Large scale monitoring of soil moisture is important for environmental systems and agriculture, with both optical and synthetic aperture radar (SAR) remote sensing having become methods of choice for repeatably imaging the Earth’s surface. Besides SAR backscatter, the repeat pass interferometric SAR (InSAR) phase is also sensitive to soil moisture changes and has been proposed as an observable for soil moisture retrieval algorithms. The phase loop sum between three interferometric SAR images, called closure phase, is of particular interest for soil moisture retrieval due to its insensitivity to topography and atmospheric changes. A specialized 3-D finite-difference time-domain simulation tool is used to simulate SAR pixels containing a variety of different soil structures. Soil parameters such as inhomogeneity size, moisture gradients and surface roughness are investigated, and the observed closure phase is compared against the predicted results from an analytical model. Further, the sensitivity of closure phase to changes in soil moisture is compared for each of the soil structures under investigation. Finally, we construct a simple moisture regression problem and show that including polarimetry can enable the regression to solve for soil structure properties.

Synthetic aperture radar (SAR) imaging of near Earth surface natural material volumes (soil, snow, vegetation) and man-made objects corresponds to an equivalent real aperture scenario with temporally short and spatially focused pulses. This equivalence holds as long as the imaged scene can be considered static within both fast (chirp duration) and slow (synthetic aperture) times. The electromagnetic backscatter recorded in a properly focused SAR image resolution cell that meets this condition has an amplitude and phase, which represents the coherent sum of the returns from individual scatterers inside the small volume represented by the cell. For some types of scatterers, the backscattering properties can be accurately characterized by analytical expressions. However, there are many interesting scattering phenomena where a purely analytical treatment leads to unrealistic simplifying assumptions. One approach to investigating these phenomena is with numerical methods such as the finite-difference time-domain method (FDTD). The FDTD method is computationally expensive, but it can model the complete physical interaction of electromagnetic waves according to Maxwell’s equations with arbitrary materials and shapes. This article describes the development of a specialized 3D FDTD software package capable of simulating the real-aperture equivalent of individual SAR image resolution cells given the spatial distribution of permittivity and conductivity within the cells. We demonstrate the usefulness of our new software tool by first recreating published 2D simulation results examining the link between soil moisture and InSAR phase, and then expanding these results to more realistic 3D soil volumes.

This study explores the effectiveness of three advanced multitemporal interferometric synthetic aperture radar (InSAR) algorithms for characterizing spatial and temporal features of the School Creek landslide in southern Tasmania, Australia: Persistent Scatterer InSAR (PSI), Quasi-PSI (Q-PSI), and Homogeneous Distributed Scatterers-InSAR (HDS-InSAR). We processed coarse-resolution Sentinel-1 and high-resolution RADARSAT-2 imagery spanning April 2019 to, respectively, December 2021 and August 2022. Both sensors satisfactorily reveal landslide extent and produce similar non-linear deformation time-series, with displacement rates of up to 10 mm/year. Nonetheless, RADARSAT-2 higher resolution enables more precise spatial characterization. The correlation coefficients between Q-PSI and HDS-InSAR deformation rates are 0.84 and 0.88 for Sentinel-1 and RADARSAT-2, respectively, and both techniques provide superior displacement-record density compared to PSI.

Snow water equivalent (SWE) is an important surface parameter for understanding a number of Earth system processes. Synthetic aperture radar (SAR) has considerable potential for measuring SWE of dry-snow because SAR can penetrate through the snow to the ground surface and is both amplitude- and phase-sensitive to refraction from the snow. Previous work on refraction-based SWE measurement by SAR has utilized the repeat-pass InSAR phase signal to estimate changes in SWE that occur between SAR acquisitions. These are subject to temporal decorrelation effects and consider only the refraction that occurs along the SAR beam center rather than the entire synthetic aperture. This study examines the refractive effect of dry-snow along the synthetic aperture and its impact on SAR image formation including defocusing and phase bias of the system impulse response. Snow phase compensation during time domain processing to recover the snow-free impulse response function (IRF) is described and demonstrated. The feasibility of using the mapdrift and image sharpness autofocus methods to estimate SWE is examined, and the effect of key system parameters on the estimation performance is derived. Experimental validation of the method was conducted by acquiring L-band data with the Simon Fraser University (SFU) Airborne SAR System over a pair of corner reflectors installed on the Kluane icefield in northwestern Canada. Results from both simulations and the icefield experiment are presented and compared including an analysis of errors affecting the estimation.

