Source Depth Research Articles

SEISMICITY of KAMCHATKA and COMMANDER ISLANDS in 2016–2017

The seismicity review of Kamchatka and surrounding territories for 2016–2017 is given. In the Kamchatka earthquake catalogue, the minimum local magnitude of completeness is MLmin=4.0, and for earth-quakes with h≥350 km under the Okhotsk sea MLmin=4.3. The Kamchatka catalogue of earthquakes with ML=3.6–7.3 for 2016–2017, published in the Appendix to this article, includes 2898 events. 191 earthquakes of the catalogue were felt in Kamchatka and surrounding areas with seismic intensity I=2–6 according to the MSK-64 scale. For all events with ML≥5.0 that occurred in 2016–2017 within the area of responsibility of Kamchatka branch of Geophysical Survey RAS, an attempt to calculate the seismic moment tensor (SMT) was made. There are 109 such events in the regional catalogue. For 102 earthquakes the SMT and depth of the equivalent point source were calculated. The calculations were performed for the SMT double-couple model using a nonlinear algorithm. The level of seismicity according to the “SOUS'09” scale in 2016 corresponds to the “background increased”, for 2017 it is also the “background increased”, but within the assessment accuracy – “high”. In 2016–2017 within the Kamchatka branch area of responsibility, an atypical pattern of the location of earthquake epicenters was observed due to the occurrence of two strong events – the Near Aleutian (Mw=7.8, July 17, 2017) and South Ozernovsky (Mw=6.6, March 29, 2017) earthquakes and their aftershock processes.

Using geophysics to follow and model the Precambrian basement terranes beneath the Caledonian nappes in Finnmark, northern Norway: A case study

The bedrock geology of Finnmark, northern Norway, consists primarily of rock complexes dating back to the Precambrian and early Mid Palaeozoic eras and comprises (1) a mid-crustal lithospheric basement which is characterised by granitoid and greenstone complexes dating from Neoarchean to late Paleoproterozoic in age; (2) the Caledonide Orogen, characterised by a series of nappes and thrust sheets. Finnmark forms part of the northern margin of the Fennoscandian Shield, which extends across northern Finland into northwestern Russia. Precambrian rock complexes are also found as thrust slices in Caledonian nappes at several places in Finnmark and Troms, as well as in tectonic windows.Recently acquired high-resolution aeromagnetic data have provided spectacular and confirmatory evidence for the continuation of the diverse Precambrian greenstone belts and granulite terranes beneath the magnetically transparent Caledonian nappes.This study aims at analysing this subsurface continuation of the terranes using potential field data and estimating the depths of the main magnetic sources beneath the nappes. In this study, several geophysical methods are used, including 3D Euler deconvolution and 2D Werner deconvolution. We have focused specifically on the depth-to-basement beneath nappes and purposefully excluded anomalies associated with shallow structures, such as dykes and faults. Different methods have produced similar results and indicate an appreciable increase in the depth of the Precambrian basement rocks from south to north. In the south, Precambrian basement rocks are either exposed on the surface southeast of the Caledonian front or reappear within tectonic windows. A prominent positive magnetic anomaly on the Porsanger Peninsula is thought to represent an extension of the anomaly ascribed to the greenstone belt of the Repparfjord Tectonic Window. A 3D model of this has also been provided. According to estimates, the depths-to-basement below nappes close to the Caledonian front range from less than 500 to 2000 m, whilst in the northern part of the study area the depths generally reach 3000 to 4000 m, and are locally estimated to range between 5000 m and 6000 m.

Influence of Range-Dependent Sound Speed Profile on Position of Convergence Zones

Based on the Wentze–Kramers–Brillouin approximation, we derive formulae to calculate the position of convergence zones in a range-dependent environment with sound speed profiles varying in linear and ellipsoidal Gaussian eddy cases. Simulation results are provided for the linear and ellipsoidal Gaussian eddy cases. Experiment data are used for calculations considering linearly varying sound speed, and the findings suitably agree with the simulation results. According to the evaluated environment, the influence of the range-dependent sound speed profile on the position of the upper and lower convergence zones for different source depths is analyzed through simulations. The corresponding results show that the influence of the sound speed profile on the position of the upper convergence zone is greater for a shallower source. In contrast, the position of the lower convergence zone for large-depth reception is less affected.

