New Radar Interferometric Time Series Analysis Toolbox Released

EZ-InSAR: An easy-to-use open-source toolbox for mapping ground surface deformation using satellite interferometric synthetic aperture radar

Satellite Interferometric Synthetic Aperture Radar (InSAR) is a space-borne geodetic technique that can map ground displacement at millimetre accuracy. Via the new era for InSAR applications provided by the Copernicus Sentinel-1 SAR satellites, several open-source software packages exist for processing SAR data. These packages enable one to obtain high-quality ground deformation maps, but still require a deep understanding of InSAR theory and the related computational tools, especially when dealing with a large stack of images. Here we present an open-source toolbox, EZ-InSAR (easy-to-use InSAR), for a user-friendly implementation of InSAR displacement time series analysis with multi-temporal SAR images. EZ-InSAR integrates the three most popular and renowned open-source tools (i.e., ISCE, StaMPS, and MintPy), to generate interferograms and displacement time series by using these state-of-art algorithms within a seamless Graphical User Interface. EZ-InSAR reduces the user's workload by automatically downloading the Sentinel-1 SAR imagery and the digital elevation model data for the user's area of interest, and by streamlining preparation of input data stacks for the time series InSAR analysis. We illustrate the EZ-InSAR processing capabilities by mapping recent ground deformation at Campi Flegrei (> 100 mm·yr^-1) and Long Valley (~ 10 mm·yr^-1) calderas with both Persistent Scatterer InSAR and Small-Baseline Subset approaches. We also validate the test results by comparing the InSAR displacements with Global Navigation Satellite System measurements at those volcanoes. Our tests indicate that the EZ-InSAR toolbox provided here can serve as a valuable contribution to the community for ground deformation monitoring and geohazard evaluation, as well as for disseminating bespoke InSAR observations for all.

InSAR Constrained Downdip and Updip Afterslip Following the 2015 Nepal Earthquake: New Insights into Moment Budget of the Main Himalayan Thrust

We use ALOS-2 and Sentinel-1 data spanning 2015–2020 to obtain the post-seismic deformation of the 2015 Mw 7.8 Nepal earthquake. ALOS-2 observations reveal that the post-seismic deformation was mainly distributed in four areas. A large-scale uplift deformation occurred in the northern subsidence area of the co-seismic deformation field, with a maximum uplift of ~80 mm within 4.5 yr after the mainshock. While in the southern coseismic uplift area, the direction of the post-seismic deformation is generally opposite to the co-seismic deformation. Additionally, two notable deformation areas are located in the region around 29° N, and near the MFT, respectively. Sentinel-1 observations reveal post-seismic uplift deformation on the north side of the co-seismic deformation field with an average rate of ~20 mm/yr in line-of-stght. The kinematic afterslip constrained by InSAR data shows that the frictional slip is distributed in both updip and downdip areas. The maximum cumulative afterslip is 0.35 m in downdip areas, and 0.2 m in the updip areas, constrained by the ALOS measurements. The stress-driven afterslip model shows that the afterslip is distributed in the downdip area with a maximum slip of 0.3 m during the first year after the earthquake. Within the 4.5 yr after the mainshock, the estimated moment released by afterslip is ~1.5174 × 1020 Nm,about 21.2% of that released by the main earthquake.

Freeze-Thaw Deformation Cycles and Temporal-Spatial Distribution of Permafrost along the Qinghai-Tibet Railway Using Multitrack InSAR Processing

The Qinghai-Tibet Railway (QTR) is the railway with the highest elevation and longest distance in the world, spanning more than 1142 km from Golmud to Lhasa across the continuous permafrost region. Due to climate change and anthropogenic activities, geological disasters such as subsidence and thermal melt collapse have occurred in the QTR embankment. To conduct the large-scale permafrost monitoring and geohazard investigation along the QTR, we collected 585 Sentinel-1A images based on the composite index model using the multitrack time-series interferometry synthetic aperture radar (MTS-InSAR) method to retrieve the surface deformation over a 3.15 × 105 km2 area along the QTR. Meanwhile, a new method for permafrost distribution mapping based on InSAR time series deformation was proposed. Finally, the seasonal deformation map and a new map of permafrost distribution along the QTR from Golmud to Lhasa were obtained. The results showed that the estimated seasonal deformation range of the 10 km buffer zone along the QTR was −50–10 mm, and the LOS deformation rate ranged from −30 to 15 mm/yr. In addition, the deformation results were validated by leveling measurements, and the range of absolute error was between 0.1 and 4.62 mm. Most of the QTR was relatively stable. Some geohazard-prone sections were detected and analyzed along the QTR. The permafrost distribution results were mostly consistent with the simulated results of Zou’s method, based on the temperature at the top of permafrost (TTOP) model. This study reveals recent deformation characteristics of the QTR, and has significant scientific implications and applicational value for ensuring the safe operation of the QTR. Moreover, our method, based on InSAR results, provides new insights for permafrost classification on the Qinghai-Tibet Plateau (QTP).

