<<

A POSSIBLE LATE IN CENTRAL NORTH AMERICA AND ITS RELATION TO THE

SUBMITTED TO THE FACULTY OF THE UNIVERSITY OF MINNESOTA BY

David Tovar Rodriguez

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

Howard Mooers, Advisor

August 2020

 2020 David Tovar

All Rights Reserved

ACKNOWLEDGEMENTS

I would like to thank my advisor Dr. Howard Mooers for his permanent support, my family, and my friends.

i

Abstract

The causes that started the Younger Dryas (YD) event remain hotly debated. Studies indicate that the drainage of into the North Ocean and south through the Mississippi River caused a considerable change in oceanic thermal currents, thus producing a decrease in global temperature. Other studies indicate that perhaps the impact of an extraterrestrial body (asteroid fragment) could have impacted the Earth 12.9 ky BP ago, triggering a of events that caused global temperature drop. The presence of high concentrations of iridium, charcoal, fullerenes, and molten glass, considered by-products of extraterrestrial impacts, have been reported in sediments of the same age; however, there is no impact structure identified so far. In this work, the Roseau structure's geomorphological features are analyzed in detail to determine if impacted layers with plastic deformation located between hard rocks and a thin layer of water might explain the particular shape of the studied structure. Geophysical data of the study area do not show gravimetric anomalies related to a possible impact structure. One hypothesis developed on this works is related to the structure's shape might be explained by atmospheric explosions dynamics due to the disintegration of material when it comes into contact with the atmosphere. Relationship between structure's diameter (D) and deformed strata thickness (h) as well as the relationship between the diameter of the projectile (d) and the depth of the water column (H ), which is considered in this study due to the geographical considerations of the area 12.9 ky ago and BP, are consistent with an extraterrestrial event. Other hypotheses, such as lake processes and glacial processes, are difficult to reconcile with the reported observations, so the impact hypothesis and its relationship with the formation of the Roseau structure are viable.

ii

TABLE OF CONTENTS

LIST OF ILLUSTRATIONS 1. INTRODUCTION 2. BACKGROUND 2.1. CONTEXTUALIZING THE YD EVENT: TIMING AND POSSIBLE TRIGGERS 2.2. THE IMPACT HYPOTHESIS AND THE CONTROVERSY 2.2.1. THE PHYSICAL AND GEOCHEMICAL EVIDENCE 2.2.2. ONSET OF THE YD AND TIMING OF CULTURAL TRANSITIONS AND 2.2.3. THE GEOMORPHOLOGICAL EVIDENCE 2.3 ARGUMENTS AGAINST ASTEROID IMPACT HYPOTHESIS 2.4 GENERAL RIDICULE 3. GLACIAL AND POSTGLACIAL HISTORY WITH EMPHASIS ON LAKE AGASSIZ 3.1 GLACIAL HISTORY 3.2 A BRIEF HISTORY OF LAKE AGASSIZ AND THE YD 3.3 STAGES OF LAKE AGASSIZ 4. DESCRIPTION AND SETTING OF THE ANOMALOUS STRUCTURE 4.1 GLACIAL ORIGIN 4.2 LACUSTRINE ORIGIN 4.3 IMPACT ORIGIN 5. DISCUSSION 6. CONCLUSIONS REFERENCES

iii

LIST OF TABLES

Table 1……………………………………………………………………………………8

Table 2……………………………………………………………………………………16

Table 3……………………………………………………………………………………33

Table 4……………………………………………………………………………………35

iv

LIST OF FIGURES

Figure 1. Lake Agazzis drainage routes………………………………….44

Figure 2. Lake Agazzis strandline point elevation data………………....45

Figure 3. Lake Agassiz lake-level hydrograph…………………………...46

Figure 4. Roseau Structure………………………………………………...47

Figure 5. Gravity data……………………………………………………….48

Figure 6. Carolina Bays……………………………………………………..49

v

1. Introduction

Few events have received as much attention from the scientific community as the

Younger Dryas. As climate was rapidly warming at the end of the last Age, this climatic anomaly plunged the Earth back into glacial conditions from 12.9 - 11.7 cal ka BP. The abrupt end of the Younger Dryas marks the beginning of the Age and the transition from the to the . The causes of this climatic event have been the subject of heated debate, and hypotheses have evolved rapidly.

Mangerud (1987) argues that an event of this magnitude requires a global trigger such as changes in ocean circulation. In addition, given the importance of the North Atlantic currents in terms of thermal equilibrium of climate, it is sensible to think that the YD event did not exclusively affect Europe; therefore it had an impact in other parts of the world, particularly in North America. This idea is reinforced by Broecker et al. (1989) suggests, through analysis of marine oxygen isotope records, that Lake Agassiz in North America had a flow of melted water to the north, which considerably reduced the salinity of the

North Atlantic, resulting in a variation in its circulation pattern and consequently a variation in the climate of the . Teller et al. (2002) show that not only large amounts of water were released from Lake Agassiz into the North Atlantic once Laurentide

Ice Sheet (LIS) began its recession. Perhaps one of the most controversial hypotheses is that raised by Firestone et al. (2007a) in which it is stated that the start of the YD is due to the impact of an asteroid with the Earth's surface. This hypothesis is based on the observation of charcoals, nanodiamonds, fullerenes, spherules, iridium anomalies, among others, in the sediments associated with Lake Agassiz.

The research reported by Firestone et al. (2007a) summarized evidence from 50+ Clovis- age archaeological sites in North America that is consistent with an extraterrestrial impact.

6

Common to all of these sites is an organic layer dating to 12.9 cal ka BP that corresponds to the of Pleistocene and marks the end of . Contained within this layer are magnetic grains with elevated iridium concentrations, magnetic microspherules, charcoal, carbon spherules, nanodiamonds, and fullerenes with extraterrestrial (ET) helium signatures. They suggest that all of these lines of evidence are consistent with an ET impact and associated biomass burning (Firestone et al.,

2007a). This assertion has been met with cautious support (Israde-Alcantara et al. 2012;

Haynes, 2008; Kennett et al., 2009), arguments against (Paquay et al. 2009; Boslough et al. 2012; Buchanan et al. 2008; Wu et al. 2013; LeCompte et al. 2012), and general ridicule

(Surovell et al. 2009; Pinter et al. 2011).

We herein describe a circular feature in northern Minnesota that has characteristics of an impact crater. In addition, the location of this feature constrains its formation to within error of 12.9 cal ka BP, and we evaluate possible sets of conditions that would allow a feature of this size and characteristics to result from a bolide impact. The feature is located a few km north of the Beach of Lake Agassiz in Roseau County, Minnesota. Because of its location, the feature is hereafter referred to as The Roseau Structure, or simply the

Structure. The Structure is a circular shallow depression with concentric rings on the NE side. The concentric rings have been wave washed during rising and falling stages of

Glacial Lake Agassiz. The location of the Structure constrains its age to after ice retreat from this area but before the drainage of Lake Agassiz. Lake Agassiz occupied the Tintah

Beach at an elevation of 359 meters until 12.9 cal ka BP at the beginning of the Younger

Dryas. At that time, the lake level dropped as the outlet of Lake Agassiz switched from the southern outlet to a lower outlet to the east (Teller, 2013) or northwest ( et al.,

2009) or neither (Lowell et al., 2013). By 11.7 cal ka BP Lake Agassiz rose again to the

Campbell Beach.

7

2. Background

2.1. Contextualizing the YD event: Timing and possible triggers

The refers to the time of maximum extension of the ice sheets during the last , approximately 21,000 BP (Table 1). Global climate was generally warming and environmental conditions in Western Europe began to resemble current conditions ca. 15,000 cal yr BP. The Younger Dryas (YD) was a brief phase (~1,300 ± 70 years long) of climate cooling in the late Pleistocene, between 12.9 and 11.7 cal ka BP (Broecker, 2010; Carlson, 2010; Lowell, 2013 ). In Blytt-Sernander climate theory, the Younger Dryas succeeds the Bölling / Allerød interstade, and precedes the Lower Holocene period. This event takes its name from the alpine flower , an arctic-alpine flower plant in the Rosaceae family.

Few other climate anomalies have been as widely studied as the YD, as this event marked the last stadial at the end of the most recent glacial period. Numerous investigations around the proposal of hypotheses that allow us to account for the event or events that could have contributed to the termination of the YD.

Timing last glacial maximum Date (ka cal BP)

Oldest Dryas (stadial) 18.5- 14.7

Bølling (interstadial) 14.7–14.1

Older Dryas (stadial) 14.1 – 13.9

Allerød (interstadial) 13.9 – 12.9

8

Younger Dryas (stadial) 12.9 – 11.7

Pre (interstadial) 10.2 - 8

Table 1. Timing of the main stadial and interstadial periods during the Last Glacial

Period

In order to contextualize the Young Dryas, here we outline the hypotheses put forward for this event. The YD has received great attention from the scientific community, and even within the press and broader public. The causes that produced this event are hotly debated and the hypotheses related to its origin are diverse and controversial.

Late Glacial climate oscillations, including the Allerød and Younger Dryas, were identified in the Paleoenvironmental record as early as the investigation of Hartz and Milthers

(1901), and Flohn (1979) explores the time scales and causes of abrupt paleoclimate events. In an effort to explain YD cooling, Mangerud (1987) summarized the possible causes of the Younger Dryas based on evidence collected in , the , and

Svalbard. He tabulates these hypotheses as: 1) Changes in the North

Current and atmospheric circulation that are dynamically related, 2 ) changes in the directions of the North Atlantic Ocean Current, 3) interactions of the North Atlantic Current with the Norwegian Sea, resulting in cooling of surface and intermediate depth water, 4) major climatic changes in Europe and the Atlantic North added to small climatic changes in other parts of the world, 5) and geographical factors that contributed to a greater amplitude of climate change.

However, thoughts on the cause of YD cooling evolved rapidly, and one of the hypotheses with the greatest amount of glaciological, geomorphological and sedimentary evidence, postulates that the North Atlantic region during the YD, is related to the diversion of glacial

9 meltwater from the Mississippi River/ to the St. Lawrence drainage, which flows into the North Atlantic Ocean (Rooth 1982). This hypothesis, fully developed by

Broecker et al. (1989), maintains that the great discharge of fresh water contributed to the

North Atlantic Ocean, resulting in a low-salinity surface layer over the North Atlantic. The cold, largely freshwater cap suppressed the formation of North Atlantic Deep Water producing a decrease in temperature and consequently a cooling of the North Atlantic region (Teller et al., 2002). Lake Agassiz, the largest glacial lake in North America during the last , is known to have dropped below the level of its southern outlet to the

Mississippi River at the beginning of the Younger Dryas (Fisher, 2003). Although the causes remain a subject of study, it is clear that not only the cooling of the North Atlantic region was affected by changing ocean circulation patterns, but also by large-scale atmospheric changes (Bakke, 2009).

Firestone et al. (2007a) suggest that a large bolide broke into multiple pieces creating multiple airbursts, possibly with some surface cratering. This event at 12.9 cal ka BP is hypothesized to have caused the Younger Dryas cooling Moore et al. (2020) has reported the presence of meltglass, nanodiamonds, charcoal, and microspherules in Abu Hureyra

(AH), Syria, which coincides with the Younger Dryas Boundary (YDB). This evidence correlates quite well with the possible impact of an asteroid with Earth 12.9 k and B.P., which would be in agreement with the hypothesis by Firestone et al. (2007a).

