Tsunami Deposits on a Paleoproterozoic Unconformity? the 2.2 Ga Yerrida Marine Transgression on the Northern Margin of the Yilgarn Craton, Western Australia

Home , Marine transgression, Stack (geology)

Journal of Marine Science and Engineering

Article Tsunami Deposits on a Paleoproterozoic Unconformity? The 2.2 Ga Yerrida Marine Transgression on the Northern Margin of the Yilgarn Craton, Western Australia

Desmond F. Lascelles 1,* and Ryan J. Lowe 2

1 Des Lascelles Consulting Geologist, 15 Bardﬁeld Way, Gosnells, WA 6110, Australia 2 School of Earth and Environment and UWA Oceans Institute, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia; [email protected] * Correspondence: [email protected]

Abstract: Large blocks and boulders of banded iron formations and massive hematite up to 40 × 27 × 6 m3 and in excess of 10,000 metric tonnes were detached from an outcrop of the Wilgie Mia Formation during the ca 2.20 Ga marine transgression at the base of the Paleoproterozoic Windplain Group and deposited in a broad band on the wave-cut surface 900 to 1200 m to the east. At the same time, sand and shingle were scoured from the sea floor, leaving remnants only on the western side of the Wilgie Mia Formation and on the eastern sides of the boulders. Evidence suggesting that the blocks were detached and transported and the sea floor scoured by a tsunami bore with a height of at least 40 m is provided by the following: (1) the deposition of the blocks indicates transportation by a unidi- rectional sub-horizontal force, whereas the smaller boulders are randomly oriented; (2) 900–1200 m separates the banded iron formation (BIF) outcrop and the blocks (3) there is an absence of the basal conglomerate between the blocks; (4) the blocks and boulders rest directly on the wave-cut surface of deeply weathered amphibolites; (5) the blocks and boulders are surrounded and overlain by Citation: Lascelles, D.F.; Lowe, R.J. fine-grained sandstone of the Windplain Group. Tsunami Deposits on a Paleoproterozoic Unconformity? The 2.2 Ga Yerrida Marine Transgression on the Northern Keywords: Paleoproterozoic; tsunami boulder deposit; marine transgression; Yilgarn Craton; Yerrida Basin; Margin of the Yilgarn Craton, Western Windplain Group Australia. J. Mar. Sci. Eng. 2021, 9, 213. https://doi.org/10.3390/ jmse9020213 1. Introduction Academic Editor: Alain Trentesaux Uplift and erosion of the northern margin of the Yilgarn Craton over between 2.7 Received: 8 January 2021 and 2.2 Ga resulted in a low relief landscape, similar to the present, underlain by granitic Accepted: 15 February 2021 rocks and minor greenstone belts that were deeply weathered to more than 200 m in places. Published: 18 February 2021 Subsidence of the craton to an elevation probably less than 10 m above the contemporary sea level was followed by a marine transgression at ca 2.20 Ga that removed the soil cover Publisher’s Note: MDPI stays neutral and eroded a wave-cut surface on the deep saprolite. with regard to jurisdictional claims in Large blocks and boulders (up to 40 × 27 × 6 m3) of banded iron formation (BIF) published maps and institutional affil- and massive hematite are exposed in a 300 m wide band stretching nearly three kilome- iations. tres within the Joiner’s Find greenstone belt, northern Yilgarn Craton, Western Australia (Figure1). The blocks rest directly on a marine erosion surface formed during the marine transgression at 2.20 Ga [1] on ultramafic rocks of the Archean Meekatharra Formation [2,3] prior to deposition of the Paleoproterozoic Windplain Group. The nearest outcrop of BIF Copyright: © 2021 by the authors. lies in a north–south trending west-facing sub-vertical outcrop 900 m to the west of the Licensee MDPI, Basel, Switzerland. blocks. A polymict conglomerate consisting of <15 cm diameter BIF, hematite, and minor This article is an open access article goethitized mafic pebbles in a quartz sand matrix lies to the west of the Archean Wilgie distributed under the terms and Mia Formation BIF outcrop. It has been scoured from the surface of the BIF and the uncon- conditions of the Creative Commons formity to the east but is also present on the eastern side of the largest blocks. The blocks Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ and boulders are surrounded and overlain by epicontinental marine sedimentary rocks of 4.0/). the Windplain Group.

J. Mar. Sci. Eng. 2021, 9, 213. https://doi.org/10.3390/jmse9020213 https://www.mdpi.com/journal/jmse J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 3 of 18

geophysics, by Golden West Resources Ltd., as part of their ore deposit reserve measurements. The final densities used in their calculations were 2.8 for weathered BIF and 4.2 for massive hematite.

3. Results The Joyner’s Find greenstone belt (JFGB) is an elongated outcrop of Archean supracrustal rocks, 42.5 km long with 10 km maximum width, extending south from the northern margin of the Yilgarn Craton, 30 km to the west of the township of Wiluna, 900 km northeast of Perth in Western Australia (Figure 1). The JFGB contains a ~2.82–2.7 Ga [10,11] sub-vertical west-facing sequence of mainly mafic to ultramafic amphibolite interbedded with two sequences of banded iron formation (BIF) passing up into siliciclastic meta-sedimentary rocks (Figure 2). The stratigraphy of the JFGB was interpreted from detailed outcrop mapping with GPS locations plotted at 1:2000 scale with units named initially after the magnetic highs—A–D from east to west. The similarity with the stratigraphy described by Van Kranendonk and Ivanic [4] for the north-eastern Murchison Do- main suggested those stratigraphic names to be adopted informally for the JFGB [2,3]. The greenstone belt is surrounded by monzogranite on three sides and is unconformably overlain by polymict conglomerate and sandstone of the Paleoproterozoic Windplain Group in the north (Table 1). Numerous schistose northerly trending shear zones, 0.10–5 m wide, occur throughout the JFGB and are typically marked by bands of chlorite or mica schist. The mica schist bands in the surrounding monzogranite, together with granitized sediments, were previously interpreted as granite gneiss [10,11]. The supracrustal rocks form part of the sub-vertical west-facing limb of a multi-kilometre scale fold preserved as a roof pendant in a large ~2.6 Ga monzogranite batholith with amphibolite facies thermal meta- J. Mar. Sci. Eng. 2021, 9, 213 morphism increasing to granulite facies near the monzogranite [3]. The base of the supra-2 of 17 crustal rocks is not seen, as the contacts of the oldest formation (Murrouli Basalt) are invariably sheared or invaded anatectically by monzogranite.

117o 120 o 123o Y Yerrida Basin er rid Ea a ra Ba hee Wiluna sin dy 26o JFGB Ba Joyner’s Find • WILUNA sin Greenstone Belt M u rch iso n Ida Fault Do main E astern Yilgarn Craton Youanmi Terrane G o ld field s Terran e o 29 Yilgarn Craton

S o u th ern Cro ss Domain PERTH 32o S o u th w est P ro vin ce N

Figure 1. SimplifiedSimplified geological map of the Yilgarn Craton showing the location of the Joyner Joyner’s’s Find greenstone greenstone belt (JFGB) and the Yerrida andand EaraheedyEaraheedy Basins, Western Australia.Australia. (Lascelles, 2014.2014. ModifiedModified after Geological Survey of Western Australia (GSWA)(GSWA) openopen accessaccess files).files).

The BIF units of the Wilgie Mia Formation form a north–south trending flat-topped ridge (C Ridge) on which the wave-cut surfaces are well preserved. The central part of the Yaloginda Formation also forms a narrow flat-topped ridge (B Ridge), but the wave-cut surfaces are less clearly defined due to the thinness of the BIF/massive hematite units. The upper and lower parts of the Yaloginda Formation consist mainly of mafic amphibolite and generally underlie slopes on either side of the ridge interrupted by resistant outcrops of thin BIF and granular iron formation (GIF) units. Parts of the Murrouli Basalt and other amphibolite outcrops are strongly goethitized and form resistant outcrops above the surrounding land surface. All sloping surfaces in the Joyner’s Find greenstone belt (JFGB) consist of pediments eroded on the saprolite and are overlain by a <30 cm thick cover of colluvium with numerous exposures of the underlying saprolite. In the absence of any other reasonable model for the transportation and deposition of the large blocks, it is hypothesized that an exceptionally large tsunami generated by an oceanic asteroid impact was responsible.

2. Methods and Materials The stratigraphy of the JFGB (Table1) was interpreted from detailed outcrop mapping with GPS locations of outcrops plotted at 1:2000 scale with units named initially after the magnetic highs; A–D from east to west. The similarity with the stratigraphy described by [4] for the north-eastern Murchison Domain suggested these stratigraphic names be adopted informally for the JFGB [2,3]. The outlines of the exposed transported blocks were initially plotted by GPS and later measured using a 50 metre tape (Table2). The longest surface exposure was considered the A axis and the greatest width the C axis. B and D axes represent the shorter sides of trapezoidal blocks. J. Mar. Sci. Eng. 2021, 9, 213 3 of 17

Table 1. Stratigraphy and tectonic history of the Joyner’s Find greenstone belt [2,3]. (Non-depositional events in italics).