Transformations of borehole magnetic data in the frequency domain and estimation of the total magnetization direction: A case study from the Mengku iron-ore deposit, Northwest China

Borehole magnetic prospecting measures the three components of the magnetic field and is sensitive to the vertical depth of the magnetic source, which plays an important role in deep mineral exploration. Magnetic field transformations in the frequency domain constitute common and important processing for ground and airborne data but are rarely applied to borehole magnetic data. Here, we deduce component transformation and magnetization-direction transformation formulas for borehole data. It is found that the transformation factors differ from those of ground and airborne data. The new formulas depend on the position of the drillholes. The transformed data are then used to estimate the total magnetization direction by computing the correlation coefficient between the transformed components and the amplitude anomaly. The advantages of this method are that the borehole data yield a local and accurate magnetization direction and the total amplitude anomaly is computed directly from the individual anomalies in the observed components. The frequency-domain transformation is tested on the synthetic data, and the estimated directions are consistent with the true values. The processing procedure is applied to borehole data collected at the Mengku iron-ore deposit in Northwest China, where magnetic surveys from 55 boreholes have been acquired. The estimated magnetization directions yield good approximations of the results from the ground magnetic data and the physical property measurements of rock samples.

Solid Mineral Potential and Geothermal Energy Reserve of Northern Basement Complex, Nigeria

Abstract in English The spectral analysis of aeromagnetic data over the study area (Northern basement complex, Nigeria) revealed the existence of two major source depths. The second segment of the spectrum layer shows shallow sources ranging in depth from 0.135 km to 0.201 km, with an average depth of 0.140 km. The depth to the basement or deeper source ranges between 1.655 km and 2.021 km, with an average depth of 1.882 km. The Precambrian basement is reflected in the deeper sources, beginning with the first segment of the power spectrum. Structural and topographic relief of the basement surface, lateral differences in basement susceptibilities, and intra-basement features such as faults and fractures contribute to the variation in basement composition. The mean thickness of sediments in the study area is represented by the D2 values obtained from the spectral plots. The depths revealed by this study appears to be reasonable and consistent with previous researchers’ findings. Tectonically active regions have a major impact on heat flows. The average heat flow in thermally normal continental regions is reported to be above 60 mW/m2. Values between 80 and 100 mW/m2 indicate a good geothermal source; values in excess of about 80–100 mW/m2 indicate anomalous geothermal conditions.

Heterogeneous seismic anisotropy beneath Madeira and Canary archipelagos revealed by local and teleseismic shear wave splitting

SUMMARYMid-plate upward mantle flow is a key component of global mantle convection, but its patterns are poorly constrained. Seismic anisotropy is the most direct way to infer mantle flow as well as melt distribution, yet the convection patterns associated with plume-like mantle upwelling are understudied due to limited seismic data coverage. Here, we investigate seismic anisotropy beneath the Madeira and Canary hotspots using a dense set of shear wave splitting observations and combining teleseismic and local events recorded by three-component broad-band and short-period seismic stations. Using a total of 26 stations in the Madeira archipelago and 43 stations around the Canary Islands, we obtain 655 high-quality measurements that reveal heterogeneous flow patterns. Although local event results are sparse around most islands, we can observe a small average of S-wave splitting times of 0.16 ± 0.01 s, which significantly increase with source depth beneath El Hierro (>20 km) and Tenerife (>38 km) up to 0.58 ± 0.01 and 0.47 ± 0.05 s. This suggests an influence of melt pocket orientation in magma reservoirs developed at uppermost-mantle depths. Likewise, anisotropy increases significantly beneath the islands with shield stage volcanism (up to 9.81 ± 1.78 per cent at El Hierro, western Canaries, against values up to 1.76 ± 0.73 per cent at Lanzarote, eastern Canaries). On average, teleseismic SKS-wave splitting delay times are large (2.19 ± 0.05 s), indicating sublithospheric mantle flow as the primary source for anisotropy in the region. In the Canaries, the western islands show significantly smaller average SKS delay times (1.93 ± 0.07 s) than the eastern ones (2.25 ± 0.11 s), which could be explained by destructive interference above the mantle upwelling. Despite complex patterns of fast polarization directions throughout both regions, some azimuthal pattern across close stations can be observed and related to present-day mantle flow and anisotropy frozen in the lithosphere since before 60 Ma. Additionally, we infer that the current presence of a mantle plume beneath the archipelagos leads to the associated complex, small-scale heterogeneous anisotropy observations.

