Sentinel-1 Research Articles

Sentinel-2 and Sentinel-1 Bare Soil Temporal Mosaics of 6-year Periods for Soil Organic Carbon Content Mapping in Central France

Satellite-based soil organic carbon content (SOC) mapping over wide regions is generally hampered by the low soil sampling density and the diversity of soil sampling periods. Some unfavorable topsoil conditions, such as high moisture, rugosity, the presence of crop residues, the limited amplitude of SOC values and the limited area of bare soil when a single image is used, are also among the influencing factors. To generate a reliable SOC map, this study addresses the use of Sentinel-2 (S2) temporal mosaics of bare soil (S2Bsoil) over 6 years jointly with soil moisture products (SMPs) derived from Sentinel 1 and 2 images, SOC measurement data and other environmental covariates derived from digital elevation models, lithology maps and airborne gamma-ray data. In this study, we explore (i) the dates and periods that are preferable to construct temporal mosaics of bare soils while accounting for soil moisture and soil management; (ii) which set of covariates is more relevant to explain the SOC variability. From four sets of covariates, the best contributing set was selected, and the median SOC content along with uncertainty at 90% prediction intervals were mapped at a 25-m resolution from quantile regression forest models. The accuracy of predictions was assessed by 10-fold cross-validation, repeated five times. The models using all the covariates had the best model performance. Airborne gamma-ray thorium, slope and S2 bands (e.g., bands 6, 7, 8, 8a) and indices (e.g., calcareous sedimentary rocks, “calcl”) from the “late winter–spring” time series were the most important covariates in this model. Our results also indicated the important role of neighboring topographic distances and oblique geographic coordinates between remote sensing data and parent material. These data contributed not only to optimizing SOC mapping performance but also provided information related to long-range gradients of SOC spatial variability, which makes sense from a pedological point of view.

A Machine Learning Approach for Mapping Chlorophyll Fluorescence at Inland Wetlands

Wetlands are a critical component of the landscape for climate mitigation, adaptation, biodiversity, and human health and prosperity. Keeping an eye on wetland vegetation is crucial due to it playing a major role in the planet’s carbon cycle and ecosystem management. By measuring the chlorophyll fluorescence (ChF) emitted by plants, we can get a precise understanding of the current state and photosynthetic activity. In this study, we applied the Extreme Gradient Boost (XGBoost) algorithm to map ChF in the Biebrza Valley, which has a unique ecosystem in Europe for peatlands, as well as highly diversified flora and fauna. Our results revealed the advantages of using a set of classifiers derived from EO Sentinel-2 (S-2) satellite image mosaics to accurately map the spatio-temporal distribution of ChF in a terrestrial landscape. The validation proved that the XGBoost algorithm is quite accurate in estimating ChF with a good determination of 0.71 and least bias of 0.012. The precision of chlorophyll fluorescence measurements is reliant upon determining the optimal S-2 satellite overpass time, which is influenced by the developmental stage of the plants at various points during the growing season. Finally, the model performance results indicated that biophysical factors are characterized by greenness- and leaf-pigment-related spectral indices. However, utilizing vegetation indices based on extended periods of remote sensing data that better capture land phenology features can improve the accuracy of mapping chlorophyll fluorescence.

Comparison of Simulated Multispectral Reflectance among Four Sensors in Land Cover Classification

Multispectral images accessible free of charge have increased significantly from the acquisitions by the wide-field-of-view (WFV) sensors onboard Gaofen-1/-6 (GF-1/-6), the Operational Land Imager (OLI) onboard Landsat 8 (L8), and the Multi-Spectral Instrument (MSI) onboard Sentinel-2 (S2). These images with medium spatial resolutions are beneficial for land-cover mapping to monitor local to global surface dynamics. Comparative analyses of the four sensors in classification were made under different scenarios with five classifiers, mainly based on the simulated multispectral reflectance from well-processed hyperspectral data. With channel reflectance, differences in classification between the L8 OLI and the S2 MSI were generally dependent on the classifier considered, although the two sensors performed similarly. Meanwhile, without channels over the shortwave infrared region, the GF-1/-6 WFVs showed inferior performances. With channel reflectance, the support vector machine (SVM) with Gaussian kernel generally outperformed other classifiers. With the SVM, on average, the GF-1/-6 WFVs and the L8 OLI had great increases (more than 15%) in overall accuracy relative to using the maximum likelihood classifier (MLC), whereas the overall accuracy improvement was about 13% for the S2 MSI. Both SVM and random forest (RF) had greater overall accuracy, which partially solved the problems of imperfect channel settings. However, under the scenario with a small number of training samples, for the GF-1/-6 WFVs, the MLC showed approximate or even better performance compared to RF. Since several factors possibly influence a classifier’s performance, attention should be paid to a comparison and selection of methods. These findings were based on the simulated multispectral reflectance with focusing on spectral channel (i.e., number of channels, spectral range of the channel, and spectral response function), whereas spatial resolution and radiometric quantization were not considered. Furthermore, a limitation of this paper was largely associated with the limited spatial coverage. More case studies should be carried out with real images over areas with different geographical and environmental backgrounds. To improve the comparability in classification among different sensors, further investigations are definitely required.

