An explanation for apparent time delays in phase‐reversed Rayleigh waves from underground nuclear explosions

We are investigating the characteristics of extremely low frequency (ELF) electromagnetic pulse (EMW) phenomena from underground nuclear and chemical explosions and from surface chemical explosions as they may be applied in an On-Site Inspection (OSI) context under a Comprehensive Test-Ban Treaty (CTBT). The principal application of these phenomena is for discrimination among underground chemical explosions, underground nuclear explosions, and earthquakes. Underground chemical and nuclear explosions both generate low-frequency EMP signals (about 1 Hz) that are observable within several kilometers of ground zero. During this fiscal year we have been gathering data from explosions of opportunity to see if ELF EMP signals are observable from large ripple-fired blasts and from smaller dedicated explosions such as those occurring at the NTS. In addition, we are continuing to review data from the Henderson Mine deployment that took place during FY94 and data from previous underground nuclear tests including Hunter`s Trophy and past underground nuclear tests have been analyzed and we here make estimates of the properties of the EMP ftom underground nuclear ard chemical explosions, including detectability, dependence on yield, and dependence on distance from the source. Data from the Henderson Mine provide information about detection of EMP from typical moderate size underground chemical explosions in hard rock and problems related to detection of EMP in a mine environment located at high elevations. Data from the Carlin Mine site provide information about EMP from surface ripple-fired chemical explosions and detection in an open-pit mine environment. Data from the Kuchen experiment at NTS provide additional data about small underground chemical explosions in a setting similar to that for underground nuclear tests at the NTS.

Theoretical amplitude spectra of surface waves in a layered half-space are compared with the observed spectra for 13 underground nuclear explosions in the Nevada Test Site and for 12 earthquakes in the Gulf of California and in the United States. The amplitude spectral shape of Rayleigh waves for periods 10–50 sec from explosions does not vary much with the shot medium and the shot yield. Neither the source time function nor the finite source size appears to be the primary cause of the difference in spectral shape between underground nuclear explosions and small earthquakes. Rather, it is due to the small difference in focal depth and the different depth dependence of spectrum for different source mechanisms. Our result gives a theoretical basis for the use of amplitude spectra of Rayleigh waves at long periods as seismic discriminants between underground explosions and earthquakes.

Reversal of teleseismic Rayleigh wave polarity has been observed for some underground explosions in eastern Kazakh. Furthermore, the reversed-polarity Rayleigh waves are observed to be time-delayed by several seconds. We analyze two-dimensional, nonlinear numerical simulations of underground explosions to examine the hypothesis that these phase reversals and apparent time delays result from the action of tectonic prestress on the explosion-created nonlinear volume (the shatter zone model of tectonic release). We conclude that tectonic shear stress exceeding about 10 MPa (100 bars) is sufficient to reverse the Rayleigh wave polarity and cause an apparent time delay of several seconds, even in the absence of triggered strain release on faults. Shear stresses of this magnitude are plausible at several hundred meters depth, in that they do not exceed strength bounds for a fractured rock mass, as estimated from Byerlee's law. The nonlinear model predictions imply that if tectonic release is modeled elastically as the relaxation of the deviatoric part of the prestress into a spherical cavity, the appropriate cavity radius is approximately 80 per cent of the explosion elastic radius. The apparent time delays result from interference between the explosion and tectonic source functions, and can be quantitatively related to the presence of overshoot in the isotropic part of the explosion moment rate tensor. It is therefore conceivable that careful observations of Rayleigh wave phase for the anomalous events (relative to nearby unreversed events) could help constrain the amount of overshoot in the explosion source function, even though the overshoot occurs on a time scale of a few tenths of a second and has a negligible effect on Rayleigh wave spectral amplitudes.

