Intermittent Small Baseline Subset (ISBAS) monitoring of land covers unfavourable for conventional C-band InSAR: proof-of-concept for peatland environments in North Wales, UK

Abstract

This paper provides a proof-of-concept for the use of the new Intermittent Small Baseline Subset (ISBAS) approach to study ground elevation changes in areas of peat and organic soils in north Wales, which are generally, unfavourable for conventional C-band interferometric applications. A stack of 53 ERS-1/2 C-band SAR scenes acquired between 1993 and 2000 in descending mode was processed with both the standard low-pass SBAS method and ISBAS. The latter revealed exceptional improvements in the coverage of ground motion solutions with respect to the standard approach. The number of identified coherent and intermittently coherent pixels increased by a factor of 26 with respect to the SBAS solution, and extended the coverage of results across unfavourable land covers, particularly for coniferous woodland, bog, acid grassland and heather. The greatest increase was achieved over coniferous woodland, which showed ISBAS/SBAS pixel density ratios above 300. Despite the intermittent nature of the ISBAS solutions, ISBAS provided velocity standard errors generally below 1-1.5 mm/yr, thus preserving good quality of the estimated ground motion rates.

Mexico City land subsidence in 2014–2015 with Sentinel-1 IW TOPS: Results using the Intermittent SBAS (ISBAS) technique

“…The respective standard errors fall between 0.24 cm/year and 3.24 cm/year, with an average of 0.94 cm/year. As expected, smaller standard errors are observed in urban compared to rural areas as supported by the cross-section profiles in figure 10e and 10f, with most of the values being less than 0.6 cm/year due to the high number of interferograms used per pixel (Cigna et al, 2014). Conversely, higher standard errors are found in non-urban areas, for instance north of Ciudad Nezahualcóyotl and in Chalco, where standard errors are 0.6 to 1.2 cm/year.…”

“…The linear ISBAS algorithm (Sowter et al, 2013;Bateson et al, 2015;Cigna et al, 2014) is very similar in structure to the linear SBAS algorithm outlined by Berardino et al (2002). The premise is that, from a stack of many SAR observations, a large number of low-resolution two-pass differential interferograms are generated, satisfying a threshold that primarily relates to a low orbital baseline.…”

“…To select the most suitable pixels for the time-series analysis and for the calculation of the LOS velocity of each pixel, we utilised the ISBAS algorithm described in Sowter et al (2013), Bateson et al (2015) and Cigna et al (2014), using 50 as the number of best coherence interferograms threshold and 0.25 as the coherence threshold. The first threshold is an indicator of the minimum number of interferograms that may be used to calculate the velocity and the coherence threshold is used to determine over which pixels velocity may be calculated.…”

“…A recent paper by Osmanoǧlu et al (2015) compared the PSI, SBAS, SqueeSAR TM (Ferretti et al, 2011) and StaMPS (Hooper et al, 2004;Hooper, 2008) techniques over the Mexico City area and also concluded that none of the methods were able to obtain deformation rates over agricultural farmlands and natural vegetation. A more recent attempt to extend a DInSAR analysis into rural classes is the Intermittent SBAS (ISBAS) (Sowter et al, 2013;Bateson et al, 2015) technique which appears particularly well-suited to low-resolution, wide-area deformation monitoring over a broad range of land classes, including grasslands, agricultural and forested cover (Cigna et al, 2014).…”

On the Use of the ISBAS Acronym in InSAR Applications. Comment on Vajedian, S.; Motagh, M.; Nilfouroushan, F. StaMPS Improvement for Deformation Analysis in Mountainous Regions: Implications for the Damavand Volcano and Mosha Fault in Alborz. Remote Sens. 2015, 7, 8323–8347

“…We offer a wide selection of references below [2][3][4][5][6][7][8], each of which clearly identifies the ISBAS acronym as Intermittent SBAS. The first peer-reviewed journal [2] in this instance was in 2013 and so the use of the acronym certainly predates the Vajedian et al reference by at least two years.…”