Era Period Group Formation Rock Type Age Quaternary Alluvium, colluvium, aeolian sand Cenozoic Ferricrete, CID, DID, silcrete Deep weathering (goethite) Uplift and erosion of Tooloo Group and exhumation of JFG and Yerrida Group To Present Sandstone, shale, stromatolitic (Earaheedy Basin) Tooloo Gp Yelma Fm 1900–1800 Ma carbonates Marine transgression Deep weathering (Goethite) Paleo-proterozoic Partial erosion of Yerrida Group exhuming southern JFGB Minor folding in foreland of Glenburgh Orogeny tilting of JFGB to north (Yerrida Basin) Windplain Gp Judarina Fm Sandstone (± conglomerate) 2015–1950 Ma Marine transgression and tsunami ~2015 Ma Extended deep weathering (hematite) Major uplift and erosion ~2500–2015 Ma D folding and faulting 4 ~2600–2500 Ma D3 major N-S shearing Monzogranite intrusion and thermal metamorphism 2637–2602 Ma D2 folding and faulting ~2700–2650 Ma Wattagee Fm Not seen 2725–2700 Ma Archean Glen Gp Medium to coarse sediments and Ryansville Fm mafic volcanic rocks (Murchison Unit C thick BIF and mafic to Supergroup) Wilgie Mia Fm 2800–2725 Ma ultramafic volcanic rocks Polelle Gp Greensleeves Fm Not seen Ultramafic, mafic and minor felsic Meekatharra Fm volcanic rocks Unit B thin BIF with mafic and Yaloginda Fm 2820–2800 Ma Norie Gp ultramafic volcanic rocks Murrouli Basalt Mafic volcanic rocks Unknown

Table 2. Measured boulders.

No. Easting ** Northing ** Long Axis Short Axis Rock Type a b c d 1 793,324 7,046,045 10.00 5.00 BIF 2 793,331 7,046,126 10.43 8.25 3.50 BIF 3 793,334 7,048,312 12.30 4.00 BIF + He 4 793,346 7,046,057 12.90 9.17 4.83 3.70 BIF 5 793,352 7,048,276 20.00 18.80 17.30 16.00 He 6 793,357 7,046,125 11.00 2.50 BIF 7 793,360 7,048,341 15.50 11.00 BIF 8 793,363 7,048,274 25.00 12.00 BIF 9 793,367 7,048,339 1.40 1.30 He 10 793,374 7,048,342 7.40 6.50 BIF 11 793,388 7,048,341 9.80 4.30 BIF 12 793,996 7,048,794 13.00 3.00 BIF 13 793,412 7,048,290 34.70 27.30 17.00 7.00 BIF + He * 14 793,434 7,048,125 ~80 ~50 Boulders * 15 793,435 7,048,283 ~70 ~50 Boulders 16 793,835 7,047,606 1.50 0.50 BIF + He 17 793,839 7,047,603 5.50 2.20 BIF + He 18 793,853 7,047,606 13.90 4.40 He 19 793,950 7,048,855 40.00 26.80 BIF 20 793,960 7,048,909 23.90 13.10 BIF 21 793,979 7,048,825 24.50 8.40 BIF 22 793,983 7,048,836 18.40 5.20 BIF 23 793,987 7,048,869 15.60 8.80 BIF 24 793,988 7,048,743 17.40 9.00 BIF 25 793,989 7,048,800 26.50 25.00 BIF 26 793,994 7,048,775 17.50 10.20 BIF 27 794,001 7,048,776 19.00 7.00 BIF J. Mar. Sci. Eng. 2021, 9, 213 4 of 17

Table 2. Cont.

No. Easting ** Northing ** Long Axis Short Axis Rock Type 28 794,001 7,048,785 3.50 1.60 BIF 29 794,007 7,048,771 11.20 ? BIF 30 794,019 7,048,783 13.00 6.50 BIF 31 794,032 7,048,771 15.80 5.50 BIF 32 794,034 7,048,785 16.00 6.80 BIF 33 794,050 7,048,312 0.60 0.60 BIF 34 794,058 7,048,316 1.40 0.90 BIF 35 794,069 7,048,316 2.00 1.00 BIF * Large groups of mixed boulders <10 m diameter in close proximity. ** GDA94 zone 50J coordinates. He = massive hematite.

We based calculations of tsunami height and velocity on the modified equations proposed by [5] for predicting the minimum flow velocity (and hence minimum wave height) required to initiate boulder transport under high-energy tsunami events, which builds on the theoretical framework originally developed by [6,7]. Given the largest boulders constrain a lower limit on the size of the tsunami [8,9], we specifically investigated the minimum velocity Umin required to move the slabs of length l = 40 m, width w = 27 m, and thickness t = 4 m over a distance of 900 m (i.e., corresponding to boulder #33 in Table2). −3 We assume a BIF boulder density ρs = 2800 kg m , (average measured value of numerous −3 samples of silicified weathered BIF from drill core), a seawater density ρw = 1020 kg m , and standard values for boulder drag and lift coefficients (Cd = 1.95 and Cl = 0.18, respectively) [5]. Given that the original state of the boulder sources is unknown, we consider two scenarios: (1) the boulders were initially fully detached, or (2) they were joint-bounded in situ. The minimum velocities for each are thus r ( − ) U = 2 ρs/ρw 1 gt = −1 min C (t2/l2)+C 26 m s and d l (1) q ( − ) U = 2 ρs/ρw 1 gt = 28 m s−1 min Cl −1 for the exposed and joint-bounded scenarios, respectively,such that in both cases Umin~30 m s . Numerous determinations of the density of the strongly weathered and silicified BIF and the massive hematite were carried out on diamond drill core, including downhole geophysics, by Golden West Resources Ltd., as part of their ore deposit reserve measurements. The final densities used in their calculations were 2.8 for weathered BIF and 4.2 for massive hematite.

3. Results The Joyner’s Find greenstone belt (JFGB) is an elongated outcrop of Archean supracrustal rocks, 42.5 km long with 10 km maximum width, extending south from the northern margin of the Yilgarn Craton, 30 km to the west of the township of Wiluna, 900 km northeast of Perth in Western Australia (Figure1). The JFGB contains a ~2.82–2.7 Ga [ 10,11] sub-vertical west-facing sequence of mainly maﬁc to ultramaﬁc amphibolite interbedded with two sequences of banded iron formation (BIF) passing up into siliciclastic meta-sedimentary rocks (Figure2). The stratigraphy of the JFGB was interpreted from detailed outcrop mapping with GPS locations plotted at 1:2000 scale with units named initially after the magnetic highs—A–D from east to west. The similarity with the stratigraphy described by Van Kranendonk and Ivanic [4] for the north-eastern Murchison Domain suggested those stratigraphic names to be adopted informally for the JFGB [2,3]. The greenstone belt is surrounded by monzogranite on three sides and is unconformably overlain by polymict conglomerate and sandstone of the Paleoproterozoic Windplain Group in the north (Table1). Numerous schistose northerly trending shear zones, 0.10–5 m wide, occur throughout the JFGB and are typically marked by bands of chlorite or mica schist. The mica J. Mar. Sci. Eng. 2021, 9, 213 5 of 17

schist bands in the surrounding monzogranite, together with granitized sediments, were previously interpreted as granite gneiss [10,11]. The supracrustal rocks form part of the sub-vertical west-facing limb of a multi-kilometre scale fold preserved as a roof pendant in a large ~2.6 Ga monzogranite batholith with amphibolite facies thermal metamorphism increasing to granulite facies near the monzogranite [3]. The base of the supracrustal rocks J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 4 of 18 is not seen, as the contacts of the oldest formation (Murrouli Basalt) are invariably sheared or invaded anatectically by monzogranite.

FigureFigure 2. 2.Block Block diagram diagram of of the the Joyner’s Joyner’s Find Find Greenstone Greenstone Belt. Belt. Stratigraphic Stratigraphic names names are are proposed proposed by by the authors. the authors.

TheThe BIF BIF sequences sequence ares are tightly tightly folded folded about about sub-vertical sub-vertical northerly northerly trending trending axial planes axial withplanes very with numerous very numerous crosscutting crosscutting normal faults;normal however, faults; however, folding andfolding faulting and faulting are rarely are visiblerarely in visible the amphibolite in the amphibolite sequences. sequences. The lower The BIF lower sequence BIF sequence(Yaloginda (Yaloginda Formation, For- B Ridge,mation, (Figure B Ridge,3) consists (Figure of 3) several consists discrete of several thin-bedded discrete thin BIF- andbedded granular BIF and iron granular formation iron (GIF)formation units interbedded (GIF) units interbedded with mafic amphibolite, with mafic amphibolite,whereas the upper whereas unit the (Wilgie upper Mia unit Formation)(Wilgie Mia outcrops Formation) as a~400 outcrops m wide as sequencea ~400 m ofwide <200 sequence m wide compositeof <200 m BIFwide bands composite with interbeddedBIF bands amphibolitewith interbe anddded chlorite amphibolite schist (C. and Ridge, chlorite Figure schist3). The (C. Wilgie Ridge, Mia Figure Formation 3). The overliesWilgie amphiboliteMia Formation with overlies numerous amphibolite bands of chloritewith numerous schist and bands minor of felsic chlorite lenses schist within and theminor Meekatharra felsic lenses Formation. within the Siliciclastic Meekatharra sedimentary Formation. rocks Siliciclastic and mafic sedimentary lavas of the overlying rocks and Glennmafic Grouplavas of are the poorly overlying exposed Glenn and Group were notare examinedpoorly exposed in detail. and were not examined in detail.The Archean rocks of the northern JFGB were buried beneath the unmetamorphosed and virtuallyThe Archean undeformed rocks of epicontinental the northern JFGB sedimentary were buried Proterozoic beneath rocks the unmetamorphosed of the Yerrida and Earaheedyand virtually Basins undeformed until exhumed epicontinental by Mid-Tertiary sedimentary to Present Proterozoic erosion. By rocks 2.20 Ga,of the the Yerrida north- ernand margin Earaheedy of the Bas Yilgarnins until Craton exhumed was eroded by Mid to-Tertiary a deeply to weathered Present erosion. arid landscape By 2.20 Ga[2,12, the] similarnorthern to margin the present of the one, Yilgarn with Craton scattered was greenstone eroded to a belts deeply separated weathered by largearid landscape areas of granitic[2,12] similar rocks. to The the deep present weathering one, with is scattered clearly indicated greenstone by belts the total separated absence by of large granitic areas pebbleof grani conglomeratetic rocks. The at deep the base weathering of the Windplain is clearly Groupindicated although by the restingtotal absence unconformably of granitic onpebble granite conglomerate over most of at the the area. base Weathered of the Windplain granitic Group rocks consistalthough of clayresting pseudomorphs unconformably of feldspar,on granite micas, over and most amphiboles of the area. with Weathered unaltered granitic quartz grains rocks andconsist are completelyof clay pseudomorp disbursedhs byof marine feldspar, erosion. micas Not, and even amphiboles core stones with were unaltered present quartz to produce grains pebbles and are indicating completely a great dis- depthbursed of weatheringby marine erosion. well below Not the even reach core of stones wave action.were present Similarly, to produce no amphibolite pebbles pebbles indicat- occuring a withingreat depth the polymict of weathering conglomerate well below associated the reach with of the wave JFGB, action. which Similarly, contains n onlyo am- stronglyphibolite goethitized pebbles occur amphibolite, within thedeeply polymict weathered conglomerate silicified BIF, associated and massive with hematite the JFGB, pebbles.which contains Further evidence only strongly of multiple goethitized cycles amphibolite, of weathering deeply shown weathered by the ore silicified deposits BIF at , JFGBand (Figuremassive4 )hematite is detailed pebbles. in [2,3 ].Further evidence of multiple cycles of weathering shown by the ore deposits at JFGB (Figure 4) is detailed in [2,3]. Part of the northern margin was depressed, forming a sag basin [13], probably associated with rifting along the northern margin of the Yilgarn Craton, onto which shallow- marine sedimentary facies of the Paleoproterozoic Windplain Group (Table 1) were deposited [13]. The marine transgression eroded the smooth wave-cut Yerrida Surface on the deeply weathered Archean rocks [2]. A second ca 1.95 Ga marine transgression, after the Capricorn Orogeny tilted the craton margin and overlying Windplain Group 12º to the north, produced a new wave-cut erosion surface (labelled the Earaheedy Surface) on

