Assessing the application of GACOS atmospheric correction for DInSAR-based mining deformation monitoring by using Sentinel-1 data in Upper Silesian Coal Basin in Poland

One of the most critical factors which determine the accuracy of deformation maps provided by Differential Synthetic Aperture Radar Interferometry (DInSAR) are atmospheric artefacts. Nowadays, one of the most popular approaches to minimize atmospheric artefacts is Generic Atmospheric Correction Onli...

Full description

Bibliographic Details
Main Authors: Kamila Pawłuszek-Filipiak, Natalia Wielgocka, Tymon Lewandowski, Damian Tondaś
Format: Article
Language:English
Published: Polish Academy of Sciences 2021-11-01
Series:Geodesy and Cartography
Subjects:
Online Access:https://journals.pan.pl/Content/121470/PDF/e10-PawluszekFilipiak_etal_2021_GEOCART-00164-2021_final.pdf
Description
Summary:One of the most critical factors which determine the accuracy of deformation maps provided by Differential Synthetic Aperture Radar Interferometry (DInSAR) are atmospheric artefacts. Nowadays, one of the most popular approaches to minimize atmospheric artefacts is Generic Atmospheric Correction Online Service for InSAR (GACOS). Nevertheless, in the literature, the authors reported various effects of GACOS correction on the deformation estimates in different study areas Therefore, this paper aims to assess the effect of GACOS correction on the accuracy of DInSAR-based deformation monitoring in USCB by using Sentinel-1 data. For the accuracy evaluation, eight Global Navigation Satellite Systems (GNSS) permanent stations, as well as five low-cost GNSS receivers were utilized. GACOS-based DInSAR products were evaluated for: (1) single interferograms in different geometries; (2) cumulative deformation maps in various geometries and (3) decomposed results delivered from GACOS-based DInSAR measurements. Generally, based on the achieved results, GACOS correction had a positive effect on the accuracy of the deformation estimates in USCB by using DInSAR approach and Sentinel-1 data in each before mentioned aspect. When considering (1), it was possible to achieve Root Mean Square Error (RMSE) below 1 cm for a single interferogram for only 20% and 26% of the ascending and descending investigated interferograms, respectively when compared with GNSS measurements. The RMSE below 2 cm was achieved by 47% and 66% of the descending and ascending interferograms, respectively.
ISSN:2300-2581