Abstract
Bajada del Diablo is located in the Northern Patagonian Massif, Chubut, Argentina. The study area includes several circular structures found in Miocene olivine basalts of the Quiñelaf Eruptive Complex and in the Late Pliocene/Early Pleistocene Pampa Sastre conglomerates. An impact origin has been proposed for these circular structures. With the aim of further investigate the proposed impact origin, topographic, gravimetric, magnetic, resistivity, palaeomagnetic and electromagnetic surveys in two circular structures (‘8’ and ‘G’) located in Pampa Sastre conglomerates and in basalts of the Quiñelaf Eruptive Complex were carried out. The new geophysical results support the hypothesis of an impact origin. However, the confirmation of such an origin through the findings of shock metamorphism evidences and/or the recovery of meteorites remains elusive.
INTRODUCTION
The actual record of known impact structures and the likely candidates that lack the key diagnostic criteria (Hawke 2004) is heavily biased towards the well-explored areas of Europe, North America and Australia, with one or two orders of magnitude lower density of craters found in the developing continents of Africa, Asia and South America. Particularly, South America has only seven described areas, some of which are only likely candidates that lack key diagnostic criteria. One of the sensible reasons for this bias is the simple fact that Earth scientists have been actively looking for impact craters in Europe, North America and Australia for more than 30 years with access to good regional geological and geophysical information (Grieve 2001).
The most important diagnostic criteria for identification and confirmation of impact structures include: (1) the presence of petrological indicators of shock metamorphism, (2) the recovery of the impactor or the detection of geochemical traces of the extraterrestrial projectile, (3) geophysical signatures and (4) characteristic crater morphology (Koeberl 1997). It is very important to remark that only the presence of shock metamorphism and the discovery of extraterrestrial projectiles (or their traces) can undoubtedly confirm an impact origin. However, geophysical observations provide important supplementary evidences (Koeberl 1997). Pilkington & Grieve (1992) established a set of general criteria that correspond to the geophysical signature of impact craters. These principles can be used to evaluate the hypothesis of impact origin of circular structures.
Bajada del Diablo is located in the Northern Patagonian Massif, Chubut, Argentina (Fig. 1a). The study area (Fig. 1b) includes several circular structures found in Miocene olivine basalts of the Quiñelaf Eruptive Complex and in the Late Pliocene/Early Pleistocene Pampa Sastre conglomerates (Acevedo et al.2009). An impact origin has been proposed for these circular structures (e.g. Acevedo et al.2009, 2012).
(a) Situation of study area in South America and Argentina. (b) Pan sharpened SPOT 5 XS3 XS2 XS1 band image of Filu-Có plateau near Bajada del Diablo showing circular structures ‘8’ and ‘G’.
In a previous contribution, Prezzi et al. (2012) reported the results of detailed land magnetic surveys carried out at two circular structures located on Pampa Sastre Fm. With the aim of further investigate the proposed impact origin of the circular structures identified in Bajada del Diablo, we carried out detailed topographic, gravimetric, magnetic, resistivity, palaeomagnetic and electromagnetic surveys in two circular structures (‘8’ and ‘G’) located in Pampa Sastre conglomerates and in basalts of the Quiñelaf Eruptive Complex, respectively.
The new geophysical results and the additional finding of microspherules in the base of circular structure ‘8’ (Orgeira et al.2016), supports the hypothesis of an impact origin. However, the confirmation of such an origin through the findings of shock metamorphism evidences and/or the recovery of meteorites remains elusive.
GEOLOGICAL SETTING
Bajada del Diablo is situated in the Northern Patagonian Massif. The Santa Anita Formation (Callovian–Oxfordian) is considered the oldest unit in the region. It is composed by limestones and shales, and it represents the sedimentary infill of a structural basin developed in Gondwana since the Triassic. Later on, the area was denuded under tropical climate. Afterward, the Araucanian movements of the Andean Cordillera Cycle increased the relief, the depth of the basin was enlarged and subsequently continental epiclastic/pyroclastic sedimentation of the Chubut Group occurred during the Albian–Turonian times (Ardolino & Franchi 1996). Upwards in the sequence, the Sarmiento Group tuffs (Oligocene and Miocene) are interbedded with lava flows of the Somún Curá Formation, which formed an extensive plateau (Ardolino 1987).
Two main units outcrop in the study area and both of them present the circular structures that are investigated in this work. They are the Quiñelaf Eruptive Complex and the Pampa Sastre Formation. The Quiñelaf Eruptive Complex is composed by trachytes, rhyolites, trachyandesites, trachybasalts and pyroclastic rocks. However, in the study area only the Final Basic Lava Facies crops out, which has an age of 22 ± 1 Ma (Ardolino & Franchi 1996). This Facies corresponds to flood basalts erupted from widely scattered vents, covers wide areas and can form parallel-stratified successions. The Final Basic Lava Facies covers all the older rocks that form the Quiñelaf Eruptive Complex, and constitutes the last effusive event of such Complex. These olivinic basalts are of dark grey colour. They are morphologically, microscopically, compositionally and geochemically very similar to the Somún Curá basalts (Remesal et al.2002). It is very important to remark that no evidences of the occurrence of explosive volcanism or hydrovolcanism in the study area were found. The Pampa Sastre Formation is integrated by conglomerate layers, compositionally and texturally very immature, with basalt boulders immersed in a coarse sandy matrix, interbedded with petrocalcic layers. Ardolino & Franchi (1996) assigned these units to the Pliocene–Early Pleistocene.
The area is structurally described by block tectonics forming uplifted and depressed areas which controlled sedimentation since the Jurassic. Main NE–SW alignments have been determined (Ardolino & Franchi 1996). A reorganization of the different regional blocks took place during the Pehuenche phase of the Andean Cycle, which enabled degradation of the landscape under an arid/semiarid cold climate. This sedimentary material was deposited in mantles composing the Pampa Sastre Formation (Pliocene–Early Pleistocene). A general elevation of the area with a relative lowering of the base level characterized the late Pleistocene times. Subsequently, erosive and sedimentary processes generated three accumulation surfaces in the area (Ardolino & Franchi 1996).
A remarkable geomorphological feature of the area is the volcanic plateau, formed by the lava flows aforementioned. Later on, fluvial and gravity processes modelled the landform, enhancing the areas occupied by the more resistant lavas. Ardolino (1987) referred to this geomorphological unit as a structural tableland, meanwhile González Díaz & Malagnino (1984) classified the Somún Curá Meseta as a volcanic structural plain.
METHODOLOGY
Detailed topography was surveyed at 354 topographic stations in the circular structure ‘8’, using a Pentax R326 total station equipment, which has an internal accuracy of 2 mm in height and horizontal components. On the other hand, in circular structure ‘G’ detailed topography was measured at 378 topographic stations. The station spacing along the surveyed profiles ranges approximately from 10 to 25 m.
20 profiles were surveyed at circular structure ‘8’ and 10 profiles were surveyed at circular structure ‘G’ with a GEM-2 small broadband electromagnetic sensor using five different frequencies (450, 1170, 3930, 13 590 and 47 010 Hz). Profile end points were fixed by GPS.
The total magnetic field was observed at 722 magnetic stations (which do not necessarily coincide with the topographic reference points) located in and outside of the circular structure ‘8’ (located in Pampa Sastre Fm) and at 734 magnetic stations located in and outside circular structure ‘G’ (located in Quiñelaf basalts), with a Geometrics 856 proton magnetometer. The spacing between stations along profiles was approximately 10 m. Three or more observations were taken at each station, and the corresponding values were averaged. Accurate position of each station was determined by GPS. Magnetic field data were corrected for the diurnal variations in the Earth's magnetic field by data of the Trelew magnetic observatory (Intermagnet UNLP, Chubut Province), then the corresponding IGRF (International Geomagnetic Reference Field) value was subtracted.
A 1.50-m-deep trench was dug into the circular structure ‘8’ across the sedimentary levels over the base. From this trench, several samples of the infilling materials were collected. Basalts boulders and sandy matrix from the rims of circular structure ‘8’ and six oriented palaeomagnetic hand samples of Quiñelaf basalts from the rims of circular structure ‘G’ were also collected. The corresponding magnetic susceptibilities were measured using a Bartington MS2 susceptibility meter. Average values of the measured magnetic susceptibilities of the sedimentary infill, Pampa Sastre conglomerates and Quiñelaf baslts are shown in Table 1. Natural remnant magnetization (NRM) was measured using a SQUID 2 G SSR cryogenic magnetometer (Table 1). Detailed alternating fields (AF) demagnetization techniques were applied. With AF up to nine demagnetization steps were performed, using a 2G600 alternating field demagnetizer up to peak fields of 140 mT. Characteristic remnant magnetizations (ChRMs) were isolated using principal components analysis (Kirschvink 1980) when linear trends of vector end points were identified and Mean Directions were obtained applying Fisher statistics (1953; Table 2). Demagnetization results were plotted on vector end point diagrams (Zijderveld 1967) and equal-angle stereographic projections.