Autonomous extraction of millimeter-scale deformation in InSAR time series using deep learning

Systematically characterizing slip behaviours on active faults is key to unraveling the physics of tectonic faulting and the interplay between slow and fast earthquakes. Interferometric Synthetic Aperture Radar (InSAR), by enabling measurement of ground deformation at a global scale every few days, may hold the key to those interactions. However, atmospheric propagation delays often exceed ground deformation of interest despite state-of-the art processing, and thus InSAR analysis requires expert interpretation and a priori knowledge of fault systems, precluding global investigations of deformation dynamics. Here, we show that a deep auto-encoder architecture tailored to untangle ground deformation from noise in InSAR time series autonomously extracts deformation signals, without prior knowledge of a fault’s location or slip behaviour. Applied to InSAR data over the North Anatolian Fault, our method reaches 2 mm detection, revealing a slow earthquake twice as extensive as previously recognized. We further explore the generalization of our approach to inflation/deflation-induced deformation, applying the same methodology to the geothermal field of Coso, California.

A deep neural network is developed to automatically extract ground deformation from Interferometric Synthetic Aperture Radar time series. Applied to data over the North Anatolian Fault, the method can detect 2 mm deformation transients and reveals a slow earthquake twice as extensive as previously recognized.

Coastal subsidence increases vulnerability to sea level rise over twenty first century in Cartagena, Caribbean Colombia

Cartagena is subsiding at a higher rate compared to that of global climate-driven sea level rise. We investigate the relative sea level rise (RSLR) and the influence of vertical land movements in Cartagena through the integration of different datasets, including tide gauge records, GPS geodetic subsidence data, and Interferometric Synthetic Aperture Radar (InSAR) observations of vertical motions. Results reveal a long-term rate (> 60 years) of RSLR of 5.98 ± 0.01 mm/yr. The last two decades exhibited an even greater rate of RSLR of 7.02 ± 0.06 mm/yr. GPS subsidence rates range between − 5.71 ± 2.18 and − 2.85 ± 0.84 mm/yr. InSAR data for the 2014–2020 period show cumulative subsidence rates of up to 72.3 mm. We find that geologically induced vertical motions represent 41% of the observed changes in RSLR and that subsidence poses a major threat to Cartagena’s preservation. The geodetic subsidence rates found would imply a further additional RSLR of 83 mm by 2050 and 225 mm by 2100. The Colombian government should plan for the future and serve as an example to similar cities across the Caribbean.

Dynamics of large effusive eruptions driven by caldera collapse

The largest effusive basaltic eruptions are associated with caldera collapse and are manifest through quasi-periodic ground displacements and moderate-size earthquakes^1-3, but the mechanism that governs their dynamics remains unclear. Here we provide a physical model that explains these processes, which accounts for both the quasi-periodic stick-slip collapse of the caldera roof and the long-term eruptive behaviour of the volcano. We show that it is the caldera collapse itself that sustains large effusive eruptions, and that triggering caldera collapse requires topography-generated pressures. The model is consistent with data from the 2018 Kīlauea eruption and allows us to estimate the properties of the plumbing system of the volcano. The results reveal that two reservoirs were active during the eruption, and place constraints on their connectivity. According to the model, the Kīlauea eruption stopped after slightly more than 60 per cent of its potential caldera collapse events, possibly owing to the presence of the second reservoir. Finally, we show that this physical framework is generally applicable to the largest instrumented caldera collapse eruptions of the past fifty years.

