The Great Lakes Are Weird, geologically speaking.
I haven’t really heard a great explanation for why they exist as they do, and I’ve been paying attention for one since 6th grade. That’s when I learned about Glacial Lake Agassiz (GLA story & maps), the draining of which carved the Minnesota River valley south of Mpls/StP into a wide enough trough that it affects local weather. (In milder storms, it may snow on one side but not the other.) GLA was larger than all the current 5 Great Lakes combined, about 260km sq. It moved north (downhill) along the retreating 1-3 km thick Laurentide ice sheet border, covering nearly a million square km in its travels. It moved and drained — rapidly, for rocks, moderately for water, and “completely”— by the end of the ice age, leaving “merely” a few Great Canadian Lakes behind. It ended NW of the current border Great Lakes and drained S via Red & MN rivers, E via GL/St Lawrence Seaway, NW along the retreating Laurentide ice sheet border, and ultimately NE into Hudson Bay when the ice sheet collapsed. But look at those paths. Most regional underlying rock in Canada is elderly precambrian shield built up on over millions of years of by shallow seas and scraped back with glaciers. There’s a border stripe running NW to SE through Agassiz with Lake Manitoba & Lake Athabasca of similar shape and scale to Lakes Erie & Ontario, and not too different from James Bay. Clearly glacial flood erosion and other shared processes explain these.
The deepest Great Lakes, Superior, Michigan, and Huron, (M & H having very similar depths), deeper than the runoff carved lakes) lack an explanation for being so different that satisfies me. (If someone else has published these details, sing out & pass on the links.) I’ve come up with an explanation, or rather a hypothesis, the technical term for a scientific guess that seems to fit known info, but wants corroboration or correction to become accepted theory, or not.
Edited 22Aug2024 to add: lake formation date “end” bracket for significant changes, more info on water vapor generally, a couple more reference links, rhetorical flourishes to better link ideas/storytell, and correcting my omission of discussing ice boulders near lakes. And a newer thought about how GL Agassiz may help explain Lake Superior with a new riff on the Lake. I was kind of expecting some of this to be a little cringe, but I’m pretty satisfied with the post — maybe not with the em dashes — including my side quest hypotheses, enough so that my seeky neurons are calming when I re-read it.
First, Niagara Falls causing erosion, between lakes Ontario and Erie, of the dolomite layer it’s on dates it to only ~13K years ago (ya). Which is also the start of the Younger Dryas Cooling Period (12,900ya) and the extinction of most North American megafauna, including sabertoothed tigers (yes, cats, I know) and people. Notably, 13Kya is Very Young geologically speaking. Geologic features can form over millions of years or in moments. I posit that the Lakes’ formation was initiated by a momentary, if momentous, event, then — gradually for humans, quickly for rocks — eroded into their current forms over ~10K years. In the weeks since I first posted, I learned that the Great Lakes have only been their current sizes for about 3000 years, so from ~13,000 to ~3000 they were still changing significantly. (I think this helps my case.)
A broader “Round or Dead” Rule: I’m coming to the conclusion that whenever there’s an abrupt, wide area geologic or biologic anomaly, or a very circular feature, impact from meteors should be ruled out rather than in. Because nothing else short of impacts, or a very massive supervolcano, causes such widespread destruction, rapidly on a continental or global scale.
(Side Quest 1: Volcanoes 🌋 are on the R/D Rule short list. The Tonga volcano threw so much water into our atmosphere, the new-ish satellite sent to orbit to monitor water vapor has its sensor maxed out. It speaks to how OOS it is that no expert thought it could go as high as it is, so the sensor doesn’t have the range. The partly underwater eruption increased the water vapor in the stratosphere by 10%. Water vapor is a greenhouse gas. Greenhouse gasses absorb heat that would otherwise escape to space, trapping it ‘within our climate’. We don’t usually worry politically about it like we do CO2 or methane, figuring our planet is mostly water so the amount in the atmosphere is from ocean evaporation and otherwise stable so it’s not something we can affect much. Turns out direct water injection will do it. Here’s a clear explanation of atmospheric water vapor’s importance.