Geophysical data provide three dimensional insights into porphyry copper systems in the Silverton caldera, Colorado, USA

The Silverton caldera in southwest Colorado, USA hosts polymetallic veins and pervasively altered rocks indicative of porphyry copper systems. Nearly a kilometer of erosion has exposed multiple levels of the hydrothermal systems from shallow lithocaps down to quartz-sericite-pyrite veins. New airborne electromagnetic and magnetic survey data are integrated with previous alteration mapping and porphyry models to show the subsurface geophysical response of shallow to deep levels of the porphyry system. Qualitative map views show lateral changes in the magnetization and resistivity of the hydrothermally altered rocks. The volcanic terrain exhibits high magnetization and high amplitude anomalies map near-surface plutonic rocks associated with porphyry systems. Magnetic susceptibility measurements on outcrops of hydrothermally altered rocks indicate magnetite content decreases upward and outward from the source intrusions where magnetic anomaly lows are observed over the lithocaps. The resistivity maps highlight hydrothermal alteration as resistivity lows with exception being rocks having propylitic alteration. Quantitative resistivity models show low resistivity zones with an apparent thickness around 50–150 m beneath quartz-sericite-pyrite veins interpreted to be the result of supergene processes that may continue today, and the calculated magnetic source depths occur near the top of this zone. The resistivity models also show rocks having propylitic, silicic, and quartz-alunite-pyrophyllite assemblages exhibit high resistivity with depth, and argillic alteration assemblages had high resistivity due to high quartz content. This integrated approach presented in a three-dimensional environment provides guidance when exploring for porphyry copper systems in less exposed terrains.

Assessment of ground deformation and seismicity in two areas of intense hydrocarbon production in the Argentinian Patagonia

The exploitation of both conventional and unconventional hydrocarbons may lead to still not well-known environmental consequences such as ground deformation and induced/triggered seismicity. Identifying and characterizing these effects is fundamental for prevention or mitigation purposes, especially when they impact populated areas. Two case studies of such effects on hydrocarbon-producing basins in Argentina, the Neuquén and the Golfo de San Jorge, are presented in this work. The intense hydrocarbon production activities in recent years and their potential link with the occurrence of two earthquakes of magnitude 4.9 and 5 near the operating well fields is assessed. A joint analysis of satellite radar interferometry and records of fluid injection and extraction demonstrate that, between 2017 and 2020, vertical ground displacements occurred in both study areas over active well fields that might indicate a correlation to hydrocarbon production activities. Coseismic deformation models of the two earthquakes constrain source depths to less than 2 km. The absence of seismicity before the beginning of the hydrocarbon activities in both areas, and the occurrence of the two largest and shallow earthquakes in the vicinity of the active well fields just after intensive production periods, points towards the potential association between both phenomena.

