Ongoing depletion of Iran’s groundwater, driven by human extraction, has contributed to 108 incidences of basin-scale land-surface subsidence covering 29,600 km² (>10 mm/yr, 1.8 %) of the country, 75 % of which correlates with agriculture. We find Karaj city, neighbouring Iran’s capital Tehran, is exposed to the steepest surface velocity gradients (angular distortion, β) caused by differential subsidence rates, with 23,000 people exposed to ‘high’ subsidence induced hazard. We further use these velocity gradients to aid identification of structural and geological controls on surface velocities of seven of Iran’s most populated cities, identifying potentially unmapped tectonic faults. We demonstrate that most of Iran’s subsidence is permanent (inelastic), with the spatial pattern of the proportion of inelastic deformation potentially depending on geology. During a recent, severe regional drought (2020–2023) we demonstrate the control of precipitation on the elastic, recoverable subsidence deformation magnitude with the elastic to inelastic deformation ratio falling from 41–44 % pre-drought to 31–36 % post-drought. We use automatically processed short baseline networks of Sentinel-1 Interferometric Synthetic Aperture Radar (InSAR) data, 2014–2022, to generate and estimate these ground displacements through time. We correct for atmospheric noise using weather model data and perform time series analysis in the satellite line-of-sight direction, serving this data through an open-access online portal. For each subsidence region, we decompose line-of-sight velocities into 100 m resolution vertical and horizontal (east-west) surface velocity fields. We use temporal Independent Component Analysis to constrain automatically and manually the inelastic and elastic components of subsidence, respectively.

We report the first detection of unrest at Socompa, Northern Chile, a stratovolcano which has recorded no eruptions since ~7,200 years ago. We measure deformation at and around Socompa using Interferometric Synthetic Aperture Radar (InSAR) observations between Jan 2018 and Oct 2021. We find that, whilst initially inactive, Socompa shows a steady uplift (17.5 mm/yr) from Dec 2019, independently recorded by near-field continuous Global Positioning System (GPS) data. The data can be fit with pressure increase in an ellipsoidal source region stretching from 1.9 to 9.5 km, with a volume change rate of ~5.8×106 m3/yr. Our observations of the onset of uplift preclude the possibility that a nearby Mw 6.8 deep intraslab earthquake on 3rd June 2020 triggered the unrest. The deformation signal we detect indicates the initiation of unrest at Socompa, after at least two decades without measurable deformation, and many thousands of years without volcanic activity.

The Main Recent Fault is a major right-lateral strike-slip fault in the western Zagros mountains of Iran. Previous studies have estimated a wide range of slip rates from both sparse GNSS (1–6 mm/yr) and geological/geomorphological (1.6–17 mm/yr) methods. None of these studies have estimated the depth to the top of the locked seismogenic zone. Characterizing this “locking depth” for the Main Recent Fault, and more accurately constraining its interseismic slip rate, are both critical for estimating the seismic hazard posed by the fault, as well as for understanding how oblique convergence is accommodated and partitioned across the Zagros. To address this important knowledge gap for the MRF, here we use 200 Sentinel-1 SAR images from the past 5 years, spanning two ascending and two descending tracks, to estimate the first InSAR-derived slip rate and locking depth for a 300 km long section of the fault. We utilise two established processing systems, LiCSAR and LiCSBAS, to produce interferograms and perform time series analysis, respectively. We constrain north-south motion using GNSS observations, decompose our InSAR line-of-sight velocities into fault-parallel and vertical motion, and fit 1-D screw dislocation models to three fault-perpendicular profiles of fault-parallel velocity, following a Bayesian approach to estimate the posterior probability distribution on the fault parameters. We estimate an interseismic slip velocity of $3.0\pm1.0$ mm/yr below a loosely constrained 18–30 km locking depth, the first such estimate for the fault, and discuss the challenges in constraining the locking depth for low magnitude interseismic signals.