Accurate monitoring of atmospheric CO₂ concentrations is essential for understanding carbon dynamics and informing climate-mitigation strategies. This study examines the seasonal variability of satellite-based XCO₂ retrievals from NASA's OCO-2 mission relative to ground-based TCCON XCO₂ measurements over ten years (2014–2024) across 17 global TCCON sites. The study distinguished between two satellite observation modes, nadir and glint. It assessed how seasonal variations in satellite-derived environmental parameters – normalized difference vegetation index (NDVI), soil surface moisture (SSM), land-surface temperature (LST), and evapotranspiration (ET) – relate to statistical characteristics of the XCO₂ bias. In addition, we evaluated the temporal dynamics of the bias across the dominant land-cover types surrounding each site. Our results identify a latitude-dependent bias in nadir retrievals across Northern Hemisphere sites, with strong associations between nadir bias and land-cover at TCCON sites located > 45°N (R² = 0.95) and between 0 and 45°N (R² = 0.79). This work provides novel insights into the temporal dynamics of satellite XCO₂ bias relative to ground measurements and demonstrates distinct behaviors across observation modes that have not been reported previously. These findings are crucial for improving the accuracy of global CO₂ mapping and for strengthening the reliability of carbon flux models.

The bidirectional reflectance distribution function (BRDF) is one of the most challenging factors in remote sensing, significantly influencing spectral measurements. Accurate BRDF measurement is crucial for understanding and correcting its impact on spectral data. However, existing approaches require specialized equipment and precise measurement setups, typically capturing data from a single point on an object's surface. Meanwhile, advances in sensor technology have led to the development of compact and user-friendly spectral cameras, thereby reducing operational complexity. Leveraging these advancements, we propose a novel, modular framework for image-based BRDF measurement that does not require specialized instruments, such as a gonioreflectometer. Our approach begins with acquiring freehand, multiview overlapping images of materials from unknown positions on a hemisphere. We then apply computer vision techniques to estimate camera locations, automatically extract tie points between overlapping images, and establish image connectivity. This connectivity enables automatic tracking of regions of interest (ROIs), reflectance measurement across multiple viewing directions, and BRDF characterization. We validated our methodology using a hyperspectral dataset. The results demonstrate its effectiveness in extracting camera positions with an error of less than two degrees compared to real-time kinematic (RTK) drone measurements, tracking ROIs with millimeter accuracy, and accurately measuring the BRDF of various land-cover types. This framework offers a robust, flexible, and accessible alternative for BRDF measurement in remote sensing applications.

Recent advancements in Global Navigation Satellite System-Interferometric Reflectometry (GNSS-IR) have enabled the extraction of environmental data by analyzing differences between direct and multipath signals. A key application is soil moisture estimation using Signal-to-Noise Ratio (SNR) measurements of reflected signals. However, these methods only provide relative moisture estimates, requiring periodic in-situ measurements to establish the minimum moisture level for each observation period. On the other hand, optical remote sensing estimates soil moisture through pixel reflectance, clearing the need for in-situ measurements but is limited by sensitivity to weather and illumination conditions, cloud cover, ground vegetation cover, and a 3–5-day satellite orbit, hindering continuous estimation. This study further develops the potential of GNSS-IR for soil moisture estimation by introducing a novel, optimized approach to enhance accuracy. We also develop a data fusion model that combines GNSS-IR's continuous, weather-independent measurements with discrete estimates from spectral remote sensing Sentinel-2 imagery. This model enables continuous soil moisture estimation without in-situ measurements. We use datasets from Valencia, Spain, and Kabri, Israel, to evaluate the methodology. Our models achieve an accuracy relative to in-situ data of approximately 0.02 [m³/m³] in soil moisture estimation, outperforming traditional methods, which have an accuracy of around 0.05 [m³/m³].

Accurately tracking flowering phenology in tree crops is critical for optimizing pollination and mitigating climate-induced yield variability. This study presents a scalable framework for monitoring almond bloom phenology across California's Central Valley using freely available Sentinel-2 multispectral satellite imagery. By leveraging the Enhanced Bloom Index (EBI) and Normalized Difference Vegetation Index (NDVI), we applied a peak-detection algorithm to over 4000 Sentinel-2 scenes, constructing standardized time-series data from 2019 to 2022 for approximately 30,000 almond orchards. We achieved a mean absolute error (MAE) of 1.9 days and 87.5% agreement with in-situ time-lapse camera observations, thereby validating the effectiveness of our remote sensing model at a regional scale. To overcome challenges posed by cloud cover effects, we aggregated spectral indices at the orchard level and applied temporal interpolation and Savitzky-Golay smoothing. The computational pipeline enabled us to generate the first daily, valley-wide maps of flowering. These maps revealed consistent spatial trends, such as earlier blooms in northern counties and substantial interannual variability influenced by climatic conditions. We further evaluated four dormancy models, Chill Hours (CH), Utah Chill Units (CU), Dynamic Chill Portions (CP), and the mechanistic Carbohydrate–Temperature (CT) model, by relating temperature-based estimates to satellite-derived bloom dates. A random forest (RF) ensemble model integrating all four models achieved an MAE of 1.48 days and an R² of 0.80, outperforming the individual models. The CT model emerged as the most informative, accounting for about 50% of the variance explained. Future work should incorporate additional sensors, high-resolution imagery, and management variables (e.g., irrigation and fertilization) to enhance model accuracy and extend applicability to diverse crops and agroecological zones. The proposed approach provides a cost-effective, scalable, and near-real-time framework for phenological monitoring and model validation. Its outputs have strong potential to inform precision agriculture, pollination scheduling, and climate-resilient crop management strategies.

This review addresses the gap between substantial research on extensive green roofs (GRs) and their limited large-scale implementation for urban stormwater management. Despite demonstrated benefits, widespread adoption remains rare due to fragmented knowledge across disciplines and practical challenges in monitoring, modelling, and performance evaluation. To support scalable deployment, the review synthesizes insights across four core areas. (I) Key design parameters are reviewed, including substrate depth, vegetation type, roof configuration, and the use of supplemental water storage modules, all of which influence retention capacity and peak flow reduction. (II) Existing hydrological models are evaluated for their ability to predict runoff across various spatial and temporal scales, highlighting opportunities for integrating process-based and data-driven approaches. (III) Monitoring strategies are examined, with an emphasis on the growing role of scalable technologies, such as remote sensing (RS) and spectral imaging, in enabling cost-effective, citywide diagnostics. (IV) Emerging directions are explored, including integrated blue green solar roofs (BGSRs) that combine vegetation, photovoltaics, and controlled drainage, alongside the development of RS-based hydrological indices for continuous performance assessment and flood risk reduction at the urban scale.

