChls. We find that the BChl-based fluorescence for wavelengths of 1000–1100 nm, assuming the spectrum of BChl b–bearing purple bacteria, could provide a suitable biosignature, but only in the absence of water cloud coverage or other strong absorbers near 1000 nm. The Chl fluorescence is weaker for several reasons, e.g., spectral blending with the VRE. The apparent reflectance excess is greatly increased in both the Chl and BChl cases around TRAPPIST-1, due to the fluorescence and stellar absorption lines. This could be a promising feature for detecting the fluorescence around ultracool red dwarfs using follow-up ground-based observations at high spectral resolution; however, this would require a long time around Sunlike stars, even for a LUVOIR-like space mission. Moreover, the simultaneous detection of fluorescence and the VRE is the key to identifying traces of photosynthesis, because absorption, reflectance, and fluorescence are physically connected. For further validation of the fluorescence detection, the nonlinear response of biological fluorescence as a function of light intensity could be considered.

A GIS‐based multicriteria decision analysis to reduce riparian vegetation hydrogeological risk and to quantify harvested biomass (Giant reed) for energetic retrieval

The vegetation development in riverbeds creates obstructions to the regular water flow with the flooding hydraulic risk increase. Giant reed (GR) (Arundo donax L.) is one of the most successful invasive species of riparian ecosystems in Mediterranean semi-arid climate conditions, with significant public administrations costs for its removal and disposal. At the meantime, due to its high biomass yield and adaption capacity to several conditions, GR is a very promising no-food crop to produce biogas by the anaerobic digestion (AD). The research activity provides for the involvement of territories belonging to the inner areas of Sicily, especially Calatino one, in which the SIMBIOSI Consortium operates an AD plant fed with agricultural by-products from other close agro-industrial companies. The aim of the study was to map and to quantify the actual spatial distribution of GR in watercourses embankments through Remote Sensing (RS) techniques applied in a Geographic Information System (GIS) environment. In this regard, a method based on the automatic supervised classification, was applied on three different combinations of spectral bands (True Color Image – TCI; Near-Infrared, Green and Blue – NGB; Vegetation Red Edge – VRE) of Sentinel-2 satellite images related to the summer season (11th August 2019), with the aim of identifying the most suitable classification to map the GR. The results showed that the VRE composition is the most accurate combination of spectral bands for the identification of GR, followed by the NGB image. The worst performance was obtained by using the TCI combination. A further elaboration was carried out combining the three classified images, in order to obtain a more accurate localization and quantification of GR. The final thematic map allowed to correctly classify GR for the 46 % of the cases, with an overall accuracy of 85.02 % and a high Kappa Coefficient of Agreement value equal to 0.81. Finally, the surface covered by GR in the study area (computed in the GIS environment) was about 2 km2 and the estimated GR biomass, available for the biomethane production, obtainable from watercourses embankments maintenance interventions would amount to about 11,780 tons year−1. The study could contribute to the development of a watercourses maintenance plan with the aim to reduce the risk of streams flooding in valley areas and at intersections with infrastructure works. Furthermore, the proposed methodology can be used by stakeholders in marginal areas, for the watercourses management, offsetting its costs through the energy use of the collected biomass.

Improving remote sensing of salinity on topsoil with crop residues using novel indices of optical and microwave bands

Remote sensing indices have been proposed to characterize soil salinity. However, the sensitivity of these indicators is unstable owing to differences in geographic environment and vegetation type. This study investigated the performance of several existing indices to estimate the salinity of topsoil with residues in southern Xinjiang, China. The results showed that these indices were not satisfactory. In order to construct an index that can be used to directly indicate soil salinity in a specific area, novel salinity indices were calculated using optical bands (blue, green, red, vegetation red edge, and shortwave infrared bands) derived from Sentinel-2 multispectral data and Sentinel-1 radar data (backscattering coefficient VV, VH). To enhance the sensitivity of the optical bands, five transformation methods (logarithmic, reciprocal, first-, second-, and third-derivative) were applied to the original spectra. Based on previous studies, statistical methods were used to construct two-, three-, and four-bands indices. One constructed three-bands index with the second-derivative transformation, called the Enhanced Residues Soil Salinity Index (ERSSI), showed the highest correlation with topsoil salinity (r = 0.65 and 0.68 in training and testing). ERSSI establishes a linear relationship in soil salinity estimation with an R2 of 0.53 and a LCCC of 0.65 in training dataset, with an R2 of 0.51 and a LCCC of 0.73 in testing dataset. And it shows contribution in random forest regression with an R2 of 0.80 and a LCCC of 0.86 in training dataset, with an R2 of 0.77 and a LCCC of 0.81 in testing dataset. The ERSSI consisted of the B, G, and SWIR1 bands, and was sensitive to salinity variations in the residues remaining in farmland soils. This study provides a novel index and method for the accurate and robust assessment and mapping of salinity in farmland covered by crop residues.