2.2. The Impact Hypothesis and the Controversy

The onset of the YD is correlated in time with a significant climate change, but also with the collapse of the and major changes in behavior such

10 as the origins of agriculture (Harris, 2003; Munro, 2003; Wright et al., 2003) and dramatic changes in cultural traditions (Waters and Stafford, 2007; Haynes, 2008; Haynes et al.,

2007; Sweatman and Tsikritsis, 2017). Therefore, the “Impact Hypothesis” (Firestone et al., 2007a), has received a great deal of scrutiny, which has come in the form of both support and dismissal.

2.2.1. The physical and geochemical evidence

More than 50 locations in North America have been reported in which a layer of dark clays with organic matter content and carbonaceous silts are observed that coincides stratigraphically with the onset of the YD at 12,900 cal ka BP (Firestone et al., 2007a).

Haynes (2008) reports that two-thirds of the geoarchaeological sites in North America that have records spanning the Pleistocene/Holocene transition are associated with “black mats” that coincide with the beginning of the YD. The mats are interpreted to relate to rapid cooling and mark the end of the land mammal age (Haynes, 2008).

Kennett et al. (2009) report the presence of shocked nanodiamonds in the YD boundary sediments, which are associated with charcoal, carbon spherules and soot consistent with biomass burning. The presence of microdiamonds can be correlated with events in which high pressures and high temperatures occur. These types of events correlate well with impact metamorphism with pressures ranging from 80-1000 kilobars and temperature ranges reaching 3000° C. Similar evidence supporting a YD impact has been reported from central Mexico (Israde-Alcantara, 2012), southern (Pino et al., 2019), and Abu

Hureyra, Syria (Moore et al., 2020).

11

Gabrielli et al. (2004) report large concentrations of iridium and platinum have been found in the GRIP () at the YDB. Similar evidence has been reported by

Moore et al. (2020) in Abu Hureyra (Syria), among which the presence of glass with a low content of magnetite and a very low percentage of water, similar to that of tectites and inconsistent with volcanic (obsidian) or anthropogenic events, stands out (French and

Koeberl, 2010). Many places in North America, Europe, and particularly in Syria (Abu

Hureyra), the presence of nanodiamonds, deformed and melted quartz grains, carbon and high contents of iridium, platinum, nickel and cobalt, elements that are present, have been reported in asteroids and meteorites in high concentrations. In particular, Moore et al.

(2020) reports that about 40% of the glass found in Abu Hureyra, shows traces of siliceous plants as well as carbon infusions, which indicates that they were formed at temperatures ranging between 1200 ° C - 1300 ° C, based on laboratory tests. In addition, the presence of magnetite, melted quartz, and chromferide in the glass found in Abu Hureyra, indicate minimum formation temperatures of 1720° ˚C, possibly reaching 2200° C and higher.

The suggestion by Firestone et al. (2007) of an extraterrestrial YD impact has renewed interest and reinterpretation of the Carolina Bays (Melton and Schriever, 1933) and high plains playas as possible impact structures (Firestone, 2009). As reported by Firestone et al. (2007a) the presence of microspherules at various sites in North America, which are enriched with titanomagnetite, at the YD boundary at 10 sites corresponding in age to

Clovis. In addition, similar titanomagnetite anomalies, with modest enrichment of nickel as well as Iridium, in sediments in 15 Carolina Bays at the YD boundary support the impact hypothesis (Firestone et al., 2007a). Firestone et al. (2007) indicates that in the 15

Carolina Bays, as well as in 8 of 9 Clovis age sites, there are peaks of presence of charcoal. Similarly, in 13 of the 15 Bays and in 6 of 9 Clovis-age sites that correspond to the YDB, spherules with the presence of charcoal were found. Additionally, they report

12 that these spherules contain nanodiamonds and fullerenes, which are extremely rare on

Earth but are found in some meteorites and impact craters (Guilmour et al., 1992; Kennett et al., 2009a, 2009b; French and Koeberl, 2010). Firestone et al. (2007a) reports sediments that belong to the YDB, have fullerenes with the presence of ET helium, which are commonly associated with meteorites and impact structures resulting from the collision of comets or asteroids with the Earth's surface (French and Koeberl, 2010). At the Carolina

Bay and other YDB study sites mentioned by Firestone et al. (2007a), several fragments with glassy textures were found, which suggests melting of the material. Furthermore, some nanodiamonds were found at these locations, being these associated to asteroid impacts as reported by Gilmour et al. (1992).

2.2.2. Onset of the YD and timing of cultural transitions and extinctions

The YD onset is correlated in time with the Clovis/Folsom cultural transition. The youngest

Clovis sites correlate to the beginning of the YD, however, Surovell et al. (2016) find that the Folsom complex dates to at least 100 years after the beginning of the YD; they acknowledge their chronology is complicated by small sample size. Therefore, Surovell et al. (2016) conclude that there is little correlation between the start and end of the Folsom complex and the onset and termination of the YD. According to Faith and Surovell (2009), the cause or cause/causes of Pleistocene megafauna collapse has/have been debated, however, the timing is correlated closely with onset of YD cooling. One hypothesis establishes that Pleistocene megafaunal extinction may be attributed to global climatic changes associated to rising CO2 levels as well as elevated temperatures and precipitation, prompted a transition from a nutrient accelerating mode to a nutrient decelerating mode in North America.

13

Although, Gill et al. (2009) present evidence that the collapse of the Pleistocene

Megafauna occurred several hundred years before the onset of the YD, which is in agreement with the idea that causal relationships between novel plant communities associated with the last in North America and megafaunal extinctions are still unclear.

Based on studies held by McDonald et al. (2006) and Schulte et al. (2010) other extinction events have been correlated with the impact of asteroids. Such is the case of the extinction that occurred in the late period in which approximately 75% of all the species on the planet became extinct. Additionally, the wide disappearance of some mammalian genera suggests that the cause of their extinction was abrupt and not gradual, as suggested by the hypotheses of climate cooling or human overkill (Firestone et al., 2007a).

2.2.3. The geomorphological evidence

Geological and geomorphological features have been cited as supporting evidence for a

YD impact. Characteristics of impact structures may resemble other geological structures; the most common of all is the circular or circular-like shape, circular deformation patterns, magnetic anomalies, among others. For example, the Carolina Bays (Firestone et al.,

2007a) and the playas of the southern High Plaines (Firestone, 2009; Firestone et al.,

2010) are interpreted to be of possible impact origin.

The description made by Prouti (1952) about the Carolina Bays, serves to have a general panorama about the structures displayed on the area, and I quote: “The terms "bay" and

"pocosin" as generally used in the Atlantic Coastal Plain refer to an area usually overgrown

14 with swamp vegetation and underlain by peat deposits. In some areas shallow lakes, with peat-filled border swamps, are also classified as bays”

According to Firestone et al. (2010) the Carolina Bays are made up of approximately

500,000 depressions with elliptical shapes that are distributed in North America, more specifically from the Atlantic coast in New Jersey to the state of Alabama. They are characterized by having depths of up to 15 meters and dimensions ranging from 50 m to

10 km in length. In the description made by Firestone (2009) and Firestone et al. (2010), the major axes of these elliptical structures are oriented to the northwest and the beach rim had the typical marks associated with YDB, including fullerenes, nanodiamonds, carbon spherules, charcoal, Soot, iridium, and glass-like carbon. As it is reported by

Melton and Schriever (1933) the formation of the Carolina Bays was originally ascribed to asteroid impacts because its uniqueness. The bays in this area are characterized by having a wide distribution and presenting elliptical shapes, parallel alignment of their major axes and high edges. This leads Melton and Schriever (1933) to suggest that rare geological processes, such as the impact of various asteroids on the Earth's surface, as a result of the disintegration of a larger asteroid in the atmosphere, could have been responsible for the formation of these structures Nonetheless, the lack of evidence regarding to meteorite material associated to these structures, suggests that other geological processes, such as eolian, marine, dissolving action of ground water, volcanism, glaciation, and submarine, might be the main cause (Melton and Schriever,

1933) .

Similar shaped basins have tentatively been identified in , , Kansas, and Nebraska (Kuzilla, 1988), but are far less abundant than the Carolina Bays. Analysis of the long axes of these oval-shaped basing indicated they are preferentially aligned

15 toward the Great Lake region, suggesting a potential impact in that region (Firestone et al., 2009).

Larger impact structures are known to be associated with other geological processes such as diapirs, tectonic deformation due to magma rise, or volcanic activity (French and

Koeberl, 2010). In such a way, in order to correctly identify and classify an impact structure, the main data related to the impact is metamorphism such as described by

Wolbach et al. (2018a, 2018b) and French and Koeberl (2010). Such characteristics are summarized in table 2.

A. Characteristic regarding to shock metamorphism and impact cratering

1. Meteorite fragments preserved

2. Chemical and isotopic projectile signatures

3. Mesoscopic features - Shatter cones

4. High-pressure (diaplectic) mineral glasses

5. High-pressure mineral phases

6. High-temperature glasses and melts

7. Planar fractures (PFs) in quartz

8. Planar deformation features (PDFs) in quartz

B. Non-diagnostic features produced by meteorite impact and by other

geological processes.

1. Circular / circular-like morphology

2. Circular structural deformation

16

3. Circular geophysical anomalies

4. Fracturing and brecciation

5. Kink banding in micas

6. Mosaicism in crystals

7. Pseudotachylite and pseudotachylitic breccias

8. Igneous rocks and glasses

9. Spherules and microspherules

10. Stratigraphic criteria – inverted layers

11. Other problematic criteria

Table 2. Shock-produced deformation effects: diagnostic and non-diagnostic. Modified from Holocombe et al. (2001) and French and Koeberl (2010).

2.3 Arguments against asteroid impact hypothesis.

Evidence for an extraterrestrial impact at the lower boundary of the YD still being highly debated. There are a number of studies that cast doubt on the impact hypothesis.

Regarding the "YD event", Firestone et al. (2007a) present an unusual variety of evidence outside the criteria accepted by the impact community. Reimold, 2007 points out that impact structure report should be reviewed in detail, as it could be compromised, incomplete or incorrect. Eventually, this information may be widely disseminated to the

17 scientific community and the general public, which could impair reliability in the affected databases where other identified impact structures are recorded.

Firestone et al. (2007a), reports iridium concentration corresponding to 117 ppb and another of 51 ppb were from separate magnetic beads and spherules, fully compatible with a non-catastrophic source of micrometeorites (Pinter and Ishman, 2008; Boslough et al., 2013).

Despite the broad acceptance of iridium anomalies because extraterrestrial impact after

Alvarez et al. (1980), authors such as Dyer et al. (1989) have found further evidence to explain iridium anomalies on sedimentary rocks, including claystones and shales. Actually, the authors found that a variety of common microorganisms found worldwide, which include heterotrophs, photoautotrophs, and chemoautotrophs, might affect the chemistry of iridium by concentrating or dissolving it. Additionally, the geochemistry of iridium must be evaluated from a microbial perspective given the permanent biological activity on sedimentary rocks that might erase, enhance or create iridium anomalies.

Iridium anomalies have also reported by Coveney et al. (1992) and Sawlowicz (1993) on shales and other sedimentary environments. According to their results, precious metals in shales, particularly on shales in China, are likely not of extraterrestrial origin given the abundances of PGE and volatile elements (Mo and Se), which differ greatly from calculated values on metallic meteorites. In fact, Sawlowicz (1993) indicates that it is possible to use the abnormal antibody values of PGE and associated elements in order to identify contributions of specific processes such as 1) extraterrestrial, 2) volcanogenic, 3) precipitation from seawater, 4) microbial, 5) exhalative-hydrothermal, 6) leaching, and 7) transport and precipitation at a redox boundary.