Humans use natural language to describe places, relying on mental processes to perceive, interpret, and communicate information about spatial relationships, environments, and navigation. Computational location-based systems strive to replicate this capability by enabling the retrieval of geographic locations from textual descriptions or queries. However, progress in this domain remains constrained by the limited availability of extensive, linguistically diverse textual datasets, which are essential for developing and evaluating robust geographic information retrieval methodologies. In this study, we conducted a review of existing geocoding datasets used for textual geolocation. Our objectives were to systematically compare these datasets, characterize their attributes, and assess their impact on retrieval performance as reported in the literature. A critical challenge we identified was the inconsistency in evaluation practices across studies, which complicates direct comparisons and underscores the need for standardized benchmarks. This review synthesizes the current landscape of geocoding datasets, offering insights for informed dataset development and fosters more consistent evaluation practices. By addressing the imperative of dataset standardization and availability, we aim to support the creation of more effective geographic information retrieval systems and establish a solid foundation for future research in this field.

Efficient Site-Specific weed management (SSWM) practice requires high-resolution data in both spatial and spectral domains. It allows spatially locating different weed species at early growth to adjust herbicide application based on weed composition and coverage. Nonetheless, such high-resolution data is not always available, and mixed pixels are likely to exist, creating a challenge to generate accurate weed maps. In this regard, Spectral Mixture Analysis (SMA), which allows exploiting subpixel information from coarse spatial resolution spectral data, can mitigate this challenge. This study assesses the potential benefits of four SMA methods for estimating weed coverage from different botanical groups. Each of the methods examined here namely, Fully Constrained Least Squares Unmixing (FCLSU), Sparse Unmixing via variable Splitting and Augmented Lagrangian (SUnSAL), Sparse Unmixing via variable Splitting and Augmented Lagrangian and Total variation (SUnSAL-TV) and the e Vectorized Code Projected Gradient Descent Unmixing (VPGDU) suggests a distinct advantage for spectral unmixing. To compare the four methods, we first established a controlled dataset that included weed species characterized by distinct botanical groups and assessed the performance of four SMA methods in estimating weed coverage and composition at various spatial resolutions. We found that SUnSAL-TV and VPGDU outperformed FCLSU and SUnSAL, with up to 13 % lower Mean Absolute Error (MAE) values. Next, we applied the comparative analysis of these four SMA methods to a multispectral field dataset involving corn and weeds. The same results trend was observed in the field study, with VPGDU as the best-performing method, with an overall MAE value lower than 12 %. These experimental outcomes demonstrate the advantages of the total variation regularization of SUnSAL-TV and the superiority of the SAM-based method, VPGDU, over other approaches, underscoring the advantage of its objective function and the significant effect of varying illumination on the results.

Urban-scale environmental performance evaluations are essential for designing cities that effectively respond to climate change and rapid urbanization. Remote Sensing (RS) technologies provide high-resolution, multi-scale, and temporal assessments across multiple interlinked environmental criteria. Despite its growing adoption in urban sustainability, a comprehensive review of RS's role in multi-criteria decision-making is still lacking. This review analyzes 124 research articles to explore RS applications in spatio-temporal analysis, impact evaluation, mitigation strategy assessment, and predictive modeling across five interconnected environmental criteria: urban air quality, urban heat, outdoor thermal comfort, building energy consumption, and solar potential. RS facilitates the integration of morphological, thermal, and meteorological data, enabling the evaluation of urban interdependence, such as the influence of urban form on air pollution dispersion, heat retention, and energy demand. Machine learning and AI-enhanced models improve air quality predictions, urban heat mitigation strategies, energy forecasting, and solar potential assessments. UAVs, LiDAR, and nanosatellite technologies further enhance real-time urban climate monitoring at finer spatial scales, supporting dynamic planning interventions. Despite challenges in data resolution, temporal coverage, and real-time monitoring, advancements in AI-driven downscaling, digital twins, and nano satellite networks continue to expand RS capabilities. By facilitating multi-criteria decision-making, RS empowers urban designers and policymakers to develop climate-adaptive, energy-efficient, and resilient cities, offering actionable insights for sustainable design and planning.

Sunflower broomrape (Orobanche cumana) poses a severe threat to sunflower crops, parasitizing their roots and hindering plant growth. Current control methods, which typically rely on uniform herbicide applications, are economically inefficient and environmentally damaging. This study investigates the use of unmanned aerial vehicle (UAV)-based multispectral imaging to detect broomrape-infected sunflowers by analyzing temporal patterns in spectral vegetation indices (VIs). Over four imaging campaigns conducted during early subsoil parasitic stages, multispectral data were collected and processed to compute ten VIs. These VIs, reflecting changes in canopy reflectance over time, were then analyzed using various machine learning models, including a pattern recognition neural network (PRNN). Results showed that the PRNN model, trained on time-series data, achieved an overall accuracy of 84.8 % and a true positive rate of 80.4 % in detecting broomrape infection, emphasizing the strength of utilizing temporal data for enhancing detection accuracies. Pixel-level reconstruction maps revealed varying spectral responses within infected canopies, highlighting the importance of accounting for this heterogeneity. This study demonstrates the potential of UAV-based multispectral imaging combined with advanced machine learning (ML) techniques for early detection of broomrape infestations in sunflower crops, offering insights for managing similar infestations in other crops.

In laser–metal processing such as Laser-based Powder Bed Fusion of Metals (PBF-LB/M), the quality of the final product is significantly influenced by the thermal behaviour of the melt pool. Accurate temperature measurements are challenging due to high temperature gradients and rapid cooling. This study introduces Multispectral Imaging (MSI) as a highly effective thermometry technique to address these demanding conditions. MSI captures multiple bands of radiation, enabling the simultaneous determination of absolute temperature and surface emissivity. This capability is crucial for ensuring high-quality outcomes in laser–metal processing such as PBF-LB/M, contingent upon robust radiometric calibration. To enhance reliability, two radiometric calibration models were developed: an empirical model based on linear sensor data correlations and an analytical model leveraging sensor behaviour. An efficient calibration was found for both models across a range of signal-to-noise ratios in the black body calibrator, where the analytical model even performs robust under a high noise level. Additionally, the temperature accuracy in the solid-state case was tested with a thermo-mechanical simulator. Results showed that the empirical and analytical models achieved mean relative errors of 2.0% and 1.6%, respectively, outperforming the state-of-the-art. Further, laser irradiation experiments highlighted the analytical model's ability to accurately determine the emissivity decrease during the solid-to-liquid transition, allowing for accurate melt pool temperature estimations. Conversely, the empirical model struggled to estimate temperature effectively in the liquid phase. This study not only proposed a robust methodology of MSI in temperature measurement but also confirmed the accuracy and reliability of the system, broadening the scope for further investigation into the intricate thermal dynamics involved in laser–metal interactions.

Satellite Image Time Series (SITS) analysis is essential for understanding environmental dynamics and forecasting future trends. This study introduces a novel spatio-temporal transformer-based model for long-term Normalized Difference Vegetation Index (NDVI) prediction, leveraging Landsat data to address the challenges of modeling complex spatial and temporal dependencies in remote sensing. The proposed framework integrates spatial, temporal, and location-specific embeddings with a transformer architecture, enabling accurate predictions across diverse temporal horizons and spatial scales. By combining a scalable embedding framework with self-supervised learning using both spatial and temporal masking, the model effectively learns robust representations without the need for extensive labeled datasets. A comprehensive evaluation on Mediterranean landscapes demonstrates the model’s ability to capture intricate spatio-temporal dependencies and predict NDVI trends effectively. This work bridges a critical gap in remote sensing by offering a unified approach for capturing intricate spatio-temporal and long-term dependencies, providing valuable insights into environmental change, and improving the analysis of dynamic ecosystems.