Assessment of Sentinel-2 and Landsat-8 OLI for Small-Scale Inland Water Quality Modeling and Monitoring Based on Handheld Hyperspectral Ground Truthing

This study investigates the best available methods for remote monitoring inland small-scale waterbodies, using remote sensing data from both Landsat-8 and Sentinel-2 satellites, utilizing a handheld hyperspectral device for ground truthing. Monitoring was conducted to evaluate water quality indicators: chlorophyll-a (Chl-a), colored dissolved organic matter (CDOM), and turbidity. Ground truthing was performed to select the most suitable atmospheric correction technique (ACT). Several ACT have been tested: dark spectrum fitting (DSF), dark object subtraction (DOS), atmospheric and topographic correction (ATCOR), and exponential extrapolation (EXP). Classical sampling was conducted first; then, the resulting concentrations were compared to those obtained using remote sensing analysis by the above-mentioned ACT. This research revealed that DOS and DSF achieved the best performance (an advantage ranging between 29% and 47%). Further, we demonstrated the appropriateness of the use of Sentinel-2 red and vegetation red edge reciprocal bands 1 / B 4 × B 6 for estimating Chl-a ( R 2 = 0.82 , RMSE = 14.52 mg / m 3 ). As for Landsat-8, red to near-infrared ratio ( B 4 / B 5 ) produced the best performing model ( R 2 = 0.71 , RMSE = 39.88 mg / m 3 ), but it did not perform as well as Sentinel-2. Regarding turbidity, the best model ( R 2 = 0.85 , RMSE = 0.87 NTU) obtained by Sentinel-2 utilized a single band (B4), while the best model (with R 2 = 0.64 , RMSE = 0.90 NTU) using Landsat-8 was performed by applying two bands ( B 1 / B 3 ). Mapping the water quality parameters using the best performance biooptical model showed the significant effect of the adjacent land on the boundary pixels compared to pixels of deeper water.

A combination of Sentinel-1 RADAR and Sentinel-2 multispectral data improves classification of morphologically similar savanna woody plants

ABSTRACT The co-existence of diverse plant forms in densely vegetated savanna environments presents a challenge when mapping species diversity using single remotely sensed data type that carries either optical or structural information. In the present study, Sentinel-1 RADAR and Sentinel-2 multispectral data were combined to classify morphologically similar woody plant species (n =27) and three coexisting land cover types using Deep Neural Network (DNN). The fused image recorded a higher overall classification accuracy (76%) than the sole use of Sentinel-2 (72%) and Sentinel-1 RADAR data (71%). Slightly more species (15) recorded accuracies exceeding 75% using fused image compared to Sentinel-2 and Sentinel-1 data (13 species >75%). Analysis of relative band contributions resulted in high importance from Sentinel-1 C-band ratio of VH/VV polarization (8.6%) as well as a select Sentinel-2 bands (Near infrared (9.86%), Shortwave (9.5%), and Vegetation red edge (8%)). Parallel to continual efforts to improve the accuracies of fused RADAR–optical data, the services of such data for regional-scale applications should be explored to inform timely biodiversity assessments.

