Summanen structure: Further geological and geophysical evidence of a meteorite impact event in Central Finland

The Summanen structure is located in Central Finland and is one of Finland's 12 known meteorite impact structures. In 2017, the discovery of Summanen was based on numerous shatter cone boulders with planar deformation features (PDFs) and a circular electromagnetic anomaly, which is 2.6 km in diameter. The site was revisited in 2020 and 2022, and shatter cone‐bearing outcrops were discovered. PDFs and feather features were identified in samples from these outcrops. A total of 38 PDF sets in 27 quartz grains resulted in rational crystallographic orientations concentrating on {10 1¯ 4}, {10 1¯ 3}, {10 1¯ 2}, and {11 2¯ 2}, implying shock pressures of 2–20 GPa. Gravity measurements were taken, and the electrical conductivity of the structure was studied. The gravimetric results revealed a circular negative anomaly of about 4 km in diameter, with an amplitude of −3.5 mGal. Excluding the gravitational effect of water and Quaternary sediments reduces the anomaly to −1.6 mGal. A bowl‐shaped conductive layer, likely containing relict saline water in the impact‐fractured bedrock, was identified to a depth of 240 m. Topographic and bathymetric data were combined to determine the impact's effect and interpret the level of erosion. Cobbles of sedimentary sand‐ and siltstones were found on the coastline of Lake Summanen. Based on their similarity to those found in the Söderfjärden impact crater with a Cambrian age, it is likely that these rocks and post‐impact infill are also of a similar age.


INTRODUCTION
The Summanen impact structure (62°39.0 0 N, 25°22.5 0 E) is located within the Paleoproterozoic Central Finland Granitoid Complex (CFGC) in the Fennoscandian Shield and is covered by Lake Summanen (Figure 1). The structure was initially discovered using low-altitude airborne geophysical data, which revealed circular electromagnetic in-phase and resistivity anomalies 2.6 km in diameter (Lerssi et al., 2007). Further field campaigns and laboratory studies conducted in 2017 led to the discovery of erratic boulders containing shatter cones and planar deformation features (PDFs), indicating that the exhibit a phenocryst-supported cumulus texture, with the phenocrysts being in contact with each other. The size of subhedral K-feldspar phenocrysts ranges from 1 to 3 cm. To the south of Lake Summanen, the granite is slightly foliated and lacks a porphyritic texture. The foliation can be interpreted as a magmatic flow structure.
Geologist Jouko Vanne from the Geological Survey of Finland (GSF) first noticed the aeroelectromagnetic anomaly (AEA) of Lake Summanen in the early 2000s. The anomaly is unique, as nearby lakes do not have such irregularities, and it is circular despite the lake's noncircular shape (Figure 1b). In addition, the Summanen anomaly is similar to AEAs associated with other Finnish lake-hosted impact craters, such as Lake Karikkoselkä (located 48 km SSW of Summanen; Lerssi et al., 2007;Pesonen et al., 1999) and Iso-Naakkima (located 105 km SE of Summanen; Elo et al., 1993). Lithologies with abnormally high electrical conductivity cause in-phase anomalies because of the basement's porous and salinerich nature. Lake Summanen is located within an area characterized by strong regional magnetic anomalies, with a general SW to NE trend, reaching up to 1000 nT. However, the lake area itself is characterized by weak magnetic relief, lacking prominent anomalies (Plado et al., 2018). This pattern is not a diagnostic feature for an impact origin, but it appears to be typical for many impact structures in Fennoscandia, particularly those that lack highly magnetic impact melt rocks, such as Jänisjärvi (Elo et al., 2000), Karikkoselkä , Keurusselkä (Raiskila et al., 2013), Suvasvesi South (Donadini et al., 2006), and Tvären (Ormö & Blomqvist, 1996).
The Lake Summanen area includes two larger islands, Summassaari and Lamposaari, situated close to the supposed rim of the crater (Plado et al., 2018; see Figure 1b). While Summassaari's basement is entirely covered by glacial and glaciofluvial sediments, Lamposaari Island hosts a few outcrops of crystalline granitic bedrock. This article reports the discovery of shatter cones and  Nironen (2003). The dashed ring outlines the aeroelectromagnetic anomaly (AEM), 2.6 km in diameter (Plado et al., 2018). Shatter cone findings, multifrequency deep sounding electromagnetic (FrEM), and gravity profiles are shown. (Color figure can be viewed at wileyonlinelibrary.com) PDFs on these outcrops, providing additional evidence for an impact origin of the Summanen structure, in addition to previous findings by Plado et al. (2018). However, the new findings, which are located in the supposed rim area of the structure, have raised questions about the structure's diameter. To address these questions and to better understand the underlying structure of the crater, a gravity survey and a multifrequency electromagnetic deepsounding (FrEM) study were carried out in 2020 and 2022.