Natural remnant magnetization, magnetic susceptibility, Koenisberger ratio and density values of the different geological units.
| Samples | Natural remanent magnetization | Susceptibility | Koenisberger | Density | ||
|---|---|---|---|---|---|---|
| Intensity (A m–1) | Inclination (°) | Declination (°) | SI | Ratio | (g cm–3) | |
| Circular structure 8 | ||||||
| Sedimentary Infill (floor) | ||||||
| 8I | – | – | – | 0.00129 | – | – |
| 8IV | – | – | – | 0.00120 | – | – |
| 8VI | – | – | – | 0.00459 | – | – |
| Pampa Sastre Conglomerates (rims) | ||||||
| PS | – | – | – | 0.00700 | – | – |
| Circular Structure G | ||||||
| Quiñefal basalts (rims) | ||||||
| bd1a (massive) | 15.11 | 10.3 | 358.4 | 0.01230 | 59.77 | 3,01 |
| bd1b (massive) | 16.49 | 13.6 | 351.3 | 0.01330 | 60.33 | 3,009 |
| bd2a (low vesicularity) | 12.85 | 18.1 | 205 | 0.01251 | 49.98 | 2,943 |
| bd2b (low vesicularity) | 14.95 | 19.4 | 197.1 | 0.01511 | 48.14 | 2,941 |
| bd3a (moderate vesicularity) | 3.64 | 23.1 | 173.6 | 0.01222 | 14.49 | 2,836 |
| bd3b (moderate vesicularity) | 2.92 | 21.4 | 178.9 | 0.01231 | 11.54 | 2,856 |
| bd4a (high vesicularity) | 17.47 | 21.4 | 257.3 | 0.01358 | 62.59 | 2,762 |
| bd5a (high vesicularity) | 8.47 | 39.9 | 40.2 | 0.01304 | 31.61 | 2,601 |
| bd5b (high vesicularity) | 6.82 | 10.4 | 94.5 | 0.01399 | 23.72 | 2,796 |
| bd6a (massive) | 8.47 | 40.6 | 82.6 | 0.01320 | 31.22 | 3,051 |
| Sarmiento Group | 0.0129 | –57.9 | 0.6 | 0.00034 | 1.85 | – |
| Samples | Natural remanent magnetization | Susceptibility | Koenisberger | Density | ||
|---|---|---|---|---|---|---|
| Intensity (A m–1) | Inclination (°) | Declination (°) | SI | Ratio | (g cm–3) | |
| Circular structure 8 | ||||||
| Sedimentary Infill (floor) | ||||||
| 8I | – | – | – | 0.00129 | – | – |
| 8IV | – | – | – | 0.00120 | – | – |
| 8VI | – | – | – | 0.00459 | – | – |
| Pampa Sastre Conglomerates (rims) | ||||||
| PS | – | – | – | 0.00700 | – | – |
| Circular Structure G | ||||||
| Quiñefal basalts (rims) | ||||||
| bd1a (massive) | 15.11 | 10.3 | 358.4 | 0.01230 | 59.77 | 3,01 |
| bd1b (massive) | 16.49 | 13.6 | 351.3 | 0.01330 | 60.33 | 3,009 |
| bd2a (low vesicularity) | 12.85 | 18.1 | 205 | 0.01251 | 49.98 | 2,943 |
| bd2b (low vesicularity) | 14.95 | 19.4 | 197.1 | 0.01511 | 48.14 | 2,941 |
| bd3a (moderate vesicularity) | 3.64 | 23.1 | 173.6 | 0.01222 | 14.49 | 2,836 |
| bd3b (moderate vesicularity) | 2.92 | 21.4 | 178.9 | 0.01231 | 11.54 | 2,856 |
| bd4a (high vesicularity) | 17.47 | 21.4 | 257.3 | 0.01358 | 62.59 | 2,762 |
| bd5a (high vesicularity) | 8.47 | 39.9 | 40.2 | 0.01304 | 31.61 | 2,601 |
| bd5b (high vesicularity) | 6.82 | 10.4 | 94.5 | 0.01399 | 23.72 | 2,796 |
| bd6a (massive) | 8.47 | 40.6 | 82.6 | 0.01320 | 31.22 | 3,051 |
| Sarmiento Group | 0.0129 | –57.9 | 0.6 | 0.00034 | 1.85 | – |
Natural remnant magnetization, magnetic susceptibility, Koenisberger ratio and density values of the different geological units.
| Samples | Natural remanent magnetization | Susceptibility | Koenisberger | Density | ||
|---|---|---|---|---|---|---|
| Intensity (A m–1) | Inclination (°) | Declination (°) | SI | Ratio | (g cm–3) | |
| Circular structure 8 | ||||||
| Sedimentary Infill (floor) | ||||||
| 8I | – | – | – | 0.00129 | – | – |
| 8IV | – | – | – | 0.00120 | – | – |
| 8VI | – | – | – | 0.00459 | – | – |
| Pampa Sastre Conglomerates (rims) | ||||||
| PS | – | – | – | 0.00700 | – | – |
| Circular Structure G | ||||||
| Quiñefal basalts (rims) | ||||||
| bd1a (massive) | 15.11 | 10.3 | 358.4 | 0.01230 | 59.77 | 3,01 |
| bd1b (massive) | 16.49 | 13.6 | 351.3 | 0.01330 | 60.33 | 3,009 |
| bd2a (low vesicularity) | 12.85 | 18.1 | 205 | 0.01251 | 49.98 | 2,943 |
| bd2b (low vesicularity) | 14.95 | 19.4 | 197.1 | 0.01511 | 48.14 | 2,941 |
| bd3a (moderate vesicularity) | 3.64 | 23.1 | 173.6 | 0.01222 | 14.49 | 2,836 |
| bd3b (moderate vesicularity) | 2.92 | 21.4 | 178.9 | 0.01231 | 11.54 | 2,856 |
| bd4a (high vesicularity) | 17.47 | 21.4 | 257.3 | 0.01358 | 62.59 | 2,762 |
| bd5a (high vesicularity) | 8.47 | 39.9 | 40.2 | 0.01304 | 31.61 | 2,601 |
| bd5b (high vesicularity) | 6.82 | 10.4 | 94.5 | 0.01399 | 23.72 | 2,796 |
| bd6a (massive) | 8.47 | 40.6 | 82.6 | 0.01320 | 31.22 | 3,051 |
| Sarmiento Group | 0.0129 | –57.9 | 0.6 | 0.00034 | 1.85 | – |
| Samples | Natural remanent magnetization | Susceptibility | Koenisberger | Density | ||
|---|---|---|---|---|---|---|
| Intensity (A m–1) | Inclination (°) | Declination (°) | SI | Ratio | (g cm–3) | |
| Circular structure 8 | ||||||
| Sedimentary Infill (floor) | ||||||
| 8I | – | – | – | 0.00129 | – | – |
| 8IV | – | – | – | 0.00120 | – | – |
| 8VI | – | – | – | 0.00459 | – | – |
| Pampa Sastre Conglomerates (rims) | ||||||
| PS | – | – | – | 0.00700 | – | – |
| Circular Structure G | ||||||
| Quiñefal basalts (rims) | ||||||
| bd1a (massive) | 15.11 | 10.3 | 358.4 | 0.01230 | 59.77 | 3,01 |
| bd1b (massive) | 16.49 | 13.6 | 351.3 | 0.01330 | 60.33 | 3,009 |
| bd2a (low vesicularity) | 12.85 | 18.1 | 205 | 0.01251 | 49.98 | 2,943 |
| bd2b (low vesicularity) | 14.95 | 19.4 | 197.1 | 0.01511 | 48.14 | 2,941 |
| bd3a (moderate vesicularity) | 3.64 | 23.1 | 173.6 | 0.01222 | 14.49 | 2,836 |
| bd3b (moderate vesicularity) | 2.92 | 21.4 | 178.9 | 0.01231 | 11.54 | 2,856 |
| bd4a (high vesicularity) | 17.47 | 21.4 | 257.3 | 0.01358 | 62.59 | 2,762 |
| bd5a (high vesicularity) | 8.47 | 39.9 | 40.2 | 0.01304 | 31.61 | 2,601 |
| bd5b (high vesicularity) | 6.82 | 10.4 | 94.5 | 0.01399 | 23.72 | 2,796 |
| bd6a (massive) | 8.47 | 40.6 | 82.6 | 0.01320 | 31.22 | 3,051 |
| Sarmiento Group | 0.0129 | –57.9 | 0.6 | 0.00034 | 1.85 | – |
In situ and tilt-corrected ChRMs, corresponding mean directions, bedding of each sampled basaltic block and results of the different fold tests carried out.