Modeling and evaluation of NAPL-impacted soil vapor intrusion facilitated by vadose zone breathing

The water table fluctuations can cause a reciprocating advection of soil gas in the vertical direction, known as vadose zone breathing. In this study, we developed a vadose zone breathing-facilitated vapor intrusion (VI) model and applied it to evaluate the influence of NAPL (non-aqueous phase liquid)-impacted soils. The developed model was first validated with data from previously conducted sand tank experiments and then used to simulate VI scenarios involving different fluctuation amplitudes and periods of the water table, soil types, building structures, and source depths. Our results suggest that within one month of vadose zone breathing (amplitude of 0.5 m and period of 4 d), the indoor pollutant concentration varies by approximately seven orders of magnitude, and the averaged concentration is nearly nine times larger than the steady-state concentration. The results also show that with the increase of mean soil particle size, the indoor pollutant concentration tends to increase and becomes more sensitive to vadose zone breathing; compared to the basement foundation, buildings with a slab-on-grade foundation are less susceptible; when the source depth exceeds 10 m, the source depth plays an insignificant influence. Our findings highlight that vadose zone breathing can promote the VI risks of buildings with a basement foundation at sites with shallow water tables, significant groundwater fluctuations, and coarse-textural soils.

Deep mantle origin of large igneous provinces and komatiites

Large igneous provinces (LIPs) resulted from intraplate magmatic events mobilizing volumes of magma up to several million cubic kilometers. LIPs and lavas with deep mantle sources have compositions ranging from komatiites found in Archean greenstone belts to basalts and picrites in Phanerozoic flood basalt and recent oceanic islands. In this study, we identify the mantle conditions appropriate to each type of lava based on an experimental study of the melting of pyrolite. The depth of the mantle source decreases from 600 to 700 km for the oldest komatiites to 100 to 300 km for picrites and basalts, and the extent of mantle melting ranges from 10 to 50%. We develop a geodynamical model that explains the origin of the hot mantle plumes capable of generating these melting P-T conditions. Within a superadiabatic temperature gradient persisting in the deep mantle, the ascent of hot mantle plumes creates excess temperatures up to 250 to 300 K by adiabatic decompression.

Advanced Potential Field Analysis Applied to Mise‐à‐la‐Masse Surveys for Leakage Detection

AbstractTraditionally interpretation of Mise‐à‐la‐masse (MALM) is limited to the visualization of equipotential contours in order to infer qualitatively the extent of the anomaly. MALM inversion algorithms rely on having a good knowledge of the electrical resistivity distribution in the subsoil. Conversely, potential imaging methods have shown their strength for several applications to quickly estimate the depth of sources even in highly heterogeneous media. In the case of the MALM method, the physics may be described by Poisson's equation. As the conductivity term is modulating the flux of current, MALM is generally referred to as a pseudo‐potential method. In this work, we have tested, for the first time, the application of the potential field theory to MALM in order to identify the current source depth. Synthetic modeling shows that the proposed algorithm is effective and efficient, using surface voltage measurements for different resistivity contrasts, anomaly depths and noise levels. We then applied the method to the real field case of a landfill leakage and showed how very different source depth estimates result from an intact or a damaged landfill liner.

Effects of Earth's gravitation and compressibility on co- and post-seismic deformations

SUMMARY In this paper, we systematically analyse the effect of Earth's gravitation and compressibility on co- and post-seismic deformations. Based on the analytical solution of dislocation Love number (DLN) in a non-gravitational and incompressible homogeneous Earth model, it was seen that reducing the density of the homogeneous Earth model can provide results equivalent to a non-gravitational model while expanding the Lamé constant of the homogeneous model gives results equivalent to those from an incompressible model. Then, we applied a non-gravitational and incompressible treatment in a layered spherical Earth model, and calculated the co- and post-seismic Green's functions for four independent sources. It is found that the effect of Earth's gravitation and compressibility are dependent on source depth and time after the event. The effect of gravitation on the co-seismic Green's function of a shallow source is much smaller than that of compressibility, while the effect of gravitation on the post-seismic Green's function over longer timescales is much greater than that of compressibility. Finally, taking the 2011 Tohoku-Oki Mw9.0 and 2013 Okhotsk Mw 8.3 earthquakes as examples, the simulation results show that post-seismic deformation studies should consider the Earth's compressibility, which is more obvious in the early post-seismic period. Likewise, when the viscosity of the study area is low, the effect of gravitation cannot be ignored. For the slip inversion of Tohoku-Oki Mw 9.0 earthquake, the incompressible Earth model leads to a smaller slip magnitude than the compressible Earth model.

