Remote Sensing — Open Access Journal

An optimal payload selection conducted in the frame of the H2020 ONION project (id 687490) is presented based on the ability to cover the observation needs of the Copernicus system in the time period 2020–2030. Payload selection is constrained by the variables that can be measured, the power consumption, and weight of the instrument, and the required accuracy and spatial resolution (horizontal or vertical). It involved 20 measurements with observation gaps according to the user requirements that were detected in the top 10 use cases in the scope of Copernicus space infrastructure, 9 potential applied technologies, and 39 available commercial platforms. Additional Earth Observation (EO) infrastructures are proposed to reduce measurements gaps, based on a weighting system that assigned high relevance for measurements associated to Marine for Weather Forecast over Polar Regions. This study concludes with a rank and mapping of the potential technologies and the suitable commercial platforms to cover most of the requirements of the top ten use cases, analyzing the Marine for Weather Forecast, Sea Ice Monitoring, Fishing Pressure, and Agriculture and Forestry: Hydric stress as the priority use cases. Full article

Irregular spatial dependency is one of the major characteristics of remote sensing images, which brings about challenges for classification tasks. Deep supervised models such as convolutional neural networks (CNNs) have shown great capacity for remote sensing image classification. However, they generally require a huge labeled training set for the fine tuning of a deep neural network. To handle the irregular spatial dependency of remote sensing images and mitigate the conflict between limited labeled samples and training demand, we design a superpixel-guided layer-wise embedding CNN (SLE-CNN) for remote sensing image classification, which can efficiently exploit the information from both labeled and unlabeled samples. With the superpixel-guided sampling strategy for unlabeled samples, we can achieve an automatic determination of the neighborhood covering for a spatial dependency system and thus adapting to real scenes of remote sensing images. In the designed network, two types of loss costs are combined for the training of CNN, i.e., supervised cross entropy and unsupervised reconstruction cost on both labeled and unlabeled samples, respectively. Our experimental results are conducted with three types of remote sensing data, including hyperspectral, multispectral, and synthetic aperture radar (SAR) images. The designed SLE-CNN achieves excellent classification performance in all cases with a limited labeled training set, suggesting its good potential for remote sensing image classification. Full article

This paper presents a proposal for a generic urban structure type (UST) scheme. Initially developed in the context of urban ecology, the UST approach is increasingly popular in the remote sensing community. However, there is no consistent and standardized UST framework. Until now, the terms land use and certain USTs are often used and described synonymously, or components of structure and use are intermingled. We suggest a generic nomenclature and a respective UST scheme that can be applied worldwide by stakeholders of different disciplines. Based on the insights of a rigorous literature analysis, we formulate a generic structural- and object-based typology, allowing for the generation of hierarchically and terminologically consistent USTs. The developed terminology exclusively focuses on morphology, urban structures and the general exterior appearance of buildings. It builds on the delimitation of spatial objects at several scales and leaves out all social aspects and land use aspects of an urban area. These underlying objects or urban artefacts and their structure- and object-related features, such as texture, patterns, shape, etc. are the core of the hierarchically structured UST scheme. Finally, the authors present a generic framework for the implementation of a remote sensing-based UST classification along with the requirements regarding sensors, data and data types. Full article

Accurate calibration of oceanic LiDAR signals is essential for the accurate retrieval of ocean optical properties. Nowadays, there are many methods for aerosol LiDAR calibration, but fewer attempts have been made to implement specific calibration methods for oceanic LiDAR. Oceanic LiDAR often has higher vertical resolution, needs greater signal dynamic range, detects several orders of magnitude lower less depth of penetration, and suffers from the effects of the air-sea interface. Therefore the calibration methods for aerosol LiDAR may not be useful for oceanic LiDAR. In this paper, we present a new simple and feasible approach for oceanic LiDAR calibration via comparison of LiDAR backscatter against calculated scatter based on iteratively bio-optical models in clear, open ocean, Type 1 water. Compared with current aerosol LiDAR calibration methods, it particularly considers geometric losses and attenuation occurring in the atmosphere-sea interface. The mean relative error percentage (MREP) of LiDAR calibration constant at two different stations was all within 0.08%. The MREP between LiDAR-retrieved backscatter, chlorophyll after using LiDAR calibration constant with inversion results of measured data were within 0.18% and 1.39%, respectively. These findings indicate that the bio-optical methods for LiDAR calibration in clear ocean water are feasible and effective. Full article

