High Resolution Satellite Sensors Research Articles

Tracking the Dynamics of Spartina alterniflora with WorldView-2/3 and Sentinel-1/2 Imagery in Zhangjiang Estuary, China

The invasion of Spartina alterniflora (S. alterniflora) has posed serious threats to the sustainability, quality and biodiversity of coastal wetlands. To safeguard coastal ecosystems, China has enacted large-scale S. alterniflora removal projects, which set the goal of effectively controlling S. alterniflora throughout China by 2025. The accurate monitoring of S. alterniflora with remote sensing is urgent and requisite for the scientific eradication, control and management of this invasive plant. In this study, we combined multi-temporal WorldView-2/3 (WV-2/3) and Sentinel-1/2 imagery to monitor the S. alterniflora dynamics before and after the S. alterniflora removal projects in Zhangjiang Estuary. We put forward a new method for S. alterniflora detection with eight-band WV-2/3 imagery. The proposed method first used NDVI to discriminate S. alterniflora from water, mud flats and mangroves based on Ostu thresholding and then used the red-edge, NIR1 and NIR2 bands and support vector machine (SVM) classifier to distinguish S. alterniflora from algae. Due to the contamination of frequent cloud cover and tidal inundation, the long revisit time of high-resolution satellite sensors and the short-term S. alterniflora removal projects, we combined Sentinel-1 SAR time series and Sentinel-2 optical imagery to monitor the S. alterniflora removal project status in 2023. The overall accuracies of the S. alterniflora detection results here are above 90%. Compared with the traditional SVM method, the proposed method achieved significantly higher identification accuracy. The S. alterniflora area was 115.19 hm2 in 2015, 152.40 hm2 in 2017 and 15.29 hm2 in 2023, respectively. The generated S. alterniflora maps clearly show the clonal growth of S. alterniflora in Zhangjiang Estuary from 2015 to 2017, and the large-scale S. alterniflora eradication project has achieved remarkable results with a removal rate of about 90% in the study area. With the continuous implementation of the “Special Action Plan for the Prevention and Control of Spartina alterniflora (2022–2025)” which aims to eliminate more than 90% of S. alterniflora in all provinces in China by 2025, the continual high-spatial resolution monitoring of S. alterniflora is crucial to control secondary invasion and restore coastal wetlands.

Enhancing detection of volcanic ash clouds from space with convolutional neural networks

Volcanic ash cloud detection is a crucial component of volcano monitoring and a valuable tool for investigating ash cloud dispersion, which is paramount for enhancing the safety of human settlements and air traffic. The latest generation of high-resolution satellite sensors (e.g., EUMETSAT MSG Spinning Enhanced Visible and InfraRed Imager, SEVIRI) provides radiometric estimates for monitoring volcanic clouds on a global scale efficiently and timely. However, these radiometric intensities are not always discriminative enough to detect volcanic ash clouds due to the spectral limitations of these instruments and the complex nature of some volcanic clouds, such as low concentration resulting in an averaged detected radiometric estimate comparable to the background. Here, we evaluate the ability of a Convolutional Neural Network (CNN) to detect and track the dispersion of volcanic ash clouds into the atmosphere, exploiting a variety of spatial and spectral intensity information mainly coming from SEVIRI Ash RGB images. We train a deep CNN model through transfer learning, and demonstrate that the trained models overcome the limitations of algorithms based solely on pixel intensity, whether traditional or machine learning, resulting in increased performance compared to other methods. We illustrate the operation of this model using the paroxysmal explosive events that occurred at Mt. Etna between 2020 and 2022.

Extraction of Coastal Levees Using U-Net Model with Visible and Topographic Images Observed by High-Resolution Satellite Sensors.