— Accurate monitoring of atmospheric carbon dioxide (CO₂₎ is essential for understanding carbon fluxes and guiding climate mitigation strategies. This study investigates the seasonal variability of the bias between satellite-based OCO-2 XCO2 observations and ground-based TCCON XCO2 measurements over nine years (2014–2023) at the Caltech TCCON station. Grouping the data by the observation month and mode (nadir or glint) enabled us to analyze the distributions of the deviations from the bias. Our results reveal distinct seasonal patterns in the bias variability. The distributions of the nadir mode observations' deviations from the mean bias exhibited relatively stable medians, ranging from -0.6 to 0.4 ppm, indicating minimal deviation from the mean bias. In contrast, distributions of the glint mode observations' deviations from the mean bias showed significant variability, with the absolute median values exceeding 1 ppm during January, March, and September. Skewness analysis highlighted the asymmetry of the data distributions and the presence of significant outliers, with the nadir mode displaying a notable positive skewness in September and the glint mode demonstrating high negative skewness during periods of elevated vegetation cover. Furthermore, seasonal vegetation dynamics, represented by monthly NDVI values, were strongly correlated with skewness in the glint mode (R² = 0.76), underscoring its sensitivity to bright conditions and surface reflectance variability. Nadir mode, in contrast, showed minimal correlation, reflecting its relative stability. These findings emphasize the importance of accounting for operational differences and environmental factors, such as vegetation cycles, when interpreting the OCO-2 data. This study provides novel insights into the temporal dynamics of satellite-ground measurement discrepancies by addressing the role of seasonal variability in OCO-2 biases. Such findings are essential for improving retrieval algorithms, enhancing the accuracy of satellite-based CO₂ monitoring, and supporting the development of reliable carbon flux models for climate policy and mitigation efforts. Future work should expand this analysis to other TCCON sites and incorporate additional environmental variables to further refine our understanding of seasonal influences on satellite biases.

Lunar cold spots are thermal anomalies linked to fresh impact craters. Detecting and understanding them provides insights into the Moon's surface evolution and thermophysical properties. This study evaluates YOLOv8 and YOLOv11 deep learning models for automating cold spot detection using Diviner radiometer data. The training dataset, derived from 128-pixel-per-degree nighttime regolith temperature maps (± 60° latitude), included 384 images with 652 annotated cold spots. Testing utilized 4,816 overlapping 512 × 512-pixel sub-images from the 2023 High-Resolution Nighttime Temperature map. In summary, Both models excelled in detecting faint anomalies, while YOLOv11 outperformed YOLOv8, achieving precision (0.85), recall (0.78), and F1 score (0.81) in cross-validation. In addition, both significantly outperformed manual methods, identifying previously undetected cold spots and enhancing our understanding of lunar thermal anomalies. These findings highlight deep learning's transformative role in planetary exploration, combining precision and scalability for efficient data analysis.

Site-specific weed management employs image data to generate maps through various methodologies that classify pixels corresponding to crop, soil, and weed. Further, many studies have focused on identifying specific weed species using spectral data. Nonetheless, the availability of open-access weed datasets remains limited. Remarkably, despite the extensive research employing hyperspectral imaging data to classify species under varying conditions, to the best of our knowledge, there are no open-access hyperspectral weed datasets. Consequently, accessible spectral weed datasets are primarily RGB or multispectral and mostly lack the temporal aspect, i.e., they contain a single measurement day. This paper introduces an open dataset for training and evaluating machine-learning methods and spectral features to classify weeds based on various biological traits. The dataset comprises 30 hyperspectral images, each containing thousands of pixels with 204 unique visible and near-infrared bands captured in a controlled environment. In addition, each scene includes a corresponding RGB image with a higher spatial resolution. We included three weed species in this dataset, representing different botanical groups and photosynthetic mechanisms. In addition, the dataset contains meticulously sampled labeled data for training and testing. The images represent a time series of the weed’s growth along its early stages, critical for precise herbicide application. We conducted an experimental evaluation to test the performance of a machine-learning approach, a deep-learning approach, and Spectral Mixture Analysis (SMA) to identify the different weed traits. In addition, we analyzed the importance of features using the random forest algorithm and evaluated the performance of the selected algorithms while using different sets of features.

Spectral and spatial resolutions are critical factors in Site-Specific Weed Management (SSWM) applications, which aim to adjust and apply herbicides according to the weed composition and coverage during early growth. Thus, a reliable SSWM requires high spectral and spatial resolution data to create informative maps indicating and spatially localizing the weed species. Besides, upscaling and improving the mapping process are essential to motivate farmers to adopt SSWM practices. However, high-resolution data in spectral and spatial domains is not always available, and mixed pixels are highly probable. Therefore, evaluating approaches that deal with such pixels that pose a significant problem for weed classification tasks is vital. In this regard, Spectral Mixture Analysis (SMA) can exploit subpixel information from coarse-resolution data. However, despite its potential, SMA has not been examined for subpixel weed mapping. In this work, we aim to evaluate the potential advantages of SMA for coverage estimation of different weed traits (weed botanical group, photosynthetic pathway, and species). For this purpose, we first created a dataset containing weed species characterized by different botanical groups and photosynthetic pathways. Then, we evaluated the performance of SMA in estimating the weed's coverage at varying growth stages and spectral and spatial resolutions. The experimental results revealed that using botanical weed groups as Endmembers (EM) is advantageous over other traits. A quantitative evaluation revealed strong correlation values between the actual and predicted weed coverage and a Mean Absolute Error (MAE) lower than 10 % for the different EM at all resolutions and growth stages. Besides, SMA outperformed the traditional classification-based approach in estimating weed coverage with up to ∼ 30 % lower Mean Absolute Error. Finally, our study establishes a basis for incorporating SMA in various SSWM practices.

In this study, we explore spectral heterogeneity within plant canopies, a characteristic often observed in stressed plants where certain leaves or intra-leaf regions exhibit stress symptoms while others remain unaffected. Considering this variability in spectral signatures holds promise for enhancing remote sensing methodologies aimed at plant stress detection. Typically, remote sensing techniques analyze the plant as a whole, potentially overlooking stress-related spectral signatures due to the inclusion of unaffected pixels. We used a clustering-based technique, which incorporates semi-supervised learning elements for tuning hyper-parameters, to differentiate spectral patterns associated with and unique to pixels from broomrape-infected (Orobanche spp. and Phelipanche spp.) carrots from unrelated patterns. Ground-based hyperspectral (400–1000 nm) images of broomrape-infected and non-infected carrot canopies were used in an agglomerative clustering procedure followed by spectral angle mapper (SAM) analysis to identify a spectral endmember indicative of broomrape infection symptoms. Pixels from this cluster constituted an average of 8.5–11.5 % from the canopies of infected plants. Subsequently, we: (a) examined the relationship between carrot leaf mineral content and the percentage of symptomatic pixels to explore stress-induced alterations creating the unique spectral signatures of infected plants; and (b) utilized the inverse mode of PROSPECT, a radiative transfer model (RTM), to derive primary plant traits from the distinct spectral data of each cluster. We found that deficits in two macro elements, phosphorous and potassium, along with two pigments, chlorophyll and carotenoid, were correlated with the symptomatic cluster in infected plants. The methodology presented in this study paves the way for further research into broomrape detection in various crop species, as well as other plant stressors.

