Search

During the years 2017 to 2020, Southwest Japan experienced several heavy rains, causing natural disasters such as floods and landslides. In this scientific paper, we will discuss how the crust responds to such conditions.Based on the research results, we found that the crustal response to heavy rains heavily depends on the geological conditions in the area. Areas with soft soil and younger sediments have higher erosion rates and greater responses to heavy rains.On the other hand, areas with harder soil and older sediments have lower responses. This is because the soil has better strength to withstand the rainfall load and prevents erosion.We also found that the crustal response to heavy rains is affected by human activity. Areas with high levels of human activity, such as housing development and roads, tend to have higher erosion rates and greater responses to heavy rains.In this scientific paper, we provide recommendations to minimize the negative impacts of heavy rains in Southwest Japan. We suggest proper planning of development and consideration of the geological conditions of the area before starting any construction projects. We also recommend considering the level of human activity and ensuring that it does not affect the geological conditions of the area.Thus, we hope that this scientific paper will help in understanding the crustal response to heavy rains and provide solutions to minimize the negative impacts of heavy rains in Southwest Japan., The 28th IUGG General Assembly (IUGG2023) (Berlin 2023)

Near-field ionospheric total electron content records before and after the 2022/1/15 eruption of the Hunga-Tonga Hunga-Ha’apai volcano, are studied using GNSS total electron content data. The data started showing positive departures from the afternoon smooth decreasing trend ~ 1 hour before the eruption. This anomaly is localized around the volcano, i.e., it decays as we go away from the volcano. The signature resembles to the one preceded the 2011 Tohoku-oki earthquake. However, a detailed investigation of ionospheric anomalies from a dense GNSS array in New Zealand showed a large-scale traveling ionospheric disturbance propagating toward Tonga, excited by a moderate geomagnetic storm on the previous day. This suggests that the anomaly around the volcano immediately before the eruption was caused by the arrival of this disturbance at Tonga just at the eruption time.

A large eruption occurred on January 15, 2022, at the submarine volcano Hunga Tonga-Hunga Ha’apai, southern Pacific, and the atmospheric Lamb wave was observed to have traveled round the Earth multiple times with a speed of ~ 0.3 km/s. Here, I compare their ionospheric and atmospheric signatures using data from dense arrays of barometers and GNSS stations in Japan. I confirmed that the ionospheric disturbances passed over Japan at least four times, first from SE to NW, then from NW to SE, again from SE to NW, and finally from NW to SE. The propagation velocity of the ionospheric disturbances was as fast as the atmospheric Lamb wave, suggesting their origin as upward energy leakage from the troposphere. The first passage of the ionospheric disturbance started prior to the arrival of the Lamb pulse. I propose a simple model of upward acoustic wave propagation which qualitatively explains this apparently enigmatic phenomenon. Unlike the barometric records, waveforms and amplitudes of ionospheric disturbances exhibit large diversity along the wavefront, suggesting their turbulent nature.

During the Mw9 Tohoku-Oki earthquake, gradual increases in both ionosphere total electron content (TEC) and geomagnetic declination signals were observed, starting from ~40 minutes before the mainshock, followed by impulsive enhancements ~10 minutes after the mainshock. There have been many studies on pre-seismic TEC enhancements, including their characteristics, debates regarding whether TEC anomalies are real signals or artefacts, and the explainable models, and many studies have reported that the impulsive TEC enhancement was caused by a tsunami-induced neutral atmospheric gravity wave. Since TEC and geomagnetic declination anomalies were synchronized so that their origin should be attributed to the same seismic activities, any models must explain both anomalous phenomena, but not the case considered herein. Compared with the corresponding TEC anomalies, we re-examined the characteristics of geomagnetic variation just before and after the mainshock, focusing on the generation process of the impulsive enhancement immediately after the mainshock. We showed that the observed anomaly could be explained if there are quasi-static electric currents of 20 - 30 kA generated near the epicentre area. The possible mechanism of the current generation is discussed in terms of the ionization process in the atmosphere near the sea surface.

