Surface Topography and Vegetation (STV) is a NASA targeted observable for maturation into an observing system architecture. STV will acquire high-resolution, global height measurements, including bare surface land topography, ice topography, vegetation structure, and shallow water bathymetry. These measurements serve a broad range of science and applications objectives that span solid earth, cryosphere, biosphere and hydrosphere disciplines. A common set of measurements could meet many of the community needs. STV objectives would be best met by new observing strategies that employ flexible multi-source and sensor measurements from a variety of orbital and sub-orbital assets. Science and application objectives would be best met by new, 3-dimensional observations from lidar, radar, and stereoimaging. Simulations, experiments, data analysis and technology development in interferometric SAR, lidar and stereo photogrammetry approaches, platform options and system architectures will all mature STV toward an observing system.

Monitoring forest ecosystems requires both, continuous time-series datasets and a comprehensive method for analyzing the same. Synthetic Aperture RADAR (SAR) data aids the processing by maintaining an uninterrupted time-series dataset. The backscatter derived from SAR data is closely related to the changes taking place in the targeted area. Therefore the changes and disturbances taking place in the forests results in corresponding variations in the backscatter values. These temporal signatures of backscatter can be used for monitoring forests with statistical change detection algorithms. In this study, Cumulative Sums of Change (CuSum) and Sequential Omnibus algorithms are used with Sentinel-1 SAR data for deforestation mapping in Central Kalimantan, Indonesia. CuSum showed an overall better performance than the sequential omnibus algorithm. It was observed that the drop in backscatter on account of deforestation was significant in case of both polarization channels. However, cross-pol channel showed better results with an average overall accuracy of 0.74.

Several studies have found that the vertical Fourier transform of lidar, interferometric Synthetic Aperture Radar (SAR), and stereo photogrammetric profiles at empirically-determined spatial frequencies enables high-performance forest aboveground biomass (AGB) estimation. Linear combinations of real and imaginary parts of Fourier transforms of Tomographic (multi-baseline) SAR (TomoSAR) profiles, from Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) airborne data, generate ~20%-precision estimates of AGB in the Saskatchewan area of Canada. We found that this 20% precision can be improved to ~15%, a factor of 30% improvement in root mean square error (RMSE) if, in addition to using Fourier transforms of the profile itself, we use Fourier transforms of the spatial, vertical derivative of the profile. The formulation of this "derivative" algorithm is the subject of this paper.

The ongoing Distributed Aperture Radar Tomographic Sensors (DARTS) project at NASA Jet Propulsion Laboratory aims to mature and demonstrate multi-static SAR measurements for fine-scale 3D imaging of surface topography, vegetation, and surface deformation and change. The project explores the use of drones as SAR platforms and integrates software-defined radar on RF system-on-chip for compact and flexible radar instruments. This paper highlights the progress in DARTS hardware development, experiments, and data processing. The recent experiments have successfully demonstrated monostatic interferometry as well as acquisition and processing of bi-static SAR imagery. By leveraging the advantages of multi-static SAR and drone-based platforms, the project aims to build a testbed for future missions design and enhanced SAR imaging capabilities for scientific applications.

This letter describes a model-based algorithm for estimating tree height and other bio-physical land parameters from time series of synthetic aperture radar (SAR) interferometric coherence and backscatter supported by sparse lidar data. The random-motion-over-ground model (RMoG) is extended to time series and revisited to capture the short- and long-term temporal coherence variability caused by motion of the scatterers and changes in the soil and canopy backscatter. The proposed retrieval algorithm estimates first the spatially slow-varying RMoG model parameters using sparse lidar data, and subsequently the spatially fast-varying model parameters such as tree height. The recently published global Sentinel-1 (S-1) interferometric coherence and backscatter data set and sparse spaceborne GEDI lidar data are used to illustrate the algorithm. Results obtained for a small region over Spain show that the temporal coherence and backscatter time series have the potential to be used for global, model-based land parameter estimation.

Recent results have highlighted the potential ability of bistatic and multistatic synthetic aperture radar (SAR) tomographers to measure vegetation structure and surface topography. However, the quality of SAR tomographic measurements with multiple platforms is impacted by the phase instability in each platform’s oscillator. The phase noise, if uncompensated, may lead to degradation in the SAR data products such as increased sidelobe levels, reduced peak amplitude of the impulse response, and low-frequency phase modulation, among others. In this work, we model and examine the effects of oscillator phase noise on tomographic SAR signals for spaceborne missions flying in formation. A synchronization process is also adopted to help mitigate oscillator phase errors by measuring and predicting relative phase offsets at prescribed temporal intervals. A simulation tool was developed to examine the point target response (PTR) as seen by realistic satellite constellations in low Earth orbit using different quality oscillators, radar configurations, and synchronization configurations. A first analysis of a multiplatform tomographic SAR mission suggests that a system without a dedicated physical link with minimal effects on the PTR may be achievable using current oscillators. Our analysis also shows that phase noise has differing effects on multistatic radar modes. Tomograms formed with a system operating in single-input–multiple-output (SIMO) mode are the most affected by an oscillator phase noise error, followed by multiple-input–multiple-output (MIMO), with negligible effects on the single-input single-output (SAR-SISO) mode. These trade studies and the simulation tool can be used to help inform the design of future multistatic radar missions.