Interferometric synthetic aperture radar (InSAR) time series analysis of natural terrain allows for characterization of long-term geophysical trends over extended areas and, in the case of distributed scatterers (DSs), is significantly enhanced by methods that exploit the full complex-valued scattering statistics. Phase-linking (PL) estimators impose a phase-closure constraint in order to estimate the temporal wrapped-phase history of a DS directly from its complex backscatter sample coherence matrix. Some PL methods, such as the SqueeSAR and maximum-likelihood-estimator of Interferometric phase (EMI) estimators, rely on knowledge of the coherence magnitude matrix. The true coherence magnitude isa prioriunknown and must therefore be estimated from the data. Bias in these estimated coherence magnitudes reduces PL performance when the true coherence magnitude is low. Many areas of the Earth are seasonally snow-covered and, for natural terrain, this leads to severe cross-season decorrelation. This poses a significant challenge for PL estimators due to bias of the near-zero cross-season coherence magnitude estimates. We introduce a clustering approach to mitigate the PL estimator bias problem that exploits the fact that in natural terrain, many DSs decorrelate similarly. This allows for averaging over large numbers of same-behaving DS, which provides robust debiasing of the coherence magnitudes used during PL. We apply our method to a RADARSAT-2 spotlight-mode InSAR dataset over a site in the western Canadian Arctic and demonstrate significant reductions ina posterioriphase variance when compared to existing PL methods.

A photogrammetry system operated simultaneously with a synthetic aperture radar (SAR) from the same aerial platform provides strong sensor fusion possibilities that can improve the accuracy of repeat pass Interferometric SAR (InSAR) processing. Motion compensation is a key step in airborne SAR/InSAR processing, and the availability of accurate external digital elevation models (DEMs) is at the heart of many InSAR applications. Eventual research goals are (1) to produce high precision photogrammetric DEMs as reference for interferometric and tomographic applications, and (2) to use photogrammetric block adjustment parameters to fine-adjust the flight trajectory for enhanced motion compensation in repeat pass InSAR. To meet these goals; as SAR is oblique looking by design, an un-conventional off-nadir photogrammetric field-of-view coinciding with the SAR swath is required for our combined system. In this paper we focus on the accuracy implications of the derived photogrammetric DEMs from off-nadir vs nadir configuration as well as the potential for trajectory refinement from the derived photogrammetric block adjustment parameters. Additionally, we carry out an accuracy comparison of our photogrammetric system with established references (Fairbanks fodar™ and WorldDEM™). Vertical photogrammetric DEMs produced with our system over a test site near Silver City, Yukon Territory, Canada were found to be accurate with a mean height difference of 0.45 m and a corresponding standard deviation of 0.78 m from the reference fodar™ system; whereas produced nadir vs. off-nadir DEMs were found in accordance with each other with a mean difference of 0.29 m and standard deviation of 0.75 m. Furthermore, the estimated flight trajectory refinement for the test data had a mean value of 0.19 m with a standard deviation of 0.09 m.

Land Use/ Land Cover (LU/LC) segmentation is a widely studied topic in the field of remote sensing. Past focus has been on independent studies either on color (RGB) and the Normalized Vegetation Index (NDVI) or on Polarimetric Synthetic Aperture Radar (PolSAR) data. In this paper we explore the fusion potential of RGB images with additional SAR and Near Infra-red (NIR) images for enhanced LU/LC segmentation through Fully-Convolutional Neural Networks (F-CNNs). F-CNNs have been extensively studied for semantic segmentation problems with U-Net and SegNet being two well-known F-CNN architectures. Both these architectures were used as references for this study. High resolution RGB, SAR and NIR images were acquired through Google Earth (GE), German Aerospace Center (DLR) and The Planet Laboratories, respectively. IR was converted to NDVI for its higher potential of segmentation of vegetations areas. Four multi-sensor configurations as input channels to the networks were studied after precise co-registration of these images, and the results were compared to individual channels for both architectures. Simon Fraser University (SFU), Burnaby Campus and its surrounding area was selected for this study due its diverse land types. The area was divided into 5 classes i.e. Roads, Buildings, Forest, Water and No class (unclassified). An overall, best accuracy of ~86% was achieved for a five-channel configuration (R+G+B+SAR+NDVI). We show that the inclusion of SAR and IR channels to RGB based network can significantly improve the performance of LU/LC segmentation.