Studies about major cultural changes and Paleoindian population decline are reported by

Buchanan (2008), showing that simulation experiments detect that, even at high site

18 destruction rates, if Paleoindians had suffered a population bottleneck, they would be visible in the summed probability distribution of the calibrated Paleoindian radiocarbon dates. Therefore, their results indicate neither Firestone et al. (2007a) hypothesis is valid nor migration to south by Paleoindians because of an asteroid impact.

The set of high-quality radiocarbon dates in the Folsom and Clovis complexes used by

Surovell et al. (2016) show a revised age range for the Folsom complex and examine its correlation with Younger Dryas, the end of Clovis, and megafauna extinction. Surovell et al. (2016) show that to refine the ages of the Folsom complex, it is best to reduce the complete set of and thus obtain more accurate dates of Folsom occupation. With the data collected and its subsequent analysis, Surovell et al. (2016) reveal that in addition to the temporal correlation between the YD and the Folsom complex, the duration of the cooling of the North Atlantic occurred between 12,850 and 11,650 years, with which in these 1200 years it is difficult to reconcile times of disappearance of

Folsom and Clovis complex; additionally, the authors show that the oldest cultural evidence dated from the Folsom complex, corresponds to 240 years after the start of YD.

Despite what was shown by Steffensen et al. (2008), which establishes a clear analysis on the survival of hunter-gatherers to the changes associated with YD, there is no clear correlation between climate change and strong cultural changes in the Folsom complex.

Waters and Stafford (2007) published a revised timeline for the complex showing that dates in Clovis appeared to be highly clustered in time, and that the duration of the Clovis complex was considerably shorter than previously believed, results that do not differ from those obtained by Surovell et al. (2016)

Paquay et al. (2009) report the presence of elements of the platinum group (PGE: Os, iridium, Ru, Rh, Pt, Pd), which together with the presence of gold particles (Au) and the proportions of 187Os / 188Os in terrestrial, lacustrine sections and marine results indicate

19 that the iridium values reported by Firestone et al. (2007a) do not coincide. Furthermore, the isotopic ratios of 187Os / 188Os in the sediments are similar to the average cortical values, indicating a non-extraterrestrial origin.

Boslough et al. (2012), propose that despite the existence of evidence that supports the asteroid impact hypothesis and its relation to the YD onset, there are not only one but several impact hypotheses that are difficult to reconcile between them. As the authors mention, the main reasons against the impact hypothesis include the lack of solid evidence showing an impact structure, as pointed out by French and Koeberl (2010), lack of temporal correlation between the supposed impact and the climatic, paleontological and archaeological changes and the failure to reproduce any of the results obtained by the authors of these hypotheses.

Reimold et al. (2007, 2014) emphasize the relevance of reviewing existing databases reporting impact craters around the world, as some of these structures do not appear to meet the basic requirements to be classified as impact structures. Likewise, the authors attribute to the lack of careful review of the structures associated with impact craters, misunderstandings not only in the academic community but in the media. Specifically in the case of the YD event, the lack of evidence of a structure that supports the reports described by Firestone et al. (2007a), makes the impact hypothesis less and less solid.

LeCompte et al. (2012) find that in three of the places reported by Firestone et al. (2007a), there are spherules deposits with chemical compositions similar to those also reported by

Firestone et al. (2007a). Despite reproducing similar results, LeCompte et al. (2012) do not assure that it is indeed material produced by the impact of an asteroid 12,900 years ago.

One of the most relevant evidences provided by Firestone et al. (2007a) to support the impact hypothesis is related to the presence of spherules found in YDB together with the

20 iridium anomaly. Wu et al. (2013) interpret the Pt/iridium relationship data found in the

Greenland Ice Sheet Project (GISP-2) and Greenland Ice Core Project (GRIP) as of terrestrial origin. Likewise, the low isotopic ratios of 187Os/188Os could be attributed to volcanic processes, although the authors do not rule out a possible extraterrestrial origin.

Despite findings reported by Wu et al. (2013) supports Firestone et al. (2007a) hypothesis, it is hard to conceal how impacts were part of a number of other impacts worldwide associated to spherules with low enrichment in PGEs.

In their analysis of platinum group elements (PGEs) and gold from continental YD sections, Paquay et al. (2009) find no evidence of elevated iridium concentrations. In addition, osmium isotopic ratios of the sediments are reported to be similar to Earth average continental crust (Paquay et al., 2009). The iridium mass concentration reported by Firestone et al. (2007a) indicates a peak of just 2.3 - 3.8 ppb and even maximum iridium values were below this amount at 10 different locations. This forces the authors to argue that the impactor had abnormally low initial Go values.

2.4 General ridicule

Discussion held by Pinter et al. (2011), describes the differences between the types of physical evidence that has been associated with the impact process. This can be divided into two large groups: the first argues that given the evidence widely rejected by the scientific community, it should not even be a matter of discussion (high concentrations of iridium, fullerenes enriched in 3He, among others). The second group is characterized by the extensive study of evidence that has been the subject of recent research, including magnetic grains, nanodiamonds, among others.

21

Pinter et al. (2011) point out that under the criteria specified by Koeberl (2007) and French and Koeberl (2010), it is considered as unequivocal evidence of impact craters the presence of shock-metamorphic effects and, in some cases, the discovery of meteorites, or traces of them. At the macroscopic scale, the presence of shatter cones and at the microscopic scale, characteristics of quartz fracturing called Planar Deformation Features

(PDF), are also evidence that corroborate the impact of an asteroid on the Earth's surface.

However, as Pinter et al. (2011) reports, none of these characteristics has been reported by Firestone et al. (2007).

Despite showing a marked difference between the evidence that has been shown has no relation to the impact hypothesis, there is another considerable amount of evidence that is worth reviewing, as it could constitute an advance in the confirmation or discard of the hypothesis proposed by Firestone et al. (2007a). There is no doubt that works should focus on verifying whether such evidence, which according to Pinter et al. (2011) would belong to the second group of tests, they could have a temporal and causal relationship of the beginning of the YD, with the impact of an asteroid 12,900 years ago.

Regarding the catastrophic event of the impact and its relationship with the YD, evidence is presented related to the presence of foraminifera in chevron-like folds and as splashes in oceanic cores associated with impact gaps. As reported by Abbott et al. (2008) CaCO3 melts are documented for the Chicxulub impact site, but these are a far cry from the "well- preserved carbonate microfossils" encased in silicate melt. Other authors specify that the presence of silica and the associated carbonate in Chicxulub, occur as "molten carbonate particles", recrystallized and decomposed calcite.

Pinter and Ishman (2008) show that recent experiments prove that it is possible to find intact grasses in silica melts as a result of impacts. However, the summary by Harris and

Schultz (2007) suggests only one mechanism in which potential grass debris is preserved

22 in impact gaps. Fully charred plant remains are documented in the geological record (e.g.

Scott, 2008), but few researchers strongly accept that unaltered plant material or the presence of foraminifera have been in direct contact with impact molten material. As mentioned by Pinter and Ishman (2008), the comments made by Firestone et al. (2007) reiterates the relationship between an impact event in North America 12.9 ka B.P. based on evidence found at 50 Clovis age sites. Due to the absence of a surface impact crater,

Firestone et al. (2007a) and Firestone et al. (2007b) report the presence of carbon such as glass, nanodiamonds and spherulites that may be associated as by-products of the impact. Furthermore, they report the presence of micrometeorites lodged in tusks (Firestone et al., 2007b).

Granulometric parameters indicate that at the end of the YD event, there was a gradual retreat of the . This added to the distribution of lake sediments and water content, show an immediate response to the absence or presence of a glacial mass (Bakke, 2009), so an extraterrestrial impact would not explain the retreat of said glacier in such an extensive time.

3. Glacial and Postglacial History with Emphasis on Lake Agassiz

3.1. Glacial History

Northern Minnesota and adjacent were extensively glaciated during the

Pleistocene, however, the modern landscape was the result of with the Last Glacial Maximum culminating about 21,000 cal years BP. The general glacial history of northwestern Minnesota is summarized by Wright (1972) and detailed glacial

23 history as it pertains to this study is summarized by Fenton et al. (1983), Mickelson et al.

(1983), Clayton et al. (1985), and Harris (1995). Ice advanced to its maximum limit along the southwestern margin of the by about 27-30,000 cal years BP

(Mooers and Lehr, 1997). The glacial history is complicated by repeated surging of the southwestern margin (Clayton et al., 1985; Lusardi et al., 2011), however, the final ice advance into northwestern Minnesota culminated at 11,700 yrs BP (13,500 cal yrs BP)

(Fenton et al., 1983; Fisher, 2003). Ice then retreated to the north of the continental divide in western Minnesota that separates drainage to the Mississippi River from drainage to the north to Hudson Bay. At this point Lake Agassiz was formed between retreating ice and the divide.

3.2. A brief history of Lake Agassiz and the YD

One of the largest lakes that has ever existed in North America was Agassiz Glacial

Lake. The lake initially formed about 13,500 years ago and expanded as ice retreated north of the continental divide in western Minnesota (Fenton et al., 1983). At that time, meltwater was ponded between the retreating Laurentide Ice Sheet and the divide.

Throughout its existence, the lake varied in size depending on the position of the and the availability of drainage outlets. At its greatest extension, the lake was

1127 km long and 402 km wide, extending from South Dakota to Canada. Most of the glacial ice had completely melted 8,000 years ago and the giant lake drained north into the newly opened Hudson Bay. Lake Agassiz drained south through Glacial River

Warren along the course of the Minnesota River to the Mississippi and to the Gulf of

Mexico (Teller and Clayton, 1983).

24

One of the explanations that has been given to Late Pleistocene cooling of the YD approximately 12,900 - 11,700 cal yrs BP, is a change in the North Atlantic meridional overturning circulation (MOC) caused by the discharge of large amounts of fresh water from Lake Agassiz. There is broad support that the YD resulted from a weakened MOC, which in turn reduced heat transport from the north and warmed the , although the relationship between causality and temporality remains a hot debate (Broecker et al., 1989; Teller et al., 2002; Breckenridge, 2015). According to

Breckenridge (2015), the main source of doubt is the uncertainty regarding the routing of the Lake Agassiz overflow at the start of the YD. During the YD, it has been proposed that Lake Agassiz drained both to the Arctic Ocean (Fisher et al., 2009; Murton et al.,

2010) through a northwest outlet, and to the Atlantic Ocean through an eastern outlet

(Fenton et al., 1983; Broecker et al., 1989; Teller et al., 2002) (Figure 1 from Murton et al., 2010). Other evidence suggests that the northwest and east outlets remained covered by the Laurentide Ice Sheet at the beginning of the YD, and that the increased flow of fresh water into the Atlantic Ocean was absent (Lowell et al., 2013).

3.3. Stages of Lake Agassiz

Multiple stages, or lake levels, are recorded by strath terraces along River Warren and associated beaches in the southern part of Lake Agassiz (Matsch, 1983; Breckenridge,

2015). However, concurrent isostatic rebound of the lake basin resulted in the formation of extensive beach complexes in the northern part of the Agassiz basin (Fenton et al.,

1983; Breckenridge, 2015). Near the southern outlet, four major strandlines are recognized. These four major lake levels also define four major stages of Lake Agassiz

(Fenton et al., 1983). The earliest phase of Lake Agassiz is the Cass Phase and is

25 associated with the highest major strandline of Lake Agassiz, the Herman level (Figure 2 from Breckenridge, 2015).

As the lake expanded in area Lake Agassiz merged with Glacial Lake Koochiching

(Nikiforoff, 1947) ushering in the Lockhart Phase of Lake Agassiz (Fenton et al., 1983).

