An Improved Multi-Sensor MTI Time-Series Fusion Method to Monitor the Subsidence of Beijing Subway Network during the Past 15 Years

Abstract

<jats:p>Land subsidence threatens the stable operation of urban rail transit, including subways. Obtaining deformation information during the entire life-cycle of a subway becomes a necessary means to guarantee urban safety. Restricted by sensor life and cost, the single-sensor Multi-temporal Interferometric Synthetic Aperture Radar (MTI) technology has been unable to meet the needs of long-term sequence, high-resolution deformation monitoring, especially of linear objects. The multi-sensor MTI time-series fusion (MMTI-TSF) techniques has been proposed to solve this problem, but rarely mentioned. In this paper, an improved MMTI-TSF is systematically explained and its limitations are discussed. Taking the Beijing Subway Network (BSN) as a case study, through cross-validation and timing verification, we find that the improved MMTI-TSF results have higher accuracy (R2 of 98% and, Root Mean Squared Error (RMSE) of 4mm), and compared with 38 leveling points, the fitting effect of the time series is good. We analyzed the characteristics of deformation along the BSN over a 15-year periods. The results suggest that there is a higher risk of instability in the eastern section of Beijing Subway Line 6 (L6). The land subsidence characteristics along the subway lines are related to its position from the subsidence center, and the edge of the subsidence center presents a segmented feature.</jats:p>

A Clustering Approach for the Analysis of InSAR Time Series: Application to the Bandung Basin (Indonesia)

“…For this reason, the authors concluded that industrial water usage from confined aquifers was likely the primary factor contributing to subsidence in the urban area of Bandung [44]. This inference is further corroborated by the fact that only the deep wells used for industrial purposes tap into the groundwater resources of the confined aquifer underlying Bandung [66]. The lower rates of subsidence in clusters 19 and 36 (Figure 11a) could reflect a difference in the type of groundwater usage.…”

“…Thus, it is possible that the lower but constant subsidence rates reflects the lack of industrial water extraction in these areas and instead could be representative of residential extraction, which coincides with the findings of [67] who investigated the subsidence phenomenon in Bandung at different spatial scales. In Bandung, extraction from the shallow aquifer for household purposes is entirely permissible and unconstrained Unfortunately, there is no monitoring program in place to track the amount of domestic groundwater extracted [66], and therefore it is not currently feasible to quantitatively assess the role that residential shallow groundwater extraction has on the subsidence in these areas. Clusters 18 and 20 exhibit a strong, linear subsidence pattern (Figure 11a).…”

“…In Bandung, extraction from the shallow aquifer for household purposes is entirely permissible and unconstrained [66]. Unfortunately, there is no monitoring program in place to track the amount of domestic groundwater extracted [66], and therefore it is not currently feasible to quantitatively assess the role that residential shallow groundwater extraction has on the subsidence in these areas.…”

Spatiotemporal heterogeneity of land subsidence in Beijing

“…As a result, Beijing becomesone of the most serious cities that suffer from land subsidence in China. The Dongbalizhuang-Dajiaoting subsidence center, the earliest-formedsubsidence centers in Beijing, experienced a subsidence rate of 124 mm/yr during the period from 2003 to 2010 19 . It is reported that wall cracks and pipeline rupturesinduced by uneven land subsidence have been found in Beijing 20 .…”

Flood Risk Assessment of Subway Systems in Metropolitan Areas under Land Subsidence Scenario: A Case Study of Beijing

“…Previous studies focused on monitoring the subsidence of subway lines, but neglected its impact on flood risk. For example, Duan et al monitored the Beijing subway network's subsidence for the past 15 years using an improved multi-sensor MTI (multi-temporal interferometric synthetic aperture radar) time-series fusion method [37]. Hu et al found that the spatial pattern of land subsidence was controlled by geological faults based on Sentinel-1 InSAR observations [38].…”