Valentin Poncos

InSAR monitoring of landslides on permafrost terrain in canada

In this study we used differential InSAR techniques to monitor landslide slide and permafrost activity at a site along the Mackenzie Valley Pipeline Corridor. Our results that motion are about 3 times more on exposed burnt slopes than the adjacent areas. The maximum activity is in September, probability related to gradual accumulated increases in soil temperatures.

InSAR Monitoring of Post-Landslide Activity

In this study we used differential InSAR techniques to monitor current post slide activity at several landslides along transportation and energy corridors. The landslide materials vary from rock debris, glacial till to permafrost alluvium. Our results show that motion is triggered by spring melt and heavy rainfall events. In the northern Mackenzie Valley pipeline corridor seasonal landslide activity is related to permafrost melt during warm summer months.

InSAR Monitoring of Permafrost Activity in the Lower Mackenzie Valley, Canada

The region of interest is located approximately in the downstream part of the Mackenzie River, Canada. Our study focuses on guidelines to process RADARSAT-1 interferometric C-band data in a permafrost environment and monitor permafrost activity and landslide motion over 18 month period. From the interferometric method (D-InSAR) on ~100 interferograms, the main conclusions are (1) a high resolution Digital Elevation Models (DEM) and a shorter satellite revisit time intervals are essential to ensure InSAR processing accuracy for small deformations associated with permafrost activity; (2) the permafrost has sustained several changes over a year of monitoring (thaw starting during end of April-mid of May, and freezing around mid October). The deformation peak is reached in the summer time between mid of May and July (up to 15 plusmn 2 mm). This method will provide a guideline for In SAR monitoring of similar areas in permafrost environment.

Point target interferometry for natural and artificial scatterers

The Coherent Targets Monitoring technique is providing superior ground deformation mapping compared with standard interferometry based on one pair of Master/Slave scenes. This is because using a larger data set it is possible to estimate and correct for additional phase error sources. Also the point targets detected as coherent targets are generally characterized by stronger signal that provides more accurate phase information than the clutter present in the rest of the scene. This paper reviews previous work in SAR noise estimation and shows the advantages of point targets versus distributed targets. Results from the Turtle Mountain/Frank Slide site show that the measured phase/deformation errors at the point target positions are within the estimated range. Corner reflectors with a high SNR are expected to improve the accuracy of the method.

Detection of flooded vegetation and measurements of water level changes using Radarsat-2

A specific double-bounce radar backscattering mechanism observed in C-band Radarsat-2 data is used to measure water level changes in areas of wetland and inundated vegetation. The backscattered SAR signal from the water - vegetation double bounce mechanism is usually much stronger than the backscattered signal from the double bounce from dry surfaces. This is because the smooth water level reflects most of the signal along a preferential direction towards the vegetation canopy, as opposed to the dry ground which scatters the signal in multiple direction. As an effect, the SAR amplitude is stronger than the one related to dry vegetation and the interferometric phase maintains coherence in time, making possible accurate phase measurements related to water level changes.

Inundation extent monitoring with smap data for carbon studies

The inundation extent is derived using brightness temperature data acquired by the L-band Soil Moisture Active Passive (SMAP) satellite, to support boreal carbon studies. Exploiting the L-band capabilities to penetrate clouds and vegetation and SMAPs 3-day revisit, the product may complement high-spatial resolution optical products in the high latitudes. The quality of the inundation extent is assessed by comparing with the following data sets: 3-m resolution maps derived using Radarsat synthetic aperture radar (SAR) data in northern Canada and multi-sensor climatology over Siberia. Initial results show encouraging comparisons. SMAP describes the seasonality of inundation more realistically compared with the climatology.

First Results with a C-band Ground Based Synthetic Aperture Radar

The paper presents the first results obtained with a C band ground based synthetic aperture radar developed by the authors. The synthetic aperture is formed by automatically moving the radar along a 3 meter long rail and scanning the scene at constant time intervals. Range resolution measurements for close range (meters) corner reflectors and the first synthetic aperture radar image obtained are presented.

Urban area monitoring by sequential patterns extracted from multi-temporal satellite data using connectivity measures

The temporal evolution of pixel values in Satellite Image Time Series (SITS) is considered criterion for the characterization, discrimination and identification of terrestrial objects and phenomena. Due to the exponential behavior of sequences number with specialization, Sequential Data Mining techniques need to be applied. The spatial aspect of the data was taken into account by the introduction of connectivity measures that characterize the pixels tendency to form objects. The conjunction of corresponding Connectivity Constraints (CC) with the Support Constraint (SC) leads to the extraction of Grouped Frequent Sequential Patterns (GFSP), a concept with proved capability for preliminary description and localization of terrestrial events. This work is focused on efficient SITS extraction of evolutions that fulfil SC and CC. Experiments performed on Bucharest urban interferometric SITS are used to illustrate the potential of the approach to find interesting evolution patterns.

SAR for surface water monitoring and public health

A specific double-bounce radar backscattering mechanism is used to measure water level changes in areas of wetland or inundated vegetation. The backscattered SAR signal from the water-vegetation double bounce mechanism is usually much stronger than the backscattered signal from the double bounce from dry surfaces. As an effect, the SAR amplitude from inundated vegetation is stronger than the one related to dry vegetation and the interferometric phase maintains coherence in time, making possible accurate phase measurements related to water level changes.

Land cover change detection using unsupervised kernel C-means and multi-temporal SAR data

Land covers and uses are dynamically being changed over the time. Detection and identification of these changes is necessary and is the first step of any study or planning for natural resource management. Synthetic Aperture Radar (SAR) imagery, thanks to its independence to weather conditions and sun illumination, is a powerful tool for these studies. In this research an unsupervised change detection framework based on the kernel-based clustering technique is presented. Kernel C-means algorithm is employed to separate the changes classes from the no-changes. This method is a non-linear algorithm which considers the contextual information. Using the kernel functions, the projecting of the data into a higher dimensional space helps to make the non-linear features more separable in a linear space. The proposed methodology has applied to dual-pol L-band SAR images acquired by the ALOS from Urmia Lake. Results show because of non-linear behavior of changed phenomenon, the algorithm leads to more reliable results.