The Greenhouse Gas Monitoring Instrument (GMI) is a short-wavelength infrared (SWIR) hyperspectral-resolution spectrometer onboard the Chinese satellite GaoFen-5 that uses a spatial heterodyne spectroscopy (SHS) interferometer to acquire interferograms. The GMI was designed to measure and study the source and sink processes of carbon dioxide and methane in the troposphere where the greenhouse effect occurs. In this study, the processing and geometric correction algorithms of the GMI Level 1 product (radiance spectrum) are introduced. The spectral quality and greenhouse gas (GHG) inversion ability of the Level 1 products are analyzed, and the results illustrate that the specifications meet the mission’s requirements. An initial evaluation of the resolution, signal-to-noise ratio (SNR), and stability of the radiance spectrum reveals that the overall function and performance are within the design objectives. A comparison between our Level 1 products and the theoretical spectrum shows that the root mean square (rms) of the residual is approximately 0.8%, and the Level 1 products of the GMI captured within five months after observations have good spectral stability characteristics (less than 0.005 cm−1 for Band 1, 0.003 cm−1 for Band 2, 0.002 cm−1 for Band 3, and 0.004 cm−1 for Band 4). These results demonstrate that the GMI payload and the processing algorithm all work well and reliably. Furthermore, based on the Level 1 products, a GHG retrieval experiment is carried out, and the results are compared with data from Total Column Carbon Observing Network (TCCON) stations. The initial comparison of the XCO2 results yields a value of 0.869 for ${R}^{2}$ (goodness of fit), 0.51 ppm for bias (mean of absolute error), and 0.53 ppm for ${sigma }$ (standard deviation of error). Similarly, the XCH4 comparison yields values of 0.841 for ${R}^{2}$ , 4.64 ppb for bias, and 4.66 ppb for $ {sigma}$ .
Recent optical remote sensing satellite missions, such as Sentinel-2 with the MultiSpectral Imager (MSI) onboard, allow the estimation of coastal water key parameters with very high spatial resolutions (down to 10 m). In this article, multiple approaches are proposed for retrieving chlorophyll-a (Chl-a) and total suspended matter (TSM) along the Adriatic and Tyrrhenian coasts in Italy, using both empirical and model-based frameworks to design regressive and neural network (NN) estimation methods. The latter proves to be more accurate on a regional scale, where standard ocean color physical models exhibit high uncertainty in their local parameterization due to the complex spectral characteristics of the observed scene. Retrieval results are encouraging for Chl-a with a coefficient of determination R2 up to 0.72 with a root-mean-square error (RMSE) of 0.33 mg m-3, using an empirical NN. The TSM algorithms exhibit higher uncertainty, mainly due to scarcity of in situ measurements and model parameterizations, with R2 = 0.52 and RMSE = 1.95 g/m3 using NNs. The bio-optical model, used for the development of model-based algorithms, shows some inadequacies in representing the inherent and apparent optical properties for the case study areas, especially considering the different spectral features between the oligotrophic Tyrrhenian Sea and the eutrophic Adriatic Sea. This study confirms the potential of Sentinel-2 MSI products for coastal water monitoring, but it also highlights key issues to be further tackled such as the atmospheric correction impact, the need of reliable in situ measurements, and possible bathymetry effects near the shores.
Based on the sea surface scattering mechanism of radio wave, compact high-frequency surface wave radar that employs the direction finding technology also shows promising potential for remotely mapping of ocean wave parameters. However, due to the low signal-to-noise ratio (SNR) of scattered echoes, the diverse external interference and clutter signals, other unresolved measurement uncertainties, the quality (such as accuracy, temporal, and spatial coverage rates) of wave maps are often limited. In this article, a novel, real-time, data quality control method is proposed to alleviate these issues. A comprehensive three-stage processing scheme is established, including the range-Doppler spectral processing, the spatial-grid processing, and the temporal-scale Kalman filtering. The first two stages aim to improve the echo signal quality and reduce the spatial gaps, respectively. The third stage is designed to mitigate the estimation error using an autoregression prediction model and to relate the observation error variance to the SNR of second-order Doppler spectral peak. A detailed verification and performance analysis between the field radar data and in situ ground truth data over one-month period is carried out, indicating that the proposed method can improve the reliability of wave maps with respect to the conventional Doppler spectral smoothing or averaging method, particularly in low sea state (i.e., low SNR) scenarios.