| In situ ChRMs | Bedding | |||||
|---|---|---|---|---|---|---|
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 86.2° | 31.8° | 0.8° | 145° | 50° | |
| bd1b | 90.4° | 45.9° | 1.4° | 145° | 50° | |
| bd2a | 133.0° | 28.0° | 1.7° | 160° | 35° | |
| bd2b | 127.2° | 31.4° | 2.1° | 160° | 35° | |
| bd3a | 166.9° | 33.5° | 0.4° | 240° | 20° | |
| bd3b | 162.2° | 37.6° | 0.2° | 240° | 20° | |
| bd4a | 227.1° | 23.7° | 0.5° | 15° | 60° | |
| bd5a | 228.7° | 69.9° | 0.5° | 70° | 30° | |
| bd5b | 233.7° | 72.1° | 0.7° | 70° | 30° | |
| bd6a | 193.4° | 67.4° | 0.3° | 90° | 15° | |
| Mean direction | D = 153.3° | I = 55.6° | R = 8.01 | K = 4.53 | α95 = 25.6° | |
| Tilt-corrected ChRMs | Bedding | |||||
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 136.8° | 63.6° | 0.8° | 145° | 50° | |
| bd1b | 169.7° | 63.7° | 1.4° | 145° | 50° | |
| bd2a | 155.7° | 37.9° | 1.7° | 160° | 35° | |
| bd2b | 153.6° | 43.8° | 2.1° | 160° | 35° | |
| bd3a | 173.4° | 52.3° | 0.4° | 240° | 20° | |
| bd3b | 167.9° | 57.0° | 0.2° | 240° | 20° | |
| bd4a | 187.3° | 38.5° | 0.5° | 15° | 60° | |
| bd5a | 189.0° | 48.7° | 0.5° | 70° | 30° | |
| bd5b | 188.2° | 51.3° | 0.7° | 70° | 30° | |
| bd6a | 188.4° | 52.7° | 0.3° | 90° | 15° | |
| Mean direction | D = 172.0° | I = 52.1° | R = 9.73 | K = 32.92 | α95 = 8.5° | |
| Fold test (95 per cent confidence level) | ||||||
| Watson & Enkin (1993) | Positive | 100 per cent unfolding | ||||
| Tauxe & Watson (1994) | Positive | 100 per cent unfolding | ||||
| Enkin (2003) | Positive | Best fitting unfolding = 119.6 ± 21.7 per cent | ||||
| In situ ChRMs | Bedding | |||||
|---|---|---|---|---|---|---|
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 86.2° | 31.8° | 0.8° | 145° | 50° | |
| bd1b | 90.4° | 45.9° | 1.4° | 145° | 50° | |
| bd2a | 133.0° | 28.0° | 1.7° | 160° | 35° | |
| bd2b | 127.2° | 31.4° | 2.1° | 160° | 35° | |
| bd3a | 166.9° | 33.5° | 0.4° | 240° | 20° | |
| bd3b | 162.2° | 37.6° | 0.2° | 240° | 20° | |
| bd4a | 227.1° | 23.7° | 0.5° | 15° | 60° | |
| bd5a | 228.7° | 69.9° | 0.5° | 70° | 30° | |
| bd5b | 233.7° | 72.1° | 0.7° | 70° | 30° | |
| bd6a | 193.4° | 67.4° | 0.3° | 90° | 15° | |
| Mean direction | D = 153.3° | I = 55.6° | R = 8.01 | K = 4.53 | α95 = 25.6° | |
| Tilt-corrected ChRMs | Bedding | |||||
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 136.8° | 63.6° | 0.8° | 145° | 50° | |
| bd1b | 169.7° | 63.7° | 1.4° | 145° | 50° | |
| bd2a | 155.7° | 37.9° | 1.7° | 160° | 35° | |
| bd2b | 153.6° | 43.8° | 2.1° | 160° | 35° | |
| bd3a | 173.4° | 52.3° | 0.4° | 240° | 20° | |
| bd3b | 167.9° | 57.0° | 0.2° | 240° | 20° | |
| bd4a | 187.3° | 38.5° | 0.5° | 15° | 60° | |
| bd5a | 189.0° | 48.7° | 0.5° | 70° | 30° | |
| bd5b | 188.2° | 51.3° | 0.7° | 70° | 30° | |
| bd6a | 188.4° | 52.7° | 0.3° | 90° | 15° | |
| Mean direction | D = 172.0° | I = 52.1° | R = 9.73 | K = 32.92 | α95 = 8.5° | |
| Fold test (95 per cent confidence level) | ||||||
| Watson & Enkin (1993) | Positive | 100 per cent unfolding | ||||
| Tauxe & Watson (1994) | Positive | 100 per cent unfolding | ||||
| Enkin (2003) | Positive | Best fitting unfolding = 119.6 ± 21.7 per cent | ||||
Note: Magnetization acquired PRIOR to tilting. ChRM, characteristic remanent magnetization; D, I, declination, inclination; MAD, mean angular deviation; Bedding, strike follows the right-hand rule; R, length of the resultant vector; α95, semi-angle of confidence; K, Fisher's precision parameter.
In situ and tilt-corrected ChRMs, corresponding mean directions, bedding of each sampled basaltic block and results of the different fold tests carried out.
| In situ ChRMs | Bedding | |||||
|---|---|---|---|---|---|---|
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 86.2° | 31.8° | 0.8° | 145° | 50° | |
| bd1b | 90.4° | 45.9° | 1.4° | 145° | 50° | |
| bd2a | 133.0° | 28.0° | 1.7° | 160° | 35° | |
| bd2b | 127.2° | 31.4° | 2.1° | 160° | 35° | |
| bd3a | 166.9° | 33.5° | 0.4° | 240° | 20° | |
| bd3b | 162.2° | 37.6° | 0.2° | 240° | 20° | |
| bd4a | 227.1° | 23.7° | 0.5° | 15° | 60° | |
| bd5a | 228.7° | 69.9° | 0.5° | 70° | 30° | |
| bd5b | 233.7° | 72.1° | 0.7° | 70° | 30° | |
| bd6a | 193.4° | 67.4° | 0.3° | 90° | 15° | |
| Mean direction | D = 153.3° | I = 55.6° | R = 8.01 | K = 4.53 | α95 = 25.6° | |
| Tilt-corrected ChRMs | Bedding | |||||
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 136.8° | 63.6° | 0.8° | 145° | 50° | |
| bd1b | 169.7° | 63.7° | 1.4° | 145° | 50° | |
| bd2a | 155.7° | 37.9° | 1.7° | 160° | 35° | |
| bd2b | 153.6° | 43.8° | 2.1° | 160° | 35° | |
| bd3a | 173.4° | 52.3° | 0.4° | 240° | 20° | |
| bd3b | 167.9° | 57.0° | 0.2° | 240° | 20° | |
| bd4a | 187.3° | 38.5° | 0.5° | 15° | 60° | |
| bd5a | 189.0° | 48.7° | 0.5° | 70° | 30° | |
| bd5b | 188.2° | 51.3° | 0.7° | 70° | 30° | |
| bd6a | 188.4° | 52.7° | 0.3° | 90° | 15° | |
| Mean direction | D = 172.0° | I = 52.1° | R = 9.73 | K = 32.92 | α95 = 8.5° | |
| Fold test (95 per cent confidence level) | ||||||
| Watson & Enkin (1993) | Positive | 100 per cent unfolding | ||||
| Tauxe & Watson (1994) | Positive | 100 per cent unfolding | ||||
| Enkin (2003) | Positive | Best fitting unfolding = 119.6 ± 21.7 per cent | ||||
| In situ ChRMs | Bedding | |||||
|---|---|---|---|---|---|---|
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 86.2° | 31.8° | 0.8° | 145° | 50° | |
| bd1b | 90.4° | 45.9° | 1.4° | 145° | 50° | |
| bd2a | 133.0° | 28.0° | 1.7° | 160° | 35° | |
| bd2b | 127.2° | 31.4° | 2.1° | 160° | 35° | |
| bd3a | 166.9° | 33.5° | 0.4° | 240° | 20° | |
| bd3b | 162.2° | 37.6° | 0.2° | 240° | 20° | |
| bd4a | 227.1° | 23.7° | 0.5° | 15° | 60° | |
| bd5a | 228.7° | 69.9° | 0.5° | 70° | 30° | |
| bd5b | 233.7° | 72.1° | 0.7° | 70° | 30° | |
| bd6a | 193.4° | 67.4° | 0.3° | 90° | 15° | |
| Mean direction | D = 153.3° | I = 55.6° | R = 8.01 | K = 4.53 | α95 = 25.6° | |
| Tilt-corrected ChRMs | Bedding | |||||
| Sample | D | I | MAD | Strike | Dip | |
| bd1a | 136.8° | 63.6° | 0.8° | 145° | 50° | |
| bd1b | 169.7° | 63.7° | 1.4° | 145° | 50° | |
| bd2a | 155.7° | 37.9° | 1.7° | 160° | 35° | |
| bd2b | 153.6° | 43.8° | 2.1° | 160° | 35° | |
| bd3a | 173.4° | 52.3° | 0.4° | 240° | 20° | |
| bd3b | 167.9° | 57.0° | 0.2° | 240° | 20° | |
| bd4a | 187.3° | 38.5° | 0.5° | 15° | 60° | |
| bd5a | 189.0° | 48.7° | 0.5° | 70° | 30° | |
| bd5b | 188.2° | 51.3° | 0.7° | 70° | 30° | |
| bd6a | 188.4° | 52.7° | 0.3° | 90° | 15° | |
| Mean direction | D = 172.0° | I = 52.1° | R = 9.73 | K = 32.92 | α95 = 8.5° | |
| Fold test (95 per cent confidence level) | ||||||
| Watson & Enkin (1993) | Positive | 100 per cent unfolding | ||||
| Tauxe & Watson (1994) | Positive | 100 per cent unfolding | ||||
| Enkin (2003) | Positive | Best fitting unfolding = 119.6 ± 21.7 per cent | ||||
Note: Magnetization acquired PRIOR to tilting. ChRM, characteristic remanent magnetization; D, I, declination, inclination; MAD, mean angular deviation; Bedding, strike follows the right-hand rule; R, length of the resultant vector; α95, semi-angle of confidence; K, Fisher's precision parameter.