From the high stand at the Herman level, the Lake level dropped in stages, first to the

Norcross level and then to the Tintah level between about 13,500 cal BP to about 13,000 cal BP (Fisher, 2003; Breckenridge, 2015). Water then dropped from the Tintah level to the Moorhead low-water phase (Teller, 1983), and ceased to drain through the southern outlet and on to the Gulf of Mexico. The timing of this drop is constrained by numerous radiocarbon and optically-stimulated luminescence (OSL) dates, however, associated error places the beginning of the Moorhead phase about 13,400 – 12, 600 cal BP. This drop has been correlated to the beginning of the Younger Dryas at 12,900 cal yrs BP.

At this time, it is suggested that Lake Agassiz either drained eastward through the Great

Lakes to the Atlantic Ocean (Broecker et al., 1989; Teller et al., 2002), NW to the Arctic

Ocean (Fisher et al., 2009), or neither (Lowell et al., 2013). Investigations by Fisher et al.

(2009) and Lowell et al. (2009), involving extensive radiocarbon dating of the northwestern and eastern outlets of Lake Agassiz, respectively, indicate that both potential outlets are too young. Breckenridge (2015) also concludes that the NW outlet could not have been open at that time. In the absence of a well-documented eastern or northwestern outlet for

Lake Agassiz, Lowell et al. (2013) suggest that the basin may have become hydrologically closed, although radiation balance and meteorological calculations over the retreating

Laurentide Ice Sheet suggest hydrologic closure is unlikely (Mooers, unpublished data).

Geochemical analysis of sediment cores near Fargo, North Dakota, did not show evidence of enrichment by evaporation at the top of the Brenna Formation associated with the

Moorhead low-water phase (Lowell et al. 2013; Liu, et. Al. 2014).

26

Although new and reanalyzed data from the Agassiz basin has emerged to restrict lake paleogeography and meltwater routing history (Fisher, 2003, 2007; Lowell et al., 2005;

Teller et al., 2005; Lepper et al., 2007), there is uncertainly of the absolute dates of the drop of Lake Agassiz to the Moorhead low-water phase, within error the Moorhead phase coincides with the onset and termination of the YD.

Given the coincident timing of the beginning of the Moorhead low-water phase and the

YD, Broecker et al. (1989) and Teller et al. (2002) suggest that the diversion of LIS meltwater from the Mississippi and Gulf of Mexico to the St. Lawrence and the North

Atlantic resulted in the changes in ocean circulation that triggered YD cooling. This conclusion is supported by the work of Levac et al. (2015) among others.

The Moorhead phase ended when the level of Lake Agassiz rose to the Campbell level

(Liu et al., 2014; Breckenridge, 2015) by about 11,500 cal yrs BP (Liuk et al., 2014; Fisher,

2003; Lowell et al., 1999). It has been speculated that the closure of the eastern outlet and the end of the Moorhead low-water phase was related to the advance of the Superior

Lobe in the Basin at 11,500 cal yrs BP (9900 BP (Hughes et al., 1978;

Lowell et al., 1999). Liu et al. (2014) (Figure 3) show the range of radiocarbon dates and their errors associated with the Moorhead low-water phase.

4. Description and setting of the anomalous structure

The anomalous structure is located in northern Minnesota, at 48 ° 55’02.22” N, 95°

52’14.08” W close to US-Canada border (Figure 4). The structure is located 8 km north of the Campbell beach (average elevation 331 meters) and 34 km north of the Tintah beach (average elevation of 361 meters). This circular-like structure has a central basin

27 lying 1-2 meters below the surrounding landscape. It is oval shaped, 10.7 km (major axis oriented predominately to the northeast) by 8.12 km (minor axis) with an average elevation of 313 m.a.s.l. The structure would be located 48 meters below the Tintah beach and 18 meters below the Campbell beach. The northeastern margin of the

Structure is characterized by a series of ridges with a spacing ranging from 72 to 155 meters; these ridges display a concentric pattern as shown on Figure 4. The ridges are 1 to a maximum of 2 meters high and are composed of silt loam to fine sandy loam soils with organic soils (peat and muck) in low swales between ridges (Web Soil Survey,

USDA).

The anomalous Structure described above is the only one of its kind in the region. It is important to now address hypotheses for its origin. Given its setting and the geological history of the landscape, the Structure must be of glacial or lacustrine origin, with the only alternative being of extraterrestrial origin.

4.1. Glacial origin

The circular-like structure may be interpreted as a result of glacial processes. These processes include subglacial meltwater flow, erosion by ice scour, formation of a lake related to melt out of buried ice, or ice marginal stagnation processes (Benn and

Evans, 2010). The presence of subglacial channels that can reach the surface through incisions, have cuts to the glacial bed or a combination of both has been reported by

Röthlisberger (1972). There are several examples of reaches of incised channels up and down in glacier beds, sometimes within the same system. There is evidence that branched channel networks occur beneath glaciers (Kamb, 1987). Subglacial channels

28 would have circular to semicircular shapes in cross section; however, the studied circular-like structure does not correspond to a cross section of a subglacial channel first because it is not associated with a glacier and second because the circular shape is in plain view and not in cross section as described by Röthlisberger (1972).

Another possibility is the presence of pits or pockmarks whose formation may be associated either by expulsion of fluid through the lake floor (pockmarks), or by the temporary grounding of an iceberg through tidally-influenced changes in lake level (pits)

(Brown et al. 2017). According to Brown et al. (2017), the pits associated with icebergs can be found in different types of sediments exposed on the surface either during glacial or post-glacial conditions. Its formation’ occurs when, in the case of sea, icebergs eventually hit the substrate either to changes in water depth or buoyancy. In the case of

Lake Agassiz, the circular-like structure, could be the fall of a fragment of the glacier of considerable size that left an imprint on the substrate. Other authors such as Bass and

Woodworth-Lynas (1988) and Newton et al. (2016) mention that both pits and structures associated with ice scour with semicircular geomorphologies have been reported on the seabed and whose causes have been reported by Brown et al. (2017). Based on satellite images near to the location of the structure, it is possible to determine that the studied structure is isolated from other circular-like structures. Spectral bands combination is suggested in order to have a better approach to identifying similar structures as well as GPR. Subglacial erosion can create overdeepenings on the glacier bed. Moran et al. (1980) describe the erosion of blocks of the glacier bed on a scale similar to the anomalous Structure. In their case, blocks of the glacier bed were detached and displaced some distance downglacier. The elongation is in the direction of ice flow, and the raised ridges are downglacier from excavated depression. In the case

29 of the anomalous Structure, the ridges are to the northeast, or upglacier and are of a very different morphology.

Another example of structures with a similar geomorphology are kettle lakes, such as

Kettle Lake, which lies within the Missouri Coteau (Clayton,1967) or Kettle in

Wisconsin (Carlson et al., 2005). Kettle lakes form when blocks of ice melt from beneath a sediment cover. Typically, kettle lakes form in groups such as most of the lakes of

Minnesota. For the Structure to be an ice block depression, it would be one of the largest in the region. This structure is characterized as resembling a bowl, reaching depths of up to 10 m (Grimm et al. 2011). For example, Lake Mille Lacs in central Minnesota is a broad, shallow ice scoured basin with raised shorelines around its margins (Wright,

1972). However, Lake Mille Lacs has a prominent moraine on the down-ice side of the basin; no such moraine exists anywhere in the vicinity of the Structure. Kettle Lake

(North Dakota) is not the only structure whose geomorphology resembles a bowl, in fact, the images show similar structures in North Dakota and Canada, which is contrasted with the null presence of structures similar to the one in northern Minnesota.

4.2. Lacustrine Origin

Lacustrine processes associated with Lake Agassiz include wave dominated shorelines, strong currents related to glacial meltwater throughput to the Lake, and sedimentation associated with shallow-water and deep-water environments. None of these lacustrine sedimentary environments are known to be associated with circular depressions. The ridges surrounding the anomalous Structure do indeed appear to be wave washed, however, likely related to the fall and rise of the water level of Lake Agassiz.

30

A possibility raised by Ken Harris (written communication, 2020), retired from the

Minnesota Geological Survey, is that the anomalous Structure is an ice-walled lake plain. Ice-walled lake plains can be of just about any size. Larson et al. (2016) describe ice-walled lake plains formed in thin stagnant ice in northeastern Minnesota. They are characterized by relative flat areas with slightly raised rims, but none are associated with multiple concentric ridges. Ice-walled lake plains require underlying stagnant ice, and it is unclear if there are any mechanisms for the preservation of stagnant blocks of ice in

Lake Agassiz. In addition, this is the only feature of its kind. Ice-walled lake plains are commonly found in groups.

4.3. Impact Origin

Impact craters are classified according to their size as follows: 1) simple craters

(approximately 15 km in diameter), complex craters (15 - 150 km), and impact basins with mound rings (150 - 1200 km) (Melosh, 1980). There are three stages in the formation process of these structures: 1) the compression , impact of the asteroid/comet with the surface of the rocky body ( or planet). At this stage, kinetic energy transfers to the surface, and the shock waves affect both the projectile (asteroid) and the surface. 2) Excavation: the removal of the material that gives way to the crater occurs and the production of the ejection material and 3) the modification stage in which the structure collapses and changes as a result of elastic and gravitational rebounds

(Melosh, 1980).

The compression stage ends once all the material released by the impact covers the meteorite. At this instant, the meteorite penetrates a certain distance into the substrate,

31 ranging from 10-3 to 10-1 s for projectile sizes ranging from 1m - 10m. The substrate material is at rest until the shock wave hits it. Similarly, the projectile continues to penetrate the substrate until the blast wave slows down. In terms of the implicit variables in the compression stage, it is possible to arrive at a solution of Hugoniot's equations for pressure, particle velocity, shock velocity and internal energies of the projectile and the substrate (Melosh, 1980; Shomaeker, 1963 ). It is essential to mention that approximately half or a little more, either of the kinetic or internal energy of the projectile, is transferred to the target, this being one of the main results of the compression process.

Experimentally, Oberbeck (1971) reports that there is a close relationship between the size and shape of the crater as a function of depth. The same explosive charge can create a small crater if it is placed very close to the surface or very deep. However, there is an ideal depth at which such loading can produce the largest craters. This relationship can be expressed as:

1/2 D = L (ρj / ρt ) ,

(1) and indicates explosion effective depth (D) and the diameter of the projectile (L), in equation (1) and the relationship between the density of the impactor (ρj) and the density of the substrate (ρt). If the projectile has a high density, for example a metallic asteroid

(7.21 - 7.71 g cm-3) (McCasuland, 2011), D will be higher if the substrate has a low density. If, on the contrary, a rocky asteroid (3 - 3.30 g cm-3) (McCasuland, 2011) impacts a high-density substrate, then D will be smaller (Melosh, 1980).

Regarding the excavation stage, Melosh (1980) highlights that, depending on the mechanical behavior of the substrate, it deforms in plastic or fragile way (ballistic fragments). Based on the calculations of Melosh (1980), the relationship between the

32 diameter of the impact crater, measured from ring to ring (D) and the thickness of the substrate that tends to deform plastically (h), it is possible to associate the corresponding morphology and structures of the crater as shown in Table 3. Additionally, the crater size is more extensive in terms of depth and diameter than the projectile that creates such a crater at a typical speed of 15 km s-1.