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 6 of 18 J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 6 of 18

Group consists of locally derived silicified BIF and hematite pebbles with minor goe- Group consists of locally derived silicified BIF and hematite pebbles with minor goe- J. Mar. Sci. Eng. 2021, 9, 213 thitized mafic pebbles in a quartz sand matrix and only occurs in the vicinity6 of 17 of the JFGB; thitized maficelsewhere pebbles in, the a quartz Finlayson sand Member matrix andsandstone only occurs rests directlyin the vicinity on the ofunconformity the JFGB; with only elsewhere, thethe Finlayson rare occurrence Member of sandstone vein quartz rests clasts directly at the on base the (Figure unconformity 2). with only the rare occurrence of vein quartz clasts at the base (Figure 2). 792,500 mE 795,000mE 792,500 mE LEGEND Paleoproterozoic 795,000mE N A LEGEND B 5f 80 Finlayson Member m 26 ˚38' Paleoproterozoic

A 0 B 5f 80 Finlayson Member Polymict conglomerate m 26 ˚38' 0

0 Transported boulders

0 0 Polymict conglomerate

, 0 Archean

, Aolian sand Transported boulders

5 Granitoid

0 cover Archean , Aolian sand Aolian sand 7 Granitoid Glen Group cover cover Wilgie Mia Formation Aolian sand Glen Group Meekatharra Formation cover Wilgie Mia Formation Yaloginda Formation 15 80 Meekatharra6d Formation Murrouli Basalt 15 80 6d Yaloginda Formation Murrouli Basalt 6b B Ridge 6b B Ridge North 5b 15

0 1 N N North 5b 15

m m

Kilometres

0 0 0 6c 1

N N

0 0

m m

, Kilometres , 0 6c26 40' 0 7 ˚ 7

0 795,0000 mE

0 0

, ,

26 40' , , 7,045,000mN 7 ˚ 7

795,000 mE 7

0 0

, 80 , 7,045,000mN

7 80 Archean 5a 6a 5d Archean 5a 6a 5d 5c 6f 5c 6f 5e 15 5e 15 80 8a6e C Ridge 80 8a6e C Ridge

, ,

0 5 26˚42'

4 4

0 Weathered

0 0

, ,

5 26 42' ˚ 7

7 4 80 4 Not mapped Weathered Granite

0 0

7 80 Not mapped 7 Granite B Ridge B Ridge 7,040,000mN 80 7,040,000mN 80 Legend LegendFinlayson Member 80

N N 80 Finlayson, Member Sandstone m Yerrida Surface

80 m

0 N

80 , Sandstone 0 m Yerrida Surface

5 m

2 0

0 4 2

26 42' 0

˚ , 4

5 Conglomerate

, 0 , Large boulder Field

7 , 4 2

26˚42' 7 4 0 119 ˚56' 119 58' Conglomerate , ˚

0 Large boulder Field 7 , 119 ˚56' 792,500 mE 119 ˚795,58' 0007 mE 792,500 mE 795,000 mE Figure 3. (A)Figure, Interpreted 3. (A), Interpretedmap of the map northern of the northernpart of the part Joyner’s of the Joyner’s Find greenstone Find greenstone belt. belt. 5a-6f5a-6f = approximate = approximate photo loca- Figure 3. (A), Interpreted map of the northern part of the Joyner’s Find greenstone belt. 5a-6f = approximate photo locations. (B). Contourphoto locations.plan showing (B). the Contour wave-cut plan erosion showing surface the and wave-cut overall erosion extent of surface conglomerate and overall and extentboulder of deposition tions. (B). Contour plan showing the wave-cut erosion surface and overall extent of conglomerate and boulder deposition (By permissionconglomerate of GWR). and boulder deposition (By permission of GWR). (By permission of GWR). C5 C4 C3 Hematite goethite 3 C2 C1 C5 C4 C3 Hematite goethite (hydrated3 zone)C2 C1 YS (hydratedES zone) F ES YS ES F ES Hematite 1 Hematite 2 ematite 1 Hematite 2 H ethite 1 Hematite goethite 2 Hematite go ethite 1 Hematite goethite 2 Hematite go

N ~15 km S N ~15 km S Proterozoic erosion surfaces Archean Saprolite (oxidized) Proterozoic erosion2.20 surfaces Ga Water table Archean SaproliteArchean (oxidized) Saprolite (unoxidized below water table) 2.20 Ga Water table2.20 Ga Weathering front Archean Saprolite Paleoproterozoic(unoxidized below saprolite water table) 2.20 Ga Weathering1.8 frontGa Weathering front Paleoproterozoic saproliteCenozoic saprolite 1.8 Ga WeatheringPresent front weathering front Cenozoic saproliteSurface hydrated zone Present weathering front Surface hydratedFinlayson zone Member Finlayson Member Figure 4. Sketch longitudinal section of JFGB showing the relationship of wave-cut erosion sur- FigureFigure 4. 4.Sketch Sketchfaces longitudinal longitudinal and paleo section -sectionregolith of of JFGBextent. JFGB showing showingSaprolite the the refers relationship relationship to the mineralogy of of wave-cut wave- cutof erosion both erosion ore surfaces sur-and banded iron faces and paleo-regolith extent. Saprolite refers to the mineralogy of both ore and banded iron and paleo-regolith extent. Saprolite refers to the mineralogy of both ore and banded iron formation (BIF) parent materials. C5-C1—ore body numbers; YS = Yerrida Surface; ES = Earaheedy Surface. Not to scale.

J. Mar. Sci. Eng. 2021, 9, 213 7 of 17

Part of the northern margin was depressed, forming a sag basin [13], probably associated with rifting along the northern margin of the Yilgarn Craton, onto which shallow- marine sedimentary facies of the Paleoproterozoic Windplain Group (Table1) were deposited [13]. The marine transgression eroded the smooth wave-cut Yerrida Surface on the deeply weathered Archean rocks [2]. A second ca 1.95 Ga marine transgression, after the Capricorn Orogeny tilted the craton margin and overlying Windplain Group 12º to the north, produced a new wave-cut erosion surface (labelled the Earaheedy Surface) on a renewed deep weathering proﬁle. An unknown thickness of sediments (Tooloo Group) was then deposited, overstepping the partially eroded Windplain group onto the Archean rocks (Figure4). Both the overlying Windplain and Tooloo Group are unmetamorphosed J. Mar. Sci. Eng. 2021, 9, x FOR PEERand REVIEW virtually undeformed. Mid-Cenozoic to Recent erosion has exhumed the Archean7 of 18 rocks of the JFGB, which have been undisturbed for nearly two billion years [2] with the 2.20 Ga and 1.95 Ga erosion surfaces well preserved on the siliciﬁed surfaces of the deeply weathered BIF units (Figures3B and5). Erosion of the exhumed BIF surfaces is typically formation (BIF) parent materials. C5-C1—ore body numbers; YS = Yerrida Surface; ES = Earaheedy restricted to minor gullies with some crumbling along the edges. Surface. Not to scale.

