Proof-of-concept for monitoring ground displacements in Tompkins County, NY using Persistent Scatterer Interferometric Synthetic Aperture Radar from the TerraSAR-X and Sentinel-1 satellites

Underground mining and the pumping of fluids, such as the proposed Cornell University Earth Source Heat Project (ESH), can result in observable displacement of the Earth’s surface that we can use to better understand the effects of those subsurface activities. Such surface movements can be monitored by ground surveying, but the process is labor intensive, limited in spatial extent, and potentially expensive. Here we test whether the established satellite monitoring of surface movements called Interferometric Synthetic Aperture Radar (InSAR) can be used in Tompkins County, NY as part of the ESH project with the goal of achieving a precision of a few mm/year over the areas of interest. We used data from two types of satellites: the TerraSAR-X and TanDEM-X (TSX) satellites of the German Space Agency ( X-band, 3.1 cm radar wavelength) and the Sentinel-1 (S1) satellites of the European Space Agency (C-band, 5.6 cm radar wavelength). We find that both data can be used to detect sub-centimeter/yr deformation rates using Persistent Scatterer Interferometry (PSInSAR). We assess the precision of the inferred rates through comparisons with limited ground survey data and between satellites. PSInSAR selects only reliable pixels, aka persistent scatterers (PS), to be analyzed at the full spatial resolution of the data. Generally, man-made objects, buildings, pipes, roads, etc., are persistent scatterers whereas vegetation cover, fields, bare soil are excluded from further analysis. An analysis with snow-/rain-free TSX data showed that while removing snow covered and rainy images increases the PS population up to two times, the points are concentrated at locations that already have denser PS points. Further, snow-/rain-free images estimate almost the same deformation behavior as the full-stack data set does. In an area of known ground subsidence above an underground mine in Lansing, NY, our analysis revealed that TSX provides more PS points compared to S1. Although both datasets show inter-annual deformation rates that agree with the in situ observations, S1 possessed a higher noise level. With this lower precision level, S1 can be a reliable monitoring tool in the ESH area if the expected deformation is larger than 4-5 mm/yr and if the deforming area extends to at least 300 meters around the drilling site. Otherwise, TSX data should be considered for ground surface monitoring in the area. Based on our comparison with ground control points, we can expect to measure deformation rates of 1-2 mm/yr with TSX PSInSAR.

Bibliographic Details

Main Authors:	Molan, Yusuf E., Pritchard, Matthew E., Lohman, Rowena B.
Format:	technical report biblioteca
Language:	en_US
Published:	2022-04-27
Online Access:	https://hdl.handle.net/1813/111250
Tags:	Add Tag No Tags, Be the first to tag this record!