SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

The El'gygytgyn impact structure in northeast Russia was drilled in 2008/2009. The 3.5 Ma old structure has a rim-to-rim diameter of 18 km and is the only known impact structure that has been formed on a siliceous volcanic target. The petrophysical, rock- and paleomagnetic properties, including attempted reorientation of samples, along the El'gygytgyn drill core were analyzed. Physical properties, such as bulk density, porosity, seismic velocity, and electrical conductivity, clearly showed the propagation of shock and the associated fracturing. The grain density, however, was probably influenced by the postimpact hydrothermal activity and/or the distribution of impact melt. The highest values of electrical conductivity coincided with higher concentrations of particular metals as indicated by Raschke et al. (2012a). The rock- and paleomagnetic investigations showed iron-titanium oxides with varying oxidation/reduction states as the main magnetic fraction in the core samples and indicated them as carriers for remanent magnetization. With few exceptions, most samples showed normal polarity of characteristic remanent magnetization and confirmed that the impact occurred after the Gauss/Gilbert (approximately 3.596 Ma) reversal. Shallower inclinations than that expected for a 3.5 Ma dipole field were probably due to impact-related block movements and/or compaction.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

Hypervelocity impacts are recognized as fundamental planetary surface forming and deforming processes. They have played a significant role in biological extinction events and, as explained by French (1998), have produced structures with high economic value. Impact structures are formed via shock metamorphic processes during which high-pressure shock waves are generated from the conversion of the kinetic energy of the impactor at the point of impact. These shock waves, with velocities which may exceed 10 km s−1 (French 1998), deform, crush, and excavate material from the target. The extent of this shock deformation is dependent on how much the force of the shock waves exceeds the compression strength of the target. As the shock waves travel away from the point of impact, the shock pressures decrease rapidly, and the waves become ordinary seismic waves near the crater rim. Reflection of the initial compression waves generates tension waves, which may further deform the target rocks if they exceed the rocks' tensile strength.

Impact structures are traced by their distinct morphology and their geophysical anomalies. With the exception of Venus, the number of known impact craters on the Earth (see http://www.passc.net/), however, is generally less than on other terrestrial planets. Furthermore, of all the 182 impact structures on Earth, only three have been found on volcanic target rocks, which is in contrast with the other terrestrial bodies, where impact structures have more or less formed solely on igneous targets. Both these contradictions can be explained by the presence of (1) the Earth's atmosphere; (2) water, which covers more than two thirds of the Earth's surface; and (3) the active geological (including tectonic) and surface processes ongoing on the planet's surface. All these things can tamper and hide the traces of impacts. To investigate impact structures as well as confirm their impact origin (see Koeberl and Milkereit 2007), laboratory studies are required. In addition, to study the deeper structure and the extent of shock penetration, subsurface samples are needed. As impact models strongly depend on the contrasts in physical properties between the impact structure, different features within the structure itself and its unshocked surroundings, knowledge of the depth relation of the physical properties, and the understanding of elastic limits and rock strength are crucial. Therefore, drilling of impact structures is important to obtain key information.

This study aims to see how far the shock effects of the El'gygytgyn structure propagated using the physical properties and their depth variation, based on samples retrieved from the El'gygytgyn drill core. Also, a general comparison of the values of siliceous volcanic rocks is given. The knowledge of physical properties will also contribute to better gravity, electromagnetic, and seismic modeling. An attempt was made to reorientate the drill core and date the impact event by paleomagnetic methods.

The El'gygytgyn Impact Structure and Drilling

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

El'gygytgyn (67°30′ N, 172°05′ E) is a complex impact structure located in the late Cretaceous Okhotsk-Chukotka Volcanic Belt, in Chukotka Peninsula, north-eastern Russia (Fig. 1a). El'gygytgyn is a well-preserved structure and is unique with its siliceous volcanic target sequence. It is, together with the 40 Ma Logancha structure (Masaitis 1999) in Russia and the 0.05 Ma Lonar structure (Sengupta et al. 1997) in India, one of only three impact structures known to have formed on volcanic target rocks on Earth. The impact origin of the El'gygytgyn crater was confirmed by the recognition of shock metamorphic effects in the volcanic rocks by Gurov et al. (1978, 1979). Based on K-Ar dating of impact melt glasses (Gurov and Gurova 1979), the age of the El'gygytgyn structure is 3.50 ± 0.50 Ma. Similar ages have been put forward by Komarov et al. (1983) and Layer (2000). The former reported a fission track age of 3.45 ± 0.15 Ma and the latter reported an age of 3.54 ± 0.04 Ma using the 40Ar/39Ar method on impact glasses.

image

Figure 1. a) Location of the El'gygytgyn impact structure. Inset at upper right is a NASA satellite image showing the location of drill core relative to central uplift and crater rim. b) Location of faults around the structure (modified from Gurov and Koeberl 2004). c) Fracture density around the crater relative to that of the unshocked surroundings is plotted against distance measured in crater radii (modified from Gurov et al. 2007).

