Eugenio Sansosti

A small-baseline approach for investigating deformations on full-resolution differential SAR interferograms

This paper presents a differential synthetic aperture radar (SAR) interferometry (DIFSAR) approach for investigating deformation phenomena on full-resolution DIFSAR interferograms. In particular, our algorithm extends the capability of the small-baseline subset (SBAS) technique that relies on small-baseline DIFSAR interferograms only and is mainly focused on investigating large-scale deformations with spatial resolutions of about 100/spl times/100 m. The proposed technique is implemented by using two different sets of data generated at low (multilook data) and full (single-look data) spatial resolution, respectively. The former is used to identify and estimate, via the conventional SBAS technique, large spatial scale deformation patterns, topographic errors in the available digital elevation model, and possible atmospheric phase artifacts; the latter allows us to detect, on the full-resolution residual phase components, structures highly coherent over time (buildings, rocks, lava, structures, etc.), as well as their height and displacements. In particular, the estimation of the temporal evolution of these local deformations is easily implemented by applying the singular value decomposition technique. The proposed algorithm has been tested with data acquired by the European Remote Sensing satellites relative to the Campania area (Italy) and validated by using geodetic measurements.

Geometrical SAR image registration

Accurate subpixel registration of synthetic aperture radar (SAR) images is an issue that is again growing interest since its initial developments related to two-pass interferometry. Recent progress in coherent (multichannel) SAR processing raises the need for accurate registration of data takes acquired with large baseline spans, high temporal coverage, and with different frequency and/or operational modes. In this paper, we discuss a SAR image-registration procedure, based on the use of external measures which allows obtaining a very accurate alignment of SAR images. The presented technique makes use of a digital elevation model and of the precise information about the acquisition flight tracks, to compute the warping functions that map the position of each pixel in the different takes, thus avoiding any approximation. The resulting algorithm is simple, robust, precise, and very efficient; as a matter of fact, it may achieve high accuracy even in critical areas, such as steep topography regions. Moreover, the availability of an analytical and exact model allows performing a detailed sensitivity analysis that can be useful in evaluating the applicability of this technique even to future high-precision satellite systems. Extensive testing, carried out on several real European Remote Sensing and ENVISAT datasets, clearly shows the effectiveness of such algorithm in registering critical SAR images

Modeling surface deformation observed with synthetic aperture radar interferometry at Campi Flegrei caldera

Satellite radar interferometry of Campi Flegrei caldera, Italy, reveals a pattern of subsidence during the period 1993–1998. Interferograms spanning the first half of the observation period (1993–1995) have a lower amplitude and average rate of subsidence than those spanning either the second half (1995–1998) or the entire period (1993–1998), consistent with observations of a slowing down or reversal of subsidence during the first half of the observation period. We calculate a time series of deformation images relative to a reference image on the basis of a least squares inversion. During the observation period the maximum subsidence progresses at a rate of roughly 38 ± 2 mm/yr, with periods of no apparent subsidence in late 1996 to early 1997. To understand the characteristics of the source, we jointly invert pairs of ascending and descending differential interferograms spanning similar time intervals (first half, second half, or entire interval) of the period 1993–1998. In each case the joint inversion fits the two unwrapped interferograms with a similar subhorizontal rectangular contracting tensile dislocation striking roughly N98°E with dimensions ? 4 × 2 km and located beneath the city of Pozzuoli at a depth of 2.5–3 km. Inversion for a spheroidal or Mogi point source also produced reasonable fits but with progressively poorer overall fits to the data, respectively. Our inversion assuming a simple source in an elastic half?space does not include the possible effects of local structure on the surface deformation, a factor that may also reduce the need for an asymmetric source. The solution we find is consistent with other studies that suggest subsidence due to hydrothermal diffusion as the primary deformation mechanism during this phase of caldera deflation.

Spotlight SAR data focusing based on a two-step processing approach

The authors present a new spotlight SAR data-focusing algorithm based on a two-step processing strategy that combines the advantages of two commonly adopted processing approaches: the efficiency of SPECAN algorithms and the precision of stripmap focusing techniques. The first step of the proposed algorithm implements a linear and space-invariant azimuth filtering that is carried out via a deramping-based technique representing a simplified version of the SPECAN approach. This operation allows the authors to perform a bulk azimuth raw data compression and to achieve a pixel spacing smaller than (or equal to) the expected azimuth resolution of the fully focused image. Thus, the azimuth spectral folding phenomenon, typically affecting the spotlight data, is overcome, and the space-variant characteristics of the stripmap system transfer function are preserved. Accordingly, the residual and precise focusing of the SAR data is achieved by applying a conventional stripmap processing procedure requiring a minor modification and implemented in the frequency domain. The extension of the proposed technique to the case of high bandwidth transmitted chirp signals is also discussed. Experiments carried out on real and simulated data confirm the validity of the presented approach, which is mainly focused on spaceborne systems.

On the generation of ERS/ENVISAT DInSAR time-series via the SBAS technique

We exploit the small baseline subset (SBAS) algorithm for generating deformation time-series from SAR data acquired by sensors with different characteristics but with the same illumination geometry. In particular, our approach is focused on the use of European Remote Sensing (ERS) and ENVISAT satellite data, the latter acquired by the Advanced Synthetic Aperture Radar sensor on the IS2 swath. The proposed solution is oriented to investigate large-scale displacements with a relatively low spatial resolution (about 100/spl times/100 m) and implements an easy but effective combination of ERS and ENVISAT multilook interferograms which benefits of the temporal overlap between the acquisitions of the two sensors. Moreover, the algorithm does not rely on specific hypothesis on the spatial or temporal characteristics of the investigated deformations. Presented results, achieved on a synthetic aperture radar dataset relevant to the Napoli city area (Italy), confirm the validity of the approach.