Multifrequency Deep Sounding Measurements
Multifrequency deep sounding (FrEM) is an electromagnetic frequency domain method used to map variations in electrical conductivity in geological structures. The technique was originally described by Reid and Macnae (1998) and Newman et al. (1989). In February 2020, the FrEM profile was run over the Summanen structure ice while the lake was ice-covered. The 3466 m long profile included 68 measurement points spaced 50 m apart, with ends extending beyond the rim of the crater (Figure 1b). The measuring equipment included a fixed, square-shaped 25 × 25 m transmitter loop and a receiver located 300 m apart during the measurement steps. The results represented the middle points between the transmitter and receiver. In all, 41 frequencies from 100 to 10 000 Hz were used. The FrEM data collected were inverted using software developed by Aittoniemi et al. (1987). A six-layer crustal model was used for modeling, with the top layer being 1 m thick and consisting of air (height of a stand) and ice. The resistivity of the uppermost layer was fixed at 1000 Ωm. The second layer corresponds to lake water with a resistivity of 400 Ωm, and the third layer represents lake bottom sediments with a resistivity of 100 Ωm (resistivity values were obtained from Lerssi et al., 2007). The thicknesses and resistivity values of the other layers were allowed to vary during the inversion and were the targets of interest.

Gravity Measurements and Modeling
During two late winter field campaigns in 2020 and 2022, the GSF measured the gravity field using a Scintrex CG-5 Autograv™ gravity meter on the lake, islands, and surrounding lands. The standard deviation in the gravity measurements was 0.036 mGal. To position the stations, a real-time kinematic GPS with a precision of 5 and 10 cm in horizontal and vertical directions, respectively, was used. The distance between individual stations along NW-SEand SW-NE-oriented profiles (Figure 1b) was 20 m. In addition, nine off-profile measurements were taken to specify the lateral configuration of the crater-associating anomaly. The observed gravity data were corrected for tidal and instrumental variations, latitude, and elevation, and a standard Bouguer reduction assuming a mean density of 2.67 g cm À3 was applied. At the individual stations on the lake, a hole was drilled into the ice, and water depth was measured to further refine the measurements.
The density and porosity of the rock samples were measured at the Petrophysical Laboratory of the University of Tartu (UT) using Archimedes' principle for both water-saturated and oven-dried samples. The obtained petrophysical results were used to interpret the two gravity profiles with the interactive forward modeling program Potent v. 4.16 (Geophysical Software Solutions Pty. Limited). To define the regional field, a second-degree polynomial surface was used, which was calculated based on observed values beyond the anomalous structure. In the model, polygonal prisms with complicated cross sections, variable off-profile lengths, and vertical side faces were utilized.

Sampling and Mineralogical Studies
A total of 14 samples were collected from three outcrops on Lamposaari island by drilling and hammering, while 12 hand samples were obtained from Lamposaari, Lannevesi, and Kyyränkylä ( Figure 1b; Table A1). Petrographic studies were conducted on 21 thin sections prepared from this material. Leica DM2500 P and DM RX polarization microscopes were used to analyze the thin sections. At the University of Tartu, microscopic evidence for shock metamorphism was investigated using a LOMO POLAM P-312 polarizing microscope equipped with a four-axis Universal stage (Fedorov, 1894). The results were plotted manually on the Wulff net with Leitz Wetzlar stereographic plate (see the method e.g., Engelhardt & Bertsch, 1969;Stöffler & Langenhorst, 1994). Projection templates were used to determine Miller-Bravais indices after Ferrière et al. (2009).
The proportions of minerals in the three sedimentary rock samples were determined by point-counting 400 points of each thin section with the SWIFT Automatic Point Counter. Photomicrographs were obtained using a Leica DFC495 camera and Leica Application Suite LAS v.4.2 software. Petrographic studies were carried out in the Department of Geology, UT and the GSF, Kuopio office.