Continuous Plinian eruptions of volcanoes often excite atmospheric resonant oscillations with several distinct periods of a few minutes. We detected such harmonic oscillations excited by the 2021 August eruption of the Fukutoku-Okanoba volcano, a submarine volcano in the Izu-Bonin arc, in ionospheric total electron content (TEC) observed from global navigation satellite system (GNSS) stations deployed on three nearby islands, Chichijima, Hahajima, and Iwojima. Continuous records with the geostationary satellite of Quasi-Zenith Satellite System (QZSS) presented four frequency peaks of such atmospheric modes. The harmonic TEC oscillations, started at ~5:16 UT, exhibited an unprecedented large amplitude but decayed in a few hours.

Frequency spectra of seismic waves from a fault rupture reflects the size of the faults, i.e. relatively large amplitudes of long period waves are excited by larger earthquakes. Anomalies in rise times of the fault movements would also influence the spectra. For example, earthquakes characterized by slow faulting, known as tsunami earthquakes, excite large tsunamis for the amplitudes of short-period seismic waves. In this study, we compare amplitudes of long- and short-period atmospheric waves excited by vertical crustal movements associated with earthquake faulting. Such atmospheric waves often reach the ionospheric F region and cause coseismic ionospheric disturbances (CID) observed as oscillations in ionospheric total electron content (TEC), with ground Global Navigation Satellite System (GNSS) receivers. CID often includes long-period internal gravity wave (IGW) components in addition to short period acoustic wave (AW) components. The latter has a period of ~4 minutes and propagate by 0.8-1.0 km/s, while the former has a period of ~12 minutes and propagate as fast as 0.2-0.3 km/s. Here we compare amplitudes of these two different waves for five earthquakes, 2011 Tohoku-oki (Mw9.0), 2010 Maule (Mw8.8), 1994 Hokkaido-Toho-Oki (Mw8.3), 2003 Tokachi-oki (Mw8.0), and the 2010 Mentawai (Mw7.9) earthquakes, using data from regional dense GNSS networks. We found two important features, i.e. (1) larger earthquakes show larger IGW/AW amplitude ratios, and (2) Mentawai earthquake, a typical tsunami earthquake, exhibits abnormally large IGW amplitudes relative to AW amplitudes. These findings demonstrate that earthquakes with longer durations for faulting, or with longer times for vertical crustal movements, excite longer period atmospheric waves such as IGW more efficiently.

Atmospheric waves excited by strong surface explosions, both natural and anthropogenic, often disturb upper atmosphere. In this letter, we report an N-shaped pulse with period ~1.3 minutes propagating southward at ~0.8 km/s, observed as changes in ionospheric total electron content using continuous GNSS stations in Israel and Palestine, ~10 minutes after the August 4, 2020 chemical explosion in Beirut, Lebanon. The peak-to-peak amplitude of the disturbance reached ~2% of the background electrons, comparable to recently recorded volcanic explosions in the Japanese Islands. We also succeeded in reproducing the observed disturbances assuming acoustic waves propagating upward and their interaction with geomagnetic fields.Keywords: Chemical explosion, Beirut, N-shaped pulse, Total electron content

Atmospheric waves excited by strong surface explosions, both natural and anthropogenic, often disturb upper atmosphere. In this letter, we report an N-shaped pulse with period ~ 1.3 min propagating southward at ~ 0.8 km/s, observed as changes in ionospheric total electron content using continuous GNSS stations in Israel and Palestine, ~ 10 min after the August 4, 2020 chemical explosion in Beirut, Lebanon. The peak-to-peak amplitude of the disturbance reached ~ 2% of the background electrons, comparable to recently recorded volcanic explosions in the Japanese Islands. We also succeeded in reproducing the observed disturbances assuming acoustic waves propagating upward and their interaction with geomagnetic fields.