When GNSS signals reflect off of the surfaces of lakes, rivers, wetlands, and other inland water bodies, the surfaces are often sufficiently smooth to produce coherent reflections. The observable produced from coherent reflections made by GNSS Reflectometry (GNSS-R) instruments exhibits particular features with respect to diffusely scattered signals by rough land and wind-driven oceans allowing detection of such smooth bodies. Several different GNSS-R coherence detection approaches have been reported in the literature and developed among the GNSS-R community over the last several years; however, the merits of each approach are difficult to compare because they are often applied to different scenarios and quantified in different ways, independently of each other. This paper provides a unified comparison of a wide variety of different GNSS-R coherence detection approaches, which is the most extensive published to date. The approaches are applied to a common data set from the NASA CYGNSS satellites that includes both the standard Level-1 DDM science product as well as raw baseband signal recordings. Additionally, simulated observables are generated with varying coherent and non-coherent reflection components to exercise algorithms over a wide range of SNRs and relative powers. Objective measures of accuracy are used to quantify the performance of each approach in the context of relative implementation complexity. Conclusions are presented on the pros/cons of the various methods as they relate to various applications such as real-time in-orbit coherence detection or post-processing on the ground.

In this paper, we assess the estimation of bio-physical land parameters from time-series of interferometric SAR coherence supported by a physical model. The random-motion-over-ground model (RMoG) is revisited to partially capture the short- and long-term temporal variability of the coherence caused by motion of the scatterers and changes in their dielectric properties. The recently-published global Sentinel-1 interferometric coherence dataset is used to compare model predictions with observations and evaluate the need for additional model assumptions or ancillary data sets. Space-borne lidar data acquired by GEDI are also considered to further constrain the parameter estimation. This work is particularly relevant to upcoming SAR missions such as NISAR and ROSE-L that will generate global and dense time-series of interferometric temporal coherence at L-band.

Recent innovations in small satellite, ultra-wideband direct RF sampling, and synchronization technologies have made multistatic and MIMO coherent SAR constellations a feasible concept for future missions. The Distributed Aperture Radar Tomographic Sensors (DARTS) mission concept at NASA JPL aims to measure Earth's surface topography and vege-tation using TomoSAR techniques. This paper describes the development of an embedded ultra-wideband next generation software defined radar (SDRadar) testbed capable of multi-band operation implemented with the Xilinx RF System on Chip (RFSoC) architecture, which features 8x 6.4 GSPS DACs and 8x 4 GSPS ADCs. The RFSoC SDRadar repre-sents a state of the art testbed for rapid prototyping of radio, radar, and synchronization technologies. We provide preliminary testing results for airborne monostatic radar imaging from a small uninhabited aerial system (sUAS), successfully demonstrating multi-band operation using first and second Nyquist zone direct RF sampling.

A novel metric for detecting coherence in global navigation satellite system reflectometry (GNSS-R) signals is presented and evaluated. It applies the Von Neumann information entropy metric for density matrices, a powerful indicator of the degree of mixing between states, coherent and incoherent, of the scene under investigation. The metric is applied to a set of raw IF data acquired by the cyclone global navigation satellite system (CYGNSS) observatories over Lake Okeechobee FL, in order to test the sensitivity of the entropy to different land cover types, including wetlands and open water. Visual comparison of results with Sentinel-1 images provides a first step in the validation of the effectiveness of entropy in detecting the presence of water covered by emergent vegetation. In addition, the entropy-based metric could be implemented on future space-based GNSS-R receivers to adapt the incoherent integration times to the observed scene, thus achieving an improvement in along-track resolution.

This article describes a projection algorithm between radar and map coordinates based on the representation of radar samples as area elements (AEs) rather than point elements. Each AE on the map grid (geographic grid) is associated with a number of radar grid samples that intersect completely or partially the AE. The association enables the geocoding (i.e., the map projection of radar imagery) with adaptive multilooking, accurately accounting for all radar samples contributing to the geocoded elements according to topography and radar geometry. By using averaging rather than interpolation, the proposed projection does not suffer from interpolation overfitting. The area-based geocoding also enables the generation of the geocoded polarimetric covariance matrix (GCOV) and geocoded synthetic aperture radar (SAR) interferograms with adaptive multilooking. Analogously, the slant-range projection of geocoded data is improved by projecting geographic grid pixels onto the radar grid according to their corresponding location based on the radar geometry without leaving gaps. This approach is used to reduce the computation time of previously published radiometric terrain correction (RTC) algorithms, performing 3.6–6.5 times faster over multilooked data and up to 26.3 times faster over single-look data. We demonstrate the strengths of the proposed area projection (AP) algorithm for RTC and geocoding using Uninhabited Aerial Vehicle SAR (UAVSAR), Sentinel-1B, and ALOS-2/PALSAR-2 data, and evaluate the results in the context of the upcoming NASA-ISRO SAR (NISAR) mission.

In this work, the sensitivity to soil moisture of different scattering mechanisms observed by an airborne polarimetric radar operating at L-band has been investigated. The main objective was assessing the capability of a fully polarimetric radar system to disentangle the change in soil moisture under different vegetation covers. We used polarimetric data collected by the NASA UA VSAR airborne radar flying over the Yucatan Lake site in Louisiana. Six overflights were analysed, and six regions of interest characterized by different vegetation covers were selected. The temporal trends of the magnitude of different scattering mechanisms, according to the Freeman-Durden and the Nonnegative Eigenvalue decompositions, as well as of the NDVI and a nearby precipitation gauge are analysed and discussed with reference to the theoretical expectations.

The sensitivity of remote sensing instruments for measuring inundation extent can vary widely. Many sensors are suitable for accurate delineation of open water extent, but in vegetated environments the vegetation canopy can obscure the presence of standing water from detection. Detecting inundation extent in these vegetated environments is especially critical for identifying flooding extent where excess surface water extends into the forests surrounding lakes and streams. In addition, cloud cover can impede timely acquisition of imagery by optical sensors. Here, we examine sensitivity of L-band Global Navigation Satellite Systems Reflectometry (GNSS-R) to flooded conditions relative to the well-known signatures of inundation by L-band SAR, and confirm that there is noticeable sensitivity of GNSS reflected signal to inundated areas, including wetlands covered by vegetation, captured by the strong response of the specular reflection by the underlying water surface.