Multifrequency Deep Sounding (FrEM) Measurements
The apparent resistivity of granitoids, which host the Summanen structure, was measured using FrEM ( Figure 2) and found to be between 5000 and 10,000 Ohm.m. However, a continuous, wavy signal with low resistivity (<100 Ohm.m) was detected at the location of the impact structure, extending all over the profile and reaching a maximum depth of 240 m. Hydrogeological studies in Finland (Kietäväinen, 2017;Nurmi et al., 1988) suggest that saline groundwater usually exists below fresh-water layers. Anomalous fracturing caused the depth of the crater-hosted low resistivity signal to be located deeper than that in the surrounding areas. Fractured bedrock and allochthonous breccias may contain saline water, resulting in low resistivities. Lithologies filled with fresh water have relatively high resistivities over low resistivity signals.

Gravity
Gravity data can provide direct insight into the densities and porosities of bedrock, allowing for the discovery of hidden forms of impact structures, including diameter, depth, and complexity. The Bouguer gravity map ( Figure 3) of Summanen reveals a circular local negative bow-shaped anomaly that extends beyond the outline of the AEA, as has been noticed in other impact structures (Pilkington & Grieve, 1992). The anomaly is superimposed on the relatively simple regional field, characterized by negative Bouguer values that exhibit a decreasing trend toward the northwest at ∼ 0.5 mGal km À1 . Unlike the regional trend, the local gravity anomaly has a shorter wavelength, with an amplitude of À3.5 mGal and a diameter of ∼ 4 km (Figure 4), which corresponds to the impact structure. Width of the anomaly at half of its amplitude is between 2.0 and 2.1 km.
The amplitude and width of the anomaly depend on the size and morphology of the structure, as well as the density contrast between impactites and target rocks (Plado et al., 1999). To describe the source bodies of the negative anomaly and study the effect of impact on the gravity field, two 2.5-dimensional models were created along the gravity profiles (Figures 3 and 4). Both models consist of four polygonal prisms to simulate (i) water body, (ii) Quaternary sediments, (iii) post-impact sediments together with allochthonous breccias, and (iv) fractured granitic basement within the unfractured target. In the NW-SE profile (Figure 4c,d), two background blocks with slightly higher density (2.71 g cm À3 ) compared to the background (2.65 g cm À3 ; the density of the background is based on the national petrophysics register of the GSF), including 102 individual density values averaged from the Summanen area and its surroundings, and vertical sides were introduced to simulate the local positive anomalies originating from density variations within the granitic bedrock. While we FIGURE 2. Resistivity model. A cross-section of resistivity variation below the measured FrEM profile from NW to SE over the Summanen impact structure. The depth of the deepest conductive layer reaches down to 240 m below sea level. The dashed vertical lines show the boundaries of the aeroelectromagnetic anomaly. (Color figure can be viewed at wileyonlinelibrary.com) FIGURE 3. Bouguer gravity map of Lake Summanen. The coastline of the lake is shown in black. Blue lines indicate the gravity measurements. The dashed ring from Plado et al. (2018) outlines the aeroelectromagnetic anomaly. A nearly rounded local gravity anomaly, overlapping almost perfectly with the aeroelectromagnetic anomaly, is associated with the Summanen impact crater. Note that the gravity profiles do not cross in the center of the crater structure. (Color figure can be viewed at wileyonlinelibrary.com) had control over the depth of the water body, the boundaries between the underlying prisms and the target were interpretational.
The thickness of the polygonal prism corresponding to Quaternary sediments was set to mirror the thickness of the water body. The non-modeled high-frequency undulations of the gravity signal ( Figure 4) originate from variations in thickness or density in the Quaternary sediments. For the density of these sediments, we used 1.95 g cm À3 based on an analogy with the Iso-Naakkima impact crater gravity model (Elo et al., 1993). Based on a few samples (Table A1) of post-impact sediments and allochthonous impact breccias, we found that the densities of these lithologies are similar and were set to 2.40 g cm À3 in the models. While the allochthonous breccias are overlaid by post-impact sediments, we did not specify the location of their boundary in the models due to almost identical density. Based on the models, the polygonal prism for post-impact sediments and allochthonous breccias is bowl shaped, with the lower boundary located at ∼ 200 m in the central region of the crater.
The deepest bowl-shaped polygonal prism was created to simulate the whole amount of fractured subautochthonous impact-brecciated basement granitoids with a density of 2.58 g cm À3 . While it is theoretically expected that densities increase as distance from the impact center increases, due to weaker shock and rarefaction waves during the cratering process and geometrical effects, the modeling software does not accommodate gradual changes in petrophysical properties. Although creating multiple can be an alternative method, it does not offer more insights into the nature of an impact crater; hence, this approach was not employed.
As previously mentioned, the central area of the crater structure exhibits a maximum anomaly of À3.5 mGal. However, it is worth noting that this anomaly is partially attributed to the presence of a water body and Quaternary sediments. By excluding the gravitational attraction of water, the maximum amplitude decreases to À2.0 mGal (as shown in Figure 4). Furthermore, removing the polygonal prisms that correspond to Quaternary sediments results in a maximum amplitude of À1.6 mGal. This value represents the net gravitational effect of the impact and ancient post-impact infill. Exclusion of gravitational effects of water and Quaternary sediments slightly reduces the width of the anomaly (half width of the anomaly decreases by a few hundred meters).