D/h ratio Main characteristics

< 4 Bowl Shaped crater. Simple carter with a depth/diameter ratio of about 0.2

< 4 , > 8 The crater displays a central mound

> 9 The crater displays a set of benches of the top of fragile layers

>> 10 Formation of Concentric craters when the bench occurs high on the crater

wall

Table 3. Summary of the main characteristics of an impact crater in terms of the D / h ratio, according to Melosh (1980).

Ormö and Lindström (2000) outline the morphology of land-target craters. Craters less that about 2-4 km wide have depths about one third to one fifth of the width and rim heights of about 4% of the diameter (Ormö and Lindström, 2000). Land target craters greater that about 4 km are generally more complex because of gravitational collapse and rebound that forms a central uplift (Ormö and Lindström, 2000).

At the Roseau Structure, the water would have been about 50 meters deep and the lacustrine and glacial sediment about 65 meters (about 200 feet) thick over

33 igneous and metamorphic bedrock. In terms of deformation, this yields a plastic layer about 115 meters thick over crystalline bedrock, which provides an ideal scenario for cratering with concentric rims (Melosh, 1980).

However, to evaluate cratering for application to the Roseau Structure there needs to be emphasis on impacts in water, even though the global cratering record is heavily biased toward terrestrial structures. An important exception is the Mjølnir crater in the Bering

Sea. The Mjølnir crater is 40 km in diameter formed by a projectile estimated to have been 1-3 km diameter, which impacted in approximately 400 meters water depth

(Shuvalov et al., 2002). Shuvalov et al. (2002) developed numerical models to evaluate the mechanical response, and the role of the unique geological setting in the cratering process. Some results of the numerical models on impact processes obtained by

Shualov et al. (2002) and described in Osinski and Piezzaro (2013) associated with strata that deform plastic and fragile indicate collapse structures, blocks delimited by normal low-angle faults and concentric rings towards the periphery of the impact craters.

Through studies such as Shuvalov (2002) and Shuvalov et al. (2002), knowledge of the impact crater formation processes underwater columns has increased significantly.

Experiments and numerical models carried out by Gault and Sonett (1982) and

Artemieva and Shuvalov, (2002) show that it is possible to have a better idea about the role of the water column and its relation to the projectile. According to Shuvalov et al.

(2002), the d / H ratio, where d is the diameter of the projectile and H is the depth of the water column, has high relevance in the process of formation of impact craters as summarized in Table 4.

34

d/H ratio Result

< 0.1 No crater is formed at seafloor

> 0.1 , < 1 Relevant results in morphology and size of the structure

If d > H only a minor direct influence of the water column

In any case, significant differences are observed on submarine impact craters

compared to their continental counterparts.

Table 4. Summary of the effects in the formation of impact craters and their relationship with the d / H ratio.

If the impactor is small compared to water depth (d/H < 0.1), there will be no disturbance of the sediment below the water column; However, if d/H > 1, the water column will have no influence on the impactor, and the crater will develop as if it were terrestrial. Impact craters, their morphology, size, and formation process are widely documented in the literature. As mentioned by Shuvalov et al. (2002), the morphology of the Mjølnir crater, located 400 meters deep under the sea, differs considerably from subaerial impact structures without a significant water column. This structure displays low topography towards the periphery, contrary to that observed in impact craters such as Meteor

Crater. Additionally, due to the plastic deformation of the weak sediments located in the first 3 km of the ocean floor, a collapse occurred at the same time as an unusual filling of the structure. A similar process could have occurred in the study area since the glacial and lacustrine sediments located deep underwater (approximately 115 m) at the time of possible impact structure formation, would also have a plastic deformation.

35

According to Sonett et al. (1991), once the impact of the extraterrestrial body with the body of water occurs, two types of waves occur, the first being related to the excavation of the body of water and the second associated to crater slosh. Assuming that the

Structure is an impact crater and considering the dimensions of the structure, a bolide with a size equivalent to 0.5 km in diameter traveling at an average speed of 25 km / s would be needed (Schmidt and Holsapple, 1982; Sonett, Pearce, and Gault, 1991).

However, according to Shuvalov (2002) and Shuvalov et al. (2002) and I quote “the processes accompanying impacts into the ocean strongly depend on the sea depth. If water depth exceeds two impactor diameters, the projectile is totally disrupted and decelerated as it passes through the water column, and only a small depression in the sea bottom is formed under the action of shock compressed and downwards accelerated water.” Additionally, Shuralov (2002) indicates: “the formation of concentric craters is not directly related to the depth of the water column. Such craters appear to form because of special underwater target lithology (layered targets). In particular, such craters are created in low strength sediments or regolith layer overlying the crystalline basement.”

Ormö and Lindström (2000) describe several impact craters in Baltoscandia that were formed in shallow-water marine environments with water depths of a few 10s of meters.

Several are on the order of the diameter of the Roseau Structure including the Lockne,

Kaluga, Granby, Kardla, Flynn Creek, and Brent craters; of these, Granby, Kardla, Flynn

Creek, and Brent craters were formed where the water depth was estimated to be less than 75 meters (Ormö and Lindström, 2000). All of these shallow-water craters are associated with slump blocks but no resurge deposits. All of these impacts, however, are suggested to have been associated with basement rock fracturing and cratering, depths of a few 100s of meters.

36

Another possibility that must be considered is that a rocky or cometary fragment detonated above Lake Agassiz in an airburst. Information on airbursts, particularly over water is difficult to find. Robertson and Mathias (2019) conduct ALE3D Hydrocode

(https://wci.llnl.gov/simulation/computer-codes/ale3d) simulations of asteroid airbursts in relation to the Tunguska event of 1908. The tree fall zone at Tunguska was over 2000 km2 in area, however, no cratering was present at the site. Robertson and Mathias

(2019) estimate the Tunguska meteor at between 50 and 100 meters diameter with airburst altitudes of 5-15 km depending on the velocity, angle, and density. et al.

(2017) conduct a numerical assessment of airbursts of rocky and cometary materials in the 10-100 meter diameter range. Their simulation show that airbursts can occur at any altitude with the energy release increasing with decreasing altitude (Collins et al., 2017).

However, it is very difficult to predict the nature of the impact or cratering associated with airburst. In addition, they emphasize the process of airburst to crater formation is not discontinuous, and there is an intermediate range where both can occur (Collins et al.,

2017).

Huss et al. (1966), reports the presence of a metallic meteorite fragment (Octahedrite) in

Anok, Minnesota (45 ° 12 'N, 93 ° 26' W), at a distance of 465 kilometers from the study area. Compositionally and genetically, this fragment is associated with a group of meteorite fragments found in Havana, Illinois, in layers whose age is estimated to be

2336 ± 250 BP (Arnold and Libby, 1951; McCoy et al., 2017) attribute its origin to the fall of meteorites in a broad region, which, due to its large vegetation cover, made it difficult to find (Carr and Sears, 1985). These fragments could be related to asteroid break-up in the atmosphere that may have formed the crater.

37

5. Discussion

The Roseau Structure in northern Minnesota has characteristics that resemble those of an impact crater. It’s nearly circular shape, bordered by ridges in a northeast direction, resembles the descriptions made by Shuvalov et al. (2002) for impact craters formed by impacting weak strata that are plastically deformed. In the study area, these strata are of lacustrine and glacial type; in turn, these lie on strong rocks that tend to deform fragilely.

Furthermore, structure formation is not consistent with the presence of glaciers due to it displays ridges as a product of washing waves. However, by the time these structures formed, there was a body of water with an average depth of 50 meters with approximately 60 m of low-density sediment overlying solid rock, which constitutes a scenario similar to Mjolnir crater in the shallow Barents Sea.

The circular-like structure displays a set of ridges that are visible at the eastern side of the structure itself. They might be associated with 1) an impact crater, which due to the high temperatures and pressures present during the crash against the earth's surface, would produce a partial melting of the substrate, which at cooling down would produce these ridges; another possibility is that the eastern side of the studied structure might suffered wind erosion, displaying similar features as has been observed in other impact craters in rocky bodies in the solar system, specifically on Mars (Raitala and Kauhanen,

1992; Zimbelman, 2010).

Firestone et al. (2007) suggest the Carolina Bays may be evidence of multiple impacts associated with the breakup of a rocky or cometary body. Melton and Schriever (1937) ruled out other “ordinary geological processes” and concluded that the Carolina Bays were most likely impact related. Although most of the Carolina Bays are smaller than the

38

Roseau Structure, their morphologies are strikingly similar including multiple concentric ovoid rings. The largest of the Carolina Bays are similar in size to the Roseau Structure.

Given the study area's geographical and geological context, it is sensible to think that the structure is the result of glacial or lacustrine processes. However, the delicacy of the structure suggests that once the ice sheet readvances in the region, the structure would disappear, making it difficult to reconcile this process with the formation of such structure due to the type of weak material. Besides, the ice flow was in the NE - SW direction throughout most of the Wisconsin Glaciation, and the ridges are in the NE direction of the study structure. At the start of the YD, Lake Agassiz was at the Tintah Level, covering the structure with a water column of at least 50 meters. Subsequently, the water level dropped and returned to reach the Campbell level at the end of the YD. On the other hand, the wave processes due to Lake Agassiz's presence in this area indicate that the structure is postglacial and, therefore, it is possible to associate it with Lake

Agassiz.

The work of Melosh (1980, 1989), Shuvalov (2002), Shuvalov et al. (2002), and Ormö and Lindström (2000) suggest that a solid impactor producing a crater the size of the

Roseau Structure would have resulted in cratering and fracturing of the basement rock.

However, gravity surveys suggest there has been no disruption of the crystalline basement (Figure 5). Little information is available in the literature regarding the impact of icy bodies.

There are no other reported similar structures in the region. This structure is the only one with this circular shape, size, and ridges on its periphery in northern Minnesota and surrounding areas. Consequently, glacial or lacustrine processes, typical of this region, are unlikely, and are ruled out to explain their origin. In terms of time and for the reasons

39 explained above, the structure must have formed after the retreat of the ice sheet covering this area and before the drainage of Lake Agassiz.

Perhaps the structure's formation is temporarily related to the start of the YD, but the quasi-quality is unclear. The reasoning mentioned above proposes that structure formed, either by the impact of an asteroid or meteorite airburst. As pointed out by Melosh et al.

(1980) and Shuralov et al. (2002), both the D/h ratio and the type of structures resulting from the plastic deformation of sedimentary material covered by a shallow column of water lying on strong rocks, are very well related to the structure present in the study area. The absence of a gravimetric anomaly in the geophysical data suggests that a meteorite airburst is a viable option, given the lithological conditions in the area, with 50+ meters of water overlying low-density sediment.

On the contrary, if it is considered that the studied structure was formed after 12.9 k BP, the impact would be subaerial, since by this time Lake Agassiz would have retreated.

This is well marked by continental evidence and in the ocean basins of a route east through the - St. Lawrence to the North Atlantic and along a northwest route through the Clearwater - Athabasca - Mackenzie system to the Arctic Ocean (Teller,

2013; Murton et al., 2010).

Whether a potential impact occurred before or after 12.9 k BP, the stratigraphic, sedimentary, geomorphological, and geochronological evidence indicates that the ice cap present at the rocks had already retreated north of the Structure location by 13,500 cal BP and therefore the location established for the possible impact structure would be ice-free. Likewise, at 12.9 ka BP the surface would not be free of water either, since it is recorded that for approximately 11,000 cal BP Lake Agassiz would have risen to reach the Campbell level (Breckenridge, 2015). Based on Shuvalov (2002), impactors can be deformed, fragmented and even decelerated during the flight through the Earth's

40 atmosphere. This effect, in turn, can influence the cratering process. This process could explain the shape of the study Structure.