← ← Sinuous grooves ↓

Shallow pothole

FigureFigure 5. 5.C C RidgeRidge outcropsoutcrops north of latitude 26 26°◦3377′33.8033.8”′′ S S.. (A (A).). BIF BIF stack stack remnant remnant with with wave wave cut cut surfacesurface dipping dipping 12 12◦°N N inin thethe foreground.foreground. AA smallsmall pocketpocket of basal conglomerateconglomerate lies on the surface ~20 m behind the camera, Yerrida Surface on Wilgie Mia Formation (looking east). (B). Wave-cut ~20 m behind the camera, Yerrida Surface on Wilgie Mia Formation (looking east). (B). Wave-cut surface on massive hematite. (C,D). Wave-cut erosion surface on BIF, C Ridge. (E). Tsunami ero- surface on massive hematite. (C,D). Wave-cut erosion surface on BIF, C Ridge. (E). Tsunami erosion sion features on wave-cut surface (sinuous grooves and shallow pothole). (F). Polymict conglom- featureserate. Hammer on wave-cut is 30 cm surface long; (sinuous map case grooves is 32 cm and long; shallow scale card pothole). ~10 cm (F long.). Polymict (Photos conglomerate. by Lascelles). Hammer is 30 cm long; map case is 32 cm long; scale card ~10 cm long. (Photos by Lascelles). A narrow outcrop of the conglomerate to the west of the Wilgie Mia Formation and other deposits of conglomerate to the east of the transported blocks and boulders are overlain by sandstone of the Finlayson Member. In between these conglomerate deposits, the sandstone rests unconformably on the Archean rocks and the basal conglomerate is almost completely absent from the surface of the BIF and adjacent amphibolites to the east but is found on the eastern side of the boulders on B Ridge North (Figure 6E). Fine-grained sandstone of the Finlayson Member was then deposited directly on the scoured surface between the large blocks (Figure 6F) surrounding and overlying the transported blocks and boulders (Figure 6B–D) passing up into ripple- marked sandstone. Thin lenses of fine BIF-rich conglomerate present near the base of the Finlayson Member to the west of the Wilgie Mia Formation outcrop were not seen in the thicker sandstone horizons on B Ridge. Detailed descriptions of the geology of the JFGB, the unconformities, erosion surfaces, and deep weathering profiles were provided in [2,3].

J. Mar. Sci. Eng. 2021, 9, 213 8 of 17

Large blocks and boulders of silicified BIF and massive hematite (up to 40 m × 27 m and 6 m thick; Table2) and individually weighing up to 18,000 tonnes (Figure6A–D) form a broad band 900 to 1200 m east of the outcrop of the Archean Wilgie Mia Formation (Figure3A,B ). A polymict conglomerate (Figures3 and5E) at the base of the Windplain Group consists of locally derived silicified BIF and hematite pebbles with minor goethitized mafic pebbles in a quartz sand matrix and only occurs in the vicinity of the JFGB; elsewhere, J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 8 of 18 the Finlayson Member sandstone rests directly on the unconformity with only the rare occurrence of vein quartz clasts at the base (Figure2).

A B

BIF block

BIF block 5m

Sst

C BIF D He He

Sst Sst Sst

Conglomerate EA F

Unconformity Sst Finlayson Member

Go unconformity

Meekatharra Fm V Meekatharra Formation Am

Figure 6. FigureB Ridge 6. NorthB Ridge outcrops. North ( outcrops.A). Large BIF (A). block Large with BIF overlying block with boulders overlying at left. boulders Dip and at left.strike Dip of BIF and is similar to outcrops on C Ridge. (B). BIF block surrounded by Finlayson Member sandstone. (C). Massive hematite block surrounded strike of BIF is similar to outcrops on C Ridge. (B). BIF block surrounded by Finlayson Member by sandstone. (D). Top of massive hematite and BIF block (s?) protruding through Finlayson Member sandstone. (E). sandstone. (C). Massive hematite block surrounded by sandstone. (D). Top of massive hematite Unconformable contact of polymict conglomerate on ultramafic amphibolite of the Meekatharra Formation. (F). Uncon- formable andcontact BIF of block Finlayson (s?) protruding Member sandstone through Finlayson on Meekatharra Member Formation sandstone. amphibolite. (E). Unconformable Map case contact is 36 cm of long; hammer is 30 polymictcm long. conglomerateAm = amphibolite; on ultramaﬁc BIF = banded amphibolite iron formation, of the Meekatharra He = hematite, Formation. Sst = sandstone. (F). Unconformable (Photos by Lascelles). contact of Finlayson Member sandstone on Meekatharra Formation amphibolite. Map case is 36 cm long; hammer is 30Table cm 2. long. Measure Amd = boulders. amphibolite; BIF = banded iron formation, He = hematite, Sst = sandstone. (Photos by Lascelles). No. Easting ** Northing ** Long Axis Short Axis Rock Type a b c d 1 793,324 7,046,045 10.00 5.00 BIF 2 793,331 7,046,126 10.43 8.25 3.50 BIF 3 793,334 7,048,312 12.30 4.00 BIF + He 4 793,346 7,046,057 12.90 9.17 4.83 3.70 BIF 5 793,352 7,048,276 20.00 18.80 17.30 16.00 He 6 793,357 7,046,125 11.00 2.50 BIF 7 793,360 7,048,341 15.50 11.00 BIF 8 793,363 7,048,274 25.00 12.00 BIF

J. Mar. Sci. Eng. 2021, 9, 213 9 of 17

A narrow outcrop of the conglomerate to the west of the Wilgie Mia Formation and other deposits of conglomerate to the east of the transported blocks and boulders are overlain by sandstone of the Finlayson Member. In between these conglomerate deposits, the sandstone rests unconformably on the Archean rocks and the basal conglomerate is almost completely absent from the surface of the BIF and adjacent amphibolites to the east but is found on the eastern side of the boulders on B Ridge North (Figure6E). Fine-grained sandstone of the Finlayson Member was then deposited directly on the scoured surface between the large blocks (Figure6F) surrounding and overlying the transported blocks and boulders (Figure6B–D) passing up into ripple- marked sandstone. Thin lenses of fine BIF-rich conglomerate present near the base of the Finlayson Member to the west of the Wilgie Mia Formation outcrop were not seen in the thicker sandstone horizons on B Ridge. Detailed descriptions of the geology of the JFGB, the unconformities, erosion surfaces, and deep weathering profiles were provided in [2,3]. Rare stacks of BIF are preserved on the erosion surface (Figure5A), and the erosion- resistant silicified BIF and massive hematite units formed flat-topped reefs (Figure5B–D) 1–2 m above the general level of the wave-cut surface on soft clay-rich mafic and ultramafic saprolite. The basal conglomerate of the Finlayson Member is confined to the JFGB and consists of rounded pebbles of BIF, hematite, and goethitized amphibolite in a quartz sand matrix (Figure5F) derived by marine erosion of the deeply weathered BIF, iron ore, and amphibolite of the JFGB and is overlain by mature sandstone (Figure6B–D). Elsewhere, throughout the northern Yilgarn Province, apart from the rare occurrences of quartz pebble conglomerate [10], the basal conglomerate is absent and sandstone of the Finlayson Member rests directly on deeply weathered Archean granitoids and greenstones (Figure6F) with rare stacks and small pockets of basal conglomerate preserved on the exhumed erosion surface on silicified BIF (Figure5A). The deposition of epicontinental sedimentary rocks of the Windplain Group was followed by minor deformation and tilting of the sequence, including the Archean basement, 12◦ to the north and a prolonged period of deep weathering. The second marine transgression at ~1.95 Ga followed the deposition and subsequent uplift of the Windplain Group and epeirogenic sedimentary rocks of the Tooloo Group were unconformably deposited on the eroded remnants of Windplain Group and over- stepped onto Archean saprolite and unweathered Archean rocks further south. Lascelles [2] described the very clear contrast between the Proterozoic saprolite and the previously unweathered Archean rocks eroded by the Earaheedy transgression. The wave-cut erosion surfaces formed by both of the marine transgressions (Figures2 and5A–E) were exhumed by Cenozoic erosion of the overlying undeformed and unmetamorphosed Paleo- proterozoic cover. These surfaces are well–preserved on the sub-vertical BIF of the Wilgie Mia Formation and less distinctly on the thinner BIF and massive hematite bands in the Yaloginda Formation and silicified weathered granite outcrops to the east of the JFGB. Deeply weathered mafic and ultramafic rocks of the Meekatharra Formation and Murrouli Basalt have typically been eroded well below the level of the unconformity, except for strongly goethitized outcrops, unless protected by the overlying transported boulders, conglomerate, and sediments of the Windplain Group (B Ridge North; Figures2 and6E,F).

unweathered Archean rocks eroded by the Earaheedy transgression. The wave-cut erosion surfaces formed by both of the marine transgressions (Figures 2 and 5A–E) were exhumed by Cenozoic erosion of the overlying undeformed and unmetamorphosed Paleo- proterozoic cover. These surfaces are well–preserved on the sub-vertical BIF of the Wilgie Mia Formation and less distinctly on the thinner BIF and massive hematite bands in the Yaloginda Formation and silicified weathered granite outcrops to the east of the JFGB. Deeply weathered mafic and ultramafic rocks of the Meekatharra Formation and Mur- rouli Basalt have typically been eroded well below the level of the unconformity, except for strongly goethitized outcrops, unless protected by the overlying transported boulders, conglomerate, and sediments of the Windplain Group (B Ridge North; Figures 2 and 6E,F).

