The physical mechanism behind TED is rooted in classical thermoelasticity. As temperature variations at the surface diffuse downward through heat conduction, they produce depth-dependent thermal strain in soil and bedrock beneath GNSS stations. This strain is transferred elastically to the ground surface and appears directly in GNSS height time series. The concept traces back to foundational work by Biot in 1956 and Berger in 1975, but its global-scale consequences for satellite geodesy have only recently become fully appreciated. Previous studies have shown that shallow subsurface thermoelastic effects contribute to seasonal height variations at some GNSS sites, and millimetre-level nonseasonal thermoelastic displacements have been detected by high-precision geodetic observations in some regions.

What makes the new dataset distinctive is its treatment of the shallow subsurface as a layered medium rather than a homogeneous one. Earlier analytical models relied on idealized elastic half-space or uniform elastic-sphere formulations with effectively depth-invariant thermophysical and elastic properties. While computationally efficient, these approaches cannot represent the attenuation and phase lag of thermal signals caused by soil-bedrock stratification, nor account for station-to-station differences in shallow subsurface conditions. The team instead implemented a full-spectrum layered finite-element model (FEM_FL) that solves the transient heat-conduction problem and computes surface displacement induced by the resulting temperature-dependent strain within a consistent numerical framework for each station independently.

The model is driven by hourly ERA5 soil-temperature forcing from the ECMWF reanalysis, using the shallowest 0–7 cm soil-temperature layer to prescribe the near-surface upper thermal boundary. Depth-dependent subsurface properties are parameterized using SoilGrids 2.0, which provides global land soil-property maps at 250 m spatial resolution across six standard depth intervals from 0 to 200 cm. Below 2 m, the model represents the medium as a homogeneous bedrock background with prescribed thermal and elastic properties. This choice reflects the depth range of available global soil data rather than an assumption that thermoelastic effects vanish at that depth, and the SoilGrids-based layering captures the main shallow subsurface contrasts relevant to the annual thermal cycle.

A key technical challenge involves temporal sampling. GNSS coordinate time series are typically provided at daily resolution, whereas near-surface temperature and the resulting thermoelastic response depend strongly on local solar time. If temperature forcing is sampled at only one fixed epoch, stations at different longitudes correspond to different local times, and part of the diurnal temperature signal may alias into longer-period components. To reduce this effect, the team sampled the hourly ERA5 record at four equally spaced epochs per day (00:00, 06:00, 12:00, and 18:00 UTC), producing four TED realizations for each station. Each realization was decomposed into annual, semiannual, and residual components, with harmonic components averaged using amplitude-phase vectors to preserve the mean magnitude and timing of the seasonal signal.

The resulting dataset reveals clear spatial patterns in thermoelastic displacement. Annual TED amplitudes are mostly within about 0.2–1.2 mm, with locally larger values exceeding 1.5 mm in regions with strong seasonal temperature contrasts. Relatively large signals are concentrated in mid-latitude continental areas, including parts of North America, Europe, East Asia, and Australia, whereas humid maritime regions and low-latitude areas generally show weaker annual responses. The annual phase shows a clear hemispheric contrast, consistent with the opposite seasonal heating cycles in the Northern and Southern Hemispheres. The percentile-based peak-to-peak variability of the full daily TED series commonly reaches 2–4 mm in continental regions and locally approaches 5–6 mm, while nonseasonal variability after removing the linear term and annual and semiannual harmonics is mostly below 1 mm but can locally reach millimetre level.

Technical evaluation of the dataset focused on three aspects: the influence of the SoilGrids-based subsurface parameterization, sensitivity to key thermal and elastic parameters, and spectral behaviour of the modelled time series. For the 2,612 stations with documented foundation-depth information, site-information-constrained solutions agree closely with SoilGrids-based solutions, with annual-amplitude differences mostly about 0.01–0.03 mm and annual-phase differences generally within 1–3 degrees. Sensitivity tests using ±10 percent perturbations in the linear thermal expansion coefficient, thermal diffusivity, and Young’s modulus indicate that the amplitude and phase of the annual TED component are stable under reasonable parameter uncertainty, with the largest sensitivity occurring mainly in dry inland regions where stronger seasonal temperature forcing also produces larger TED signals.

Spectral analysis of the TED series shows a dominant annual peak at 1 cycle per year and a weaker semiannual peak near 2 cycles per year in all latitude bands, consistent with the temperature-driven origin of the modelled displacement. Spectral power is also present at higher frequencies, indicating that the released series retain nonseasonal variability rather than representing only fitted annual and semiannual terms. The annual peak is stronger in mid- and high-latitude bands than in low-latitude bands, reflecting the stronger seasonal temperature cycle outside the tropics. Regional spectra show similar features, with North America, Europe, and East Asia generally showing stronger annual and semiannual peaks, whereas Australia and some lower-latitude or maritime regions show weaker seasonal power.

When applied alone to GNSS height time series, the TED correction produces positive weighted root-mean-square (WRMS) reductions at about 79 percent of the assessed stations, with most improvements within 3–5 percent and the largest reductions reaching about 10 percent. Some stations show negative WRMS reductions, locally around −2 to −5 percent, indicating that the TED-only correction does not improve every GNSS height series. This behaviour is expected because TED is only one part of the non-tectonic vertical displacement observed by GNSS, and its amplitude is usually smaller than hydrological loading and many local site-dependent signals. Even so, the spatial pattern suggests that TED explains part of the GNSS height variability, especially in continental regions with stronger seasonal temperature forcing.

The released product represents subsurface thermoelastic deformation only and does not include thermal expansion of above-ground structures such as concrete pillars, steel masts, rooftop monuments, or antenna-support structures. These monument-related effects can be important at individual stations and are not included because globally complete information on monument height, material, geometry, anchoring condition, and local thermal environment is not available for most GNSS stations. The TED series should therefore be interpreted as a subsurface thermoelastic correction rather than a complete correction for all temperature-related site effects. The dataset is publicly available on Zenodo and through the Ground Motion Data Service platform, with files organized by station and including date tags, vertical TED displacements in millimetres, station coordinates, record start and end dates, and dataset version information. Future updates may address snow, soil moisture, and frozen ground processes, and may consider sub-daily products after further validation with appropriate GNSS solutions.