Coastal levees play a role in protecting coastal areas from storm surges and high waves, and they provide important input information for inundation damage simulations. However, coastal levee data with uniformity and sufficient accuracy for inundation simulations are not always well developed. Against this background, this study proposed a method to extract coastal levees by inputting high spatial resolution optical satellite image products (RGB images, digital surface models (DSMs), and slope images that can be generated from DSM images), which have high data availability at the locations and times required for simulation, into a deep learning model. The model is based on U-Net, and post-processing for noise removal was introduced to further improve its accuracy. We also proposed a method to calculate levee height using a local maximum filter by giving DSM values to the extracted levee pixels. The validation was conducted in the coastal area of Ibaraki Prefecture in Japan as a test area. The levee mask images for training were manually created by combining these data with satellite images and Google Street View, because the levee GIS data created by the Ibaraki Prefectural Government were incomplete in some parts. First, the deep learning models were compared and evaluated, and it was shown that U-Net was more accurate than Pix2Pix and BBS-Net in identifying levees. Next, three cases of input images were evaluated: (Case 1) RGB image only, (Case 2) RGB and DSM images, and (Case 3) RGB, DSM, and slope images. Case 3 was found to be the most accurate, with an average Matthews correlation coefficient of 0.674. The effectiveness of noise removal post-processing was also demonstrated. In addition, an example of the calculation of levee heights was presented and evaluated for validity. In conclusion, this method was shown to be effective in extracting coastal levees. The evaluation of generalizability and use in actual inundation simulations are future tasks.

Downscaling Land Surface Temperature Derived from Microwave Observations with the Super-Resolution Reconstruction Method: A Case Study in the CONUS

Optical sensors cannot penetrate clouds and can cause serious missing data problems in optical-based Land Surface Temperature (LST) products. Under cloudy conditions, microwave observations are usually utilized to derive the land surface temperature. However, microwave sensors usually have coarse spatial resolutions. High-Resolution (HR) LST data products are usually desired for many applications. Instead of developing and launching new high-resolution satellite sensors for LST observations, a more economical and practical way is to develop proper methodologies to derive high-resolution LSTs from available Low-Resolution (LR) datasets. This study explores different algorithms to downscale low-resolution LST data to a high resolution. The existing regression-based downscaling methods usually require simultaneous observations and ancillary data. The Super-Resolution Reconstruction (SRR) method developed for traditional image enhancement can be applicable to high-resolution LST generation. For the first time, we adapted the SRR method for LST data. We specifically built a unique database of LSTs for the example-based SRR method. After deriving the LST data from the coarse-resolution passive microwave observations, the AMSR-E at 25 km and/or AMSR-2 at 10 km, we developed an algorithm to downscale them to a 1 km spatial resolution with the SRR method. The SRR downscaling algorithm can be implemented to obtain high-resolution LSTs without auxiliary data or any concurrent observations. The high-resolution LSTs are validated and evaluated with the ground measurements from the Surface Radiation (SURFRAD) Budget Network. The results demonstrate that the downscaled microwave LSTs have a high correlation coefficient of over 0.92, a small bias of less than 0.5 K, but a large Root Mean Square Error (RMSE) of about 4 K, which is similar to the original microwave LST, so the errors in the downscaled LST could have been inherited from the original microwave LSTs. The validation results also indicate that the example-based method shows a better performance than the self-similarity-based algorithm.

Land cover and crop types mapping using different spatial resolution imagery in a Mediterranean irrigated area.

Crop type identification is critical for agricultural sustainability policy development and environmental assessments. Therefore, it is important to obtain their spatial distribution via different approaches. Medium-, high- and very high-resolution optical satellite sensors are efficient tools for acquiring this information, particularly for challenging studies such as those conducted in heterogeneous agricultural fields. This research examined the ability of four multitemporal datasets (Sentinel-1-SAR (S1), Sentinel-2-MSI (S2), RapidEye (RE), and PlanetScope (PS)) to identify land cover and crop types (LCCT) in a Mediterranean irrigated area. To map LCCT distribution, a supervised pixel-based classification is adopted using Support Vector Machine with a radial basis function kernel (SVMRB) and Random Forest (RF). Thus, LCCT maps were generated into three levels, including six (Level I), ten (Level II), and fourteen (Level III) classes. Overall, the findings revealed high overall accuracies of >92%, >83%, and > 81% for Level I, Level II, and Level III, respectively, except for Sentinel-1. It was found that accuracy improves considerably when the number of classes decreases, especially when cropland or non-cropland classes are grouped into one. Furthermore, there was a similarity in performance between S2 alone and S1S2. PlanetScope LCCT classifications outperform other sensors. In addition, the present study demonstrated that SVM achieved better performances against RF and can thereby effectively extract LCCT information from high-resolution imagery as PlanetScope.