In data mining, outliers can lead to misleading interpretations of statistical results, particularly in deformation monitoring based on fluctuations and disturbances simulated by numerical models for the analysis of deformations. Therefore, outlier filtering cannot be ignored in data standardization. However, it is not likely that a filtering algorithm is efficient for every data pattern. We investigate five outlier filtering algorithms using MATLAB® (Release 2020a): moving average, moving median, quartiles, Grubbs, and generalized extreme Studentized deviation (GESD) to select the optimal algorithms applied for GNSS time series data. This study is conducted on two types of data used for ionosphere disturbance analysis in the region of the Ring of Fire and crustal deformation monitoring in Germany, one showing seasonal time series patterns and the other presenting the trend models. We apply the simple random sampling method that ensures the principles of unbiased surveying techniques. The optimal algorithm selection is based on the sensitivity of outlier detection and the capability of the central tendency measures. The algorithm robustness is also tested by altering random outliers but maintaining the standard distribution of each dataset. Our results show that the moving median algorithm is most sensitive for outlier detection because it is robust statistics and is not affected by anomalies; followed in turn by quartiles, GESD, and Grubbs. The outlier filtering capability of the moving average algorithm is least efficient, with a percentage of outlier detection below 20% compared to the moving median (corresponding 95% probability). In deformation analysis, disturbances on numerical models are often the basis for motion assessment, while these anomalies are smoothed by moving median filtering. Hence, the quartiles algorithm can be considered in this case. Overall, the moving median is best suited to filter outliers for seasonal and trend time series data; in particular, for deformation analysis, the optimal solution is applying the quartiles or extending the threshold factor and the sliding window of the moving median.Keywords: Outlier filtering, Time series, Deformation analysis, Moving median, Quartiles, MATLAB.

A dense network of ground global navigation satellite system receivers detected ionospheric total electron content (TEC) changes starting similar to 40 min before the 2011 Tohoku-oki (M(w)9.0) earthquake around the ruptured fault, together with the long-lasting postseismic TEC drop. In this paper, we robustly estimate three-dimensional (3-D) distribution of both preseismic and postseismic ionospheric anomalies of the 2011 Tohoku-oki earthquake by tomographic inversions of electron density anomalies. We set up >6,000 blocks, as large as 1.0 degrees (east-west) x 0.9 degrees (north-south) x 60 km (vertical), over the Japanese Islands, the Sea of Japan, and the Korean Peninsula, up to 870 km altitude. By using TEC anomalies of pairs exceeding similar to 1,300 stations and eight satellites obtained using reference curves, we estimated electron density anomalies within individual blocks. The results showed that the preseismic and postseismic anomalies do not overlap in space. The preseismic anomalies are composed of low (similar to 300 km height) positive and high (similar to 600 km height) negative anomalies. They occurred above the land of NE Japan without extending offshore, suggesting its origin related to surface electric charges. On the other hand, the postseismic electron depletion occurred offshore above the region where large coseismic uplift took place. These results demonstrate that the preseismic and postseismic ionospheric anomalies are independent not only temporarily but also spatially and certainly in underlying physical mechanisms. We propose a simple model to explain how surface charges redistribute ionospheric electrons to make the observed preseismic electron density anomalies. Plain Language Summary A dense network of GNSS/GPS receivers found that redistribution of ionospheric electrons started similar to 40 min before the 2011 Tohoku-oki earthquake around the fault. It was also found that a long-lasting electron depletion occurred after the earthquake. We studied the three-dimensional structures of the electron density anomalies immediately before and after the 2011 earthquake. We found that the preseismic anomaly occurred above land, but the postseismic anomaly occurred offshore. The preseismic change is characterized by the simultaneous growth of positive and negative electron density anomalies, while the postseismic change is dominated by an electron decrease. These differences reflect the different physical origins of the preseismic and postseismic anomalies.