Subject of Research: Temperature-driven thermoelastic displacement of the shallow subsurface and its effect on GNSS station height measurements

Article Title: TED: a global temperature-driven thermoelastic displacement dataset for GNSS reference stations (2000–2023)

Article References: Lu, R., Li, Z., Ye, L., Yuan, P., & Feng, Y. (2026). TED: a global temperature-driven thermoelastic displacement dataset for GNSS reference stations (2000–2023). Earth System Science Data, 18(10), 7575-7591. https://doi.org/10.5194/essd-18-7575-2026

Image Credits: AI Generated

DOI: 10.5194/essd-18-7575-2026

Keywords: GNSS, thermoelastic displacement, TED dataset, ERA5, SoilGrids, finite-element model, vertical land motion, reference frame, geodesy, soil temperature, height time series, crustal deformation

The study, published in Communications Earth & Environment, was led by Rikuto Fukushima of Stanford University, together with Masayuki Kano of Kyoto University and the Japan Agency for Marine-Earth Science and Technology, Kazuro Hirahara of RIKEN and Kagawa University, and Makiko Ohtani of Kyoto University. Their target was the 2010 Bungo slow slip event, a well-documented episode of silent fault motion that struck the Bungo Channel region between the islands of Shikoku and Kyushu. What the team found was a striking picture of frictional heterogeneity: the slow slip event nucleated in a localized patch off southwest Shikoku, where frictional properties differ sharply from their surroundings, and then propagated westward along the plate boundary.

To understand why this matters, it helps to grasp the long-standing puzzle of fault behavior. Laboratory experiments on rock samples have shown that faults obey friction laws — mathematical rules describing how resistance to sliding depends on slip rate, contact time, and other variables. When modelers feed these laboratory-derived laws into simulations of fault slip, the simulations naturally produce a rich diversity of behavior: some patches rupture rapidly as earthquakes, others creep slowly, and still others remain locked for centuries, accumulating the strain that eventually powers the largest quakes. The agreement between such simulations and real observations has long supported the hypothesis that the observed diversity of slip — from fast earthquakes to slow transient events — arises from spatial variations in frictional properties stamped into the fault by geology.

But there has been a stubborn gap between that elegant hypothesis and hard data. Confirming it quantitatively would require taking real geodetic observations — the millimeter-to-centimeter shifts of GPS stations as an unseen fault slips below — and working backwards to infer the frictional parameters that must have produced them. That inverse problem is brutally difficult. A fault plane is a two-dimensional surface, and frictional properties vary across it point by point, meaning the unknowns number in the thousands or more. Traditional optimization methods buckle under this high-dimensional search space, so most studies have settled for estimating only the distribution of fault slip itself, leaving the underlying friction physics unconstrained. The new work was designed precisely to close that gap.

The team’s solution was to deploy physics-informed neural networks, or PINNs, a technique that has swept through computational physics in recent years. In a conventional neural network application, the network learns patterns purely from data. In a PINN, the network is additionally constrained to obey the governing equations of the physical system — in this case, the equations of quasi-dynamic fault slip based on rate- and state-dependent friction laws. The researchers used the neural network to represent the spatially variable frictional properties on the fault surface. Because a neural network can encode smooth, structured fields with relatively few internal parameters, it tames the high-dimensional optimization problem that had defeated earlier approaches. The network’s outputs are fed into the fault mechanics equations, and the mismatch between the resulting predicted surface displacements and the actual GPS observations drives the training, pulling the estimated friction field toward one that is simultaneously mechanically consistent and consistent with the data.

Applying this framework to the 2010 Bungo slow slip event, the researchers estimated the heterogeneous frictional parameters along the subducting plate interface beneath southwest Japan. The resulting picture showed frictional properties varying systematically along the fault, with a localized region off southwest Shikoku standing out as the nucleation zone where the slow slip event began. From there, the event propagated westward, a pattern that emerged naturally from the estimated friction distribution. Crucially, the calculated fault slip produced by those frictional properties successfully reproduced the spatio-temporal pattern of the observed surface displacements — not just how much the ground moved, but the sequence and geography of that motion through time.

This dual consistency is what sets the result apart. Many previous slip models could fit geodetic data, but they said nothing about whether the inferred slip was physically achievable given plausible fault friction. Conversely, forward simulations of fault cycles could produce realistic slow slip events, but only for assumed friction distributions, not ones verified against observations. The PINN-based inversion delivers a model that is validated on both fronts: it is a mechanically consistent fault slip model, grounded in laboratory-derived friction physics, and it is quantitatively checked against real measurements. In effect, the team has demonstrated that the frictional heterogeneity hypothesis — the idea that patchy friction generates the spectrum of slip behavior — can be tested against data rather than merely assumed.

The study did not stop at reconstructing the past. The researchers also used their estimated friction model to forecast the future evolution of fault slip, and in doing so they uncovered a practical lesson for the emerging field of earthquake forecasting. The accuracy of such forecasts depends heavily on how much observational history is assimilated into the model. By assimilating observations spanning multiple slow slip event cycles, rather than a single event, the forecasts become substantially more reliable. This finding carries a clear message for monitoring networks: continuous, long-term geodetic observation is not just an archive of past behavior but an essential ingredient for anticipating future fault motion.

