Model-based forest height inversion from single- and multi-baseline polarimetric synthetic aperture radar interferometry (Pol-InSAR) data is today well-established. One of the critical performance points is the parameterization of the vertical reflectivity profile. In this article, the parameterization of the vertical reflectivity profile using a tomographic reconstruction is proposed. To mitigate the limitations in the tomographic reconstruction induced by the limited vertical resolution especially for the ground-scattering component, the ground and volume contributions are separated using the sum of Kronecker products (SKPs) decomposition. For the forest height inversion, the Pol-InSAR line inclination angle is proposed, which, unlike the absolute coherence and the phase center, is invariant to the presence of a residual Dirac-like ground-scattering contribution. The inversion performance is addressed and compared for both single- and multi-baseline cases, in the absence and presence of temporal decorrelation. In all cases, the ground elevation is estimated from the tomographically reconstructed vertical reflectivity profiles and its accuracy directly influences the achieved inversion performance. The proposed methodology establishes a link between interferometric and tomographic measurements and is therefore relevant for missions that allow the implementation of both techniques. ESA’s BIOMASS is one such mission, and as such is being used as a test case to discuss different implementation scenarios. The different inversion strategies are applied to P-band campaign data acquired in the frame of the AFRISAR 2016 experiment and validated against reference measurements.

The advent of computerized medical recording systems in healthcare facilities has made data retrieval tasks easier, compared to manual recording. Nevertheless, the potential of the information contained within medical records remains largely untapped, mostly due to the time and effort required to extract data from unstructured documents. Natural Language Processing (NLP) represents a promising solution to this challenge, as it enables the use of automated text-mining tools for clinical practitioners. In this work, we present the architecture of the Virtual Dementia Institute (IVD), a consortium of sixteen Italian hospitals, using the NLP Extraction and Management Tool (NEMT), a (semi-) automated end-to-end pipeline that extracts relevant information from clinical documents and stores it in a centralized REDCap database. After defining a common Case Report Form (CRF) across the IVD hospitals, we implemented NEMT, the core of which is a Question Answering Bot (QABot) based on a modern NLP model. This QABot is fine-tuned on thousands of examples from IVD centers. Detailed descriptions of the process to define a common minimum dataset, Inter-Annotator Agreement calculated on clinical documents, and NEMT results are provided. The best QABot performance show an Exact Match score (EM) of 78.1%, a F1-score of 84.7%, a Lenient Accuracy (LAcc) of 0.834, and a Mean Reciprocal Rank (MRR) of 0.810. EM and F1 scores outperform the same metrics obtained with ChatGPTv3.5 (68.9% and 52.5%, respectively). With NEMT the IVD has been able to populate a database that will contain data from thousands of Italian patients, all screened with the same procedure. NEMT represents an efficient tool that paves the way for medical information extraction and exploitation for new research studies.

This paper provides an overview of the state of the art and an outlook on future developments of spaceborne Synthetic Aperture Radar (SAR) systems with multi-baseline imaging capability, such as 3D differential SAR interferometry (3D-DinSAR), polarimetric SAR interferometry (Pol-InSAR), tomography (TomoSAR), and holography (HoloSAR). The goal is to fill the multidimensional data space with additional information from images with different spatial and/or temporal baselines.

ESA Forest Carbon Monitoring project (FCM) is developing Earth Observation based, user-centric approaches for forest carbon monitoring. Forest carbon accounting based on forest inventory requires precise and timely estimation of forest variables at various spatial levels accompanied by verifiable uncertainty information. In this paper, we present the algorithm trade-off and selection approach and preliminary results of the algorithm intercomparison exercise in the FCM project. The studies were performed over 7 European test sites located in Finland, Ireland, Romania, Spain and Switzerland, and one tropical forest site in Peru. EO datasets were represented by Sentinel-1, Sentinel-2, TanDEM-X and ALOS-2 PALSAR-2 imagery. Examined approaches include popular parametric and SAR/InSAR scattering physics based approaches, and nonparametric and machine learning approaches such as k-NN, random forests, support vector regression.

The combination of TanDEM-X interferometric measurements with GEDI lidar full waveform measurements can provide continuous high-resolution forest height maps at global scale with sufficient accuracy without using external information about the underlyingtopography. In previous studies, the GEDI lidar full waveforms have been used to provide an approximation of the TanDEM-X X-band (radar) vertical reflectivity function in the height inversion of an entire TanDEM-X scene. This framework has been applied to the whole the whole Brazilian Amazon, and the obtained results are presented and analyzed in this paper. More than 12,000 TanDEM-X scenes and 250 millions GEDI lidar measurements have been processed.have.

