A Framework for Correcting Ionospheric Artifacts and Atmospheric Effects to Generate High Accuracy InSAR DEM

Zhiwei Liu,Haiqiang Fu,Tingying Zuo,Jianjun Zhu,Cui Zhou

doi:10.3390/rs12020318

Abstract

Repeat-pass interferometric synthetic aperture radar is a well-established technology for generating digital elevation models (DEMs). However, the interferogram usually has ionospheric and atmospheric effects, which reduces the DEM accuracy. In this paper, by introducing dual-polarization interferograms, a new approach is proposed to mitigate the ionospheric and atmospheric errors of the interferometric synthetic aperture radar (InSAR) data. The proposed method consists of two parts. First, the range split-spectrum method is applied to compensate for the ionospheric artifacts. Then, a multiresolution correlation analysis between dual-polarization InSAR interferograms is employed to remove the identical atmospheric phases, since the atmospheric delay is independent of SAR polarizations. The corrected interferogram can be used for DEM extraction. Validation experiments, using the ALOS-1 PALSAR interferometric pairs covering the study areas in Hawaii and Lebanon of the U.S.A., show that the proposed method can effectively reduce the ionospheric artifacts and atmospheric effects, and improve the accuracy of the InSAR-derived DEMs by 64.9% and 31.7% for the study sites in Hawaii and Lebanon of the U.S.A., respectively, compared with traditional correction methods. In addition, the assessment of the resulting DEMs also includes comparisons with the high-precision Ice, Cloud, and land Elevation Satellite-2 (ICESat-2) altimetry data. The results show that the selection of reference data will not affect the validation results.

Highlights

The interferometric synthetic aperture radar (InSAR) technology is a powerful remote sensing tool for generating digital elevation models (DEMs), owing to its day-and-night” and all-weather image acquisition capability, as well as its fine spatial resolution [1]
The results show that the elevation difference of the proposed approach is significantly smaller than those of the other two methods, indicating that the proposed method can effectively remove the ionospheric artifacts and atmospheric effects on the interferograms, and can effectively improve the accuracy of InSAR topographic mapping
This could have resulted from many reasons: (1) besides the errors addressed in this paper, geometrical distortions can significantly reduce the DEM accuracy, which has not been considered by the proposed method; (2) the ICESat-2 point distribution is too sparse to describe the DEM error well in spatial

Summary

Introduction

The interferometric synthetic aperture radar (InSAR) technology is a powerful remote sensing tool for generating digital elevation models (DEMs), owing to its day-and-night” and all-weather image acquisition capability, as well as its fine spatial resolution [1]. The various noise sources inherent in InSAR technology, such as ionospheric artifacts and atmospheric effects, greatly reduce the accuracy of topographic mapping. For the repeat-pass InSAR systems, the atmospheric effect is the dominant noise source. A spatial and temporal change of 20% in the relative humidity of the troposphere can result in an elevation error of about 100 m [2]. Mitigating the ionospheric artifacts and atmospheric effects that are confused with the topographic phase is a great challenge in high-precision InSAR topographic mapping

Objectives

Methods

Discussion

Conclusion