We employ data from the second generation of SAR systems, e.g., the Italian COSMO-SkyMed (CSK) constellation, the German TanDEM-X formation and the European Sentinel-1A/B constellation to monitor the characteristics of grounding line migration using short repeat-time interferometry and accurate InSAR DEM in the Amundsen Sea Embayment (ASE), West Antarctica. The ASE is a marine-based ice sheet with a retrograde bed containing enough ice to raise global sea level by 120 cm. Several studies have inferred the mechanical properties of portions of ASE using observationally constrained numerical models, but these studies offer only temporal snapshots of basal mechanics owing to a dearth of observational time series. Prior attempts of grounding lines mapping have been limited because few space-borne SAR missions offer the short-term repeat pass capability required to map the differential vertical displacement of floating ice at tidal frequencies with sufficient detail to resolve grounding line boundaries in areas of fast ice deformation. Using 1-day CSK repeat pass data and TanDEM-X DEMs, we collected frequent, high-resolution grounding line measurements of Pine Island (PIG), Thwaites, Kohler and Smith glaciers spanning 2015-2016. We compare the results with ERS data spanning 1996-2011, and Sentinel-1A 2014-2015 data. We observe an ongoing, rapid 2 km/yr grounding line retreat on Smith, 0.5 km/yr retreat on Pope, ongoing 1 km/yr retreat on Thwaites and PIG and a slight re-advance on Kohler since 2011. On PIG, the data reveal rapid subsidence along the glacier flanks, significantly more than in 1996/2000. We do not observe similar patterns on the other glaciers.

The processing of spaceborne Synthetic Aperture Radar (SAR) range and azimuth measurements according to geodetic standards enables accurate observations of dedicated point targets (corner reflectors, CRs) across SAR missions with different frequencies. This approach, also called imaging geodesy, relies on precisely determined orbits and corrects the inconsistencies introduced by approximations in SAR image processing, the disturbances of atmosphere (path delays) and solid Earth tides in order to obtain consistent measurements. If these requirements are fulfilled, the range and azimuth measurements from different pass geometries can be jointly processed for the global 3D positions of the reflectors, which is even possible when using data of different SAR missions.

The TerraSAR-X mission (X-band, 9.65 GHz) operated by DLR is well known for its centimeter level measurement and positioning capability, but the combination with other missions becomes necessary if only small amount of data are available. ESA’s Sentinel-1 mission (C-Band, 5.04 GHz) can achieve a similar observation quality, but is partly limited by the coarser resolution of the standard image product (Interferometric wide-swath product). However, this can be largely compensated by the larger amount of available Sentinel-1 data and therefore it makes sense to combine the measurements of both missions. We will demonstrate how this can be done in order to retrieve centimeter-level accurate CR coordinates based on our tests carried out in Australia.

During the past five years, Australia has considerably expanded its terrestrial SAR infrastructure. This infrastructure now offers a reflector array located in Queensland that consists of 40 trihedral CRs covering about 130 x 130 kilometers, as well as two trihedral CRs at the geodetic station of Yarragadee Geodetic Observatory in Western Australia. The CRs of both sites have accurate reference coordinates in the International Terrestrial Reference Frame (ITRF). The array has been surveyed with GNSS, whereas the CRs at Yarragadee have been integrated into the local station network by measuring the local ties, which makes them ideal testbeds for our imaging geodesy methods. The sites are regularly captured by Sentinel-1 and TerraSAR-X, which ensures sufficient amount of image data for the experiments. We will use the CRs at Yarragadee to confirm the consistency of both SAR missions and to demonstrate the combination method, and then assess the positioning on a larger scale by using the data from the Australian CR array.

In the framework of this project we investigated the capability of Satellite SAR Interferometry for the temporal and spatial prediction of sinking phenomena due to the evolution of underground cavities and potentially interfering with the water and sewerage system in the urban area of Rome.
In order to support a local company to develop a plan for monitoring and prevention of hydrogeological instability events in the city, high-resolution X-band COSMO-SkyMed (Italian Space Agency) and medium-resolution C-band Sentinel-1 (European Space Agency) data have been used.
In particular, both Ascending and Descending orbital geometry have been used for COSMO-SkyMed (covering the period from 2011 to 2017) and Sentinel-1 (covering the period from 2014 to 2017).
The study area includes the southeastern portion of Rome in a densely urbanized area. This area is well-known, as it is affected by the occurrence of several sinkholes, especially during the last decades, mainly due to the collapse of tunnels (e.g. electrical, sewage) or mining galleries.
In addition to the InSAR analyses, several stratigraphic logs and location of the historical sinkholes and cavities have been collected. Based on the available stratigraphic data, we built a 3D geological model of the area and studied the relationship between subsoil and cavities occurrence, obtaining the Sinkhole Susceptibility Index Map.
The A-DInSAR results allowed to predict the spatial occurrence of sinking phenomena. In particular, we have been able, through the clustering analysis of measurement points, to identify specific portion of the territory potentially interested by sinking phenomena where to focus the attention.
By combining multi-frequency C- and X- band A-DInSAR results, the subsurface geology (in terms of thickness and depth of lithological units), the historical subsurface excavation tracks and the Sinkhole Susceptibility Index Map, an operational procedure able to identify the potential occurrence of cavities and sinkholes was developed.
This case study represents an interesting example of application of A-DInSAR, not only from a scientific point of view, but also because it proves that high- and medium- resolution satellite missions are a valuable resource for local companies that deal with the management of water and sewerage frameworks, that can benefit from high accuracy, spatially dense and fairly low-cost monitoring data.