Potential of mapping dissolved oxygen in the Little Miami River using Sentinel-2 images and machine learning algorithms

Dissolved oxygen (DO) is one of the most significant indicators of water quality in the inland water systems. Maintaining DO monitoring systems is time and money consuming since in situ DO data with high spatial and temporal resolution is needed to design appropriate coastal management plans. In this study, we mapped the spatiotemporal changes of DO in the Little Miami River (LMR) using 10-m Sentinel-2 images. We trained two machine learning algorithms – Random Forest (RF) and Support Vector Machine (SVM) – to predict DO concentrations using spectral predictors derived from the satellite images. Moreover, we calculated several metrics, which include Root Mean Squared Error (RMSE), Amount of Variance Explained (AVE), Coefficient of Efficiency (COE), and Normalized Mean Bias (NMB) to assess the performance of the models and accuracy of the DO maps. Our results showed a good agreement between modeled and measured DO concentrations with minimal residual errors ranging between 0.201 mg/L and 0.241 mg/L. Furthermore, RF and SVM revealed to be the reliable and effective algorithms to estimate DO concentrations with bands 5 (B5, vegetation red edge) and 8 (B8, NIR) as important predictors. We found the DO concentrations in the LMR to be in good to excellent range (7.3 mg/L to 14.7 mg/L for August and 10.8 mg/L to 14.5 mg/L for October), based on the minimum criterion for warm water habitat in the Ohio Water Quality Standards (WQS). Our findings have provided spatial insight to the current status of DO and the success of the management steps taken to manage and prevent eutrophic problems in the LMR.

Water Information Extraction Based on Multi-Model RF Algorithm and Sentinel-2 Image Data

For the Sentinel-2 multispectral satellite image remote sensing data, due to the rich spatial information, the traditional water body extraction methods cannot meet the needs of practical applications. In this study, a random forest-based RF_16 optimal combination model algorithm is proposed to extract water bodies. The research process uses Sentinel-2 multispectral satellite images and DEM data as the basic data, collected 24 characteristic variable indicators (B2, B3, B4, B8, B11, B12, NDVI, MSAVI, B5, B6, B7, B8A, NDI45, MCARI, REIP, S2REP, IRECI, PSSRa, NDWI, MNDWI, LSWI, DEM, SLOPE, SLOPE ASPECT), and constructed four combined models with different input variables. After analysis, it was determined that RF_16 was the optimal combination for extracting water body information in the study area. Model. The results show that: (1) The characteristic variables that have an important impact on the accuracy of the model are the improved normalized difference water index (MNDWI), band B2 (Blue), normalized water index (NDWI), B4 (Red), B3 (Green), and band B5 (Vegetation Red-Edge 1); (2) The water extraction accuracy of the optimal combined model RF_16 can reach 93.16%, and the Kappa coefficient is 0.8214. The overall accuracy is 0.12% better than the traditional Relief F algorithm. The RF_16 method based on the optimal combination model of random forest is an effective means to obtain high-precision water body information in the study area. It can effectively reduce the “salt and pepper effect” and the influence of mixed pixels such as water and shadows on the water extraction accuracy.

Normalized Difference Enhanced Sand Index for desert sand dunes detection using Sentinel-2 and Landsat 8 OLI data, application to the north of Figuig, Morocco

This paper discusses developing a novel spectral index for desert sand dunes detection using four bands from Sentinel-2 data. This index, called Normalized Difference Enhanced Sand Index (NDESI), uses the blue band (492 nm), the red (664 nm) or the Vegetation Red Edge (VRE) band (782 nm), and two of the Shortwave Infrared (SWIR) bands (1613 and 2202 nm) to generate a grayscale image, where DN values are within −2 to 2. The lowest and highest values represent water bodies and sand dunes, respectively. A threshold calculation technique was used to generate a unique threshold for each image. This index application on several images covering the northern region of Figuig city seems to perfectly highlight sand dunes all over the image, with an overall accuracy of 97.9% and a kappa coefficient of 83.4%. This index is adapted for Landsat 8 data and can track sand dunes change over time.