Soil moisture content estimation is essential for soil quality assurance before undertaking construction operations. Existing methodologies for estimating soil moisture content are often associated with significant costs and considerable time requirements. This study introduces a pragmatic, cost-effective, accurate, and prompt indirect method for estimating soil moisture content using a multispectral imaging based regression neural network. An in-house built, site-testing capable multispectral imaging system operating in the range from 365 nm to 940 nm is used to obtain multispectral images of soil samples. Image preprocessing techniques such as dark current subtraction and histogram equalization are used to mitigate the random noise and surface imperfections in the images. A regression neural network is utilized to estimate the soil moisture content with R2 value of 0.9987. Further testing with validation data gives R2 value of 0.9922 ensuring the suitability of this method for real-world applications.

Early detection of non-optimal weed control is now a priority to ensure herbicide efficacy. This study aimed to evaluate the potential of hyperspectral imaging (HSI) for early detection of the effects of glyphosate and glufosinate on weeds. Specific features (bands and vegetation indices [VIs]) were extracted as indicators for glyphosate and glufosinate efficacy. Black nightshade (Solanum nigrum L.) was used as the model weed and treated with glyphosate or glufosinate at the fourth-leaf stage. Plants were imaged in the laboratory at 6, 24, 48, 72, and 96 hours after treatment (HAT) with a 204-wavelength (400–1000 nm) hyperspectral camera. The impact of the herbicide treatments on the spectral reflectance values was analyzed using the two-sided Mann–Whitney U test, followed by classification with a machine learning (ML) model applied on the full spectrum and on 12 VIs. In addition, the contribution of the different wavelengths (features) to classification accuracy was assessed using a feature selection process. For glufosinate, 95% classification accuracy was observed as early as 6 HAT, with four features from the green region required. For glyphosate, four features from the red, red-edge, and green regions were used to achieve 88% classification accuracy at 24 HAT. The accuracy of VIs-based classification was generally lower than that of the full spectrum classification accuracy. Above 85% classification accuracy was achieved only at later imaging campaigns, starting at 48 HAT. This study thus demonstrates that non-optimal application of glyphosate and glufosinate can indeed be detected using spectral imaging.

While extensive research addressed the Bidirectional Reflectance Distribution Function (BRDF) effect and approaches to correct it, too few works have considered the influence of spectral mixture on the correction results. This work studies the BRDF effect in spectral data and presents an approach to correct its undesired impact, considering the likely presence of mixed pixels. We propose an unmixing-based semiempirical model, incorporating endmembers’ (EMs) fractions within the data correction. To evaluate the performance of the proposed methodology, we conducted experiments with laboratory and aerial hyperspectral image data. The outcomes from all experiments reveal vital insights into the influence of the spectral mixture on the BRDF correction accuracy. Nevertheless, the results clearly show that the accuracy of the corrected reflectance significantly improved using the proposed model for reducing the BRDF effect, regardless of the pixel's microtopography arrangement. Most importantly, a quantitative assessment of the difference in reflectance values within the overlapping region between two aerial images shows that the unmixing-based model outperforms the commonly used one. While the traditional method reduces the influence of the BRDF in the corrected data by a factor of two, with an average Mean Absolute Error (MAE) of≈2.5%, the proposed approach reduces it by a factor of four with an average MAE≈1.2%.

Fine airborne particles (diameter <2.5 μm; PM_2.5) are recognized as a major threat to human health due to their physicochemical properties: composition, size, shape, etc. However, normally only size-fraction-specific particle concentrations are monitored. Interestingly, although the aerosol type is reported as part of the aerosol optical depth retrieval from satellite observations, it has not been utilized, to date, as an auxiliary information/co-variate for PM_2.5 prediction. We developed Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) models that account for this information when predicting surface PM_2.5. The models take as input only widely available data: satellite aerosol products with full cover and surface meteorological data. Distinct models were developed for AOD of specific aerosol types. Both the RF and XGBoost models performed well, showing moderate-to-high cross-validated adjusted R² (RF: 0.753–0.909; XGBoost: 0.741–0.903), depending on the aerosol type and other covariates. The weighted performance of the specific aerosol-type models was higher than of the RF and XGBoost baseline models, where all the AOD retrievals were used together (the common practice). Our approach can provide improved risk estimates due to exposure to PM_2.5, better resolved radiative forcing calculations, and tailored abatement surveillance of specific pollutants/sources.

Recent climate shifts significantly reduced major crops' yield worldwide and challenged our understanding of the physiology of deciduous trees during dormancy, which is highly affected by warmer winters. Adapting to climatic changes requires making rapid decisions regarding crops and accelerating related studies. Therefore, we need to enhance the ability to monitor tree crops on a large scale. In this regard, earth-observing missions are greatly beneficial for such monitoring. We aimed to develop an approach for monitoring the flowering phenology of almonds using remote sensing applications in California. To do so, we used multi-spectral satellite images to create time series of the Enhanced Bloom Index (EBI) for thousands of almond orchards across California Central Valley. Then, we converted these time series to maps showing the bloom progression of almond orchards across California. Using these maps, we determined the time and duration of the almond bloom in different parts of California.

Quantitative evidence for the fundamental spectral shape similarity existing between different soils is presented utilizing Pearson's correlation and SAM (Spectral Angle Mapper) analysis in-between signatures representing thousands of soil samples collected from different environments around the world. A universal quadratic soil line (UQSL) was found to highly fit these representative signatures as well as spectra from the ISRIC (International Soil Reference and Information Center) global database including soil samples collected from 58 countries. K-means clustering of ISRIC spectral reflectance deviations from the UQSL facilitated identification of 5 characteristic deviation patterns. Significant correlation was found for a set of 1000 soil samples between changes in multiple chemical properties and corresponding reflectance changes between these clusters. Future availability of large global soil data bases may allow improved determination of the UQSL and generalization of its deviation patterns relationships with PBC (Physical, Biological and Chemical) properties for wide areas monitoring of soil conditions.