Mapping Sea Grass Coverage of Tanjung Benoa Bali Using Medium Resolution Satellite Imagery Sentinel 2B

Seagrass beds are important components of a coastal ecosystem. This ecosystem serves as the primer producers of the water food chain, habitat for marine biota, produces organic carbon, and indirectly contributes to the economic well-being of coastal communities. However, the ecosystem is vulnerable to damage caused by natural factors and human activities. The objectives of this study were, firstly to identify the distribution of seagrass beds in Tanjung Benoa using Sentinel 2B satellite imagery and secondly to compare classification results from two different approaches namely pixel-based image classification and object-based image classification. Accuracy-test was carried out using field data reference of 195 sample points in the form of a 10 m X 10 m transect. The image pre-processing process was conducted with Bottom of Atmosphere (BoA) correction using the Dark Object Subtraction (DOS) method. Furthermore, the water column correction was performed using the Depth Invariant Index (DII) and the Lyzenga algorithm. The mapping results showed that the area of seagrass beds in the shallow waters of Tanjung Benoa reaches 242.99 ha. There were seven seagrass species in the study area, with an average cover of 75%. The accuracy of object-based image classification was higher than that of pixel-based classification with a difference up to 25% for six classes classification and 15% for two classes classification. Excellent results for classifying seagrasses based on cover density can be obtained when high-resolution satellite imagery and OBIA are combined with the SVM or Fuzzy Logic algorithm.

Building Change Detection with Deep Learning by Fusing Spectral and Texture Features of Multisource Remote Sensing Images: A GF-1 and Sentinel 2B Data Case

Building change detection is an important task in the remote sensing field, and the powerful feature extraction ability of the deep neural network model shows strong advantages in this task. However, the datasets used for this study are mostly three-band high-resolution remote sensing images from a single data source, and few spectral features limit the development of building change detection from multisource remote sensing images. To investigate the influence of spectral and texture features on the effect of building change detection based on deep learning, a multisource building change detection dataset (MS-HS BCD dataset) is produced in this paper using GF-1 high-resolution remote sensing images and Sentinel-2B multispectral remote sensing images. According to the different resolutions of each Sentinel-2B band, eight different multisource spectral data combinations are designed, and six advanced network models are selected for the experiments. After adding multisource spectral and texture feature data, the results show that the detection effects of the six networks improve to different degrees. Taking the MSF-Net network as an example, the F1-score and IOU improved by 0.67% and 1.09%, respectively, compared with high-resolution images, and by 7.57% and 6.21% compared with multispectral images.

Synergistic use of Sentinel-1, Sentinel-2, and Landsat 8 in predicting forest variables

Forest variables measurement is indispensable to supervising the dynamic of forest carbon storage. As commonly quantified forest variables, canopy cover (CC), diameter at breast height (DBH), and tree height (TH) were the predicted objects in our study. Recently, remote sensing technology, with high efficiency and wide temporal and spatial coverage, has been extensively used in practice. However, integrating open access satellite resources Sentinel-1 (S1), Sentinel-2 (S2), and Landsat 8 (L8) to quantify forest CC, DBH, and TH has not been extensively examined in the literature. In this study, we explored the potential of individual satellite bands/indices and the integration of different sensors in predicting mean CC, mean DBH, and mean TH in the plots dominated by four tree species (total area 102,887.6 ha). We acquired the satellite patterns of S1, S2, and L8 covered our research plots in 2017, and extracted the mean pixel values of the spectral images for every plot we detected. Specifically, seven sets of features derived from the satellite images of S1, S2, and L8 were used to predict CC, DBH, and TH for the survey plots. The results mainly indicated that group S2 + L8 (RMSEr (the relative root mean square error) = 17.9%, 21.0%, 23.0%) or S2 + L8 + S1 (RMSEr = 17. 8%, 21.0%, 23.1%) achieved the best accuracy for CC, DBH, and TH of all plots. Additionally, group S2 + L8 outperformed the other groups within two sensors and was superior to the individual sensors. Adding S1 synthetic aperture radar (SAR, the sensor carried by S1) had no evident improvement or deviation for most models, nevertheless, numerous bands/indices of SAR occupied important positions in variable importance and were been selected by stepwise regression, which denoted that it is available and valuable for forest parameters of some dominant tree species. Moreover, shortwave infrared (SWIR) bands played a crucial role in the three forest parameters prediction during the rainy season. SWIR bands (RMSEr = 20.5%), NIR bands (RMSEr = 24.1%), and red-edge bands (RMSEr = 27.1%) exerted the most significant implication for predicting CC, DBH, and TH apiece as univariate models. Green bands and NIR also occupied more positions of higher variable importance in multivariate models for estimating DBH and TH, while SWIR took up more for CC in all dominant species. Furthermore, non-parametric models outperformed the parametric model on the whole. In the meanwhile, the best models also were influenced by the data analysis methods.In a nutshell, a combination of multiple satellite sensors was proposed for forest variable selection, especially different optical sensors. SAR data were also inspired to be pitched in related forest measurement work, but it requires some data processing before applying to the research in many cases.