Download figure to PowerPoint

The El'gygytgyn impact structure has a flat-floored circular basin with a rim-to-rim diameter of 18 km (Gurov and Koeberl 2004). In the crater floor, which has a diameter of about 15 km, is the up to 170 m deep, 12 km diameter Lake El'gygytgyn. The lake, which is situated southeast of the center of the crater (Fig. 1b), is surrounded by wide lacustrine terraces on the western side (Gurov et al. 2007). Deposits within these lacustrine terraces on the crater floor are the only source of impact rocks at the recent erosional level (Gurov and Koeberl 2004). Gurov et al. (2007) have shown that the impactites include material eroded from the rims and from impact ejecta. The rim of the crater (Fig. 1a) is on average 140 m higher than the level of the lake and 180 m higher than the surrounding area. The crater rim has steep inner walls and gentle outer slopes. It has been estimated that the original height of the rim crest was 230 m (Gurov et al. 2007). Gurov and Yamnichenko (1995) have shown the presence of an outer rim with an average height of 14 m. This outer rim is separated from the main rim by a small depression in which some streams exist and the flow is concentric to the crater rim. Seismic studies (Gebhardt et al. 2006) show the evidence for a central uplift (Fig. 1a) and brecciation of the target rock to a depth of 500–600 m below the lake floor. Postimpact sediments, with a total thickness of 320–340 m, fill the crater depression. The El'gygytgyn impact structure is surrounded by a complex system of faults (Gurov and Koeberl 2004). Short radial faults predominate (Fig. 1b), with the highest fault density within the inner walls of the structure as shown in Fig. 1c. Gurov and Koeberl (2004) reported that megablock breccia zones occur at the northern and inner western slopes. Ignimbrites and tuffs of rhyolites and dacites are the main rock types of the crater basement, with the rare presence of andesite and andesitic tuffs (Gurov and Koeberl 2004).

The El'gygytgyn structure was drilled by the International Continental Scientific Drilling Program (ICDP) between November 2008 and February 2009. The drill core 5011-1 C (Melles et al. 2011) was taken from the southeast of the structure, just beside the location of the central uplift (Fig. 1a) demonstrated by Gebhardt et al. (2006) and Niessen et al. (2007). The total length of the core was 517 m. The upper part of the drill core (0–315 m; 0 at the lake floor) comprised postimpact lacustrine sediments and was used in paleoclimate studies. The lower 202 m of the core comprised the impactites used in impact-related studies. Due to the complex logistics and extreme conditions, it was only possible to retrieve one impact core and as a result, it was not possible in this study to make direct comparisons between the physical properties of samples from different locations around the impact structure.

The lithological profile of the impactite section of the drill core consists of four main sections (Raschke et al. 2012b). The first, from 315–328 m, is the reworked suevite (RS), which is composed of a mixture of lacustrine sediments and strongly shocked rock fragments of various target lithologies. Below this (328–391 m) is a sequence of polymict impact breccia (PIB), consisting of crystalline clasts and melt particles in a clast-rich, fine-grained groundmass. Within this section are three volcanic block inclusions. The upper bedrock (UB) section is between 391 and 421 m and is made of two different volcanic rocks. The upper subunit (approximately 1 m thick) has basaltic composition, while the second has rhyodacitic composition and is strongly brecciated. The lowermost section (421–517 m) is the lower bedrock (LB), consisting of homogenous welded ignimbrite. Near the top (approximately 422 m) of the LB is a less than 0.5 m thick dacitic volcanic block. Also, at approximately 471 m depth is a less than 1 m thick block of polymict impact breccia.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

The sampling for impact-related studies took place in May 2010 at the Natural History Museum in Berlin. Based on variation in lithology, mineralogy, texture, and structure, 92 samples were selected for this study with a sample interval of about 2 m.

Petrophysical, rock- and paleomagnetic properties were measured at the Solid Earth Geophysics Laboratory of the University of Helsinki (Pesonen 2006). The samples were drilled to obtain smaller cylinders (approximately 2.5 cm in diameter and 2.3 cm high) or cut into cubes (with sides approximately 2.5 cm). Special care was taken with the fragile samples to avoid disintegration due to water immersion. Whenever possible, alternative methods of measuring, described below, were used for those samples. Because limitations were caused by the fragility of some samples, and based on the results from the initial measurements of 10 samples, a total of 50 samples were used with a roughly equal sampling interval of approximately 4 m.

Wet and dry densities were determined by weighing water-saturated and oven-dried samples in air and in water. Porosities were calculated from these measurements. For very fragile samples, which could not be exposed to water, dry bulk density measurements were made using tiny glass beads (20–50 μm in diameter) instead of water (Consolmagno and Britt 1998). As it was possible to make more dry bulk density measurements than wet bulk density measurements, the “bulk density” mentioned henceforth is the dry bulk density.

Seismic velocities of wet samples were measured with ultrasonic devices (e.g., Lassila et al. 2010). First, only uncompressed p-wave velocity (vp) measurements were made followed by both vp and shear-wave velocity (vs) measurements under in situ like pressure (up to 15 MPa) conditions. The estimates of the in situ pressures were calculated from the sample depths based on the measured densities. Some samples were crushed under the laboratory loading, thus presenting limitations in the measurement of seismic velocities. The reflection coefficients between lithological units (e.g., PIB/UB) were calculated from uncompressed vp values and bulk densities using standard techniques.