The implications reach well beyond the Bungo Channel. Slow slip events are now recognized along many of the world’s subduction zones, including Cascadia in the Pacific Northwest, Mexico, Chile, New Zealand, and Alaska. In several of these regions, slow slip episodes have been linked to heightened stress on adjacent locked patches capable of producing large earthquakes — most notably in the lead-up to the 2011 Tohoku-oki earthquake in Japan, where precursory slow slip near the hypocenter has been documented. If physics-informed inversion can map the frictional architecture of these faults from routine GPS data, it could eventually help scientists identify which slow slip events are routine and which signal dangerous stress transfer toward locked, earthquake-ready patches.

The methodological advance may prove just as consequential as the specific findings. High-dimensional inverse problems pervade geophysics, from imaging Earth’s interior to modeling magma chambers and groundwater flow. By showing that a neural network can serve as a compact, physically constrained representation of a complex field like fault friction — and that training it against both physical laws and observational data yields a solution previously out of reach — the study opens a template that other researchers can adapt. The work was supported by Japan’s MEXT STAR-E and STAR-E NEXT projects, the Earthquake Research Institute at the University of Tokyo, and JSPS KAKENHI grants, reflecting a sustained national investment in earthquake science and machine learning.

For a field that has spent half a century bridging laboratory rock mechanics and field observation, the arrival of a tool that finally connects the two at scale is a genuine milestone. The 2010 Bungo slow slip event slipped silently, registering only as a gentle drift of GPS antennas across Shikoku and Kyushu. Yet within that quiet drift lay the frictional blueprint of the fault itself — and now, with physics-informed deep learning, scientists have learned to read it. The same approach, applied to longer and richer observational records, promises not only a deeper understanding of how faults work but a step toward the long-sought goal of physics-based fault slip forecasting, where the slow whispers of a fault today inform the assessment of its roar tomorrow.

Subject of Research: Physics-informed deep learning estimation of spatially variable fault friction from geodetic observations of slow slip events

Article Title: Physics-informed deep learning links geodetic data and fault friction

Article References: Fukushima, R., Kano, M., Hirahara, K., & Ohtani, M. (2026). Physics-informed deep learning links geodetic data and fault friction. Communications Earth & Environment. https://doi.org/10.1038/s43247-026-04138-y

Image Credits: AI Generated

DOI: 10.1038/s43247-026-04138-y

Keywords: physics-informed neural networks, slow slip events, fault friction, geodesy, subduction zone, Bungo Channel, rate-and-state friction, machine learning, seismology, earthquake forecasting, southwest Japan, fault mechanics

The oldest of the three records came from Mossel Bay, on the north coast of the largest island of Spitsbergen, during the Swedish polar expedition led by the celebrated explorer Adolf Erik Nordenskiöld. In October and November 1872, the crew of the brig Gladan cut a small square hole in the sea ice and passed a graduated pole through it, fixed at one end in the seabed and standing four or five feet above the ice at the other. A wooden float slid up and down the pole with the tide, and sailors recorded readings every half hour at the changing of the guard. In spring 1873, personnel at the Polhem base repeated the exercise in shallower water, taking hourly measurements from mid-February to late April. The measurements were made in Swedish feet and decimal inches, units slightly smaller than their British equivalents, and the analysis was eventually published by August Wijkander in 1889, years after the expedition returned.

The second record is tied to one of the most haunting episodes in polar history. In June and July 1897, officers and sailors of the gunboat Svensksund read a simple tide board on the shore of Île des Danois at Port Virgo, on the northwest coast of Spitsbergen, during the second Svalbard expedition of Salomon August Andrée, the engineer who later that same year vanished while attempting to reach the North Pole by balloon. A book of the tabulated observations was handed to the physicist Vilhelm Carlheim-Gyllensköld by a sailor named Dahlgren, and the hourly values found their way into his 1905 report. The record spans just over a month, with several gaps filled by graphical interpolation, and the researchers note some confusion in the original documents over whether the times were recorded in local apparent or mean time, an ambiguity they show to be of little practical consequence.

The third and technically most extraordinary record was made at Sorgfjord, on the north coast of Spitsbergen, during the Swedish-Russian Arc-of-Meridian expedition of 1899 to 1902. That campaign, one of the largest geodetic projects of its kind ever attempted, stretched a chain of triangulation points across mountain tops spanning more than four degrees of latitude, in order to measure the length of a meridian arc and determine how much the spherical Earth is flattened at the poles. The Swedish wintering party at Sorgfjord was led by the geodesist Edvard Jäderin, who built much of the expedition’s equipment himself. Because the height of every point in a triangulation network depends on knowing the height of a baseline above mean sea level, a tide gauge was essential, and Jäderin installed one on 26 March 1900 after the expedition’s intended modern gauge, designed by the Finnish geodesist Alfred Petrelius, failed to arrive in time.

What Jäderin improvised has no known precedent or successor. A strong wire, roughly 2.25 millimetres in diameter, was anchored at the foot of a post on land, passed through a hasp on a second weighted post, ran above the recording apparatus, and extended out to a pulley on a tripod frozen into the floating bay ice some 80 to 85 metres from shore. A stone weighing 75 to 100 kilograms hung from the wire below the pulley. Because the ratio of distances along the wire was fixed at one to 17.22, the vertical motion of the ice as the tide rose and fell was reproduced on a chart recorder, traced by a crayon on a thin red copper wire, at 17.22 times magnification. The clock-driven cylinder turned at just 2.3 millimetres per hour, and charts were replaced every Monday at ten in the morning. Datum control came from levelling sights on the ice and from dipping measurements through a hole near the tripod, tied back to benchmarks on land.

Against all expectations, the contraption worked. It produced hourly sea level values spanning 105 consecutive days, from late March to mid-July 1900, with only occasional interruptions of a few hours to several days, during which interpolated values were printed in italics in the original report. Woodworth and Aarup subjected the digitised record to modern harmonic analysis using software that fits 27 independent tidal constituents, appropriate for short records. The resulting amplitudes and phase lags for the two main semidiurnal constituents, M2 and S2, and the two main diurnal ones, K1 and O1, agree closely both with Carlheim-Gyllensköld’s own hand calculations and with the values long listed in the Admiralty Tide Tables. The residuals of the analysis show remarkably little non-tidal variability, as expected for the calm spring and summer months, with only a couple of short suspicious episodes that appear to stem from chart recording or digitisation errors.

