永久散射体雷达干涉(PSI)已被广泛应用于城市缓慢沉降监测,但针对显著性地表沉降监测及PSI适用性的探讨较少。本文选取天津市与廊坊市城郊结合部的典型工业化城镇为研究区,分别以2007至2010年间的23幅L波段(23.6 cm) ALOS PALSAR影像和2007至2009年间的23幅C波段(5.6 cm) ENVISAT ASAR影像为数据源,使用基于相位空间相关性分析的PSI方法进行时空沉降提取,并对两个平台的结果进行对比分析和交叉验证。在此基础上,结合长短波SAR对沉降敏感性不同的特征探讨该PSI方法应用于显著沉降解算的可靠性,并结合地质条件及地下水开采信息对实验区沉降的时空分布进行深入分析和解释。研究表明,两平台的沉降结果十分吻合,在沉降漏斗区二者均方根误差(RMSE)为6.5 mm/a;实验区地表沉降呈现出显著的非均匀性,最大沉降速率超过200 mm/a,造成严重沉降的主要原因为地下水的大量开采;证实了PSI方法在显著地表沉降监测方面的有效性和可靠性。

Persistent Scatterer Interferometry(PSI) is widely used to monitor the slowly accumulated land subsidence in urban areas. However, a few explorations concentrated on monitoring the significant subsidence by PSI and its applicability in such areas. In this study, typical industrial towns(subsidence troughs) located around the boundary between Tianjin and Langfang cities were chosen as the study areas for subsidence detection and analysis with multiband Synthetic Aperture Radar(SAR) images to evaluate the potential and applicability of PSI to monitor significant ground subsidence. Land subsidence was extracted using PSI through phase spatial correlation analysis and three dimensional phase-unwrapping approaches. In total, 23 L-band(wavelength is 23.6 cm) Phased Array type L-band Synthetic Aperture Radar(PALSAR) images acquired between 2007 and 2010, and 23 C-band(wavelength of 5.6 cm) Advanced Synthetic Aperture Radar(ASAR) images acquired between 2007 and 2009 were selected for multiband data processing. The results derived from the two types of SAR images were comprehensively and mutually compared and validated. In addition, the reliability of the PSI method for significant subsidence detection was explored by considering the different sensitivities to subsidence of the PALSAR and ASAR systems. The spatiotemporal distribution of the subsidence in the study area was also analyzed with respect to the geological settings and groundwater exploitation information. The derived subsidence in the study area presents a significant inhomogeneous pattern, and the subsidence trend derived from the two types of SAR images ranges between 0 and 210 mm/a. The root mean squared error is 6.5 mm/a between the two types of PSI subsidence results in the subsidence bowl area. Further exploitation indicates that such significant subsidence is mainly caused by excessive exploitation of groundwater. Moreover, the subsidence magnitude is highly related to different land-use categories, such as factories, residential quarters, railways, highways, and farmland, indicating anisotropy of groundwater usage in these areas. Comparison analysis demonstrates that the PALSAR and ASAR PSI results are in good agreement and are consistent with previous studies, although the two data sets have different spatial and temporal resolutions, wavelengths, incidence angles, and time spans. The above results and analysis indicate that the PSI method used in our experiment is applicable to different SAR systems. The PSI method shows good reliability and applicability to monitor long-term significant ground subsidence in areas with sparse PS distributions. Therefore, the PSI method used in this study can provide scientific and effective technical assistance for ground subsidence monitoring in such areas. The results(subsidence rate and time series) of the PSI method can serve as assistant information when constituting measures to control groundwater exploitation.