The measurement of the electrical conductivity of wet samples was made with a four-point conductivity instrument, part of the RISTO-5 petrophysical measurement setup (Pesonen 2006). Conductivity values at 1 and 1000 Hz were obtained, with the results of the latter presented below. As many samples could not be soaked in water, it was not possible to get electrical conductivity data for all samples.

Magnetic susceptibility (χ) measurements were made with the use of a RISTO-5 kappabridge and/or an AGICO KLY-3S kappabridge. The KLY-3S kappabridge was also used to measure the anisotropy and the temperature dependence of the susceptibility in argon gas. Hysteresis properties were measured using a Princeton vibrating sample magnetometer (VSM).

Paleomagnetic measurements, including the natural remanent magnetism (NRM) and the demagnetization of the NRM to yield its components, were performed with the use of a 2G–Superconducting Rock Magnetometer (SQUID). The samples were demagnetized in an alternating field (AF) up to 160 mT, in increments of 2.5 up to 5 mT, then in increments of 5 up to 20 mT, and finally in increments of 10 up to 160 mT. Also, thermal demagnetization was performed using Magnetic Measurements Thermal Demagnetizer furnace from ambient temperature to 600 °C.

An estimate of the relative magnetization of magnetically soft and hard minerals was made using the SQUID and a Magnetic Measurements Pulse Magnetizer. Samples were demagnetized with the SQUID and then progressively magnetized along the +z direction with the pulse magnetizer up to 5 T, to saturate the isothermal remanent magnetization (SIRMz). The samples were then exposed to a field of 200 or 300 mT in the opposite (−z) direction, and their new magnetizations were measured (see 'Results' section).

The El'gygytgyn drill core was unoriented, having only the vertical up-down markings to yield inclination data. Due to breaks in the core, such as those from core loss, the common fiducial mark for relative declination could not be applied. To get the declinations, the core was reoriented in the laboratory using the method of Järvelä et al. (1995) based on the data from AF demagnetization of NRM. The low coercivity viscous remanent magnetization (VRM) and the intermediate to high coercivity characteristic remanent magnetization (ChRM) components of the NRM were identified, and the declination of ChRM of the samples was then rotated so that the declination of the VRM aligned with the present (averaged over the last 100 y) geomagnetic field. The re-orientation method could not be applied to samples without VRM components. Thermal NRM demagnetization data were unstable, revealing no VRM. The ChRM for these samples was found by finding the Fisher mean of endpoints. “Inclination-only” paleomagnetic analysis of all demagnetized samples was performed and comparisons with the reoriented data were made.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

Petrophysical Properties

Results of various petrophysical properties are given in Fig. 2 and Table 1. The bulk densities in upper part of the drill core, from the reworked suevites (RS) through the polymict breccias (PIB) until the upper bedrock section (UB), show a clear increase with depth from less than 1000 to approximately 2200 kg m−3, while porosities decrease from approximately 50 to 10%. The bulk density and porosity in the lowermost part of the core, the lower bedrock (LB) section, however, indicate relatively constant values, at approximately 2500 kg m−3 and 7% respectively, with only minor changes with depth. This marked increase in bulk density coupled with the decrease in porosity is indicative of progressive shock decay with depth, and thus the formation mechanism of the structure. The grain density decreased through the length of the core as the result of the decrease in melt content (Raschke et al. 2012b) with depth and/or postimpact hydrothermal activity. Chemical analysis by Raschke et al. (2012a) showed that the percentage proportions of the oxides of lighter elements including silica, sodium, potassium, and aluminum, increased with depth, while those of heavier elements such as titanium and iron, decreased with depth. This is in agreement with the observed relationship between depth and grain density.

Table 1. Petrophysical measurements from the El'gygytgyn drill core, and measurements of unshocked acid volcanics from Iceland and Kamchatka
 nρb kg m−3ρg kg m−3Φ %χ 10−6 SINRM mAm−1Q 10−3
  1. = number of samples, ρb = bulk density (dry density for El'gygytgyn samples), ρg = grain density, Φ = porosity, χ = magnetic susceptibility, NRM = natural remanent magnetization, Q = Koenigsberger ratio.

  2. Data for Iceland volcanites after Airo (1990), and for Kamchatka volcanites after Frolova (2008).

  3. a

    Denotes the number of samples for which at least one of the physical properties listed was measured. (Some of the physical properties were not measured for some samples).

  4. b

    One sample only.

  5. c

    Calculated in this work.

El'gygytgyn lithologies
Reworked suevite5a1409284642.62594329.13899.4
Polymict impact breccia6a2088289715.8102328.5551.1
Basaltic volcanics12235  92600.20.7
Dacitic volcanics7a21952645b3.4b111320.20.7
Welded ignimbrite25a245426467.27851415.91571.5
Iceland volcanites
Dacite22385  39364002455.6c
Andesite362893  4465461003300.8c
Rhyolite672366  668011003979.0c
Ignimbrite42227  1853110001309.9c
Kamchatka volcanites
Ignimbrites82100  13000  
image

Figure 2. Variation in physical properties along the drill core. (Lithological profile modified from Raschke et al. 2012b.)