Relative gravity measurements were made using a ZLS Burris Standard gravity meter, 39 stations were surveyed in circular structure ‘8’ and 51 stations were surveyed in circular structure ‘G’, with an equidistance of 15 m and a precision of 0.01 mGal. The survey was not tied to an absolute gravity station. The geographic location of each one of the measured gravity stations was determined with the Pentax R326 total station equipment. Current standards for reduction of observed gravity to Bouguer anomaly established by the U.S. Geological Survey (USGS) and the North American Gravity Database Committee were applied through the use of the spreadsheet for reduction of raw data to the Bouguer anomaly developed by Holom & Oldow (2007). With the additional use of solid earth tide and terrain corrections, the complete Bouguer anomaly was calculated. The spreadsheet of Holom & Oldow (2007) calculates the corrections for instrument drift, as well as the DC shift (i.e. constant value added or subtracted to observed gravity values) for multiple-day gravity surveys. To do these corrections relative gravity was measured in the same base station several times a day during the survey. The average instrument drift correction in each station was of approximately of 0.01 mGal. In order to calculate height and Bouguer spherical cap corrections, the height above the WGS84 reference ellipsoid for each station was calculated from the measurements made with the Pentax R326 total station equipment with a precision of 0.05 m and the geoid model World EGM96 15' (http://earth-info.nga.mil/GandG/wgs84/gravitymod/egm96/binary/binarygeoid.html). Earth tide correction was calculated with the software TSoft version 2.1.15 and entered in the spreadsheet of Holom & Oldow (2007). Finally, terrain correction was performed using as input the topographic grid obtained from the detailed topography survey and the software Oasis Montaj (Geosoft Inc.) and also included in the spreadsheet of Holom & Oldow (2007).
Grids of the different measured values and calculated anomalies were produced applying ‘minimum curvature’, a geostatistical gridding method. In case of the magnetic and complete Bouguer anomalies as well as topographic data, grid cell sizes of approximately 10 m × 10 m were used. For the electromagnetic data the grid cell size used was of approximately 5 m × 5 m.
3-D Euler deconvolution was applied to magnetic and complete Bouguer anomalies in order to provide preliminary information of the depth distribution of the sources. This method is based on Euler's homogeneity equation (Reid et al.1990, 2003; Salem & Smith 2005). For the study area, we calculated Euler source points with SI = 1 for magnetic data and SI = 0 for gravity data and window sizes of 10, 12 and 15 m using the software Oasis Montaj (Geosoft Inc.).
We also analysed the curvature of the magnetic and gravimetric signals. Curvature is a property of a field which describes how bent is it at a particular point (by quantifying the field deviation from a tangent surface; Roberts 2001). Therefore, curvature quantifies surface deviation from a perfect plane. Roberts (2001) determined the most useful curvatures as those defined by orthogonal planes to the surface which termed normal curvatures. Normal curvatures can be combined defining important curvature attributes. Calculation of curvature attributes (Roberts 2001) has been applied in 3-D seismic, medicine and to the study of gravity and magnetic fields (Schmidt & Götze 2003; Riller et al.2006; Prezzi & Lince Klinger 2010). Curvature attributes allow to highlight a particular aspect or property of a surface, which otherwise would be difficult or impossible to observe. Curvature attributes are generated by the JAVA code ‘Curvature’ by S. Schmidt (personal communication). The dip angle attribute, which is the angle between the vector normal to the studied surface (i.e. complete Bouguer or magnetic anomaly in this contribution) at a point on the surface and the vertical at the same point (Roberts 2001), was calculated. This attribute emphasizes lineaments and contacts which separate geological domains with different densities or susceptibilities. The most negative curvature of the magnetic anomalies was also calculated. Such attribute shows the most negative values of all possible normal curvatures, thus enhancing the visibility of single lineaments in the surface, with a polygonal appearance (Roberts 2001). This set of curvature methods was selected with the aim of highlighting the possible existence of susceptibility or density contrasts between the rim, the base and the outer sector of the circular structures.
Analytic signal, vertical derivatives and tilt derivatives of magnetic and complete Bouguer anomalies were also calculated in order to facilitate the location of shallow contacts between different geological units. The magnetic anomalies were reduced to the pole using the software Oasis Montaj (Geosoft Inc.). Pole reduction transforms the magnetic field to a fictitious magnetic pole, centring the magnetic anomalies on the causing bodies.
An application of Poisson's relation allows the conversion of the total-field magnetic anomaly into the gravity anomaly that would be observed if the magnetization distribution were to be replaced with an identical density distribution. The result of this conversion is called pseudo-gravity and has several important applications. In practice, gravity and magnetic anomalies often arise from different sources, so the application of such a process is limited to special situations. However, a pseudo-gravity anomaly, calculated from the measured magnetic field, can be compared directly with measurements of the gravity field. Such comparisons might help to build an interpretation of the shape and size of the source, or at least permit an investigation of the ratio M/d and how it varies within the source, where d is the density and M is the intensity of magnetization. In this study, pseudo-gravity anomalies were estimated from the magnetic anomalies obtained for circular structures ‘8’ and ‘G’.
3-D density models were developed in order to explore the nature of the bodies responsible for the detected gravity anomalies. The average density of the Quiñelaf basalts was determined in the laboratory (Table 1). The modelling software IGMAS+ was used (Götze 1978; Götze & Lahmeyer 1988). IGMAS+ essentially uses triangulated polyhedrons to approximate areas of constant density and/or susceptibility within the Earth's crust and mantle. The numerical algorithms were developed by Götze (1984) and permit the calculation of gravity field and its gradients as well as magnetic field components and components of remnant and induced magnetic fields in one program step. The software package IGMAS+ eases the 3-D interpretation of gravity and magnetic data bases. It uses an interoperable 3-D Geoinformation System (IOGIS) and its functions (e.g. data queries, interoperability, visualization and interdisciplinary interpretations in an object-oriented data environment) to integrate other geophysical models, information, and data from both geophysics and geology (Schmidt & Götze 1999; Breunig et al.2000). The initial geometries of the 3-D modelled bodies are pre-defined by the user on a series of parallel vertical cross-sections. The automated triangulation of model surfaces between the parallel vertical cross-sections, allows the construction of complicated model geometries. The optimal fit between observed and calculated anomalies is achieved during forward modelling by iterative geometry changes introduced by the user, according to the constraining information incorporated in the IOGIS. Following, the user manually modifies the geometry of the different bodies composing the 3-D density model, in view of the additional geological and geophysical information integrated in the model. The final model should minimize the differences between observed and calculated Bouguer anomalies (i.e. residual anomalies). Residual anomalies should have lower amplitude than the estimated error in the observed Bouguer anomaly and should show a tight concentration around zero with low standard deviation.
Four vertical electrical soundings (VES) were carried out in and outside circular structure ‘8’ along a profile using a digital resistivity meter GEOMETER MPX-400 (Ponti Electronics) and a Schlumberger electrode array. The position of each one of the VES points was determined using GPS. The potential electrodes separation was of 3 m and the maximum current electrode separation was of 80 m. An inverse 2-D model of resistivity distribution along the profile was numerically obtained in the form of a simple rectangular cell model using the software RES2DINV of Geotomo (Loke 1996–2002, 2001). This program allows an estimation of the cells resistivity (model parameters) that adjusts the quantities measured at surface, within certain discrepancy. At first, the quantities derived from the field measurements are presented in the form of a pseudo-section, a contour diagram in which the apparent resistivity values are assigned to a predefined location according to the array type used (Telford et al.1990). During the inversion routine the initial model parameters are modified and improved by solving a least-squares equation (Lines & Treitel 1984). This Gauss–Newton equation determines the corresponding change in the model parameters that should reduce the sum of squares of the discrepancies between the apparent resistivity values calculated from the cell model and the apparent resistivity values deduced from the real field data, while also minimizing the change in the model parameters between iterations. The discrepancy between the calculated values of apparent resistivity and those inferred from field data are expressed through the rms. Finally, a contour diagram showing the resistivity distribution of the modelled profile is obtained. Topography information in the cell model was incorporated by means of a distorted finite-element mesh: all nodes along the same vertical line are shifted the same amount according to the elevation of the ground surface (uniformly distorted grid).
RESULTS
Circular structure ‘8’
The circular structure ‘8’ has a simple geometry, it is bowl-shaped with a rim diameter of approximately 300 m and a maximum apparent depth of 15 m (Fig. 2). The interior has been partially filled in by debris flows from the rims and wind-blown sands (Acevedo et al.2009).
Detailed topography of circular structure ‘8’. Black dots: topographic stations.
The magnetic anomalies (Fig. 3) show a slightly negative and relatively flat signal (approximately of −20 to −130 nT) at the bottom of the structure. On the other hand, near the rim, high-amplitude and localized (short wavelength) anomalies ranging from 900 to −880 nT are observed (Fig. 3). Such large amplitude and short wavelength anomalies are mostly of dipolar character, some of them are normal (Fig. 3) while others are reverse (Fig. 3). Furthermore, dipolar anomalies showing vertical axis rotations of variable sense and magnitude were detected (Fig. 3). The presence of reverse and rotated dipolar anomalies indicates that at least some of the detected anomalies are dominated by the remnant magnetization instead of the induced one. This type of short wavelength and high-amplitude anomalies are not detected out of the circular structure.
Magnetic anomaly detected in and out of circular structure ‘8’. White dots: magnetic stations, black lines: topography contour lines.
In Fig. 4 we present the detailed topography together with the corresponding Euler solutions considering a SI of 1 and a window size of 10 m. In order to solve Euler's equation, horizontal and vertical gradients of potential field data are required, which were calculated in the frequency domain (Gunn 1975). It can be observed that most of the solutions are located below the rim, at maximum depths of approximately 10–15 m (Fig. 4). These maximum depths of Euler solutions would be in agreement with an average apparent depth around 10–15 m documented for circular structure ‘8’ (Fig. 2). However, it should be noticed that Euler source points are mathematical fictions and do not point to real geological point masses. Consequently, it should be taken into account that Euler calculations could not actually portray real distributions of magnetic material.
Euler solutions corresponding to the magnetic anomaly of circular structure ‘8’, calculated using a SI = 1. Shaded relief is shown in grey scale.