Polarimetric SAR Interferometry (Pol-InSAR) is a SAR remote sensing discipline with unique and powerful applications related to the vertical structure of natural and man-made volume scatterers. The coherent combination of single- or multi -baseline interferograms acquired at different polarisations provides sensitivity to the vertical distribution of scattering processes and allows their characterisation by using the associated (volume) interferometric coherences.

Synthetic aperture radar tomography (TomoSAR) at lower frequencies allows the reconstruction of the 3-D radar reflectivity of volume scatterers allowing access to their physical 3-D structure by means of multiangular SAR acquisitions. The performance of the reconstruction critically depends on the number and (spatial) distribution of the tomographic acquisitions (tracks). This dependence is addressed in this letter with respect to forest applications (volume scatters) at L-band. The letter discusses the optimum definition of tomographic configurations based on the peak sidelobe level (PSL) of the point spread function (PSF). For demonstration, a tomographic data set consisting of 15 acquisitions acquired by the DLR’s F-SAR system at L-band over the Traunstein test site in Germany is used, complemented by airborne LiDAR measurements. Three different reconstruction algorithms (Fourier beamforming, Capon beamforming, and compressive sensing) are implemented and compared to each other for scenarios with a reduced number of acquisitions. Although the limitation of the specific forest, the results show the potential of using the PSL of the PSF to define tomographic configurations optimized for forest structure applications.

The TerraSAR-X add on for Digital Elevation Measurement (TanDEM-X) mission provides Interferometric Synthetic Aperture Radar (InSAR) wall-to-wall data (not sparse) at high resolution and at global scale. In addition, the NASA Global Ecosystem Dynamics Investigation (GEDI) is a new spaceborne system that provides (from 51.6°N and 51.6°S) sparse measurements (not images) through LiDAR waveforms. Both systems are sensitivity to the canopy structure such as the forest height but with their own limitations. The TanDEM-X single polarization (HH) interferometric coherence magnitude at X-band provides a continuous mapping of the forest while GEDI provides accurate (but sparse) measurements of the forest. In this paper a methodology of how to combine both systems to estimated forest height is presented and applied to more than 900 TanDEM-X scenes over Gabon in Africa. The forest height results over an area of 1° by 1° are shown and compared respect to GEDI. Finally, a wall-to-wall forest map over the entire country of Gabon is presented as an example of large scale mapping towards a potential global (entire earth) forest height map.

The realistic parameterization of the underlying (vertical) radar reflectivity profile is critical for a model-based inversion of forest height from polarimetric interferometric (Pol-InSAR) data. Indeed, an appropriate parameterization not only affects the final estimation performance, but also enables the inversion from a reduced observation space in terms of number of baselines and / or polarizations. Here, we investigate the possibility of using a full tomographic profile to parameterize the inversion. The proposed methodology is demonstrated and validated using Pol-InSAR and tomographic data acquired in the framework of relevant airborne campaigns.

Polarimetric SAR Interferometry (Pol-InSAR) is a SAR remote sensing discipline with unique and powerful applications related to the vertical structure of natural and man-made volume scatterers. The coherent combination of single- or multi -baseline interferograms acquired at different polarisations provides sensitivity to the vertical distribution of scattering processes and allows their characterisation by using the associated (volume) interferometric coherences [1]–[5].

Active microwave sensors, such as synthetic aperture radars (SARs), offer all-weather and daylight independent operability which is of great advantage for monitoring polar regions, where extreme environmental conditions and long period of darkness strongly limit other kinds of sensors. In addition, microwaves allow penetrating into dry snow and ice, making SAR measurements sensitive to the subsurface structure of glaciers and ice sheets. However, the retrieval of glacier subsurface parameters from SAR observations remains difficult due to their sensitivity to a large number of factors, including snow and ice properties, presence of layers, etc. The objective of this study is to attempt advancing the understanding of SAR measurements of glaciers and ice sheets. A combined analysis of polarimetric and tomographic measurements is carried out to derive a 3-D characterization of the scattering mechanisms occurring in a glacier subsurface scenario. The investigation exploits a fully-polarimetric tomographic airborne dataset, acquired over Greenland by the DLR's F-SAR system in the frame of the ARCTIC15 campaign.

TanDEM-X experiments have shown that forest height can be estimated with single polarization X-band interferometric coherences. An external digital terrain model (DTM) not only allows to use both coherence magnitude and phase information, but also to overcome X-band penetration limitations. However, DTM information is not available for large areas. Using coherence magnitudes makes height inversion feasible, but it requires a model relating coherence to height. Here we report an experiment using the X-band local phase center variations. Results over a tropical forest site show that in those stands in which the low X-band penetration is not a limitation, the there is a good correlation between the obtained TanDEM-X heights and the heights from Lidar measurements.