Download figure to PowerPoint

Seismic velocities were mainly measured for depths below 420 m (Fig. 2). The uncompressed vp values increased with depth due to the decrease in porosity down the drill core, while the compressed vp and vs values decreased with depth and reflect the lowering grain density. In general, the uncompressed vp values are in agreement with those of Niessen et al. (2007), while the compressed vp values are higher. It is likely that the higher values are due to the more effective closing of cracks and fractures under laboratory compression, than in in situ conditions. While compressed seismic velocity data do not exist for the upper sections, uncompressed vp values were used to get the reflection coefficients for each section boundary (Fig. 2), with the UB/LB boundary clearly having the highest reflection coefficient (= 0.74, for UB/LB).

As shown in Fig. 2, the electrical conductivity also decreased with depth. This trend may also reflect the reduction of cracks and fractures further down the drill core. Apart from the two measurements from the very porous RS section, where ionic conduction would have been dominant, the highest conductivity values were at approximately 420 m depth. This coincided with an enrichment of various metal elements (Raschke et al. 2012a), including iron, titanium, calcium, and aluminum. The increase in the conductivity at approximately 470 m depth also coincided with higher elemental compositions of the same metals. The electrical conductivity hints at postimpact hydrothermal activity, which would have influenced the zones of metal enrichment.

Rock-Magnetism

The susceptibility values were lower in the upper part of the drill core (Fig. 2) and ranged between 100 and 15000 μSI, with most samples having values of over 5000 μSI. The lowest average values were in the RS (approximately 2600 μSI) and the PIB (approximately 1000 μSI) sections as seen in Table 1. The highest values (approximately 11000 μSI) were in the largely dacitic UB section and decreased in the ignimbrites (approximately 8000 μSI) of the LB section.

The variation in susceptibility with temperature (Fig. 3) shows that Curie temperatures of the samples were just above 500 °C, with well-defined Hopkinson's peaks for most samples. Samples from the UB (Fig. 3c) had less well-defined Hopkinson's peaks than other samples. This suggests that the main magnetic minerals present in the samples were low titanium titanomagnetites. Susceptibility temperature plots also show the presence of magnetic minerals with Curie temperatures of over 600 °C. These minerals are interpreted as titanohematites.

image

Figure 3. Temperature dependence of susceptibility (χ) normalized to its peak value (χp). Black (gray) curves denote heating (cooling) curves. The original susceptibility in μSI is given in parentheses next to the depth value. The dashed and dotted lines show Curie temperatures, interpreted from the first inflection point after Hopkinson's peaks, indicating titanomagnetites (between 500–530 °C) and titanohematites (between 600 and 630°C), respectively.

Download figure to PowerPoint

The intensity of NRM varied along the drill core, but two distinct trends were observed. One group of samples had values of between 10−3 and 10−1 mAm−1 and another had values between 10 and 103 mAm−1 (Fig. 2). Samples with lower NRM were present in all sections above 460 m. Samples with higher values were mainly from the LB section, while being largely or completely absent in PIB and UB sections, respectively. This trend is visible also in the mean NRM value for each lithology; the ignimbrites had the highest mean value (415.9 mAm−1), followed by the RS (329.1 mAm−1), while the PIB, basaltic, and dacitic lithologies had mean NRM values of 28.5, 0.2, and 0.2 mAm−1, respectively (Table 1). Q values followed the trends of NRM, with two noticeable groups (Fig. 2): One group with Q = 10−4 to 10−2 and the other with Q = 1–10).

Thermal demagnetization showed different blocking temperatures in different samples (Fig. 4). For most samples, 50% of the magnetism was removed between 300 and 400°C. The blocking temperatures of the magnetic carriers were mostly low to intermediate, indicating titanomagnetites, but minerals with high blocking temperatures, such as titanohematites, were also present. AF demagnetization (Fig. 4) showed that generally, 50% of the samples' magnetism was removed by fields of less than 50 mT, while some magnetism remained even after fields of 160 mT were applied. The lower coercive magnetic carriers are interpreted as titanomagnetites, while the highly coercive ones interpreted as titanohematites.

image

Figure 4. Demagnetization of the NRM. Stereographic (Lambert equal area) projections are left, intensity decay curves are in the middle, and orthogonal (Zijderveld) vector projections are right. For the stereo plots, closed (open) squares represent downward (upward) vectors. For the orthogonal plots, closed (open) diamonds are points on the horizontal (vertical) plane. PEF is the present Earth's field.

Download figure to PowerPoint

Hysteresis and isothermal remanent magnetization (IRM) measurements show that squareness ratios decreased with depth along the core (Fig. 5). Coercivities of magnetic saturation ranged between 5 and 25 mT, with the magnetic grain size for all samples plotting on the boundary of single domain/pseudo–single domain region of the Day plot for magnetite. The highest squareness ratios were in the PIB section where the presence of titanohematites is interpreted. That the highest values were approximately 0.2 can be explained by the relative magnetizations of the various magnetic minerals. The IRM data (Fig. 5) show that the magnetization due to titanohematites is far more influential in the PIB compared with other sections, but titanomagnetites still form the largest fraction of magnetic minerals present.

image

Figure 5. Variation in the squareness ratio (solid black line) and the coercivity of magnetic saturation (dashed gray line) with the depth (left), saturation IRM curves along +z-axis (center), and estimated magnetization carried by titanohematites (right).