The study also untangles a century of confusion in the official tidal literature. The Admiralty Tide Tables values for Sorgfjord, long assumed to derive from a 1930s reanalysis, were in fact computed far earlier by the United States Coast and Geodetic Survey for Robert Harris’s 1911 monograph on Arctic tides. Harris, comparing his values with Carlheim-Gyllensköld’s, mistakenly declared the latter evidently erroneous, apparently failing to realise that the Swedish phase lags were referred to Greenwich while the American ones were local values. A similar mix-up affected the Port Virgo constants, which passed through Harris and later through the Norwegian oceanographers Kjær and Fjeldstad into the modern tables, carrying with them an erroneous S2 phase lag. For Mossel Bay, the researchers identified a 180-degree error in the diurnal phase lags of Wijkander’s original analysis, a mistake noted by George Darwin as early as 1889, and conclude that the original values, once corrected, are the more reliable.

To place the historic measurements in context, the authors compared their computed constants with FES2022b, a leading global ocean tide model built on more than three decades of precise satellite altimetry and advanced hydrodynamic modelling. The agreement is excellent for the semidiurnal tides at all three sites, and good for most of the diurnal components, with the largest discrepancy a difference of about 20 degrees in the O1 phase lag at Mossel Bay for a constituent only centimetres in amplitude. The model also reveals the physics of the region: the M2 tidal wave rotates clockwise around the archipelago, its amplitude falling and its phase lag increasing as one travels east along the north coast of Spitsbergen, while the diurnal tides are comparatively uniform, apart from an amphidromic feature near the southern tip of Spitsbergen. Crucially, the match between the 1872, 1897 and 1900 observations and the modern model suggests there have been no large changes in the tide in this part of the Arctic over the past century and a half.

The researchers are careful to note that the records carry scientific value precisely because tide gauge coverage of the high Arctic remains sparse even today. Shorter modern recordings exist at Sorgfjord, one week in 1938 and three weeks in 2014, and at Port Virgo in 2008 and 2009, but none approaches the length of the historic series. The paper also points to other lost opportunities, including an Italian Hydrographic Institute gauge established in 1928 at Kings Bay in support of Umberto Nobile’s airship flights, whose benchmarks might still survive and could, with new measurements, allow long-term sea level change to be investigated. Above all, the authors pay tribute to the dedicated and often highly qualified participants who took such care with the most basic equipment in brutal conditions, from Jäderin’s wire-and-crayon machine to the sailors reading floats through holes in the ice. Their rescued data, now freely available in digital form, demonstrate that the patient recovery of historical measurements can still pay real dividends for present-day ocean science.

Subject of Research: Historical Arctic tide gauge records from Svalbard and their modern tidal analysis

Article Title: Three historic tide gauge records from Svalbard

Article References: Woodworth, P. L., & Aarup, T. (2026). Three historic tide gauge records from Svalbard. History of Geo- and Space Sciences, 17(1), 1-12. https://doi.org/10.5194/hgss-17-1-2026

Image Credits: AI Generated

DOI: 10.5194/hgss-17-1-2026

Keywords: tide gauges, Svalbard, Arctic, sea level, tidal analysis, harmonic constants, data archaeology, polar exploration, geodesy, FES2022b, ocean tides, history of science

The study, published in the journal Solid Earth, analyzed Very Long Baseline Interferometry, or VLBI, observations spanning 1980 to 2025. VLBI works by pointing networks of radio telescopes scattered across the globe at the same distant quasars, the brilliant active nuclei of galaxies billions of light-years away. Because quasars are so remote, they serve as almost perfectly fixed reference points in the sky. By measuring the tiny differences in the arrival times of each quasar’s radio wavefront at different telescopes, geodesists can determine Earth’s orientation in space with extraordinary precision, down to microarcseconds, an angular unit equivalent to resolving a coin on the Moon. That stability is what makes VLBI the only technique capable of tracking Earth’s precession and nutation, the slow conical drift and periodic nodding of the spin axis, against a truly inertial celestial frame.

Nutation is the key to probing the deep interior. The gravitational tug-of-war between Earth and the Sun, Moon, and planets on a perfectly rigid, spherical planet can be calculated with exquisite accuracy. Any deviation between that idealized prediction and what VLBI actually observes must be caused by the real planet’s imperfections: its flattened, fluid-filled, deformable interior, the sloshing of ocean tides, and the elasticity of its rocks. In the language of geophysics, these deviations are captured by a set of quantities called Basic Earth Parameters, which encode the planet’s dynamical ellipticity, the deformability of its layers, and the coupling strengths at the boundaries between the mantle, the liquid outer core, and the solid inner core. Estimating them accurately is equivalent to performing a tomographic scan of the planet’s deepest reaches using nothing but its rotation.

The conventional nutation model in use today, known as MHB2000 and adopted by the International Astronomical Union in 2003, was built on VLBI data covering only two decades, from 1980 to 1999, and fitted with classical least-squares methods. The new work, led by Yuting Cheng, builds on a Bayesian approach pioneered in 2008 but modernizes it substantially. Instead of linearizing the problem, the team used an ensemble Markov Chain Monte Carlo sampler, a statistical engine that deploys hundreds of parallel random walkers to explore the full probability landscape of the parameters. This approach, implemented with the Python package emcee, does not require approximating the relationship between observations and parameters, and it yields not just single best-fit numbers but complete probability distributions, with honest uncertainties attached.