The aim of the paper is the experimentation of innovative multi-temporal/multi-frequency differential SAR tomographic techniques for the detection of anomalous deformations on man-made structures that could be a premonitory to the disastrous events.
Specific innovative Differential SAR Tomographic (D-TomoSAR) techniques able to provide information related to possible displacements and deformations with the accuracy of few millimiters are implemented, including a proper structural model which takes into account the particular geometrical configuration of observed objects (bridges, buildings, etc…).
The algorithm is based on the detection of proper control points of the object under investigation, which is particularly suited to analysis of deformation and to the detection of anomalous dynamics. The algorithm is a proper modification of the D-TomoSAR algorithm already developed by the authors for building monitoring. The main innovative issues of the proposed algorithm are, respect to the state of the art methods present in literature, are the following:
- Separation of the contribution of reliable multiple scatterers laying in the same range-azimuth resolution cell and estimation of their deformation and thermal dilation. This feature is very important when in the area under observation different strongly reflecting structures (bridges, buildings, railway, etc.) are contributing to the backscattering in the same range-azimuth resolution cell. Interferometric techniques are not able to separate their different contribution and then would lead to inaccurate estimations;
- Inclusion of a simple structural model allowing the approximated description of the object under investigation dynamics and the identification of the most important control points to be monitored;
- Selection of the most reliable pixels in the image and evaluation of the expected estimation accuracy;
- Analysis and identification of possible premonitory events, if they have occurred, for the definition of risk and early warning conditions.
The analysis is conducted based on multi-temporal and multi-frequency data acquired over the area of Genoa, Italy. Concerning the frequency, the algorithm exploits both C-band Sentinel 1 data and X-band Cosmo-SkyMed ones. Concerning the temporal extension, the data sets have an overall temporal baseline of more than 3 years.
The final output of the implemented algorithm is a map of possible spatial displacements of the structure under investigation and its thermal dilatation occurred during different years


. [1] Budillon, A. Evangelista, G. Schirinzi. “Three-Dimensional SAR Focusing From Multipass Signals Using Compressive Sampling”, IEEE Trans. Geosci. Remote Sensing, 49, p. 488-499, 2011.
. [2] Budillon, G. Schirinzi. “GLRT Based on Support Estimation for Multiple Scatterers Detection in SAR Tomography“, IEEE J. of Selected Topics in Applied Earth Obs. and Remote Sens., 9, pp. 1086-1094, 2016.
. [3] Budillon, A. Johnsy, G. Schirinzi. “A Fast support detector for superresolution localization of multiple scatterers in SAR tomography”. IEEE J. of Selected Topics in Applied Earth Obs. and Remote Sens., 10, pp. 2768-2779, 2017.
. [4] Budillon, A. Johnsy, G. Schirinzi. “Extension of a fast GLRT algorithm to 5D SAR tomography of Urban areas”, Remote Sensing 9, 844, 2017.

Temporal decorrelation is known as a critical parameter for the design of spaceborne missions, even for the longest wavelengths currently operable from space or in a close future with the upcoming Biomass mission at P-band.
In this framework, tower-based campaigns have been funded by ESA and CNES in order to obtain valuable time series from which temporal decorrelation can be calculated,considering time interval from 15 min to several months.
Among key results supported by both TropiScat (2011-2013) and AfriScat (2015-2017) campaigns in French Guiana and Ghana respectively, the statistical analysis support the choice made for Biomass of helio-synchronous orbits at 6h and repeat passes shorter than 3 days.
In addition to the P-band results, both campaigns also enabled to acquire time series at L-band with the same acquisition scenarios, hence the possible comparisons between temporal decorrelation for the two frequencies. Beyond the expected stronger temporal decorrelation at L-band, the gap with P-band results is shown to be closely related to seasonal effects, and non-linearly dependent on the temporal baseline.
Supported by the physical basis of the MIPERS-4D forward model which accounts for dielectric changes and convective motion, a semi-empirical relation between temporal decorrelation at P and L bands has been derived as function of temporal baseline.
Cross-validation between both TropiScat and AfriScat sites shows 95% confidence interval lower than 0.1, which strongly support the prediction of P-band temporal decorrelation from current or upcoming zero baseline L-band missions, and paves the way for future synergies across spaceborne missions.