Airborne data processing and 3D modeling of magnetic structures between Nador region and the Chaffarine Islands: Geological significances

The Cap of Trois Fourches-Nador-Chaffarine Islands zone (CTF-N-CI) located in the Moroccan Oriental Rif is recognized by Neogene-Quaternary volcanic outcrops. Those later formed volcanic cones and are considered the main source of the huge Nador magnetic anomaly. This study aims to enhance the magnetic context using geophysical techniques and identify the geometry of magnetic sources by performing 3D modeling using Talwani method. We have achieved the following findings: (1) The elaborated residual magnetic map (Rs) that shows three aligned magnetic anomalies. They are oriented W-E and register a maximum intensity of 380 nT. Applying reduction to the pole to (RTP) the Rs map makes the anomalies on the top of their body sources. Thus, we have found that those anomalies have resulted essentially from the Gourougou volcano and Chaffarine Islands volcanoes outcropping in the study area. Except for the middle Pm anomaly, which is related to volcanic rocks of the Kariate Arekmane drillhole. (2) Performing horizontal derivative (HD) technique leads to establishing the edges of the source bodies. Hence, the inversed map of the magnetization distribution proves that the three bodies are homogeneous and probably splintered from a unique mass. (3) The downward up continuation (DC) to different heights (from 500 to 2000 m) indicates the source's depth. Thus, we have demonstrated that the common source could reach 0.6 km of altitude. (4) Finally, we have constructed a 3D Talwani model of the volcanic edifices. The released 3D model demonstrates that the anomalies are related to a very extensive mass of 80 km in length, 20 km in width, 1.3 km in thickness. This model supports that the Neogene-Quaternary volcanism of the study area was fissural and related to the W-E fracture zone of up to 80 km in length. This volcanism was also affected by the sinistral Nekor fault, which probably made the Trois Fourches Cap body isolated from the other volcanic edifices.Previously mentioned elements probably prove the simultaneous establishment of this large magnetic structure. Moreover, they emphasize the crucial role of the magmatic phenomena which occurred in Northeastern Morocco at the beginning of the Messinian.

Fluid flowrates and compositions and water–rock interaction in the Hikurangi margin forearc, New Zealand

The decrease in interseismic coupling and slip rate deficit from south to north, from locked to slipping, of the Hikurangi margin is attributed from interpretation of geophysical data to more fluid-saturated sediments in the north. To assess this interpretation, fluid flowrates were measured in springs of the subaerial forearc, structures correlated with flowrates and the fluids geochemically and isotopically evaluated to understand their origins and source depths. Extrapolations from on-ground measurements show that ∼95% of aqueous fluids and ∼98% of gas are expelled to the surface in the slipping zone, within the northern subaerial forearc. The relative fluid residence time, from the equilibration of CH4 and CO2 and measured ascent rates, is 3x shorter in the slipping zone suggesting less constricted migration pathways than in the locked zone. Fluids are expelled not only from the sediments that produce Na–Cl fluids and thermogenic gas, but also from the oceanic igneous mafic slab that releases high-Cl fluids and abiotic CH4 from serpentinization, giving rise to Na–Ca–Cl fluids. Chloride-rich surface emissions consist of ∼70% by volume Na–Cl and ∼30% Na–Ca–Cl fluids. Higher rates of fluid expulsion in the north coincide with higher permeability and greater production of aqueous fluids from clay dehydration, serpentinization and a higher degree of diagenesis in sedimentary rocks. Mantle gas, subduction interface fluids and porewaters have the most direct access to the surface in the slipping Center, where the degree of water–rock interaction and homogeneity of water compositions suggest more recent fluid generation than other sectors in the subaerial forearc. Porewaters and thermogenic gas equilibrate at 110° + 20 °C with a slight increase in temperature from south to north. Overall, ∼120 Tg/a water is subducted in the Hikurangi margin and <0.5 Tg/a CH4 is being emitted to the atmosphere. This study shows the added impact of waters of serpentinization on the overall fluid compositions, aqueous fluid subduction budget, extent of fluid distribution in the oceanic slab and, indirectly, the temperature estimate of 185° + 25 °C for fluids ascending from the subduction interface in the Hikurangi margin.