The advent of smaller SAR satellites and cheaper access to space is bringing the notion of a multistatic SAR constellation into the realm of feasibility. Researchers at JPL are studying a Distributed Aperture Radar Tomographic Sensors (DARTS) mission concept intended to measure Earth's surface topography and vegetation using TomoSAR techniques. This paper describes progress on the airborne testbed for the DARTS study. The testbed is the union of a software-defined radio that implements a radar and synchronization link together with a small uninhabited aerial system (sUAS) that serves as a platform with precise control of the observation geometry. Initial experiments have demonstrated successful multi-sensor synchronization as well as acquisition and processing of monostatic SAR imagery.

Distributed Aperture Radar Tomographic Sensors (DARTS) is a mission concept being studied at the NASA Jet Propulsion Laboratory in collaboration with the California Institute of Technology to enable global and repeated imaging of surface topography and three-dimensional vegetation structure using single-pass tomographic SAR technique. The observing system consists of a distributed formation of multiple small synthetic aperture radar platforms deployed in space with variable distances to achieve look angle diversity and sensitivity to the vertical distribution of vegetation components. Our goal is to identify the optimal system configuration starting from documented community needs and mature the critical technologies that lead to a viable implementation of DARTS. Here, we provide an overview of DARTS and describe our approach for designing and demonstrating single-pass SAR tomographic systems as part of an on-going funded NASA Instrument Incubator Program effort.

We present analytic expressions for the clock phase error power spectral density (PSD) resulting from a previously reported decentralized distributed wireless synchronization protocol acting on independent sensor oscillators. We provide an overview of oscillator phase noise error modelling and examine the effects of the wireless synchronization protocol and the resulting synchronized clock phase noise PSDs. We present results from both simulation and experiment to validate the expressions derived.

GNSS Reflectometry (GNSS-R) measurements are very sensitive to the presence of inland waters such as wetlands, floods, rivers and lakes. This paper reviews the basic characteristics of a GNSS-R ‘water detection’ research product, including resolution and temporal sampling of wetlands, and discusses the main known sources of errors. Additionally, a summary of GNSS-R applicability to the study of lakes is provided.

SAR tomographic methods have proven extremely adept at measuring vegetation vertical structure at a variety of wavelengths including L and P-bands. The three dimensional structure of vegetation and its changes resulting from either natural or anthropogenic causes are key parameters in monitoring ecosystems. The NASA/JPL UAVSAR collected data at L and P-bands at Delta Junction, Alaska in September of 2017 whereas the NASA/JPL UAVSAR and DLR F-SAR acquired data at the BERMS site near Saskatoon, Canada on August 19 and 23 of 2018 respectively. Tomographic data sets were collected at L-band and P-band by the NASA/JPL UAVSAR at Delta Junction and at L-band at BERMS and DLR F-SAR acquired data at L-band and S-band. Ground truth data sets and lidar data from the NASA LVIS system were also acquired at BERMS. We compare L and P tomography at Delta Junction and L-band and S-band tomography from the two systems to each other and to the lidar data sets at BERMS. These data are then used to estimate biomass and assess spatial gradients in the canopy vertical structure. We also compare our data with simulated boreal forest data to assess the sensitivity to the data collection geometry and canopy parameters.

The forthcoming launch of the NASA-ISRO Synthetic Aperture Radar (NISAR) mission will open the path to a new type of L-band measurements constituted by dense time-series of polarimetric backscatter and interferometric coherence with unprecedented spatial and temporal sampling. Here, we start the development of a theoretical framework that links L-band backscatter time-series with interferometric coherence measurements. The water-cloud-model (WCM) and an extended version of the random-motion-over-ground (RMoG) model are adopted to express radar measurements in terms of forest above-ground biomass and tree height. Time-series data collected during the 2019 UAVSAR AM-PM campaign in Southeastern United States are used to evaluate the correlation of various PolSAR- and InSAR-derived parameters with field-measured above-ground biomass.

The enduring activity of CYGNSS and TDS-1 observatories has provided clear evidence that ground reflected GNSS signal can be exploited for mapping of geophysical quantities in climate and global monitoring applications. One main challenge is to find out useful variables for determining the level of coherence of the scattered wave in the presence of complex natural landscapes such as water basins, river floods, mixed ice and water. In this study, we approach the concept of wave coherence by analyzing the persistence of signal energy across principal directions, determined via generalized eigenvalue decomposition of the DDM delay waveform correlation matrix. An example shows that the dimensional spread of the eigenvalues is finely sensitive to coherence of the ground reflected wave.

This article presents a projection algorithm based on the representation of radar samples as area elements, rather than point elements as traditionally done in previous works. Each area element in the geographic grid (geogrid) is associated with a set of samples in the radar grid that intersect completely or partially the area element according to the topography and the radar geometry. Accurate geocoding with adaptive multi-looking is achieved by successively assigning the weighted average of the radar samples to the corresponding geogrid elements. Analogously, the slant-range projection of geocoded data is improved by projecting the geogrid pixels onto the radar grid according to their projected area. When our slant-range projection approach is used within previously-published radiometric terrain correction (RTC) algorithms, the processing time is significantly reduced, performing 4.2 to 6.5 times faster over multi-looked data and up to 16.7 over single-look data. We demonstrate the strength of the area projection algorithm for RTC and geocoding using UAVSAR and Sentinel-1 data, and evaluate the results in the context of the upcoming NISAR mission.