Waves much shorter than those measured by operational systems make significant contribution to the ocean surface roughness. This article describes a method to obtain the L-band tilting ocean surface roughness using wind speed and windsea dominant wave period coupled with a wind-wave spectrum model. Examples are presented with wind and dominant wave data from ocean buoys and hurricane hunters. Several related issues are discussed: high-frequency wave spectrum, integration limit, swell contribution, and measurements in extreme winds: 1) it is well known since the 1970s that with stationary sensors, extending the frequency range in measuring elevation spectrum does not yield useful short-wave information because of the low signal level and large Doppler frequency shift involved in measuring short waves. 2) Low-pass mean square slopes (LPMSSs) integrated to 5 and 11 rad/m are computed to quantify their difference as a function of wind speed and inverse wave age (IWA). The normalized difference decreases with increasing wind speed and decreasing IWA. 3) Swell contribution to the L-band LPMSS is almost negligible for wind speed greater than 5 m/s (less than 5% in 99% of observations). In low-wind conditions (wind speed less than 5 m/s), the swell contribution is difficult to assess because of inaccuracy in identifying the weak windsea system. 4) The coarse resolution in National Data Buoy Center (NDBC) wave spectra causes large data scatter in the computed LPMSS in very high winds (greater than 20 m/s). A mitigating solution is offered.
A new technique for automated retrieval of ice concentration from RADARSAT-2 dual-polarization HH-HV ScanSAR Wide images for subsequent assimilation in ice numerical models is presented. First, we extended our previously introduced ice and water detection approach operating at a 2.05 km x 2.05 km spatial scale to a set of 19 different spatial scales ranging from 2.05 km (41 pixels) down to 0.25 km (5 pixels). As the spatial resolution was increased, the overall accuracy of ice and water detection stayed at a very high level across all scales (between 99.5% and 99.8%), but the number of water retrievals substantially dropped. Second, we designed an approach for estimating ice concentration in a 2 km × 2 km (40 × 40 pixels) area consisting of 64 5 × 5 pixel blocks. The 5 × 5 pixel blocks which are initially classified as unknowns are iteratively combined in clusters with effective spatial scales larger than 5 pixels. The clusters are further classified as ice or water using the ice probability model corresponding to the effective spatial scale. The 40 x 40 pixel area becomes populated with high-resolution (5 x 5 pixels) ice and water retrievals, and the ice concentration is estimated based on the number of ice and water retrievals. The proposed approach produces a much better agreement with the Canadian Ice Service Image Analysis ice concentrations (rootmean-square error (RMSE) = 2.2%) compared to the original 2-km ice/water detection approach (RMSE = 19.9%). The developed technique will be adapted to the RADARSAT Constellation Mission data for data assimilation in Environment and Climate Change Canada Regional Ice-Ocean Prediction System.
Process-based ecosystem models are increasingly used to simulate the effects of a changing environment on vegetation growth in the past, present, and future. To improve the simulation, the multimodel ensemble mean (MME) and ensemble Bayesian model averaging (EBMA) methods are often used in optimizing the integration of ecosystem model ensemble. These two methods were compared with four other optimization techniques, including genetic algorithm (GA), particle swarm optimization (PSO), cuckoo search (CS), and interior-point method (IPM), to evaluate their efficiency in this article. Here, we focused on eight commonly used ecosystem models to simulate vegetation growth, represented by the growing season leaf area index (LAIgs), collected globally from 2000 to 2014. The performances of the multimodel ensembles and individual models were compared using the satellite-observed LAI products as the reference. Generally, ensemble simulations provide more accurate estimates than individual models. There were significant performance differences among the six tested methods. The IPM ensemble model simulated LAIgs more accurately than the other tested models, as the reduction in the root-mean-square error was 84.99% higher than the MME results and 61.50% higher than the EBMA results. Thus, IPM optimization can reproduce LAIgs trends accurately for 91.62% of the global vegetated area, which is double the area of the results from MME. Furthermore, the contributions and uncertainties of the individual models in the final simulated IPM LAIgs changes indicated that the best individual model (CABLE) showed the greatest area fraction for the maximum IPM weight (32.49%), especially in the low-lalitude to midlatitude areas.