4. Transported Block and Boulder Deposits The blocks, boulders, and conglomerate forming a band <3 km wide, 900 to 1200 m east of the Wilgie Mia Formation outcrop, were deposited on the smooth wave-cut erosion surface of deeply weathered mafic and ultramafic amphibolite and schist of the Meeka- tharra Formation. The bedding strike and dip of the largest blocks is oriented north–south similar, to the outcrops on C Ridge (Figure 3A) and was first mapped by the author as in J. Mar. Sci. Eng. 2021, 9, 213 situ Wilgie Mia Formation BIF. The smaller boulders have more random orientations10, ofcast- 17 ing doubt on their being in situ. Furthermore, aeromagnetic imagery (Figure 7) shows no evidence of underlying BIF [2,3], and exposures at the base of the blocks indicated that they were isolated from each other and underlain by deeply weathered mafic and ultra- eredmafic maﬁc amphibolite and ultramaﬁc and schist amphibolite of the Meekatharra and schist ofFormation. the Meekatharra Only a few Formation. direct measure- Only a fewments direct of the measurements block thicknesses of the could block be thicknesses taken, (Table could 2) as bethe taken, bases were (Table typically2) as the covered bases wereby scree, typically conglomerate covered by or scree,sandstone, conglomerate but thickness or sandstone, of 5–6 m could but thickness be reasonably of 5–6 estimated m could beby reasonably the distance estimated from the by unconformity the distance from to the the upper unconformity surface of to the the larger upper blocks surface (Figure of the larger5A). blocks (Figure5A).

792,500 mE 795,000 mE 119˚56' 119o 58' N

0 26˚38' ,

0 1 2

kms

m Large

5 , boulder

4 26˚40'

0 , field

, 7 26˚42' B Ridge C Ridge

FigureFigure 7. 7.Aeromagnetic Aeromagnetic imageimage (Total (Total magnetic magnetic intensity intensity (TMI) (TMI) 1st 1st vertical vertical derivative) derivative) of of part part of of the the JFGB showing location of boulders (black outline). JFGB showing location of boulders (black outline).

The upper surface of the blocks was typically flat with maximum relief elevations of ~1 m. Although absent from the erosion surface on C Ridge and for >1 km to the east, a narrow outcrop of polymict conglomerate is well exposed for a distance of ~3 km on weathered amphibolite to the west of the Wilgie Mia Formation BIF (Figures3 and5E). The polymict conglomerate is also exposed on B Ridge North for ~1 km to the east of a large block of BIF with sporadic exposures associated with BIF and hematite blocks to the north. Between the blocks, the unconformity is overlain by sandstone with no trace of polymict conglomerate at the base. The Finlayson Member disconformably overlies the conglomerate with a small outlier of sandstone resting directly on the BIF at C Ridge. A single layer of slightly irregular spheroids of granulitic metaquartzite was found near the base of the strongly weathered granular quartz sandstone (Figure8A) with a few irregular blebs of devitrified glass attached to the spheroids (Figure8B). SEM and electron microprobe examination (J. Muhling, pers. comm. 2012) indicated that the glass had a similar composition to the metaquartzite grains. Quartz grains in the metaquartzite show parallel planes of fluid inclusions that are not present in the sandstone grains (Figure8C,D). The metaquartzite spheroids appear to be of exotic origin with no known local correlatives. It is suggested that they could possibly be ejecta from an impact crater. J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 11 of 18

The upper surface of the blocks was typically flat with maximum relief elevations of ~1 m. Although absent from the erosion surface on C Ridge and for >1 km to the east, a narrow outcrop of polymict conglomerate is well exposed for a distance of ~3 km on weathered amphibolite to the west of the Wilgie Mia Formation BIF (Figures 3 and 5E). The polymict conglomerate is also exposed on B Ridge North for ~1 km to the east of a large block of BIF with sporadic exposures associated with BIF and hematite blocks to the north. Between the blocks, the unconformity is overlain by sandstone with no trace of polymict conglomerate at the base. The Finlayson Member disconformably overlies the conglomerate with a small outlier of sandstone resting directly on the BIF at C Ridge. A single layer of slightly irregular spheroids of granulitic metaquartzite was found near the base of the strongly weathered granular quartz sandstone (Figure 8A) with a few irregular blebs of devitrified glass attached to the spheroids (Figure 8B). SEM and electron microprobe examination (J. Muhling, pers. comm. 2012) indicated that the glass had a similar composition to the metaquartzite grains. Quartz grains in the metaquartzite show

J. Mar. Sci. Eng. 2021, 9, 213 parallel planes of fluid inclusions that are not present in the sandstone grains11 of (Figure 17 8C,D). The metaquartzite spheroids appear to be of exotic origin with no known local correlatives. It is suggested that they could possibly be ejecta from an impact crater.

A B

Sst Dvg Mqt

500 m

100μm

FigureFigure 8. 8.(A (A).) Layer. Layer of spheroidsof spheroids on sandstone,on sandstone, B Ridge B Ridge North. North. (B). Portion (B). Portion of granulitic of granulitic metaquartzite metaquartzite spheroid (left of dashed boundary) with attached bleb of devitrified glass (dotted spheroid (left of dashed boundary) with attached bleb of devitrified glass (dotted boundary) in boundary) in granular loosely cemented sandstone matrix. (C). Enlarged view of metaquartzite granular loosely cemented sandstone matrix. (C). Enlarged view of metaquartzite with planar with planar fluid inclusion tracts in quartz (arrowed) (plane-polarized transmitted light). (D). fluid inclusion tracts in quartz (arrowed) (plane-polarized transmitted light). (D). Same view with Same view with crossed polarizers. GPS unit is 15 cm long. Dvg = devitrified glass; Mqt = granu- crossed polarizers. GPS unit is 15 cm long. Dvg = devitrified glass; Mqt = granulitic metaquartzite; litic metaquartzite; Sst = sandstone. Sst = sandstone.

5.5. Discussion Discussion TheThe large large blocks blocks were were derived derived from from the the Wilgie Wilgie Mia Mia Formation Formation as none as none of the of BIFthe unitsBIF units inin the the Yaloginda Yaloginda Formation Formation are are thick thick enough enough to provideto provide such such large large blocks, blocks and, and several several possiblepossible scenarios scenarios were were considered considered to to explain explain how how the the BIF BIF and and massive massive hematite hematite blocks blocks werewere transported transported at at least least 900 900 m m to to the the east east from from C RidgeC Ridge (Figure (Figure2) on 2) the on Yerridathe Yerrida Surface. Surface. PossiblePossible agencies agencies for for the the transportation transportation of suchof such large large blocks blocks range range from from low-angle low-angle thrust thrust faults,faults, mass mass wasting, wasting, debris debris flows, flows, glacial glacial transport, transport or, tsunami;or tsunami however,; however, the wave-cutthe wave-cut ◦ surfacesurface is is extremely extremely flat, flat, with with the the exception exception of rareof rare stack stack remnants, remnants, dipping dipping 12 to12° the to the north, and shows very little variation in elevation between the boulder deposits and the erosion surface on C Ridge (Figure3B). The large size of the blocks makes them significantly beyond the capacity of tidal surges and storms to move. Only an extreme flood from an exceptionally large tsunami would have the power to move 18,000-tonne masses over ~1 km. A low-angle thrust is discounted as the stacks rise from the unconformity surface (Figure5A), and the boulders and conglomerate rest directly on the unconformity with no intervening breccia or slick- ensides. A debris flow across the slope could not transport massive material horizontally across such a gently sloping surface. Similarly, mass wasting requires a large difference in altitude for gravity to transport the blocks over one kilometre, which was not present in a low-lying deeply weathered landscape. Numerous boulders lie adjacent to the BIF outcrops on C Ridge at the foot of small cliffs, where they are clearly derived by mass wasting but never extend more than a few metres from the outcrop. At Mt Gibson and Koolyanobbing, for example, sub-vertical BIF outcrops form pronounced ridges with deposits formed by mass wastage extending no more than a few metres from the outcrop. In neither case were BIF boulders found more than 10 m from the outcrop. The stacks show that at least ~3 m was eroded locally by the transgression, but the maximum cannot be J. Mar. Sci. Eng. 2021, 9, 213 12 of 17

much greater or unsilicified saprolite would have been exposed and the surfaces could not have been preserved. The Hamersley Province has many exposures of shallow dipping (0–40◦) BIF forming pronounced cliffs outcropping up to 100 m above adjacent valleys. In most cases, their scree deposits contain numerous large boulders at the top, diminishing in size towards the foot of the slope. Only rarely can large blocks and boulders be carried downslope due to sliding or slumping of scree, due in most cases to increased water saturation. Near Giles point in the Ophthalmia Range, a large 2 m block of BIF had been carried to the foot of the slope ~50 m from the base of the cliff over 30 m above. At Sliding Mountain in the central part of the Hamersley Ranges and elsewhere, large-scale mass wasting is such that Brockman Formation dipping parallel to the slope has been undercut by fluvial erosion, allowing oversized slabs to break away and slide down towards the valley. This movement is invariably down a steep hillside and does not pass beyond the foot of the slope. The southernmost blocks at JFGB form part of the cuesta on the opposite side of a valley. At Hope Downs, a ravine that was probably the result of tunnel erosion or cavity collapse, cutting through Marra Mamba Iron Formation was blocked by large <4 m boulders of BIF, probably derived from the roof of the cavity and more or less in situ, that the underlying stream was unable to move. In all these cases, a large difference in elevation is involved, yet transportation is limited to tens of metres. At JFGB, there was little or no difference in elevation of the essentially flat Yerrida Surface, yet the blocks were transported 900 m from their source. No evidence of glacial deposition such as tillites or other glacial features are known from near the JFGB or elsewhere in the Yilgarn Province in the Paleoproterozoic period. The Meteorite Bore Member (MBM) of the Kungarra Formation lies on the Pilbara Craton that was an unknown distance from the Yilgarn Craton at the time and has been dated at ~2.31 Ga [14] over 100 Ma earlier. The well-preserved erosion surfaces, boulders, and conglomerate pebbles show no trace of glacial striae. The Yilgarn Craton was a low-lying deeply weathered landscape at 2.20 Ga, clearly shown by the absence of unweathered granite below the unconformity and basal conglomerate. This is evidence of a long period of weathering and stability prior to the transgression. Ice cover would have removed the soft regolith. If the blocks had been moved a mere 1 km to the east by an ice front, then the conglomerate would constitute a moraine left behind as the ice sheet melted and should have covered the whole area and included granite clasts. Since the ice cannot reverse direction, there is no possibility for BIF and hematite pebbles to be deposited to the west of the BIF outcrop, and therefore the polymict conglomerate cannot be part of a moraine. Several lines of evidence support a theory of tsunami deposition. The situation of the blocks, their transportation and deposition, and the scouring of the sea floor are inexplicable by processes other than an extraordinarily large flood during the transgression. The larger slabs show no evidence of rolling or sliding and appear to have been picked up and then dropped in a linear deposit parallel to the source (i.e., in a single action by the onset of the largest bore). They could not be further moved by backwash or subsequent waves, unlike the smaller boulders (Table2), which were probably deposited as irregularly oriented clumps [8]. The basal polymict conglomerate was preserved on the western side of the Wilgie Mia Formation, which acted like a groyne but, apart from rare pockets of polymict conglomerate in potholes on the BIF erosion surface, was completely scoured from the surface of the BIF and the sea floor to the east except where deposited in the lee of the largest blocks (Figure9). The smaller boulders may have been loose debris on the sea floor, and rare stacks may have had fractured tops that were dislodged by the tsunami, but the major source of the giant blocks was probably due to the plucking of joint-bounded blocks from the eastern side of the BIF outcrop (cf. 5). The probable source of the larger slabs can be seen in rectangular notches along the eastern side of the Wilgie Mia Formation outcrop (Figure9), and bedrock sculpturing similar to that described by Bryant and Young [15] is visible on the eastern edge of the wave-cut erosion surface (Figure5E), although such sculpturing by tsunami is J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 13 of 18