Ionospheric delay is a significant error source in multi-GNSS positioning. We present different processing strategies to fully exploit the ionospheric delay effects on multi-frequency and multi-GNSS positioning performance, including standard point positioning (SPP) and precise point positioning (PPP) scenarios. Datasets collected from 10 stations over thirty consecutive days provided by multi-GNSS experiment (MGEX) stations were used for single-frequency SPP/PPP and dual-frequency PPP tests with quad-constellation signals. The experimental results show that for single-frequency SPP, the Global Ionosphere Maps (GIMs) correction achieves the best accuracy, and the accuracy of the Neustrelitz TEC model (NTCM) solution is better than that of the broadcast ionospheric model (BIM) in the E and U components. Eliminating ionospheric parameters by observation combination is equivalent to estimating the parameters in PPP. Compared with the single-frequency uncombined (UC) approach, the average convergence time of PPP with the external ionospheric models is reduced. The improvement in BIM-, NTCM- and GIM-constrained quad-constellation L2 single-frequency PPP was 15.2%, 24.8% and 28.6%, respectively. The improvement in convergence time of dual-frequency PPP with ionospheric models was different for different constellations and the GLONASS-only solution showed the least improvement. The improvement in the convergence time of BIM-, NTCM- and GIM-constrained quad-constellation L1/L2 dual-frequency PPP was 5.2%, 6.2% and 8.5%, respectively, compared with the UC solution. The positioning accuracy of PPP is slightly better with the ionosphere constraint and the performance of the GIM-constrained PPP is the best. The combination of multi-GNSS can effectively improve the positioning performance. Full article

Flash flood, one of the most devastating weather-related hazards in the world, has become more and more frequent in past decades. For the purpose of flood mitigation, it is necessary to understand the distribution of flash flood risk. In this study, artificial intelligence (Least squares support vector machine: LSSVM) and classical canonical method (Logistic regression: LR) are used to assess the flash flood risk in the Yunnan Province based on historical flash flood records and 13 meteorological, topographical, hydrological and anthropological factors. Results indicate that: (1) the LSSVM with Radial basis function (RBF) Kernel works the best (Accuracy = 0.79) and the LR is the worst (Accuracy = 0.75) in testing; (2) flash flood risk distribution identified by the LSSVM in Yunnan province is near normal distribution; (3) the high-risk areas are mainly concentrated in the central and southeastern regions, where with a large curve number; and (4) the impact factors contributing the flash flood risk map from higher to low are: Curve number > Digital elevation > Slope > River density > Flash Flood preventions > Topographic Wetness Index > annual maximum 24 h precipitation > annual maximum 3 h precipitation. Full article

The OLI (Operational Land Imager) sensor on Landsat-8 has the potential to meet the requirements of remote sensing of water color. However, the optical properties of inland waters are more complex than those of oceanic waters, and inland atmospheric correction presents additional challenges. We examined the performance of atmospheric correction (AC) methods for remote sensing over three highly turbid or hypereutrophic inland waters in China: Lake Hongze, Lake Chaohu, and Lake Taihu. Four water-AC algorithms (SWIR (Short Wave Infrared), EXP (Exponential Extrapolation), DSF (Dark Spectrum Fitting), and MUMM (Management Unit Mathematics Models)) and three land-AC algorithms (FLAASH (Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes), 6SV (a version of Second Simulation of the Satellite Signal in the Solar Spectrum), and QUAC (Quick Atmospheric Correction)) were assessed using Landsat-8 OLI data and concurrent in situ data. The results showed that the EXP (and DSF) together with 6SV algorithms provided the best estimates of the remote sensing reflectance (R_rs) and band ratios in water-AC algorithms and land-AC algorithms, respectively. AC algorithms showed a discriminating accuracy for different water types (turbid waters, in-water algae waters, and floating bloom waters). For turbid waters, EXP gave the best R_rs in visible bands. For the in-water algae and floating bloom waters, however, all water-algorithms failed due to an inappropriate aerosol model and non-zero reflectance at 1609 nm. The results of the study show the improvements that can be achieved considering SWIR bands and using band ratios, and the need for further development of AC algorithms for complex aquatic and atmospheric conditions, typical of inland waters. Full article