Fine-scale mapping of Spartina alterniflora-invaded mangrove forests with multi-temporal WorldView-Sentinel-2 data fusion

Mangrove forests, a critical component of blue carbon ecosystems, make a significant contribution to mitigating climate change and maintaining biodiversity. But in China, they have been suffering from severe invasion by Spartina alterniflora (S. alterniflora). Accurate information on spatial distribution of mangroves and S. alterniflora is a prerequisite to protecting indigenous mangrove ecosystems and managing S. alterniflora invasion. However, the classification performance is often constrained by two major issues: (1) A single-date image, regardless of acquisition time, hardly provides optimal spectral separability among evergreen mangroves, deciduous S. alterniflora and mudflats; (2) Due to the gradual invasion process, the mixtures between land cover classes are prevalent in moderate-resolution images. The use of high-resolution (finer than 2 m) image time series could effectively solve these two issues. However, due to frequent cloud cover in coastal areas and long revisit time of high-resolution satellite sensors, the collection of high-resolution image time series is still a great challenge, let alone substantial expenses. In this study, we proposed a new multi-temporal classification approach to fine-scale mapping of exotic S. alterniflora and native mangroves using synthetic high-resolution image time series derived from spatiotemporal fusion of WorldView-2 and Sentinel-2 data. A total of six synthetic WorldView-2-like images in different phenological periods were produced by the Flexible Spatiotemporal DAta Fusion (FSDAF) method. With the combination of the original and synthetic images, both types of image-based multi-temporal features and pixel-based image compositing features were extracted and evaluated in the following Random Forest classifications. Monte Carlo cross-validation was adopted to provide robust accuracy estimates, and a spatially explicit confidence map was generated to show classification uncertainty and potential errors. In order to further demonstrate the necessity of generating synthetic WorldView-2 images, the entire classification process was repeated using Sentinel-2 images at 10-m spatial resolution and the comparison between fine- and moderate-resolution classification results was made in terms of classification accuracies, classification maps and fractional abundances. Our results show that the spectral profiles of S. alterniflora and mangroves generated from the synthetic WorldView-2 images were consistent with the ones from the paired Sentinel-2 images. Meanwhile, the spatial details of high landscape heterogeneity, unrecognizable in the Sentinel-2 images, were revealed by the synthetic WorldView-2 images. Compared to the single-date classification with the original WorldView-2 image, most of multi-temporal classifications using both the original and synthetic images provided more accurate results, with the overall accuracy raised by up to 3.9%. Among various temporal features adopted in this study, the use of the image-based, stacked spectral reflectance and vegetation index features of all original and synthetic images produced the best classification performance, increasing the producer's and user's accuracies of mangroves, S. alterniflora and mudflats by 2.5%–7.1%. Finally, the Sentinel-2 classification accuracies were generally lower than the WorldView-2 counterparts across all temporal features, with a 6% difference in the highest overall accuracy between the two. Prediction errors of the moderate-resolution classification were often found in heterogeneous landscapes, where S. alterniflora was mixed with mangroves or mudflats. Considering the availability of moderate-resolution data with high revisit frequencies, such as Sentinel-2A/B and Landsat 8/9, future fine-scale mapping exercises might benefit from the multi-temporal classification approach with spatiotemporal fusion of moderate- and high-resolution data, alleviating the problem of the scarcity of high-resolution data in cloudy areas.