The pathogenesis of wheat powdery mildew (WPM) is from the bottom to upper layers of the plant, and the vertical observation angle limits the early monitoring of WPM status. Multiangle remote sensing could effectively extract spatial structural information from different plant layers. The objectives of this study were to improve the monitoring accuracy of WPM severity and to screen suitable observation angles by developing a novel vegetation index. The monitoring capacities of the novel parameters [Normalized Powdery Mildew Index (NPMI) and Ratio Powdery Mildew Index (RPMI)] and 14 optimized traditional spectral parameters were compared at 13 observation angles and different angle ranges. The correlation between all spectral parameters and disease severity was superior in the forward observation direction than in the backward observation direction in the principal plane of the Sun, and the correlation between the two observation directions decreased with an increase in observation angle. The new spectral parameter suitable for inversion of disease index was RPMI (R744/R762 - 0.5 × R710), and the best optimal observation angle was +10°, with a 0.852 coefficient, which was 31.74% higher than that of the two-band spectral parameter, R744/R762. The fitting accuracy of the new parameter in the range of 0° to +30° in the forward direction was 0.704. RPMI could not only improve the monitoring accuracy of powdery mildew severity at a single angle but also achieve a more stable monitoring accuracy in the 0° to +30° range in the forward direction, which significantly expands the application scope of remote sensing monitoring and enhances the flexibility of the technology in actual production environments.
Prediction of large-scale water-related natural disasters such as droughts, floods, wildfires, landslides, and dust outbreaks can benefit from the high spatial resolution soil moisture (SM) data of satellite and modeled products because antecedent SM conditions in the topsoil layer govern the partitioning of precipitation into infiltration and runoff. SM data retrieved from Soil Moisture Active Passive (SMAP) have proved to be an effective method of monitoring SM content at different spatial resolutions: 1) radiometer-based product gridded at 36 km; 2) radiometer-only enhanced posting product gridded at 9 km; and 3) SMAP/Sentinel-1A/B products at 3 and 1 km. In this article, we focused on 9-, 3-, and 1-km SM products: three products were validated against in situ data using conventional and triple collocation analysis (TCA) statistics and were then merged with a Noah-Multiparameterization version-3.6 (NoahMP36) land surface model (LSM). An exponential filter and a cumulative density function (CDF) were applied for further evaluation of the three SM products, and the maximize-R method was applied to combine SMAP and NoahMP36 SM data. CDF-matched 9-, 3-, and 1-km SMAP SM data showed reliable performance: R and ubRMSD values of the CDF-matched SMAP products were 0.658, 0.626, and 0.570 and 0.049, 0.053, and 0.055 m3/m3, respectively. When SMAP and NoahMP36 were combined, the R-values for the 9-, 3-, and 1-km SMAP SM data were greatly improved: R-values were 0.825, 0.804, and 0.795, and ubRMSDs were 0.034, 0.036, and 0.037 m3/m3, respectively. These results indicate the potential uses of SMAP/Sentinel data for improving regional-scale SM estimates and for creating further applications of LSMs with improved accuracy.