Download figure to PowerPoint

The rock-magnetic datasets suggest that lower susceptibility values in the upper sections of the drill core were not due to differences in magnetic grain size or a smaller amount of magnetic minerals, as the percentage proportions of iron and titanium oxides were highest in the upper sections (Raschke et al. 2012a). The results indicate that there was more significant postimpact hydrothermal activity in the PIB and UB sections compared with the LB section. Thus, it is more likely that the stronger hydrothermal activity in upper sections (Raschke et al. 2012b) oxidized the titanomagnetites more than in the LB section. The low susceptibility and NRM values in most of the PIB and UB sections were the result of titanohematization or titanomaghemitization. This is supported by the greater variation in the anisotropy of susceptibility in the upper sections (Fig. 2), as these sections would have undergone more alteration.

Paleomagnetism

AF demagnetization of the samples revealed three main magnetic components. These were initially interpreted as a normal ChRM component, which was dominant, a reverse ChRM component, which was sporadic, and a VRM component. Most samples had both a VRM component and either a normal or a reverse ChRM component. Some samples had superimposed normal and reverse ChRM components, but no VRM (Tables 2 and 3). The sample 373.0 (PIB), however, revealed no discernible components. The normal ChRM components had higher intensities than the reverse ChRM components. All VRM components had normal polarity, hinting to the orientation of the present field, and were used to reorientate the drill core as explained before. The thermally demagnetized samples gave unstable vector components and the ChRM of these samples was found by calculating the Fisher mean of the vector endpoints (Table 3; Fig. 4).

Table 2. Paleomagnetic reorientation of core samples from the El'gygytgyn impact structure (67°30′ N 172°5′ E)
SampleLithologyChRMinitialVRMinitialChRMre-orientated
DIDIDI
  1. Geomagnetic field for 67°30′ N 172°5′ E averaged over the last 100 years: D = 3.8°, I = 74.6°

  2. ChRM = characteristic remanent magnetization, VRM = viscous remanent magnetization, D = declination, I = inclination, TS = reworked suevite, BV = basaltic volcanics, DV = dacitic volcanics, WI = welded ignimbrite, N = number of specimens, k, α95 = precision parameters, λ = paleolatitude.

  3. The declination of the re-oriented ChRM has been found by subtracting the declination of the VRM from that of the initial ChRM, and then adding the declination of the averaged (over the last 100 y) geomagnetic field of El'gygytgyn.

  4. Mean reorientations are only calculated for sets of three or more samples.

Normal polarity
320.9RS50.276.07.669.946.476.0
326.4bRS0.585.61.273.83.185.6
390.8BV256.260.2191.247.168.860.2
407.9DV211.560.0223.275.1352.160.0
421.7WI336.042.412.663.4327.242.4
423.0WI137.034.9176.068.0324.834.9
425.3WI219.540.9242.677.9340.740.9
429.5WI97.853.6340.776.2120.953.6
437.7WI274.441.919.778.0258.541.9
445.5WI8.456.5340.177.232.156.5
470.2WI341.248.941.469.7303.648.9
481.5WI76.768.534.579.246.068.5
489.9WI84.555.121.285.067.155.1
516.6WI42.231.433.579.512.531.4
Mean ChRMreorientated: D = 16°, I = 67°, N = 14, = 6, α95 = 18°.
λ: 50°
For WI: Mean ChRMre-orientated: D = 19°, I = 62°, N = 10, = 5, α95 = 24°.
λ: 43°
Reverse polarity
351.4PIB127.1−40.8156.427.0334.5−40.8
414.1DV200.7−43.8177.115.027.4−43.8
442.6WI94.8−46.2116.6−81.4342.0−46.2
Mean ChRMreorientated: D = 354°, I = −46°, N = 3, = 2, α95 = 73°
Table 3. Inclination only analysis of paleomagnetic data from the El'gygytgyn impact structure (67°30′ N 172°5′ E)
AF demagnetization measurementsThermal demagnetization measurements
SampleLithologyChRMinitial ISampleLithologyChRMinitial I
  1. a

    Denotes samples with both normal and reverse ChRM components.

  2. ChRM = characteristic remanent magnetization, D = declination, I = inclination, RS = reworked suevite, PIB = polymict impact breccia, BV = basaltic volcanics, DV = dacitic volcanics, WI = welded ignimbrite, N = number of specimens, k, α95 = precision parameters, λ = paleolatitude.