Zircon U-Pb ages and Hf isotope characteristics of granitic rocks from the Shapinggou molybdenum deposit, Dabie Shan, China

The giant Shapinggou porphyry molybdenum deposit in Jinzhai county, Anhui province is located in eastern Dabie Shan, but its genesis, genetic type and tectonic setting are still debated. New zircon U − Pb age data presented in this contribution show that the granitic rocks at Shapinggou area have been episodically formed at ∼ 135 Ma, ∼127 Ma, ∼117 Ma and ∼ 113 Ma in Early Cretaceous, and can be classified into the early-stage (135–127 Ma) granodiorite and monzogranite plutons, and the late-stage (117–113 Ma) quartz-orthophyre stock and cryptoexplosive breccia pipes. Mo mineralization is closely associated with the late-stage shallow-seated granitic stock or pipes. Granitic rocks of both stages have high contents of SiO2, K2O + Na2O and Al2O3, low contents of TiO2, CaO and MgO, showing a metaluminous to peraluminous, high-K calc-alkaline to shoshonite affinity. They are generally featured by enrichments of LILE (Rb, U, Th) and LREE, and depletions of HFSE (Nb, Ta, Ti) and HREE, with high (La/Yb)N ratios. The early-stage granitic plutons show trace element geochemical signatures of partial melting of over-thickened continental crust (>50 km), whereas element geochemistry of the late-stage small-sized granitic intrusions suggests a shallow source depth (ca. 35 km) under post-collisional extension setting. The Hf isotope data indicate that the Shapinggou granitic rocks are originated from a crustal source predominated by the Kuanping Group mixed with other components such as the mantle. The southward subduction and partial melting of the southern North China Block (including Kuanping Group) beneath the Dabie Shan lead to the generation of the Shapinggou porphyry Mo system in the post-collisional extension setting. Hence the Shapinggou Mo system distinguishes for its crustal source and post-collisional tectonic setting from the porphyry Mo deposits of Climax- or Endako-types, and should belong to collision-type or Dabie-type porphyry Mo deposit.

Dimension-scaling relationships of Pachmarhi dyke swarm and their implications on Deccan magma emplacement

The Continental flood basalts of the Deccan Volcanic Province in India contain three major dyke swarms, namely the Narmada-Satpura-Tapi (N-S-T), the Western Coastal and the Nasik-Pune dyke swarm. The current study area is around Pachmarhi town, Madhya Pradesh, which marks the eastern end of the N-S-T dyke swarm. There are ∼244 mappable doleritic and basaltic dykes around Pachmarhi. The shortest and longest dykes in the area are 140 m and 22 km respectively. The mean length of all the studied dykes is ∼5.15 km. The dykes are in general shorter than those exposed in the western end of the NST swarm in the Dhule-Nandurbar area in Maharashtra. The thickness of the Pachmarhi dykes varies between 3.5 m and 35 m. The dykes exhibit a preferred orientation of N82°E that is parallel to the general trend of the Narmada-Son Lineament (NSL) and perpendicular to the N-S trending paleo extension. In this paper, we present detailed documentation of the scaling relationships of Pachmarhi dyke dimensions, calculate magmatic overpressures during emplacement of dykes and therefore infer magma source depth. Further, we compare the inferred magma source depths among Pachmarhi and Dhule-Nandurbar dykes of Deccan with Early-Neoarchean to Late-Paleoproterozoic Newer dolerite dykes of the Singhbhum craton. Finally, we discuss some possible scenarios of magma emplacement mechanisms for Pachmarhi dykes and relate them to the larger geodynamics of Deccan volcanism.