Petrography
The studied samples can be classified into three groups: (1) shatter cone-carrying granites, (2) lithic allochthonous breccias, and (3) sedimentary rocks. Sample details are available in the Table A1. Clasts of breccia and shatter cone-carrying rocks have undergone a pre-impact regional Svekofennian metamorphism, generally in greenschist facies (Hölttä & Heilimo, 2017). The studied samples are moderately altered, related to regional metamorphism or synÀ/post-impact hydrothermal processes. Typical alteration features are sericitization and chloritization. Chlorite is a common secondary mineral and alteration product of hydrothermal fluids. Hematite is frequently present as a pigment and occurs as an accessory mineral. It could represent a weathering product derived from ironbearing minerals in oxygenated environments or a primary precipitate in hydrothermal environments. The iron coating, indicating oxic conditions, has likely developed during the last few thousand years (Lerssi et al., 2007).

Shatter Cone-Carrying Granites
In 2019, while conducting fieldwork on Lamposaari Island (Figure 1b), in situ shatter cones were discovered at three closely located granitoid outcrops (designated I, II, III). The outcrops are situated in the NE part of the island, at the center of a small peninsula. Despite being covered in soil and vegetation, the outcrops are shallow and not significantly different from the surrounding topography. Outcrops I and II are 40 m apart, and the distance between outcrops I and III is 9 m. Notably, these outcrops are located ∼ 350 m outside of the present aeroelectromagnetic rim.
The bedrock in Lamposaari consists of coarse-to medium-grained K-feldspar porphyritic granite with occasional aplitic fine-grained granitic veins. Shatter cones have primarily developed in the porphyritic granitoid, with curved and converging cone surfaces and variable striation orientations (Figure 5a,c). Fracturing of the bedrock has occurred along the shatter cone surfaces, with curved and occasionally straight surfaces, but no observable shearing (Figure 5b,d). The fractured surfaces are covered with red-brown iron oxides (hematite; Figure 5b,c) and bluish-black Fe-Mnhydroxides ( Figure 5a).
The aplitic and porphyritic varieties of shatter conecarrying granites comprise quartz, alkali K-feldspar, plagioclase, biotite, and hornblende ( Figure 6a). The porphyritic granite contains K-feldspar phenocrysts measuring 1-3 cm and plagioclase as phenocrysts measuring 0.5-1 cm in size. Apatite and zircon occur as accessory minerals. The surfaces of shatter cones exhibit intense weathering and alteration of K-feldspar, plagioclase, and biotite up to a depth of 2 mm depth from the shatter cone surface (Figure 6b). Kink-banding and alteration to chlorite are common in biotite (Figure 6c,d), while hematite occurs in a botryoidal form and fills the pores and fractures in the minerals (Figure 6b,d).
Occasionally, quartz appears conspicuously white due to the presence of tiny two-phase gas-liquid fluid inclusions that scatter white light (Figure 6e). Plagioclase contains sericite and is fractured (Figure 6f).