  3. Mean reorientations are only calculated for sets of five or more samples.

Normal polarity
320.9RS76.0325.0RS85.2
326.4aaRS75.0328.8PIB82.4
326.4bRS85.6334.7PIB31.3
339.9aPIB72.3410.8DV71.1
390.8BV60.2411.7DV55.0
407.9DV60.0427.8WI73.4
421.7WI42.4459.4WI54.1
423.0WI34.9   
425.3WI40.9   
429.5WI53.6   
437.7WI41.9   
445.5WI56.5   
460.8aWI44.2   
470.2WI48.9   
481.5WI68.5   
489.9WI55.1   
492.5aWI51.1   
505.8aWI54.1   
516.6WI31.4   
Mean ChRM: I = 55°, N = 19, = 30, α95 = 6°.Mean ChRM: I = 64°, N = 7, = 25, α95 = 14°.
λ: 36° λ: 46°
For WI: Mean ChRM: I = 48°, N = 12, = 26, α95 = 8°.   
λ: 29°   
Reverse polarity
326.4aaRS−54.3454.9WI−55.4
339.9aPIB−39.8   
351.4PIB−40.8   
381.6PIB−71.1   
414.1DV−43.8   
442.6WI−46.2   
460.8aWI−79.0   
492.5aWI−59.2   
505.8aWI−57.0   
Mean ChRM: I = −54°, N = 9, = 25, α95 = 9°.   
λ: −35°   

The reorientation was unsuccessful (Table 2; Fig. 4), with the reoriented inclinations and especially the declinations remaining scattered. Therefore, “inclination-only” analysis of both AF and thermal demagnetized data was used, but this too gave much lower paleolatitudes than would be predicted for the 3.5 Ma El'gygytgyn impact event. Only the normal polarity ChRM from the RS and upper part of the PIB (Tables 2 and 3) gave as steep inclinations as would be expected. The observed inclinations may be explained by the drill core being taken from a block, which moved during the impact and/or postimpact and which may have been compacted.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

Several impact-induced effects were observed by petrophysical and rock-magnetic properties of the El'gygytgyn drill core. Comparison between El'gygytgyn dacitic rocks from the UB section and the unshocked dacites from Iceland (Airo 1990) show slightly lower bulk densities for drill core samples (Table 1). This may be due to differences in lithological and mineral/chemical composition, as well as due to differences in age (Ladygin and Frolova 2001) or due to shock-induced effects, such as fracturing, in the El'gygytgyn samples. The latter interpretation is supported by this study, which clearly showed the shock led to marked increases in bulk density and the decrease in porosity in the UB section. Unlike the dacites, the ignimbrites from the LB section of the drill core have higher bulk densities than ignimbrites from Kamchatka (Frolova 2008) and from Iceland (Airo 1990). This may again be due to lithological differences or may reflect compaction due to the impact. The comparison of magnetic properties of this study to unshocked lithologies show that susceptibility values of ignimbrites from the El'gygytgyn drill core are lower than in Icelandic and Kamchatka ignimbrites (Airo 1990; Frolova 2008), while dacites from the drill core have higher susceptibility than those from Iceland (Table 1). Although the differences in susceptibilities are noticeable, they do not differ by orders of magnitude as is the case of NRM, which is much lower in the dacites from the drill core compared with those from Iceland. This clearly indicates the presence of impact effects by either shock demagnetization or the maghemitization due to postimpact hydrothermal activity in the UB. Samples with weak Q values reflect shock or shock-induced alterations.

Two onsets of impact-induced changes were observed in petrophysical profiles (Fig. 2). First, a distinct shift from rapid changes in bulk density and porosity, as well as susceptibility, to relatively minor variations with depth, occurs close to the UB/LB boundary (see Raschke et al. 2012b). The second change, at around 470 m, seems to coincide with the lithological boundary between an upper and a lower LB (ULB and LLB). Below ULB/LLB boundary, the bulk densities seize to decrease, while the distinct decrease in electrical conductivity turns into an increase. Both these boundaries coincide with significant increases in relative iron oxide concentration (Raschke et al. 2012a), while the latter boundary also coincides with a break in the welded ignimbrite. Seismic data support a boundary within the LB, as velocities in the ULB show a clear increase with depth, while those in the LLB show a clear decrease with depth (Fig. 2). Niessen et al. (2007) suggested that brecciation existed up to 600 m below the lake floor, using a value of over 5 km s−1 for vp in the crystalline target. The physical properties in this study clearly show that strong brecciation ends by approximately 420 m depth (UB/LB boundary), while shock effects completely decay at approximately 470 m (ULB/LLB boundary). Thus, the weakly shocked ULB should be included in any further seismic study of the structure, with its vp value between the uncompressed 3.2 km s−1 (same as in the PIB) and the uncompressed 4.8 km s−1 values measured here. Like the petrophysical data, the magnetic data hint at a boundary between an ULB and a LLB. The ULB is characterized by the magnetic properties of titanomagnetites with some maghemitization, whereas the LLB shows no maghemitization at all. In the LB, the anisotropy of susceptibility is more constant than in the other sections, but even within the LB, it is more stable in the LLB than in the ULB (Fig 2).