According to data provided by George H. Shaw (personal communication, 2020), gravimetric anomalies are not significant in the study area (Figure 5). Although the geophysical evidence associated with an impact crater indicates variations in the gravitational field close to these structures, as a result of the thinning of the crust

(Osisnski and Pierazzo, 2013), it is crucial to indicate that this region has glacial and lacustrine sediments lying on metamorphic and igneous rocks of the Precambrian. Both types of lithologies have different rheologies, so the mechanical response to impact would leave structures whose geophysical signature is different from typical subaerial craters.

If the impact structure hypothesis is plausible, it is prudent to point out that an asteroid fragment has impacted northern Minnesota as a result of the main asteroid break up once go through the atmosphere. As pointed out by Passey and Melosh (1980), meteoroids (fragments of asteroids that vary in size from dust particles to blocks) once pass through Earth's atmosphere at speeds of several kilometers per second, are subject to such stresses and pressures thus producing their fragmentation. Some of these fragments survive the descent and can reach the surface while others disintegrate in the atmosphere at heights that vary between 70 and 100 kilometers in height for fragments between 0.1 - 10 cm in diameter and between 4 to 40 kilometers in height for larger fragments (Passey and Melosh, 1980; Popova, 2004; Gritsevich, 2008; Silber,

2018). Passey and Melosh (1980) indicate that, if the size of the impact crater is greater than 1 kilometer in diameter, the separation of the meteoroids is only a fraction of its diameter, so the impact crater formed by the meteoroid would be indistinguishable from that generated by the main impactor.

41

Considering that the crater has an approximate diameter of 6 km, based on the equations of Melosh (1980) and under the lithological and rheological parameters explained by Shuvalov (2002) and Shuvalov et al. (2002), the necessary dimensions of the projectile to form an impact crater with the diameter mentioned

D = 8 km, h = 100 m

D/h = 80

Consequently, this value exceeds the relationship established by Melosh (1980) whose main characteristic is the generation of concentric rings such as they are observed towards the northeast of the structure.

According to the parameters established by Shuvalov et al. (2002), the water column of approximately 60 meters deep and a small fragment, has the following relationship:

Water column (H) = 60 m

Diameter of the projectile (d) = 10 m

If d/H = 0.16 d/H > 0.1

With these parameters, it could be said that an asteroid fragment could have impacted in this region generating the studied structure.

6. Conclusions

Based on the geological, geophysical, and geographic observations of the study area and the characteristics of the structure in northern Minnesota, it is possible that under the circumstances given approximately 12,900 years ago, a meteorite impact was the

42 process that formed the Roseau Structure. The geophysical data (Figure 5), no gravity anomaly, is inconsistent with an impact of a rocky or metallic body, as a solid body would need to be of substantial size to generate a 6+ km diameter impact structure.

However, the response of an icy body is more difficult to predict; there is a paucity of studies that analyze the impact of a low-density body. Similarly, there is little information on how an airburst over water might manifest itself in the lake bottom. Analysis of the

Tunguska Event (Robertson and Mathias, 2019). The Tunguska Event is accepted to be caused by an airburst of a stony meteoroid at an altitude of about 5-10 km. It flattened trees over an area of approximately 2000 km2. The Roseau Structure is on the order of

30 km2 in area, but would have been beneath 50+ meters of water. Therefore, it is not possible to determine with any degree of confidence if this is an impact structure that related to that reported by Firestone et al. (2007). However, a glacial or lacustrine origin in difficult to conceive. Further impact-related evidence from nearby environmental archives would help in this Younger Dryas event reconstruction.

43

Figure 1. Large proglacial lakes along retreating Laurentide Ice Sheet at 12.65–12.75 cal. kyr BP, near the start of Younger Dryas. Three outlets are indicated: (1) to the northwest along the to the Arctic Ocean, (2) to the east along the St

Lawrence River to the North Atlantic Ocean, and (3) to the south along the river

Mississippi to the Gulf of Mexico. The distribution of glacial ice (white), proglacial lakes

(dark blue), and land (gray) is for reference. Modified from Murton, J. et al., 2010.

44

Figure 2. Summary graph of strandline point elevation data. Outlet sills (#1-14) are shown as vertical bars and best fit lines for the lowest Tintah and Campbell levels are included (Breckenridge, 2015).

45

Figure 3. Lake Agassiz lake-level hydrograph. Gray bars represent radiocarbon-dated peat deposits from Liu et al. 2014.

46

Figure 4. The Roseau Structure in Northern Minnesota. The shape of the structure is mainly elliptical and some ridges are located towards Northeast of the structure (thin white solid lines).

47

Figure 5. Inverse-distance-weighted contour map of gravity data for portions of Roseau County, MN. Bold black ellipse indicates the location of the Structure. Light blue dots represent locations of gravity measurements. Data are from the Minnesota Geological

Survey Geophysical Database (https://mngs-umn.opendata.arcgis.com/app/1a79082164d74c1f89f0fb31217fa5b7).

48

Figure 6. Satellite image form some of the "Carolina Bays" in Horry County, SC. The elliptical shape and the orientation of such structures are clearly observed from an aerial view.

49

References:

1. Abbott, D. H., Bryant, E. A., Gusiakov, V., Masse, W. B., & Breger, D. 2008. Impacts, mega-tsunami, and other extraordinary claims. Comment. GSA Today, 18(6), 12. DOI: 10.1130/GSATG9C.1 2. Alvarez, L. W., Alvarez, W., Asaro, F., and Michel, H. V. (1980). Extraterrestrial cause for the Cretaceous-Tertiary extinction. Science, 208(4448), 1095-1108. DOI: 10.1126/science.208.4448.1095 3. Arnold, J. R., & Libby, W. F. (1951). Radiocarbon dates. Science, 113(2927), 111-120. 4. Bakke, J., Lie, Ø., Heegaard, E., Dokken, T., Haug, G.H., Birks, H.H., Dulski, P., and Nilsen, T. 2009. Rapid oceanic and atmospheric changes during the Younger Dryas cold period. Nature Geoscience. v. 2. p. 202-205; doi:10.1038/ngeo439. 5. Bass, D.W. and Woodworth-Lynas, C., 1988. Iceberg crater marks on the sea floor, Labrador Shelf. Marine Geology, 79(3-4), pp.243-260. https://doi.org/10.1016/0025-3227(88)90041-2 6. Benn, D. I., & , D. J. A. (2010). Glaciers and Glaciation. Hodder Education. London, UK, 802 pp. 7. Boslough, M., and other 15 authors. 2012. Arguments and Evidence Against a Younger Dryas Impact Event. , Landscapes, and Civilizations. AGU. doi:10.1029/2012GM001209. 8. Breckenridge, A. 2007. The Lake Superior stratigraphy and implications for eastern Lake Agassiz outflow from 10,700 to 8900 cal ybp (9.5–8.0 14C ka). Palaeogeography, Palaeoclimatology, Palaeoecology. v. 246. p. 45-61; doi.org/10.1016/j.palaeo.2006.10.026. 9. Breckenridge, A. 2015. The Tintah-Campbell gap and implications for glacial Lake Agassiz drainage during the Younger Dryas cold interval. Science Reviews. v. 117. p. 124-134, doi.org/10.1016/j.quascirev.2015.04.009. 10. Broecker, W. S., Kennett, J. P., Flower, B. P., Teller, J. T., Trumbore, S., Bonani, G., & Wolfli, W. (1989). Routing of meltwater from the Laurentide Ice Sheet during the Younger Dryas cold episode. Nature, 341(6240), 318. 11. Broecker, W.S., Denton, G.H, Edwards, R.L., Cheng, H., Alley, R.B., and Putnam, A.E. 2010. Putting the Younger Dryas cold event into context. Quaternary Science Reviews. v. 29, Issues 9–10. p. 1078-1081; doi.org/10.1016/j.quascirev.2010.02.019. 12. Brown, C.S., Newton, A.M., Huuse, M. and Buckley, F., 2017. Iceberg scours, pits, and pockmarks in the North Falkland Basin. Marine Geology, 386, pp.140- 152. https://doi.org/10.1016/j.margeo.2017.03.001. 13. Buchanan, B., Collard, M. and Edinborough, K. 2008. Paleoindian demography and the extraterrestrial impact hypothesis. PNAS. v.105 (33). p. 11651-11654; doi:10.1073/pnas.0803762105.

50

14. Carlson, A. E., Mickelson, D. M., Principato, S. M., & Chapel, D. M. (2005). The genesis of the northern Kettle Moraine, Wisconsin. Geomorphology, 67(3-4), 365-374. 15. Carlson, A. E., , P.U., and Hostetler, S.W. 2009. Comment: Radiocarbon deglaciation chronology of the Thunder Bay, Ontario area and implications for ice sheet retreat patterns. Quaternary Science Reviews. v. 28. p. 2546-2547. 16. Carlson, A.E. 2010. What caused the Younger Dryas Cold Event? Geology. v. 38, p. 383-384; doi:10.1130/focus042010.1. 17. Carr, C., & Sears, D. W. (1985). Toward an analysis of the exchange of meteoritic iron in the middle woodland. Southeastern , 79-92.

18. Clayton, L. (1967). Stagnant-glacier features of the Missouri Coteau in North Dakota. North Dakota Geological Survey Miscellaneous Series, 30, 25-46. 19. Clayton, L., TELLER, J. T., & ATTIG, J. W. (1985). Surging of the southwestern part of the Laurentide Ice Sheet. Boreas, 14(3), 235-241. 20. Collins, G. S., Lynch, E., McAdam, R., & Davison, T. M. (2017). A numerical assessment of simple airblast models of impact airbursts. Meteoritics & Planetary Science, 52(8), 1542-1560. 21. Collins, G. S., Artemieva, N., Wünnemann, K., Bland, P. A., Reimold, W. U. and Koeberl, C. 2008. Evidence that Lake Cheko is not an impact crater. Terra Nova, 20: 165–168. doi:10.1111/j.1365-3121.2008.00791.x 22. Coveney Jr, R. M., Murowchick, J. B., Grauch, R. I., Glascock, M. D., & Denison, J. R. 1992. Gold and platinum in shales with evidence against extraterrestrial sources of metals. Chemical Geology, 99(1-3), 101-114. https://doi.org/10.1016/0009-2541(92)90033-2. 23. Daulton, T.L., Pinter, N., and Scott, A. 2010. No evidence of nanodiamonds in Younger–Dryas sediments to support an impact event. PNAS. v. 107 (37). p. 16043-16047; doi:10.1073/pnas.1003904107. 24. Dyer, B. D., Lyalikova, N. N., Murray, D., Doyle, M., Kolesov, G. M., & Krumbein, W. E. (1989). Role for microorganisms in the formation of iridium anomalies. Geology, 17(11), 1036-1039. doi.org/10.1130/0091- 7613(1989)017<1036:RFMITF>2.3.CO;2. 25. Faith, J. T. (2011). Late Pleistocene climate change, nutrient cycling, and the megafaunal extinctions in North America. Quaternary Science Reviews, 30(13- 14), 1675-1680. 26. Faith, J. T., & Surovell, T. A. (2009). Synchronous extinction of North America's Pleistocene mammals. Proceedings of the National Academy of Sciences, 106(49), 20641-20645. 27. Fay, H., 2002. Formation of kettle holes following a glacial outburst flood (jökulhlaup), Skeiðarársandur. The extremes of the extremes: extraordinary floods. (271), p.205. 28. Fenton, M. M., Moran, S. R., Teller, J. T., & Clayton, L. (1983). Quaternary stratigraphy and history in the southern part of the Lake Agassiz basin. In Glacial Lake Agassiz (pp. 49-74). Geological Association of Canada Special Publication 26.