Satellite-Based Estimation of Soil Moisture Content in Croplands: A Case Study in Golestan Province, North of Iran

Soil moisture content (SMC) plays a critical role in soil science via its influences on agriculture, water resources management, and climate conditions. There is broad interest in finding relationships between groundwater recharge, soil characteristics, and plant properties for the quantification of SMC. The objective of this study was to assess the potential of optical satellite imagery for estimating the SMC over cropland areas. For this purpose, we collected 394 soil samples as targets in Gonbad-e Kavus in the Golestan province in the north of Iran, where a variety of crop types are cultivated. As input data, we first computed several spectral indices from Sentinel 2 (S2) and Landsat 8 (L8) images, such as the Normalized Difference Water Index (NDWI), Modified Normalized Difference Water Index (MNDWI), and Normalized Difference Salinity Index (NDSI), and then analyzed their relationships with surveyed SMC using four machine learning regression algorithms: random forests (RFs), XGBoost, extra tree decision (EDT), and support vector machine (SVM). Results revealed a high and rather similar correlation between the spectral indices and measured SMC values for both S2 and L8 data. The EDT regression algorithm yielded the highest accuracy, with an R2 = 0.82, MAE = 3.74, and RMSE = 1.08 for S2 and R2 = 0.88, RMSE = 2.42, and MAE = 1.08 for L8 images. Results also revealed that MNDWI, NDWI, and NDSI responded most sensitively to SMC estimation.

The Development of Dark Hyperspectral Absolute Calibration Model Using Extended Pseudo Invariant Calibration Sites at a Global Scale: Dark EPICS-Global

This research aimed to develop a novel dark hyperspectral absolute calibration (DAHAC) model using stable dark targets of “Global Cluster-36” (GC-36), one of the clusters from the “300 Class Global Classification”. The stable dark sites were identified from GC-36 called “Dark EPICS-Global” covering the surface types viz. dark rock, volcanic area, and dark sand. The Dark EPICS-Global shows a temporal variation of 0.02 unit reflectance. This work used the Landsat-8 (L8) Operational Land Imager (OLI), Sentinel-2A (S2A) Multispectral Instrument (MSI), and Earth Observing One (EO-1) Hyperion data for the DAHAC model development, where well-calibrated L8 and S2A were used as the reference sensors, while EO-1 Hyperion with a 10 nm spectral resolution was used as a hyperspectral library. The dark hyperspectral dataset (DaHD) was generated by combining the normalized hyperspectral profile of L8 and S2A for the DAHAC model development. The DAHAC model developed in this study takes into account the solar zenith and azimuth angles, as well as the view zenith and azimuth angles in Cartesian coordinates form. This model is capable of predicting TOA reflectance in all existing spectral bands of any sensor. The DAHAC model was then validated with the Landsat-7 (L7), Landsat-9 (L9), and Sentinel-2B (S2B) satellites from their launch dates to March 2022. These satellite sensors vary in terms of their spectral resolution, equatorial crossing time, spatial resolution, etc. The comparison between the DAHAC model and satellite measurements showed an accuracy within 0.01 unit reflectance across the overall spectral band. The proposed DAHAC model uncertainty level was determined using Monte Carlo simulation and found to be 0.04 and 0.05 unit reflectance for the VNIR and SWIR channels, respectively. The DAHAC model double ratio was used as a tool to perform the inter-comparison between two satellites. The sensor inter-comparison results for L8 and L9 showed a 2% difference and 1% for S2A and S2B across all spectral bands.