Several lines of evidence support a theory of tsunami deposition. The situation of the blocks, their transportation and deposition, and the scouring of the sea floor are inexplicable by processes other than an extraordinarily large flood during the transgression. The larger slabs show no evidence of rolling or sliding and appear to have been picked up and then dropped in a linear deposit parallel to the source (i.e., in a single action by the onset of the largest bore). They could not be further moved by backwash or subsequent waves, unlike the smaller boulders (Table 2), which were probably deposited as irregularly oriented clumps [8]. The basal polymict conglomerate was preserved on the western side of the Wilgie Mia Formation, which acted like a groyne but, apart from rare pockets of polymict conglomerate in potholes on the BIF erosion surface, was completely scoured from the surface of the BIF and the sea floor to the east except where deposited in the lee of the largest blocks (Figure 9). The smaller boulders may have been loose debris on the sea floor, and rare stacks may have had fractured tops that were dislodged by the tsunami, but the major source of the giant blocks was probably due to the plucking of joint-bounded blocks from the east- J. Mar. Sci. Eng. 2021, 9, 213 ern side of the BIF outcrop (cf. 5). The probable source of the larger slabs can be seen13 of in 17 rectangular notches along the eastern side of the Wilgie Mia Formation outcrop (Figure 9), and bedrock sculpturing similar to that described by Bryant and Young [15] is visible on the eastern edge of the wave-cut erosion surface (Figure 5E), although such sculpturing bydisputed tsunami by is later disputed authors. by Thelater fine authors. quartz The sand fine that quartz was deposited sand that directlywas deposited on the wave-cut directly onsurface the wave between-cut andsurface over between the boulder and andover conglomerate the boulder depositsand conglomerate was probably deposits carried was by probablythe tsunami carried bore and by the deposited tsunami immediately bore and deposited by the flood. immediately The metaquartzite by the flood. spheroids The metaquartzitewere apparently spheroids also deposited were apparently during this also period. deposited The during proposed this tsunami period. depositsThe proposed were tsunamioverlain deposits by the thick were ripple-marked overlain by the sandstone thick ripple of the-marked Finlayson sandstone Member of the at the Finlayson base of Memberthe Windplain at the Group. base of The the sandstone Windplain with Group. thin The lenses sandstone of fine-grained with thin BIF lenses conglomerate of fine- grainedoverlying BIF the conglomerate conglomerate overlying on the westthe conglomerate side of the Wilgie on the Mia west Formation side of the outcrop Wilgie wasMia Formationprobably formed outcrop by was normal probably marine formed processes by normal after the marine tsunami. processes after the tsunami. Before tsunami Before

After tsunami

Deposition of sandstone After

Present outcrops N Tsunami direction Legend Sandstone Conglomerate Glen Group Wilgie Mia Fm (BIF) Meekatharra Fm FigureFigure 9. Simplistic schematicschematic diagrams diagrams of theof the sea sea floor floor before before and afterand theafter tsunami. the tsunami. Left, cross-sections. Left, cross-sections. Right, block Right diagrams., block diagrams. The marine transgression that formed the wave-cut erosion surface is comparable to theThe Holocene marine transgression marine transgression that formed that the occurred wave-cut in the erosion English surface Channel is comparable [16–18] and to thethe Holocene North Sea marine as sea levels transgression rose following that occurred the Pleistocene in the English ice ages. Channel Wave erosion [16–18] occurred and the Northdown toSea a maximumas sea levels wave rose depth following of ~10 the m, Pleistocene forming a wave-cut ice ages. erosion Wave erosion surface inoccurred deeply downweathered to a maximum rocks on which wave sediments depth of ~10 were m, deposited forming a as wave the sea-cut level erosion continued surface to in rise deeply and weatheredthe shoreline rocks moved on which inland. sediments were deposited as the sea level continued to rise and the shorelineThe landscape moved ofinland. the Yilgarn Craton at the time of the first marine transgression was alsoThe landscape deeply weathered of the Yilgarn to a depth Craton of at <200 the mtime [2] of and the generally first marine flat transgression with silicified was BIF, alsosilcreted deeply granite, weathered and massive to a depth hematite of <200 providing m [2] andminor generally relief. Naturally, flat with no silicified trace of BIF, the shoreline at the time of the tsunami exists as the transgression continued for at least 100 km to the south of the JFGB, and it must be realized that marine transgressions take place over tens to hundreds of thousands of years, depending on the distance covered, the elevation of the eroding landscape and possibly the rate of subsidence of the continent. Neither the blocks nor the conglomerate show any sign of reworking by marine erosion prior to the deposition of the sandstone, of which the lowest layer would have been deposited from the flood. However, no transported boulders occur south of latitude 26◦37033.8” S (Figure3A), suggesting that was the limit of deposition below wave base and deposits further south were reworked by the progressive marine erosion. The northern and eastern limits of the boulder field are obscured by alluvial and aeolian deposits.

5.1. Estimation of the Flood Velocity and Height Although extreme storm waves have been known to move large boulders (e.g., up to two metres diameter) many tens of meters [19], they tend to be much less efﬁcient in transporting larger boulders (>2 m) over 100s of meters, for several reasons. Firstly, maximum storm wave heights Hmax in the deep ocean are limited by their maximum steepness Hmax/L~1/7, where L is the wave length, which tends to limit wind-wave heights to the order of 10 m, even under extreme conditions. Secondly, the short period of J. Mar. Sci. Eng. 2021, 9, 213 14 of 17

storm waves also implies they cannot sustain high velocities in a single direction to move 40 m wide blocks over the distances of ~1000 m that were observed. Observations of tsunami boulder deposits on land have proven very useful in providing robust estimates of the height and maximum flow velocities associated with modern tsunami waves e.g., [5,6], but very few studies of extremely large bores have been done. In the present case of sea floor deposits, it is assumed that the shallow water on the shelf was dragged along by the flood and that the size of the boulders and their displacement from the Wilgie Mia Formation outcrop (Figures2 and6) can provide a plausible estimate of the magnitude of the flood. At the time when the event occurred, the unconformity representing the sea floor was already eroded to the wave base by the marine transgression, and the deposits left by the tsunami on the sea floor were undisturbed by further wave action or further erosion of the sea floor, implying the local water depth h was >10 m. It is most probable that the marine transgression inundated the continent from the north as the Yerrida Basin formed 50–60 km to the northwest [10], but all traces of coastline, coastal deposits, and onshore deposits were reworked south of latitude 26◦37033.8” S as the marine transgression progressed. In addition, high energy flow in tsunamis generally lasts for minutes [8]. The average flow velocity required to transport the boulders in full-suspension over ~1 km would be of order 10 m s−1, which is also compatible with the minimum instantaneous flow velocity (~30 m s−1) required to initiate the transport.

5.2. Speculation on Possible Cause of the Flood A catastrophic event could produce a tsunami of this magnitude (flood > 40 m in height). It is unlikely that a massive rock slump could generate a flood with the velocity required to transport blocks of over 6000 m3 in the region of 20,000 tonnes for approximately 1 km, particularly with the low-altitude differential of the extant landscape [20]. A massive rock slump on the continental slope would only generate low-velocity tsunami waves. An extremely large earthquake on the sea floor with an instant displacement of >30 m could generate waves of sufficient magnitude to displace the blocks, but in neither case would a high-velocity splash be formed to cause a high velocity bore that would be required to scour sand and shingle from the sea floor. The most probable cause would ultimately be a large oceanic asteroid impact, but a perfunctory search of the Juderina Formation failed to discover any impact spherulites. The layer of granulitic metaquartzite spheroids (Figure6A,B) with the devitrified glass blebs and the planar fluid inclusion trails that may possibly be shock-induced (Figure6C,D ) are suggestive but inconclusive evidence of impact ejecta, and no further evidence was found.

The layer of granulitic metaquartzite spheroids (Figure 6A,B) with the devitrified glass blebs and the planar fluid inclusion trails that may possibly be shock-induced (Figure 6C,D) are suggestive but inconclusive evidence of impact ejecta, and no further evidence was found.