Reduced to the pole magnetic anomalies, curvature attributes, the analytic signal and the first vertical derivative of the magnetic anomalies are shown in the Supporting Information. The complete Bouguer anomaly display a circular pattern with a relative minimum value of approximately –1.6 mGal, which correlates with the bottom of the structure (Fig. 5). This negative anomaly indicates the existence of a mass deficit beneath the floor of the structure ‘8’.
Complete Bouguer anomaly detected in circular structure ‘8’. White dots: gravity stations, black lines: topography contour lines.
Fig. 6 shows the detailed topography along with the corresponding Euler solutions considering a SI of 0 and a window size of 10 m. It can be observed that most of the solutions are located below the circular structure rim, at depths greater than 35 m (Fig. 6). These maximum depths of Euler solutions would be in accordance with the average apparent thickness of Pampa Sastre conglomerates in the study area.
Euler solutions corresponding to the complete Bouguer anomaly of circular structure ‘8’, calculated using a SI = 0. Shaded relief is shown in grey scale.
Curvature attributes, the tilt derivative and the first vertical derivative of the complete Bouguer anomaly are shown in the Supporting Information.
A 3-D forward gravity model was developed in order to test the existence of density contrasts between the rim and the bottom of the circular structure. These variations could cause the detected complete Bouguer anomaly (Fig. 5). An initial thickness of approximately 35 m and average densities of 2.6 and 2.7 g cm–3 were used to model the Pampa Sastre conglomerates. Thickness estimate is in accordance with the expected lateral thickness variation of Pampa Sastre Formation in the study area. From the trench dug in the centre of the structure several samples of the infilling materials were collected and the complete absence of basalt boulders and blocks was determined. Average densities of 1.6 and 2.2 g cm–3 were used to model infilling sediments. Below Pampa Sastre Formation, the presence of the Sarmiento Group was modelled assuming a density of 2.2 g cm–3. In order to reduce the ambiguity and the uncertainty inherent to the gravimetric method, certain aspects of the density structure during the forward modelling were fixed. Thereby, the densities assigned to each one of the modelled bodies were predetermined and were not changed to attain a better fit of measured and modelled anomalies. Our model is composed of 17 parallel E–W sections extending down to 100 m. Fig. 7 shows six of such sections, displaying measured and calculated anomalies and the corresponding subsurface density structure. The forward modelled 3-D density structure thoroughly reproduces the measured gravity field. The residual complete Bouguer anomaly depicts the differences between measured and modelled anomalies (Fig. 8). Only less than ∼5 per cent of the study area presents residual anomalies greater than ±0.2 mGal (Fig. 8). These anomalies are of very short wave length and not systematically distributed across the map. Residual anomaly values are tightly concentrated around 0 mGal, with a standard deviation of 0.21 mGal and a correlation coefficient of 0.903. These results suggest that our model satisfactorily represents the density distribution below circular structure ‘8’. A 3-D view of the measured complete Bouguer anomaly and the modelled 3-D density structure is showed in Fig. 9.
Parallel E–W vertical sections composing our 3-D gravity model of circular structure ‘8’. Densities in g cm−3, red bodies: Pampa Sastre conglomerates, pale blue bodies: infilling sediments.
(a) Measured complete Bouguer anomaly in circular structure ‘8’, (b) modelled complete Bouguer anomaly in circular structure ‘8’, (c) residual complete Bouguer anomaly in circular structure ‘8’.
3-D view of the 3-D density model of circular structure ‘8’, showing measured complete Bouguer anomaly.
A pseudo-gravity anomaly was calculated from the measured magnetic anomalies considering an average density contrast of 0.5 g cm–3 (in accordance with the densities used in our 3-D modelling) and an average magnetization of 150 nT (calculated from the average measured magnetic susceptibilities: Table 1; Fig. 10). This pseudo-gravity anomaly is very much alike to the measured complete Bouguer anomaly (Fig. 5). It has a tendency to form a circular pattern and also shows a relative minimum of approximately −1 mGal (which approximately coincides with the centre of the circular structure).
Pseudo-gravity calculated from magnetic anomaly of circular structure ‘8’. Black lines: topography contour lines.
For all used frequencies, the electromagnetic profiles show near surface lower apparent electrical conductivities in the structure bottom, while the rim possess notably higher values. Fig. 11 shows the results obtained using a frequency of 3950 Hz. Apparent conductivities range between 3600 and 3750 mS m–1 in the structure floor, while at the rim apparent conductivities ranging between 3751 and 4100 mS m–1 were registered.
Apparent electrical conductivity of circular structure ‘8’ obtained using a frequency of 3950 Hz. Black lines: topography contour lines.
Fig. 12 displays the location of the four vertical electrical soundings (VES) carried out in the centre, the rim and outside of the circular structure ‘8’. These four VES were integrated to produce a NW–SE 2-D cross-section. The corresponding measured and calculated apparent resistivity pseudo-sections and the resulting resistivity inverse model with a vertical exaggeration of 1.73 can be seen in Fig. 13. The resistivity inverse model considering the topography and applying the same vertical exaggeration is presented in Fig. 14. From this model the existence of a very low resistivity anomalous zone beneath the centre of the circular structure, with resistivity values ranging between ∼50 and 250 ohm.m and extending down to at least 20 m depth, is identified. On the other hand, below the circular structure rim very high resistivities ranging between ∼ 800 and 1500 ohm.m are observed.
Detailed topography of circular structure ‘8’ showing the location of the four VES carried out. Black diamonds: location of each VES.
Top panel: measured apparent resistivity pseudo-section. Middle panel: calculated apparent resistivity pseudo-section. Bottom panel: inverse resistivity model section. Black arrows: location of each VES in circular structure ‘8’. Vertical exaggeration: 1.73.
Inverse resistivity model section of circular structure ‘8’ including detailed topography. Vertical exaggeration: 1.73.
Circular structure ‘G’
The circular structure ‘G’ has a simple geometry, it is bowl-shaped with a rim diameter of approximately 300 m and a maximum apparent depth of 27 m (Fig. 15).
Detailed topography of circular structure ‘G’. Black dots: topographic stations.
The magnetic anomalies (Fig. 16) show a relatively flat signal (approximately of −200 to 500 nT) at the circular structure bottom, which is a distinctive feature of impact related structures. On the other hand, in the structure rim, high-amplitude anomalies ranging from 2400 to −2900 nT are observed (Fig. 16). Such large amplitude anomalies appear to have a dipolar character (Fig. 16). Considering the Koenisberger ratios calculated for the samples of Quiñelaf basalts obtained in the rim of the circular structure (Table 1), the detected magnetic anomalies would be dominated by the NRM. While most of the isolated NRMs have low positive inclinations, the declinations are notably dispersed ranging between 40° and 358°. This last fact is reflected in the varying orientations of the dipolar magnetic anomalies (Fig. 16). High-amplitude anomalies are not detected out of the circular structure (Fig. 16).
Magnetic anomaly detected in and out of circular structure ‘G’. White dots: magnetic stations, black lines: topography contour lines.
In Fig. 17 we present the detailed topography with the corresponding Euler solutions considering to a SI of 1 and a window size of 10 m. The majority of the solutions are located below the circular structure rim, at maximum depths of approximately 15–25 m (Fig. 17). These maximum depths of Euler solutions would be also in accordance with the average apparent depth of circa 20–25 m documented for circular structure ‘G’ (Fig. 15).
Euler solutions corresponding to the magnetic anomaly of circular structure ‘G’, calculated using a SI = 1. Shaded relief is shown in grey scale.
Reduced to the pole magnetic anomalies, curvature attributes, analytic signal and the first vertical derivative of the magnetic anomalies are shown in Supporting Information.
The complete Bouguer anomaly exhibits a circular pattern with a relative maximum value of ∼1.0 mGal, correlated with the circular structure floor (Fig. 18). This positive anomaly indicates the existence of a mass excess below the base of the circular structure ‘G’, in contradiction with the negative anomalies generally detected in impact structures (Pilkington & Grieve 1992).
Complete Bouguer anomaly detected in circular structure ‘G’. White dots: gravity stations, black lines: topography contour lines.
In Fig. 19, we present the detailed topography along with the corresponding Euler solutions considering a SI of 0 and a window size of 10 m. It can be observed, as with the magnetic Euler results, that most of the solutions are located below the circular structure rim, at depths greater than 25 m (Fig. 19). These maximum depths of Euler solutions would be likewise in accordance with the average apparent thickness of Quiñelaf basalts in the study area.
Euler solutions corresponding to the complete Bouguer anomaly of circular structure ‘G’, calculated using a SI = 0. Shaded relief is shown in grey scale.
Curvature attributes, the tilt derivative and the first vertical derivative of the complete Bouguer anomaly are shown in the Supporting Information.
As with structure ‘8’ a 3-D forward gravity model was developed in order to assess the existence of density contrasts between the circular structure rim and floor, which could cause the complete Bouguer anomaly detected (Fig. 18). An initial thickness of approximately 25 m was used to model Quiñelaf basalts, in accordance with the estimated average thickness in the study area. Densities of 2.5, 2.7 and 2.8 g cm–3 were used to model Quiñelaf basalts located along the rims of circular structure ‘G’. On the other hand, a density of 3.0 g cm–3 was used for basalts located in the centre of the circular structure. These densities correspond to average values obtained from laboratory measurements of different samples of Quiñelaf basalts collected along the circular structure rim (Table 1). Massive basalts have an average density of ∼3.0 g cm–3, while vesicular basalts have average densities ranging between ∼2.6 and 2.9 g cm–3 depending on the degree of vesicularity (Table 1). Below Quiñelaf basalts, the presence of the Sarmiento Group was modelled assuming a density of 2.2 g cm–3. As previously discussed for crater ‘8’, the densities assigned to the modelled bodies were not changed in order to properly account for the gravimetric method uncertainty. The model is composed of 21 parallel E–W sections extending down to 100 m. Fig. 20 presents six of such sections, showing measured and calculated anomalies and the corresponding subsurface density structure. 3-D density structure replicates the measured gravity field. The residual complete Bouguer anomaly depicts the differences between measured and modelled anomalies (Fig. 21). Only less than ∼5 per cent of the study area presents residual anomalies greater than ±0.02 mGal (Fig. 21). As with structure ‘8’, such anomalies are characterized by a very short wave length and their distribution is not systematical within the map. Residual anomaly values are focused around 0 mGal, with a standard deviation of 0.02 mGal and a correlation coefficient of 0.995. According to these results, the presented model reasonably denotes the density distribution below circular structure ‘G’. A 3-D view of the measured complete Bouguer anomaly and the modelled 3-D density structure is showed in Fig. 22.