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.

Deep learning (DL) methods have in recent years yielded impressive results in medical imaging, with the potential to function as clinical aid to radiologists. However, DL models in medical imaging are often trained on public research cohorts with images acquired with a single scanner or with strict protocol harmonization, which is not representative of a clinical setting. The aim of this study was to investigate how well a DL model performs in unseen clinical datasets-collected with different scanners, protocols and disease populations-and whether more heterogeneous training data improves generalization. In total, 3117 MRI scans of brains from multiple dementia research cohorts and memory clinics, that had been visually rated by a neuroradiologist according to Scheltens' scale of medial temporal atrophy (MTA), were included in this study. By training multiple versions of a convolutional neural network on different subsets of this data to predict MTA ratings, we assessed the impact of including images from a wider distribution during training had on performance in external memory clinic data. Our results showed that our model generalized well to datasets acquired with similar protocols as the training data, but substantially worse in clinical cohorts with visibly different tissue contrasts in the images. This implies that future DL studies investigating performance in out-of-distribution (OOD) MRI data need to assess multiple external cohorts for reliable results. Further, by including data from a wider range of scanners and protocols the performance improved in OOD data, which suggests that more heterogeneous training data makes the model generalize better. To conclude, this is the most comprehensive study to date investigating the domain shift in deep learning on MRI data, and we advocate rigorous evaluation of DL models on clinical data prior to being certified for deployment.

Forest structure is a key parameter for forest applications, but it is difficult to be estimated at the required spatial and temporal scales. In this context, synthetic aperture radar Tomography (TomoSAR) that allows, at lower frequencies, the 3-D imaging of natural volume scatterers with high spatial and temporal resolution may be a game changer. The aim of this article is to evaluate three TomoSAR algorithms, Fourier beamforming (FB), Capon beamforming (CB), and compressive sensing (CS) with respect to their performance in the reconstruction of the 3-D forest reflectivity. The implications of volumetric forest scattering, as well as the temporal decorrelation of scatterers, are analyzed. The algorithms are compared on a set of simulated scenarios and then evaluated on an experimental L-band data set composed by four acquisition dates, each one consisting of five tomographic tracks. The data were acquired in 2014, within a time span of two months, over the Traunstein forest (Germany) using the F-SAR system. Additionally, discrete airborne Lidar has been used for a qualitative evaluation. The results indicate that the CS reconstruction is, for many practical cases, superior when compared to FB or CB reconstructions as they achieve higher vertical resolution, especially in cases with a lower number of acquisitions and complex forest scenarios. By combining acquisitions performed at different days, the effect of temporal decorrelation on each algorithm for two different tomographic implementations (repeat-pass vs. single-pass) has been assessed. The results indicate that simultaneously acquired image pairs allow a better reconstruction of the 3-D forest reflectivity.

NASA's Global Ecosystem Dynamics Investigation (GEDI) waveform lidar is expected to provide unprecedented measurements of forest structure and biomass in tropical and temperate environments. In order to bridge the limitations induced by the ground sampling of the GEDI waveforms, and to obtain enhanced forest structure estimates, the potential of combining TanDEM-X (high resolution) singlepass interferometric coherences and lidar waveforms is currently investigated. In this work, a combination framework based on the ability of lidar and TanDEM-X measurements to express physical forest structure by means of appropriate indices is discussed. In particular, commonalities and complementarities between the different measurements are addressed by means of experimental results obtained in temperate and tropical forest sites in which comparisons among structure indices from lidar, T anDEM-X and field inventories can be established.

The objective of this paper is to address the penetration capabilities of different synthetic aperture radar (SAR) frequencies in forest scenarios in terms of "visibility" of the ground. First, the ratio between the ground and the volume scattering powers is estimated from multi-baseline polarimetric interferometric (Pol-InSAR) data across different forest types. Afterwards, the resulting performance in the estimation of the ground height is characterized by comparing algorithms based on either the identification of the ground scatterer in tomographic profiles, or on the inversion of a model. This analysis is supported by experimental results obtained with multi-frequency, multibaseline Pol-InSAR data acquired by the DLR's E-/F-SAR airborne platforms over different forest types.

The need of large scale observations with high temporal frequency has promoted airborne and satellite remote sensing techniques for glaciological applications. In particular, active microwave sensors, such as synthetic aperture radars (SARs), offer all-weather and daylight independent operability which is of great advantage at high latitudes, where extreme environmental conditions and long period of darkness strongly limit other kinds of sensors. Moreover, longer wavelengths allow to penetrate significantly into dry snow and ice, interacting with surface as well as subsurface features. On the one hand, this makes SAR measurements suitable to investigate the subsurface structure of glaciers and ice sheets. On the other hand, the complex interaction of microwaves with the subsurface layers makes the interpretation of SAR measurements challenging. This study investigates the potential of SAR techniques to retrieve information about glacier subsurface. SAR polarimetry and tomography are used to gain a 3-D characterization of the scattering scenario of the K-transect, a site located in the ablation zone of Greenland. For this, a fully-polarimetric tomographic airborne dataset, acquired by the DLR's F-SAR system in the frame of the ARCTIC15 campaign, is exploited.