Two technical upgrades proved decisive. First, the team modeled the amplitude of the Free Core Nutation, a natural rotational wobble of the Earth with a period of about 430 days that arises because the planet holds a fluid, flattened core inside a deformable mantle. This mode is excited by poorly understood processes, possibly atmospheric and oceanic loading or abrupt changes in the core’s magnetic field known as geomagnetic jerks, so its strength must be estimated from the data itself. Earlier work represented its variation with straight-line segments; the new study replaced those with smooth cubic splines, which dramatically reduced the tendency of the statistical sampler to get stuck on multiple contradictory solutions. Second, the researchers updated the ocean tidal corrections using the FES2014 global ocean tide atlas, dropping an empirical scaling factor of 0.7 that had been patched into the 2000 model to compensate for crude ocean maps. Modern tidal modeling, it turns out, no longer needs that fudge factor.

The results, consistent across five independent VLBI data series processed by different analysis centers, carry profound geophysical implications. The imaginary part of the coupling constant at the core-mantle boundary, which measures dissipative braking between the liquid core and the mantle, came out larger in absolute value than the 2000 estimate, approaching the two-sigma boundary of the old model. Electromagnetic coupling alone, the standard explanation, struggles to account for it: matching the observed value would require a radial magnetic field of roughly 0.75 millitesla at the boundary and implausibly high, uniform electrical conductivity throughout the lower mantle. The authors point instead to recent theoretical work on topographic coupling, in which the liquid core flows over an irregular, bumpy boundary and internal waves exert a form of drag, a mechanism that could supply the missing dissipation without demanding exotic mantle properties.

The inner core delivered its own surprise. The real part of the coupling constant at the inner core boundary, which reflects how rigidly the solid inner core is attached to the surrounding liquid, came out at roughly half the value in the conventional model. Under a purely electromagnetic interpretation, that weaker coupling implies a magnetic field at the inner core boundary closer to what numerical geodynamo simulations predict, between 0.1 and 0.4 tesla. But the finding also resonates with startling recent seismic discoveries. Independent studies of earthquake waves passing through the planet’s center have suggested that the inner core may deform viscously on timescales of years, and that a mushy, partially molten zone may exist at its boundary. If the inner core boundary is not a surface of hydrostatic equilibrium, as nutation theory has long assumed, the theoretical scaffolding underneath these estimates needs revision.

Even the planet’s bulk deformability showed cracks in the old framework. The compliances that describe how the whole Earth and its fluid core yield under tidal forcing are normally derived from a seismic reference model and extrapolated from seismic frequencies, seconds to hours, up to the daily nutation band using a power law. In the new inversion, one compliance matched expectations while the other, governing the fluid core, came out significantly smaller. Because both are corrected by the same anelastic scaling, a shift in one without the other cannot be fixed by tweaking the extrapolation exponent, pointing to deeper flaws either in the reference Earth model or in the anelastic correction scheme itself. The estimated dynamical ellipticity of the whole Earth, meanwhile, sits at the edge of the old model’s one-sigma range, implying that the core’s equatorial bulge may exceed its hydrostatic value by the equivalent of 370 to 500 meters of radius.

For the researchers, the pattern across so many parameters points in one direction: nutation theory itself is due for a major overhaul, one that integrates realistic core-mantle coupling, fluid dynamics of the outer core, and revised models of the inner core’s behavior into a single consistent framework. The stakes extend beyond pure science. Earth orientation parameters underpin the celestial and terrestrial reference frames on which satellite navigation, spacecraft tracking, and deep-space navigation all depend, so refining them sharpens humanity’s collective aim across the solar system. And every improvement in the precision of these rotation-based probes brings geophysicists closer to answering some of the oldest questions about the engine at the heart of our planet, an engine we cannot visit, cannot drill to, and can only read through the gentle, patient wobble of the world itself.

Subject of Research: Estimation of Basic Earth Parameters and core-mantle coupling from Bayesian inversion of VLBI nutation observations

Article Title: Basic Earth Parameters from VLBI observations using Bayesian inversions in the time domain: updated insights of the Earth's interior

Article References: Cheng, Y., Dehant, V., Rivoldini, A., Rekier, J., & Bizouard, C. (2026). Basic Earth Parameters from VLBI observations using Bayesian inversions in the time domain: updated insights of the Earth's interior. Solid Earth, 17(5), 789-801. https://doi.org/10.5194/se-17-789-2026

Image Credits: AI Generated

DOI: 10.5194/se-17-789-2026

Keywords: VLBI, Earth rotation, nutation, core-mantle boundary, inner core, Bayesian inversion, MCMC, Free Core Nutation, geodesy, ocean tides, dynamical ellipticity, Earth's interior

The controversy began when Li and colleagues compared two published InSAR datasets covering the central Gulf Coast, one produced by Ohenhen and colleagues in 2024 and another by Wang and colleagues, also in 2024. Both datasets reported broadly similar regional averages of land motion, yet when Li’s team compared them pixel by pixel outside urban areas, the spatial correlation collapsed to a coefficient of determination of just 0.05, essentially no agreement at all. From this, the authors concluded that InSAR is unreliable in vegetated coastal settings for rates below five millimeters per year, and they recommended that such satellite-derived velocities be interpreted with utmost caution in regions lacking dense ground-truth networks. Taken at face value, that verdict threatened to sideline radar mapping precisely where it is most needed: the marshy, rapidly deforming coastal lowlands.

Shirzaei’s team does not dispute that reproducibility matters, and they endorse the call for harmonized processing frameworks and systematic cross-validation. But they contend that the apparent disagreement between the two datasets can be traced almost entirely to how the comparison was constructed. The first and most consequential problem, they argue, is a temporal one that the original paper never acknowledged. The two products were built from non-overlapping observational windows: the Ohenhen dataset spans 2007 to 2020, combining C-band Sentinel-1 radar with L-band ALOS observations and Global Navigation Satellite System data in a joint stochastic inversion, while the Wang dataset covers only 2017 to 2020 using Sentinel-1 alone. In a landscape where subsidence rates are known to change over time, comparing products from different epochs is not a like-for-like test.