51

29. Firestone, R.B. and other 25 authors. 2007a. Evidence for an extraterrestrial impact 12,900 years ago that contributed to the megafaunal extinctions and the Younger Dryas cooling. PNAS. v. 104 no. 41 p.16016–16021; doi: 10.1073/pnas.0706977104. 30. Firestone, R.B., West, A., Zsolt, S., Zsolt, R., and Hagstrum, J.T., 2007b, Micrometeorite impacts in Beringian mammoth tusks and a skull: Eos (Transactions, American Geophysical Union) v. 88, no. 52, Abstract U23A-0865, http://ie.lbl.gov/Mammoth/Impact.html. 31. Firestone, R. B. 2009. The case for the Younger Dryas extraterrestrial impact event: mammoth, megafauna and Clovis extinction. Journal of Cosmology. 2, 256-285. https://escholarship.org/uc/item/8fj3d8mc 32. Firestone, R.B., West, A., Revay, Z., Hagstrum, J.T., Belgya, T., Que Hee, S.S., Smith, A.R., 2010. Analysis of the Younger Dryas impact layer. Journal of Siberian Federal University, Engineering & Technologies. 3, 30–62. https://escholarship.org/uc/item/03q2r98x 33. Fisher, T.G. 2003. Chronology of glacial Lake Agassiz meltwater routed to the Gulf of Mexico. Quaternary Research. v. 59. p. 271-276; doi.org/10.1016/S0033- 5894(03)00011-5. 34. Fisher, T.G. 2005. Strandline analysis in the southern basin of glacial Lake Agassiz, Minnesota and North and South Dakota, USA. GSA Bulletin. v. 117 no. 11-12 p. 1481-1496; doi: 10.1130/B25752.1. 35. Fisher, TG., and Lowell, T.V. 2006. Questioning the age of the Moorhead Phase in the glacial Lake Agassiz basin. Quaternary Science Reviews. v. 25. p. 2688- 2691; doi.org/10.1016/j.quascirev.2006.05.007. 36. Fisher, T. G., Yansa, C. H., Lowell, T. V., Lepper, K., Hajdas, I., and Ashworth, A. 2008. The chronology, climate, and confusion of the Moorhead Phase of glacial Lake Agassiz: new results from the Ojata Beach, North Dakota, USA. Quaternary Science Reviews. v. 27(11-12). p. 1124-1135; doi: 10.1016/j.quascirev.2008.02.010. 37. Fisher, T.G., Waterson, N., Lowell, T.V., and Hajdas, I. 2009. Deglaciation ages and meltwater routing in the Fort McMurray region, northeastern Alberta and northwestern Saskatchewan, Canada. Quaternary Science Reviews. v. 28. p. 1608-1624; doi.org/10.1016/j.quascirev.2009.02.003. 38. Fisher, T.G. and Lowell, T.V., 2012. Testing northwest drainage from Lake Agassiz using extant ice margin and strandline data. Quaternary International, 260, pp.106-114. https://doi.org/10.1016/j.quaint.2011.09.018. 39. Flohn, H. (1979). On time scales and causes of abrupt paleoclimatic events. Quaternary Research, 12(1), 135-149. 40. French, B.M., and Koeberl, C. 2010. The convincing identification of terrestrial meteorite impact structures: What works, what doesn't, and why. Earth-Science Reviews. v. 98. p. 123-170; doi.org/10.1016/j.earscirev.2009.10.009. 41. Gabrielli, P., and other 12 authors. 2004. Meteoric smoke fallout over the Holocene revealed by iridium and platinum in Greenland ice. Nature, 432(7020), 1011-1014. https://doi.org/10.1038/nature03137.

52

42. Gilmour, I., Russell, S.S., Arden, J.W., Lee, M.R., Franchi, I.A. and Pillinger, C.T. 1992. Terrestrial carbon and isotopic ratios from Cretaceous-Tertiary boundary nanodiamonds. Science, 258(5088), pp.1624-1626. 43. Grimm, E.C., Donovan, J.J. and Brown, K.J., 2011. A high-resolution record of climate variability and landscape response from Kettle Lake, northern Great Plains, North America. Quaternary Science Reviews, 30(19-20), pp.2626-2650. https://doi.org/10.1016/j.quascirev.2011.05.015 44. Gritsevich, M.I., 2008. The Pribram, Lost City, Innisfree, and Neuschwanstein falls: an analysis of the atmospheric trajectories. Solar System Research, 42(5), pp.372-390. https://doi.org/10.1134/S003809460805002X 45. Harris, D. R. (2003). Climatic change and the beginnings of agriculture: the case of the Younger Dryas. In Evolution on Planet Earth (pp. 379-394). Academic Press. 46. Harris, K.L. (1995). Quaternary geology-Southern Red River Valley, Minnesota [Part A]. Minnesota Geological Survey. Retrieved from the University of Minnesota Digital Conservancy, http://hdl.handle.net/11299/59764. 47. Haynes, C.V., Jr. 2008. Younger Dryas “black mats” and the Rancholabrean termination in North America. PNAS. v. 105 no. 18. p. 6520–6525; doi: 10.1073/pnas.0800560105. 48. Hills, J.G. and Goda, P. 1993. The fragmentation of small asteroids in the atmosphere. Astronomical Journal. v. 105, no. 3, p. 1114-1144. 49. Hobbs, H.C., 1983. Drainage relationship of Glacial Lake Aitkin and Upham and early Lake Agassiz in northeastern Minnesota. In: Teller, J.T., Clayton, L. (Eds.), Glacial Lake Agassiz, pp. 245e259. Geological Association of Canada Special Paper 26. 50. Holcombe, T. L., Warren, J. S., Reid, D. F., Virden, W. T., & Divins, D. L. 2001. Small rimmed depression in : An impact crater? Journal of Great Lakes Research, 27(4), 510-517. https://doi.org/10.1016/S0380-1330(01)70664- 8 51. Hughes, J.D., and Merry, W.J. 1978. Marquette Buried Forest9,850 years old. American Association for the Advancement ofScience, Abstract for Annual Meeting, 12–14 February 1978. 52. Israde-Alcántara, I., and other 15 authors. 2012. Evidence from central Mexico supporting the Younger Dryas extraterrestrial impact hypothesis. PNAS. v. 109 (13) E738-E747; doi:10.1073/pnas.1110614109. 53. James T. Teller, J.T. 2013. Lake Agassiz during the Younger Dryas. Quaternary Research. v. 80. p. 361-369; doi.org/10.1016/j.yqres.2013.06.011. 54. Kamb, B. (1987). Glacier surge mechanism based on linked cavity configuration of the basal water conduit system. Journal of Geophysical Research: Solid Earth, 92(B9), 9083-9100. 55. Kennet, D.J., and other 16 authors. 2009a. Shock-synthesized hexagonal diamonds in Younger Dryas boundary sediments. PNAS. v. 106 (31). p. 12623- 12628; doi:10.1073/pnas.0906374106

53

56. Kennet, D.J., Kennet, J.P., West, A., Mercer, C., Que Hee, S.S., Bennet, L., Bunch, T.E., Sellers, M., and Wolbach, W.S. 2009b. Nanodiamonds in the Younger Dryas boundary sediment layer. Science. v. 323. Issue 5910. p. 94; doi: 10.1126/science.1162819. 57. Koeberl, C., 2007. The geochemistry and cosmochemistry of impacts, In: Davis, A. (Ed.), online edition. Treatise of Geochemistry, vol. 1. Elsevier, New York, NY, pp. 1.28.1–1.28.52. doi:10.1016/B978-008043751-4/00228-5. 58. Kuzilla M.S. (1988) Genesis and morphology of soils in and around large depressions in County, Nebraska, Thesis, University of Nebraska, Lincoln. 59. Larson, P.C., Mooers, H.D., and Meyer, M.S. (2016) Glacial Geology of the Laurentian Uplands. In: Institute of Lake Superior Geology, Field Trip Guidebook, Vol. 62, Part 2, 1-13. 60. LeCompte, M.A., Goodyear, A.C., Demitroff, M.N., Batchelor, D., Vogel, E.K., Mooney, C., Rock, B.N., and Seidel, A.W. 2012. Independent evaluation of conflicting microspherule results from different investigations of the Younger Dryas impact hypothesis. PNAS. v. 109 (44), E2960- E2969. doi:10.1073/pnas.1208603109. 61. Lepper, K., Fisher, T.G., Hajdas, I., and Lowell, T.V. 2007. Ages for the Big Stone Moraine and the oldest beaches of glacial Lake Agassiz: Implications for deglaciation chronology. Geology. v. 35 no. 7 p. 667-670; doi: 10.1130/G23665A.1. 62. Lepper, K., Buell, A.W., Fisher, T.G., and Lowell, T.V. 2013. A chronology for glacial Lake Agassiz shorelines along Upham's namesake transect. Quaternary Research. v. 80. p. 88-98; doi.org/10.1016/j.yqres.2013.02.002. 63. Liu, X., Fisher, T. G., Lepper, K., & Lowell, T. V. (2014). Geochemical characteristics of glacial Lake Agassiz sediments and new ages for the Moorhead Phase at Fargo, North Dakota, USA. Canadian Journal of Earth Sciences, 51(9), 850-861. 64. Lowell, T.V., Fisher, T.G., Hajdas, I., Glover, K., Loope, H., and Henry, T. 2009. Radiocarbon deglaciation chronology of the Thunder Bay, Ontario area and implications for ice sheet retreat patterns. Quaternary Science Reviews. v. 28. p. 1597-1607; doi.org/10.1016/j.quascirev.2009.02.025. 65. Lowell, T.V., and Fisher, T.G. 2009. Reply to comments by Carlson et al., (2009) on Radiocarbon glaciation chronology of the Thunder Bay, Ontario area and implications for ice sheet retreat patterns. Quaternary Science Reviews. v. 28. p. 254-2550. 66. Lowell, T.V., Applegate, P.J., Fisher, T.G., and Lepper, K. 2013. What caused the low-water phase of glacial Lake Agassiz? Quaternary Research. v. 80. p. 370-382; doi.org/10.1016/j.yqres.2013.06.002. 67. Lusardi, B. A., Jennings, C. E., & Harris, K. L. (2011). Provenance of Des Moines lobe till records ice‐stream catchment evolution during Laurentide deglaciation. Boreas, 40(4), 585-597. 68. Mangerud, J. (1987). The Allerød/Younger Dryas boundary. In Abrupt Climatic Change (pp. 163-171). Springer, Dordrecht. https://doi.org/10.1007/978-94-009- 3993-6_16.