Due to the proliferation of GPS-enabled devices, vast amounts of trajectory datasets are being collected every day. Analyzing this data efficiently and at scale is a major challenge. Several different types of spatio-temporal queries are used to analyze these datasets. One important query is the distance query on trajectory data which, given a query distance D, a point P and a time span T, finds all trajectories within D of P during T. This query is frequently used in traffic analysis and numerous other applications.

This article demonstrates how 3-D vegetation structure can be approximated by interferometric synthetic aperture radar (InSAR) backscatter-height histograms. Single-look backscatter measurements are plotted against the InSAR phase height and are aggregated spatially over a forest patch to form a 3-D histogram, referred to as InSAR backscatter-height histogram or simply InSAR histogram. InSAR histograms resemble LiDAR waveforms, suggesting that existing algorithms used to retrieve canopy height and ground topography from radar tomograms or LiDAR waveforms can be applied to InSAR histograms. Three algorithms are evaluated to generate maps of digital terrain, surface, and canopy height models: Gaussian decomposition, quantile, and backscatter threshold. Full-polarimetric L-band uninhabited aerial vehicle synthetic aperture radar (UAVSAR) data collected over the Gabonese Lopé National Park during the 2016 AfriSAR campaign are used to illustrate and compare the performance of the algorithms for the HH, HV, VV, HH+VV, and HH-VV polarimetric channels. Results show that radar-derived maps using the InSAR histograms differ by 4 m (top-canopy), 5 m (terrain), and 6 m (forest height) in terms of average root-mean-square errors (RMSEs) from standard maps derived from full-waveform laser, vegetation, and ice sensor (LVIS) LiDAR measurements.

We report observations of the three-dimensional polarimetric covariance matrix derived from interferometric synthetic aperture radar (InSAR) histograms. Single-look covariance samples plotted versus InSAR phase are aggregated spatially over a forest patch to form a three-dimensional histogram. To first order, InSAR backscatter histograms resemble lidar waveforms, suggesting that this type of measurement can be used as a proxy for horizontal and vertical tree structure. In this paper, we extend the formulation of the InSAR backscatter histograms to the covariance matrix and discuss possible new applications. Full-polarimetric L-band and P-band UAVSAR (Uninhabited Aerial Vehicle Synthetic Aperture Radar) data collected over Delta Junction in Alaska during the Arctic-Boreal Vulnerability Experiment (ABoVE) are used to illustrate the results. Validation is conducted with the full-waveform LVIS (Laser, Vegetation and Ice Sensor) lidar instrument. Important parameters such as the alpha and the entropy will be extracted along the tree vertical direction and illustrated in the final paper.

Spaceborne GNSS Reflectometry (GNSS-R) measurements have shown strong coherent scattering over inland waters. It has been recognized that GNSS-R could be utilized for monitoring the global surface water distribution by making dynamic maps of wetlands as well as rapid response to flood events. Using the strength of the reflected signals, one can make maps that reveal the presence of water over land. In this paper, we used simulations to analyze the accuracy of these maps. The CYGNSS End-to-end Simulator (E2ES) was extended to include coherent scattering in the heterogeneous scenes where the region around the specular point is composed of both land and water in complex geometries. The simulation is then used to evaluate the accuracy of a simple fractional water in footprint approach to mapping wetland extent. We find that scattering from outside the first Fresnel zone and CYGNSS measurement processing effects significantly impact the accuracy of this approach. However, the accuracy can be improved by combining multiple measurements into a gridded map.

We employ dual-polarimetric P-band interferometric synthetic aperture radar (InSAR) backscatter-phase histograms of the Amazon rainforest of Urucu to estimate the elevation of the forest terrain. High-resolution single-look phase and phase-height histograms weighted by backscatter intensity are projected in the elevation direction and are combined in the range and azimuth directions to form a three-dimensional backscatter-phase histogram. Profiles of the resulting histograms resemble forest reflectivity. Higher ground backscatter is observed in the P-HH compared to the P-HV profile, in agreement with scattering models. Results suggest that the peak of the difference between the normalized P-HH and P-HV profiles can be used as an estimator for the terrain elevation from a model-free approach. We qualitatively compare the results to traditional P-band interferometry and polarimet-ric InSAR (PolInSAR) random-volume over ground (RVoG) three-stage approach over five different baselines 20m, 65m, 80m, 100m, and 145m.

SAR tomographic methods have proven extremely adept at measuring vegetation vertical structure at a variety of wavelengths including L and P-bands [2]. Measuring the three dimensional structure of vegetation and its changes resulting from either natural or anthropogenic causes are key parameters in monitoring ecosystems. The NASA/JPL UAVSAR system has deployed to multiple sites including Alaska over the last several years to conduct tomographic SAR observations at L-band and P-band. This talk will provide a brief overview of a tomographic SAR experiment conducted in the boreal forests of Alaska in August and September of 2017 at both L and P-bands. This site consists mostly of relatively short vegetation with mean height less than 20 m and maximal height less than 25 m. It is sparse compared with previous temperate and tropical forest tomographic observations made by UAVSAR. These observations provides a unique data set to compare tomographic data at L and P-bands for this type of biome.