Tonga side quest cont… The amount of water vapor in the atmosphere is largely stable but the distribution isn’t. We do monitor amounts and distribution closely for weather and other science, hence the satellite. Hotter air holds exponentially more water than colder air. The amount at 90°F (32°C) is about double the amount held at 70°F (21°C).” Surface air is hotter than high atmospheric air; equatorial air is hotter than arctic air. Air movement causes hot and cold air meet up. The hot wet air dumps water as fog, rain, or ice when it cools off. Hotter air can keep gathering moisture longer before it hits cool air and dumps it. Shorter rain cycles in moderate temperatures bring useful soaking rains. Long cycles in hotter air bring droughts and monsoons. Changing agricultural area rain conditions from soaking rains to monsoons isn’t ideal for growing reliable food. Just saying.
Side Quest 1A: Water vapor is mostly gone by 5km, which explains somewhat why Denver is high desert, and the western plains don’t get much water. The ~4.2km high Rocky Mountains can collect water from the coast that might get over or through that top ~0.8km in strong weather patterns, but air cools as it rises, holding less, so it precipitates out, leaving the water/snow closer to the mountain peaks. As it continues E, it descends and gets warmer but mixes with already dry air, so rain stops falling… until it runs unto hot, wet Gulf air heading NE near the Mississippi.
SQ 1 resumes: Only 1% of atmospheric water vapor is in the stratosphere. But it can still trap escaping heat. It just went up by 10%, thanks to the Tonga 🌋. Now we have both extra water & a sudden blip of more greenhouse gas in our atmosphere. We know hotter air holds more water. We know that methane released from arctic permafrost takes a full year cycle to migrate to the southern hemisphere. After the volcano, there were “suddenly wiping out whole suburbs” floods in Australia. A year later, and there are similar floods in northern hemispheres, including enough water and snow to partially relieve the California drought. I’m not suggesting that only Tonga water is coming down. My hypothesis is that the extra heat from the extra water vapor also increased the water carrying capacity of the atmosphere a bit which can be modeled by increasing the effective diameter of the atmospheric layers that hold water, maybe by a cm, maybe by a meter. And maybe that’s enough more extra water than usual to cause massive flooding, first in similar latitudes, then the following year or two (so far…) in all latitudes. So volcanoes can and do cause worldwide issues, but it takes [more than a Southern Hemisphere volcano you can hear in Alaska] to cause a continent wide extinction.
SQ 2: Guesstimating the volume of water in 1cm or 1m of expanded troposphere, a [back of the] Taco bell bag calculation. Assuming (nonexistent) uniformity & perfect spheres, Rearth =6378km, top of troposphere at 9km, effective R=6425km, at 200mbar, -40°C, mixing ratio of 0.002g vapor per 1kg dry air, and an olympic swimming pool (OSP) having 2.5X10^6 L of water in it…
adding to the height to the troposphere
+ 1cm ➡️ 8.9X10^7 L or ~35 OSP of water
or
+1m ➡️ 8.9X10^9 L or ~3500 OSP of water
and while I calculated with the numbers for 50° latitude, as a ballpark figure, most of the precipitated water would hit the tropics. So, ballpark, every +1cm the effective height (diameter) the troposphere expands, we get 35 olympic swimming pools worth of water that could come down as extra rain. Add +1m (to 6425km), and 3500 OSPs come raining down on, say, Pakistan.
/Side quests about atmospheric water vapor are now done.
Back to Great Lakes Formation Hypothesis.
I’m on team “Glacial ice impact theory” for what happened 12,900ya. Briefly, a pretty sizeable meteor 2-3km) hit Michigan. Likely there were concurrent fragments of varying sizes, similar to how Shoemaker-Levy-9 broke up before crashing into Jupiter. But a main piece impacted on or directly over Michigan. Trick was, at the time, there was a mile of ice covering it, so the meteor or its💥 burst hit the ice sheet, not the ground rock directly.
