Abstract

Abstract. Recently, high-resolution commercial SAR satellites with several meters of resolutions are widely utilized for various applications and disaster monitoring is one of the commonly applied areas. The information about the flooding situation and ground displacement was rapidly announced to the public after the Great East Japan Earthquake 2011. One of the studies reported the displacement in Tohoku region by the pixel matching methodology using both pre- and post- event TerraSAR-X data, and the validated accuracy was about 30 cm at the GEONET reference points. In order to discuss the spatial distribution of the displacement, we need to evaluate the relative accuracy of the displacement in addition to the absolute accuracy. In the previous studies, our study team evaluated the absolute 2D geo-location accuracy of the TerraSAR-X ortho-rectified EEC product for both flat and mountain areas. Therefore, the purpose of the current study was to evaluate the spatial and temporal relative geo-location accuracies of the product by considering the displacement of the fixed point as the relative geo-location accuracy. Firstly, by utilizing TerraSAR-X StripMap dataset, the pixel matching method for estimating the displacement with sub-pixel level was developed. Secondly, the validity of the method was confirmed by comparing with GEONET data. We confirmed that the accuracy of the displacement for X and Y direction was in agreement with the previous studies. Subsequently, the methodology was applied to 20 pairs of data set for areas of Tokyo Ota-ku and Kawasaki-shi, and the displacement of each pair was evaluated. It was revealed that the time series displacement rate had the seasonal trend and seemed to be related to atmospheric delay.

Highlights

  • The high-resolution Synthetic Aperture Radar (SAR) satellites with several meters resolution for the commercial usages are in operation for the past several years, and the information that we obtain from the images are different due to the specification of the sensors

  • This study evaluated the relative geo-location accuracy of the TerraSAR-X ortho-rectified Ellipsoid Corrected (EEC) product by the pixel matching methodology using the pairs of intensity images

  • According to the analysis using 20 pairs of images, it was revealed that the spatial deviation of the displacement was less than 10 cm for both X and Y direction, and the temporal variation was larger than spatial one

Read more

Summary

Introduction

The high-resolution Synthetic Aperture Radar (SAR) satellites with several meters resolution for the commercial usages are in operation for the past several years, and the information that we obtain from the images are different due to the specification of the sensors. For example, we can select various wavelength sensors, such as TerraSAR-X (Xband SAR with 3.1 cm wavelength), Radarsat-2 (C-band SAR with 5.6 cm wavelength), and ALOS PALSAR-2 (L-band SAR with 23.6 cm wavelength). Some studies combined those different types of sensors to obtain more variable information (Baghdadi et al, 2009; Dong et al, 2011). InSAR may not always be able to observe deformation depending on vegetation

Objectives
Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call