Global declines of offshore gas flaring inadequate to meet the 2030 goal

Monitoring of gas flaring (GF)—the burning of natural gas associated with oil extraction—over most oil- and gas-producing areas is challenging due to high costs or difficult field investigations. As GF contributes to both global warming and air pollution, an up-to-date picture (locations, emissions and trends) of global offshore GF can help countries’ energy decarbonization efforts substantially. Although high-resolution satellite sensors regularly capture high-temperature signals from GF, retrieving spatially explicit information and estimating GF volumes from petabyte images remain challenging. Here we developed a monitoring framework and compiled a 20 m resolution inventory of offshore GF sites by analysing ~8.53 million Sentinel-2 images. A robust model (R2 > 0.99) was established to estimate offshore GF volumes from Sentinel-2 metrics. Our findings reveal that Sentinel-2 can pinpoint global offshore GFs to support scientifically sound decision-making; a vital few (~20%) sites are responsible for >80% of offshore GF volumes, calling for more targeted regulations; and the offshore GF volumes have declined by 26.4% from 2016 to 2021. We conclude that the Zero Routine Flaring Initiative by the Global Gas Flaring Reduction Partnership committing governments and oil companies to end routine flaring by no later than 2030 could be 5 years behind schedule. Monitoring of gas flaring (GF) can be expensive and practically difficult, but better information about global offshore GF is needed to inform decarbonization policies. This study presents a monitoring framework, a detailed inventory of offshore GF sites and estimates of GF volumes globally.

A preliminary assessment of Sentinel-3 ocean and land color instrument data for the estimation of chlorophyll-a concentration using bio-optical methods in Annaba Bay and El Kala’s coast (Algerian Basin)

The use of bio-optical methods and high spatial and spectral resolution satellite sensors has enabled the estimation of chlorophyll-a (Chl-a) on a large spatiotemporal scale. However, their application on a regional scale necessitates validation and, in certain cases, calibration, especially when it comes to coastal waters. This study was carried out in the eastern Algerian coastal waters to evaluate the relevance of Sentinel-3 satellite data in Annaba Bay and El Kala’s coast. Sentinel-3 OLCI (Ocean and Land Color Instrument) Level-2 products (OC4Me and Chl-a NN) and the Chl-a generated by applying the C2RCC processor on Level-1B (Chl-a C2RCC) were compared to the measured Chl-a at fixed locations. In addition, nine other Chl-a algorithms have been tested. Finally, six of those later were locally calibrated. The results demonstrate the good performance of the initially available products (Chl-a C2RCC and Chl-a NN), followed by the blue–green algorithms (OC4, OC5, OC6, MedOC4, and ADOC4), whereas the near-infrared (NIR) algorithms (2B-OLCI, 3B-OLCI, and G2B) show underwhelming performance and very poor prediction of the Chl-a concentration. Even though the OC4Me has the best correlation (r = 0.9820), it also has the highest MAE, RMSD, and BIAS (0.628, 1.264, and 0.5696, respectively). The fit of the NIR methods has been significantly refined. Nevertheless, their performance is still inadequate for the estimation of the Chl-a in our region. The performance of the calibrated OCx was improved, with the OC6-AW at the top of the list (r = 0.9881, MAE = 0.08). The latter outperforms the existing ones, which improves the estimation of Chl-a based on the Sentinel-3 OLCI data in our study area.

EMS-CDNet: an efficient multi-scale-fusion change detection network for very high-resolution remote sensing images

ABSTRACT Remote sensing image change detection (RSICD) is an essential measure for monitoring the earth’s surface changes. In recent years, the explosive growth of very high-resolution (VHR) satellite sensors and the booming innovations in deep learning technology have significantly boosted RSICD development. However, most of the current RSICD models focus on locating accurate change areas while ignoring the efficiency of their method, which limits the practical application of RSICD models, especially for large-scale and emergency RSICD tasks. In this paper, we propose an Efficient Multi-scale-fusion Change Detection Network (EMS-CDNet) for bi-temporal RSICD tasks. Our EMS-CDNet pays more attention to the model’s inference speed and the accuracy-efficiency trade-off rather than only pursuing detection accuracy. We designed a multi-scale fusion module for EMS-CDNet, which adopts multi-scale and multi-branch operations to extract multi-scale features simultaneously and aggregate features at different feature levels. In addition to EMS-CDNet’s ability to achieve sufficient feature extraction, the multi-scale image input within the designed module alleviates the influence of image registration errors in practical applications, thereby strengthening EMS-CDNet’s value for practical RSICD tasks. We also integrated a novel partition unit in EMS-CDNet to lighten the model while maintaining the detection ability of small targets, thus shortening its processing time without a severe accuracy decrease. We conducted experiments on two state-of-the-art (SOTA) public RSICD datasets and our own collected dataset. The public datasets were utilized to comparatively measure the overall accuracy and efficiency measurement of EMS-CDNet, and the dataset of images we collected was used to observe EMS-CDNet’s performance under the influence of image registration errors. Our experimental results show that EMS-CDNet achieved a better accuracy-efficiency trade-off than the SOTA public datasets methods. For example, EMS-CDNet reduced the inference time by about 33% while maintaining identical detection accuracy to CLNet (the optimal method among the comparison methods). Furthermore, EMS-CDNet achieved higher accuracy on our collected dataset, with an F1 of 74% and mIoU of 0.806, demonstrating its robustness to image registration errors and showing its value for practical RSICD applications.