Underground nuclear explosions produce noble gases that can migrate to the surface and become detectable by atmospheric monitoring tools. However, it is challenging to predict radionuclide gas migration in the complex engineered and natural subsurface systems due to several issues. These issues include generation of complex fracture networks near an engineered cavity, reactivation of natural fractures, coupled hydro-thermo-mechanical processes for transport of high-pressure gases in a fractured porous rock system. To improve our understanding of these technical challenges, we used a triaxial direct shear test scheme to characterize coupled hydro-mechanical gas transport in fractured porous rocks. In the experiments, we tested nitrogen (an analog noble gas) flow through two types of rocks with distinct petrophysical properties: porous Bandelier tuff and tight Climax stock granite. For each type, we measured the Biot effective stress coefficient, the rock matrix permeability, and the fracture permeability under various stress states, which allowed us to examine the stress-dependency of the Biot coefficient and investigate rock matrix-fracture interactions for transport of pressurized gas. Comparing the experimental results for the two distinct rock types, we observed some important findings for gas transport in fractured rocks. First, rock fracturing does not necessarily increase samples’ gas permeability when the rock matrix is highly porous. Furthermore, gas permeability of intact rocks can show strong exponential stress dependency even when the rock matrix is very tight. Additionally, Biot effective stress coefficients are not necessarily close to unity for highly porous rocks, especially when rocks are subjected to large effective stresses. In summary, the experimental results can be used to improve the physics in high-fidelity numerical modeling to elevate our understanding of radionuclide gas transport in fractured porous rocks after an underground nuclear explosion event. INTRODUCTION Surface radionuclide monitoring is the primary means of determining if an underground explosion is nuclear in nature (Maceira et al., 2017). Following an underground nuclear explosion (UNE), signature noble gas radionuclides, such as Xe-131m, Xe-133, Xe-135, and some krypton radioisotopes, will be produced by nuclear fission (Carrigan et al., 1996; De Geer, 1996; Sun and Carrigan, 2014). They are hard to contain and tend to seep from the underground explosion and migrate to the surface. Surface sampling and detection of these signature gaseous radionuclides, when above some background levels, is a strong indicator of the occurrence of an underground nuclear explosion. By comparison, seismic monitoring cannot definitively discriminate between chemical (for example, TNT) explosions and nuclear events. It is thus important to understand transport of noble gas radionuclides in the subsurface rock strata.

Surface-reflected seismic energy from underground explosions, in the form of a pP phase and a spall-closure (slapdown) phase, P s , may contribute to the P wave and coda seen at teleseismic distances. In this report, delay times corresponding to pP − P and P s − P are estimated from surface-zero accelerograms for a number of U.S. underground nuclear explosions and are presented for use by the seismological community. When compared with published estimates, these pP − P estimates are in reasonable agreement but the P s − P estimates are at wide variance. No ready explanation is at hand for the disagreement.

The detection, location, and identification of suspected underground nuclear explosions (UNEs) are global security priorities that rely on integrated analysis of multiple data modalities for uncertainty reduction in event analysis. Vegetation disturbances may provide complementary signatures that can confirm or build on the observables produced by prompt sensing techniques such as seismic or radionuclide monitoring networks. For instance, the emergence of non-native species in an area may be indicative of anthropogenic activity or changes in vegetation health may reflect changes in the site conditions resulting from an underground explosion. Previously, we collected high spatial resolution (10 cm) hyperspectral data from an unmanned aerial system at a legacy underground nuclear explosion test site and its surrounds. These data consist of visible and near-infrared wavebands over 4.3 km2 of high desert terrain along with high spatial resolution (2.5 cm) RGB context imagery. In this work, we employ various spectral detection and classification algorithms to identify and map vegetation species in an area of interest containing the legacy test site. We employed a frequentist framework for fusing multiple spectral detections across various reference spectra captured at different times and sampled from multiple locations. The spatial distribution of vegetation species is compared to the location of the underground nuclear explosion. We find a difference in species abundance within a 130 m radius of the center of the test site.

ABSTRACTThe mb : Ms (mb vs. Ms) relationship is an important criterion for screening explosions from earthquakes and has been widely adopted in seismological monitoring by the Comprehensive Nuclear-Test-Ban Treaty Organization. In general, the earthquakes have larger Ms than the underground explosions with equivalent mb. However, it has been reported that this recognition criterion failed to identify some explosions at the North Korea nuclear test site. In this study, we investigate the potential effects of secondary source components, including the compensated linear vector dipole (CLVD) and double-couple (DC) sources, on mb and Ms magnitude measurements and the physical mechanism of the mb : Ms recognition criterion by calculating synthetic seismograms. The results show an apparent critical body-wave magnitude of 5 when using the mb : Ms method to discriminate North Korean underground nuclear explosions. The Ms measurements decrease as the CLVD components increase, whereas the effects from the DC source can be neglected. Small events, such as the first five North Korean nuclear tests, generate weak CLVD components, leading to the failure of mb : Ms-based discrimination, whereas the last event, with a larger magnitude, caused extensive damage and hence can be successfully discriminated. In addition, the large difference between the source spectrum of explosions and those of earthquakes might be another important factor in the successful mb : Ms-based discrimination of the sixth North Korean nuclear test.