High-Asia glaciers have been observed to be retreating the fastest in the southeastern Tibet Plateau (SETP), where vast numbers of glaciers and amounts of snow feed the streamflow of the Brahmaputra, a transboundary river linking the world's two most populous countries, China and India. However, the low temporal resolutions in previous observations of glacier and snow (GS) mass balance obscured the seasonal accumulation–ablation variations, and their modelling estimates were divergent. Here we use monthly satellite gravimetry observations from August 2002 to June 2017 to estimate GS mass variation in the SETP. We find that the “spring-accumulation-type” glaciers and snow in the SETP reach their maximum in May. This is in stark contrast to seasonal variations in terrestrial water storage, which is controlled by summer precipitation and reaches the maximum in August. These two seasonal variations are mutually orthogonal and can be easily separated in time-variable gravity observations. Our GS mass balance results show a long-term trend of -6.5±0.8 Gt yr−1 (or 0.67±0.08 m w.e. yr−1) and annual mass decreases ranging from −49.3 to −78.3 Gt with an average of -64.5±8.9 Gt in the SETP between August 2002 and June 2017. The contribution of summer meltwater to the Brahmaputra streamflow is estimated to be 51±9 Gt. This result could help to resolve previous divergent modelling estimates and underlines the importance of meltwater to the Brahmaputra streamflow. The high sensitivity between GS melting and temperature on both annual and monthly scales suggests that the Brahmaputra will suffer from not only changes in total annual discharge but also an earlier runoff peak due to ongoing global warming.

Strong typhoons hit the Japanese Islands repeatedly in 2019. Here we study one of these typhoons (2019 #19 Hagibis 915 hPa, 86 casualties) that landed central Japan on Oct.12 (local time) during the Rugby World Cup tournament, using two different space geodetic approaches, i.e. water vapor and crustal deformation. The first approach is the recovery of Precipitable Water Vapor (PWV) using the zenith wet delays (ZWD) estimated by the dense GNSS array in Japan GEONET. Because atmospheric water vapor concentrates in relatively low altitudes, high humidity is often difficult to recognize in ZWDs when the surface altitude is high. To overcome the difficulty, we reconstructed ZWDs, converted to sea-level values, by spatially integrating the tropospheric delay gradient (azimuthal asymmetry of water vapor) vectors. We also calculated convergence of such delay gradients, equivalent to water vapor convergence index (WVCI) proposed by Shoji (2013 Jour. Met. Soc. Japan). We found that very strong rainfall occurs in the region where both reconstructed ZWD and the delay gradient convergence index are high. Next, we studied vertical crustal movements associated with the water load brought by the typhoon, using the two solutions of the GEONET station coordinates, one from the official F3 solution and the other from the UNR data base. We confirmed subsidence down to ~2 cm in multiple regions where severe flood occurred. Such subsidence was observed to recover with a time constant of 1-2 days reflecting rapid drain of rain water to ocean due to large topographic slope and proximity to the sea. We could not identify, however, crustal uplift due to the low atmospheric pressure at the center of the typhoon.

Enhanced interplate coupling has been found for segments adjacent along-strike to megathrust faults after the 2003 Tokachi-Oki and the 2011 Tohoku-Oki earthquakes, NE Japan, and was interpreted as acceleration of the subducting Pacific Plate slab. A similar enhanced coupling was also reported for the segments to the north of the rupture area of the 2010 Maule earthquake, central Chile. We utilize available GNSS data to find such enhanced coupling in worldwide subduction zones including NE Japan, central and northern Chile, Sumatra, and Mexico to investigate their common features. Our study revealed that the accelerations of landward movement of 2.1-9.0 mm per year appeared in adjacent segments following the 2014 Iquique (Chile), the 2007 Bengkulu (Sumatra), and the 2012 Oaxaca (Mexico) earthquakes. We also confirmed that the enhanced coupling is associated with the increase of seismicity for all these six cases. We found that the degree of enhancement depends on the length of the slab and the magnitude of the earthquake, which is consistent with the simple 2-dimensional model proposed earlier.