Detecting and mapping the spatial distribution of Chromoleana odorata invasions in communal areas of South Africa using Sentinel-2 multispectral remotely sensed data

Alien Invasive Plants (AIPs) have been proclaimed to be a significant non-climatic driver of global change. The Siam weed (Chromoleana odorata) is highly invasive in South African rangelands, with serious environmental, agricultural and socio-economic consequences. Therefore, a comprehensive analysis of C.odorata's spatial distribution is necessary for implementing relevant mitigation and management approaches to vulnerable and invaded landscapes. Remotely sensed data offer a viable opportunity for detecting and mapping AIPs spatial extents. Hence, this study sought to compare the value of algorithms (i.e. maximum likelihood and random forest) in detecting and mapping C.odorata's spatial distribution in relation to other land use land cover classes (LULC) using the freely available Sentinel-2 multispectral image data. The findings of the study revealed that the vegetation red edge and near-infrared spectral bands of Sentinel-2 multispectral imagery (MSI) sensor were the most important spectral variables for discriminating C.odorata from other land cover classes. The random forest algorithm yielded the highest overall classification accuracy of 83%, outperforming the traditional maximum likelihood classifier (75%) in classifying C.odorata's spatial distribution. Overall, results demonstrate that the combination of freely available Sentinel-2 MSI satellite data and machine learning random forest algorithm produces the highest accuracies for detecting the spatial distribution of C.odorata, particularly in heterogeneous environments. These results could be beneficial to sustainable rangeland management and the adoption of site-specific mitigation approaches in areas invaded by C.odorata.

PERFORMANCE ASSESSMENT OF CLASSIFICATION ALGORITHMS FOR LANDUSE / LANDCOVER CHANGE USING SENTINEL 2 DATA – A CASE STUDY OF TIRUPPUR

Abstract. In the history of mankind, one of the vibrant geographical phenomena is urbanization. The urbanization process is characterized by the expansion of the city from the core to peripheral areas which includes economic development, social, political forces and population density. Very rapid urbanization in the highly populated country like India, which changes natural land cover into urban land use, which is unavoidable. However, the study region Tiruppur is known as the knitwear capital of India that induces urban development in the region which results in the modification of the natural land cover. For understating the interaction between the natural landscape and human activities, land use and land cover (LULC) is considered as the important indicator. Research on land-use and land cover changes using remote sensing technology has a long history to evident. The advancement in the Remote Sensing and GIS techniques provide the fine resolution of data sets to proceed. Sentinel-2B imagery was chosen for this study for two main reasons one is that compare to Landsat imagery it has a high spatial resolution of 10 m and its radiometry includes three vegetation red edge bands. These two characteristics make the Sentinel-2B data appealing for LULC mapping. Different types of classification algorithms have been used to perform land use and land cover mapping. The study aims to create land use and land cover classification by making a comparison between different algorithms in Tiruppur by using Sentinel-2B satellite imagery. The commonly known classification algorithms, K-means, IsoData, support vector machines (SVMs), and maximum likelihood (ML) classification are adopted for investigation. This is followed by the selection of training pixels from the remaining classes to perform and compare different supervised learning algorithms for the first- and second-level classification in terms of accuracy rates. Accuracy was assessed through metrics derived from an error matrix, but primarily overall accuracy and kappa coefficient was used in allocating algorithm hierarchy. Finally, after the comparison, the highly accurate algorithm was suggested for the mapping of urban areas. The highest overall accuracy and kappa coefficient was produced by support vector machine (SVM) is due to the algorithm’s relatively small number of complex decision boundaries. The results are helpful to understand the performance of the classification algorithm for the future studies.

Diurnal Variability and Detectability of Vegetation Red Edge of Earth-like Exoplanets

Abstract The vegetation red edge (VRE) is a unique spectral fingerprint of light-harvesting vegetation on Earth and provides a robust remote detectable surface biosignature of exoplanets. To improve the detectability and sensitivity, we have studied the diurnal variability of VRE in the disk-integrated spectra of Earth and also Earth analogs in the case of different observing geometry conditions. Simulation results show that the VRE index varies from <−0.4 to >0.6 at a diurnal timescale for both present and also Late Triassic Earth, and the maximum variation of VRE in 1 day changes by >3 times with different observing geometry conditions. This means that the extraterrestrial light-harvesting vegetation (even if it really exists) will not be efficiently detectable without proper observing geometry conditions and time, especially in the case of the exoplanets covered with thick clouds. The VRE temporal variation curve can also be used to retrieve the cloud cover fraction and continent distribution of exoplanets with relatively high precision. Several observational strategies are proposed to detect the light-harvesting vegetation and retrieve the planetary information from the planet’s VRE variation signals, and a mock observation is also demonstrated.