This study examines uncertainties in the retrieval of the Aerosol Optical Depth (AOD) for different aerosol types, which are obtained from different satellite-borne aerosol retrieval products over North Africa, California, Germany, and India and Pakistan in the years 2007–2019. In particular, we compared the aerosol types reported as part of the AOD retrieval from MODIS/MAIAC and CALIOP, with the latter reporting richer aerosol types than the former, and from the Ozone Monitoring Instrument (OMI) and MODIS Deep Blue (DB), which retrieve aerosol products at a lower spatial resolution than MODIS/MAIAC. Whereas MODIS and OMI provide aerosol products nearly every day over of the study areas, CALIOP has only a limited surface footprint, which limits using its data products together with aerosol products from other platforms for, e.g., estimation of surface particulate matter (PM) concentrations. In general, CALIOP and MAIAC AOD showed good agreement with the AERONET AOD (r: 0.708, 0.883; RMSE: 0.317, 0.123, respectively), but both CALIOP and MAIAC AOD retrievals were overestimated (36–57%) with respect to the AERONET AOD. The aerosol type reported by CALIOP (an active sensor) and by MODIS/MAIAC (a passive sensor) were examined against aerosol types derived from a combination of satellite data products retrieved by MODIS/DB (Angstrom Exponent, AE) and OMI (Aerosols Index, AI, the aerosol absorption at the UV band). Together, the OMI-DB (AI-AE) classification, which has wide spatiotemporal cover, unlike aerosol types reported by CALIOP or derived from AERONET measurements, was examined as auxiliary data for a better interpretation of the MAIAC aerosol type classification. Our results suggest that the systematic differences we found between CALIOP and MODIS/MAIAC AOD were closely related to the reported aerosol types. Hence, accounting for the aerosol type may be useful when predicting surface PM and may allow for the improved quantification of the broader environmental impacts of aerosols, including on air pollution and haze, visibility, climate change and radiative forcing, and human health.

One of the most challenging effects of remote sensing is landcover materials' Bidirectional Reflectance Distribution Function (BRDF). A wide range of approaches and measuring methods address the BRDF in various studies. However, there is a requirement for an accurate measurement setup and costly special equipment. Furthermore, the measurements and calculations are applied to model the BRDF for a single point on the object's surface. Considering these limitations, we propose a new modular framework and methodology for measuring, modeling, and analyzing the BRDF without the need for unique instruments. Instead, we suggest acquiring multiple overlapping images in a simple and time-saving way, sampling the desired object's Region Of Interest (ROI) in one image and automatically tracking it in the other images. Experimental results using laboratory data acquired under controlled conditions clearly show the advantages of our framework in retrieving the camera positions, tracking ROIs in the different images, and accurately measuring the BRDF of various land-cover types. Moreover, we observed the variability of the obtained measurements before and after applying the kernel-driven approach to minimize the BRDF effect. The results show that the applied correction reduces this variability significantly, indicating the high accuracy of measuring the directional reflectance using the proposed approach.

Site-specific weed management (SSWM) is a precise and resource-efficient approach that can result in more productive and sustainable agricultural practices. SSWM requires weed maps, in which the vegetation-related pixels are segmented from the soil and other substances and then classified into crops and different weed species. Such classification with a high spatial resolution is significant for SSWM since preventing economic losses due to weeds requires making management decisions at meter scales. In this regard, hyperspectral sensors can capture leaf anatomy and biochemistry variations, suggesting many advantages for weed classification. However, the typical tradeoff between spectral and spatial resolution poses a challenge for applying hyperspectral imaging in large scales and scenarios of high densities and tiny seedlings at early growth due to mixed pixels. Mixture analysis methods were previously demonstrated to offer opportunities for dealing with mixed pixels in vegetation ecology and agriculture. Nonetheless, they were not widely utilized for weed classification. This study aims to reveal the impact of the spectral mixture on classification results using supervised classification, spectral unmixing, and spatial analysis. We attempted to characterize how the spectral mixture of different weed species and soil at different growth stages affects classification results. Our results suggest that spectral mixtures are probably a significant factor driving misclassifications when classifying weed species. Their effect can be characterized by spatial analysis and fractions obtained by spectral unmixing. We assume that the subpixel information provided by the fraction maps may add information about the spectral mixture that can assist in interpreting misclassification pixels alongside the widely used confusion matrix. This contribution is highly relevant at coarser spatial resolutions.

The spectral mixture analysis (SMA) plays a vital role in spectral data analysis and extraction of subpixel information. However, this technique provides only quantitative information regarding the materials' abundance fractions within the pixel. On the other hand, the Bidirectional Reflectance Distribution Function (BRDF) indicates that sub-pixel topography affects the surface's directional reflection to a large extent. Unfortunately, despite the high importance of the BRDF effect and the SMA in remote sensing, only very few research works addressed their mutual influence. Thus, in this work, we propose a study that addresses this mutual influence and suggests an approach for extracting sub-pixel topographic information from mixed pixels. For this purpose, we conducted two multiview imaging experiments under controlled conditions using artificial mixed surfaces. Each surface type is made of two materials and has a varying structural pattern. Then we measured the BiConical Reflectance Factor (BCRF) of each surface from various viewing zenith angles. Next, we applied spectral unmixing to estimate the abundance fraction of three endmembers (EMs) in each surface's pattern. Finally, we tested the relationship between the sup-pixel topography and the fraction variation vs. the multiple imaging directions. The first experiment results showed that multiview spectral measurements allow the separability between surfaces combining the same materials' composition but with different sub-pixel structural arrangements. Moreover, such separability is more accurate in the fraction space than in reflectance space. Besides, and most importantly, the second experiment revealed exciting outcomes regarding the relationship between the sub-pixel topographic feature and the variation of the EM fraction vs. the imaging viewing direction. Specifically, we showed a high correlation between the EMs' fractions and the height of a repetitive element within the sub-pixel topography with a determination coefficient that reaches 0.89.

Numerous natural surfaces observed using Remote Sensing do not reflect light as Lambertian surfaces. Instead, their reflection is highly dependent on two main directions: the direction of the light source and the observation viewing angle, which characterize the Bidirectional Reflectance Distribution Function (BRDF). The BRDF is one of the challenging main effects of remote sensing. Thus, studying the BRDF of various land cover surfaces is essential, and researchers invest many efforts to fulfill this objective. However, measuring the BRDF is tricky and requires unique instruments, e.g., the Gonioreflectometer. Unfortunately, the availability of such instruments is deficient, and they are costly and hard to maintain. Considering these limitations, we present a study and a new approach for measuring the BRDF of surfaces with a camera-aided spectroradiometer that simultaneously acquires an RGB image from the sensor location beside the spectral measurement. Then, we feed the Structure From Motion (SFM) process with the RGM images to retrieve the sensor locations. Next, we convert the sensor locations into the quantities needed for the BRDF measurement, i.e., zenith angles and distances relative to the measured sample. Finally, we apply a set of measurements under controlled conditions in a dark room designed for hyperspectral remote sensing studies to evaluate the proposed methodology. In particular, we experimented with three different material surfaces. The results clearly show the highly accurate sensor position derived by SFM, providing zenith angles and distance from the scene's center with mean errors around one degree and 2.5 centimeters, respectively. In addition, the obtained spectra tell that the proposed approach is suitable for multiangular measurements of reflected light and studying the BRDF.