Crop Phenology Modelling Using Proximal and Satellite Sensor Data

Understanding crop phenology is crucial for predicting crop yields and identifying potential risks to food security. The objective was to investigate the effectiveness of satellite sensor data, compared to field observations and proximal sensing, in detecting crop phenological stages. Time series data from 122 winter wheat, 99 silage maize, and 77 late potato fields were analyzed during 2015–2017. The spectral signals derived from Digital Hemispherical Photographs (DHP), Disaster Monitoring Constellation (DMC), and Sentinel-2 (S2) were crop-specific and sensor-independent. Models fitted to sensor-derived fAPAR (fraction of absorbed photosynthetically active radiation) demonstrated a higher goodness of fit as compared to fCover (fraction of vegetation cover), with the best model fits obtained for maize, followed by wheat and potato. S2-derived fAPAR showed decreasing variability as the growing season progressed. The use of a double sigmoid model fit allowed defining inflection points corresponding to stem elongation (upward sigmoid) and senescence (downward sigmoid), while the upward endpoint corresponded to canopy closure and the maximum values to flowering and fruit development. Furthermore, increasing the frequency of sensor revisits is beneficial for detecting short-duration crop phenological stages. The results have implications for data assimilation to improve crop yield forecasting and agri-environmental modeling.

Identification of tea plantations in typical plateau areas with the combination of Sentinel-1/2 optical and radar remote sensing data based on feature selection algorithm

ABSTRACT Efficiently and accurately identifying the spatial distribution of tea plantations in the subtropical plateau regions of southwest China is of great significance for ecological and environmental protection. However, the lands of those regions are fragmented with complex vegetation types. Moreover, there is much cloudy and rainy weather over those areas, making it very difficult to identify tea plantations using only optical remote sensing data. In order to solve these problems, this paper uses Sentinel-1 (S1) Synthetic Aperture Radar (SAR) data and Sentinel-2 (S2) optical data to design seven classification feature combinations to explore the influence of red edge features, radar features and texture features on the identification accuracy of tea plantations. The feasibility of Jeffreys-Matusita distance (JM) feature selection and Recursive Feature Elimination (RFE) feature selection algorithm to find the optimal feature combination is verified, and the distribution of tea plantations in the study area is acquired by using the object-oriented random forest algorithm. The study shows that (1) the combination of SAR data and optical data can effectively improve the identification accuracy of tea plantations. (2) S2 red edge features and S1 radar features can significantly improve the accuracy of the identification results of tea plantations. (3) After applying the JM distance and RFE feature selection algorithms, the producer’s accuracy of tea plantations is improved by 1.39% and 2.38%, and the user’s accuracy is improved by 1.02% and 1.3%, respectively, compared with the identification of all features. The overall accuracy of the random forest algorithm combined with RFE is 93.43%. This study proposes the application of feature selection algorithms in identification of tea plantations, which improves accuracy and increases efficiency while minimizing redundant features and provides an effective approach to identify tea plantations in cloudy and rainy areas in the subtropical plateau of southern China.

Incorporation of Fused Remote Sensing Imagery to Enhance Soil Organic Carbon Spatial Prediction in an Agricultural Area in Yellow River Basin, China

To overcome spatial, spectral and temporal constraints of different remote sensing products, data fusion is a good technique to improve the prediction capability of soil prediction models. However, few studies have analyzed the effects of image fusion on digital soil mapping (DSM) models. This research fused multispectral (MS) and panchromatic Landsat 8 (L8) bands, and MS Sentinel 2 (S2) and panchromatic L8 bands using the Brovey, Intensity–Hue–Saturation and Gram–Schmidt methods in an agricultural area in Yellow River Basin, China. To analyze the effects of image fusion on DSM models, various SOC prediction models derived from remote sensing image datasets were established by the random forest method. Soil salinity indices and spectral reflectance from all the remote sensing data had relatively strong negative correlations with SOC, and vegetation indices and water indices from all the remote sensing data had relatively strong positive correlations with SOC. Soil moisture and vegetation were the main controlling factors of the SOC spatial pattern in the study area. More spectral indices derived from pansharpened L8 and fused S2–L8 images by all three image fusion methods had stronger relationships with SOC compared with those from MS L8 and MS S2, respectively. All the SOC models established by pansharpened L8 and fused S2–L8 images had higher prediction accuracy than those established by MS L8 and MS S2, respectively. The fusion between S2 and L8 bands had stronger effects on enhancing the prediction accuracy of SOC models compared with the fusion between panchromatic and MS L8 bands. It is concluded that digital soil mapping and image fusion can be utilized to increase the prediction performance of SOC spatial prediction models.