Synthetic Aperture Radar Tomography techniques provide 3D information of the forest due to the ability of microwaves to penetrate through vegetation. Recent studies link the radar 3D information to forest 3D structure in order to translate the tomographic results to an ecological interpretation. However, due to the undersampled nature of tomographic acquisitions, the number and distribution of acquisitions can change the estimated 3D radar reflectivity and as a consequence the forest structure estimates. This paper explores the results for radar as well as for forest structure for different number and distribution of acquisitions in order to analyse the potential and limitations of the estimation of forest structure in future space borne scenarios where the number and distribution of acquisitions will be suboptimal. In this context, the paper analyses a tomographic campaign of 15 tracks acquired over Traunstein (Germany) in 2017 together with ground measurements and Lidar.

Recent TanDEM-X experiments have shown that the limited penetration capability at X-band in forest volumes allow the estimation of the height variability of the top canopy layer, which can be used as a proxy to the horizontal structure (i.e. heterogeneity), by using high resolution digital elevation models (DEMs). However, the use of an external digital terrain model (DTM) is necessary to separate the (high resolution) canopy height variations from the topographic ones. In this work, the possibility of compensating terrain topographic variation by using a low resolution TanDEM-X DEM instead of an external DTM is investigated. The results show that the use of a reference DEM with a resolution on the order of 100 m allows to compensate the terrain-induced topographic variations and to preserve the information on forest horizontal heterogeneity at a large extent.

In this paper, polarimetric multi-baseline (MB) Synthetic Aperture Radar (SAR) Interferometry data are used to estimate the polarimetric ground component under vegetation. However, the solution of the applied separation algorithm is not unique and depends on the constraints in the regularization. First, the effect of this non-uniqueness is analyzed and then exploited to isolate a ground component with minimized influence of depolarizing scattering mechanisms. Using experimental MB SAR data acquired by DLR's airborne sensor F-SAR, the polarimetric entropy and mean alpha angle of the isolated ground component are compared to the original polarimetry of the full image. Finally, the ground polarimetry is interpreted for changing soil moisture vegetation conditions in corn. To this purpose, three dates are compared characterized by 1) a change in soil moisture, 2) a change in vegetation cover or 3) a simultaneous change of soil moisture and vegetation cover.

Polarimetric SAR Interferometry (Pol-InSAR) is a SAR remote sensing discipline with unique and powerful applications related to the vertical structure of natural and man-made volume scatterers. The coherent combination of single- or multi-baseline interferograms acquired at different polarisations provides sensitivity to the vertical distribution of scattering processes and allows their characterisation by using the associated (volume) interferometric coherences [1]-[5].

Space borne implementations of SAR tomography (TomoSAR) for forest volumes are particularly challenging, as the estimation of 3D reflectivity has to be carried out relying on a small number of acquisitions (or interferometric coherences) and/or non-uniform baseline distributions. A way to overcome this shortcoming is to approximate the reflectivity profile by a weighted series of functions, as for example proposed in the case of Polarization Coherence Tomography (PCT). In this case the individual weights can be estimated from a rather small number of interferometric coherences. In this work, we investigate how conventional TomoSAR can be linked to PCT by means of a suitable function basis. A real data analysis is carried out on TomoSAR reflectivity profiles obtained from a data set acquired in an airborne campaign over a temperate forest.

The aim of this paper is to interpret and characterize the changes of the 3-D polarimetric scattering signatures of agricultural crops at C-band and to relate them to temporal changes of the soil and plant parameters. For this, a time series of multibaseline (MB) synthetic aperture radar (SAR) data acquired at C-band by the airborne F-SAR system of the German Aerospace Center over the Wallerfing test site in Germany was analyzed. The availability of MB SAR data enables the resolution of scattering contributions in height by means of SAR tomography. The tomographic profiles at different polarizations were analyzed regarding temporal changes for different crop types. First, it was investigated if the center of mass (CoM) of the vertical reflectivity profiles as a single parameter enables the tracking of changes in soil and vegetation. The results show that the vertical reflectivity profiles and their CoM do not allow resolving the ambiguity if a change originates from soil or vegetation dynamics as expected. Thus, the scattering contributions from ground and volume were separated in height, using a filtering approach, and used for the estimation of the ground and volume scattering powers by means of covariance matching. Comparing the outputs with coincident ground measurements showed that dielectric as well as geometric changes in the vegetation are traceable by the separated ground and volume powers. Finally, the estimated powers were analyzed with respect to orientation effects, i.e., to polarimetric anisotropic behavior. They were found to be not significant for the crops under study at C-band.