Parallel E–W vertical sections composing our 3-D gravity model of circular structure ‘G’. Densities in g cm–3, green bodies: Quiñelaf basalts, brown body: shocked Quiñelaf basalts and infilling sediments.
(a) Measured complete Bouguer anomaly in circular structure ‘G’, (b) modelled complete Bouguer anomaly in circular structure ‘G’, (c) residual complete Bouguer anomaly in circular structure ‘G’.
3-D view of the 3-D density model of circular structure ‘G’, showing measured complete Bouguer anomaly.
A pseudo-gravity anomaly was calculated from the measured magnetic anomalies considering an average density contrast of 0.2 g cm–3 (in accordance with the densities used in our 3-D modelling) and an average magnetization of 200 nT (calculated from the average measured magnetic susceptibilities: Table 1; Fig. 23). This pseudo-gravity anomaly is very much alike to the measured complete Bouguer anomaly (Fig. 18). It can be observed that the pseudo-gravity anomaly depicts a relative maximum of ∼1.2 mGal (approximately matching the centre of the circular structure) with a tendency to form a circular pattern.
Pseudo-gravity calculated from magnetic anomaly of circular structure ‘G’. Black lines: topography contour lines.
For all used frequencies, the electromagnetic profiles show near surface lower apparent electrical conductivities at the structure bottom, while the rim present notably high values analogous to the signal of structure ‘8’. Fig. 24 shows the results obtained using a frequency of 3950 Hz. Apparent conductivities range between 2000 and 5000 mS m–1 in the circular structure floor, whereas in the rim apparent conductivities ranging between 5000 and 8150 mS m–1 were recorded.
Apparent electrical conductivity of circular structure ‘G’ obtained using a frequency of 3950 Hz. Black lines: topography contour lines.
The attitude of Quiñelaf basalts vary along the rims of circular structure ‘G’. In the outer slope of the rim, a series of basalt blocks whose strike is parallel to the rim with outward dips were identified (Table 2). Thus, six oriented hand palaeomagnetic samples were collected (Fig. 25) with the aim of testing if such attitudes are a primary or secondary feature (i.e. if the basalts were emplaced with its actual attitude, or if their actual attitude was the consequence of tectonic processes that affected the basalts after their emplacement). ChRMs were isolated for each AF demagnetized specimen obtained from the different hand samples applying principal component analysis. Table 2 shows the in situ ChRMs (Fig. 26a) and the corresponding mean direction with its statistical parameters. The Tilt-corrected ChRMs (Fig. 26b) and the corresponding Mean Direction with its statistical parameters are also presented in Table 2. Tilt-corrected ChRMs are significantly better grouped. Distinct fold tests (Watson & Enkin 1993; Tauxe & Watson 1994; Enkin 2003) were performed by rotating the blocks to palaeohorizontal (rotation axis = bedding strike, rotation angle = bedding dip). At 95 per cent confidence level the isolated magnetization is pretectonic (i.e. it was acquired prior to the tilting of the different blocks; Table 2). Thus, the actual attitude of the distinct basalt blocks is the consequence of the processes that affected the basalts after their emplacement. Despite the fact that our palaeomagnetic data set is spatially and temporally too limited to average out secular variation, the Tilt-corrected mean direction (Table 2) is indistinguishable of the reverse actual dipolar field direction at 95 per cent confidence level (Dec. 180°; Inc. 61.6°). These results show that the primary magnetization, acquired during cooling of the lavas, was preserved in all the sampled blocks.
Detailed topography of circular structure ‘G’ showing the location of the collected palaeomagnetic samples and the strike of each sampled basaltic block.
Equal angle lower hemisphere stereoplot showing (a) the in situ isolated ChRMs, (b) the tilt-corrected isolated ChRMs. Black symbols: positive inclination.
DISCUSSION
Terrestrial impact craters have a set of geophysical signatures which are the consequence of changes in the physical and lithological properties of rocks generated by the shock waves, high pressures, high temperatures and the cratering process. Geophysical characteristics of impact craters have been compiled and summarized by Pilkington & Grieve (1992), Hawke (2004), etc. The most noticeable gravity anomaly associated with terrestrial impact structures is generally a negative anomaly. These gravity lows are generally more or less circular and extend to, or slightly beyond, the rim of the structure (Pilkington & Grieve 1992). Their origin is related to density contrasts induced by fracturing and brecciation of the target rocks and low-density sedimentary infill of the topographic depression in uneroded structures. Magnetic anomalies associated with many impact structures are generally more complex than gravity anomalies, and reflect the broad range variation within the magnetic properties of rocks. The dominant effect in many structures is a magnetic low or subdued zone in the centre of the crater. Like with gravity, the magnetic signature does not reflect a one-to-one correspondence between the cross-sectional shape of the anomaly and the morphology of the impact structure. The causes of magnetic lows at impact structures are not as clear (Grieve & Pilkington 1996). The impact process undoubtedly results in a reduction in the magnetization intensity of the target material. Geoelectrical methods indicate resistivity lows coinciding with the extent of the potential field anomalies. The conductivity of rocks is heavily dependent on their water content. The degree of fragmentation determines the amount and distribution of fluids within the rock and hence its electrical properties (Pilkington & Grieve 1992). The large increase in conductivity of fluid-filled fractured material results in electrical methods being potentially useful in mapping the structure of impact craters.
Circular structure ‘8’
The magnetic, gravimetric and resistivity signatures of this circular structure concur with the ones generally expected for impact structures (Pilkington & Grieve 1992). Moreover, most Euler solutions are located below the circular structure rim. Curvature attributes, analytic signal, vertical gradients, etc. also show circular patterns matching the circular structure rim. Calculated pseudo-gravity from magnetic anomalies is very much alike the measured complete Bouguer anomaly. This similarity suggests that the magnetic and gravity anomalies would arise from sources with analogous shape and size, both of them being caused by the same body.
In the circular structure floor, negative complete Bouguer anomalies, low amplitude magnetic anomalies and very low electrical resistivities were detected. These results suggest that in the circular structure bottom, Pampa Sastre conglomerate would be at least partially absent and sediments without basalt boulders and blocks would have been deposited as infilling material. On the contrary, the circular structure rim exhibit high-amplitude localized magnetic anomalies, higher electrical resistivities and maximum complete Bouguer anomalies, which could be related to the anomalous accumulation of basalt boulders and blocks. Moreover, the distribution of apparent electrical conductivity obtained by means of electromagnetic methods would reveal the existence of a more resistive sedimentary infilling in the centre of the circular structure ‘8’, compared to the more conductive Pampa Sastre conglomerate that crops out along the rim. This resistivity distribution is exactly the contrary than the one portrayed by the VESs. This difference can be explained considering the different penetration depth of these two methods. The four VES provide information about resistivity of rocks deeper than 1 m, while electromagnetic data corresponds to near surface materials. The high electrical conductivity zone beneath the centre of the circular structure ‘8’ imaged by the resistivity inverse model to at least 20 m depth would be related to fracturing, which increases porosity and permeability allowing a current to be carried by ions in pore fluids. In this connection, Hawke (2004) noticed that high conductivity would be expected for allochthonous breccias and post-impact sedimentary fill deposits in impact related structures.
The anomalous accumulation of basalt blocks and boulders of up to 90 cm diameter at the rim of the circular structure ‘8’ and the fact that such boulders were not observed in the trench dug in the circular structure floor cannot be explained considering any known geomorphological process such as deflation or fluvial removal. Neither can it be attributed to the occurrence of hydrovolcanism or phreatomagmatism, as no volcanic activity in the study area took place after the deposition of Pampa Sastre conglomerates (Ardolino & Franchi 1996; Acevedo et al.2009). Evidence of magmatic activity has not been found anywhere in the study area for this period (Ardolino & Franchi 1996; Acevedo et al.2009). Volcanic eruptions commonly have associated lava flows or other eruptive features such as pyroclastic or phreatomagmatic deposits, whereas in these structures the previously mentioned characteristics are completely absent.
Our 3-D gravity model suggests that Pampa Sastre conglomerates could have been ejected and/or displaced during excavation by the impact of an extraterrestrial projectile. Taking into account the large diameter and weight for the majority of the basalt boulders and blocks conforming Pampa Sastre conglomerates, it could be reasonably expected that such materials would have been deposited not much further than the rim. The trench dug in the bottom of circular structure ‘8’ revealed that basaltic conglomerates of the Pampa Sastre Formation are absent. Furthermore, Orgeira et al. (2016) found microspherules of extraterrestrial origin in an anomalous concentration in the samples of the infilling sediments collected from this trench. Such microspherules exhibit different surface morphologies (Fig. 27) and are I-type (Orgeira et al.2016), composed of magnetite, wüstite and metallic iron.