The Virginia Tech group quantified just how much that epoch mismatch matters. When each radar product is validated against GNSS measurements over its own observational window, performance is strong; when validated against the mismatched window, residual standard deviations inflate by roughly 19 to 22 percent. In other words, a meaningful share of the reported disagreement between the two satellite products simply reflects the fact that they were watching the ground during different years, not that either instrument was failing. The comment paper presents this as a reductio ad absurdum of the original critique’s logic: if inter-product spread defines a caution threshold, then the threshold itself is an artifact of methodology, not a property of the Earth.

The second major concern involves spatial resolution. Li’s team resampled the Ohenhen dataset from its native 50-meter pixel spacing to a one-kilometer grid to match the coarser Wang product, computing median values within a 500-meter search radius around each coarse pixel. That aggregation, Shirzaei’s team calculates, represents a roughly 400-fold reduction in spatial information density, discarding exactly the sub-kilometer structure that the fine-scale dataset was designed to resolve, including localized signals from differential sediment compaction, fault-zone deformation, and fluid extraction. The original authors justified the resampling by noting that different search radii produced only minor differences, but the commenters counter that this merely shows the averaging had already erased the fine-scale variability at the smallest radius tested. Until the comparison is repeated at native resolution, they argue, the headline correlation of 0.05 cannot be attributed to radar incoherence in vegetated landscapes rather than to the resampling itself.

The validation strategy also drew sharp criticism. Li’s team progressively filtered a pool of 110 candidate GNSS stations down to roughly 20 usable sites, excluding Holocene-deposit locations, all Texas stations, and sites near active fluid extraction wells. The surviving network was concentrated in Pleistocene upland settings, by design the most stable and geologically quiet parts of the coast. But that is precisely the problem, the commenters argue: demonstrating that radar disagrees with GNSS at stable upland sites says little about performance in the Holocene coastal lowlands, where subsidence is fastest, vegetation is densest, and the datasets are actually meant to be used. By contrast, the original Ohenhen validation employed 157 GNSS stations spanning the full range of Gulf Coast environments and yielded a residual standard deviation of just 1.5 millimeters per year.

To settle the question independently, Shirzaei’s team assembled their own benchmark: daily vertical displacement time series from 88 GNSS stations of the Nevada Geodetic Laboratory within the same study domain, processed to remove offsets, outliers, and common-mode errors. Validating the Ohenhen dataset at its native 50-meter resolution against this network produced a residual standard deviation of 1.6 millimeters per year with a mean residual of only minus 0.1 millimeters per year, closely consistent with the original published uncertainty. The Wang dataset, evaluated over its own window and resolution, achieved a residual standard deviation of 2.3 millimeters per year. Both products, in short, perform robustly when tested on their own terms, in direct contrast to the characterization offered by the original critique.

The comment paper then turns the original critique’s own logic against itself. The five-millimeter-per-year caution threshold was derived from the 95th percentile of absolute differences between the two radar products in urban areas, then extrapolated to vegetated settings where noise is known to be worse. But applying the same reasoning to a GNSS-versus-GNSS comparison across the two observational windows yields an equivalent threshold of 3.7 millimeters per year, below which the original paper’s own glacial isostatic adjustment estimate of minus 1.2 millimeters per year falls unambiguously. By the critique’s own standard, its own benchmark measurement would be untrustworthy. The commenters also note an internal contradiction: the paper anchors its analysis on a GNSS rate of minus 1.2 millimeters per year while simultaneously declaring satellite rates in that range unresolvable, without explaining why one instrument is exempt from the caution it imposes on the other.

Beyond the specific dispute, the comment paper offers a broader set of best practices for validating geodetic deformation products, distinguishing validation against independent ground truth from mere benchmarking between products, insisting on native-resolution evaluation, demanding spatially representative validation networks, separating accuracy from precision, and requiring per-pixel uncertainty to be propagated from interferometric phase variance through every processing step rather than estimated after the fact. They also outline a statistical framework, from Bland-Altman agreement analysis to spatial autocorrelation tests, designed to prevent comparison artifacts from being mistaken for product deficiencies. The editorial board of Earth Observation highlighted the exchange as an important moment for the entire vertical land motion community.

The bottom line for coastal communities is reassuring: when evaluated at native resolution, over its own observational window, and against a geographically representative ground-truth network, satellite radar mapping of coastal subsidence performs robustly at the millimeter-per-year level. The apparent crisis of reproducibility documented in the original critique was, in substantial part, a consequence of the comparison’s construction rather than of the measurements themselves. As sea levels rise and Gulf Coast cities sink, the researchers argue, the wiser investment is in rigorous intercomparison infrastructure, matched epochs, matched resolutions, and principled uncertainty analysis, rather than blanket caution thresholds that risk discounting valid, policy-relevant datasets on the basis of a flawed test.

Subject of Research: Validation of InSAR-derived surface-elevation change rates along the central U.S. Gulf Coast

Article Title: Comment on: “Evaluating InSAR-derived rates of surface-elevation change along the central U.S. Gulf Coast” by Li et al. (2026)

Article References: Shirzaei, M., Ohenhen, L., Atkins, C., Dasho, O., Sadhasivam, N., Kamaraj, N. P., Onyike, F., Olorunsaye, O., Oyedele, E. O., & Werth, S. (2026). Comment on: “Evaluating InSAR-derived rates of surface-elevation change along the central U.S. Gulf Coast” by Li et al. (2026). Earth Observation, 1(1), 145-154. https://doi.org/10.5194/eo-1-145-2026

Image Credits: AI Generated

DOI: 10.5194/eo-1-145-2026

Keywords: InSAR, land subsidence, Gulf Coast, GNSS, satellite radar, sea-level rise, vertical land motion, coastal monitoring, geodesy, validation methods, Sentinel-1, uncertainty quantification

The appeal of the roughness hypothesis has always been its visual plausibility. On bathymetric maps, the regions that hosted the largest recorded events, such as the 2004 Sumatra-Andaman earthquake, the 2011 Tohoku-oki earthquake and the 1960 Chile earthquake, seemed to correspond with segments of the trench where the incoming plate appeared comparatively smooth, while rougher margins like those off parts of Central America or Japan’s older, sediment-starved trenches appeared to host only smaller ruptures. The hypothesis gained traction because it offered a forecasting shortcut: map the bumps on the incoming seafloor before it subducts, and you have a proxy for how big the next earthquake might be. Given how difficult it is to observe the actual fault surface at depth, a measurable feature of the incoming plate seemed like a gift to hazard assessors.