Agricultural drought is a frequent global phenomenon. Solar-induced chlorophyll fluorescence (SIF) is a by-product of photosynthesis that can be used to monitor vegetation growth and agricultural drought. The global 0.05° spatial resolution data set has been obtained using the data-driven algorithm method. However, the broken farmland is not conducive to regional agricultural drought monitoring. Hence, 0.05° SIF products should be downscaled. On this basis, a convolutional neural network (CNN) downscaled work was conducted in this article to obtain 0.008° spatial resolution SIF results. The downscaled SIF and land surface temperature (LST) data were used to establish the temperature fluorescence dryness index (TFDI). The new TFDI was subsequently used for monitoring agricultural drought in Henan province (China) during the corn-growing season (from June to October 2013-2017). Results showed that the downscaled SIF data exhibit a good correlation with gross primary productivity (GPP) from the Moderate Resolution Imaging Spectroradiometer (MODIS) than 0.05° SIF products. During the study period, the soil moisture fluctuation corresponded well with precipitation, and the value of TFDI had an opposite fluctuation with soil moisture. Meanwhile, the annual averaged TFDI had a high correlation with summer corn yield (R = -0.84). In conclusion, the SIF results through the CNN-based downscaled method were reliable, and the new TFDI was suitable for region agricultural drought monitoring.
Accurate land use/land cover (LULC) mapping over a large area is essential to environmentally sustainable development. Recently, the Chinese government established a new national economic zone called the Xiongan New Area, and along with the upcoming large-scale urban construction, this area will inevitably experience a dramatic LULC change, which will threaten the local ecological balance. In this article, we proposed a two-stage approach for LULC mapping in the Xiongan New Area ahead of the forthcoming dense urban construction. The first stage is to obtain base-class maps through a supervised imagery classification. Specifically, we designed a new object-based framework consisting of automatic image segmentation, pixel-based probabilistic estimation, and area-weighted probability statistics for Sentinel-2 multiresolution imagery classification. The second stage is an LULC map refinement process in which the temporal features of each land use category are extracted to refine the LULC classification. Through the implementation of the proposed two-stage approach, an LULC map containing permanent water, temporal water, natural vegetation, barren land, built-up land, and cropland categories can be produced. Through an accuracy assessment, the proposed multiresolution imagery classification method achieved the highest overall accuracy of 88.58% and an average accuracy (AA) of 87.78% compared with conventional classification methods. After obtaining the refined LULC map, we find that the current Xiongan New Area is in a less developed state, that is, cropland accounts for the highest proportion of 51.59%, which is followed by natural vegetation (22.38%) and built-up land (15.69%).
In this article, an efficient algorithm for the reconstruction of a 1-D random rough surface profile separating two lossy dielectric half-spaces is presented. First, the general scattering problem is formulated by the use of surface integral equations (SIEs). Then, the synthetic scattering field data are obtained through the use of these conventional SIEs. In the inverse problem, the same SIEs together with the data equation are solved in an iterative fashion to reconstruct the surface variation. In the numerical implementation, the so-called ill-posed inverse problem is regularized in the sense of Tikhonov, and a least squares solution is obtained by the use of appropriate basis functions. A very detailed numerical assessment of the presented approach is provided which shows that the method is very effective and promising.
The lunar surface has complex geomorphic characteristics. Since the lunar terrain entropy can reflect the amount of geomorphic information contained in the lunar terrain, this article uses the ratio of the elevation value of a local point on the lunar surface to the total elevation value of the neighborhood to calculate the local terrain entropy value of the Moon. Then, the hierarchical polishing splines algorithm is proposed to construct the digital terrain entropy model (DTEM) of the Moon, wherein the new algorithm produces a sequence of functions based on a hierarchy of coarse-to-fine control lattices to generate the modeling function, which has good modeling performance. Using the proposed algorithm, multiscale DTEMs of the Moon are constructed based on square moving windows with different sizes. From the lunar DTEMs, it can be found that the lunar terrain entropy is sensitive to the size of the square moving window and the resolution of lunar DEM, and the high-resolution lunar DTEM with suitable moving window can well show topographical variations. In addition, the lunar terrain entropy distribution models are created based on the lunar DTEMs, which is significantly important to the study of the lunar terrain entropy distribution law. Besides, two terrain parameters, i.e., surface roughness and surface slope, are selected to show that the geomorphic characteristics of the Moon can be well reflected by the lunar terrain entropy.