Synthetic aperture radar (SAR) tomography (TomoSAR) allows the reconstruction of the vertical distribution of the power backscattered by natural volumes by combining multiple SAR images acquired with slightly different incidence angles. Being a “radar” quantity, the profile depends on the radar frequency, polarization, and acquisition geometry, the 3-D distribution of the scattering elements and their dielectric properties. The characterization of each one of these factors is crucial to enable the extraction of physical 3-D structure attributes from TomoSAR profiles. The objective of this paper is to investigate how the vertical distribution of the backscattered power at L-band is affected by seasonal- and weather-induced changes. Radiometric (e.g., ground and volume powers) and geometric (e.g., center of mass of the volume-only profiles and phase centers of the volume scattering layers) parameters have been estimated under different weather and season conditions and compared. Then, TomoSAR data sets affected by dielectric nonstationarity (i.e., variability) have been considered in order to assess the invariance degree of each radiometric and geometric parameters. This analysis has been carried out by processing four L-band airborne TomoSAR data sets acquired before and after a rainfall and in spring and autumn over the Traunstein forest (south of Germany).

AfriSAR is an ESA-funded airborne P- and L-band SAR campaign over the African tropical forests of Gabon carried out by ONERA (July 2015) and DLR (February 2016). The different acquisitions were designed in order to collect multibaseline fully polarimetric data allowing the inversion of key forest vertical structure-based parameters, like e.g. forest height and high resolution reflectivity profiles. Results of processing and parameter inversion with the DLR's F-SAR data are shown in this paper at both P- and L-band.

The purpose of this paper is to investigate how SAR Tomography (TomoSAR) can be used to estimate ground polarimetric covariances in forest scenarios in order to increase the amount of information that can be extracted from SAR data stacks is beyond the vertical reflectivity profiles. Under the hypothesis of a two-layer model composed by ground and volume contributions, an algorithm is proposed that makes use of an a priori knowledge of the ground height. The performance in the estimation of the ground polarimetry is investigated with respect to frequency and accuracy of the ground height estimated from the TomoSAR stack. This analysis is carried out by processing an L- and a P-band TomoSAR data stacks acquired by the DLR's E-SAR sensor over the temperate forest of Traunstein.

TanDEM-X forms together with TerraSAR-X the first single-pass polarimetric interferometer in space. This allows for the first time the acquisition and analysis of Single-, Dual-, and Quad-Pol-InSAR data without the disturbing effect of temporal decorrelation globally. For this reason, the exploration of TanDEM-X data for forestry is constantly increasing especially concerning forest height estimation, biomass classification and structure characterization. This paper reports the results of recent experiments aimed at investigating the potentials of TanDEM-X in characterizing quantitatively the spatial variability of the canopy top and phase center height, which is a proxy to horizontal structure. It is shown that such characterization can allow to differentiate among e.g. different successional and / disturbance stages in tropical forests.

In forest and agricultural scattering scenarios, the backscattered synthetic aperture radar (SAR) signature consists, depending on the frequency, of the superposition of ground and volume scattering contributions. Using multibaseline SAR data, SAR tomography techniques allow resolving contributions occurring at different heights. Two algorithms for the separation of ground and volume scattering are compared with respect to their ability to provide a coherent volume component that can be further used for parameter inversion, both of them requiring only the a priori known ground topography. Once the volume-only coherences are available, the total ground and volume scattering powers are estimated by means of a least squares fitting. The objective of this letter is to quantitatively evaluate the performance of this estimation by means of a Monte Carlo analysis with simulated data focusing on the impact of vertical resolution, errors in the knowledge of the ground topography and phase calibration residuals.

A two-layer model composed by ground and volume contributions has been proven suitable to describe the 3-D backscattering signatures of forest scenarios in a number of experiments. Under this hypothesis, the purpose of this letter is to investigate how synthetic aperture radar tomography (TomoSAR) can be used to estimate ground and volume polarimetric covariances and with which performance. An algorithm which is able to overcome the intrinsic ambiguity in the estimation problem is proposed, and it is shown to be a reliable alternative to the poorly performing full-rank Capon beamformer for estimating the ground polarimetric covariances. This performance improvement can be achieved, for instance, if an a priori knowledge of the ground topography (or an accurate estimate of it) is available. This analysis is carried out by processing an L-band TomoSAR stack acquired by the DLR's E-SAR sensor over the temperate forest site of Traunstein.