Actively growing anticlines beneath catania from the distal motion of Mount Etna's Decollement measured by SAR interferometry and GPS

We used SAR interferometry and GPS to measure the rise of an anticline beneath the urban area of Catania (Italy), which originates from outward thrusting above the basal decollement of Etna volcano. The anticline grows at a rate of over 0.015 m a−1 of SE thrusting and about 0.007 m a−1 of relative uplift. By relating this growth to the simultaneous extension measured at the summit of the volcano, we demonstrate the occurrence of active volcanic spreading. This process may exert a control on eruptions, earthquakes, sector collapses, and landslides.

Deformation and eruptions at Mt. Etna (Italy): A lesson from 15 years of observations

This work was partly funded by INGV and the Italian DPC and was supported by ASI, the Preview Project and CRdC-AMRA. DPC-INGV Flank project providing the funds for the publication fees.

The September 26, 1997 Colfiorito, Italy, earthquakes: Modeled coseismic surface displacement from SAR interferometry and GPS

The largest events of the 1997 Umbria-Marche seismic sequence were the two September 26 earthquakes of Mw = 5.7 (00:33 GMT) and Mw = 6.0 (09:40 GMT), which caused severe damage and ground cracks in a wide area around the epicenters. We created an ERS-SAR differential interferogram, where nine fringes are visible in and around the Colfiorito basin, corresponding to 25 cm of coseismic surface displacement. GPS data show a maximum horizontal displacement of (14±1.8) cm and a maximum subsidence of (24±3) cm. We used these geodetic data and the seismological parameters to estimate geometry and slip distribution on the fault planes. Modeled fault depths and maximum slip amplitudes are 6.5 km and 47 cm for the first event and 7 km and 72 cm for the second one, in good agreement with those derived from the seismological data.

The use of IFSAR and classical geodetic techniques for caldera unrest episodes: application to the Campi Flegrei uplift event of 2000

Abstract Campi Flegrei caldera has a long history of large surface deformation, with displacements of several meters in the 1970s and early 1980s. Its location within a densely populated urban area underscores the importance of understanding the relationship between large and episodic deformation events and their source mechanisms. The primary observable of the caldera’s activity is its surface deformation. Classical geodetic approaches such as leveling, have been complemented by the more advanced measurements of the Global Positioning System (GPS) and Synthetic Aperture Radar (SAR) interferometry. In this work we focus on the Campi Flegrei caldera uplift event that occurred from early spring to late summer 2000. Our goal is to highlight the potential to integrate interferometric SAR (IFSAR), GPS, and classical leveling data for ground deformation studies and source modeling. We compare models for the deformation source constrained by inversion of the differential IFSAR data (DIFSAR) with the model’s prediction for the GPS and leveling data. Resolution of possible changes in the source mechanism for Campi Flegrei caldera are limited by differences in the temporal, spatial and deformation component strengths of each data set. In the future, overcoming these data deficiencies will be important for resolving the dynamics of volcano systems and for volcanic hazard mitigation.

Modeling coseismic displacements resulting from SAR interferometry and GPS measurements during the 1997 Umbria-Marche seismic sequence

In this study we analyse coseismic GPS displacements and DInSAR data to constrain a dislocation model for the three largest earthquakes of the 1997 Umbria-Marche seismic sequence. The first two events, which occurred on September 26 at 00:33 GMT (Mw 5.7) and 09:40 GMT (Mw 6.0) respectively, are investigated using both GPS displacements and DInSAR interferograms. We discuss and compare the results of previous studies which separately modeled a smaller subset of geodetic data. We provide a dislocation model for these two earthquakes which fits well both GPS and DInSAR data and agrees with the results of seismological and geological investigations. The first event consists of a unilateral rupture towards the southeast with a uniform dislocation. The strike, rake and dip angles are those resulting from the CMT solution. The second event consists of an unilateral rupture towards the northwest and a variable slip distribution on the fault plane. The strike and the rake are consistent with the CMT solution, but the dip angle has been slightly modified to improve the simultaneous fit of GPS and DInSAR data. While the second rupture (09:40 GMT) arrived very close to the surface, the fit to geodetic data shows that the first rupture (00:33 GMT) is deeper (2 km), despite the more evident surface geological effects. The analysis of new SAR interferograms allows the identification of a 5–6 cm additional displacement caused by the October 3 (Mw 5.2) and 6 (Mw 5.4) seismic events.We use data from a new DInSAR interferogram to model the displacement field of the Sellano earthquake of October 14, 1997. For this event significant GPS measurements were not available. We tested two different fault plane geometries: a blind, planar fault (top depth = 2.4 km), and a curved (listric) fault reaching the surface. The two models provide a generally similar fit to the data, and show that most of the slip was released at depths greater than 2.4 km along a gently dipping (40°–45°) fault surface. They also show that a unilateral rupture does not allow fitting the interferometric fringes since there is evident surface deformation to the northwest of the hypocenter. Moreover, we suggest that the concentration of high residuals in the southern part of our uniform slip model may in fact indicate a certain slip variability in this area.We conclude that, despite the moderate magnitudes and the lack of significant surface faulting, the space geodetic data allowed to constrain dislocation models giving new insights in the rupture process of the three largest events of the sequence.