Transient heat transfer characteristics of submerged heat source under seismic excitation

In this experimental based study, experiments are carried out to investigate the transient heat transfer behavior from the horizontal heat source submerged in the partial filled water pool including the effect of submergence depth of heat source. Further, experiments are conducted and analyses have been made to investigate the heat transfer behavior under the seismic environment along with the consideration of seismic excitation parameters: natural frequency of liquid and amplitude of displacement.The transient heat transfer coefficient for natural convection and boiling region are evaluated with and without excitation cases. Experimental heat transfer coefficients were compared with the available correlations and found inadequate. Therefore, new correlations have been proposed for transient heat transfer coefficient for each region. For the seismic condition, separate correlation has been proposed to predict the vibrational heat transfer coefficient in case when strong sloshing occurs with wave breaking at the free surface. Proposed correlations are predicted quite satisfactory results.

Prediction of the Stress Wave Amplification Factor of a Spherical Blast Source Using Numerical Simulations

A typical blast wave attenuation curve presents a relationship between Peak Particle Velocity (PPV) at the surface of a geologic profile and distance. As the stress wave is amplified at the free-field boundary, the attenuation curve at the surface is always larger than the within media profile curve. Measurements are made at the rock's surface and test blasts are always conducted to ensure the safety of underground existing structures. In order to design underground blasting, the recorded PPVs are then reduced by a factor of 2. In this paper, particle velocity amplification was studied by using numerical simulation, and the difference between PPV at the surface and within media profiles is quantified. The amplification factor depends upon source depth, incidence angle, and Poisson’s ratio of the media. It is calculated as the ratio of the magnitude of PPV at the surface of the media to the within media profile. According to the parametric study, the amplification factor for a uniform medium increases with increasing source depth, while the amplification factor decreases with increasing Poisson’s ratio. Considering a three-layer model with a source depth of 30m, the amplification factor is high for low incident angles and low for higher incident angles. The range varies between 1.5 to 2.1.

Learning both a value and uncertainty label for seabed parameters from ship noise data

Because the seabed impacts sound propagation in the ocean, machine learning is being used for both seabed classification and to obtain estimates of individual seabed properties. This paper proposes a method to simultaneously estimate these properties and an associated uncertainty label. A residual neural network is trained and validated using synthetic ship noise spectrograms generated with a range-independent normal mode sound propagation model and a ship noise source spectrum. The data set includes 140 seabeds: In each, the top sediment layer has a random thickness and properties randomly chosen from five sets of bounds, which roughly correspond to clay, mud, sand, silt, and gravel. For each of the 140 sediments, a random selection of 405 combinations of ship speed, closest-point-of-approach range, and source depth are used resulting in 22k data samples. Each data sample is labeled with the true values of the sediment parameters as well as a label for the uncertainty level of each. The uncertainty levels are obtained from the Fisher information and qualify of the information content in the ship spectrogram about each parameter. Examples of how the residual neural network learns to perform regression for the parameter value and the uncertainty level will be presented.

Extensions to Lisa Zurk’s depth-based signal separation method

Lisa Zurk and her colleagues introduced a depth-based signal separation method that uses interference between the direct and surface-reflected CW signals which are received on a bottom-moored vertical line array in deep water [Asilomar Conference on Signals, Systems and Computers, IEEE, 2130–2312, 2013, J. Acoust. Soc. Am. 133(4), EL320-EL325, 2013]. In this work, her method was extended to broadband signals for instantaneous depth-based signal separation and range estimation. The broadband method was successfully tested by using a single phone or a pair of phones, deployed in the Northern Gulf of Mexico. A broadband (250–800 Hz) source was towed at 20 m, 50 m, and 100 m depths and received at four bottom-moored Environmental Acoustic Recording Systems (EARS). Source depth estimation and localization by trilateration was successfully demonstrated. Realizing the lack of comprehensive CW and broadband data sets, two experiments were planned for late 2022 and mid 2023 in collaboration with Lisa Zurk. These experiments will still be conducted to demonstrate her depth-based signal separation method experimentally. [Work supported by the ONR.]