Yet the new analysis finds that the correlation does not survive careful, systematic testing. Rather than selecting a few celebrated case studies, the researchers assembled a globally consistent dataset of subduction interface roughness measurements and paired them with the maximum magnitudes that each margin has produced, constrained by both the instrumental record and, where available, longer historical and paleoseismic evidence. When roughness is quantified uniformly, using the same topographic measures of relief and spectral character of the incoming plate across all margins, the expected relationship between rough interfaces and smaller maximum earthquakes largely dissolves. Margins with prominently rough seafloor have still hosted very large ruptures, and some comparatively smooth margins have not delivered the giant events the hypothesis predicts.

The authors point to several reasons why the roughness argument fails as a control on maximum magnitude. First, roughness measured on the seafloor before subduction is a poor guide to roughness on the fault at seismogenic depths. As the plate descends, sediments smear into the topographic lows, hydrothermal alteration and mineralization smooth the interface, and the accretionary wedge redistributes material. A seamount that looks like a formidable asperity at the trench may be largely underplated or subducted within a weaker sedimentary layer by the time it reaches the depth range where most seismic moment is released. The fault that ruptures in a great earthquake is therefore not the same rough surface that geodesists and marine geologists can map from shipboard sonar.

Second, and more fundamentally, the maximum magnitude of an earthquake is a geometric property of how large a patch can rupture in a single event, and that is governed by the lateral and downdip extent of coherent locking, the segmentation of the plate boundary, and the accumulated slip deficit, not necessarily by centimeter- to kilometer-scale frictional heterogeneity. A rupture can jump or creep past small roughness elements if the surrounding fault is sufficiently stressed and strongly coupled. Conversely, a smooth interface segmented by major structural boundaries such as fracture zones or tear faults may still be unable to host a rupture longer than a few hundred kilometers. In other words, the features that actually arrest or release ruptures may be structural discontinuities with tens of kilometers of offset, not the relief on the incoming plate.

The study also re-examines the physical reasoning behind the roughness hypothesis itself. Laboratory and theoretical work shows that roughness influences frictional behavior most strongly at small scales, affecting the onset of slip and the distribution of aftershocks, but its effect on the total possible rupture area at the scale of a magnitude 9 earthquake is weak. Scale considerations matter: an earthquake of magnitude 9 ruptures a fault patch on the order of 1,000 kilometers long. Bumps a few kilometers across are simply too small to act as persistent barriers to a rupture front carrying enormous elastic strain energy. The comparison is often made to tearing a sheet of paper: small wrinkles in the paper do not determine where the tear stops if the sheet is being pulled hard enough.

The implications for seismic hazard assessment are significant and uncomfortable. If roughness cannot be used as a proxy for maximum magnitude, then several regional hazard models that factor seafloor roughness into estimates of maximum credible earthquakes may need revision. The study suggests that hazard assessors should weight other evidence more heavily, including geodetic measurements of plate coupling, the distribution of past ruptures inferred from historical accounts, tsunami deposits and coral microatolls, and the structural segmentation of the margin. It also argues against using roughness to declare any margin incapable of producing a giant earthquake, a conclusion with direct consequences for coastal communities and infrastructure planning along subduction margins worldwide.

The researchers emphasize that their findings do not make roughness irrelevant. Interface roughness still shapes where within a rupture the largest slip occurs, how strong ground shaking is distributed, and possibly the frequency of smaller-to-moderate events. What the study removes is the assumption that roughness imposes a hard ceiling on rupture size. Distinguishing between influences on the distribution of slip and controls on maximum magnitude is, the authors argue, an essential refinement that the field has too often glossed over. Roughness may sculpt the earthquake, but it does not appear to cap it.

The work also speaks to a broader lesson in earthquake science: the danger of building predictive frameworks on a handful of spectacular case studies. The great earthquakes of the past century are few, and any global comparison involving fewer than a dozen giant events is statistically fragile. By compiling a comprehensive, uniformly processed dataset, the new study reduces the risk of overfitting a narrative to memorable examples. The result is a more sober picture of subduction zones: nearly any of them, given enough time to accumulate strain, may be capable of producing ruptures larger than their instrumental records suggest, and the size of the next great earthquake may be far less predictable from the shape of the seafloor than scientists once hoped.

Subject of Research: The relationship between subduction interface roughness and maximum earthquake magnitude

Article Title: Maximum earthquake magnitude unlikely to be controlled by subduction interface roughness

Article References: Yang, X., Bell, R. E., Whittaker, A. C., Xu, H., Han, X., Knowlson, A. R., & Locher, V. A. (2026). Maximum earthquake magnitude unlikely to be controlled by subduction interface roughness. Nature Geoscience. https://doi.org/10.1038/s41561-026-02093-z

Image Credits: AI Generated

DOI: 10.1038/s41561-026-02093-z

Keywords: subduction zones, earthquake magnitude, fault roughness, seismic hazard, plate boundaries, rupture segmentation, great earthquakes, seismology, geodesy, marine geophysics, plate coupling, Nature Geoscience

The core of the argument is deceptively simple. GNSS-based deformation analysis is no longer an emerging technique. Over the past three decades, it has evolved from a promising geodetic experiment into a foundational piece of global observing infrastructure, underpinning everything from earthquake hazard assessment to volcanic monitoring, sea-level studies, and even hydrological drought detection. Landmark studies, such as the 2003 use of one-hertz GPS data to capture ground motions during the Denali fault earthquake, demonstrated decades ago that satellite geodesy could record seismic waves directly. When a technology reaches this level of maturity, the paper contends, claims of methodological novelty deserve especially careful scrutiny, because the vocabulary of innovation can obscure what a study genuinely contributes.

