Surveys in Geophysics最新文献

The mission of Advanced Ionospheric Probe (AIP) onboard FORMOSAT-5 (F5) satellite is to detect pre-earthquake ionospheric anomalies (PEIAs) and observe ionospheric space weather. F5/AIP plasma quantities in the nighttime of 22:30 LT (local time) and the total electron content (TEC) of the global ionosphere map (GIM) are used to study PEIAs of an M7.3 earthquake in the Iran–Iraq border area on 12 November 2017, as well as signatures of two magnetic storms on 7 and 21–22 November 2017. Statistical analyses of the median base and one sample test are employed to find the characteristics of temporal PEIAs in GIM TEC over the Iran–Iraq area. The anomalous increases of the GIM TEC and F5/AIP ion density over the epicenter area on 3–4 November (day 9–8 before the M7.3 earthquake) agree with the temporal PEIA characteristics that the significant TEC increase frequently appears on day 14–6 before 53 M ≥ 5.5 earthquakes in the area during 1999–2016. The spatial analyses together with odds studies show that the PEIAs frequently appear specifically over the epicenter day 9–8 before the M7.3 earthquake and day 10–9 before a M6.1 earthquake on 1 December, while proponent TEC increases occur at worldwide high latitudes on the two magnetic storm days. The F5/AIP ion velocity uncovers that the PEIAs of the two earthquakes are caused by associated eastward electric fields, and the two positive storm signatures are due to the prompt penetration electric fields.

Very Long Period (VLP) signals with periods longer than 2 s may occur during eruptive or quiet phases at volcanoes of all types (shield and stratovolcanoes with calderas, as well as other stratovolcanoes) and are inherently connected to fluid movement within the plumbing system. This is supported by observations at several volcanoes that indicate a correlation between gas emissions and VLPs, as well as deformation episodes due to melt accumulation and migration that are followed by the occurrence of VLPs. Moment tensors of VLPs are usually characterized by large volumetric components of either positive or negative sign along with possibly the presence of single forces that may result from the exchange of linear momentum between the seismic source and the Earth. VLPs may occur during a variety of volcanological processes such as caldera collapse, phreatic eruptions, vulcanian eruptions, strombolian activity, and rockfalls at lava lakes. Physical mechanisms that can generate VLPs include the inflation and deflation of magma chambers and cracks, the movement of gas slugs through conduits, and the restoration of gravitational equilibrium in the plumbing system after explosive degassing or rockfalls in lava lakes. Our understanding of VLPs is expected to greatly improve in the future by the use of new instrumentation, such as Distributed Acoustic Sensing, that will provide a much denser temporal and spatial sampling of the seismic wavefield. This vast quantity of data will then require time efficient and objective processing that can be achieved through the use of machine learning algorithms.

The potential of blended seismic acquisition to improve acquisition efficiency and cut acquisition costs is still open, particularly with efficient deblending algorithms to provide accurate deblended data for subsequent processing procedures. In recent years, deep learning algorithms, particularly supervised algorithms, have drawn much attention over conventional deblending algorithms due to their ability to nonlinearly characterize seismic data and achieve more accurate deblended results. Supervised algorithms require large amounts of labeled data for training, yet accurate labels are rarely accessible in field cases. We present a self-supervised multistep deblending framework that does not require clean labels and can characterize the decreasing blending noise level quantitatively in a flexible multistep manner. To achieve this, we leverage the coherence similarity of the common shot gathers (CSGs) and the common receiver gathers (CRGs) after pseudo-deblending. The CSGs are used to construct the training data adaptively, where the raw CSGs are regarded as the label with the corresponding artificially pseudo-deblended data as the initial training input. We employ different networks to quantitatively characterize decreasing blending noise levels in multiple steps for accurate deblending with the help of a blending noise estimation–subtraction strategy. The training of one network can be efficiently initialized by transfer learning from the optimized parameters of the previous network. The optimized parameters trained on CSGs are used to deblend all CRGs of the raw pseudo-deblended data in a multistep manner. Tests on synthetic and field data validate the proposed self-supervised multistep deblending algorithm, which outperforms the multilevel blending noise strategy.