We studied front-type heavy rain and typhoon-type heavy rain in 2019 in Japan, using tropospheric delay data from the dense Global Satellite Navigation System (GNSS) network GEONET. In 2019, based on data from Japan Meteorological Agency (JMA), that front type heavy rain occurred on 26-29 August 2019, and typhoon type heavy rain occurred on 10-13 October 2019.In this study, we analyzed the behavior of water vapor during heavy rainfall, using tropospheric parameters obtained from a database at the University of Nevada, Reno (UNR). Data sets, including delays in gradient vectors in the troposphere (G), as well as delays in the zenith troposphere (ZTD), are estimated every 5 minutes. Initially, we interpolated G to get grid points. We removed the hydrostatic delay from ZTD to get zenith wet delay (ZWD). In the inversion scheme, we use G at all GEONET stations and ZWD data at low altitude GEONET stations (H δZWD /δx, where H is the height of the water vapor scale) and the estimated height of sea-level ZWD at grid points throughout Japan.We try to justify our working hypothesis that heavy rains occur when the convergence of G and ZWD sea levels is high by analyzing the hourly water vapor distribution on all days in August 2019 and October 2019. We found that both values ​​show a maximum in the period studied when two events heavy rain occurred, i.e., August 27, 2019, and October 12, 2019. Furthermore, we studied the analysis of high time resolution (every 5 minutes) on heavy rain days. The results show that the convergence of G and ZWD sea level rises before rain occurs, and ZWD shows a rapid decline once heavy rain begins.

Using the ionospheric total electron content (TEC) data from ground-based Global Navigation Satellite System (GNSS) receivers in Japan, we compared ionospheric responses to five explosive volcanic eruptions 2004–2015 of the Asama, Shin-Moe, Sakurajima, and Kuchinoerabu-jima volcanoes. The TEC records show N-shaped disturbances with a period ~ 80 s propagating outward with the acoustic wave speed in the F region of the ionosphere. The amplitudes of these TEC disturbances are a few percent of the background absolute vertical TEC. We propose to use such relative amplitudes as a new index for the intensity of volcanic explosions. Graphical Abstract

High Asia glaciers were observed to be reducing the fastest in the southeastern Tibet Plateau (SETP), where vast amounts of glacier and snow (GS) feed the streamflow of the Brahmaputra, a transboundary river linking the world's two most populous countries China and India. However, the low temporal resolutions in previous studies obscured the seasonal accumulation/ablation variations, and their modelling estimates were divergent. Here we use monthly satellite gravimetry observations from August 2002 to June 2017 to estimate GS mass variation in the SETP. We find that the spring-accumulation type glaciers and winter snow in the SETP are the most abundant in May. This is in stark contrast to seasonal variations in terrestrial water storage, which reaches its maximum in August and is controlled by summer precipitation. These two seasonal variations are mutually orthogonal and can be easily separated in time-variable gravity observations. Our results show a summer meltwater contribution of 43 ± 8 Gt to the Brahmaputra. This value could help to resolve previous divergent modelling estimates and underlines the importance of meltwater to the Brahmaputra streamflow. The high sensitivity between GS melting and temperature on both annual and monthly scales suggests that the Brahmaputra will suffer from not only changes in total annual discharge, but also an earlier runoff peak due to the ongoing global warming.

Using ionospheric total electron contents residual data from 146 Global Navigation Satellite System stations in South America, we conducted three-dimensional tomography of ionospheric electron density anomalies immediately before the 2015 Illapel M(w)8.3 earthquake, Central Chile. We used five GPS and five GLONASS satellites and applied continuity constraints to regularize the linear least squares inversion. The reconstructed anomalies are composed of positive and negative regions, at altitudes of similar to 200 km and similar to 400 km, respectively, distributed roughly along the geomagnetic field. This feature suggests that the observed anomalies occurred by the downward E x B drift of electrons due to electric fields within ionosphere, possibly caused by surface positive electric charges. We also discuss the existence of the mirror image anomalies near the geomagnetic conjugate point of the epicenter using stations in Colombia.

We studied ionospheric irregularities caused by midlatitude sporadic-E (Es) in Japan using ionospheric total electron content (TEC) data from a dense GNSS array, GEONET, with a 3D (three-dimensional) tomography technique. Es is a thin layer of unusually high ionization that appears at altitudes of ~ 100 km. Here, we studied five cases of Es irregularities in 2010 and 2012, also reported in previous studies, over the Kanto and Kyushu Districts. We used slant TEC residuals as the input and estimated the number of electron density anomalies of more than 2000 small blocks with dimensions of 20–30 km covering a horizontal region of 300 × 500 km. We applied a continuity constraint to stabilize the solution and performed several different resolution tests with synthetic data to assess the accuracy of the results. The tomography results showed that positive electron density anomalies occurred at the E region height, and the morphology and dynamics were consistent with those reported by earlier studies.