The soil moisture content (SMC) is one key parameter in many environmental processes. It influences hydrological and agricultural processes, runoff generation, drought development and many other processes. Furthermore, it influences the climate system through atmospheric feedbacks. Good knowledge on the spatial and temporal distribution of this parameter is of high interest for many applications and to better understand the water cycle (Beniston et al., 2012).
The sensitivity of radar to SMC is well established and is, already today, exploited in several coarse scale products like the ones based on ASCAT, or SMOS. These products are available with spatial resolutions between 25 and 50 km. In the last few years the interest towards the estimation of SMC at finer scales (< 1 km), using active microwave technologies like SAR, has grown. Especially since the launch of Sentinel-1, there is an increasing capability to monitor the same areas with a frequent revisiting time.
In this study the focus is on alpine areas where reliable measurements require high spatial and temporal resolution. For this purpose, we explore the synergetic use of SAR sensors at L-Band, C-Band, and X-Band to understand how they can work in a virtual constellation with two main objectives:
- With images acquired on the same day, at similar times, multi-frequency approaches can provide information on different part of the observed scene, e.g. C-band is more related to vegetation properties and structure, while L-band can penetrate the canopy and provide information on the underlying soil (Notarnicola & Posa, 2007).
- When acquisitions are spaced further apart, the data can be combined in a virtual constellation to increase the temporal resolution. In this case it is interesting to assess the accuracy of the different bands in retrieving the target variables (Bindlish & Barros, 2000).
To fulfill these two objectives, an algorithm that integrates multi-frequency SAR data and optical data to estimate SMC in vegetated areas has been developed. The algorithm relies on a Support Vector Regression (SVR) retrieval approach (Pasolli et. 2015, Padovano et al., 2017). The results are presented based on two case studies. The area of interested in located in South Tyrol (italy), mainly composed by pastures and meadows. It is equipped with fixed stations for soil moisture measurements. Moreover, on the days of SAR acquisitions, other field measurements were collected.
The first case study focuses on the joint use of Sentinel 1 (C-Band), Sentinel 2 (optical), and ALOS 2 (L-Band) imagery, acquired on the same day. The proposed algorithm has been tested in the following configurations:
- Sentinel 1
- ALOS 2
- Sentinel 1, ALOS 2
- Sentinel 1, ALOS 2, DEM
- Sentinel 1, Sentinel 2, ALOS 2, DEM
The preliminary results indicate that:
• L band HH and HV backscattering produce SMC estimates with the following performances R²= 0.33 – RMSE= 7% while only C-band VV and HV backscattering coefficients have slightly worse performances: R²= 0.25 – RMSE= 6%. Adding to L band information about local Incidence angle and DEM parameters (slope, height, and aspect), the performances improve notably: R²= 0.80 – RMSE= 4%. In an area with a complex topography elevation, aspect and slope information are relevant to improve SVR performances;
• L and C band combined information with topography features, in an area mainly composed by pastures and meadows, show these SVR performances: R²= 0.58 – RMSE= 6%.
• L and C band backscattering coefficients plus optical reflectance information lead to a stable and reliable model (R²=0.83 –RMSE=4%);
• A preliminary test for the retrieval of Vegetation Water Content (VWC) indicates that ALOS 2 L-Band backscattering, combined with L-Band simulated backscattering, leads to a reliable model to compute VWC, without any optical information.

For the second case-study, RADARSAT 2 (RS2) and COSMO-SkyMed (CSK) images (C-Band and X-Band, respectively), acquired on different days, have been exploited in order to understand how to properly combine them. The retrieval with only RS2 provides the best R² (0.91). The joint use of CSK (July 21) and RS2 (18 July) data slightly decreases R² (0.86), due to the combination of images from different dates. However, the MAE (1.99%) and RMSE (3.24%) are the lowest. By adding the backscatter coefficient of the CSK data of July 22 to the SVR model, the performances do not improve, probably due to two rain events, which occurred between the two CSK acquisitions. CSK alone provides slightly worse results with respect to RS2 (R²=0.86, MAE=2.95%, RMSE=3.96%)
A third case study is under study where Sentinel1, ALOS 2, and COSMO-SkyMed images are jointly exploited. If available, SAOCOM images will be exploited over selected Argentinean test sites, jointly with Sentinel1 and COSMO-SkyMed images.


References:
M. Beniston, Impacts of climatic change on water and associated economic activities in the Swiss Alps, Journal of Hydrology, 412-413, 291–296, 2012. doi:10.1016/j.jhydrol.2010.06.046

R. Bindlish and A. Barros, Multifrequency soil moisture inversion from SAR measurements with the use of IEM., Remote Sens. Environ., 2000.

C. Notarnicola, F.Posa, Inferring vegetation water content from C and L band images, IEEE Transactions on Geoscience and Remote Sensing, vol.45, no.10, October 2007, p.3165-3171.

A. Padovano, F. Greifeneder, R. Colombo, G. Cuozzo and . Notarnicola, Integration of SAR multi-frequency and optical data for the retrieval of soil moisture and vegetation water content, 4th Satellite Soil Moisture Validation and Application Workshop , 19-20 Sept. 2017, Wien (AT).

L. Pasolli, C. Notarnicola, G. Bertoldi, L. Bruzzone, R. Remelgado, F. Greifeneder, G. Niedrist, S. Della Chiesa, U. Tappeiner and M. Zebisch, Estimation of soil moisture in mountain areas using SVR technique applied to multiscale active radar images at C band, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 01/2015; 8(1):262-283.