Monitoring Dryland Trees With Remote Sensing. Part A: Beyond CORONA—Historical HEXAGON Satellite Imagery as a New Data Source for Mapping Open-Canopy Woodlands on the Tree Level

Monitoring woody cover by remote sensing is considered a key methodology towards sustainable management of trees in dryland forests. However, while modern very high resolution satellite (VHRS) sensors allow woodland mapping at the individual tree level, the historical perspective is often hindered by lack of appropriate image data. In this first study employing the newly accessible historical HEXAGON KH-9 stereo-panoramic camera images for environmental research, we propose their use for mapping trees in open-canopy conditions. The 2–4 feet resolution panchromatic HEXAGON satellite photographs were taken 1971–1986 within the American reconnaissance programs that are better known to the scientific community for their lower-resolution CORONA images. Our aim is to evaluate the potential of combining historical CORONA and HEXAGON with recent WorldView VHRS imagery for retrospective woodland change mapping on the tree level. We mapped all trees on 30 1-ha test sites in open-canopy argan woodlands in Morocco in the field and from the VHRS imagery for estimating changes of tree density and size between 1967/1972 and 2018. Prior to image interpretation, we used simulations based on unmanned aerial system (UAS) imagery for exemplarily examining the role of illumination, viewing geometry and image resolution on the appearance of trees and their shadows in the historical panchromatic images. We show that understanding these parameters is imperative for correct detection and size-estimation of tree crowns. Our results confirm that tree maps derived solely from VHRS image analysis generally underestimate the number of small trees and trees in clumped-canopy groups. Nevertheless, HEXAGON images compare remarkably well with WorldView images and have much higher tree-mapping potential than CORONA. By classifying the trees in three sizes, we were able to measure tree-cover changes on an ordinal scale. Although we found no clear trend of forest degradation or recovery, our argan forest sites show varying patterns of change, which are further analysed in Part B of our study. We conclude that the HEXAGON stereo-panoramic camera images, of which 670,000 worldwide will soon be available, open exciting opportunities for retrospective monitoring of trees in open-canopy conditions and other woody vegetation patterns back into the 1980s and 1970s.

IMPLICITY: CITY MODELING FROM SATELLITE IMAGES WITH DEEP IMPLICIT OCCUPANCY FIELDS

Abstract. High-resolution optical satellite sensors, combined with dense stereo algorithms, have made it possible to reconstruct 3D city models from space. However, these models are, in practice, rather noisy and tend to miss small geometric features that are clearly visible in the images. We argue that one reason for the limited quality may be a too early, heuristic reduction of the triangulated 3D point cloud to an explicit height field or surface mesh. To make full use of the point cloud and the underlying images, we introduce IMPLICITY, a neural representation of the 3D scene as an implicit, continuous occupancy field, driven by learned embeddings of the point cloud and a stereo pair of ortho-photos. We show that this representation enables the extraction of high-quality DSMs: with image resolution 0.5 m, IMPLICITY reaches a median height error of ≈0.7m and outperforms competing methods, especially w.r.t. building reconstruction, featuring intricate roof details, smooth surfaces, and straight, regular outlines.