The objective of this paper is to model and study the sensitivity of bistatic microwave scattering versus changes in wetland characteristics as observed by a GNSS-R satellite system such as CYGNSS. We develop a simplified scattering model starting from the Water Cloud Model traditionally used in monostatic radar problems. Vegetation is idealized as a cloud of randomly oriented scattering elements over a rough surface representing either soil or water. The bistatic scattering coefficient is modeled as the incoherent sum of soil, water and vegetation scattering weighted by the fraction of each contribution within the CYGNSS footprint. The model is tested against CYGNSS observations across the Everglades National Park for which high-resolution land-cover and water depth maps are available. We show that our simplified model is able to capture to first order the variability of bistatic scattering versus changes in water depth and water fraction. This effort is a step forward towards the development of an effective algorithm to map the dynamic state of tropical wetlands and other regions subject to flooding using CYGNSS measurements.

We study the temporal variability of soil and vegetation backscatter at L-band using a dense airborne time-series. Backscatter is assumed to change over time due to diurnal variations in soil and canopy water content as well as precipitations. A two-layer SAR backscattere model traditionally used for above-ground biomass retrieval is augmented here with the time dimension in order to guide the data analysis. The model is informed by examining a 5-year time-series of 32 L-band polarimetric UAVSAR images acquired over a vegetated area near the Sacramento Delta in California. Our initial results reported in this paper show that the temporal variability of soil and vegetation backscatter in absence of precipitation events fits well a lognormal probability distribution with mean and standard deviation related to each other. Characterizing the diurnal, seasonal and interannual variability of L-band backscatter may be critical for the successful estimation of ecosystem variables from dense time-series to be acquired globally by the NISAR mission.

This paper investigates the benefits of integrating polarimetric radar variables with LiDAR samples using Support Vector Machine (SVM) to estimate forest canopy height. Multiple polarimetric variables are required as an input to ensure consistent height retrieval performance across a broad range of forest heights. We train the SVM with LiDAR samples and different polarimetric variables based on 5000 samples (less than 1% of the full subset) collected across the images using a stratified random sampling approach. The trained SVM was applied to the rest of the image using the same variables but excluding the LiDAR samples. The estimated height was cross validated versus LiDAR-derived height (RH100) yielding overall good accuracy with r2=0.86 and RMSE = 6.8 m.

The InSAR Scientific Computing Environment (ISCE) was first developed under the NASA Advanced Information Systems Technology as a flexible, extensible object-oriented framework for Interferometric Synthetic Aperture Radar (InSAR) processing. The ISCE framework uses Python 3 at the workflow level, controlling modules of compiled code for functional processing, and managing inputs, outputs, and other flow control services. The currently released version, called ISCE 2.1, is distributed to the research community through the Western North America InSAR Consortium under a research license. The ISCE team is working on the next generation of the code in order to prepare for the NASA-ISRO SAR (NISAR) mission operational processing. Innovations in this code include augmentation or conversion of the custom Python framework elements in ISCE with the Pyre framework, new workflows for interferometric and polarimetric stack processing, a more intuitive and graphically based user interface, and flow control for hybrid computing environments including CPU/GPU clusters, logging and error tracking facilities, and new more efficient computational modules that exploit graphical processor units (GPUs) when available. The ISCE 3.0 framework is designed to work in an operational environment as well as on a single user's laptop or compute cluster, with services to discover capabilities and scale computations accordingly.

InSAR coherence is a promising parameter for land-cover classification and mapping. The ESA SEOM SInCohMap project is devised to test and analyze multi-temporal InSAR coherence potentialities exploiting dense multitemporal data from the Sentinel-1 constellation. In the framework of the project, this paper shows the first classification results using machine learning algorithms over a two-year period of InSAR coherence data. The evaluation is performed on the test site of Doñana (Seville, Southwestern Spain), mainly an agricultural area where different land covers can be identified. Classification results exploiting InSAR coherence shows accuracies around 80 % for this site.

This paper presents a new damage-mapping algorithm based on coherence images estimated from multitemporal polarimetric-interferometric synthetic aperture radar (SAR) data. The interferometric coherence has been restricted in the conventional damage-mapping approaches because the decorrelation sources are too complicated to interpret accurately and temporal decorrelation effects caused by slowly occurring natural changes and disaster events are often coupled together. To overcome these limitations, we formulate a coherence model that accounts for temporal decorrelation in two simplified layers, ground and volume layers, for long-temporal repeat-pass scenarios with zero spatial baseline. The model parameters include: 1) ground-to-volume ratio, a factor to determine the relative scattering contribution of ground and volume layers; 2) temporally correlated change, which captures the exponentially decaying behavior of coherence with time; and 3) temporally uncorrelated change, which is associated with random temporal changes. We estimate the model parameters in three steps: coherence optimization, interferometric pair-invariant parameter estimation, and interferometric pair-variant parameter estimation. To isolate the effects of disaster events from background natural changes, we calculate the probability density functions of historical change pixel by pixel and produce a probability map of damage. We tested the algorithm with uninhabited aerial vehicle data acquired from 2009 to 2015 for mapping the area damaged by the 2015 Lake Fire in California. Based on performance evaluation using receiver operating characteristic curves for optimized coherences and averaged probability maps, the proposed method reduced the false alarm from 0.25 to 0.07 when the probability of detection was 0.85 compared to coherence products alone.

Background The AfriSAR campaign was a joint NASA and European Space Agency airborne campaign conducted in Gabon in support of the upcoming ESA BIOMASS, NASA-ISRO Synthetic Aperture Radar (NISAR) and NASA Global Ecosystem Dynamics Initiative (GEDI) missions. The aim of the campaign was to collect ground, airborne SAR and airborne Lidar data for the development and evaluation of forest structure and biomass retrieval algorithms. The campaign consisted of two deployments, the first in 2015 with the ONERA SETHI SAR system and the second in 2016 with the NASA LVIS (Land Vegetation and Ice Sensor) Lidar, the NASA L-band UAVSAR and the DLR F-SAR. In addition, field teams from the Gabon ANPN (Agence Nationale des Parcs Nationaux), University College London and NASA were collecting ground data. Here we focus on the 2016 NASA contributions to campaign.