Sunflower broomrape (Orobanche cumana) is a root parasitic weed that severely limits sunflower yield in large areas of Europe and Asia. Early detection of the parasite can facilitate site-specific control of this weed. However, most of its life-cycle takes place in the soil sub-surface and by the time that O. cumana shoots emerge, the damage to the crop is irreversible. The main aim of this study was to evaluate the potential use of hyperspectral imaging for the early detection of parasitism by monitoring changes in spectra obtained from the host plants. A field experiment was conducted on infested and non-infested sunflower plants, imaged by a ground-based hyperspectral camera at two early parasitism stages that are relevant for herbicide application. A logistic regression model was used to classify infected and non-infected plants, 31 and 38 days after sunflower planting, with 76 and 89% accuracy, respectively. A partial dataset, containing only 10 spectral bands of the hyperspectral dataset, gave 69 and 82% accuracy, indicating the potential of multi-spectral sensors for the detection task. Sampling pixels from specific sunflower leaf segments improved the classification compared to non-specific sampling. This study thus contributes to establishing a basis for future development of site-specific weed management of O. cumana and of other broomrape species.

This work examines the impact of different environmental attributes on the uncertainty in satellite-based Aerosol Optical Depth (AOD) retrieval against the benchmark Aerosol Robotic Network (AERONET) AOD measurements at 21 sites across North Africa, California and Germany, in the years 2007–2017. As a first step, we studied the effects of spatial averaging the Multi-Angle Implementation of Atmospheric Correction (MAIAC) AOD retrievals, and of temporal averaging the AERONET AOD around the satellite (Aqua) overpass, on the agreement between the two products. AERONET AOD averaging over a time-window of ±15 min around the satellite overpass and the 1 × 1 km² spatial grid of MAIAC were found to provide the best AOD retrieval performance. Next, MAIAC AOD were stratified according to different co-measured environmental attributes (aerosol loading, dominant particle size, vegetation cover, and prevailing particle type) and analyzed against the AERONET AOD. The envelope of the expected retrieval error varied considerably among different environmental attributes categories, with more accurate AOD retrievals obtained over highly vegetated areas (i.e. less surface reflectance) than over arid areas. Moreover, the retrieval accuracy was found to be sensitive to the aerosol loading and particle size, with a large bias between the MAIAC and AERONET AOD during high aerosol loading of coarse particles. In addition, the retrieval accuracy of MAIAC AOD was found to depend on the aerosol type due to the aerosol model assumptions regarding their optical properties.

Hyperspectral imaging is crucial for a variety of land-cover mapping and analyzing tasks. The available large number of reflected light measurements along a wide range of wavelengths allows for distinguishing between different materials under various conditions. Though, several effects bear an undesired variability within hyperspectral images and increase the complexity of interpreting such data. Two of the most significant effects in this regard are the BRDF and the spectral mixture. Due to the first, the acquisitions geometrical and viewing conditions influences the measured spectral signature of a surface to a large extent. On the other hand, because of the typical low spatial resolution of remotely sensed images, each pixel can contain more than one material. Despite much research addressing either the BRDF effect and ways to correct it or the spectral unmixing, too few works considered these two effects' mutual influence. In this work, we study the BRDF of mixed pixels and present preliminary insights of testing a strategy to correct its undesired impact on the data by incorporating the EMs fractions within an unmixing-based semi-empirical correction model. Experimental results using real laboratory data acquired under controlled conditions clearly show the significant improvement of the corrected reflectance results through the proposed model.

We propose a new methodology for enhancing the spatial resolution of unsupervised classification through a fusion of multispectral and visible images. The new method, DFuSIAL-C (Data Fusion through Spatial Information-Aided Learning for Classification), relies on automatically extracted invariant points (IPs), assumed to have the same land cover type in the two data sources. In contrast to typical methods, DFuSIAL-C does not require a full spatial, spectral, and temporal overlapping between the data sources and allows for the fusion of data from different sensors. An evaluation of the proposed method, compared to a state-of-the-art pansharpening fusion method, is carried out using Landsat-8 and Sentinel-2 images. Our experimental results show that the DFuSIAL-C obtains unsupervised classification maps with a significantly enhanced spatial resolution and an overall accuracy (OA) of 85%. Furthermore, we show that the proposed method is preferable when full overlapping is not available due to the acquisition by different instruments.

We propose an unmixing framework for enhancing endmember fraction maps using a combination of spectral and visible images. The new method, data fusion through spatial information-aided learning (DFuSIAL), is based on a learning process for the fusion of a multispectral image of low spatial resolution and a visible RGB image of high spatial resolution. Unlike commonly used methods, DFuSIAL allows for fusing data from different sensors. To achieve this objective, we apply a learning process using automatically extracted invariant points, which are assumed to have the same land cover type in both images. First, we estimate the fraction maps of a set of endmembers for the spectral image. Then, we train a spatial-features aided neural network (SFFAN) to learn the relationship between the fractions, the visible bands, and rotation-invariant spatial features for learning (RISFLs) that we extract from the RGB image. Our experiments show that the proposed DFuSIAL method obtains fraction maps with significantly enhanced spatial resolution and an average mean absolute error between 2% and 4% compared to the reference ground truth. Furthermore, it is shown that the proposed method is preferable to other examined state-of-the-art methods, especially when data is obtained from different instruments and in cases with missing-data pixels.

Spectral unmixing is a key tool for a reliable quantitative analysis of remotely sensed data. The process is used to extract subpixel information by estimating the fractional abundances that correspond to pure signatures, known as endmembers (EMs). In standard techniques, the unmixing problem is solved for each pixel individually, relying only on spectral information. Recent studies show that incorporating the image’s spatial information enhances the accuracy of the unmixing results. In this chapter, we present a new methodology for the reconstruction of the fraction abundances from spectral images with a high percentage of corrupted pixels. This is achieved based on a modification of the spectral unmixing method called Gaussian-based spatially adaptive unmixing (GBSAU). Besides, we present a summarized review of the existing spatially adaptive methods.

Training a deep neural network for classification constitutes a major problem in remote sensing due to the lack of adequate field data. Acquiring high-resolution ground truth (GT) by human interpretation is both cost-ineffective and inconsistent. We propose, instead, to utilize high-resolution, hyperspectral images for solving this problem, by unmixing these images to obtain reliable GT for training a deep network. Specifically, we simulate GT from high-resolution, hyperspectral FENIX images, and use it for training a convolutional neural network (CNN) for pixel-based classification. We show how the model can be transferred successfully to classify new mid-resolution VENµS imagery.

Lava flows pose a hazard in volcanic environments and reset ecosystem development. A succession of dated lava flows provides the possibility to estimate the direction and rates of ecosystem development and can be used to predict future development. We examine plant succession, soil development and soil carbon (C) accretion on the historical (post 874 AD) lava flows formed by the Hekla volcano in south Iceland. Vegetation and soil measurements were conducted all around the volcano reflecting the diverse vegetation communities on the lavas, climatic conditions around Hekla mountain and various intensities in deposition of loose material. Multivariate analysis was used to identify groups with similar vegetation composition and patterns in the vegetation. The association of vegetation and soil parameters with lava age, mean annual temperature, mean annual precipitation and soil accumulation rate (SAR) was analysed. Soil carbon concentration increased with increasing lava age becoming comparable to concentrations found on the prehistoric lavas. The combination of a sub-Arctic climate, gradual soil thickening due to input of loose material and the specific properties of volcanic soils allow for continuing accumulation of soil carbon in the soil profile. Four successional stages were identified: initial colonization and cover coalescence (ICC) of Racomitrium lanuginosum and Stereocaulon spp. (lavas <70 years of age); secondary colonization (SC) – R. lanuginosum dominance (170−700 years); vascular plant dominance (VPD) (>600 years); and highland conditions/retrogression (H/R) by tephra deposition (70−860 years). The long time span of the SC stage indicates arrested development by the thick R. lanuginosum moss mat. The progression from SC into VPD was linked to age of the lava flows and soil depth, which was significantly deeper within the VPD stage. Birch was growing on lavas over 600 years old indicating the development towards birch woodland, the climax ecosystem in Iceland.

