Geohazard assessment of Mexico City's Metro system derived from SAR interferometry observations.

Land subsidence rates in Mexico City reach 500 mm/year, causing progressive damage to the city's core infrastructure, including the Metro system. A deadly overpass collapse in 2021, along a Metro line that had operated for less than 10 years, brought subsidence-related structural damage to the attention of the system's authorities and led to major repairs to two of the twelve Metro lines. Still, the need for quantifying the magnitude and extent of subsidence affecting the Metro system's widespread infrastructure prevails. Using a wealth of satellite radar interferometry observations, levelling surveys, subsurface profiles, linear gradient and differential displacement analyses, and structural-engineering parameters, we assess the vulnerability of the Metro system's street-level and elevated segments to land subsidence. Our results reveal that high subsidence velocity gradients occur over sharp transitional zones between stable and fast-subsiding areas, reaching values of 1 × 10 - 3 year - 1 , resulting in slope changes up to 3.5% over a 20-year period and differential displacements between columns. Our findings suggest locations where the consequences of subsidence have compromised the train's braking safety design, increased railway flooding hazard, produced railway bending, and reduced the conceived 50-year service life of the Metro's elevated overpasses.

Country-scale assessment of urban areas, population, and households exposed to land subsidence using Sentinel-1 InSAR, and GPS time series.

The increased need for water resources in urban sprawls and intense droughts has forced more aggressive groundwater extraction resulting in numerous urban areas undergoing land subsidence. In most cases, only some large metropolitan areas have been well-characterized for subsidence. However, there is no existing country-wide assessment of urban areas, population, and households exposed to this process. This research showcases a methodology to systematically evaluate urban localities with land subsidence higher than - 2.8 cm/year throughout Mexico. We used Interferometric Synthetic Aperture Radar (InSAR) tools with a dataset of 4611 scenes from European Space Agency's Sentinel-1 A/B SAR sensors acquired from descending orbits from September 2018 through October 2019. This dataset was processed at a supercomputer using InSAR Scientific Computing Environment and the Miami InSAR Time Series software in Python software. The quality and calibration of the resulting velocity maps are assessed through a large-scale comparison with observations from 100 continuous GPS sites throughout Mexico. Our results show that an urban area of 3797 km2, 6.9 million households, and 17% of the total population in Mexico is exposed to subsidence velocities of faster than - 2.8 cm/year, in more than 853 urban localities within 29 land subsidence regions. We also confirm previous global potential estimations of subsidence occurrence in low relief areas over unconsolidated deposits and where groundwater aquifers are under stress. The presented research demonstrates the capabilities for surveying urban areas exposed to land subsidence at a country-scale level by combining Sentinel-1 velocities with spatial national census data. Supplementary Information: The online version contains supplementary material available at 10.1007/s11069-023-06259-5.

Earthquake swarms and slow slip on a sliver fault in the Mexican subduction zone.

The Mexican subduction zone is an ideal location for studying subduction processes due to the short trench-to-coast distances that bring broad portions of the seismogenic and transition zones of the plate interface inland. Using a recently generated seismicity catalog from a local network in Oaxaca, we identified 20 swarms of earthquakes (M < 5) from 2006 to 2012. Swarms outline what appears to be a steeply dipping structure in the overriding plate, indicative of an origin other than the plate interface. This steeply dipping structure corresponds to the northern boundary of the Xolapa terrane. In addition, we observed an interesting characteristic of slow slip events (SSEs) where they showed a shift from trenchward motion toward an along-strike direction at coastal GPS sites. A majority of the swarms were found to correspond in time to the along-strike shift. We propose that swarms and SSEs are occurring on a sliver fault that allows the oblique convergence to be partitioned into trench-perpendicular motion on the subduction interface and trench-parallel motion on the sliver fault. The resistivity structure surrounding the sliver fault suggests that SSEs and swarms of earthquakes occur due to high fluid content in the fault zone. We propose that the sliver fault provides a natural pathway for buoyant fluids attempting to migrate upward after being released from the downgoing plate. Thus, sliver faults could be responsible for the downdip end of the seismogenic zone by creating drier conditions on the subduction interface trenchward of the sliver fault, promoting fast-slip seismogenic rupture behavior.