5.3. Oceanic Asteroid Impacts Four very large asteroids, tens of kilometres in diameter (Vredefort, Sudbury, Shoe- maker and Yarrabubba), impacted on land during the Paleoproterozoic era. At present, approximately 71% of the Earth’s surface is covered by oceans, whereas during the Paleo- proterozoic, continents were typically smaller with an estimated 90% of the Earth’s surface covered by ocean. Statistically, ten times as many asteroids could have impacted the ocean. Most of the ejecta from an ocean floor crater would either be hurled into space or into the surrounding water column or fall back onto the ocean floor. Unfortunately, since the onset of plate tectonics, oceanic crust is subducted and all the evidence of ancient oceanic impacts is destroyed. Several authors [21–27] have calculated the effects of ocean impacts by small asteroids (300–400 m diameter). These calculations were based largely on the calculations of Van Dorn [28,29] with respect to underwater explosions and concluded that they would generate comparatively small tsunami waves. Furthermore, due to the Van Dorn effect, their force would be dissipated before reaching shore. Melosh [24] claimed that the water displaced by a vertical impact would pile up near the rim of the crater (Figure 10). All studies have concentrated on the waves generated by the collapse of the ocean cavity caused by the passage of the bolide but have not considered the effect of splash caused by the displacement of water at impact. The Van Dorn [24] study involved waves generated J. Mar. Sci. Eng. 2021, 9, x FOR PEERby REVIEW underwater explosives, in which the water displaced by the explosion fell back into16 theof 18 J. Mar. Sci. Eng. 2021, 9, 213 15 of 17 cavity, from which it was assumed that the splash generated by asteroid impacts behaved similarly. However, they did not calculate such effects for bolides with a magnitude greaterthe depth than at the the depth point of of the impact, water wouldor non -alsovertical cause impacts. a massive displacement of water. In causedadditionA simple by to the the passageexperiment size of of the the readily impactor, bolide shows but the have thedimensions notcontrasting considered of theeffect thesplash of effect a largewould of splash asteroid depend caused bolide. on bythe Athevelocity 4 cm displacement diameter of the impact spheroid of water and (golf at the impact. ball)angle dropped Theof incidence Van from Dorn (Figurea [height24] study 11). of 50 For involved cm impacts into wavesa large with generatedtray a high filled ve- withbylocity underwater water and 5high cm explosives, deepangle cre of atedincidence in whicha largely, theit is vertical water probable displaced splash that withsome by theminor of explosionthe airborne displaced fell dropletsback water into wouldup the to 10cavity,be cm vaporized from from the which and point itthus wasof impactexplode assumed (Figure laterally that 10B the in) splash. allHowever, directions generated when, cre by aating 7.5 asteroid cm a craterdiameter impacts much rock behaved greater was droppedsimilarly.than the from However,diam theeter same of they the didheight impactor not the calculate (Figureeffect, such was 10A effectsspectacular). By comparison, for bolides with witha a large low a magnitude anglevolume of ofincidence greater water ejectedthanwould the as impel depth far as the often thedisplaced times water the or water diameter non-vertical in the of thedirection impacts. rock (Figure of the impact.11). The water displaced by the impact would travel away from the impact over the surface of the ocean at hundreds A of meters per second—not as a wave but as a moving mass with its own momentum (Fig- T T T T 0 ure 12). Collapse1 ⁭ of the cavity in2 the ocean so produced3 would then generate waves with a height approaching the depth of the ocean and velocity of hundreds of metres per second. Water travelling over the surface of the ocean as a flood would be slowed by friction B with the underlying water, butT the momentum wouldT also be transferred and the surface T0 T1 2 3 water dragged along until reaching the shelf, where drag would then be caused by contact with the sea floor, inducing the removal of loose sediment from the sea floor. The amount of friction involved is roughly constant for any size of bore, but the larger the bore, the less reduction in velocity over a given distance. Thus, whereas a small impact producing a flood a few meters deep would be rapidly slowed, water and sea floor drag would have Sea water Sea Floor minimal effect on floods of 50 m height and above. The flood from the impact of an aster- Figure 10. Tsunami generation.oid kilometres (A) Underwater in diameter explosion. would (Balso) Impact drag of the small shallow bolide. shelf To == time water of alongimpact. , producing a massive flow capable of scouring loose debris from the sea floor and transporting the large blocksA simplelargeas seen asteroid, experiment at Joyner’s greater readily Find. than showsThe two Van the kilometres Dorn contrasting effect across applies effect in of diameterto a largewaves asteroid assumingthat go bolide. from ocean the A depth4rotary cm diameter in motion the Paleoproterozoic spheroidof a wave (golf to forward ball)of two dropped kilometres, translation from would aafter height breakinginitially of 50 cmdisplace and into does a a large volume not trayapply of ﬁlled wa- to a terwithstrong (diameter water flood 5 cmof of the deepwater asteroid created that timesis aalready largely depth in vertical of forward the ocean) splashing atmotion with a velocity minor (Figure airborneapproaching 12). Contrary droplets the veloc- upto the to ity10opinion cmof the from asteroidof the[24] point, anat hundredsasteroid of impact greater of (Figure meters in 10perdiameterB). second However, than (the the whensplash). depth a 7.5Similarly of cm the diameter ocean,, a much of rock smallwhich waser > bolidedropped100,00 striking0 occur from withinthe sea same the floor heightasteroid on a the continental belt, effect, would was shelf, cause spectacular provided a disastrous with it is flood agreater large in volumeadditionin diameter ofto watersevere than ejectedearthquakes as far asand ten atmospheric times the diameter disruption of thedevastating rock (Figure to coastal 11). communities in its path.

T0 T1

T T2 3

Figure 11. Tsunami generation by a large impactor. To. Impact creates a major splash. T1. Splash Figure 11. Tsunami generation by a large impactor. To. Impact creates a major splash. T1. Splash forms a bore travelling away from the impact. T2. Closure of cavity after bore has moved rapidly forms a bore travelling away from the impact. T2. Closure of cavity after bore has moved rapidly away from the site generates tsunami waves T3. away from the site generates tsunami waves T3. A large→ asteroid,Bore greater→ than two kilometres across in diameter assuming ocean depth in the Paleoproterozoic of two kilometres,→ would initially displace a volume of water (diameter→ of→ the asteroid→ → times depth of the ocean) at a velocity approaching the velocity of the asteroid at hundreds of meters per second (the splash). Similarly, a much smaller bolide striking the sea ﬂoor on a continental shelf, provided it is greater in diameter than Figure 12. Contrasting effects of bore and waves on reaching shallow water. the depth at the point of impact, would also cause a massive displacement of water. In addition6. Conclusion to thessize of the impactor, the dimensions of the splash would depend on the velocity of the impact and the angle of incidence (Figure 11). For impacts with a high velocityOf andthe usual high angleproxies of incidence,for identifying it is probableHolocene that paleo some-tsunamis of the displaced [29], only waterthe grain would size beof vaporizedthe deposits and and thus the explode erosion laterally of the sea in allfloor directions, are available creating for athe crater identification much greater of a than very theancient diameter flood of deposit the impactor that occurred (Figure 10duringA). By a comparison,marine transgressions. a low angle Howev of incidenceer, due would to the impel the displaced water in the direction of the impact. The water displaced by the impact

J. Mar. Sci. Eng. 2021, 9, x FOR PEER REVIEW 16 of 18

the depth at the point of impact, would also cause a massive displacement of water. In addition to the size of the impactor, the dimensions of the splash would depend on the velocity of the impact and the angle of incidence (Figure 11). For impacts with a high velocity and high angle of incidence, it is probable that some of the displaced water would be vaporized and thus explode laterally in all directions, creating a crater much greater than the diameter of the impactor (Figure 10A). By comparison, a low angle of incidence would impel the displaced water in the direction of the impact. The water displaced by the impact would travel away from the impact over the surface of the ocean at hundreds of meters per second—not as a wave but as a moving mass with its own momentum (Fig- ure 12). Collapse of the cavity in the ocean so produced would then generate waves with a height approaching the depth of the ocean and velocity of hundreds of metres per second. Water travelling over the surface of the ocean as a flood would be slowed by friction with the underlying water, but the momentum would also be transferred and the surface water dragged along until reaching the shelf, where drag would then be caused by contact with the sea floor, inducing the removal of loose sediment from the sea floor. The amount of friction involved is roughly constant for any size of bore, but the larger the bore, the less reduction in velocity over a given distance. Thus, whereas a small impact producing a flood a few meters deep would be rapidly slowed, water and sea floor drag would have minimal effect on floods of 50 m height and above. The flood from the impact of an asteroid kilometres in diameter would also drag the shallow shelf water along, producing a massive flow capable of scouring loose debris from the sea floor and transporting the large blocks as seen at Joyner’s Find. The Van Dorn effect applies to waves that go from the rotary motion of a wave to forward translation after breaking and does not apply to a strong flood of water that is already in forwarding motion (Figure 12). Contrary to the opinion of [24], an asteroid greater in diameter than the depth of the ocean, of which > 100,000 occur within the asteroid belt, would cause a disastrous flood in addition to severe earthquakes and atmospheric disruption devastating to coastal communities in its path.

T0 T1

T J. Mar. Sci. Eng. 2021, 9, 213 T2 3 16 of 17

would travel away from the impact over the surface of the ocean at hundreds of meters per second—not as a wave but as a moving mass with its own momentum (Figure 12). Figure 11. Tsunami generation by a large impactor. To. Impact creates a major splash. T1. Splash Collapse of the cavity in the ocean so produced would then generate waves with a height forms a bore travelling away from the impact. T2. Closure of cavity after bore has moved rapidly approaching the depth of the ocean and velocity of hundreds of metres per second. away from the site generates tsunami waves T3. → Bore → → → → → →

FigureFigure 12.12. ContrastingContrasting effectseffects ofof borebore andand waveswaves onon reachingreaching shallowshallow water.water.