Global food demand will increase over the next few decades, and sustainable agricultural intensification on current cropland may be a preferred option to meet this demand. Mapping cropping intensity with remote sensing data is of great importance for agricultural production, food security, and agricultural sustainability in the context of global climate change. However, there are some challenges in large-scale cropping intensity mapping. First, existing indicators are too coarse, and fine indicators for measuring cropping intensity are lacking. Second, the regional, intra-class variations detected in time-series remote sensing data across vast areas represent environment-related clusters for each cropping intensity level. However, few existing studies have taken into account the intra-class variations caused by varied crop patterns, crop phenology, and geographical differentiation. In this research, we first presented a new definition, a normalized cropping intensity index (CII), to quantify cropping intensity precisely. We then proposed a Bayesian network model fusing prior knowledge (BNPK) to address the issue of intra-class variations when mapping CII over large areas. This method can fuse regional differentiation factors as prior knowledge into the model to reduce the uncertainty. Experiments on five sample areas covering the main grain-producing areas of mainland China proved the effectiveness of the model. Our research proposes the framework of obtain a CII map with both a finer spatial resolution and a fine temporal resolution at a national scale. Full article

Quiescent volcanoes dissipate a large part of their thermal energy through hot soils and ground degassing mainly in restricted areas called Diffuse Degassing Structures. La Solfatara crater represents the main spot of thermal release for the Campi Flegrei volcano (Italy) despite its reduced dimensions with regards to the whole caldera. The purpose of this study was to develop a method to measure thermal energy release extrapolating it from the ground surface temperature. We used imaging from thermal cameras at short distances (1 m) to obtain a mapping of areas with thermal anomalies and a measure of their temperatures. We built a conceptual model of the energy release from the ground to atmosphere, which well fits the experimental data taken in the La Solfatara crater. Using our model and data, we could estimate the average heat flux in a portion of the crater as $q_{avg}=220\pm 40{\mathrm{W}/\mathrm{m}}^2$, compatible with other measurements in literature. Full article

The Lianjiang Plain in China and ancient villages distributed within the plain are under the potential threat of surface motion change, but no effective monitoring strategy currently exists. Distributed Scatterer InSAR (DSInSAR) provides a new high-resolution method for the precise detection of surface motion change. In contrast to the first-generation of time-series InSAR methodology, the distributed scatterer-based method focuses both on pointwise targets with high phase stability and distributed targets with moderate coherence, the latter of which is more suitable for the comprehensive environment of the Lianjiang Plain. In this paper, we present the first study of surface motion change detection in the Lianjiang Plain, China. Two data stacks, including 54 and 29 images from Sentinel-1A adjacent orbits, are used to retrieve time-series surface motion changes for the Lianjiang Plain from 2015 to 2018. The consistency of measurement has been cross-validated between adjacent orbit results with a statistically significant determination coefficient of 0.92. The temporal evolution of representative measuring points indicates three subzones with varied surface patterns: Eastern Puning (Zone A) in a slight elastic rebound phase with a moderate deformation rate (0–40 mm/year), Chaonan (Zone B) in a substantial subsidence phase with a strong deformation rate (−140–0 mm/year), and Chaoyang (Zone C) in a homogeneous and stable situation (−10–10 mm/year). The spatial distribution of these zones suggests a combined change dynamic and a strong concordance of factors impacting surface motion change. Human activities, especially groundwater exploitation, dominate the subsidence pattern, and natural conditions act as a supplementary inducement by providing a hazard-prone environment. The qualitative and quantitative analysis of spatial and temporal details in this study provides a basis for systematic surface motion monitoring, cultural heritage protection and groundwater resources management. Full article