54

69. Melosh, H. J. (1980). Cratering mechanics—Observational, experimental, and theoretical. Annual Review of Earth and Planetary Sciences, 8(1), 65-93. doi:10.1146/annurev.ea.08.050180.000433

70. Melton, F.A. and Schriever, W., 1933. The Carolina" Bays": Are They Meteorite Scars?. The Journal of Geology, 41(1), pp.52-66. 71. McCausland, P. J., Samson, C., and McLeod, T. (2011). Determination of bulk density for small meteorite fragments via visible light 3‐D laser imaging. Meteoritics & Planetary Science, 46(8), 1097-1109. https://doi.org/10.1111/j.1945-5100.2011.01217.x

72. McCoy, T. J., Marquardt, A. E., Wasson, J. T., Ash, R. D., & Vicenzi, E. P. (2017). The Anoka, Minnesota iron meteorite as parent to Hopewell meteoritic metal beads from Havana, Illinois. Journal of Archaeological Science, 81, 13-22. https://doi.org/10.1016/j.jas.2017.03.003

73. McDonald, I., Irvine, G J., de Vos, E., Gale, A.S. and Reimold, W.U., 2006. Geochemical search for impact signatures in possible impact-generated units associated with the -Cretaceous Boundary in southern England and northern France. Biological Processes Associated with Impact Events. C. Cockell, I. Gilmour and C. Koeberl, Springer Berlin Heidelberg, 257-286 74. Melosh, H.J. 1989. Impact Cratering. New York: Oxford University Press, 245 pp. 75. Moore, A.M.T., Kennett, J.P., Napier, W.M. et al. Evidence of Cosmic Impact at Abu Hureyra, Syria at the Younger Dryas Onset (~12.8 ka): High-temperature melting at >2200 °C. Sci Rep 10, 4185 (2020). https://doi.org/10.1038/s41598- 020-60867-w 76. Munro, N. D. (2003). Small game, the Younger Dryas, and the transition to agriculture in the southern Levant. Mitteilungen der Gesellschaft für Urgeschichte, 12(4), 47-71. 77. Murton, J.B., Bateman, M.D., Dallimore, S.R., Teller, J.T., and Yang, Z. 2010. Identification of Younger Dryas outburst flood path from Lake Agassiz to the Arctic Ocean. 78. Newton, A.M., Huuse, M. and Brocklehurst, S.H., 2016. Buried iceberg scours reveal reduced North Atlantic Current during the stage 12 deglacial. Nature communications, 7(1), pp.1-7. doi: 10.1038/ncomms10927. 79. Oberbeck, V. R. (1971). Laboratory simulation of impact cratering with high explosives. Journal of Geophysical Research, 76(23), 5732-5749. https://doi.org/10.1029/JB076i023p05732.

80. Osinski, G. R., & Pierazzo, E. (2013). Impact cratering: Processes and products. Impact Cratering, Wiley-Blacwell. 81. Paquay, F., Goderis, S., Ravizza, G., Vanhaeck, F., Boyd, M., Surovell, T.A., Holiday, V.T., Haynes, C.V. Jr., and Claeys, P. 2009. Absence of geochemical evidence for an impact event at the Bølling–Allerød/Younger Dryas transition. PNAS. v. 106 (51). p. 21505-21510; doi:10.1073/pnas.0908874106.

55

82. Passey, Q., and Melosh, H.J. 1980. Effects of atmospheric breakup on crater field formation. Icarus. v. 42. p. 211-233. 83. Popova, O., 2004. Meteoroid ablation models. Earth, Moon, and Planets, 95(1-4), pp.303-319. https://doi.org/10.1007/s11038-005-9026-x 84. Pinter, N., and Ishman, S.E. 2008. Impacts, mega-tsunamis, and other extraordinary claims. GSA Today. Reply. p. e14. doi: 10.1130/GSATG13Y.1 85. Pinter, N., Scott, A.C., Daulton, T.L., Podoll, A., Koeberl, C., Anderson, S.R., and Ishman, S.E. 2011. The Younger Dryas impact hypothesis: A requiem. Earth- Science Reviews. v.106, Issues 3–4. p. 247-264; doi.org/10.1016/j.earscirev.2011.02.005. 86. Prouty, W. F. 1952. Carolina bays and their origin. Geological Society of America Bulletin, 63(2), 167-224. 87. Raitala, J. and Kauhanen, K., 1992. Ridge systems related to Martian impact craters. Earth, Moon, and Planets, 58(1), pp.65-78. https://doi.org/10.1007/BF00058074. 88. Rampino, M. R., & Caldeira, K. 2017. Correlation of the largest craters, stratigraphic impact signatures, and extinction events over the past 250 Myr. Geoscience Frontiers, 8(6), 1241-1245. https://doi.org/10.1016/j.gsf.2017.03.002 89. Reimold, W. U. 2007. The Impact Crater Bandwagon (Some problems with the terrestrial impact cratering record). Meteoritics & Planetary Science, 42: 1467– 1472. doi:10.1111/j.1945-5100.2007.tb00585.x 90. Reimold, W. U., Ferrière, L., , A. and Koeberl, C. 2014. Impact controversies: Impact recognition criteria and related issues. Meteorit Planet Sci, 49: 723–731. doi:10.1111/maps.12284 91. Rooth, C., 1982. Hydrology and Ocean Circulation. Progress in Oceanography. 11: 131-145. doi.org/10.1016/0079-6611(82)90006-4. 92. Röthlisberger, H. 1972. Water pressure in intra- and subglacial channels. Journal of Glaciology. 11, 177–203. DOI: https://doi.org/10.3189/S0022143000022188. 93. Sawlowicz, Z. 1993. Iridium and other platinum-group elements as geochemical markers in sedimentary environments. Palaeogeography, Palaeoclimatology, Palaeoecology, 104(1-4), 253-270. https://doi.org/10.1016/0031-0182(93)90136- 7 94. Schmidt, R.M. and Holsapple, K.A., 1982. Estimates of crater size for large-body impact: Gravity-scaling results. In Geological Implications of Impacts of Large Asteroids and Comets on the Earth (Vol. 190, pp. 93-102). The Geological Society of America Boulder. 95. Schmieder, M., and Buchner, E. 2008. Dating impact craters: Palaeogeographic versus isotopic and stratigraphic methods – a brief case study. Geological Magazine. v.145(4). p. 586-590. doi:10.1017/S0016756808005049. 96. Shoemaker, E. M. 1963. Impact mechanics at Meteor Crater, Arizona. In The Moon, Meteorites, and Comets, ed. B. M. Middlehurst, G. P. Kuiper, 4: 301-36. Chicago & London. Univ. Chicago Press. 810 pp

56

97. Schulte, P., Alegret, L., Arenillas, I., Arz, J.A., Barton, P.J., Bown, P.R., Bralower, T.J., Christeson, G.L., Claeys, P., Cockell, C.S. and Collins, G.S. 2010. The Chicxulub asteroid impact and mass extinction at the Cretaceous-Paleogene boundary. Science, 327(5970), pp.1214-1218. 98. Scott, A.C., Pinter, N., Collinson, M.E., Hardiman, M., Anderson, R.S., Brain, A.P.R., Smith, S., Marone, F., and Stampanoni, M. 2010. Fungus, not comet or catastrophe, accounts for carbonaceous spherules in the Younger Dryas “impact layer.” Geoph. Res. Lett., v. 37, L14302, doi:10.1029/2010GL043345. 99. Shakun, J.D., and Carlson, A.E. 2010. A global perspective on Last Glacial Maximum to Holocene climate change. Quaternary Science Reviews. v. 29. Issues 15–16. p. 1801-1816; doi.org/10.1016/j.quascirev.2010.03.016. 100. Shuvalov, V., Dypvik, H., & Tsikalas, F. (2002). Numerical simulations of the Mjølnir marine impact crater. Journal of Geophysical Research: Planets, 107(E7), 1-1. doi:10.1029/2001je001698

101. Silber, E.A., Boslough, M., Hocking, W.K., Gritsevich, M. and Whitaker, R.W., 2018. Physics of meteor generated shock waves in the earth’s atmosphere–a review. Advances in Space Research, 62(3), pp.489-532. https://doi.org/10.1016/j.asr.2018.05.010 102. Sonett, C.P., Pearce, S.J. and Gault, D.E., 1991. The oceanic impact of large objects. Advances in Space Research, 11(6), pp.77-86. https://doi.org/10.1016/0273-1177(91)90234-B 103. Steffensen, J. P., Andersen, K. K., Bigler, M., Clausen, H. B., Dahl-Jensen, D., , H., ... & Masson-Delmotte, V. 2008. High-resolution Greenland ice core data show happens in few years. Science. 321(5889), 680-684. DOI: 10.1126/science.1157707 104. Surovell, T.A, Holliday, V.T., Gingerich, J.A.M., Ketron, C., Haynes, C.V., Jr., Hilman, I., Wagner, D., Johnson, E., and Claeys, P. 2009. An independent evaluation of the Younger Dryas extraterrestrial impact hypothesis. PNAS. v. 106 (43). p. 18155-18158. doi:10.1073/pnas.0907857106. 105. Surovell, T.A., Boyd, J.R., Haynes Jr, C.V. and Hodgins, G.W., 2016. On the dating of the Folsom Complex and its correlation with the Younger Dryas, the end of Clovis, and megafaunal extinction. PaleoAmerica, 2(2), pp.81-89. doi: 10.1080/20555563.2016.1174559 106. Svetsov, V.V., Nemtchinov, I.V., and Teterev, A.V. 1995. Disintegration of Large Meteoroids in Earth's Atmosphere: Theoretical Models. Icarus. v. 116. p. 131- 153; doi.org/10.1006/icar.1995.1116. 107. Sweatman, M. B., & Tsikritsis, D. (2017). Decoding Göbekli Tepe with archaeoastronomy: What does the say? Mediterranean Archaeology & Archaeometry, 17(1). 108. Tarasov, L., and Peltier, W.R. 2006. A calibrated deglacial drainage chronology for the North American continent: evidence of an Arctic trigger for the Younger Dryas. Quaternary Science Reviews. v. 25. p. 659-688, doi.org/10.1016/j.quascirev.2005.12.006.

57

109. Teller, J. T., & Clayton, L. (1983). Glacial Lake Agassiz: Geological Association of Canada Special Publication No. 26. 110. Teller, J.T., Leverington, D.W., and Mann, J.D. 2002. Freshwater outbursts to the oceans from glacial Lake Agassiz and their role in climate change during the last deglaciation. Quaternary Science Reviews. v. 21. p. 879-887; doi.org/10.1016/S0277-3791(01)00145-7. 111. Teller, J.T., 2013. Lake Agassiz during the Younger Dryas. Quaternary Research, 80(3), pp.361-369. https://doi.org/10.1016/j.yqres.2013.06.011. 112. Tian, H., Scryves, D., and Claeys, Ph. 2011. Nanodiamonds do not provide unique evidence for a Younger Dryas impact. PNAS. v. 108 (1). p. 40- 44; doi:10.1073/pnas.1007695108. 113. Waters, M. R., & Stafford, T. W. (2007). Redefining the age of Clovis: implications for the peopling of the Americas. Science, 315(5815), 1122-1126. DOI: 10.1126/science.1137166 114. Wolbach, W. S., and oher 26 authors. 2018a. Extraordinary biomass-burning episode and impact winter triggered by the Younger Dryas cosmic impact∼ 12,800 years ago. 1. Ice cores and glaciers. The Journal of Geology, 126(2), 165-184. 115. Wolbach, W. S., and other 30 authors. 2018b. Extraordinary biomass-burning episode and impact winter triggered by the Younger Dryas cosmic impact∼ 12,800 years ago. 2. Lake, marine, and terrestrial sediments. The Journal of Geology, 126(2), 185-205. https://doi.org/10.1086/695704 116. Wright, H. E., Thorpe, J. L., Mackay, A., Battarbee, R., Birks, J., & Oldfield, F. (2003). Climatic change and the origin of agriculture in the . Global change in the Holocene, 49-62. 117. Wu, Y., Sharma, M., Lecompte, M.A., Demitroff, M.N., and Landis, J. 2013. Origin and provenance of spherules and magnetic grins at the Younger Dryas boundary. PNAS. v. 110 (38). E3557-66. doi: 10.1073/pnas.1304059110. 118. Zimbelman, J.R., 2010. Transverse aeolian ridges on Mars: First results from HiRISE images. Geomorphology, 121(1-2), pp.22-29. https://doi.org/10.1016/j.geomorph.2009.05.012.

58