This study aims to evaluate the potential of coherent change detection using multi-temporal polarimetric interferometric SAR data. One of the limitations in damage area extraction is that the decorrelation caused by the disaster is commonly coupled with the natural changes. Also, the interpretation of the coherence is troublesome and requires the coherence model. The approach used in this study is based on the modified coherence model for a case of long-temporal and zero-spatial baseline. The inversion processes yield the temporal decorrelation contributions of ground and volume, respectively, having the physical meaning. We additionally applied simple statistical probability estimation method for the damage area to isolate the contributions of disaster from the natural changes. In this study, we used UAVSAR data acquired over Lake fire which occurred in June, 2015.

We present a new algorithm for detecting forest disturbances from a pair of dual-polarimetric synthetic aperture radar (SAR) data. The algorithm uses the mean dual-polarimetric alpha angle in conjunction with its probability distribution to isolate forest structural changes from statistical noise fluctuations. In contrast to radar backscatter, the dual-polarimetric alpha angle is estimated from the complex 2-by-2 covariance matrix and it is less sensitive to variations in soil moisture and terrain slope. Here we derive the statistical properties of the alpha angle for distribute targets, and show how these properties can be applied to detect the burned area of the 2009 Station Fire (CA) from L-band ALOS-1 data. The proposed algorithm can be applicable to future NASA-ISRO SAR (NISAR) time-series to achieve automated global mapping of forest disturbances.

We present our progress results of SAR tomographic imaging using L-band NASA/JPL UAVSAR data collected in Gabon during the 2016 AfriSAR campaign. Several tomographic experiments were conducted in February 2016 over four different sites with a broad diversity of vegetation types, soil characteristics and weather conditions. Here we describe the campaign objectives and report on the status of the UAVSAR tomographic processor for retrieving the 3D structure of forests. We discuss several algorithms, including stack formation, phase calibration and structure retrieval. The availability of NASA/GSFC LVIS waveforms enables cross-comparison of the radar-derived structure with the lidar-derived structure. Results are reported for the Lopé National Park and demonstrate the maturity of the 3D UAVSAR tomographic processing for ecosystem science and applications.

We improved the repeat-pass InSAR processing capability for the L-band UAVSAR airborne synthetic aperture radar in order to support time-series analysis of repeat zero-baseline observations as well as multiple baseline observations for TomoSAR imaging. This new capability enabled us to conduct tomographic experiments in Gabon during the AfriSAR deployment in support of vegetation structure studies. For the GLISTIN-A Ka-band radar, we streamlined the radar operations and implemented a robust production processor that will routinely generate topographic data products in order to support large-scale science campaigns. The new capabilities were put to test in support of the Oceans Melting Glacier Greenland campaign in March 2016.

The NASA/JPL UAVSAR system was deployed to Europe in the May-June 2015 to collect data in support of experiments in Iceland, Norway and Germany. The deployment in Germany was focused on PolInSAR and tomographic data collections at the Traunstein Forest and in the Munich urban area. In addition data were collected at Kaufbeuren, the DLR calibration site, where several surveyed corner reflectors were available for imaging. We describe the experiment design, data collections and present some preliminary results from these experiments.

Detection of changes induced by major events such as earthquakes, flooding and volcanic eruptions from interferometric SAR data is difficult due to the coupled effects with temporal decorrelation caused by natural phenomena such as rain, snow, wind and seasonal changes. In this study, we aim to separate the decorrelation caused by natural changes from the one caused by the major event by analyzing the coherence behavior using a temporal decorrelation model. We formulated the temporal decorrelation model that accounts for the random motion and dielectric changes. By applying the model into the multi-temporal coherence before the event, we extracted the temporal decorrelation components induced by natural phenomena. Based on the extracted parameters, their decorrelation probabilities related to natural changes were estimated in canopy and ground. The model parameters are also extracted from the interferometric SAR data acquired across the event. We compared probabilities between the natural phenomena and the certain event in order to assign the changed regions. Pixels with cumulative probabilities greater than 80% are selected as changed due to the event. A case study for detecting volcanic ash during the eruption of the Shinmoedake volcano in January 2011 was carried out using L-band Advanced Land Observation Satellite (ALOS) PALSAR data.

PLANT (Polarimetric-interferometric Lab and Analysis Tools) is a new collection of software tools developed at the Jet Propulsion Laboratory to support processing and analysis of Synthetic Aperture Radar (SAR) data for ecosystem and land-cover/land-use change science and applications. PLANT inherits code components from the Interferometric Scientific Computing Environment (ISCE) to generate high-resolution, coregistered polarimetric-interferometric SLC stacks from Level-0/1 data for a variety of airborne and spaceborne sensors. The goal is to provide the ecosystem and land-cover/land-use change communities with rigorous and efficient tools to perform multi-temporal, polarimetric and tomographic analyses in order to generate calibrated, geocoded and mosaicked Level-2 and Level-3 products (e.g., maps of above-ground biomass and forest disturbance). In this paper we introduce the capabilities of PLANT and report first results obtained with the tools developed up to date.