Multisensor Satellite Data and Field Studies for Unravelling the Structural Evolution and Gold Metallogeny of the Gerf Ophiolitic Nappe, Eastern Desert, Egypt

The gold mineralization located in the southern Eastern Desert of Egypt mostly occurs in characteristic geologic and structural settings. The gold-bearing quartz veins and the alteration zones are confined to the ductile shear zones between the highly deformed ophiolitic blocks, sheared metavolcanics, and gabbro-diorite rocks. The present study attempts to integrate multisensor remotely sensed data, structural analysis, and field investigation in unraveling the geologic and structural controls of gold mineralization in the Gabal Gerf area. Multispectral optical sensors of Landsat-8 OLI/TIRS (L8) and Sentinel-2B (S2B) were processed to map the lithologic rock units in the study area. Image processing algorithms including false color composite (FCC), band ratio (BR), principal component analysis (PCA), minimum noise fraction (MNF), and Maximum Likelihood Classifier (MLC) were effective in producing a comprehensive geologic map of the area. The mafic index (MI) = (B13-0.9147) × (B10-1.4366) of ASTER (A) thermal bands and a combined band ratio of S2B and ASTER of (S2B3+A9)/(S2B12+A8) were dramatically successful in discriminating the ophiolitic assemblage, that are considered the favorable lithology for the gold mineralization. Three alteration zones of argillic, phyllic and propylitic were spatially recognized using the mineral indices and constrained energy minimization (CEM) approach to ASTER data. The datasets of ALSO PALSAR and Sentinel-1B were subjected to PCA and filtering to extract the lineaments and their spatial densities in the area. Furthermore, the structural analysis revealed that the area has been subjected to three main phases of deformation; (i) NE-SW convergence and sinistral transpression (D2); (ii) ~E-W far field compressional regime (D3), and (iii) extensional tectonics and terrane exhumation (D4). The gold-bearing quartz veins in several occurrences are controlled by D2 and D3 shear zones that cut heterogeneously deformed serpentinites, sheared metavolcanic rocks and gabbro-diorite intrusions. The information extracted from remotely sensed data, structural interpretation and fieldwork were used to produce a gold mineralization potential zones map which was verified by reference and field observations. The present study demonstrates the remote sensing capabilities for the identification of alteration zones and structural controls of the gold mineralization in highly deformed ophiolitic regions.

Appropriateness of Land and Crop Selection for an Area in Iraq Al-Salamiyat

Al-Salamiyat project, which is situated inside the administrative boundaries of Baghdad Governorate and has a total size estimated at 2621 hectares, required a study to assess the lands for cereal crops (wheat, barley, and maize). The study region was visited and examined repeatedly in order to set up and gather data; take soil models; observe the land cover; the level of soil salinity; the kind of agricultural use; and the planted crops. The internodes project was covered by satellite imagery retrieved from the satellite sentinel (2B) Al-Mashhad (SMC) for the month of March 2021. The soil model’s physical and chemical characteristics were estimated. The analyses’ findings demonstrated that the research area’s soils had a degree of interaction (PH), as estimated values for this characteristic ranged from 7.05 for soil model No. 8 to 7.9 for model No. 11, or neutral to medium basal. Because most of the study area’s lands are used for agriculture and have a high percentage of vegetative cover, the majority of the soils there contain modest amounts to moderately moderate salt. Carbonate minerals factor: According to Sys et alsuggestions.’s from 1993, the carbonate minerals factor estimations for wheat and barley harvests were 95% for all soils of the study area, and the final estimates for maize yield ranged from 60% to 85%. According to the findings, the estimations of the amount of exchanged sodium in the study area’s soils were a non-specific factor for the growth of the chosen crops and were given an estimate higher than 95%.. The evaluation of the soil’s properties for the research area’s specified crops showed that the (S1) variety is extremely suited for growing wheat, barley, and yellow corn due to the absence of determinants, while the (S2) class is moderate and acceptable for growing wheat and maize due to high salinity.