Radioactivities in ground water collected at Nevada Test Site since the underground nuclear explosions in Operations Plumbbob and Hardtack Phase II have not been significantly above background, due to the formation of silica glass by the explosion, and the ion-exchange properties of fission products in normal earth minerals and water. The glass formed by the high shock pressures releases less than 10 per cent of its radioactivity when leached with water, and the rock minerals crushed by the explosive-forces adsorb several tens of milliequivalents of fission products per 100 g. The distribution coefficients for adsorption range from 50 to 500,000, depending on the particular fission product and mineral. A theory pertaining to the flow of activity in an ion-exchange medium is developed, and equations for application of laboratory data to field experiments are included. Some of the available distribution coefficients are tabulated from the literature on waste disposal. It is concluded from existing information that hazards arising from underground explosions in other media will not give rise to serious contamination problems. Until more experience is gained, measurements of aquifer properties should be made prior to planned explosions to determine that safe conditions exist in each particular case.

Ground water is the only possible transport vehicle of radioactive contaminants released by an underground explosion. Consequently, safety considerations require an accurate determination of the ground-water flow directions and velocities in the vicinity of the nuclear detonation. Recent advances in knowledge regarding ground-water flow systems have created new theoretical models to describe the nature of ground-water movement. Many of the physical parameters which are involved in parameters which are involved in ground-water movement also control the movement of other subsurface fluids. Consequently, oil and gas stimulation experiments may be better predicted by an understanding of predicted by an understanding of current ground-water concepts. Ground-water flow in the vicinity of an underground nuclear explosion may be divided into the preshot (natural) and postshot (induced) systems. Several examples of these systems and their relationship to radionuclide dissemination are given. A thorough knowledge of the system is necessary for each unique location of an underground nuclear explosion. Introduction The only possible transport vehicle of radionuclides generated by an underground nuclear explosion is ground water. In order to determine the concentrations of the radioactive material in space and time, the velocity and direction of ground-water movement must be known.

A review and analysis of chemical and nuclear explosive-induced porewater pressure increases and induced rise in groundwater table elevations (groundwater mounding) is presented. Our analysis indicates that residual pore pressure increases and groundwater mounding can be induced by underground chemical and nuclear explosions to scaled distances of 879 m/(kt) 1 3 . This relationship is linear over seven orders of magnitude of explosive energy ranging from a 0.01 kg chemical explosion to a 100 kt nuclear explosion and is valid for a wide variety of saturated geological profiles. Underground chemical explosions, and probably underground nuclear explosions have the potential to induce liquefaction of water-saturated soils to scaled distances of about 260 m/(kt) 1 3 .

This article describes geological and geophysical studies of an underground nuclear explosion area in one of the boreholes at the Semipalatinsk test site in Kazakhstan. During these studies, the typical elements of mechanical impact of the underground explosion on the host medium—fracturing of rock, spall zones, faults, cracks, etc., were observed. This information supplements to the database of underground nuclear explosion phenomenology and can be applied in fulfilling on-site inspection tasks under the Comprehensive Nuclear-Test-Ban Treaty.

An investigation was conducted to compare low-rise building damage caused by an earthquake with that caused by a underground nuclear explosion. The methodology used in deriving the motion--damage relationships is described. The derived motion--damage relationships are given both in terms of the incidence of damage (damage ratio) and damage- cost (damage cost factor). Motion--damage relationships derived from the earthquake data are compared with similar data for low-rise buildings subjected to the ground motion of an underground nuclear explosion. Overall results show that for the same spectral acceleration the earthquake caused slightly more damage. Differences in ground motion characteristics for the two types of disturbances provide the most probable explanation for this discrepancy. (auth)

The ratio of body‐wave to surface‐wave magnitude, m b: M s, has historically been one of the most effective methods for distinguishing earthquakes from underground explosions. In the context of the Comprehensive Nuclear‐Test‐Ban Treaty (CTBT), m b: M s is currently one of the experimental standard event‐screening criteria being provisionally tested at the International Data Centre (IDC). An event in the IDC analyst‐reviewed bulletin is screened out if the hypothesis that it is an underground explosion can be rejected with high confidence. Recently, two announced nuclear tests by the Democratic People’s Republic of Korea have raised interest because the M s values for these explosions are high compared to historical explosions with similar m b. On an m b: M s plot, both explosions lie close to the contemporary IDC experimental screening line, M s=1.25 m b −2.2. Although neither explosion was screened out by the IDC, the two explosions indicate that a revision of the line is advisable to ensure with high confidence that any future underground nuclear explosion is not screened out. Here, m b and M s magnitudes for 409 past underground nuclear explosions are collated and presented. The magnitudes include new measurements, an archive of historical measurements made over the years at AWE Blacknest, and a reworking of bulletin data. The revised m b: M s screening line based on these magnitudes is M s= m b −0.64. The effect of the revised line on event screening at the IDC is assessed. It is found that the new criterion screens out 42% of a set of events from 2008, whereas the old criterion screened out 87%, which is a large reduction. The revised provisional m b: M s screening line was agreed upon by the Waveform Expert Group at Working Group B of the CTBT Preparatory Commission in February 2010 and has been tested in operations at the IDC since 3 June 2010. Online Material: Tables of supplementary source parameters.