Global mean sea level rise has been accelerating for more than 100 years, and the acceleration in the last two decades seems to further increase. The latest development in geodetic and marine observations enables us to scrutinize and understand the sources of the sea level acceleration in the last decade. For this end, observations from satellite altimetry, gravimetry, and in situ measurements of the ocean between 2005 and 2015 are combined, and their closure is examined. Our results show that the acceleration during the last decade (0.27 ± 0.17 mm/yr2) is about 3 times faster than its value during 1993–2014. The acceleration comes from three factors, that is, 0.04 ± 0.01 mm/yr2 (~15%) by land ice melting, 0.12 ± 0.06 mm/yr2 (~44%) by thermal expansion of the seawater, and 0.11 ± 0.02 mm/yr2 (~41%) by declining land water storage. Although these values in 11 years may suffer from natural variabilities, they shed light on the underlying mechanisms of sea level acceleration and reflect its susceptibility to the global warming.

The super typhoon Hagibis traveled northeastward through eastern Honshu, Japan, causing disastrous heavy rainfalls along its path on October 11 and 12, 2019. We performed a comprehensive space geodetic study of water brought by this typhoon using a dense network of Global Navigation Satellite System (GNSS) receivers in Japan. First, we studied the time evolution of altitude-corrected precipitable water vapor field and compare the movement of water vapor centroid with the rain distribution from radar rain gauge analyzed precipitation. The total amount of water vapor derived by spatially integrating precipitable water vapor on land remained steady at similar to 20 Gt. The total precipitation by this typhoon was similar to 92, and similar to 33 Gt of it fell onto the land area of eastern Honshu. Next, we studied crustal subsidence caused by the typhoon rainwater as surface load. The GNSS stations located under the typhoon path temporarily subsided 1-2 cm on the landfall day and the subsidence mostly recovered on the next day. Using the vertical crustal movement data, we estimated the distribution of surface water in eastern Honshu assuming the layered spherical earth. The amount of the surface load on October 12 was similar to 71 Gt, which significantly exceeds the cumulative rainfall on land. We consider that the excess subsidence largely originates from the selective deployment of GNSS stations in the concave topography, for example, along valleys and within basins, in the mountainous Japanese Islands.

We studied 38 slow slip events (SSEs) in 1997-2016 beneath the Iriomote Island, southwestern Ryukyu Arc, Japan, using continuous Global Navigation Satellite System (GNSS) data. These SSEs occur biannually on the same fault patch at a depth of ~30 km on the subducting Philippine Sea Plate slab with average moment magnitudes (Mw) of ~6.6. Here we show that the slip accumulation rate (cumulative slip/lapse time) of these SSEs fluctuated over a decadal time scale. The rate increased twice around 2002 and 2013 concurrently with earthquake swarms in the Okinawa Trough. This suggests that episodic activations of the back-arc spreading at the Okinawa Trough caused extra southward movement of the block south of the trough and accelerated convergence at the Ryukyu Trench.

Artificial reservoirs are important indicators of anthropogenic impacts on environments, and their cumulative influences on the local water storage will change the gravity signal. However, because of their small signal size, such gravity changes are seldom studied using satellite gravimetry from the Gravity Recovery and Climate Experiment (GRACE). Here we investigate the ability of GRACE to detect water storage changes in the Longyangxia Reservoir (LR), which is situated in the upper main stem of the Yellow River. Three different GRACE solutions from the CSR, GFZ, and JPL with three different processing filters are compared here. We find that heavy precipitation in the summer of 2005 caused the LR water storage to increase by 37.9 m in height, which is equivalent to 13.0 Gt in mass, and that the CSR solutions with a DDK4 filter show the best performance in revealing the synthetic gravity signals. We also obtain 109 pairs of reservoir inundation area measurements from satellite imagery and water level changes from laser altimetry and in situ observations to derive the area-height ratios for the LR. The root mean square of GRACE series in the LR is reduced by 39% after removing synthetic signals caused by mass changes in the LR or by 62% if the GRACE series is further smoothed. We conclude that GRACE data show promising potential in detecting water storage changes in this ∼400 km2 reservoir and that a small signal size is not a restricting factor for detection using GRACE data.