Improved Real-Time Natural Hazard Monitoring Using Automated DInSAR Time Series

As part of the collaborative GeoSciFramework project, we are establising a monitoring system for the Yellowstone volcanic area that integrates multiple geodetic and seismic data sets into an advanced cyber-infrastructure framework that will enable real-time streaming data analytics and machine learning and allow us to better characterize associated long- and short-term hazards. The goal is to continuously ingest both remote sensing (GNSS, DInSAR) and ground-based (seismic, thermal and gas observations, strainmeter, tiltmeter and gravity measurements) data and query and analyse them in near-real time. In this study, we focus on DInSAR data processing and the effects from using various atmospheric corrections and real-time orbits on the automated processing and results. We find that the atmospheric correction provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) is currently the most optimal for automated DInSAR processing and that the use of real-time orbits is sufficient for the early-warning application in question. We show analysis of atmospheric corrections and using real-time orbits in a test case over the Kilauea volcanic area in Hawaii. Finally, using these findings, we present results of displacement time series in the Yellowstone area between May 2018 and October 2019, which are in good agreement with GNSS data where available. These results will contribute to a baseline model that will be the basis of a future early-warning system that will be continuously updated with new DInSAR data acquisitions.

Adjacent-Track InSAR Processing for Large-Scale Land Subsidence Monitoring in the Hebei Plain

Large-scale land subsidence has threatened the safety of the Hebei Plain in China. For tens of thousands of square kilometers of the Hebei Plain, large-scale subsidence monitoring is still one of the most difficult problems to be solved. In this paper, we employed the small baseline subset (SBAS) and NSBAS technique to monitor the land subsidence in the Hebei Plain (45,000 km2). The 166 Sentinel-1A data of adjacent-track 40 and 142 collected from May 2017 to May 2019 were used to generate the average deformation velocity and deformation time-series. A novel data fusion flow for the generation of land subsidence velocity of adjacent-track is presented and tested, named as the fusion of time-series interferometric synthetic aperture radar (TS-InSAR) results of adjacent-track using synthetic aperture radar amplitude images (FTASA). A cross-comparison analysis between the two tracks results and two TS-InSAR results was carried out. In addition, the deformation results were validated by leveling measurements and benchmarks on bedrock results, reaching a precision 9 mm/year. Twenty-six typical subsidence bowls were identified in Handan, Xingtai, Shijiazhuang, Hengshui, Cangzhou, and Baoding. An average annual subsidence velocity over −79 mm/year was observed in Gaoyang County of Baoding City. Through the cause analysis of the typical subsidence bowls, the results showed that the shallow and deep groundwater funnels, three different land use types over the building construction, industrial area, and dense residential area, and faults had high spatial correlation related to land subsidence bowls.

Multi-Temporal InSAR Parallel Processing for Sentinel-1 Large-Scale Surface Deformation Mapping

Interferometric synthetic aperture radar (InSAR) has achieved great success in various geodetic applications, and its potential for ground deformation measurements on the large scale has attracted increasingly more attention in recent years. The increasing number of synthetic aperture radar (SAR) satellite systems have steadily provided a large amount of SAR data. Among these systems, the Sentinel-1 mission can be considered a milestone in the development of InSAR techniques, offering new opportunities to monitor global surface deformation with high precision, due to its wide coverage, short revisit time, and free access. However, conventional InSAR techniques have encountered great challenges in large-scale InSAR processing over wide areas because of the large computational burden and complexity. In this work, we present a novel parallel computing-based coherent scatterer InSAR (P-CSInSAR) method for automatic and efficient generation of deformation results from Sentinel-1 raw data. To achieve high parallelization performance for the overall InSAR processing chain, parallelization strategies at different levels have been adopted in the P-CSInSAR method, which has been fully addressed in this work. To evaluate the efficiency and accuracy of the proposed method, P-CSInSAR has been tested on the North China Plain regions with three adjacent frames of Sentinel-1 images, and the deformation results have been validated by GPS measurements. The experimental results confirm the effectiveness of the proposed parallel computing-based P-CSInSAR method. The proposed method can also play an important role in exploiting Sentinel-1 InSAR big data for disaster prevention and reduction.