Mapping the Eucalyptus spp woodlots in communal areas of Southern Africa using Sentinel-2 Multi-Spectral Imager data for hydrological applications

Invasive alien plant species (IAPs) drive anthropogenic global environmental change by threatening native species communities and ecosystems. For instance, the Eucalyptus species are notoriously known for releasing toxic chemicals which impedes the growth of indigenous plants which usually results in severe erosion. It is, therefore, important and necessary to quantify and understand the spatial extent and distribution of these IAPs to identify critical hotspots for restoration. This study, therefore, tested the utility of Sentinel 2 Multispectral Imager (MSI) data in distinguishing Eucalyptus community-woodlots from other land cover types with an aid of the Support Vector Machines (SVM) algorithm. The results show that eucalyptus species were successfully discriminated from other land cover types with an overall accuracy (OA) of 85.07%. Eucalyptus covered an area of 2378.08 ha, constituting 11.6% of the study area. Furthermore, about 23% of Eucalyptus wood plots were found at a closer range of 1 km from the rivers. Sentinel 2 MSI Band 3 (green), 7 (vegetation red edge) and 8A (narrow near-infrared (NIR)) proved to be effective in characterizing the spatial distribution and extent of Eucalyptus wood plots. The findings of this study serve as a footstool towards the establishment of a framework for monitoring these IAPs and understanding their impact on the diversity of indigenous tree species.

Mapping changes in artisanal and small-scale mining (ASM) landscape using machine and deep learning algorithms. - a proxy evaluation of the 2017 ban on ASM in Ghana

Artisanal and Small-Scale Mining (ASM) landscapes form integral part of the Land use land cover (LULC) in the developing worlds. However, the spatial, spectral, and temporal footprints of ASM present some challenges for using most of the freely available optical satellite sensors for change analysis. The challenge is even profound in tropical West African countries like Ghana where there is prolonged cloud cover. Whiles very few studies have used Sentinel-2 data to map change analysis in ASM landscape, none examined the contribution of individual S2 bands to the ASM classifications. Also, despite the capabilities of Machine Learning (ML) and Deep Learning (DL) models for LULC classifications, few studies have compared the performances of different classifiers in mapping ASM landscape. This study utilized Sentinel-2 data, four ML and DL models (Artificial Neural Network –ANN, Random Forest – RF, Support Vector Machines –SVM, a pixel-based Convolutional Neural Network-CNN) and image segmentation to examine the performance of S2 bands and ML and DL algorithms for change analysis in ASM landscape, with the Birim Basin in Ghana as a study area. The result of the change analysis was used to assess changes in LULC during the recent ban on the expansion of ASM in the country. It was found out that ANN is a better classifier of ASM achieving the highest overall accuracy (OA) of 99.80% on the segmented Sentinel-2 bands. The study also found out that the Band 5 Vegetation Red Edge (VRE) 1 contributed most to classifying ASM, with the segmented VRE 1 being superlative over the other predictors. In terms of expansion, ASM increased by 59.17 km2 within the period of the study (January 2017 to December 2018), suggesting that ASM still took place under the watch of the ban. The classification results showed that most of the peripheral of forest and farmland have been converted to ASM with little disturbance within the interior of the forest reserves. The study revealed that, the ban was yielding very little or no results due to a number of policy deficiencies including low staff strength, lack of logistics and low remuneration. Enforcement of legal instruments against ASM and farming activities within the forest reserves, improvement in the monitoring systems and intensification of public education on the value of forest and the need to protect it are some of the major recommendations that could control encroachment on the forest reserves.