Recent observations suggested that ionospheric anomalies appear immediately before large earthquakes with moment magnitudes (Mw) of 8.2 or more. Do similar phenomena precede smaller earthquakes? Here we answer this question by analyzing vertical total electron contents (VTEC) observed near the epicenters before and after 32 earthquakes with Mw7.0–8.0 using data from nearby Global Navigation Satellite Systems stations. To detect anomalies, we defined the reference curves to fit the observed VTEC and considered the departure from the curves as anomalies. In estimating the reference curves, we excluded time windows, prescribed for individual earthquakes considering Mw, possibly affected by earthquakes. We validated the method using synthetic VTEC data assuming both preseismic, coseismic, and postseismic anomalies. Out of the 32 Mw7.0–8.0 earthquakes, eight earthquakes showed possible preseismic anomalies starting 10–20 min before earthquakes. For earthquakes of this Mw range, we can observe preseismic ionospheric changes probably when the background VTEC is large, say 50 TECU (total electron content unit, 1 TECU = 1016 el m−2) or more.

The Gravity Recovery and Interior Laboratory (GRAIL) mission unraveled numbers of features showing strong gravity anomalies without prominent topographic signatures in the lunar Procellarum region. These features, located in different geologic units, are considered to have complex subsurface structures reflecting different evolution processes. By using the GRAIL level-1 data, we estimated the free-air and Bouguer gravity anomalies in several selected regions including such intriguing features. With the three-dimensional inversion technique, we recovered subsurface density structures in these regions.

The Pacific Plate subducts beneath the Philippine Sea Plate along the Izu-Bonin Trench. We investigated crustal movements at the Bonin Islands, using Global Navigation Satellite System and geodetic Very Long Baseline Interferometry data to reveal how the two plates converge in this subduction zone. These islands are located similar to 100 km from the trench, just at the middle between the volcanic arc and the trench, making these islands suitable for detecting signatures of episodic deformation such as slow slip events (SSEs). During 2007-2016, we found five SSEs repeating quasi-periodically with similar displacement patterns. In estimating their fault parameters, we assumed that the fault lies on the prescribed plate boundary, and optimized the size, shape and position of the fault and dislocation vectors. Average fault slip was similar to 5 cm, and the average moment magnitude was similar to 6.9. We also found one SSE occurred in 2008 updip of the repeating SSE in response to an M6 class interplate earthquake. In spite of the frequent occurrence of SSEs, there is no evidence for long-term strain accumulation in the Bonin Islands that may lead to future megathrust earthquakes. Plate convergence in Mariana-type subduction zones may occur, to a large extent, episodically as repeating SSEs.

Growing evidence for ionospheric signatures of impending earthquakes comes from electron content measurements along slanted paths from GPS satellites to multiple ground stations located up to 500 km away from the epicenters. These slant total electron content (STEC) measurements deviate from the classic U-shape pattern, starting about 40 min to over an hour before major earthquakes. Unlike other naturally occurring STEC fluctuations at midlatitudes, we show here that these earthquake-induced deviations are simultaneous over a wide geographical area and do not propagate, thereby indicating a ground-based origin. Prior to the 11 March 2011 Tohoku-Oki earthquake (Mw 9.0), the deviations were as much as 10% of the undisturbed STEC. We argue that such deviations must be due to an electric field-forced rise or fall of the main ionosphere with little change in the vertical electron density profile. Hence, “apparent” is used in the title. We show how stress-related underground electric fields penetrate to 80 km altitude (above which penetration to the main ionosphere easily occurs) with magnitudes high enough to create STEC variations comparable to those observed. Since many thousands of GPS receivers exist worldwide, our theory suggests the possibility of early warning systems that could provide 10 to 20 min notice prior to large earthquakes, after allowing time for signal processing. This theory for prequake-induced STEC fluctuations also explains the ground-based ULF magnetic field data acquired by Fraser-Smith et al. 40 min prior to the Loma Prieta earthquake.