Small-Baseline Approach for Monitoring the Freezing and Thawing Deformation of Permafrost on the Beiluhe Basin, Tibetan Plateau Using TerraSAR-X and Sentinel-1 Data

The dynamic changes of the thawing and freezing processes of the active layer cause seasonal subsidence and uplift over a large area on the Qinghai–Tibet Plateau due to ongoing climate warming. To analyze and investigate the seasonal freeze–thaw process of the active layer, we employ the new small baseline subset (NSBAS) technique based on a piecewise displacement model, including seasonal deformation, as well as linear and residual deformation trends, to retrieve the surface deformation of the Beiluhe basin. We collect 35 Sentinel-1 images with a 12 days revisit time and 9 TerraSAR-X images with less-than two month revisit time from 2018 to 2019 to analyze the type of the amplitude of seasonal oscillation of different ground targets on the Beiluhe basin in detail. The Sentinel-1 results show that the amplitude of seasonal deformation is between −62.50 mm and 11.50 mm, and the linear deformation rate ranges from −24.50 mm/yr to 5.00 mm/yr (2018–2019) in the study area. The deformation trends in the Qinghai–Tibet Railway (QTR) and Qinghai–Tibet Highway (QTH) regions are stable, ranging from −18.00 mm to 6 mm. The InSAR results of Sentinel-1 and TerraSAR-X data show that seasonal deformation trends are consistent, exhibiting good correlations 0.78 and 0.84, and the seasonal and linear deformation rates of different ground targets are clearly different on the Beiluhe basin. Additionally, there are different time lags between the maximum freezing uplift or thawing subsidence and the maximum or minimum temperature for the different ground target areas. The deformation values of the alpine meadow and floodplain areas are higher compared with the alpine desert and barren areas, and the time lags of the freezing and thawing periods based on the Sentinel-1 results are longest in the alpine desert area, that is, 86 days and 65 days, respectively. Our research has important reference significance for the seasonal dynamic monitoring of different types of seasonal deformation and the extensive investigations of permafrost in Qinghai Tibet Plateau.

The dynamics of large silicic systems from satellite remote sensing observations: the intriguing case of Domuyo volcano, Argentina

Silicic magmatic systems are the most dangerous volcanoes on Earth, capable of large and catastrophic eruptions, yet their low eruptive frequency makes it challenging to interpret their short-term unrest. Here we present a decade-plus analysis that integrates, for the first time, time series of satellite interferometric synthetic aperture radar (InSAR) surface deformation and satellite thermal infrared edifice-scale surface warming at a large silicic system: Domuyo volcano, in Argentina. We find that deformation and warming are highly correlated, and depending on the sign and lag between the time series, either shallow sealing or magma influx could drive Domuyo’s ongoing inflation (~ 0.15 m/year; from an InSAR-derived tabular source, ~ 11 × 8 × 1 km; ~ 6.5 km depth; ~ 0.037 km³/year volume-change rate) and warming (0.3–0.4 °C/year). This study shows the potential that combined satellite surface deformation and edifice-scale surface warming time series have on assessing the physical mechanisms of silicic volcanic systems and for constraining deterministic models.

Radar Interferometry: 20 Years of Development in Time Series Techniques and Future Perspectives

The research and improvement of methods to be used for deformation measurements from space is a challenge. From the previous 20 years, time series Synthetic Aperture Radar (SAR) interferometry techniques have proved for their ability to provide millimeter-scale deformation measurements over time. This paper aims to provide a review of such techniques developed in the last twenty years. We first recall the background of interferometric SAR (InSAR). We then provide an overview of the InSAR time series methods developed in the literature, describing their principles and advancements. Finally, we highlight challenges and future perspectives of the InSAR in the Big Data era.

LiCSBAS: An Open-Source InSAR Time Series Analysis Package Integrated with the LiCSAR Automated Sentinel-1 InSAR Processor

For the past five years, the 2-satellite Sentinel-1 constellation has provided abundant and useful Synthetic Aperture Radar (SAR) data, which have the potential to reveal global ground surface deformation at high spatial and temporal resolutions. However, for most users, fully exploiting the large amount of associated data is challenging, especially over wide areas. To help address this challenge, we have developed LiCSBAS, an open-source SAR interferometry (InSAR) time series analysis package that integrates with the automated Sentinel-1 InSAR processor (LiCSAR). LiCSBAS utilizes freely available LiCSAR products, and users can save processing time and disk space while obtaining the results of InSAR time series analysis. In the LiCSBAS processing scheme, interferograms with many unwrapping errors are automatically identified by loop closure and removed. Reliable time series and velocities are derived with the aid of masking using several noise indices. The easy implementation of atmospheric corrections to reduce noise is achieved with the Generic Atmospheric Correction Online Service for InSAR (GACOS). Using case studies in southern Tohoku and the Echigo Plain, Japan, we demonstrate that LiCSBAS applied to LiCSAR products can detect both large-scale (>100 km) and localized (~km) relative displacements with an accuracy of <1 cm/epoch and ~2 mm/yr. We detect displacements with different temporal characteristics, including linear, periodic, and episodic, in Niigata, Ojiya, and Sanjo City, respectively. LiCSBAS and LiCSAR products facilitate greater exploitation of globally available and abundant SAR datasets and enhance their applications for scientific research and societal benefit.