Lithic Breccia and Sedimentary Rocks
During the 2020 field trip to Kyyränkylä and Lannevesi (Figure 1b), various boulders were discovered, including lithic allochthonous breccias, intensely fractured porphyritic granites, one sandy siltstone, and two sandstones. The light brown sandy siltstone (sample SHHI-2021-4) is composed of lithified particles with a bimodal grain size ranging from silt size to medium sand size, 50 μm-1 mm (Figure 9a). Quartz (96%) is the primary mineral, with carbonate serving as a cement material. Sample SHHI-2021-3 is a sandstone with a graded texture, mainly composed of rounded quartz grains (84%) and K-feldspar (14%) with grain sizes varying from 0.5 to 2 mm ( Figure 9b). Hematite (2%) occurs as a pigment mineral. Sandstone sample SHHI-2021-5 primarily consists of rounded quartz grains (93%) with a grain size of up to 1 mm, and the texture is slightly graded (Figure 9c). In general, the silt-and sandstones are mature and composed mostly of quartz, with a minor amount of other minerals.
The allochthonous breccia sample (SHHI-2021-1) is monomictic, composed of lithic clasts up to 2 cm in a fine-grained matrix. The clasts are made up of aplitic granite with K-feldspar (microcline), plagioclase, quartz, and biotite. The matrix is comprised of quartz and K-feldspar fragments ranging 0.5-2 mm in size. Hematite is abundant in the matrix, giving it a reddish coloration (Figure 9d,e).

DISCUSSION
The available data indicate that Summanen is a wellpreserved simple crater caused by a hypervelocity impact. This conclusion is supported by petrographic observations of shocked material and geophysical anomalies typical of impact structures worldwide.
Shatter cones in Summanen have been observed beyond the present aeroelectromagnetical outline of the crater structure which is uncommon but not unique. The presence of shatter cones outside of 2.6 km is evidence that the original structure must have been larger. Similar in situ position shatter cones have been found at the edge and beyond the topographic depression in the Karikkoselkä (2.4 km in diameter) impact structure . Although the original diameter has been reduced by erosion since the formation, no evidence of a much larger diameter and complex morphology has been found. Thus, the formation of shatter cones is related to the shock pressures prevailing at the depth of the present erosional level.
Analysis of topographic and bathymetric data ( Figure 10) indicates that the Summanen structure lacks a clearly defined circular depression or rim feature due to reshaping by glaciers. However, it is apparent that the pre-Quaternary cavity has been partially filled with glacial material and eroded. Bathymetric data reveal a depression on the southeast side of the geophysically derived ring (marked by a dotted black line in Figure 10), which reaches a depth of 20-40 m. This depression follows the outline of the AEA and is curved on the southeastern side. It is likely that this side was subject to abrasion by the latest (Weichselian) ice flow in the direction of 120°. Erratic boulders of impactites and postimpact fill found at the southeastern side from the lake are likely part of breccia from the inner slope and sedimentary layers from the bottom of the structure. The supposed post-impact infill, silt-and sandstones, are relatively soft and loose, hinting at a very local transport.
Silt-and sandstones have only been found near the lake SE from the crater. The closest similar find is associated with the Söderfjärden impact crater (Laurén et al., 1978), located 200 km WNW from Summanen. In Söderfjärden, sandstones have been dated to be of early Cambrian age based on micropaleontology (Tynni, 1978). Although our attempts to date the Summanenassociated material micropaleontologically failed due to the absence of acritarchs, we suggest a Cambrian age based on analogy. The sandstones are unlikely to be Mesoproterozoic (Jotnian) in age since they do not exhibit any recrystallization or other metamorphic features characteristic of Jotnian sandstones. Despite the silt-and sandstones being loose, they contain a significant amount of quartz, suggesting that the material traveled a long distance before settling in the Summanen depression. Therefore, it is likely that the Summanen structure was covered by sea during some period in the Cambrian era. The topographic features were compared to the geophysical data to help to determine the structure's size and interpret the erosion level using the model presented by Hall et al. (2021). Aeroelectromagnetic and gravity data suggest a simple impact structure with a maximum depth (true bottom) of 240 m from the present surface. Applying the calculations in Hall et al. (2021), it appears that ∼ 90 m of material has been eroded from the paleoelevation height at the time of impact. Adding this value to the current depth of 240 m and comparing the sum (330 m) to the present diameter of the crater (2.6 km; presumably the original diameter was larger), the structure is anomalously shallow (depth/diameter ∼ 0.13; see Wünnemann & Ivanov, 2003). This may indicate either a deeper erosional level or a partial uplift of the crater floor during the modification stage. However, if we assume that 90 m of material has eroded since the formation of the crust (maximum possible age of the impact is 1880 Ma), then the maximum rate of erosion would be 0.05 m Myr À1 . Considering the Cambrian age, the long-term maximum rate of erosion is estimated to be between 0.17 and 0.19 m Myr À1 , which is consistent with other estimations, such as Suvasvesi South and Saarijärvi (Hall et al., 2021).
