Validation of the Intermittent SBAS (ISBAS) DInSAR algorithm over the gas reservoirs of North Holland, The Netherlands
Publication Date
July 11, 2016
Creator
Abstract
Ground motion as a consequence of natural geological processes or due to anthropogenic influences, is a considerable problem and concern to a variety of people and groups. These include the mining industry, insurance companies, government and not least the general population. Increased knowledge of subsidence patterns would benefit all parties involved; however, often techniques used such as levelling and GPS have spatial or temporal shortfalls. Utilizing remotely sensed data and Differential Interferometric Synthetic Aperture Radar (DInSAR), the shortcomings of traditional measurement can be overcome, producing precise maps over wide expanses and over many years. DInSAR utilizes stacks of Synthetic Aperture Radar (SAR) imagery of the same area to derive ground motion rates with millimetric precision. A significant limitation of these techniques, be it utilizing small baseline or persistent scatterer methods, is the density of coverage it can achieve over dynamic rural environments. Interferometric analysis is not possible over large sections of the scene which can inhibit the identification of subsidence patterns, impeding the techniques efficacy. The Intermittent Small Baseline Subset (ISBAS) is an improved version of the established SBAS algorithm, designed to improve coverage over rural settings by considering intermittent coherence. 52 ERS-2 (SAR) C-band images (1995-2000) and 75 ENVISAT (ASAR) C-band images (2003-2010) of North Holland were processed using the ISBAS technique. Up to 15 times as many point velocities were returned than when using the conventional SBAS analysis. The measurement of linear velocities in rural areas enabled a more complete identification subsidence, offering an improvement over the original SBAS technique. The dominant feature was an area of subsidence in the area to the West of Alkmaar, attributed to natural gas production with subsidence rates in the region of 3 mm/year. Linear displacement velocities were validated with respect to the first order system of levelling benchmarks which form the Dutch ordinance level. It established that ISBAS products were accurate to within 1.29 mm/year and 1.42 mm/year for the ERS and ENVISAT data sets respectively. These error budgets were comparable to results found in the Terrafirma validation project, where the same site was studied using conventional persistent scatterer approaches.
Item Type
ethesis
Thesis Type
MScRes
Supervisors
Subjects (LC)
Associated Schools / Departments
School of Geography
eprints ID
33615
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
MSc (Res) Thesis David Gee.pdf
Type
Full-text
Description
Examined
Size
10.08 MB
Format
Adobe PDF
Checksum (MD5)
dc9d07fb004b3645deb92ab5525c4db3