In this study, we develop a single-baseline Pol-InSAR approach for estimating the ground-to-volume ratios at different polarizations. We then apply it to time series of Pol-InSAR measurements over agricultural crops in C-Band to assess the relationship between these estimated Pol-InSAR parameters and plant biophyical properties such as structure, biomass and water content. The assessment reveals that the temporal patterns of the Pol-InSAR estimated ground-to-volume ratios are influenced by vegetation water content variations. The dependence of these temporal patterns on polarization indicates the role of crop structural properties in the Pol-InSAR measurements.

In this paper, we introduce a finite-difference time-domain simulator that accurately models the interaction of microwaves with realistic soils, specifically from spaceborne interferometric synthetic aperture radar (InSAR). The modeled soils are characterized by surface roughness, correlation length, bulk moisture content, vertical moisture gradient, and small air-filled-void content. Simulation results include both backscatter and interferometric phase, and we are particularly interested in assessing the potential of the latter as a proxy for soil moisture. We find that differences in homogeneous bulk moisture result in only small phase differences (30? for HH and > 50? for VV when the soil moisture is varied from 3% to 30% in the uppermost 2 cm of the soil. Phase changes of this magnitude are easily detectable by spaceborne InSAR techniques. While a strong phase response to a change in mean bulk moisture is common to vertical moisture gradient and small air-filled-void cases, their corresponding backscatter responses are very different. A vertical moisture gradient makes the backscatter response dramatically flatter compared with the case of uniform moisture; in contrast, the introduction of air-filled voids barely alters the backscatter. Thus, it may be possible to infer near-surface soil-structure parameters such as vertical gradients or fractions of voids and inhomogeneities from combined SAR phase and backscatter data. Future SAR sensors could be optimized for this purpose. Prior theoretical work based on the assumption of vertically uniform soil-moisture distributions may need to be adjusted, and the lack of a theory that accommodates more complex soil structures may explain why backscatter inversions have yet to result in a viable operational system.

Interferometric point target analysis using Gamma Remote Sensing's IPTA module has been applied to two interferometric stacks of RADARSAT-1 FINE mode data over the Belridge/Lost Hills oil fields in California. The stacks (/spl sim/30 images each) are both from descending orbits and span the same 2 year time period (2002/02-2004/02) but incidence angles are different (F1 vs. F3F). An IPTA analysis is carried out on each stack separately and results are then compared in terms of achieved density of the scatterers and residual errors of the motion analysis. We verify the accuracy of our methods indirectly by studying the subtle difference in line-of-sight motion caused by the different incidence angles of the data stacks (38.4 vs. 43.1 degrees at the study site).

Locust infestations of semiarid areas are preceded by well defined soil moisture conditions. Therefore the potential of outbreaks can be partially predicted by accurate mapping of soil moisture on the field-scale (/spl sim/100 m). In this manner, Radarsat International has run a successful locust outbreak prediction campaign in Kazakhstan (LIMIS). This was based on Radarsat-1 backscatter images combined with simultaneous soil moisture ground truth. Although, as shown by LIMIS, soil moisture maps can be successfully derived from RADARSAT-1 imagery using high quality ground measurements, this method has disadvantages. The main problem is the high cost and difficulty associated with collecting sufficient ground measurements. To overcome this problem, soil moisture maps can be derived without the use of ground measurements from multiple RADARSAT-1 images, each with a different incidence angle. This work presents encouraging research results towards an automated ground-measurement free system based on multi-angle SAR data inversion.