Time-Series InSAR Monitoring of Permafrost Freeze-Thaw Seasonal Displacement over Qinghai–Tibetan Plateau Using Sentinel-1 Data

Permafrost is widely distributed in the Tibetan Plateau. Seasonal freeze–thaw cycles of permafrost result in upward and downward surface displacement. Multitemporal interferometric synthetic aperture radar (MT-InSAR) observations provide an effective method for monitoring permafrost displacement under difficult terrain and climatic conditions. In this study, a seasonal sinusoidal model-based new small baselines subset (NSBAS) chain was adopted to obtain a deformation time series. An experimental study was carried out using 33 scenes of Sentinel-1 data (S-1) from 28 November 2017 to 29 December 2018 with frequent revisit (12 days) observations. The spatial and temporal characteristics of the surface displacements variation combined with different types of surface land cover, elevation and surface temperature factors were analyzed. The results revealed that the seasonal changes observed in the time series of ground movements, induced by freeze–thaw cycles were observed on flat surfaces of sedimentary basins and mountainous areas with gentle slopes. The estimated seasonal oscillations ranged from 2 mm to 30 mm, which were smaller in Alpine deserts than in Alpine meadows. In particular, there were significant systematic differences in seasonal surface deformation between areas near mountains and sedimentary basins. It was also found that the time series of deformation was consistent with the variation of surface temperature. Based on soil moisture active/passive (SMAP) L4 surface and root zone soil moisture data, the deformation analysis influenced by soil moisture factors was also carried out. The comprehensive analysis of deformation results and auxiliary data (elevation, soil moisture and surface temperature et al.) provides important insights for the monitoring of the seasonal freeze-thaw cycles in the Tibetan Plateau.

A shift from drought to extreme rainfall drives a stable landslide to catastrophic failure

The addition of water on or below the earth’s surface generates changes in stress that can trigger both stable and unstable sliding of landslides and faults. While these sliding behaviours are well-described by commonly used mechanical models developed from laboratory testing (e.g., critical-state soil mechanics and rate-and-state friction), less is known about the field-scale environmental conditions or kinematic behaviours that occur during the transition from stable to unstable sliding. Here we use radar interferometry (InSAR) and a simple 1D hydrological model to characterize 8 years of stable sliding of the Mud Creek landslide, California, USA, prior to its rapid acceleration and catastrophic failure on May 20, 2017. Our results suggest a large increase in pore-fluid pressure occurred during a shift from historic drought to record rainfall that triggered a large increase in velocity and drove slip localization, overcoming the stabilizing mechanisms that had previously inhibited landslide acceleration. Given the predicted increase in precipitation extremes with a warming climate, we expect it to become more common for landslides to transition from stable to unstable motion, and therefore a better assessment of this destabilization process is required to prevent loss of life and infrastructure.

Audit of stored strain energy and extent of future earthquake rupture in central Himalaya

The deadly 25 April 2015 Gorkha earthquake (M_w = 7.8) and aftershocks have partially released the accumulated interseismic strain along the Main Himalayan Thrust (MHT). Postseismic deformation associated with this earthquake is mainly confined to the north of the rupture. This suggests possible occurrence of future large events towards west or south, where MHT is locked. Asperities arising due to heterogeneity in the stress-strain patterns are believed to play a major role in controlling the coseismic rupture propagation. We determine interseismic coupling along the MHT and spatial variations in total strain rate using two decades of GPS, InSAR and sprit leveling data. Further, b-values derived from the seismicity data are used to identify zones of stress accumulation. We demonstrate that the 2015 earthquake ruptured an asperity which hosted high strain and stress accumulation prior to the event. A similar asperity towards west of the epicenter with unreleased strain energy is identified. This could spawn a future large earthquake akin in magnitude to the 2015 Gorkha event. These findings compel a revisit of the seismic hazard assessment of the central Himalaya.