The domain adaptation of satellite images has recently gained increasing attention to overcome the limited generalization abilities of machine learning models when segmenting large-scale satellite images. Most of the existing approaches seek for adapting the model from one domain to another. However, such single-source and single-target setting prevents the methods from being scalable solutions since, nowadays, multiple sources and target domains having different data distributions are usually available. Besides, the continuous proliferation of satellite images necessitates the classifiers to adapt to continuously increasing data. We propose a novel approach, coined DAugNet, for unsupervised, multisource, multitarget, and life-long domain adaptation of satellite images. It consists of a classifier and a data augmentor. The data augmentor, which is a shallow network, is able to perform style transfer between multiple satellite images in an unsupervised manner, even when new data are added over time. In each training iteration, it provides the classifier with diversified data, which makes the classifier robust to large data distribution difference between the domains. Our extensive experiments prove that DAugNet significantly better generalizes to new geographic locations than the existing approaches.
Satellite images from the same scene observed over time can be composed in an image stack, which could be modeled as a 3-D cube. To handle this type of remote sensing data, on the one side, unidimensional dynamical models have been considered, modeling each pixel separately along the time (pixel-based approach), and exploring the temporal correlation. On the other side, 2-D approaches have been considered to process each image at one date, exploring the spatial correlation. In this article, we propose a new 3-D autoregressive (AR) (3-D-AR) model useful for multitemporal image interpretation exploring the correlation in three dimensions altogether. The 3-D-AR model is statistically defined, and a robust parameter estimation method is discussed. The tools for filtering, forecasting, and detecting anomalies are also introduced. A Monte Carlo simulation study is performed to evaluate the finite signal length performance of the robust estimation and its sensitivity to outliers. The proposed model is applied to a multitemporal normalized difference vegetation index (NDVI) image stack for filtering, prediction, and anomaly detection purposes. The numerical results show the importance of the proposed 3-D-AR model for spatiotemporal remote sensing data interpretation.
In distributed passive radar, optimization of the receiver placement attracts much attention recently to improve the performance. Whereas most research optimizes receiver placement for best coverage or localization performance, the effect of one performance optimization on the other is ignored. In this article, the theoretical analysis on the conflict between the coverage and localization performance optimization is presented. Then, a joint coverage and localization-based receiver placement optimization is formulated, in which the detection probability and the localization accuracy are chosen as the corresponding metrics. To solve this optimization problem, a multiobjective neighborhood search algorithm with multineighborhood structure is proposed, by which a set of Pareto solutions corresponding to the optimal receiver placements is obtained. Compared with the traditional neighborhood search algorithms, the proposed algorithm is able to provide better receiver placements approaching the global Pareto optimum, which is validated by the simulations.
VIPR (vapor in-cloud profiling radar) is a tunable G-band radar designed for humidity and cloud remote sensing. VIPR uses all-solid-state components and operates in a frequency-modulated continuous-wave (FMCW) radar mode, offering a transmit power of 200–300 mW. Its typical chirp bandwidth of 10 MHz over a center-frequency tuning span of 167–174.8 GHz results in a nominal range resolution of 15 m. The radar’s measured noise figure over the transmit band is between 7.4 and 10.4 dB, depending on its frequency and hardware configuration, and its calculated antenna gain is 58 dB. These parameters mean that with typical 1 ms chirp times, single-pulse cloud reflectivities as low as −26 dBZ are detectable with unity signal-to-noise at 5 km. Experimentally, radar returns from ice clouds above 10 km in height have been observed from the ground. VIPR’s absolute sensitivity was validated using a spherical metal target in the radar antenna’s far-field, and a G-band switch has been implemented in an RF calibration loop for periodic recalibration. The radar achieves high sensitivity with thermal noise limited detection both by virtue of its low-noise RF architecture and by using a quasioptical duplexing method that preserves ultrahigh transmit/receive isolation despite operation in an FMCW mode with a single primary antenna shared by the transmitter and receiver.