AfriSAR is an ESA-funded airborne P-band SAR campaign over the African tropical forests of Gabon that is being carried out by ONERA (July 2015) and DLR (February 2016) in support of the development of the geophysical algorithms of the future BIOMASS mission. Multibaseline fully-polarimetric acquisitions have been designed over four test sites in order to further develop and validate algorithms for the estimation of forest upper canopy height, 3-D structure and biomass by implementing PolSAR (SAR Polarimetry), Pol-InSAR (Polarimetric SAR Interferometry) and TomoSAR (SAR Tomography) P-band measurements. In this paper first data processing results will be shown.

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.

Tandem-L is a highly innovative SAR satellite mission for the global observation of dynamic processes on the Earth's surface with hitherto unknown quality and resolution. Thanks to its novel imaging techniques and its unprecedented acquisition capacity, Tandem-L will deliver urgently needed information for the solution of pressing scientific questions in the areas of the biosphere, geosphere, cryosphere and hydrosphere. The feasibility of Tandem-L has been analyzed and confirmed in the scope of a phase A study, which has been conducted in close cooperation between the German Aerospace Center (DLR) and the German space industry. This paper provides an overview of the Tandem-L mission concept and summarizes the actual development status.

This paper provides an overview of the future development of spaceborne SAR systems with multi-baseline imaging capability like polarimetric SAR interferometry (PolInSAR), tomography (TomoSAR) and holography (HoloSAR). The goal is to fill the multi-dimensional data space with additional information from acquisitions having different spatial or temporal baselines. Multi-baseline imaging opens the door for a new class of image products in spaceborne SAR. Well-known examples are across-track and along-track interferometry, which allow the measurement of surface topography, ground deformation, ocean currents as well as glacier movements. While across-track and along-track interferometry are well established techniques and have been widely used by current spaceborne SAR systems, PolInSAR, TomoSAR and HoloSAR are emerging techniques which are shaping the future development of spaceborne SAR. New mission concepts for multi-static SAR configurations with distributed and sparse arrays will pave the way for this development.

Polarimetric SAR Interferometry (Pol-InSAR) is a SAR remote sensing discipline with unique and powerful applications related to the vertical structure of natural and man-made volume scatterers. The coherent combination of singleor multi-baseline interferograms acquired at different polarisations provides sensitivity to the vertical distribution of scattering processes and allows their characterisation by using the associated (volume) interferometric coherences [1][2][3][4][5]. Pol-InSAR is today a well established technique that promises a break-through in solving essential radar remote sensing problems. Indeed, structural parameters of volume scatterers in the biosphere and cryosphere such as vegetation height, structure, biomass, snow depth, and ice layering are today critical inputs for ecological process modeling and enable monitoring and understanding of eco-system change. In this paper we review the actual status of PolInSAR techniques and applications in forestry, agriculture and cryosphere; assess results from actual space (TanDEM-X) and air-borne (F-SAR) campaigns and experiments at different frequencies; and finally discuss new potential applications and challenges.

Systems based on Synthetic Aperture Radar Tomography at low frequencies offer 3D imaging capabilities, appropriate for forest monitoring. However the extraction of an ecologically meaningful measure of forest structure from the 3D reflectivity is not straightforward and several considerations need to be carefully taken into account, in order to avoid misinterpretations of the nature of the information reflected in the tomograms. Besides, it should be noted that the methodology employed in the TomoSAR inversion has a significant effect on the overall performance of the TomoSAR system to estimate forest structure. In this framework, this paper discusses the potential and limitations of TomoSAR systems for forest structure estimation.

In this study application results are presented derived from multi-parametric SAR observations covering five different thematic domains: forest, agriculture, ocean, urban and cryosphere. In total 21 application products have been selected and described. Their application on different data sets, space- and airborne sensors, was demonstrated and can independently be reproduced by any scientist. The results and algorithms are available soon through Springer.

In this paper, agricultural crop volumes are analysed using (polarimetric) tomographic SAR methodologies. A procedure for the separation of the ground and volume multibaseline coherences is proposed. This separation is used in particular to investigate the polarization dependency of the vegetation vertical structure in order to get first insights about orientation effects. This analysis has been carried out for different crops and on different development stages. The presented results have been obtained by processing a multibaseline fully polarimetric data set purposely acquired at C-band by the DLR's airborne sensor F-SAR in 2014.