Deep Learning Based Thin Cloud Removal Fusing Vegetation Red Edge and Short Wave Infrared Spectral Information for Sentinel-2A Imagery

Thin clouds seriously affect the availability of optical remote sensing images, especially in visible bands. Short-wave infrared (SWIR) bands are less influenced by thin clouds, but usually have lower spatial resolution than visible (Vis) bands in high spatial resolution remote sensing images (e.g., in Sentinel-2A/B, CBERS04, ZY-1 02D and HJ-1B satellites). Most cloud removal methods do not take advantage of the spectral information available in SWIR bands, which are less affected by clouds, to restore the background information tainted by thin clouds in Vis bands. In this paper, we propose CR-MSS, a novel deep learning-based thin cloud removal method that takes the SWIR and vegetation red edge (VRE) bands as inputs in addition to visible/near infrared (Vis/NIR) bands, in order to improve cloud removal in Sentinel-2 visible bands. Contrary to some traditional and deep learning-based cloud removal methods, which use manually designed rescaling algorithm to handle bands at different resolutions, CR-MSS uses convolutional layers to automatically process bands at different resolution. CR-MSS has two input/output branches that are designed to process Vis/NIR and VRE/SWIR, respectively. Firstly, Vis/NIR cloudy bands are down-sampled by a convolutional layer to low spatial resolution features, which are then concatenated with the corresponding features extracted from VRE/SWIR bands. Secondly, the concatenated features are put into a fusion tunnel to down-sample and fuse the spectral information from Vis/NIR and VRE/SWIR bands. Third, a decomposition tunnel is designed to up-sample and decompose the fused features. Finally, a transpose convolutional layer is used to up-sample the feature maps to the resolution of input Vis/NIR bands. CR-MSS was trained on 28 real Sentinel-2A image pairs over the globe, and tested separately on eight real cloud image pairs and eight simulated cloud image pairs. The average SSIM values (Structural Similarity Index Measurement) for CR-MSS results on Vis/NIR bands over all testing images were 0.69, 0.71, 0.77, and 0.81, respectively, which was on average 1.74% higher than the best baseline method. The visual results on real Sentinel-2 images demonstrate that CR-MSS can produce more realistic cloud and cloud shadow removal results than baseline methods.

Vegetation Characterization at Community Level Using Sentinel-2 Satellite Data and Random Forest Classifier in Western Himalayan Foothills, Uttarakhand

Characterization of forests is an important aspect for the management and conservation of biodiversity and satellite remote sensing provides opportunity to rapidly assess biodiversity patterns across large geographical areas. The present study aimed to characterize vegetation in a landscape with heterogenous topography using medium resolution Sentinel-2 satellite data and random forest (RF) machine learning algorithm. The study area falls in the Nandhaur landscape of western Himalayan foothills and forms an important part of the Terai Arc Landscape. Sentinel-2 satellite data of January, April and May months was used along with SRTM Digital Elevation Model. Primary field data was collected for dominant vegetation community characteristics and used for training and validation of the model in RF. A total of eleven forest vegetation communities was characterized with an accuracy of 70–87% for individual imageries as well as in collocation of images. Satellite image of April was found most useful for discriminating forest communities with high spectral separability among communities. Variable important analysis revealed Vegetation Red Edge, Narrow NIR and NIR bands were useful for discriminating communities with maximum information along with SWIR and Red bands. Study found that collocation of the multi-date images was useful for classifying communities with higher accuracies. This attempt at characterizing heterogeneous vegetation communities with advanced machine learning tools in Himalayan foothills showed encouraging outcomes which can be used in future studies for monitoring and management of key habitats in this area.

Performance of random forest and buffer analysis of Sentinel-2 data for modelling soil salinity in the Lower-Cheliff plain (Algeria)