Left-hand panel: interlocking equant FeO-rich crystals display a polygonal texture at the surface of spherule 8.21. Spherule diameter is 150 μm. Right-hand panel: enhanced view of the spherule 8.21 surface showing sharp boundaries between individual crystal grains. Field of view: 56 × 40 μm.
Circular structure ‘G’
The magnetic and palaeomagnetic results of this circular structure agree with the one generally expected for impact structures (Pilkington & Grieve 1992). Most Euler solutions are located beneath the circular structure rim. Curvature attributes, analytic signal, vertical gradients, etc., also show circular patterns coinciding with the circular structure rim. Calculated pseudo-gravity from magnetic anomalies resembles the measured complete Bouguer anomaly, suggesting that the magnetic and gravity anomalies would be generated by sources with similar shape and size. That is to say, the same body would cause the measured magnetic and gravimetric anomalies. The distribution of near surface apparent electrical conductivity obtained during the electromagnetic survey would indicate the presence of a more resistive sedimentary infilling in the centre of the circular structure ‘G’, compared to the more conductive basaltic rims.
Complete Bouguer anomaly is positive, contrary to what would be expected for an impact structure (Pilkington & Grieve 1992). Such positive complete Bouguer anomaly suggests the existence of an excess of mass beneath the circular structure's floor. On the other hand, the structure rims exhibit minimum complete Bouguer anomalies, suggesting the existence of a mass deficit. This mass deficit could be related to the anomalous accumulation of basalt blocks in the rims. Such build-up of angular blocks leaves many empty spaces between them, which are filled with air and sand, diminishing the density of the circular structure rim with respect to the circular structure floor.
Impact craters are clearly distinguishable from other landforms because they are almost perfectly circular, with an upraised rim. However, this does not mean that any of such circular-shaped structures is an impact crater. Volcanic calderas, maars (maar-diatreme), scoria cones, spatter cones, tuff rings, tuff cones and rootless cones could show similar features, however they present associated eruptive features such as pyroclastic or phreatomagmatic deposits (bombs, scoria, lapilli, ash-fall layers, tuffs and surges), hydrothermalism, etc., that are completely absent in circular structure ‘G’. Moreover, geophysical signatures of maars and phreatomagmatic structures are notably distinct to the ones obtained in our study. Kroner et al. (2006), Mrlina et al. (2009), Matthes et al. (2010) and Schmidt et al. (2013) investigated an eroded relic of a maar volcano (diatreme) in Ebersbrunn (Germany) with a diameter of approximately 2 km. Mrlina et al. (2009) reported a circular gravity minimum of –2.5 mGal, a circular magnetic maximum ranging between 300 and 1000 nT and a circular maximum electrical conductivity of around 15 mS m–1 in the centre of the structure. The maximum magnetic and conductivity anomalies are surrounded by minimum values ranging between −100 and 250 nT and 10 and 1 mS m–1. A 3-D model was developed by Mrlina et al. (2009) for Ebersbrunn diatreme, consisting of a symmetrical cone shaped body with 82° dipping walls with lower density and a higher magnetic susceptibility relative to the surrounding volcaniclastic rocks in the centre of the circular structure. Blaikie et al. (2014) carried out a geophysical study of maars in Australia. They detected simple geophysical signals consisting of long wavelength gravity and magnetic lows over the craters, with diameters ranging between 200 and 2000 m. Montesinos et al. (2003) published a gravity study in Terceira Island (Azores). On the Bouguer anomaly map an alignment of gravity minima of ∼–3 mGal was observed. Gravity inversion was carried out by Montesinos et al. (2003). The obtained model showed that the gravity minima correspond to negative density contrasts, which coincide with a volcanic alignment of scoria cones. These low densities are due to the scoria cones deposits: highly vesiculated material of basaltic nature associated with moderate explosive activity of strombolian type and perhaps breccias-type deposits on the conduit system, while the surroundings are basaltic lava flows with higher densities (Montesinos et al.2003). Risso et al. (2015) investigated eroded remnants of scoria cones occurring in the centre of ‘circular depressions’ in basaltic lava flows of the Llancanelo Volcanic Field, Argentina. Gravity, magnetic and palaeomagnetic studies were performed in Las Bombas volcano. The complete Bouguer anomaly presented a minimum value ∼ of –5 mGal, correlating perfectly with the centre of the circular structure and indicating the existence of a mass deficit. A dipolar magnetic anomaly, closely associated with the centre of the structure and the remnant scoria cone, was detected. It presents a maximum ∼ of 1700 nT and a minimum ∼ of –100 nT, being of normal polarity. Risso et al. (2015) obtained oriented hand samples of the basaltic flow surrounding the eroded scoria cone for its palaeomagnetic study. The isolated magnetizations showed positive and negative inclinations and a noticeable dispersion. Contrary to the results obtained for the circular structure ‘G’, the tilt corrected final mean direction calculated by Risso et al. (2015) presented a larger confidence interval than the uncorrected one.
Rootless cones correspond to pyroclastic accumulations emplaced through interaction of lava with surficial or near-surface water in lacustrine, glacial riverine outwash plains or marshy environments (Burr et al.2009). These structures were studied by Thorarinsson (1953) at the type locality in Iceland. In such locality many hundreds of pyroclastic structures formed where a lava flow invaded the shallow Lake Mývatn. Bruno et al. (2006) performed a geospatial analysis of cone alignment in the Raudholar locality in Iceland, revealing not strong linear alignment. These cones sit on the tops of lava surfaces, do not appear to have lava issuing from them and are grouped in clusters (Sheth et al.2004). Rootless cones are primarily composed of volcanic bombs, scoria and tephra interfingering with lacustrine muds, with internal stratification with reverse grading (Burr et al.2009). More than a single episode of explosivity is required to form a rootless cone, generating nested or overlapping cones (Thorarinsson 1953). It is important to remark that the circular structures indentified in Bajada del Diablo do not overlap. Moreover, no pyroclastic deposits or evidences of hydrovolcanism were found in circular structure ‘G’. Keszthelyi (2005) documented that for reasonable permeabilities, the water has to be less than a few centimetres under the surface and the substrate has to present a water saturation of up to 60 per cent to allow enough pressure to build and produce explosions. This last point is particularly important, and in the case of our study area water availability was very low during the emplacement of Quiñelaf Final Basic Lava Facies at 22 ± 1 Ma. Bellosi & González (2010) investigated the palaeopedological characteristics of the middle Cenozoic Sarmiento Formation at Gran Barranca, Chubut Province, approximately 350 km to the southwest of Bajada del Diablo. Particularly, they determined that the occurrence of a retreat of arboreal vegetation and concomitant expansion of shrubs was registered by the Colhue-Huapi Member of Sarmiento Formation during the Aquitanian (23.03–20.44 Ma). This trend was intensified in the late Aquitanian leading into the Burdigalian (20.44–15.97 Ma), with the predominance of xerophytic and halophytic shrubs and herbs and the development of palm savannas (Barreda & Bellosi 2003; Barreda & Palazzesi 2007; Iglesias et al.2011). Consequently, the lower section of the Colhue-Huapi Member (23.03–20.44 Ma) records an interval of subhumid seasonal conditions (MAP 650–900 mm), and a landscape dominated by wooded grasslands and riparian forests. On the other hand, the interpreted palaeoenvironment of the upper section (20.44–15.97 Ma) would be loessic shrubby grassland, with a semiarid to arid (MAP 650–400 mm) and probably cooler climate. Furthermore, Sheth et al. (2004) proposed that Jabalpur craters in the NE Deccan (India) are rootless cones. Srivastava et al. (2004) carried out magnetic and gravimetric surveys of these craters. These authors did not detect isolated circular or elliptical magnetic or gravimetric anomalies associated to them.
Geophysical anomalies coincident with the ones obtained in this study were reported for various impact craters. The Wolfe Creek Meteorite Crater is an 880 m diameter impact structure (O’Neill & Heine 2005), formed in Devonian sandstones of the eastern Canning Basin (Hawke 2003), with minor quartzite and scattered nickel laterite. O’Neill & Heine (2005) detected a large moat-like complete Bouguer anomaly low along the crater rim. This feature is similar to that observed for Meteor Crater (Regan & Hinze 1975), which is in part due to a seismically imaged subrim fracture zone, and the gravity anomaly over Wolfe Creek crater probably represents a similar situation (O’Neill & Heine 2005). The remaining rim anomaly is most likely due to the low density of the fractured sandstone, and an extensive fracture zone extending to some depth beneath the crater rim. O’Neill & Heine (2005) modelled the complete Bouguer anomaly considering the presence of lower density fractured sandstones beneath the rim, compared to the compacted, clay-rich sediments filling the centre of the crater. Pilkington & Grieve (1992) remarked that reduced densities due to fracturing of autochthonous rocks beneath the crater floor do not appear to contribute significantly to the gravity anomaly in simple craters. Negative gravity anomalies detected in simple craters are largely due to the low density sediment infilling the craters, particularly in those cases where lakes were developed inside the craters (i.e. Lonar crater, Lake Tüttensee, Bosumtwi crater, etc.).