Earth/rock-fill dams and embankments are the main water retaining structures in hydraulic projects, and they can effectively resist floods and are of great significance for protecting people's lives and property. Leakage is a common problem in these structures. Investigation activities, including geotechnical, geoelectric, and tracing methods, are required to locate the leakage path and provide a basis for risk mitigation and reinforcement. These three methods provide information on different leakage characteristics, uncertainties, and spatiotemporal distributions. This work first introduces the micro-mechanism of internal erosion and then, provides a site case base for leakage investigation of earth/rock-fill dams and embankments from all over the world. For each investigation method, the basic principle, investigation process, data interpretation, and future potential are summarized. It should be emphasized that geotechnical, geoelectric, and tracing methods are placed on an equal level to assist dam managers and researchers in selecting the most appropriate method to assess dam leakage against specific geological backgrounds and structural types. Finally, the advantages, disadvantages, and applicable conditions of each investigation method are compared. The role of surface investigation methods and internal investigation methods in different stages of leakage is explained. The application of combined methods is discussed at four levels, and a new combined method is proposed.

Two computational strategies for the evaluation of the spherical harmonic coefficients of the gravitational potential due to a generally shaped homogeneous polyhedral source are examined in detail. The techniques are implemented numerically for the known asteroid shape models of Eros and Didymos. The aim of the investigation is to quantify specific numerical aspects of the two algorithms, such as the accuracy of the techniques compared to a closed analytical solution for varying distance between source and computation point, the band-limited spectral analysis of the obtained spherical harmonic models and the convergence behavior of the corresponding series expansion in the vicinity of the characteristic Brillouin sphere. From a computational point of view, the line integral approach demands approximately three times the CPU time of Werner’s method. The two sets of spherical harmonic coefficients are 100% correlated up to degree 45 for Eros and up to degree 49 for Didymos. Approaching degree 100, the correlation by degree decreases by 0.0004% for Eros and by 0.004% for Didymos, the corresponding values for the correlation by order being 0.0002% and 0.304%. Inside the Brillouin sphere and approaching its boundary, the numerical agreement of the gravitational potential between the line integral method and the analytical solution is at the 1E-4 level, while with Werner’s approach at the 1E-7 level. At a distance of 33.5 km outside the Brillouin sphere for Eros and 2.2 km for Didymos, both methods are identical, reaching an agreement level with the analytical solution of 1E-11 level for Eros and 1E-14 for Didymos. In terms of spherical harmonic representation, the series defined by the line integral approach converges faster to the analytical value for the gravitational potential by 4 degrees.

The paper summarizes the issues related to relationships between the PC index and magnetic disturbances: threshold level of the PC index required for the disturbances beginning, delay time in response of magnetic substorms and storms to the PC index growth, relation of PC index to magnetospheric field-aligned currents in course of substorm, different types of magnetic substorms (isolated, expanded, delayed, sawtooth) and magnetic storms (classic, pulsed and composite) and their relation to different regularities in the PC index alterations, linear dependence of the substorm and storm intensities on value of the preceding of PC index, special features of magnetic activity in the winter and summer polar caps, variations of PC index and magnetic disturbances in course of the 23/24 solar activity cycles. New aspects that have arisen due to the PC index application are concerned with the threshold-dependent mode of the substorm development and regular repeateness of sawtooth substorms occurring under conditions of steady powerful E_KL field. The experimental results examined in the paper are indicative that the PC index serves as an indicator of the solar wind energy which comes in the magnetosphere and then realizes in the form of magnetosphere disturbances. This paper follows the review of Troshichev (Front Astron Space Sci 9:1069470, 2022), where the relationships between the solar wind electric field E_KL and PC index have been examined.