The main aim of this paper is to test the use of multi-temporal differential interferometric synthetic aperture radar (DInSAR) techniques as a tool for geological and geomorphological surveys in urban areas, where anthropogenic features often completely obliterate landforms and surficial deposits. In the last two decades, multi-temporal DInSAR techniques have been extensively applied to many topics of Geosciences, especially in geohazard analysis and risks assessment, but few attempts have been made in using differential subsidence for geological and geomorphological mapping. With this aim, interferometric data of an urbanized sector of the Venetian-Friulian Plain were considered. The data derive by permanent scatterers InSAR processing of synthetic aperture radar (SAR) images acquired by ERS 1/2, ENVISAT, COSMO SKY-Med and Sentinel-1 missions from 1992 to 2017. The obtained velocity maps identify, with high accuracy, the border of a fluvial incised valley formed after the last glacial maximum (LGM) and filled by unconsolidated Holocene deposits. These consist of lagoon and fluvial sediments that are affected by a much higher subsidence than the surrounding LGM deposits forming the external plain. Displacement time-series of localized sectors inside the post-LGM incision allowed the causes of vertical movements to be explored, which consist of the consolidation of recent deposits, due to the loading of new structures and infrastructures, and the exploitation of the shallow phreatic aquifer. Full article

The global availability of high spatial resolution images makes mapping tree species distribution possible for better management of forest resources. Previous research mainly focused on mapping single tree species, but information about the spatial distribution of all kinds of trees, especially plantations, is often required. This research aims to identify suitable variables and algorithms for classifying land cover, forest, and tree species. Bi-temporal ZiYuan-3 multispectral and stereo images were used. Spectral responses and textures from multispectral imagery, canopy height features from bi-temporal stereo imagery, and slope and elevation from the stereo-derived digital surface model data were examined through comparative analysis of six classification algorithms including maximum likelihood classifier (MLC), k-nearest neighbor (kNN), decision tree (DT), random forest (RF), artificial neural network (ANN), and support vector machine (SVM). The results showed that use of multiple source data—spectral bands, vegetation indices, textures, and topographic factors—considerably improved land-cover and forest classification accuracies compared to spectral bands alone, which the highest overall accuracy of 84.5% for land cover classes was from the SVM, and, of 89.2% for forest classes, was from the MLC. The combination of leaf-on and leaf-off seasonal images further improved classification accuracies by 7.8% to 15.0% for land cover classes and by 6.0% to 11.8% for forest classes compared to single season spectral image. The combination of multiple source data also improved land cover classification by 3.7% to 15.5% and forest classification by 1.0% to 12.7% compared to the spectral image alone. MLC provided better land-cover and forest classification accuracies than machine learning algorithms when spectral data alone were used. However, some machine learning approaches such as RF and SVM provided better performance than MLC when multiple data sources were used. Further addition of canopy height features into multiple source data had no or limited effects in improving land-cover or forest classification, but improved classification accuracies of some tree species such as birch and Mongolia scotch pine. Considering tree species classification, Chinese pine, Mongolia scotch pine, red pine, aspen and elm, and other broadleaf trees as having classification accuracies of over 92%, and larch and birch have relatively low accuracies of 87.3% and 84.5%. However, these high classification accuracies are from different data sources and classification algorithms, and no one classification algorithm provided the best accuracy for all tree species classes. This research implies the same data source and the classification algorithm cannot provide the best classification results for different land cover classes. It is necessary to develop a comprehensive classification procedure using an expert-based approach or hierarchical-based classification approach that can employ specific data variables and algorithm for each tree species class. Full article