Continuous development, experimentation and validation activities with airborne campaigns pushed towards the proposal of spaceborne missions that could implement SAR Tomography (TomoSAR) on a global scale. Together with innovative SAR imaging modes, TomoSAR plays an essential role in forest mapping for the characterization of 3-D structure and its dynamics at high spatial and temporal resolutions. In this framework, the characterization of the structure information content of multibaseline (MB) data stacks is of fundamental importance in order to shape both system design and acquisition concepts. In this paper, MB acquisition tradeoffs are analysed in representative forest scenarios fixing a given performance in the estimation of the parameters characterizing the vertical structure. In addition, the benefit against temporal decorrelation of repeated MB single-pass interferometric acquisitions is quantified with respect to repeat-pass acquisitions, and the role of polarization is discussed.

Synthetic Aperture Radar Tomography provides the 3D reflectivity of the observed scene. Hence, in a forest scenario, it reflects information relevant to forest structure. However, how and to which extent this information is reflected and which are possible ways to quantify it are still open questions. This paper explores the link between ecological measures of forest structure and the 3D spatial distribution of the peaks in the reflectivity profiles at L-band and, with this, proposes measures for forest structure estimation from tomographic SAR data.

Forest height is an important parameter for the allometric estimation of above-ground forest biomass (AGB). However, variable forest stand densities limit the performance of the allometric estimation of AGB from height measurements alone. Recently, the use of vertical forest structure information as an indicator for the variation of stand density has been proposed and used to improve the allometric estimation of AGB from height measurements. In this letter, the use of vertical radar reflectivity profiles at L-band obtained from SAR tomography, as a proxy for vertical forest structure for the allometric estimation of AGB, is investigated. L-band reflectivity profiles, which are reconstructed from data at different polarizations (HH and HV) and acquired under “moist” and “dry” weather conditions, are investigated. The proposed allometric AGB estimator increases the correlation factor from 0.60 to 0.81 and reduces the root-mean-square error from 50.25 to 36.30 Mg/ha when compared with the AGB estimation from forest height alone. The effect of polarization and weather conditions on the AGB estimation performance is discussed.

One common method to estimate biomass is measuring forest height and applying allometric equations to get forest biomass. However, changing forest density or forest structure bias the known allometric relations. Remote sensing systems like SAR or LIDAR allow to measure vertical forest structure. In this paper the value of vertical forest structure information for biomass inversion is investigated. First, vertical biomass profiles are calculated from forest inventory data. Then, a vertical structure descriptor based on Legendre polynomials is suggested and its sensitivity to biomass is evaluated. In a second step, this descriptor is used to describe SAR vertical reflectivity profiles. Then, a biomass estimation algorithm is developed. This is a case study based on inventory data from the Traunstein test site, a temperate mixed forest, located in the southeast of Germany.

Synthetic aperture radar (SAR) waves have the capability of penetrating into forest volumes. Multibaseline (MB) polarimetric-interferometric SAR (PolInSAR) systems can retrieve information about the vertical distribution of the radar backscattered power and to relate it to the physical 3-D forest structure of the vegetation. Radar backscattering from forests can change in time due to different weather conditions, to seasonality, and to logging or disturbances. In this paper, experiments with L-band and X-band data are presented for a first characterization of the related effects in the estimated 3-D forest structure parameters.

TanDEM-X (TDX) forms with TerraSAR-X (TSX) the first single-pass synthetic aperture radar (SAR) interferometer in space with polarimetric capabilities. The availability of such system allows for the first time the acquisition and analysis of X-band Pol-InSAR data from space without the disturbing effect of temporal decorrelation. After two years of mission, time series with variable baseline over the same forest sites are available, allowing to (1) explore their information content, (2) assess penetration capabilities, (3) assess scattering model assumptions, and (4) estimate vertical structure and monitor its dynamics. This paper discusses for the first time the potential of estimating forest vertical structure from spaceborne single-pass interferometers, extending classical tomographic concepts. Results of first experiments with TSX/TDX multibaseline Pol-InSAR data acquired over the Tapajos national forest (Brazil) are shown. Especially regarding tropical forests, potentials and applications of X-band for forest structure monitoring will also be discussed.

Forest characterization and biomass estimation by means of remote sensing systems are nowadays “hot topics” within the remote sensing community, given their importance in the terrestrial carbon budget. In fact, forest vertical structure is a key variable for assessing biodiversity and structural degradation and/or regeneration. Moreover, the (vertical) structure information is important as it can allow the development of accurate and robust (alometric) estimators of the forest biomass. In this paper, potentials and challenges of forest vertical structure estimation with low frequency multibaseline polarimetric synthetic aperture radar are reviewed and discussed.

Synthetic aperture radar (SAR) systems in L-band and P-band are characterized by deep penetration capabilities into volumes, enabling new opportunities for the radar remote sensing of forests. In the last years, the interest has been continuously growing in the estimation of the sub-canopy topography, especially by exploiting multibaseline (possibly polarimetric) SAR data. This work intends to contribute on this topic by presenting further experiments with real L-band data about ground topography estimation and by quantifying the obtained performance. Different forest scenarios are considered. Potentials and limitations are analyzed with particular reference to a multibaseline relaxation-based algorithm.