That’s not my hypothesis/theory, I just find it the most sensible explanation. , A. Zamora has good videos on it, and further explains how a ~2km meteor hitting a 1-3km thick ice sheet would detonate the ice and spew ice boulders over a large percentage of the North American continent, wreaking havoc. These ice boulders, where they hit soils with high water tables, created ballistic impact craters, which then mostly filled in leaving conic outlines as rims. He posits the Carolina Bays & Nebraska rainwater basins formed this way. I believe his assertion best fits the data, and as more data is found, it supports his case. For instance, inverted stratigraphy on the crater rims dates to the YD period.
I think we need to look at more than the bays & basins. I’ve been looking at maps of Michigan for a few years now, seeing map after map showing some kind of concentric circle centered there that’s much larger than a tiny crater. What if the ice sheet caused the very large concentric deformation that formed Mainland Michigan itself? Without an impact/explosion event on/over the ice sheet, I don’t think it would have its massive lake borders, maybe no Lake Michigan at all. The escarpment is (semi) circular (to the north) and is gigantic. I want to know why the circular dolomite cap formed there and not in other places with the same geology & same glaciation. Can Dolomite form rapidly with a shock while under pressure? I think it’s likely but I don’t insist on this to support my hypothesis. Dolomitization occurs by swapping in CaMg for Ca in limestone, but which & how many circumstances support its formation is unclear and contested. I don’t know from rocks in the wild, but explosions are often used to get energy for chemical reactions, and shocking fluids can cause dissolved minerals to suddenly crystallize and precipitate out of solution. A water saturated limestone layer high in Mg, under a km of ice, getting slammed by the impacted ice could have been the initiating force, or the pressure followed my substantial ground tremors. At any rate, there’s salt for days capped by limestone and dolomite over mainland Michigan. And the salt caves are dry enough to mine.
One thing that’s hard for me to wrap my mind around is underground water but it’s critical to my (admittedly brand new) hypothesis. Suffice it to say, subsurface stone can be porous enough for flowing underground rivers, or it could collect in some areas more than others. Other stones can absorb water and mineral water solutions. The Great Lakes area was frequently covered in shallow seas for millions of years. There are extensive midwestern aquifers. But in the Michigan and Minnesota areas, we get deep lakes. The Great Lakes are actually pretty young, geologically, they’re extreme examples of their type. So why do they exist as they are instead of being smaller and shallower, or just remaining as aquifer covered with 10,000 lakes?? I have thoughts on that.
That underground water? It’s located in holes in rocks (or absorbed into rocks under pressure). The rocks carry the load of the land above them while the water sits there or flows or does its thing. BUT if the water containing rock is disturbed — shocked, rattled, vibrated, etc — and that rock and porosity distribution changes, the effect is liquefaction. When the rattled rock compacts, it no longer supports the land above it; the land sinks. The weight of the sinking land forces the underground water out. (For a fictional example, read LMBujold’s Captain Vorpatril’s Alliance, a comedic sci/fi novel I enjoy enough to reread.)
That displaced water finds its way sideways or up to the surface. Here is groundwater being forced up during an earthquake. ) This happens with earthquakes, pile driving, or simply stomping on beach sand enough to get water to pool at the surface. There are findable geologic remnants of this process at all scales, like fractals. The water can come up in fairly round tubes, or it can come up in long cracks. Spoiler alert, those tubes are, or can grow to be, sinkholes. After the event, depending on what the ground is made of and what minerals are dissolved or suspended in the water there, the rock that fills those newly opened holes and cracks over time is often softer or harder than the original rock surrounding the scars. After erosion of the crack shaped scars, or their surrounds, there are either cracks or walls left. Think vertical mudflat, drying out. After the erosion of tubular scars, either pits or towers remain. The round remains may erode more or less at the center, , due to what precipitated out or otherwise filled the sinkhole like cream in a Twinkie.
Under Michigan are layers and layers upon layers of rock deposited in shallow seas, and that rock can/could still hold water. But that rock is also under neighboring states that don’t have a round dolomite cap on them or lake borders. Remember where I said the Carolina Bays were filled in? Yeah, I thought not. I also didn’t pay much attention to that part after I heard the explanation once. But the mechanism for the bats filling in almost as soon as they formed was seismic vibration. The seismic activity of the impact liquified some area of land, ice boulders hit some liquified land, formed oval (conic section) craters, then the seismic shaking of millions of tons of ice boulders relentlessly pelting half a continent, and all the dire bears and people thereon, made the land shake enough that the bays mostly filled back in, resettling the liquidized ground, leaving only bays with raised rims and flat, level centers.