Uganda is one of the poorest nations in the world. It is important to obtain accurate, timely data on socio-economic characteristics sub-nationally, so as to target poverty reduction strategies to those most in need. Many studies have demonstrated that nighttime lights (NTL) can be used to measure human activities. Nevertheless, the methods developed from these studies (1) suffer from coarse resolutions, (2) fail to capture the nonlinearity and multi-scale variability of geospatial data, and (3) perform poorly for agriculture-dependent regions. This study proposes a new enhanced light intensity model (ELIM) to estimate the gross domestic product (GDP) for sub-national units within Uganda. This model is developed by combining the NTL data from the Suomi National Polar-orbiting Partnership Visible Infrared Imaging Radiometer Suite (NPP-VIIRS), the population data from the Global Human Settlement Layer (GHSL), and information on agricultural production and market prices across several commodity types. This resulted in a gridded dataset for Uganda’s GDP at sub-national levels, to capture the spatial heterogeneity in the economic activity. Full article

This paper proposes a method of combining and visualizing multiple lithological indices derived from Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) data and topographical information derived from digital elevation model (DEM) data in a single color image that can be easily interpreted from a geological point of view. For the purposes of mapping silicate rocks, carbonate rocks, and clay minerals in hydrothermal alteration zones, two new indices derived from ASTER thermal infrared emissivity data were developed to identify silicate rocks, and existing indices were adopted to indicate the distribution of carbonate rocks and the species and amounts of clay mineral. In addition, another new method was developed to visualize the topography from DEM data. The lithological indices and topographical information were integrated using the hue–saturation–value (HSV) color model. The resultant integrated image was evaluated by field survey and through comparison with the results of previous studies in the Cuprite and Goldfield areas, Nevada, USA. It was confirmed that the proposed method can be used to visualize geological information and that the resulting images can easily be interpreted from a geological point of view. Full article

The objective of this study was to develop a decision-based methodology, focused on data fusion for wetland classification based on surface water hydroperiod and associated riparian (transitional area between aquatic and upland zones) vegetation community attributes. Multi-temporal, multi-mode data were examined from airborne Lidar (Teledyne Optech, Inc., Toronto, ON, Canada, Titan), synthetic aperture radar (Radarsat-2, single and quad polarization), and optical (SPOT) sensors with near-coincident acquisition dates. Results were compared with 31 field measurement points for six wetlands at riparian transition zones and surface water extents in the Utikuma Regional Study Area (URSA). The methodology was repeated in the Peace-Athabasca Delta (PAD) to determine the transferability of the methods to other boreal environments. Water mask frequency analysis showed accuracies of 93% to 97%, and kappa values of 0.8–0.9 when compared to optical data. Concordance results comparing the semi-permanent/permanent hydroperiod between 2015 and 2016 were found to be 98% similar, suggesting little change in wetland surface water extent between these two years. The results illustrate that the decision-based methodology and data fusion could be applied to a wide range of boreal wetland types and, so far, is not geographically limited. This provides a platform for land use permitting, reclamation monitoring, and wetland regulation in a region of rapid development and uncertainty due to climate change. The methodology offers an innovative time series-based boreal wetland classification approach using data fusion of multiple remote sensing data sources. Full article

The sea surface essentially contains multiscale roughness with capillary waves of many sizes riding on large-scale waves that are also of many sizes. It is instructive to exploit the effect of radar frequency and observation geometry on the effective roughness scales responsible for radar backscattering so that the scattering mechanism and the scattering source can be better understood and quantitated. Based on common sea spectra and a theoretical scattering model, an attempt is made to attain the above objective. Model predictions, with selective roughness scales, are compared with wide validation data, including L-band radar observations, and predictions from C-band and Ku-band empirical models: geophysical model function (CMOD7) and NASA scatterometer (NSCAT-4) for C- and Ku-bands at different incident angles. Numerical results indicate that effective roughness scales for radar backscattering vary with radar frequency and incidence angle and are related to a portion of sea spectral components; the low limit of which is linearly proportional to the Bragg wavenumber determined by frequency and incidence angle, and the scale factor of the linear relationship is about 0.05. In addition, the root mean square (RMS) height and the correlation length of the effective roughness (i.e., scattering source) derived from the effective roughness decrease gradually as incident angle increases. In particular, the correlation length also linearly depends on the effective wavelength with a coefficient of 3.2. Moreover, these two coefficients are both independent of wind speed, radar frequency, and incident angle. These findings also reveal the essential properties of the spectral components contributing to radar backscattering and its variation with radar frequency and incident angle. Full article