Optimum Feature and Classifier Selection for Accurate Urban Land Use/Cover Mapping from Very High Resolution Satellite Imagery

Feature selection to reduce redundancies for efficient classification is necessary but usually time consuming and challenging. This paper proposed a comprehensive analysis for optimum feature selection and the most efficient classifier for accurate urban area mapping. To this end, 136 multiscale textural features alongside a panchromatic band were initially extracted from WorldView-2, GeoEye-3, and QuickBird satellite images. The wrapper-based and filter-based feature selection were implemented to optimally select the best ten percent of the primary features from the initial feature set. Then, machine leaning algorithms such as artificial neural network (ANN), support vector machine (SVM), and random forest (RF) classifiers were utilized to evaluate the efficiency of these selected features and select the most efficient classifier. The achieved optimum feature set was validated using two other images of WorldView-3 and Pleiades. The experiments revealed that RF, particle swarm optimization (PSO), and neighborhood component analysis (NCA) resulted in the most efficient classifier and wrapper-based and filter-based methods, respectively. While ANN and SVM’s process time depended on the number of input features, RF was significantly resistant to the criterion. Dissimilarity, contrast, and correlation features played the greatest contributing role in the classification performance among the textural features used in this study. These trials showed that the feature number could be reduced optimally to 14 from 137; these optimally selected features, alongside the RF classifier, can produce an F1-measure of about 0.90 for different images from five very high resolution satellite sensors for various urban geographical landscapes. These results successfully achieve our goal of assisting users by eliminating the task of optimal feature selection and classifier, thereby increasing the efficiency of urban land use/cover classification from very high resolution images. This optimal feature selection can also significantly reduce the high computational load of the feature-engineering phase in the machine and deep learning approaches.

Mapping South America’s Drylands through Remote Sensing—A Review of the Methodological Trends and Current Challenges

The scientific grasp of the distribution and dynamics of land use and land cover (LULC) changes in South America is still limited. This is especially true for the continent’s hyperarid, arid, semiarid, and dry subhumid zones, collectively known as drylands, which are under-represented ecosystems that are highly threatened by climate change and human activity. Maps of LULC in drylands are, thus, essential in order to investigate their vulnerability to both natural and anthropogenic impacts. This paper comprehensively reviewed existing mapping initiatives of South America’s drylands to discuss the main knowledge gaps, as well as central methodological trends and challenges, for advancing our understanding of LULC dynamics in these fragile ecosystems. Our review centered on five essential aspects of remote-sensing-based LULC mapping: scale, datasets, classification techniques, number of classes (legends), and validation protocols. The results indicated that the Landsat sensor dataset was the most frequently used, followed by AVHRR and MODIS, and no studies used recently available high-resolution satellite sensors. Machine learning algorithms emerged as a broadly employed methodology for land cover classification in South America. Still, such advancement in classification methods did not yet reflect in the upsurge of detailed mapping of dryland vegetation types and functional groups. Among the 23 mapping initiatives, the number of LULC classes in their respective legends varied from 6 to 39, with 1 to 14 classes representing drylands. Validation protocols included fieldwork and automatic processes with sampling strategies ranging from solely random to stratified approaches. Finally, we discussed the opportunities and challenges for advancing research on desertification, climate change, fire mapping, and the resilience of dryland populations. By and large, multi-level studies for dryland vegetation mapping are still lacking.

Comparative Analysis of Several Typical Landsat 8 OLI Cloud Detection Methods

The Landsat 8 Operational Land Imager (OLI) is a high-resolution satellite sensor that is carried on the next-generation Landsat which was launched by the National Aeronautics and Space Administration (NASA) in 2013. Compared to the previous land observation satellites in this series, Landsat 8 OLI has been further optimized in regard to its band setting and data acquisition frequency. The application scope of the data and the ability of information extraction have been further expanded and enhanced. However, the existence of clouds reduces the efficiency and the quality of satellite data. Therefore, in order to identify cloud pixels with a high level of accuracy and efficiency, a variety of cloud detection methods have been developed for Landsat 8 OLI sensors. Distinct treatment methods have been adopted for identifying different surface backgrounds and various types of cloud characteristics in order to improve the accuracy of cloud detection. The present study compares and analyzes the detection results from five typical cloud detection algorithms for different cloud types over different ground types. In addition, it discusses the various cloud detection algorithms used for different types of adaptive backgrounds and clouds in order to shed light upon the comprehensive application of different methods. Subsequently, these findings may provide a reference for the development of an algorithm with a higher precision of cloud detection.