Detection of changes caused by major events-such as earthquakes, volcanic eruptions, and floods-from interferometric synthetic aperture radar (SAR) data is challenging because of the coupled effects with temporal decorrelation caused by natural phenomena, including rain, snow, wind, and seasonal changes. The coupled effect of major events and natural phenomena sometimes leads to misinterpretation of interferometric coherence maps and often degrades the performance of change detection algorithms. To differentiate decorrelation sources caused by natural changes from those caused by an event of interest, we formulated a temporal decorrelation model that accounts for the random motion of canopy elements, temporally correlated dielectric changes, and temporally uncorrelated dielectric changes of canopy and ground. The model parameters are extracted from the interferometric pairs associated with natural changes in canopy and ground using the proposed temporal decorrelation model. In addition, the cumulative distribution functions of the temporally uncorrelated model parameters, which are associated with natural changes in canopy and ground, are estimated from interferometric pairs acquired before the event. Model parameters are also extracted from interferometric SAR data acquired across the event and compared with the cumulative probabilities of natural changes in order to calculate the probability of a major event. Subsequently, pixels with cumulative probabilities greater than 75% are marked as changed due to the event. A case study for detecting volcanic ash during the eruption of the Shinmoedake volcano in January 2011 was carried out using L-band Advanced Land Observation Satellite PALSAR data.

This paper addresses the important yet unresolved problem of estimating forest properties from polarimetric-interferometric radar images affected by temporal decorrelation. We approach the problem by formulating a physical model of the polarimetric-interferometric coherence that incorporates both volumetric and temporal decorrelation effects. The model is termed random-motion-over-ground (RMoG) model, as it combines the random-volume-over-ground (RVoG) model with a Gaussian-statistic motion model of the canopy elements. Key features of the RMoG model are: 1) temporal decorrelation depends on the vertical structure of forests; 2) volumetric and temporal coherences are not separable as simple multiplicative factors; and 3) temporal decorrelation is complex-valued and changes with wave polarization. This third feature is particularly important as it allows compensating for unknown levels of temporal decorrelation using multiple polarimetric channels. To estimate model parameters such as tree height and canopy motion, we propose an algorithm that minimizes the least square distance between model predictions and complex coherence observations. The algorithm was applied to L-band NASA's Uninhabited Aerial Vehicle Synthetic Aperture Radar data acquired over the Harvard Forest (Massachussetts, USA). We found that the RMS difference at stand level between estimated RMoG-model tree height and NASA's lidar Laser Vegetation and Ice Sensor tree height was within 12% of the lidar-derived height, which improved significantly the RMS difference of 37% obtained using the RVoG model and ignoring temporal decorrelation. This result contributes to our ability of estimating forest biomass using in-orbit and forthcoming polarimetric-interferometric radar missions.

In this letter we propose a three-baseline approach to the extraction of forest tree height from synthetic aperture radar data. Three polarimetric-interferometric pairs are used to constrain a physical model that relates forest parameters to multiple repeat-pass coherence observations. The observations may be performed by a dual, compact or full polarimetric radar, and may be affected by distinct levels of temporal decorrelation. Here, we present the theoretical framework based on the random-motion-over-ground model, and describe an algorithm to extract tree height from the data. The performance of the algorithm is illustrated with L-band airborne data collected by the German Aerospace Center in the frame of the BIOSAR2008 campaign. The proposed method provides height estimates in good agreement with lidar measurements and can be applied to data to be collected by forthcoming polarimetric-interferometric spaceborne missions.

Quantification of the various components of the carbon cycle budget is key to improved climate modeling and projecting anthropogenic affects on climate in the future. Estimating the levels of above ground biomass contained in the world's forests that comprise 86% of the planet's above ground carbon and monitoring the rate of change to these standing stocks resulting from both natural and anthropogenic disturbances is necessary to solving the carbon cycle sink. Remote sensing is the only viable means of obtaining a global inventory of forest biomass at the hectare scale. The most promising means of obtaining remotely sensed biomass measurements involve using either lidar or radar measurements of vegetation structure coupled with allometric relationships. We have collected repeat-pass L-band fully polarimetric radar data at multiple spatial and temporal baselines to investigate the tree height and structure measurements using polarimetric interferometry techniques. This paper will discuss this experiment and comparison with lidar data.

In this paper we show our first POLINSAR results using the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) developed by the Jet Propulsion Laboratory (JPL). UAVSAR is a L-band repeat-pass polarimetric and interferometric system designed for measuring vegetation structure and monitoring crustal deformations. In order to extract canopy height from POLINSAR data and account for temporal decorrelation, we formulate a physical model of the temporal-volumetric coherence, random motion over ground (RMOG) model. Canopy height extracted from single-baseline UAVSAR data using the RMOG model is shown to be in agreement with canopy height measured by the Land, Vegetation, and Ice Sensor (LVIS) lidar.

The world forest systems are dynamic and play an integral role in the Earth's carbon budget. Monitoring of these valuable assets is being mandated by the international community. The requirement of global forest inventories suggests that a global measurement methodology should be adopted and that a verification and validation strategy should be accepted. The wide areas and varied forest types that need monitoring suggest the use of airborne remote sensing assets even for verification or cross validation of the varied global forest measurement methodologies. Airborne SAR systems that operate at appropriate frequencies, e.g., L-band or P-band, can provide useful forest information. The simplest forest product to generate would be a forest/non-forest classification map that can be robustly generated using radar polarimetric or interferometric systems. More elaborated products like forest classification or biomass maps require more sophisticated mapping algorithms and potentially ancillary data sets in order to obtain robust results. In this paper we examine the potential for airborne mapping system to obtain these type of forest mapping products and the limitations and accuracy of such systems.

This paper describes a physical model of the temporal changes that occur in vegetated land surfaces observed by a repeat-pass radar interferometer. We assume the temporal changes to be caused by a Gaussian-statistic motion of the vegetation elements, with motion variance changing along the vertical direction. We show that the temporal correlation between two interferometric radar signals is affected by the structural parameters of the vegetation, such as canopy height, and varies with the wave polarization. We validate the model using L-band data acquired by the Jet Propulsion Laboratory with the Uninhabited Aerial Vehicle Synthetic Aperture Radar airborne radar. This work provides new insights into the role of temporal decorrelation in interferometric radar applications.