Synergy of Sentinel-1 and Sentinel-2 Time Series for Cloud-Free Vegetation Water Content Mapping with Multi-Output Gaussian Processes

Optical Earth Observation is often limited by weather conditions such as cloudiness. Radar sensors have the potential to overcome these limitations, however, due to the complex radar-surface interaction, the retrieving of crop biophysical variables using this technology remains an open challenge. Aiming to simultaneously benefit from the optical domain background and the all-weather imagery provided by radar systems, we propose a data fusion approach focused on the cross-correlation between radar and optical data streams. To do so, we analyzed several multiple-output Gaussian processes (MOGP) models and their ability to fuse efficiently Sentinel-1 (S1) Radar Vegetation Index (RVI) and Sentinel-2 (S2) vegetation water content (VWC) time series over a dry agri-environment in southern Argentina. MOGP models not only exploit the auto-correlations of S1 and S2 data streams independently but also the inter-channel cross-correlations. The S1 RVI and S2 VWC time series at the selected study sites being the inputs of the MOGP models proved to be closely correlated. Regarding the set of assessed models, the Convolutional Gaussian model (CONV) delivered noteworthy accurate data fusion results over winter wheat croplands belonging to the 2020 and 2021 campaigns (NRMSEwheat2020 = 16.1%; NRMSEwheat2021 = 10.1%). Posteriorly, we removed S2 observations from the S1 & S2 dataset corresponding to the complete phenological cycles of winter wheat from September to the end of December to simulate the presence of clouds in the scenes and applied the CONV model at the pixel level to reconstruct spatiotemporally-latent VWC maps. After applying the fusion strategy, the phenology of winter wheat was successfully recovered in the absence of optical data. Strong correlations were obtained between S2 VWC and S1 & S2 MOGP VWC reconstructed maps for the assessment dates (R2¯wheat−2020 = 0.95, R2¯wheat−2021 = 0.96). Altogether, the fusion of S1 SAR and S2 optical EO data streams with MOGP offers a powerful innovative approach for cropland trait monitoring over cloudy high-latitude regions.

Soil moisture estimates at 1 km resolution making a synergistic use of Sentinel data

Abstract. Very high-resolution (∼10–100 m) surface soil moisture (SM) observations are important for applications in agriculture, among other purposes. This is the original goal of the S2MP (Sentinel-1/Sentinel-2-Derived Soil Moisture Product) algorithm, which was designed to retrieve surface SM at the agricultural plot scale by simultaneously using Sentinel-1 (S1) backscatter coefficients and Sentinel-2 (S2) NDVI (Normalized Difference Vegetation Index) as inputs to a neural network trained with Water Cloud Model simulations. However, for many applications, including hydrology and climate impact assessment at regional level, large maps with a high resolution (HR) of around 1 km are already a significant improvement with respect to most of the publicly available SM datasets, which have resolutions of about 25 km. In this study, the S2MP algorithm was adapted to work at 1 km resolution and extended from croplands to herbaceous vegetation types. A target resolution of 1 km also allows the evaluation of the interest in using NDVI derived from Sentinel-3 (S3) instead of S2. Two sets of SM maps at 1 km resolution were produced with S2MP over six regions of ∼104 km2 in Spain, Tunisia, North America, Australia, and the southwest and southeast regions of France for the whole year of 2019. The first set was derived from the combination of S1 and S2 data (S1 + S2 maps), while the second one was derived from the combination of S1 and S3 (S1 + S3 maps). S1 + S2 and S1 + S3 SM maps were compared to each other, to those of the 1 km resolution Copernicus Global Land Service (CGLS) SM and Soil Water Index (SWI) datasets, and to those of the Soil Moisture Active Passive (SMAP) + S1 product. The S2MP S1 + S2 and S1 + S3 SM maps are in very good agreement in terms of correlation (R≥0.9), bias (≤0.04 m3 m−3), and standard deviation of the difference (SDD≤0.03 m3 m−3) over the six domains investigated in this study. In a second step, the S1 + S3 S2MP maps were compared to the other HR maps. S1 + S3 SM maps are well correlated to the CGLS SM maps (R∼0.7–0.8), but the correlations with respect to the other HR maps (CGLS SWI and SMAP + S1) drop significantly over many areas of the six domains investigated in this study. The highest correlations between the HR maps were found over croplands and when the 1 km pixels have a very homogeneous land cover. The bias among the different maps was found to be significant over some areas of the six domains, reaching values of ±0.1 m3 m−3. The S1 + S3 maps show a lower SDD with respect to CGLS maps (≤0.06 m3 m−3) than with respect to the SMAP + S1 maps (≤0.1 m3 m−3) for all the six domains. Finally, all the HR datasets (S1 + S2, S1 + S3, CGLS, and SMAP + S1) were also compared to in situ measurements from five networks across five countries, along with coarse-resolution (CR) SM products from SMAP, SMOS, and the European Space Agency Climate Change Initiative (CCI). While all the CR and HR products show different bias and SDD, the HR products show lower correlations than the CR ones with respect to in situ measurements. The discrepancies in between the different HR datasets, except for the more simple land cover conditions (homogeneous pixels with croplands) and the lower performances with respect to in situ measurement than coarse-resolution datasets, show the remaining challenges for large-scale HR SM mapping.