Validation of Landsat 8 high resolution Sea Surface Temperature using surfers

Nearshore coastal waters are highly dynamic in both space and time. They can be difficult to sample using conventional methods due to their shallow depth, tidal variability, and the presence of strong currents and breaking waves. High resolution satellite sensors can be used to provide synoptic views of Surface Temperature (ST), but the performance of such ST products in the nearshore zone is poorly understood. Close to the shoreline, the ST pixels can be influenced by mixed composition of water and land, as a result of the sensor’s spatial resolution. This can cause thermal adjacency effects due to the highly different diurnal temperature cycles of water bodies and land. Previously, temperature data collected during surfing sessions has been proposed for validation of moderate resolution (1 km pixel size) satellite ST products. In this paper we use surfing temperature data to validate three high resolution (100 m resampled to 30 m pixel size) ST products derived from the Thermal InfraRed Sensor (TIRS) on board Landsat 8 (L8). ST was derived from Collection 1 and 2 Level 1 data (C1L1 and C2L1) using the Thermal Atmospheric Correction Tool (TACT), and was obtained from the standard Collection 2 Level 2 product (USGS C2L2). This study represents one of the first evaluations of the new C2 products, both L1 and L2, released by USGS at the end of 2020. Using automated matchup and image quality control, 88 matchups between L8/TIRS and surfers were identified, distributed across the North-Western semihemisphere. The unbiased Root Mean Squared Difference (uRMSD) between satellite and in situ measurements was generally ¡ 2 K, with warm biases (Mean Average Difference, MAD) of 1.7 K (USGS C2L2), 1.3 K (TACT C1L1) and 0.8 K (TACT C2L1). Large interquartile ranges of ST in 5 × 5 satellite pixels around the matchup location were found for several images, especially for the summer matchups around the Californian coast. By filtering on target stability the number of matchups reduced to 31, which halved the uRMSD across the three methods (to around 1.1K), MAD were much lower, i.e. 1.1 K (USGS C2L2), 0.6 K (TACT C1L1), and 0.2 K (TACT C2L1). The larger biases of the C2L2 product compared to TACT C2L1 are caused as a result of: (1) a lower emissivity value for water targets used in USGS C2L2, and (2) differences in atmospheric parameter retrieval, mainly from differences in upwelling atmospheric radiance and lower atmospheric transmittance retrieved by USGS C2L2. Additionally, tiling artefacts are present in the C2L2 product, which originate from a coarser atmospheric correction process. Overall, the L8/TIRS derived ST product compares well with in situ measurements made while surfing, and we found the best performing ST product for nearshore coastal waters to be the Collection 2 Level 1 data processed with TACT.

Forest Conservation with Deep Learning: A Deeper Understanding of Human Geography around the Betampona Nature Reserve, Madagascar

Documenting the impacts of climate change and human activities on tropical rainforests is imperative for protecting tropical biodiversity and for better implementation of REDD+ and UN Sustainable Development Goals. Recent advances in very high-resolution satellite sensor systems (i.e., WorldView-3), computing power, and machine learning (ML) have provided improved mapping of fine-scale changes in the tropics. However, approaches so far focused on feature extraction or the extensive tuning of ML parameters, hindering the potential of ML in forest conservation mapping by not using textural information, which is found to be powerful for many applications. Additionally, the contribution of shortwave infrared (SWIR) bands in forest cover mapping is unknown. The objectives were to develop end-to-end mapping of the tropical forest using fully convolution neural networks (FCNNs) with WorldView-3 (WV-3) imagery and to evaluate human impact on the environment using the Betampona Nature Reserve (BNR) in Madagascar as the test site. FCNN (U-Net) using spatial/textural information was implemented and compared with feature-fed pixel-based methods including Support Vector Machine (SVM), Random Forest (RF), and Deep Neural Network (DNN). Results show that the FCNN model outperformed other models with an accuracy of 90.9%, while SVM, RF, and DNN provided accuracies of 88.6%, 84.8%, and 86.6%, respectively. When SWIR bands were excluded from the input data, FCNN provided superior performance over other methods with a 1.87% decrease in accuracy, while the accuracies of other models—SVM, RF, and DNN—decreased by 5.42%, 3.18%, and 8.55%, respectively. Spatial–temporal analysis showed a 0.7% increase in Evergreen Forest within the BNR and a 32% increase in tree cover within residential areas likely due to forest regeneration and conservation efforts. Other effects of conservation efforts are also discussed.