ABSTRACT Modelling approaches are becoming an efficient tool in the forecasting of the salinization spread impact on the local and global scales. In the Lower-Cheliff plain, updating the information on soil salinity expansion is more than required as it continues to damage the agricultural environment in there. Through this study, we adopted an artificial methodology that consists of a combination of buffer analysis using Sentinel-2 data and machine learning modelling to assess their aptitude in the prediction and mapping of soil salinity. The adopted random forest (RF) algorithm included the reflectance information from of the bands from Blue (), Green (), Red (), Near infrared (NIR), Vegetation red-edge (VRE) and Shortwave Infrared (SWIR), optimized with the geospatial buffering based on the 91 soil random samples collected during the summer of 2019 and measured for the Electrical Conductivity (EC) in the laboratory. The outputs from the geospatial buffering refined the goodness of the correlation between field data and the variables set from bands reflectance and selected salinity indices. The obtained coefficient of determination (R 2) with Multiple Linear Regression (MLR) and Partial Least Square regression (PLS) models proved an improvement in the multivariable prediction of soil salinity with the optimized Sentinel-2 data (R 2 = 0.61 and 0.68 + RMSE of 3.36 and 2.87 dS m−1 respectively), compared to the results from the 2015 multiple regression model, Given the high values of R 2 = 0.77 and 0.95 and the low values of RMSE = 2.29 and 1.18 dS m−1 respectively; the RF regressor was very efficient in predicting EC either with all variables or with selected variables from the variable importance measure based on the mean decrease accuracy (MDA). The random forest classifier performed top of the range classification with an overall Accuracy (OA) about 99%, estimating that over 75% of the study area surface is suffering from the salinization extent.

A Multi Sensor Approach to Forest Type Mapping for Advancing Monitoring of Sustainable Development Goals (SDG) in Myanmar

Monitoring forests is important for measuring overall success of the 2030 Agenda because forests play an essential role in meeting many Sustainable Development Goals (SDG), especially SDG 15. Our study evaluates the contribution of three satellite data sources (Landsat-8, Sentinel-2 and Sentinel-1) for mapping diverse forest types in Myanmar. This assessment is especially important because Myanmar is currently revising its classification system for forests and it is critical that these new forest types can be accurately mapped and monitored over time using satellite imagery. Our results show that using a combination of Sentinel-1 and Sentinel-2 yields the highest accuracy (89.6% ± 0.16 percentage point(pp)), followed by Sentinel-2 alone (87.97% ± 0.11 pp) and Landsat-8 (82.68% ± 0.13 pp). The higher spatial resolution of Sentinel-2 Blue, Green, Red, Narrow Near Infrared and Short Wave Infrared bands enhances accuracy by 4.83% compared to Landsat-8. The addition of the Sentinel-2 Near Infrared and three Vegetation Red Edge bands further improve accuracy by 0.46% compared to using only Sentinel-2 Blue, Green, Red, Narrow Near Infrared and Short Wave Infrared bands. Adding the radar information from Sentinel-1 further increases the accuracy by 1.63%. We were able to map the two major forest types, Upper Moist and Upper Dry Mixed Deciduous Forest, which comprise 90% of our study area. Accuracies for these forest types ranged from 77 to 96% depending on the sensors used, demonstrating the feasibility of using satellite data to map forest categories from a newly revised classification system. Our results advance the ongoing development of the National Forest Monitoring System (NFMS) by the Myanmar Forest Department and United Nations-Food and Agriculture Organization (UN-FAO) and facilitates future monitoring of progress towards the SDGs.

A NEW INDEX FOR IDENTIFYING WATER BODY FROM SENTINEL-2 SATELLITE REMOTE SENSING IMAGERY

Abstract. Water body plays an irreplaceable role in the global ecosystem and climate system. Sentinel-2 is a new satellite data with higher spatial and spectral resolution. Through analysing spectral characteristics of Sentinel-2 satellite imagery, the brightness of water body in vegetation red edge band and shortwave infrared band showe sharply different than that of the not water body. Therefore, a new type of water index SWI (Sentinel-2 Water Index) was proposed by combing those two bands. Four representative water types, which included Taihu Lake, the Yangtze River Estuary, the ChaKa Salt Lake and the Chain Lake, were selected as experimental areas. Normalized difference water index (NDWI) and Sentinel-2 Water Index (SWI) with Otsu method were employed to extract water body. The results showed that overall accuracy and Kappa coefficient of SWI were higher than that of NDWI and SWI was efficient index to rapidly and accurately extract water for Sentinel-2 data. Therefore, SWI had application potential for larger scale water mapping in the future.