Hyperspectral images (HSIs) data that is typically presented in 3-D format offers an opportunity for 3-D networks to extract spectral and spatial features simultaneously. In this paper, we propose a novel end-to-end 3-D dense convolutional network with spectral-wise attention mechanism (MSDN-SA) for HSI classification. The proposed MSDN-SA exploits 3-D dilated convolutions to simultaneously capture the spectral and spatial features at different scales, and densely connects all 3-D feature maps with each other. In addition, a spectral-wise attention mechanism is introduced to enhance the distinguishability of spectral features, which improves the classification performance of the trained models. Experimental results on three HSI datasets demonstrate that our MSDN-SA achieves competitive performance for HSI classification. Full article

Lakes in arid regions play an important role in regional water cycles and are a vital economic resource, but can fluctuate widely in area and volume. This study demonstrates the use of a multisensor satellite remote sensing method for the comprehensive monitoring of lake surface areas, water levels, and volume for the Toshka Lakes in southern Egypt, from lake formation in 1998 to mid-2017. Two spectral water indices were used to construct a daily time-series of surface area from the Advanced Very High Resolution Radiometer (AVHRR) and the Moderate Resolution Imaging Spectroradiometer (MODIS), validated by higher-resolution Landsat images. Water levels were obtained from analysis of digital elevation models from the Shuttle Radar Topography Mission (SRTM) and the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), validated with ICESat Geoscience Laser Altimeter System (GLAS) laser altimetry. Total lake volume peaked at 26.54 × 10⁹ m³ in December 2001, and declined to 0.76 × 10⁹ m³ by August 2017. Evaporation accounted for approximately 86% of the loss, and groundwater recharge accounted for 14%. Without additional inflows, the last remaining lake will likely disappear between 2020 and 2022. The Enhanced Lake Index, a water index equivalent to the Enhanced Vegetation Index, was found to have lower noise levels than the Normalized Difference Lake Index. The results show that multi-platform satellite remote sensing provides an efficient method for monitoring the hydrology of lakes. Full article

Collection of spectroradiometric measurements with associated biophysical variables is an essential part of the development and validation of optical remote sensing vegetation products. However, their quality can only be assessed in the subsequent analysis, and often there is a need for collecting extra data, e.g., to fill in gaps. To generate empirical-like surface reflectance data of vegetated surfaces, we propose to exploit emulation, i.e., reconstruction of spectral measurements through statistical learning. We evaluated emulation against classical interpolation methods using an empirical field dataset with associated hyperspectral spaceborne CHRIS and airborne HyMap reflectance spectra, to produce synthetic CHRIS and HyMap reflectance spectra for any combination of input biophysical variables. Results indicate that: (1) emulation produces surface reflectance data more accurately than interpolation when validating against a separate part of the field dataset; and (2) emulation produces the spectra multiple times (tens to hundreds) faster than interpolation. This technique opens various data processing opportunities, e.g., emulators not only allow rapidly producing large synthetic spectral datasets, but they can also speed up computationally intensive processing routines such as synthetic scene generation. To demonstrate this, emulators were run to simulate hyperspectral imagery based on input maps of a few biophysical variables coming from CHRIS, HyMap and Sentinel-2 (S2). The emulators produced spaceborne CHRIS-like and airborne HyMap-like surface reflectance imagery in the order of seconds, thereby approximating the spectra of vegetated surfaces sufficiently similar to the reference images. Similarly, it took a few minutes to produce a hyperspectral data cube with a spatial texture of S2 and a spectral resolution of HyMap. Full article