A new methodology is proposed for the enhancement of endmember (EMs) fractions' maps. The new method, termed DFNeFE (data fusion through neural-network for fraction estimation), is based on the fusion of a multispectral image, with low spatial resolution (LSR) and a visible RGB image, with high spatial restitution (HSR), through a back propagation neural network (BPNN). First, the fraction maps of a set of EMs are estimated for the spectral image using an accurate unmixing method. Then spatial statistical features (SSFs) are extracted from both images and a BPNN is trained to learn the relationship between the fractions, the visible bands of the HSR image and the SSFs based on invariant points (IPs) which are assumed to have the same land cover type in both the multispectral and visible images. Using an automatic method for IP extraction, we can also apply our method to images that are not co-registered. An evaluation of the proposed method, is carried out using a real data set with two spectral images acquired by Landsat -8 and Sentine1-2 satellites, and an RGB image available in Google Earth.

Spectral unmixing provides, for each pixel in the image, an estimated vector of fractional abundances that correspond to pure signatures, known as endmembers (EMs). Standard unmixing techniques rely only on spectral information and each pixel is solved as an individual entity. Recent studies show that incorporating the image's spatial information enhances accuracy of the unmixing results. Spatial information may allow better selection of relevant EMs, for each pixel, rather than utilizing all potential EMs in solving the spectral mixing problem. To implement this approach, we developed a new method for spatially adaptive spectral unmixing called Gaussian-based spatially adaptive unmixing (GBSAU). GBSAU fits for each EM, a surface by a series of spatial anisotropic 2D Gaussians whose sum represents the EM's fraction distribution over the whole image. These analytical surfaces facilitate a sparse solution for the unmixing process by spatial localization of EMs, which is then used to determine an adaptive subset of actual EMs for each pixel. The performance of our novel method was compared with that of the state-of-art spatially adaptive unimximg method, sparse unmixing via variable splitting augmented Lagrangian and total variation (SUnSAL-TV) as well as with two ordinary non-spatial methods, sparse unmixing by variable splitting and augmented Lagrangian (SUnSAL) and vectorized code projected gradient descent unmixing (VPGDU). The comparison was carried out on both simulated and real hyperspectral images. The results obtained with GBSAU indicated a significant improvement in the overall accuracy of the unmixing process compared with both spatially adaptive and ordinary methods. Quantitatively, GBSAU reduces the average mean absolute error (MAE) of the results by ∼15%, for cases with SNR = 30 db and 20 db, and by ∼30% for cases with SNR = 10 db and 5 db.

The empirical line (EL) calibration method is commonly used for atmospheric correction of remotely sensed spectral images and recovery of surface reflectance. The current EL-based methods are applicable to calibrate only single images. Therefore, the use of the EL calibration is impractical for imaging campaigns, where many (partially overlapped) images are acquired to cover a large area. In addition, the EL results are unconstrained and an undesired reflectance with negative values or larger than 100% can be obtained. In this paper, we use the standard EL model to formulate a new generalized empirical line (GEL) model. Based on the GEL, we present a novel method for simultaneous and constrained calibration of multiple images. This new method allows for calibration through multiple image constrained empirical line (MIcEL) and three additional calibration modes. Given a set of images, we use the available ground targets and automatically extracted tie points between overlapping images to calibrate all the images in the set simultaneously. Quantitative and visual assessments of the proposed method were carried out relatively to the off-the-shelf method quick atmospheric correction (QUAC), using real hyperspectral images and field measurements. The results clearly show the superiority of MIcEL with respect to the minimization of the difference between the reflectance values of the same object in different overlapping images. An assessment of the absolute accuracy, with respect to 11 field measurement points, shows that the accuracy of MIcEL, with an average mean absolute error (MAE) of ∼11%, is comparable with respect to the QUAC.

Low-cost air quality sensors offer high-resolution spatiotemporal measurements that can be used for air resources management and exposure estimation. Yet, such sensors require frequent calibration to provide reliable data, since even after a laboratory calibration they might not report correct values when they are deployed in the field, due to interference with other pollutants, as a result of sensitivity to environmental conditions and due to sensor aging and drift. Field calibration has been suggested as a means for overcoming these limitations, with the common strategy involving periodical collocations of the sensors at an air quality monitoring station. However, the cost and complexity involved in relocating numerous sensor nodes back and forth, and the loss of data during the repeated calibration periods make this strategy inefficient. This work examines an alternative approach, a node-to-node (N2N) calibration, where only one sensor in each chain is directly calibrated against the reference measurements and the rest of the sensors are calibrated sequentially one against the other while they are deployed and collocated in pairs. The calibration can be performed multiple times as a routine procedure. This procedure minimizes the total number of sensor relocations, and enables calibration while simultaneously collecting data at the deployment sites. We studied N2N chain calibration and the propagation of the calibration error analytically, computationally and experimentally. The in-situ N2N calibration is shown to be generic and applicable for different pollutants, sensing technologies, sensor platforms, chain lengths, and sensor order within the chain. In particular, we show that chain calibration of three nodes, each calibrated for a week, propagate calibration errors that are similar to those found in direct field calibration. Hence, N2N calibration is shown to be suitable for calibration of distributed sensor networks. Node-to-node calibration is proposed as a general method for field calibration of wireless distributed air-quality sensor networks.

Lava flow thicknesses, volumes, and effusion rates provide essential information for understanding the behavior of eruptions and their associated deformation signals. Preeruption and posteruption elevation models were generated from historical stereo photographs to produce the lava flow thickness maps for the last five eruptions at Hekla volcano, Iceland. These results provide precise estimation of lava bulk volumes: V_1947–1948 = 0.742 ± 0.138 km³, V₁₉₇₀ = 0.205 ± 0.012 km³, V_1980–1981 = 0.169 ± 0.016 km³, V₁₉₉₁ = 0.241 ± 0.019 km³, and V₂₀₀₀ = 0.095 ± 0.005 km³ and reveal variable production rate through the 20th century. These new volumes improve the linear correlation between erupted volume and coeruption tilt change, indicating that tilt may be used to determine eruption volume. During eruptions the active vents migrate 325–480 m downhill, suggesting rough excess pressures of 8–12 MPa and that the gradient of this excess pressure increases from 0.4 to 11 Pa s⁻¹ during the 20th century. We suggest that this is related to increased resistance along the eruptive conduit.