The QAA-RGB: A universal three-band absorption and backscattering retrieval algorithm for high resolution satellite sensors. Development and implementation in ACOLITE

We present the “QAA-RGB”, a semianalytical algorithm that retrieves total absorption, particle backscattering, as well as the diffuse attenuation coefficient and the Secchi disk depth, from eighteen metre and decametre satellite sensors, including present and heritage Landsat's data, Sentinel 2 at 10 m and an array of commercial satellites such as PlanetScope, Pléiades and Worldview. The QAA-RGB is a minimal version of the Quasi-Analytical Algorithm (QAA), and therefore keeps its robustness and general validity across different water types. It ingests remote-sensing reflectance at only three bands, centred on red, green and blue wavelengths. Retrievals of non-water absorption at 555 nm (anw(555)) are satisfactory, with a standard deviation of σ ~ 50% up to anw(555) < 2 m−1, based on in situ datasets. Careful calibration results in unbiased retrievals across all sensors. Validation to independent in situ data is also provided. The QAA-RGB has been implemented in the ACOLITE generic processor, and is openly accessible to the community. Consistency of near-coincident observations from different satellite sensors is demonstrated.

Spectral decomposition of remote sensing reflectance variance due to the spatial variability from ocean color and high-resolution satellite sensors

The variability of the remote sensing reflectance, Rrs, now routinely retrieved from ocean color (OC) and high spatial resolution sensors, is often used to characterize water variability due to changes in inherent optical properties of the water body. At the same time, Rrs is partially variable because of uncertainties in its retrieval in the process of atmospheric correction. Using data from SNPP-VIIRS and Landsat-8 OLI sensors, the contribution of the main components to the variance of Rrs due to its spatial variability is determined based on a model in which variances were considered proportional to the mean values of the corresponding components. It is shown that there is practically no spatial variability in the open ocean waters and water variability is proportional to the spatial resolution of the sensor in coastal waters. Variances due to surface effects, inaccuracies of aerosol models, and sunglint can contribute significantly to Rrs variance, which characterizes Rrs spatial variability, with variances due to the water variability itself often being significantly smaller.

UAV and High Resolution Satellite Mapping of Forage Lichen (Cladonia spp.) in a Rocky Canadian Shield Landscape

Reindeer lichens (Cladonia spp.) are an important food source for woodland and barren ground caribou herds. In this study, we assessed Cladonia classification accuracy in a rocky, Canadian Shield landscape near Yellowknife, Northwest Territories using both Unmanned Aerial Vehicle (UAV) sensors and high-resolution satellite sensors. At the UAV scale, random forest classifications derived from a multispectral, visible-near infrared sensor (Micasense Altum) had an average 5% higher accuracy for mapping Cladonia (i.e., 95.5%) than when using a conventional color RGB camera (DJI Phantom 4 RTK). We aggregated Altum lichen classifications from three 5 ha study sites to train random forest regression models of fractional lichen cover using predictor features from WorldView-3 and Planet CubeSat satellite imagery. WorldView models at 6 m resolution had an average 6.8% RMSE (R 2 = 0.61) when tested at independent study sites and outperformed the 6 m Planet models, which had a 9.9% RMSE (R 2 = 0.34). These satellite results are comparable to previous lichen mapping studies focusing on woodlands, but the small cover of Cladonia in our study area (11.6% or 16.8% within the barren portions) results in a high relative RMSE (62.2%) expressed as a proportion of mean lichen cover.

An approach for building rooftop border extraction from very high-resolution satellite images

The development of a new Very High-Resolution (VHR) satellite sensor provides opportunities for mapping more details. The building extraction process is a challenging problem due to the misclassification between buildings and similar spectral objects. The main aim of this study is to introduce an approach to building rooftops border extraction. In this approach, a new index to highlight building areas using the capabilities of the WorldView-2 eight-bands VHR satellite images is presented. The developed index is then integrated into a complete approach to extract borders between buildings and their corresponding shadows. The building’s rooftops boundaries are determined using a novel algorithm. The proposed approach is applied to five study areas and compared with three recent approaches for building extraction. The results of the comparison have illustrated that the proposed approach achieves the highest overall accuracy (average of 92.20%) and has the ability for detecting building rooftops separated from their façades.