Prior to any processing of multibaseline (MB) synthetic aperture radar (SAR) data stacks, a MB phase calibration is necessary to compensate for phase contributions due to platform motions and/or atmospheric propagation delays. Classical calibration methods rely on the detection of point-like scatterers. However, especially in natural scenarios, their final calibration performance could be impaired by the nature of the scattering and by the typical low number of baselines. In this paper, we propose a calibration method based on the minimization of the entropy of the vertical profile of the backscattered power. This allows to potentially exploit the MB SAR signal independently of the nature of the scattering. The proposed method has been tested by processing simulated and real airborne datasets of a forest stand.

Interest is growing in the application of coherent processing of synthetic aperture radar (SAR) data to the monitoring of complex urban or infrastructure areas. However, such scenarios are characterized by the layover phenomenon, in the presence of which conventional interferometric SAR techniques degrade or cannot operate. As a consequence, to monitor reliably a high number of ground structures, the identification, i.e., the detection and height and deformation velocity estimation, of both single and multiple scatterers interfering in the same SAR cell can be a key step. This issue is addressed here by means of differential tomography (Diff-Tomo), a recent multibaseline-multitemporal generalized interferometric framework which allows to resolve multiple moving scatterers at different heights in the same cell. In particular, superresolution adaptive Diff-Tomo is extensively tested and augmented with a new information extraction algorithm for the automated identification of the multiple scatterers. Experiments have been carried out with real C-band spaceborne data over urban areas; corresponding results are shown and discussed.

SAR Tomography (Tomo-SAR) is an experimental advanced coherent data combination mode allowing full 3-D imaging of volumetric and layover scatterers from a multibaseline (MB) synthetic aperture radar (SAR) data stack. However, the linear Fourier-based Tomo-SAR is generally affected by unsatisfactory imaging quality due to a typically low number of baselines with irregular spatial distribution. Recently, to improve the elevation focusing technique, a sector interpolation approach has been proposed by the authors, in which a set of uniform baseline data is recovered from the available non-uniform one by exploiting the a priori information about the extension of a height sector which contains the scatterers. In this work, first experiments are presented of sector interpolated Tomo-SAR carried out with real spaceborne MB SAR data acquired over the Cinecittà area of the city of Rome.

In the last few years, the interest is increasing in the interferometric processing of multibaseline/multitemporal SAR data from complex urban or infrastructure areas. In order to locate and monitor a high number of ground structures with the lowest signal misinterpretation, the identification, i.e. the detection and height and deformation velocity estimation, of both single and multiple layover scatterers is an important step. This issue is addressed here by extensively experimenting the technique of adaptive differential tomography, a recent interferometric framework which allows to resolve multiple moving scatterers at different heights in the same SAR cell. To this aim, adaptive differential tomography is augmented with an automated information extraction algorithm. The technique has been applied to real C-band spaceborne data over an urban area. Corresponding results are discussed.

This work is focused on the detection of single and multiple scatterers in multiview/multitemporal SAR data in order to locate and monitor a high number of ground structures with low signal misinterpretation. This issue is addressed here by combining amplitude and phase data, differently from common techniques for the detection of single scatterers which rely on phase coherence measures. Experiments with real satellite C-band data are presented with both full resolution and multilook processing.

Multibaseline (MB) synthetic aperture radar (SAR) tomography is a promising mode of SAR interferometry, allowing full 3-D imaging of volumetric and layover scatterers in place of a single elevation estimation capability for each SAR cell However, Fourier-based MB SAR tomography is generally affected by unsatisfactory imaging quality due to a typically low number of baselines with irregular distribution. In this paper, we improve the basic elevation focusing technique by reconstructing a set of uniform baselines data exploiting in the interpolation step the ancillary information about the extension of a height sector which contains all the scatterers. Thisaprioriinformation can be derived from the knowledge of the kind of the observed scenario (e.g., forest or urban). To demonstrate the concept, an imaging enhancement analysis is carried out by simulation.

Multi-Dimensional (MultiD) SAR imaging is a modern technique, based on coherent SAR data combination, aimed to space (full-3D) and space deformation-velocity (4D) analysis. It extends the concept of SAR interferometry and differential interferometry and offers new options for the analysis and monitoring of ground scenes. With this regard, we discuss the current status and the results obtained by processing ERS real data, we investigate perspectives related to the next generation multi-static satellite formations, and we show some sample results regarding 3D and 4D theoretical performance bounds.