Development of InSAR Neutral Atmospheric Delay Correction Model by Use of GNSS ZTD and Its Horizontal Gradient
Development of InSAR Neutral Atmospheric Delay Correction Model by Use of GNSS ZTD and Its Horizontal Gradient
复制标题
利用GNSS ZTD及其水平梯度建立InSAR中性大气延迟校正模型
DOI：
10.1109/tgrs.2022.3188988
复制
发表时间：
2022
影响因子：
8.2
通讯作者：
Kinoshita Yohei
中科院分区：
文献类型：
--
作者：
森祐紀; 外田智千; 宮本知治; 池田剛;森祐紀,久保友明,本田陸人,岩里拓弥,肥後祐司,丹下慶範;久保友明,本田陸人,後藤佑太,森悠一郎,森祐紀,岩里拓弥,肥後祐司,宮原正明;Masashi Ogiso;小木曽 仁;Masashi Ogiso;Kinoshita Yohei
干涉型合成孔径雷达（InSAR）由于微波传播延迟效应，经常受到大气干扰，使地表位移探测精度限制在厘米数量级以上。本文利用全球导航卫星系统（GNSS）天顶总延迟（ZTD）及其水平梯度数据，建立了一种新的InSAR中性大气延迟校正模型。该模型首先利用最小二乘法从GNSS ZTD和梯度观测数据中检索出规则网格化的海平面ZTD分布和线性高度依赖关系。然后，将网格化的ZTD投影到InSAR坐标上以校正中性大气延迟。通过将修正模型应用于日本关东平原的l波段先进陆地观测卫星2号(ALOS-2)/相控阵型l波段合成孔径雷达2号（PALSAR-2）扫描干涉图，对修正模型的性能进行了评估。校正结果表明，采用所提出的延迟校正后，相位标准差平均降低33.87%。通过与InSAR模式的通用大气校正在线服务（GACOS）和日本区域中尺度模式（MSM）的校正进行比较，本文提出的基于gnss的模式在我的测试案例中表现优于其他模式。灵敏度测试表明，在可用GNSS台站较少的情况下，加入延迟梯度可以提高延迟再现性。尽管所提出的校正模型的适用性取决于目标区域内可用GNSS站的数量，但所提出的模型具有有效缓解中性大气延迟和提高小幅度地表位移检测能力的潜力。
Interferometric synthetic aperture radar (InSAR) often suffers from atmospheric disturbances due to the microwave propagation delay effect, which limits the surface displacement detection accuracy to an order of centimeters or more. Here, I developed a new neutral atmospheric delay correction model for InSAR by using the global navigation satellite system (GNSS) zenith total delay (ZTD) and its horizontal gradient data. The proposed model at first retrieves the regularly gridded ZTD distribution at sea level and the linear height dependence from GNSS ZTD and gradient observations by the least-squares method. Then, the gridded ZTD is projected onto the InSAR coordinate to correct the neutral atmospheric delay. I evaluated the correction model performance by applying it to the L-band Advanced Land Observing Satellite 2 (ALOS-2)/Phased Array-type L-band Synthetic Aperture Radar 2 (PALSAR-2) ScanSAR interferograms over the Kanto plain in Japan. The correction result showed that, by applying the proposed delay correction, the phase standard deviation decreased by 33.87% on average. By comparing it with the Generic Atmospheric Correction Online Service (GACOS) for the InSAR model and the correction by the Japanese regional mesoscale model (MSM), the proposed GNSS-based model outperformed others in my test case. The sensitivity test indicated that including the delay gradient could improve delay reproducibility under situations with fewer available GNSS stations. Although the proposed correction model’s applicability depends on the number of available GNSS stations in the area of interest, the proposed model has the potential to effectively mitigate the neutral atmospheric delay and improve the detection ability for small-amplitude surface displacements.