Topographic lineaments dominate the area around the lake (Figure 10), with various orientations that are interpreted here as originating from regional fracture zones. Some Quaternary formations, such as eskers, are oriented in the direction of movement of the ice sheet (NW-SE), topographical features parallel to this direction tend to be emphasized. However, no impact-related radial or concentric lineaments are visible, likely due to their relatively small size and Quaternary cover.
The FrEM method was originally developed for deep mineral exploration but has also proven successful in mapping fracture zones. This method can provide information on the near-surface layers (overburden and the uppermost rock layers) as well as the deeper levels of bedrock, down to a depth of 1500 m. FrEM is capable of investigating 3D geophysical conductivity contrasts to identify good conductors from more resistant surroundings. In Summanen, FrEM has revealed the presence of a good conductor within the structure, as shown in Figure 2. We propose that the increased conductivity is due to saline fluids filling the impact-caused fractures within the impact breccias. Deep saline groundwaters are commonly found in old craton areas worldwide, and various estimates of their origin have been proposed, including seawater evaporation or freezing followed by infiltration, as well as water-rock interaction. Estimates of their age range from thousands of years to several hundred million years. In Finland, deep saline groundwater is attributed to longterm water-rock interaction lasting tens of millions of years (Kietäväinen, 2017). Generally, brackish or saline groundwater exists below a freshwater layer, with a thickness of a few tens of meters in coastal areas to several hundred meters in inland regions (Nurmi et al., 1988).
Asteroid collisions can have a lasting effect on surface water and groundwater of the target rock, even at great depths (e.g., Chicxulub; Perry et al., 1995, Tswaing, South Africa;McCaffrey & Harris, 1996;Lonar, India;Jhingran & Rao, 1954). It is likely that a layer of saturated saline fluids has formed on the bottom of the Summanen crater due to long-term water-rock interaction in a low-temperature hydrothermal process. The nearest salty groundwater deposit, Kivetty, is located 20 km NE of Summanen Lake. The deposit starts at depths below 200-300 m in granitoid bedrock. In Kivetty, the interaction between the rock and water has caused increased salinity in the upper parts of the bedrock, and it has been suggested by Anttila et al. (1999) that the origin of these salts may derive from ancient hydrothermal activity. It is reasonable to assume that a similar history has deposited the Summanen groundwater.
The study of trapped saline groundwaters in impact craters has been limited. In Finland, a similar layer has been found related to the Lake Karikkoselkä impact crater . It is recognized that impact structures can contain significant groundwater reservoirs (e.g., Lappajärvi in Finland). The age of water is an important factor that affects its chemistry, especially in deep groundwater where the oldest waters have a distinct composition due to rock-water interactions. Exploring deep groundwater in impact structures could provide valuable insights into the hydrological, biological, and chemical evolution of water, and offer new targets for research in these areas.
The age of the Summanen impact crater has not been defined yet, but it must be younger than the target rock age of 1.88 Ga. The presence of sedimentary rocks and the maximum rate of erosion suggest a younger age, possibly up to the Cambrian period. Hematite is commonly found along fractures within the quartz that contains PDFs. If these fractures are related to the impact, then U-Pb geochronology of hematite using LAICPMS could be a feasible method. A seismic study, combined with numerical modeling and drilling, would provide a more comprehensive view of the crater's interior.

CONCLUSIONS
The Summanen structure is a simple impact structure that has been eroded over time and is estimated to be from the late Paleoproterozoic to Cambrian era. The bathymetrical data of the structure do not show typical characteristics of an impact crater; however, a bowlshaped depression can be observed with a low resistivity and circular negative gravity anomaly. Evidence of shock pressures ranging from 2 to 20 GPa at the rim of the structure is indicated by in situ shatter cones and PDFs in quartz that are parallel to {1014}, {1013}, {1012}, and {1122} orientations, providing unquestionable evidence of the impact origin of the Summanen structure.   Porphyritic granite with a coarse-grained texture, composed of quartz, Kfeldspar (microcline), plagioclase, and a minor amount of biotite. The biotite is altered into chlorite, and kink-bands are commonly observed (see Figure 6c). Hematite, in a botryoidal form, fills the pores and microcracks in the minerals (see  Sandstone with a graded texture. The sample is composed of rounded quartz grains (84%) and K-feldspar (14%) with a grain size of 0.5-2 mm ( Figure 9b) ρ w = 2.55 g cm À3 ρ g = 2.66 g cm À3 P = 6.2%