The Pretoria Saltpan crater is located in the south of the Bushveld Complex (Brandt et al.1994). It is an almost circular 1 km diameter structure formed in 2 Ga old granites. Magnetic surveys (Brandt et al.1994) detected only local anomalies on the northern inner slopes of the rim, which would be caused by lamprophyric dykes. Brandt et al. (1994) highlighted that these observations excluded the possibility that any magnetic volcanic material or pipe existed near the crater centre beneath the lake sediments. A negative Bouguer anomaly centred on the crater and reaching a minimum of –3.2 mGal was detected (Brandt et al.1994). Brandt et al. (1994) stated that the low density lake muds that filled the crater floor were the primary contributors to the observed anomaly. From the study of drill cores, these authors determined the existence of a zone of fractured granite with interbedded granite sands beneath the lake sediments. However, such zones of granite breccias and sands contributed only slightly to the negative anomaly (Brandt et al.1994). The authors also investigated a 400 m diameter circular crater located 3 km to the southwest of the main crater. A slight positive Bouguer anomaly of 0.3 mGal was obtained in the centre of the structure. Brandt et al. (1994) proposed that this slight gravity high could be the result of a dense sedimentary fill in a fractured crater floor. Concurrently, a trench in the centre of this crater revealed the presence of a calcrete layer at 0.76 m depth.
Impact melt lithologies can occur as allochthonous coherent layered melt sheets in the impact crater fill or as dykes, veins and veins networks in the autochthonous crater basement. During crater formation, phase changes can occur in the target rocks due to shock waves. Shock deformation can produce phase transitions to high pressure polymorphs of minerals through solid state processes (Koeberl 1997). Several minerals form high pressure phases (Stöffler 1972), which have higher densities than the original ones. Stishovite (4.23 g cm–3) and coesite (2.93 g cm–3) forms from quartz (2.65 g cm–3), jadeite (3.24 g cm–3) forms from plagioclase (2.63–2.76 g cm–3), majorite (3.67 g cm–3) forms from pyroxene (3.20–3.52 g cm–3) and ringwoodite (3.90 g cm–3) forms from olivine (3.22–4.34 g cm–3; Koeberl 1997). Allwardt et al. (2004) studied high pressure aluminosilicate glasses and densification in basaltic magmas. These authors determined that densities of glasses quenched from simple basalt-like melts at 10 GPa pressures are 16 per cent higher than those quenched at ambient pressure. Lee et al. (2012) explored the structures of shock compressed silicate glass with a common basaltic composition. These authors determined that the studied sample increased its density by ∼30 per cent. Lee et al. (2012) microscopically constrained the magnitude of impact events involving natural basalts on Earth and planetary surfaces. Taking into account all the above mentioned, we suggest that the positive complete Bouguer anomaly detected at the bottom of circular structure ‘G’ could be explained by means of: (1) the anomalous accumulation of fractured blocks of basalts in the rims of the structure and/or, (2) the existence of calcrete and/or compacted, clay-rich sediments filling the centre of the structure in the absence of a lake and/or (3) the presence of allochthonous coherent layered melt sheets in the circular structure fill or as dykes, veins and veins networks in the autochthonous circular structure basement.
In an impact event the target rocks are affected in a characteristic way, particularly the structure of the impacted area can help to distinguish volcanic structures from impact craters (Koeberl 1997). In the outer walls of circular structure ‘G’ the existence of blocks of basaltic flows that strike parallel to the rim crest and dip outwards was registered. The same structure has been identified and studied by Kumar (2005) and Louzada et al. (2008) in Lonar crater. Lonar crater is a simple, bowl-shaped, near-circular impact crater in the 65 Ma Deccan Volcanic Province in India (Kumar 2005). Five individual flows of massive, vesicular and amygdaloidal basalts are exposed on the inner wall of Lonar crater (Kumar 2005). Louzada et al. (2008) carried out a palaeomagnetic study in Lonar crater, obtaining results very similar to the ones here presented. These authors sampled the rims of Lonar crater in upturned beds dipping out of the crater, in folded beds below the fold hinge dipping out of the crater and in overturned rim blocks. Louzada et al. (2008) isolated a high temperature component and also applied the fold test (McElhinny 1964), which was passed at 100 per cent unfolding. These authors interpreted the high temperature component, which was similar to the mean Deccan direction, as a primary magnetization acquired during cooling of the Deccan lavas approximately 65 Ma.
Rajasekhar & Mishra (2005) documented the existence of almost circular gravity and magnetic anomalies over Lonar lake of almost –2.25 mGal and 550 nT, respectively. Circular gravity and magnetic anomalies are also observed in some cases of volcanic plugs, but their amplitudes and signs differ considerably ranging from a few mGal (<10 mGal) negative gravity anomalies reported for impact craters to 60–80 mGal anomalies over volcanic plugs, especially those reported from the Deccan trap (Rajasekhar & Mishra 2005). Magnetic anomalies observed over volcanic plugs related to Deccan trap activity are also of higher order (800–1000 nT) compared to the ones recorded in the Lonar lake (Rajasekhar & Mishra 2005). The higher amplitude of gravity and magnetic anomalies due to volcanic plugs compared to those obtained as a result of cratering processes, can be attributed to the greater depth extent of volcanic plugs and the higher contrasts in the physical parameters (Rajasekhar & Mishra 2005). While impact craters are circular structures without deep roots, in volcanic structures the disturbances emerge from great depth (Koeberl 1997).
Finally, and to conclude the discussion of our results, we would like to remark that no explosive deposit was found that could establish a hydro or phreatomagmatic genesis for the circular structure ‘G’. This field observation, the detected geophysical anomalies and also the fissural and basaltic nature of the volcanism in the studied area suggest an impact origin.
CONCLUSIONS
The 3-D forward gravity model developed for circular structure ‘8’ suggests that Pampa Sastre conglomerates could have been ejected and/or displaced during excavation by the impact of an extraterrestrial projectile.
The 3-D forward gravity model developed for circular structure ‘G’ suggests the possible existence of: (1) anomalous accumulation of fractured blocks of basalts in the rims of the structure and/or, (2) calcrete and/or compacted, clay-rich sediments filling the centre of the structure in the absence of a lake and/or, (3) allochthonous coherent layered melt sheets in the circular structure fill or as dykes, veins and veins networks in the autochthonous circular structure basement.
The palaeomagnetic results obtained in circular structure ‘G’ show that the primary magnetization, acquired during cooling of the lavas, was preserved in all the sampled blocks. Thus, the actual attitude of the distinct rim basalt blocks is a consequence of a process that affected the basalts after their emplacement, consistently with an impact.
The detected magnetic, gravimetric and resistivity anomalies, the extraterrestrial spherules found in an anomalous concentration in the sedimentary infill of the circular structure ‘8’, the lack of evidences of the occurrence of hydrovolcanism or phreatomagmatism and the geomorphologic characteristics of the studied circular structures together with the palaeoclimatic conditions would support an impact origin.
However, the confirmation of an impact origin through the findings of shock metamorphism evidences and/or the recovery of meteorites was not yet accomplished.
Bajada del Diablo should be envisaged as a focus of further research, which could provide novel information about impact events, associated processes and their evidences.
We are very grateful to the Editor Dr Mark Everett and to the reviewer Dr Stephanie Werner for their thorough corrections and comments that helped us to greatly improve our manuscript. Diurnal variation of the geomagnetic field was generously provided by Prof Julio César Gianibelli, Nicolás Quaglino and Sebastián Pelliciuoli from Trelew Magnetic Observatory – Intermagnet, UNLP. Financial support was granted by the Argentine National Agency of Scientific and Technological Promotion (ANPCyT; PICT 2013-1950), the National Geographic/Waitt, the CONICET (PIP0573 and PIP 2011/13 00416) and Buenos Aires University (UBACyT 2014–2017 20020130100146BA). Authors thank the kind and friendly relation with landowner (Mr Valle), manager (Mr Heiken) and foreman (Mr Ibáñez) of Estancia La Primera, where the team spent their days during field work in this very remote area of Patagonia.
REFERENCES
SUPPORTING INFORMATION
Additional Supporting Information may be found in the online version of this paper:
Figure 1. Circular structure ‘8’ magnetic anomaly reduced to the pole. Black lines: topography contour lines.
Figure 2. Dip angle curvature attribute of circular structure ‘8’ magnetic anomaly. Black lines: topography contour lines.
Figure 3. Most negative curvature attribute of circular structure ‘8’ magnetic anomaly. Black lines: topography contour lines.
Figure 4. Analytic signal of circular structure ‘8’ magnetic anomaly. Black lines: topography contour lines.
Figure 5. First vertical derivative of circular structure ‘8’ magnetic anomaly. Black lines: topography contour lines.
Figure 6. Dip angle curvature attribute of circular structure ‘8’ complete Bouguer anomaly. Black lines: topography contour lines.
Figure 7. Tilt derivative of circular structure ‘8’ complete Bouguer anomaly. Black lines: topography contour lines.
Figure 8. First vertical derivative of circular structure ‘8’ complete Bouguer anomaly. Black lines: topography contour lines.
Figure 9. Circular structure ‘G’ magnetic anomaly reduced to the pole. Black lines: topography contour lines.
Figure 10. Dip angle curvature attribute of circular structure ‘G’ magnetic anomaly. Black lines: topography contour lines.
Figure 11. Most negative curvature attribute of circular structure ‘G’ magnetic anomaly. Black lines: topography contour lines.
Figure 12. Analytic signal of circular structure ‘G’ magnetic anomaly. Black lines: topography contour lines.
Figure 13. First vertical derivative of circular structure ‘G’ magnetic anomaly. Black lines: topography contour lines.
Figure 14. Dip angle curvature attribute of circular structure ‘G’ complete Bouguer anomaly. Black lines: topography contour lines.
Figure 15. Tilt derivative of circular structure ‘G’ complete Bouguer anomaly. Black lines: topography contour lines.
Figure 16. First vertical derivative of circular structure ‘G’ complete Bouguer anomaly. Black lines: topography contour lines.
Please note: Oxford University Press is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the paper.