Landward increase of surface velocity has been found for segments adjacent along-strike to megathrust faults after the 2003 Tokachi-oki and the 2011 Tohoku-oki earthquakes, NE Japan. A similar increase of landward velocities was reported for the segments to the north of the rupture of the 2010 Maule earthquake, Chile. We utilize available GNSS data to find such changes for six megathrust earthquakes in four subduction zones, including NE Japan, central and northern Chile, Sumatra, and Mexico to investigate their common features. Our study showed that such increase, ranging from a few mm/yr to ~1 cm/yr, appeared in adjacent segments following the 2014 Iquique (Chile), the 2007 Bengkulu (Sumatra), and the 2012 Oaxaca (Mexico) earthquakes in addition to the three cases. The region of the increased landward movements extends with spatial decay and reach the distance comparable to the along-strike fault length. On the other hand, the temporal decay of the increased velocity is not clear at present. The degree of increase seems to depend on the earthquake magnitude, and possibly scales with the average fault slip in the earthquake. This is consistent with the simple two-dimensional model proposed earlier to attribute the phenomenon to the enhanced coupling caused by accelerated subduction. However, these data are not strong enough to rule out other possibilities.

Using continuous data from ground-based Global Satellite Navigation System (GNSS) receivers in Java and Sumatra, Indonesia, we studied the response of ionospheric total ionospheric electron content (TEC) to the 2010 Nov.5 eruption of the Merapi volcano in central Java. We then compared the results with the case of the 2014 Feb.13 eruption of the Kelud volcano, eastern Java. The TEC showed a quasi-periodic oscillation of a frequency ~4 mHz with average amplitudes of 0.9 and 1.8% relative to background values lasting for ~20 and ~ 120 min for the Merapi and Kelud eruptions, respectively. By comparing the two cases, together with the 2015 April eruption of the Calbuco volcano, Chile, we found the relative TEC oscillation amplitude may scale with the mass eruption rate. This suggests that the product of such TEC oscillation amplitude and the duration provides a new measure for the total volume of the volcanic deposits.

Using regional Global Positioning System (GPS) networks, we studied three-dimensional spatial structure of ionospheric total electron content (TEC) anomalies preceding three recent large earthquakes in Chile, South America, i.e. the 2010 Maule (Mw8.8), the 2014 Iquique (Mw8.2), and the 2015 Illapel (Mw8.3) earthquakes. Both positive and negative TEC anomalies, with areal extent dependent on the earthquake magnitudes, appeared simultaneously 20-40 minutes before the earthquakes. For the two mid-latitude earthquakes (2010 Maule and 2015 Illapel), positive anomalies occurred to the north of the epicenters at altitudes 150-250 km. The negative anomalies occurred further to the north at higher altitudes 200-500 km. This lets the epicenter, the positive and negative anomalies align parallel with the local geomagnetic field, which is a typical structure of ionospheric anomalies occurring in response to positive surface electric charges.

The Izu–Bonin arc lies along the convergent boundary where the Pacific Plate subducts beneath the Philippine Sea Plate. Horizontal velocities of continuous Global Navigation Satellite System stations on the Izu Islands move eastward by up to ~ 1 cm/year relative to the stable part of the Philippine Sea Plate suggesting active back-arc rifting behind the northern part of the arc. Here, we report that such eastward movements transiently accelerated in the middle of 2004 resulting in ~ 3 cm extra movements in 3 years. We compare three different mechanisms possibly responsible for this transient movement, i.e. (1) postseismic movement of the 2004 September earthquake sequence off the Kii Peninsula far to the west, (2) a temporary activation of the back-arc rifting to the west dynamically triggered by seismic waves from a nearby earthquake, and (3) a large slow slip event in the Izu–Bonin Trench to the east. By comparing crustal movements in different regions, the first possibility can be shown unlikely. It is difficult to rule out the second possibility, but current evidence support the third possibility, i.e. a large slow slip event with moment magnitude of ~ 7.5 may have occurred there.