Hazard Implications of the 2016 Mw 5.0 Cushing, OK Earthquake from a Joint Analysis of Damage and InSAR Data

The Cushing Hub in Oklahoma, one of the largest oil storage facilities in the world, is federally designated as critical national infrastructure. In 2014, the formerly aseismic city of Cushing experienced a Mw 4.0 and 4.3 induced earthquake sequence due to wastewater injection. Since then, an M4+ earthquake sequence has occurred annually (October 2014, September 2015, November 2016). Thus far, damage to critical infrastructure has been minimal; however, a larger earthquake could pose significant risk to the Cushing Hub. In addition to inducing earthquakes, wastewater injection also threatens the Cushing Hub through gradual surface uplift. To characterize the impact of wastewater injection on critical infrastructure, we use Differential Interferometric Synthetic Aperture Radar (DInSAR), a satellite radar technique, to observe ground surface displacement in Cushing before and during the induced Mw 5.0 event. Here, we process interferograms of Single Look Complex (SLC) radar data from the European Space Agency (ESA) Sentinel-1A satellite. The preearthquake interferograms are used to create a time series of cumulative surface displacement, while the coseismic interferograms are used to invert for earthquake source characteristics. The time series of surface displacement reveals 4–5.5 cm of uplift across Cushing over 17 months. The coseismic interferogram inversion suggests that the 2016 Mw 5.0 earthquake is shallower than estimated from seismic inversions alone. This shallower source depth should be taken into account in future hazard assessments for regional infrastructure. In addition, monitoring of surface deformation near wastewater injection wells can be used to characterize the subsurface dynamics and implement measures to mitigate damage to critical installations.

Mapping Damage-Affected Areas after Natural Hazard Events Using Sentinel-1 Coherence Time Series

The emergence of the Sentinel-1A and 1B satellites now offers freely available and widely accessible Synthetic Aperture Radar (SAR) data. Near-global coverage and rapid repeat time (6–12 days) gives Sentinel-1 data the potential to be widely used for monitoring the Earth’s surface. Subtle land-cover and land surface changes can affect the phase and amplitude of the C-band SAR signal, and thus the coherence between two images collected before and after such changes. Analysis of SAR coherence therefore serves as a rapidly deployable and powerful tool to track both seasonal changes and rapid surface disturbances following natural disasters. An advantage of using Sentinel-1 C-band radar data is the ability to easily construct time series of coherence for a region of interest at low cost. In this paper, we propose a new method for Potentially Affected Area (PAA) detection following a natural hazard event. Based on the coherence time series, the proposed method (1) determines the natural variability of coherence within each pixel in the region of interest, accounting for factors such as seasonality and the inherent noise of variable surfaces; and (2) compares pixel-by-pixel syn-event coherence to temporal coherence distributions to determine where statistically significant coherence loss has occurred. The user can determine to what degree the syn-event coherence value (e.g., 1st, 5th percentile of pre-event distribution) constitutes a PAA, and integrate pertinent regional data, such as population density, to rank and prioritise PAAs. We apply the method to two case studies, Sarpol-e, Iran following the 2017 Iran-Iraq earthquake, and a landslide-prone region of NW Argentina, to demonstrate how rapid identification and interpretation of potentially affected areas can be performed shortly following a natural hazard event.

Sentinel-1 InSAR Measurements of Elevation Changes over Yedoma Uplands on Sobo-Sise Island, Lena Delta

Yedoma—extremely ice-rich permafrost with massive ice wedges formed during the Late Pleistocene—is vulnerable to thawing and degradation under climate warming. Thawing of ice-rich Yedoma results in lowering of surface elevations. Quantitative knowledge about surface elevation changes helps us to understand the freeze-thaw processes of the active layer and the potential degradation of Yedoma deposits. In this study, we use C-band Sentinel-1 InSAR measurements to map the elevation changes over ice-rich Yedoma uplands on Sobo-Sise Island, Lena Delta with frequent revisit observations (as short as six or 12 days). We observe significant seasonal thaw subsidence during summer months and heterogeneous inter-annual elevation changes from 2016–17. We also observe interesting patterns of stronger seasonal thaw subsidence on elevated flat Yedoma uplands by comparing to the surrounding Yedoma slopes. Inter-annual analyses from 2016–17 suggest that our observed positive surface elevation changes are likely caused by the delayed progression of the thaw season in 2017, associated with mean annual air temperature fluctuations.