The utility of airborne hyperspectral and satellite multispectral images in identifying Natura 2000 non-forest habitats for conservation purposes

Aerial hyperspectral and multispectral satellite data are the two most commonly used datasets to identify natural and semi-natural vegetation. However, there is no documented analysis based on data from several areas concerning the difference in the classification accuracy of non-forest Natura 2000 habitat with the use of aerial hyperspectral and satellite multispectral data. Also, there is no recommendation, on which habitat can be classified with sufficient accuracy using free multispectral images. This study aimed to analyse the difference in classification accuracy of Natura 2000 habitats representing: meadows, grasslands, heaths and mires between data with different spectral resolutions and the results utility for nature conservation compared to conventional maps. The analysis was conducted in five study areas in Poland. The classification was performed on multispectral Sentinel-2 (S2) and hyperspectral HySpex (HS) images using the Random Forest algorithm. Based on the results, it can be stated that the use of HS data resulted in higher classification accuracy, on average 0.14, than using S2 images, regardless of the area of the habitat. However, the difference in accuracy was not constant, varying by area and habitat characterisation. Greater differences in accuracy were observed for areas where habitats were characterised by high α-diversity or β-diversity. The HS and S2 data make it possible to create maps that provide a great deal of new knowledge about the distribution of Natura 2000 habitats, which is necessary for the management of protected areas. The obtained results indicate that by using S2 images it is possible to identify, at a satisfactory level, alluvial meadows and grassland. For heaths and mires, using HS data improved the results, but it is also possible to acquire general distribution of these classes, whereas HS images are obligatory for mapping salt, Molinia and lowland hay meadows.

Optimizing Wheat Yield Prediction Integrating Data from Sentinel-1 and Sentinel-2 with CatBoost Algorithm

Accurately estimating wheat yield is crucial for informed decision making in precision agriculture (PA) and improving crop management. In recent years, optical satellite-derived vegetation indices (Vis), such as Sentinel-2 (S2), have become widely used, but the availability of images depends on the weather conditions. For its part, Sentinel-1 (S1) backscatter data are less used in agriculture due to its complicated interpretation and processing, but is not impacted by weather. This study investigates the potential benefits of combining S1 and S2 data and evaluates the performance of the categorical boosting (CatBoost) algorithm in crop yield estimation. The study was conducted utilizing dense yield data from a yield monitor, obtained from 39 wheat (Triticum spp. L.) fields. The study analyzed three S2 images corresponding to different crop growth stages (GS) GS30, GS39-49, and GS69-75, and 13 Vis commonly used for wheat yield estimation were calculated for each image. In addition, three S1 images that were temporally close to the S2 images were acquired, and the vertical-vertical (VV) and vertical-horizontal (VH) backscatter were calculated. The performance of the CatBoost algorithm was compared to that of multiple linear regression (MLR), support vector machine (SVM), and random forest (RF) algorithms in crop yield estimation. The results showed that the combination of S1 and S2 data with the CatBoost algorithm produced a yield prediction with a root mean squared error (RMSE) of 0.24 t ha−1, a relative RMSE (rRMSE) 3.46% and an R2 of 0.95. The result indicates a decrease of 30% in RMSE when compared to using S2 alone. However, when this algorithm was used to estimate the yield of a whole plot, leveraging information from the surrounding plots, the mean absolute error (MAE) was 0.31 t ha−1 which means a mean error of 4.38%. Accurate wheat yield estimation with a spatial resolution of 10 m becomes feasible when utilizing satellite data combined with CatBoost.