The paleomagnetic data indicate intermediate to high AF coercivities of the ChRM and suggest that ChRM components are either thermal remanent magnetization (TRM) or chemical remanent magnetization (CRM). They are unlikely to be shock remanent magnetization (SRM) as SRM is mostly restricted to the low coercivity fraction of minerals capable of carrying remanence (Henkel et al. 2002). The El'gygytgyn impact occurred close to the Gauss/Gilbert (R [RIGHTWARDS ARROW] N) reversal, which is currently dated at 3.596 Ma (Ogg and Smith 2004). Using the 3.54 ± 0.04 Ma date given by Layer (2000) would imply that it occurred after the Gauss/Gilbert reversal in a period of normal polarity. Based on personal communication (2012) with Norbert Nowaczyk from GFZ Potsdam, the well-stratified postimpact sediment layers just above the impactite section (RS) of the core showed a distinct normal polarity, and there is evidence that these sediments were deposited soon after the impact. In the RS, the inclinations of the NRM closely resemble those in the sediment layers. Here, the variation in the rock-magnetic properties (including NRM) in the PIB and UB were explained as the result of oxidation due to postimpact hydrothermal activity. If this is correct, the acquisition of the normal ChRM with the higher inclinations in the upper sections was likely to be impact related. Therefore, it is favored that the higher inclination (consistent with the impact age), normal polarity ChRM in the RS, PIB, and UB sections have a reverse polarity overprint and an impact-related origin. The observed reduced inclinations and scattered declinations are due to block movements and compaction. Alternatively, both normal and interpreted reverse ChRM components may have a preimpact origin. The pattern of dominant normal polarity interspaced with the interpreted reverse component exists all the way down the core, and at the lowermost depths, there were only very weak shock effects and little evidence of hydrothermal activity.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

The upper part of the core is characterized by lower bulk densities and higher porosities, while values for these parameters became quite stable in the LB section. This pattern supports that in the upper sections, there was greater shock propagation, which decreased with depth and was weak or completely absent in the LLB. Grain densities had a slight but constant decrease with depth, in agreement with a decrease in both the amount of impact melt (Raschke et al. 2012b), and in the concentrations of the oxides of the heavier metals (Raschke et al. 2012a) down the core. Compressed seismic velocities decreased with depth and are probably related to the decrease in grain density, whereas uncompressed vp values increased down the core, probably due to fewer cracks at greater depths. The fracture distribution may also account for the observed decrease in electrical conductivity with depth. Peaks in electrical conductivity values coincided with higher concentrations of metal oxides.

Temperature–susceptibility data reveal the presence of low titanium iron-titanium oxides. Postimpact hydrothermal activity oxidized titanomagnetites into titanohematites and/or titanomaghemites in the PIB and UB, while in the LB, the ferrimagnetic titanomagnetites showed little oxidation. The differences in these phases of iron-titanium oxides were reflected in the rock-magnetic properties (anisotropy of susceptibility, susceptibility, NRM, and Q-ratio). The variation in these properties supported the lithological profile by Raschke et al. (2012b). Additionally, the physical properties and rock-magnetic data hint at a boundary within the LB.

Paleomagnetic data gave three main components. A low coercivity component was interpreted as VRM. Two medium-to-high coercivity components were interpreted as ChRM, one with normal polarity (dominant) and the other, which was much less common, with an interpreted reverse polarity. The inclinations of the normal polarity ChRM were steeper in the upper sections of the core, and the origin of the normal ChRM in these sections was considered impact related. This was in agreement with impact occurrence after the Gauss/Gilbert reversal. The attempt to reorientate the core produced scattered declinations and shallower than expected inclinations. This may have been a result of the core comprising material added at different times after the impact, impact-related block movements and/or compaction.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References