So my hypothesis is that when the meteor hit the Laurentide ice sheet over Michigan, it caused a shock that destabilized the underlying water bearing rock layers. (And maybe shock crystalized the dolomite but that’s not necessary for this.) Those water filled layers liquified and collapsed under the weight of the ice. Given the thickness of the ice sheet, it sank as a large piece rather than splitting up piecemeal, leaving mainland Michigan as a plateau of sorts. The water had to escape somewhere, so I posit it went sideways until it was able to surface. I think the gigantic aquifer formerly undeneath it was reduced and the water came up in a ring around most of the crater, forming Lake Michigan & Lake Huron.
The largest ice boulders would have landed, piled in a ring around the Saginaw bay impact area. That secondary impact event (of multiple ice boulders raining down) could have extended the pressure out in a larger ring, blocking water from coming up in the central area, or they could have landed preferentially in the (now) lake areas, making piles of uncertain composition — glacial ice contains rocks and soil it picks up as it grows, leaving some astonishing and chaotic heaps of debris when the ice melts. Tumbling that debris laden ice around even more could increase the chaos of the remains or create subtle terrain buildup. It would be something to keep an eye out for while investigating primary issues, but the concentric rings of Lake Huron invite explanation.
Supporting this is the pattern of sinkholes. the pattern of sinkholes. Like how the cenotes in the Yucatan peninsula have strong ties to the Chicxulub crater, marking a distant ring around it where subsurface earth was fractured but took a while to erode. Inside the escarpment, and mostly towards the lake borders, Michigan gets sinkholes where the underground water seeps into the lakes.
We’re pretty sure there weren’t similar lakes to the Great Lakes before these formed. It was largely flattened land or shallow seas covering vast swaths. So what changed? An impact could do it The impact distorted area is larger than normal because it was distributed thru a massive ice sheet. And while a lot of the ice sheet got blasted to kingdom come, as it were, and distributed to at least Nebraska and Georgia, I can imagine a scenario where the most fragile rock of the aquifer would collapse wholesale with a major impact PLUS the weight of an ice sheet. Just an impact wouldn’t distribute the water so far radially. Just ice sheet advances and retreats and the lakes would be shallower and have the rectilinear karst like features of the region. But shock destabilize it & squash it with ice before it can rebound? The water squirts sideways before it can come up. And it’s a lot of water. All that liquified land roiled, making it easy to wash away more ragged edges with the post glacial floods (and all that meltwater from the ice boulders). What’s left is very unstable patches of very wet quicksand like land that grew into a big, rounder lake like Michigan, and concentric rings of erosion in Huron.
(Updated, re Superior. Keep in mind the ice sheet was sinking the land to the NE so as it retreated, melt waters mostly ran that way across/into Superior before being shunted toward Michigan. I now don’t think there was a 2nd equivalent impact a bit west of the main impact at Saginaw bay. Disturbances at or near Lake Superior were likely secondary effects and runoff. I think it’s deeper, perhaps due to Glacial Lake Agassiz filling it with excess water runoff, ice boulders, and maybe ice bergs. (Possibly it’s composed of 2-3 larger initial lakes. Two spurs draining NE towards Hudson Bay, another running NW-SE along the stripe of terrain housing Canada’s great lakes, including Erie and Ontario. The underwater terrain is scraped mostly NE in the deeps. My nascent opinion is that water, ice, and rock slurry was trying to drain down toward Hudson Bay, got blocked by the ice sheet, and had to bank SE, at least closer to the surface, which maybe connected 2 pits, (or bays or trenches). It might have gotten water from Agassiz after the main impact damaged the ice sheet and lobbed a bunch of ice boulders into the lake like a frozen margarita. That sloshy run off could have made it to Superior, sloshed around a bit more, trying to escape N or to Michigan, where, for a while at least, it may have met with an accumulation of ice boulders. By the time it emptied the overflow, it wouldn’t have unduly carved up Erie on the way out. This is not the main thrust of my hypothesis, but the formation of Superior can’t be overlooked despite my focus being on the rounded, smoother shapes of Huron and Michigan.)