Rapid early-season maize mapping without crop labels

Maize (Zea mays), the second most-produced crop worldwide, serves as the cornerstone for global food security and human livelihood. Early-season maize mapping benefits maize production forecasting and other pre-harvest decision-making applications. However, most existing early-season mapping efforts rely heavily on either the current-year or historical crop labels to train classifiers, limiting the potential applications to new regions lacking crop labels. To explore the possibility of maize mapping only using satellite data in the early season, we proposed a Multi-temporal Gaussian Mixture Model (MGMM) to map maize planting areas without any crop labels. A chlorophyll content relevant proxy, named the Red-edge position (REP), was selected as model input, based on the truth that summer maize tends to show a higher chlorophyll content than other summer crops (e.g., soybean, cotton, peanut, sunflowers, etc.) in the peak season. The novel early-season mapping framework using the REP-based MGMM (MGMM-REP) was applied in four diverse areas (Iowa and Georgia in the US, Heilongjiang province (HLJ) in China, and Grand-Est in France). The MGMM-REP could generate maize maps more than two months before harvest with reasonable accuracy (F1 ≥ 77%) using all the available Sentinel-2 (S2) images and the Google Earth Engine platform (GEE). Our early-season maps agreed well with the existing crop maps and official statistics. The correlation coefficient (R) of the maize acreage between our early-season maps and statistics was higher than 0.94. The high inter-class difference of REP between maize and other summer crops could increase the F1 score by 2–47% compared to the other commonly used Vegetation indices (VIs). Since MGMM-REP does not rely on crop labels, it had the potential to be transferred to label-scarce maize-cropped regions and contribute to the international commodity trade and food security forecast.

Deep ResU-Net Convolutional Neural Networks Segmentation for Smallholder Paddy Rice Mapping Using Sentinel 1 SAR and Sentinel 2 Optical Imagery

Rice is a globally significant staple food crop. Therefore, it is crucial to have adequate tools for monitoring changes in the extent of rice paddy cultivation. Such a system would require a sustainable and operational workflow that employs open-source medium to high spatial and temporal resolution satellite imagery and efficient classification techniques. This study used similar phenological data from Sentinel-2 (S2) optical and Sentinel-1 (S1) Synthetic Aperture Radar (SAR) satellite imagery to identify paddy rice distribution with deep learning (DL) techniques. Using Google Earth Engine (GEE) and U-Net Convolutional Neural Networks (CNN) segmentation, a workflow that accurately delineates smallholder paddy rice fields using multi-temporal S1 SAR and S2 optical imagery was investigated. The study′s accuracy assessment results showed that the optimal dataset for paddy rice mapping was a fusion of S2 multispectral bands (visible and near infra-red (VNIR), red edge (RE) and short-wave infrared (SWIR)), and S1-SAR dual polarization bands (VH and VV) captured within the crop growing season (i.e., vegetative, reproductive, and ripening). Compared to the random forest (RF) classification, the DL model (i.e., ResU-Net) had an overall accuracy of 94% (three percent higher than the RF prediction). The ResU-Net paddy rice prediction had an F1-Score of 0.92 compared to 0.84 for the RF classification generated using 500 trees in the model. Using the optimal U-Net classified paddy rice maps for the dates analyzed (i.e., 2016–2020), a change detection analysis over two epochs (2016 to 2018 and 2018 to 2020) provided a better understanding of the spatial–temporal dynamics of paddy rice agriculture in the study area. The results indicated that 377,895 and 8551 hectares of paddy rice fields were converted to other land-use over the first (2016–2018) and second (2018–2020) epochs. These statistics provided valuable insight into the paddy rice field distribution changes across the selected districts analyzed. The proposed DL framework has the potential to be upscaled and transferred to other regions. The results indicated that the approach could accurately identify paddy rice fields locally, improve decision making, and support food security in the region.

Multispectral remote sensing of potential groundwater dependent vegetation in the greater Floristic region of the Western Cape, South Africa

ABSTRACT Groundwater dependent vegetation (GDV) is increasingly threatened by the transformation of the natural environment. An understanding of the nature of GDV at the appropriate scale helps environmental managers make suitable decisions. This study assesses the potential for mapping the distribution of GDV within the Heuningnes Catchment using multispectral remotely sensed data (i.e., Landsat 8 (L8) and Sentinel 2 (S2)), the derived vegetation indices (Normalised Difference Vegetation Index (NDVI) and the Soil-Adjusted Vegetation Index (SAVI)) and in-situ data. The GDV distribution maps were produced by integrating vegetation productivity, landcover, and topographic layers as GDV indicators. The findings of the study revealed that the spectral indices had a significant influence on the sensor’s GDV classification performance. Specifically, the S2-derived SAVI mapped the GDV areas with the highest overall accuracy (97%), followed by the S2-derived NDVI, with an accuracy of 95%. Comparatively, the L8(NDVI) GDV map was achieved with an overall accuracy of 92% and the L8(SAVI) map had an overall accuracy of 96%. The estimated coverage of potential GDV within the catchment ranges between 2.34 and 2.60%. This work demonstrated the capabilities of a combined remote sensing and GIS methodological framework, which can improve our knowledge on GDV.