Through-the-wall radar imaging is a sensing technology that can be used by first responders to see through obscure barriers during search-and-rescue missions or deployed by law enforcement and military personnel to maintain situational awareness during tactical operations. However, the strong reflections from the front wall and other obstacles render the detection of stationary targets very difficult. In this article, a learning-based approach is proposed to mitigate the effect of the wall and background clutter. A sparse autoencoder with a low-rank projection is developed to mitigate the wall clutter and recover the target signal. The weights of the proposed autoencoder are determined by solving an augmented Lagrange multiplier optimization problem, and the regularization parameters are estimated using the Bayesian optimization technique. Experiments using real data from a stepped-frequency radar were conducted to illustrate its effectiveness for wall clutter removal. The results show that the proposed method achieves superior performance compared with the existing approaches.
Multistatic radar is a promising option for the low-cost collection of multiple-Doppler weather observations. However, due to the use of low-directivity antennas at the receivers in these systems, they typically suffer from extremely high two-way sidelobe levels compared to monostatic radars. Doppler velocity estimation biases induced by sidelobe contamination have proved to be a significant obstacle to more widespread adoption of this technology. It has been noted in the existing literature that the technique of sidelobe whitening, first developed for use in monostatic systems, has the potential to mitigate this issue. However, the existing sidelobe whitening algorithm is not suitable for use in this application, as it is only capable of achieving whitening in the two-way antenna pattern, whereas an algorithm to be used in the multistatic case must be able to achieve this result in the transmit pattern alone. This article proposes an alternate sidelobe whitening technique based on the method of alternating projections which allows for effective whitening in the one-way antenna pattern. A multistatic weather radar time-series simulator is used in conjunction with numerical weather prediction data to demonstrate the effectiveness of this method in mitigating multiple-Doppler measurement biases.
A frequency-modulated continuous-wave (FMCW) radar, operated at the central frequency of 27.75 MHz and the bandwidth of 300 kHz has been established on the seashore near the Taichung harbor ( $24^{circ } 18.591^prime $ N, $120^{circ } 31.389^prime $ E), Taiwan. Sixteen vertical dipole antennas were located linearly and attached with 16 receiving channels. One purpose of the radar is to monitor the ships that navigate toward, away, and around the harbor. In this article, we applied the radar beamforming methods that transform the temporal radar signals as brightness on the 2-D range-azimuthal domain, making the ship echoes visible directly on the spatial domain. Three beamformers, linear Fourier, directionally constrained minimum power (DCMP), and norm-constrained DCMP (NC-DCMP) algorithms, were employed to produce range–angle (RA) brightness distribution that is different from the conventionally used range–Doppler (RD) spectra in ship detection. Both DCMP and NC-DCMP are adaptive beamforming methods. With the auxiliary of a band-stop filter to suppress the sea echoes, the NC-DCMP beamformer was demonstrated to surpass the other two beamformers and could provide more visible ship echoes in the RA brightness distribution. Automatic Identification System (AIS) information was also used to validate the radar-determined ship locations from the RA brightness distribution. Although some ships having the AIS information were not observed clearly by the radar, the radar detected some targets without AIS information.
This article shows how the array of corner reflectors (CRs) in Queensland, Australia, together with highly accurate geodetic synthetic aperture radar (SAR) techniques—also called imaging geodesy—can be used to measure the absolute and relative geometric fidelity of SAR missions. We describe, in detail, the end-to-end methodology and apply it to TerraSAR-X Stripmap (SM) and ScanSAR (SC) data and to Sentinel-1 interferometric wide swath (IW) data. Geometric distortions within images that are caused by commonly used SAR processor approximations are explained, and we show how to correct them during postprocessing. Our results, supported by the analysis of 140 images across the different SAR modes and using the 40 reflectors of the array, confirm our methodology and achieve the limits predicted by theory for both Sentinel-1 and TerraSAR-X. After our corrections, the Sentinel-1 residual errors are 6 cm in range and 26 cm in azimuth, including all error sources. The findings are confirmed by the mutual independent processing carried out at University of Zurich (UZH) and German Aerospace Center (DLR). This represents an improvement of the geolocation accuracy by approximately a factor of four in range and a factor of two in azimuth compared with the standard Sentinel-1 products. The TerraSAR-X results are even better. The achieved geolocation accuracy now approaches that of the global navigation satellite system (GNSS)-based survey of the CRs positions, which highlights the potential of the end-to-end SAR methodology for imaging geodesy.