Adding - the standard explanation for Minnesota’s 10,000 lakes is that the glaciers deposited an uneven amount of debris when they receded, leaving many depressions to be filled with water. That’s probably mostly true But what if some lakes in MN & WI & the UP were bolstered by seismic shifting in underground aquifers when the overlying ice sheet vibrated from a meteor impact? The temporary force pushing michigan water to the edges wouldn’t be there, though the overlying ice sheet still was. The effect would be more evenly distributed water rising up rather than clumped around a central node. And possibly the ice boulders helped with residual rock boulder distribution.
Another piece of evidence about the meteor is fire. Huh? I thought this was all ice, all the time. No. The YD layer has a platinum spike indicative of impact, but also a layer of “black mat” which, some people report, smells like fire when initially dug up. This black mat is the destruction of the surface biomass that was so monumental it didn’t even properly decay fully. A large meteor coming in from the southwest to Saginaw could have heated the atmosphere enough to start fires before blasting the areas with ice boulders, thus pulping whatever is still there. If the shock wave thru the ice and into the ground went mostly NE plus some on the sides, it would explain why michigan isn’t an island in my hypothesis. It explains why Lakes Michigan and Huron have similar depths and form most of a ring. The incoming meteor’s direction wouldn’t have prevented surface ice from blasting back the way it came, as evidenced by most having circular craters, but the direction it slammed into the ice sheet would matter a bit more for under ice and under ground force propagation.
Anyway, watch geology videos. We’ve figured out a lot of wacky stuff in a very short amount of time, and it’s fascinating. By short time, I mean My high school geology teacher had a professor who stubbornly refused to believe plate tectonics existed. Today, any kid with an internet connection to satellite maps can look and say “wow, Africa fits *right there*, I wonder what happened?”
Side Quest the 3rd: While I’m here speaking heresy I will also record my stance that the tidal effects of the moon keep our planet’s core heated enough to preserve the magnetic field that protects our atmosphere and oceans from being blown away by solar effects like what may have happened on Mars.
Cautionary wording: No one in any of the links I posted or allude to has stated that they think a meteor caused, in effect, great clastic dikes that became the Great Lakes Huron & Michigan. So far as I know, it’s just me. So if you think my assertion is heretical, bother me, not the people detailing geologic processes from other places or proposing different effects, as I put these pieces together myself. (Some people have quite a lot of ego invested in earth sciences. My ego wants a good explanation more than I want to be the one to figure out the explanation.) If you think there’s something to my hypothesis, or really not, let me know where I can find more information. Because “retreating glaciers did it” isn’t good enough. The glaciers retreated all over the place and there aren’t circular features near Lake Manitoba. Those lakes had access to more water yet wound up shallower. I think the Glacial ice impact hypothesis is sound enough to be theory Working from there, the soil liquefaction and compression in the region makes sense, So yeah, Great Clastic Dikes plus subsequent erosion and some potential overflow from Agassiz helping dig out Superior.
But I dare you to look at Michigan and not see the ⭕️ . (Sorry but I don’t currently know how to add pictures stored on my phone to the blog.) Then I dare you to find other geologic features the size of a state that are circles but aren’t related to impact events.
Side Quest the 4th: My heart still clutches when I think they there used to be buffalo herds the size of Rhode Island roaming the plains. And my ancestors came over and wiped them out for sport But working through the impact and subsequent blanketing by ice boulders, plus immediate climate destabilization, North America lost most of its people, all its native camels & horses, most mastadons, all the dire bears, all the dire wolves, & all the sabertoothed tigers. Maybe buffalo herds that size were only possible relatively recently (last ~13,000 years) in the absence of co-evolved dire predators. In a land nearly denuded of large mammals, the buffalo made it. Until the US decided conquest was easier than cooperation, and all but killed them off.
Freeze, burn, bam, giant hail, jiggle jiggle, squirt, wash the debris away for 10K years, & there you have them, Great Lakes.