The InSAR Scientific Computing Environment (ISCE) is a flexible, extensible software tool designed for the end-to-end processing and analysis of synthetic aperture radar data. ISCE inherits the core of the ROIPAC interferometric tool, but contains improvements at all levels of the radar processing chain, including a modular and extensible architecture, new focusing approach, better geocoding of the data, handling of multi-polarization data, radiometric calibration, and estimation and correction of ionospheric effects. In this paper we describe the characteristics of ISCE with emphasis on the ionospheric modules. To detect ionospheric anomalies, ISCE implements the Faraday rotation method using quad-polarimetric images, and the split-spectrum technique using interferometric single-, dual- and quad-polarimetric images. The ability to generate co-registered time series of quad-polarimetric images makes ISCE also an ideal tool to be used for polarimetric-interferometric radar applications.

We model the temporal decorrelation in volumetric media imaged by a repeat-pass SAR interferometer by using a temporal correlation function that varies with depth. An expression of this function is proposed and based on the Brownian motion of the canopy and soil elements. The spatial and temporal correlation terms are merged in a single coherence model that includes a large class of decorrelation effects, such as those induced by changes in the structure of the medium. We discuss the effects of the temporal correlation function and its implications on the parameters estimation using the POLINSAR random volume over ground model.

Due to their large scale of observation and their relatively high revisiting frequency, spaceborne SAR systems offer interesting possibilities for the systematic monitoring of urban areas. Several techniques have been developed to analyze urban areas from single-polarization spaceborne SAR data, based on the statistical properties of the reflectivity of such complex media and its spatial variations (texture). The reduced resolution of the data, compared to the airborne SAR case, is a particularly limiting factor. Polarization diversity offers an interesting and powerful alternative mean to detect and characterize urban areas. In this paper, we propose to use po-larimetric SAR acquired by the ALOS sensor at L band, to monitor urban areas. The proposed technique uses three complementary approaches to discriminate urban structures using detectors adapted to the complex polarimetric features of this medium, to isolate specific coherent responses from a Time-Frequency analysis of the coherent SAR signal, and finally to characterize built-up areas from the coherence properties of their Polarimetric and Interferometric SAR (POL-inSAR) response.

In this paper we discuss some aspects of the forest height estimation using Polarimetric and Interferometric (POLINSAR) SAR data. Three main issues limit the inversion of the POLINSAR coherence from repeat-pass POLINSAR systems: temporal decorrelation, terrain slope distortions and effects of wave penetration. We show that, if temporal decorrelation is not severe, the distortions due to terrain slope can be removed and the wave penetration can be compensated using the predictions of the scattering simulator PolSARProSIM. A detailed procedure that applies to any POLINSAR data is presented and illustrated using ALOS/PALSAR data and the SRTM digital elevation model (DEM).

GeoSAR is a unique dual-band, interferometric SAR (DBInSAR) sensor capable of collecting single-pass, X-band (VV) and P-band (HH) interferometric data simultaneously. In this paper we examine the dependence of the P-band HH interferometric phase centre height upon forest and terrain parameters. We develop a simple model for P-band GeoSAR observations, and use the model to show how the elevation in P-band HH phase centre height above true ground height is related to the volume-to-ground scattering ratio. GeoSAR is not fully-polarimetric, but records cross-polar (HV) returns at P-band (although not interferometrically). We conjecture that these returns are dominated by direct-volume scattering and related to the direct-volume HH backscatter. We use this relationship to model the dependence of the P-band HH DTM height upon the HV/HH ratio, and the difference in X-band DEM with P-band DTM heights. The relationships are examined using simulated forest InSAR data, and a model is proposed for ground-height and tree-height estimation using DBInSAR that does not require full polarimetry.

The objective of this paper is to make a review of the current status of the PolSARpro v4.0 Software (Polarimetric SAR Data Processing and Educational Toolbox), developed under contract to ESA by a consortium comprising I.E.T.R at the University of Rennes 1, AELc, DLR-HR and Dr mark Williams from Adelaide. The objective of this current project is to provide Educational Software that offers a tool for self-education in the field of Polarimetric SAR data analysis at University level and a comprehensive suite of functions for the scientific exploitation of fully and partially polarimetric multi-data sets and the development of applications for such data. The PolSARpro v4.0 Software establishes a foundation for the exploitation of Polarimetric techniques for scientific developments and stimulates research and applications developments using PolSAR and PolInSAR data.

This paper presents a new approach to the exploitation of polarimetric and interferometric (POLINSAR) data for the quantitative estimation of the forest height and, in general, of forest parameters. Our procedure aims to match simulations from the coherent scattering model PolSARProSim and real measurables. First, a parametric model analysis is used to study the dependence of the interferometric coherence versus forest height, canopy density and terrain slope. Secondly, we show the separation of scattering phase centers on PALSAR data acquired over the Amazon forest. Finally, we perform a preliminary forest height inversion based on the sensitivity of coherence versus forest height. The same procedure can be applied to a compact polarimetric (CP) data-set. In this case, we introduce a full POLINSAR (FP) reconstruction algorithm based on symmetry properties of geophysical media.

Polarimetric SAR interferometric data can provide estimates of forest biomass density. There are different approaches to deal with the inversion problem, such as neural networks and the traditional optimal estimation approach. This paper presents a study to evaluate their performance by means of quantitative indexes addressing both the computation time and the retrieval accuracy. Better forest parameters estimates have been obtained when neural networks algorithms were used.