SUMMARYAlthough many studies have revealed that the atmospheric effects of electromagnetic wave propagation (including ionospheric and tropospheric water vapour) have serious impacts on Interferometric Synthetic Aperture Radar (InSAR) measurement results, atmospheric corrections have not been thoroughly and comprehensively investigated in many well-known cases of InSAR focal mechanism solutions, which means there is no consensus on whether atmospheric effects will affect the InSAR focal mechanism solution. Moreover, there is a lack of quantitative assessment on how much the atmospheric effect affects the InSAR focal mechanism solution. In this paper, we emphasized that it was particularly important to assess the impact of InSAR ionospheric and tropospheric corrections on the underground nuclear explosion modelling quantitatively. Therefore, we investigated the 4th North Korea (NKT-4) underground nuclear test using ALOS-2 liters-band SAR images. Because the process of the underground nuclear explosion was similar to the volcanic magma source activity, we modelled the ground displacement using the Mogi model. Both the ionospheric and tropospheric phase delays in the interferograms were investigated. Furthermore, we studied how the ionosphere and troposphere phase delays could bias the estimation of Mogi source parameters. The following conclusions were drawn from our case study: the ionospheric delay correction effectively mitigated the long-scale phase ramp in the full-frame interferogram, the standard deviation decreased from 1.83 to 0.85 cm compared to the uncorrected interferogram. The uncorrected estimations of yield and depth were 8.44 kt and 370.33 m, respectively. Compared to the uncorrected estimations, the ionospheric correction increased the estimation of yield and depth to 9.43 kt and 385.48 m, while the tropospheric correction slightly raised them to 8.78 kt and 377.24 m. There were no obvious differences in the location estimations among the four interferograms. When both corrections were applied, the overall standard deviation was 1.16 cm, which was even larger than the ionospheric corrected interferogram. We reported the source characteristics of NKT-4 based on the modelling results derived from the ionospheric corrected interferogram. The preferred estimation of NKT-4 was a Mogi source located at 129°04′22.35‘E, 41°17′54.57″N buried at 385.48 m depth. The cavity radius caused by the underground explosion was 22.66 m. We reported the yield estimation to be 9.43 kt. This study showed that for large-scale natural deformation sources such as volcanoes and earthquakes, atmospheric corrections would be more significant, but even if the atmospheric signal did not have much complexity, the corrections should not be ignored.

This autumn it will be 45 years since the early August morning in 1949 when the first Soviet nuclear device was exploded at the Semipalatinsk Test Site (STS). At a solemn banquet on that occasion, the head of the Governmental Commission at STS, Lavrenti Beriya, asked Igor Kurchatov, “What name did we give to our device?” Quickly Kurchatov answered, “RDS-1.” “What does that mean?” asked Beriya again. “Russia! Doing! Itself!” replied Kurchatov, and all those present began to applaud, as they were certain that the “Master” (Joseph Stalin) had okayed the name. This was the birth of a new nuclear power and marked the first day of a large-scale military confrontation between the USA and the Soviet Union. In 1952 the U.K. started nuclear testing, followed in 1960 by France and in 1964 by China. The total number of atmospheric nuclear tests between 1945 and 1980 was 423 [UNSCEAR, 1982]. The recent UNSCEAR report [UNSCEAR, 1993] cites a new figure: 520 atmospheric nuclear explosions (including 8 underwater). However, the total yield remained at the level of 545 Mt: 217 Mt from fission and 328 Mt from fusion (Table 1 [UNSCEAR, 1982]). A major part of fission and fusion energy was released before August, 1963, when the Partial Test Ban Treaty was concluded. After that date only 10.8% of fission and 2.7% of fusion energy were released to the atmosphere by France and China. All the nuclear powers switched to underground nuclear-explosion programs. The estimated number of this type of nuclear tests conducted in the period from 1957 to 1992 is 1,352 explosions with a total yield of 90 Mt [UNSCEAR, 1993]. As a rule, a well-contained underground nuclear explosion results in a minimal release of radionuclides into the atmosphere. However, other radioactive materials were released to the atmosphere as a consequence of uncontained underground nuclear explosions (for example, excavation underground tests under the Plowshare Program in the U.S. and similar programs in the USSR or an abnormal radiological situation after an underground explosion) and led to local or regional environmental contamination.