We are very grateful to the Nordenskiöld Samfundet for providing funding toward this study. Special thanks to the ICDP and partner organizations, such as the DOSECC, for the drilling of the impact structure and providing data on the core. Ilkka Lassila kindly assisted in the measurement and analysis of the seismic velocities. Satu Mertanen and Matti Kauranne helped with sample preparation. Meri-Liisa Airo and Julia Frolova gave access to the Finnish and Russian petrophysical databases, respectively. The contributions of Tomas Kohout are appreciated. We also express gratitude to Steven Alexander, who contributed to the editing of the text. We greatly acknowledge Jüri Plado and an anonymous reviewer for their reviews, which, together with the comments of Christian Koeberl, have considerably improved the original manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. The El'gygytgyn Impact Structure and Drilling
  5. Methods
  6. Results
  7. Discussion
  8. Conclusions
  9. Acknowledgments
  10. Editorial Handling
  11. References
  • Airo M.-L. 1990. Susceptibilities, intensities of remanence and densities of Precambrian and recent volcanic rocks: Examples from Iceland and northern Fennoscandia. Geological Survey of Finland, Report of Investigation 93.
  • Consolmagno G. J. and Britt D. T. 1998. The density and porosity of meteorites from the Vatican collection. Meteoritics & Planetary Science 33:12311241.
  • French B. M. 1998. Traces of catastrophe: A handbook of shock-metamorphic effects in terrestrial meteorite impact structures. LPI Contribution No. 954. Houston, Texas: Lunar and Planetary Institute. 120 p.
  • Frolova Y. V. 2008. Specific features in the composition, structure and properties of volcaniclastic rocks. Moscow University Geological Bulletin 63:2837.
  • Gebhardt A. C., Niessen F., and Kopsch C. 2006. Central ring structure identified in one of the world's best-preserved impact structures. Geology 34:145148.
  • Gurov E. P. and Gurova E. P. 1979. Stages of shock metamorphism of volcanic rocks of acid composition on example of the El'gygytgyn crater (Chukotka). Doklady Academii Nauki 249:11971201.
  • Gurov E. P. and Koeberl C. 2004. Shocked rocks and impact glasses from the El'gygytgyn impact structure (Russia). Meteoritics & Planetary Science 39:14951508.
  • Gurov E. P. and Yamnichenko A. Y. 1995. Morphology of rim of complex terrestrial craters (abstract). 26th Lunar and Planetary Science Conference. p. 533.
  • Gurov E. P., Valter A. A., Gurova E. P., and Serebrennikov A. I. 1978. Impact meteorite crater El'gygytgyn in Chukotka. Doklady Academii Nauki 240:14071410.
  • Gurov E. P., Gurova E. P., and Rakitskaya R. B. 1979a. Stishovite and coesite in shock-metamorphosed rocks of the El'gygytgyn crater in Chukotka. Doklady Academii Nauki 248:213216.
  • Gurov E. P., Valter A. A., Gurova E. P., and Kotlovskaya F. I. 1979b. El'gygytgyn impact crater, Chukotka: Shock metamorphism of volcanic rocks (abstract). 10th Lunar and Planetary Science Conference. p. 479.
  • Gurov E. P., Koeberl C., and Yamnichenko A. 2007. El'gygytgyn impact crater, Russia: Structure, tectonics, and morphology. Meteoritics & Planetary Science 42:307319.
  • Henkel H., Reimold W. U., and Koeberl C. 2002. Magnetic and gravity model of the Morokweng impact structure. Journal of Applied Geophysics 49:129147.
  • Järvelä J., Pesonen L. J., and Pietarinen H. 1995. On palaeomagnetism and petrophysics of the Iso Naakkima impact structure southeastern Finland. Open File Report Q19/29. 1/3232/95/1, Laboratory for Palaeomagnetism, Department of Geophysics, Geological Survey of Finland, Espoo. 43 p.
  • Koeberl C. and Milkereit B. 2007. Continental drilling and the study of impact craters and processes—An ICDP perspective. In Continental scientific drilling, edited by Harms U., Koeberl C., and Zoback M. D. Heidelberg: Springer. pp. 95161.
  • Komarov A. N., Koltsova T. V., Gurova E. P., and Gurov E. P. 1983. Determination of the El'gygytgyn impactites age by fission track method. Doklady Academii Nauki UkrSSR. 10:1311.
  • Ladygin V. M. and Frolova J. V. 2001. The stages of rock petrophysical property alteration during the earth's evolution. Proceedings of the 6th Nordic Symposium on Petrophysics. pp. 19.
  • Lassila I. J., Karlqvist R., Elbra T., Gates F. K., Pesonen L. J., and Haeggström E. 2010. Ultrasonic velocity of the upper gneiss series rocks from the Outokumpu deep drill hole, Fennoscandian shield––Comparing uniaxial to triaxial loading. Journal of Applied Geophysics 72:178183.
  • Layer P. W. 2000. Argon-40/Argon-39 age of the El'gygytgyn event, Chukotka, Russia. Meteoritics & Planetary Science 35:591599.
  • Masaitis V. L. 1999. Impact structures of northeastern Eurasia: The territories of Russian and adjacent countries. Meteoritics & Planetary Science 34:691711.
  • Melles M., Brigham-Grette J., Minyuk P., Koeberl C., Andreev A., Cook T., Fedorov G., Gebhardt C., Haltia-Hovi E., Kukkonen M., Nowaczyk N., Schwamborn G., Wennrich V., and the El'gygytgyn Scientific Party. 2011. The Lake El'gygytgyn scientific drilling project—Conquering Arctic challenges through continental drilling. Scientific Drilling 11:2940.
  • Niessen F., Gebhardt A. C., Kopsch C., and Wagner B. 2007. Seismic investigation of the El'gygytgyn impact crater lake (Central Chukotka, NE Siberia): Preliminary results. Journal of Paleolimnology 37:4963.
  • Ogg J.G. and Smith A.G. 2004. The geomagnetic polarity time scale. In A geological timescale 2004, edited by Gradstein F., Ogg J., and Smith A. Cambridge, UK: Cambridge University Press. pp. 6386.
  • Pesonen L. J. 2006. Kiinteän maan geofysiikan uudet laboratoriot Kumpulassa. Geologi 58:144148.
  • Raschke U., Reimold W. U., and Schmitt R. S. 2012a. Petrography and geochemistry of the impactites of the ICDP drill core from Lake El'gygytgyn, NE Russia. Meteoritics & Planetary Science, doi:10.1111/maps.12087.
  • Raschke U., Reimold W. U., Zaag P., Pitarello L., and Koeberl C. 2012b. Lithostratigraphy of the impactite and bedrock section in ICDP drill core D1c from the El'gygytgyn impact crater, Russia. Meteoritics & Planetary Science, doi:10.1111/maps.12072.
  • Sengupta D., Bhandari N., and Watanabe S. 1997. Formation age of Lonar meteor crater, India. Revista de Fisica Aplicada e Instrumentaçao 12:17.