To move beyond anecdote, Almeida conducted a systematic bibliometric search of the Scopus database, targeting articles published between 2020 and 2025 that related to GNSS-based crustal deformation. The search returned 445 studies, a figure that itself illustrates how productive and crowded the field has become. Among those hundreds of papers, 34 articles explicitly used terminology associated with novelty or innovation in their titles, abstracts, or keywords. Those 34 studies were then read and individually classified according to the dimension in which the claim of novelty was actually made, producing a five-part taxonomy of what scientists mean when they call their work new.

The five categories are worth spelling out, because they map directly onto different kinds of scientific value. Category A covers genuine methodological novelty: new algorithms, new processing strategies, or new mathematical frameworks. Category B describes the integration of multiple data sources or methodologies, for example combining GNSS time series with machine learning techniques or fusing satellite positioning with other geophysical observations. Category C captures new scientific applications of established methods, such as repurposing GPS deformation measurements to detect hydrological droughts or assess flood potential. Category D is the new regional case study, applying well-tested tools in a geographic area where they had not previously been used. Category E, finally, covers new datasets or observation networks, the quiet infrastructural contributions that make future science possible.

The results of the classification carry a pointed message. Exactly half of the 34 articles, seventeen papers, were classified as presenting methodological novelty in the strict sense. The other half claimed novelty primarily through new applications, new geographic contexts, integration of existing data streams, or new observational contributions. In other words, when researchers in this mature field reach for the language of innovation, they are as likely to be describing the skillful application, extension, or combination of established methods as they are to be describing a genuinely new technique. Both kinds of contribution are scientifically valuable, the paper stresses, but they are conceptually distinct, and blurring them distorts how readers, reviewers, and funders perceive the state of the field.

The technical substance behind many of the non-methodological papers illustrates the point concretely. Recent studies have used GNSS-derived terrestrial water storage anomalies to detect extreme hydrological drought in the Poyang Lake basin, characterized droughts in Brazil with multiscale GNSS indices, and constrained water storage changes in Yunnan, China, by combining GNSS with GRACE satellite gravimetry. Others have applied machine learning to detect geodynamic anomalies in GNSS time series, introduced sparse modeling into geodetic data inversion to estimate strain-rate fields, or fused GPS displacements with seismic observations to interpret earthquake sequences in Iceland. In each case, the underlying measurement technique and much of the analytical machinery were already established; what changed was the scientific question, the region, or the combination of data sources.

Why does this distinction matter so much? The commentary draws on a long-running debate in innovation studies, citing work that has struggled for decades with the definitional quagmire surrounding terms like innovation and novelty, and on scholarship about responsible language in scientific writing. Words are not neutral in science communication. When every applied study describes itself as innovative, reviewers and editors lose the ability to discriminate between a genuine methodological advance and a competent regional application of a thirty-year-old technique. The innovation narrative, repeated often enough, also misrepresents the maturity of the field itself, making GNSS-based deformation analysis appear earlier in its developmental arc than it actually is. The United Nations Global Geodetic Centre of Excellence’s recent baseline maturity assessment of the geodesy profession provides the broader institutional backdrop: geodesy is now essential infrastructure, and its literature should reflect that reality.

There are practical stakes beyond semantics. Peer review is built on the premise that claims can be evaluated against what a manuscript actually delivers. If a paper promises a novel method but delivers a new regional case study of an existing method, the review process becomes harder, the eventual readers are potentially misled, and the incremental contributions that genuinely advance a mature field risk being undervalued precisely because they were marketed as something they are not. Conversely, the paper argues, precise language would promote balanced recognition: methodological advances would stand out more clearly, while applied, integrative, and observational contributions would receive honest credit for the real and often substantial value they provide. Better terminology, in this view, is not pedantry but a form of scientific quality control.

The study also touches on a question increasingly asked across science: how should novelty be measured at all? A recent Nature comment has called for finding ways to quantify novelty in scientific publications, and Almeida’s five-category classification offers one practical template for doing so within a specific technical domain. By reading the actual contributions of papers rather than their advertised language, the approach shows that the distribution of novelty types can be mapped empirically. Applied more widely, such taxonomies could help journals, databases, and assessment exercises describe research more accurately, and could give young scientists a more honest picture of the many legitimate ways to contribute to a mature discipline, beyond the narrow pursuit of the new.

The commentary ends where the field itself now stands. GNSS-based deformation analysis has delivered an extraordinary record of Earth’s moving surface, and the coming years will see that record extended by denser networks, longer time series, machine-learning-assisted analysis, and integration with complementary observing systems. Methodological innovation will certainly continue, as the seventeen papers in the strict category demonstrate. But the mature phase of a science is defined as much by its patient applications as by its breakthroughs, and the language of the literature should say so. Choosing the right word, the paper suggests, is one of the cheapest and most powerful improvements any researcher can make: it sharpens communication, protects the review process, and gives both breakthrough methods and steady incremental progress the distinct recognition each deserves.

Subject of Research: Innovation and novelty claims in GNSS-based crustal deformation research

Article Title: Rethinking innovation narratives in mature GNSS-based deformation analysis

Article References: Almeida, Y. D. M. (2026). Rethinking innovation narratives in mature GNSS-based deformation analysis. Earth Science Informatics, 19(10), Article 183. https://doi.org/10.1007/s12145-026-02240-5

Image Credits: AI Generated

DOI: 10.1007/s12145-026-02240-5

Keywords: GNSS, GPS, crustal deformation, geodesy, innovation, novelty, scientific communication, bibliometrics, Earth Science Informatics, time series, Rethinking, narratives