— Hekla volcano is known to have erupted at least 23 times in historical time (last 1100 years); often producing mixed eruptions of tephra and lava. The lava flow volumes from the 20th century have amounted 80% to almost 100% of the entire erupted volume. Therefore, evaluating the extent and volume of individual lava flows is very important when assessing the historical productivity of Hekla volcano. Here we present new maps of the historical lava flow fields at Hekla in a digital format. The maps were produced at a scale of 1:2000–10000 using a catalogue of orthophotos since 1945, acquired before and after each of the last five eruptions, combined with field observation of stratigraphy, soil profiles, tephra layers and vegetation cover. The new lava flow maps significantly improve the historical eruptive history of Hekla, prior to the 1947 eruption. The historical lava flow fields from Hekla cover ∼233 km² and the lavas reach up to 16 km from Hekla volcano. Flow lengths up to 20 km are known, though lava flows only travelled up to 8–9 km from Hekla in the last 250 years. Identified historical vents are distributed between 0 and 16 km from Hekla volcano and vents are known to have migrated up to 5 km away from Hekla during eruptions. We have remapped the lava flow fields around Hekla and assigned the identified flow fields to 16 eruptions. In addition, ca. 60 unidentified lava units, which may be of historical age, have been mapped. It is expected that some of these units are from known historical Hekla eruptions such as the 1222, 1341, 1510, 1597, 1636 and potentially even from the previously excluded eruptions such as 1436/1439.

The empirical line (EL) calibration is commonly used for atmospheric correction of remotely sensed spectral images and recovery of surface reflectance. Current methods for EL calibration are applied to single image using two (or more) reference targets. Considering cases with large number of (partially overlapped) images, only few scenes will include reference targets. Moreover, applying the estimated calibration coefficients of one image to other images can cause wrong results. Accordingly, the use of EL calibration is impractical for these cases. In this paper, we present a novel method for a simultaneous calibration of multiple images, which is called multiple image constrained empirical line (MIcEL). We present a generalized EL model that provide constrained results and is adaptable for large number of images. Given a set of images, we use available reference targets and tie points between overlapping images to calibrate all the images in the set simultaneously. Tie points are automatically extracted using scale-invariant feature transform (SIFT) method. Accuracy assessment of the MIcEL was carried out using real hyperspectral images and field measurements. The performance of MIcEL was compared to the quick atmospheric correction (QUAC) method. the results show that (comparable with respect to QUAC) the absolute accuracy of the MIcEL, with respect to filed measurements, is ∼ ± 11%.

We present, in this paper, a new methodology for spectral unmixing, where a vector of fractions, corresponding to a set of endmembers (EMs), is estimated for each pixel in the image. The process first provides an initial estimate of the fraction vector, followed by an iterative procedure that converges to an optimal solution. Specifically, projected gradient descent (PGD) optimization is applied to (a variant of) the spectral angle mapper objective function, so as to significantly reduce the estimation error due to amplitude (i.e., magnitude) variations in EM spectra, caused by the illumination change effect. To improve the computational efficiency of our method over a commonly used gradient descent technique, we have analytically derived the objective function's gradient and the optimal step size (used in each iteration). To gain further improvement, we have implemented our unmixing module via code vectorization, where the entire process is 'folded' into a single loop, and the fractions for all of the pixels are solved simultaneously. We call this new parallel scheme vectorized code PGD unmixing (VPGDU). VPGDU has the advantage of solving (simultaneously) an independent optimization problem per image pixel, exactly as other pixelwise algorithms, but significantly faster. Its performance was compared with the commonly used fully constrained least squares unmixing (FCLSU), the generalized bilinear model (GBM) method for hyperspectral unmixng, and the fast state-of-the-art methods, sparse unmixing by variable splitting and augmented Lagrangian (SUnSAL) and collaborative SUnSAL (CLSUnSAL) based on the alternating direction method of multipliers. Considering all of the prospective EMs of a scene at each pixel (i.e., without a priori knowledge which/how many EMs are actually present in a given pixel), we demonstrate that the accuracy due to VPGDU is considerably higher than that obtained by FCLSU, GBM, SUnSAL, and CLSUnSAL under varying illumination, and is, otherwise, comparable with respect to these methods. However, while our method is significantly faster than FCLSU and GBM, it is slower than SUnSAL and CLSUnSAL by roughly an order of magnitude.

Recent developments in sensory and communication technologies have made the development of portable air-quality (AQ) micro-sensing units (MSUs) feasible. These MSUs allow AQ measurements in many new applications, such as ambulatory exposure analyses and citizen science. Typically, the performance of these devices is assessed using the mean error or correlation coefficients with respect to a laboratory equipment. However, these criteria do not represent how such sensors perform outside of laboratory conditions in large-scale field applications, and do not cover all aspects of possible differences in performance between the sensor-based and standardized equipment, or changes in performance over time. This paper presents a comprehensive Sensor Evaluation Toolbox (SET) for evaluating AQ MSUs by a range of criteria, to better assess their performance in varied applications and environments. Within the SET are included four new schemes for evaluating sensors' capability to: locate pollution sources; represent the pollution level on a coarse scale; capture the high temporal variability of the observed pollutant and their reliability. Each of the evaluation criteria allows for assessing sensors' performance in a different way, together constituting a holistic evaluation of the suitability and usability of the sensors in a wide range of applications. Application of the SET on measurements acquired by 25 MSUs deployed in eight cities across Europe showed that the suggested schemes facilitates a comprehensive cross platform analysis that can be used to determine and compare the sensors' performance. The SET was implemented in R and the code is available on the first author's website.

Performing standard unmixing of a hyperspectral image, while taking into account all of the potential endmembers (EMs) in a pixel, is known to be prone to error. Instead, determining first the set of EMs that actually reside in each pixel, leads to enhanced unmixing results. This important insight for achieving higher unmixing accuracy can be exploited efficiently by extracting relevant spatial information from a given image. In this work, we present a new method for spatially adaptive spectral unmixing, called the Gaussian based spatially adaptive unmixing (GBSAU) method. GBSAU takes advantage of the spatial arrangement of the image pixels and their spectral relations in order to determine an actual subset of EMs per pixel. It is based on spatial localization of the EMs by fitting, for each EM, the parameters of the series of spatial Gaussians whose sum represents the EM's fraction surface over the image.

A novel unmixing methodology is presented, searching for a fraction combination of end-members (EMs) that reconstructs the integrated source signal. The search starts with computing an initially estimated unmixing solution and then assesses combinations selected at random within an envelope surrounding this estimated solution. From each of these combinations, it then progresses iteratively along a path of neighboring combinations, so as to minimize the spectral angle between the corresponding (integrated) signatures and the source signal, until reaching a satisfactory solution. The new iterative fraction combination search (IFCS) was compared to the standard least squares unmixing (LSU). An assessment of both methods was conducted with a real Airborne Visible/Infrared Imaging Spectrometer image and nine synthetic images generated by randomly selecting fractions for two up to ten EMs derived from this real image. Considering all these EMs for the unmixing solution (not knowing specifically which or how many of them are actually mixed at each pixel), the IFCS method performed considerably better than LSU.