6. ConclusionWater travellings over the surface of the ocean as a flood would be slowed by friction with theOf the underlying usual proxies water, for but identif the momentumying Holocene would paleo also-tsunamis be transferred [29], only and the the grain surface size waterof the draggeddeposits alongand the until erosion reaching of the the sea shelf, floor where are available drag would for the then identification be caused by of contact a very withancient the flood sea floor, deposit inducing that occurred the removal during of loose a marine sediment transgressions. from the sea Howev floor.er, The due amount to the of friction involved is roughly constant for any size of bore, but the larger the bore, the less reduction in velocity over a given distance. Thus, whereas a small impact producing a flood a few meters deep would be rapidly slowed, water and sea floor drag would have minimal effect on floods of 50 m height and above. The flood from the impact of an asteroid kilometres in diameter would also drag the shallow shelf water along, producing a massive flow capable of scouring loose debris from the sea floor and transporting the large blocks as seen at Joyner’s Find. The Van Dorn effect applies to waves that go from the rotary motion of a wave to forward translation after breaking and does not apply to a strong flood of water that is already in forwarding motion (Figure 12). Contrary to the opinion of [24], an asteroid greater in diameter than the depth of the ocean, of which >100,000 occur within the asteroid belt, would cause a disastrous flood in addition to severe earthquakes and atmospheric disruption devastating to coastal communities in its path.

6. Conclusions Of the usual proxies for identifying Holocene paleo-tsunamis [29], only the grain size of the deposits and the erosion of the sea floor are available for the identification of a very ancient flood deposit that occurred during a marine transgressions. However, due to the absence of any other feasible explanation, it is proposed that the blocks, boulders, and conglomerate were transported by a flood produced by the oceanic impact of a large asteroid hundreds to thousands of meters in diameter. Further research is suggested to match individual blocks with gaps in the Wilgie Mia Formation and identify additional tsunami erosion features [8,16] in an outcrop of the Wilgie Mia Formation. In addition, it must be established how much of the Finlayson Member at Joyner’s Find was deposited by the tsunami flood, how much of the ripple marked sandstone could have been deposited by subsequent tsunami waves, and just where the quartzite spheroids were deposited. Detailed examination of the basal conglomerate and the base of the Finlayson Member may also disclose impact spherulites or elevated iridium levels.

Author Contributions: Conceptualization, D.F.L.; Methodology, D.F.L., R.J.L.: Validation, D.F.L.; investigation, D.F.L.; data curation, D.F.L.; Writing, D.F.L.; All authors have read and agreed to the published version of the manuscript. Funding: This research rceived no external funding. The APC was funded by D.F.L. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Not applicable. J. Mar. Sci. Eng. 2021, 9, 213 17 of 17

Acknowledgments: The discovery was made during a 1:2000 geological mapping project for Golden West Resources Ltd., who gave permission to publish the work. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of Interest: The authors declare no conflict of interest.

References 1. Occhipinti, S.; Hocking, R.; Lindsay, M.; Aitken, A.; Copp, I.; Jones, J.; Sheppard, S.; Pirajno, F.; Metelka, V. Paleoproterozoic basin development on the northern Yilgarn Craton, Western Australia. Precambrian Res. 2017, 300, 121–140. [CrossRef] 2. Lascelles, D.F. Paleoproterozoic regolith and wave cut platforms preserved on the northern margin of the Yilgarn Craton, Western Australia. Aust. J. Earth Sci. 2014, 61, 843–863. [CrossRef] 3. Lascelles, D.F.; Tsiokos, D.S. Microplaty hematite ore in the Yilgarn Province of Western Australia: The geology and genesis of the Wiluna West iron ore deposits. Ore Geol. Rev. 2014, 66, 309–333. [CrossRef] 4. Van Kranendonk, M.J.; Ivanic, T.J. A new lithostratigraphic scheme for the north-eastern Murchison Domain, Yilgarn Craton. In Geological Survey of Western Australia Annual Review; 2007–2008; GSWA: Perth, Australia, 2008; pp. 35–53. 5. Nandasena, N.A.K.; Paris, R.; Tanaka, N. Reassessment of hydrodynamic equations: Minimum flow velocity to initiate boulder transport by high energy events (storms, tsunamis). Mar. Geol. 2011, 281, 70–84. [CrossRef] 6. Nott, J.F. Waves, coastal boulder deposits and the importance of the pre-transport setting. Earth Planet. Sci. Lett. 2003, 210, 269–276. [CrossRef] 7. Nott, J.F. Extremely -energy wave deposits inside the Great Barrier Reef, Australia: Determinin high g the cause-tsunami or tropical cyclone. Mar. Geol. 1997, 141, 193–207. [CrossRef] 8. Bryant, E.A. Tsunami. In The Underrated Hazard; Cambridge University Press: Cambridge, UK, 2001; 320p. 9. Felton, E.A.; Crook, K.A.W. Evaluating the impacts of huge waves on rocky shorelines: An essay review of the book ‘Tsunami—the underrated hazard’. Mar. Geol. 2003, 197, 1–12. [CrossRef] 10. Elias, M.; Bunting, J.A.; Wharton, P.H.; Glengarry, W.A. Western Australia Geological Survey. In 1:250,000 Series Explanatory Notes; GSWA: Perth, Australia, 1982; 27p. 11. Ferdinando, D.D. Western Australia Geological Survey, 1:100,000 Geological Series Explanatory Notes. In Geology of the Merewether 1:100,000 Sheet; GSWA: Perth, Australia, 2002; 24p. 12. Fairbridge, R.W.; Finkl Charles, W., Jr. Cratonic erosion unconformities and peneplains. J. Geol. 1980, 88, 69–86. [CrossRef] 13. Pirajno, F.; Jones, J.A.; Hocking, R.M.; Halilovic, J. Geology and tectonic evolution of Palaeoproterozoic basins of the eastern Capricorn Orogen, Western Australia. Precambrian Res. 2004, 128, 315–342. [CrossRef] 14. Philippot, P.; Ávila, J.N.; Killingsworth, B.A.; Tessalina, S.; Baton, F.; Caquineau, T.; Muller, É.; Pecoits, E.; Cartigny, P.; Lalonde, S.V.; et al. Globally asynchronous sulphur isotope signals require re-definition of the Great Oxidation Event. Nat. Commun. 2018, 9, 2245. [CrossRef][PubMed] 15. Bryant, E.A.; Young, R.W. Bedrock-sculpturing by tsunami, south coast New South Wales, Australia. J. Geol. 1996, 104, 565–583. [CrossRef] 16. Rankin, W.J.; May, R.E. Geotechnical properties. In Engineering Geology of the Channel Tunnel; Thomas Telford Publishing: London, UK, 1996; pp. 144–173. 17. Wood, C.J.; Shephard-Thorn, E.R.; Harris, C.S. Stratigraphy and Regional Geology. In Engineering Geology of the Channel Tunnel; Thomas Telford Publishing: London, UK, 1996; pp. 29–56. 18. Lorang, M.S. A wave-competence approach to distinguish between boulder and megaclast deposits due to storm waves versus tsunamis. Mar. Geol. 2011, 283, 90–97. [CrossRef] 19. Løvholt, F.; Pedersen, G.; Harbitz, C.B.; Glimsdal, S.; Kim, J. On the characteristics of landslide tsunamis. Phil. Trans. R. Soc. A 2015, 373.[CrossRef][PubMed] 20. Crawford, D.A.; Mader, C.L. Modelling asteroid impact and tsunami. Sci. Tsunami Hazard. 1998, 16, 21–30. 21. Gisler, G.; Weaver, R.; Mader, C.; Gittings, M.L. Two-and three-dimensional simulations of asteroid ocean impacts. Sci. Tsunami Hazard. 2003, 27, 119–134. [CrossRef] 22. Gisler, R.G. Tsunami simulations. Annu. Rev. Fluid Mech. 2008, 40, 71–90. [CrossRef] 23. Melosh, H.J. Impact generated tsunamis: An over-rated hazard. In Proceedings of the 34th Annual Lunar and Planetary Science Conference, League City, TX, USA, 17–21 March 2003; p. 2013. 24. Korycansky, D.G.; Lynett, P.J. Offshore breaking of impact tsunami: The Van Dorn effect revisited. Geophys. Res. Lett. 2005, 32, L10608. [CrossRef] 25. Weiss, R.; Wünnemann, K.; Bahlburg, H. Numerical modelling of generation, propagation and run-up of tsunamis caused by oceanic impacts: Model strategy and technical solutions. Geophys. J. Int. 2006, 167, 77–88. [CrossRef] 26. Wünnemann, K.; Weiss, R. The meteorite impact-induced tsunami hazard. Phil. Trans. R. Soc. A 2015, 373, 20140381. [CrossRef] [PubMed] 27. Van Dorn, W.G.; Le Méhauté, B.; Hwang, L.-S. Handbook of Explosion Generated Water Waves; Report TC-130; Tetra Tech Inc.: Pasadena, CA, USA, 1968. 28. Le Mehaute, B.; Wang, S. Water Waves Generated by Underwater Explosion; World Scientific: Singapore, 1996. 29. Goff, J.; Chagué-Goff, C.; Nichol, S.; Jaffe, B.; Dominey-Howes, D. Progress in palaeotsunami research. Sediment. Geol. 2012, 243–244, 70–88. [CrossRef]