Planetary Sciences News

Four Planetary Landscapes That Scientists Can’t Explain

These are just a handful of the hundreds of mysterious features across our solar neighborhood that beg to be studied closer.

By

Take a tour of the solar system, and you’ll find myriad mysteries. How old are Saturn’s rings? What carved out Mercury’s hollows? What created Iapetus’s weird ridge?

This year at the American Geophysical Union’s annual Fall Meeting in New Orleans, La., several scientists dedicated a poster session to just a pinch of solar system puzzles—including strange landscapes on our own planet.

“We thought it would be interesting to create a risk-free session” where scientists can “come with a totally off-the-wall idea and everyone will want to chat about it,” said Angela Stickle, a planetary scientist at Johns Hopkins University’s Applied Physics Laboratory in Laurel, Md., and co-convener of the session.

Perhaps some new collaborations will come out of the discussion, Stickle added. Maybe an Earth scientist will see an image of a strange feature on Mars and think, “Wait, I’ve seen something like that on Earth!” and a new partnership will flourish.

Here are four of these mysterious terrains, along with possible explanations for them. For more of the solar system’s wacky, unexplained morphology, check out the poster session “If You See Something, Say Something: Exploring the Weird and Wonderful Features of the Solar System Posters” today from 8:00 a.m. to 12:20 p.m.

Brainy Mars?

NASA’s Mars Reconnaissance Orbiter snapped this image of Mars’s so-called brain from several hundred kilometers away.
NASA’s Mars Reconnaissance Orbiter snapped this image of Mars’s so-called brain from several hundred kilometers away. Scientists estimate that the dark troughs are about 10 meters wide. Credit: NASA/JPL-Caltech/Univ. of Arizona

One curious landscape spotted on Mars is a vast expanse known as the “brain terrain.” Scientists spotted the landscape in 2013 as they began to study Arcadia Planitia, a vast, smooth plain in Mars’s northern midlatitudes. Some scientists posit that the roughly 10-meter-wide dark troughs are places where ice has been lost to sublimation—the process by which solid ice skips the liquid stage and just evaporates—while the bright spots still contain ice, said Nathan Williams, a planetary scientist at NASA’s Jet Propulsion Laboratory in Pasadena, Calif., and a poster presenter.

If this explanation is true, more questions abound. “How old is the ice?” Williams wonders. “What implications does that have on the climate history of Mars?”

Most important of all, “in the future, could ice from the brain terrain be used as an in situ resource for drinking, fuel, and/or agriculture?” Williams asks.

Willy Wonka and the Taffy Factory

Taffy-pull terrain
Image of the mysterious taffy pull terrain taken by the Mars Global Surveyor orbiter. The image covers an area 3 kilometers across. Credit: NASA/JPL

Meanwhile, half a planet away in the southern midlatitudes lies Mars’s taffy pull terrain, or simply “taffy terrain,” which looks like a swirling mass of the gooey candy trapped in time. Scientists found the terrain in the basin of a 2,300-kilometer-wide impact crater called Hellas Planitia and nowhere else.

“The taffy terrain bears some resemblance to submarine salt domes in the Gulf of Mexico, glacial deposits with mixed ash in Iceland, or chalk formation in Egypt’s white desert,” Laura Kerber, a planetary scientist at NASA’s Jet Propulsion Laboratory, wrote in her poster abstract.

However, scientists have “never seen anything like it, and it’s only on one place on Mars, so something weird is going on there,” said Tanya Harrison, a planetary scientist at Arizona State University who wasn’t involved in this particular poster.

The leading theory behind the taffy terrain’s origins involves landscapes containing different types of rock—some easy to erode, some harder to erode. As time passes, more easily eroded rock moves away, leaving behind the weird, flow-like pattern on the harder rocks, Harrison explained.

Carolina Bays

Carolina Bays
An aerial view of several Carolina Bays in North Carolina. The largest bay in this image stretches up to 3 kilometers long. Credit: George Howard

Our own blue planet isn’t lacking mysterious terrains. On the United States’ east coast, for example, stretching from New Jersey to Florida, hundreds of unexplained circular depressions pock the landscape. These depressions, which range in length from 180 meters to around 20 kilometers, are the Carolina bays—so named because a large cluster is found in the Carolinas. All of them are aligned northwest to southeast. And no one has a solid explanation for how they formed.

“Tens of thousands of these landforms [exist] around and amongst millions of people, and yet no serious geological undertakings have been made in the past 40 years to investigate them,” said Michael Davias, a researcher at Cintos Research, an independent group of citizen scientists, and presenter of a poster.

Theories of formation range from swarms of meteorites to wind, Davias said. Scientists seem to be divided into two camps: those who think some external force like impacts created the depressions and those who think the depressions formed from wind- or water-related erosion.

Bright Streaks

Dione’s bright streaks
NASA’s Cassini spacecraft took this image of Saturn’s 1,123-kilometer-wide moon Dione in 2008. Note the bright streaks in the lower right-hand region of the moon. Credit: NASA/JPL/Space Science Institute

One face of Saturn’s icy moon Dione is crisscrossed with unexplainable bright streaks. The Cassini spacecraft first showed scientists these streaks on Dione, as well as on the larger moon Rhea. Scientists have yet to figure out their origins.

When a mystery like these bright streaks pops up, “it makes you rise to the challenge” to get to the bottom of it, said Emily Martin, a planetary scientist at the National Air and Space Museum in Washington, D. C., and presenter of a poster on the topic.

The leading theory behind the bright streaks involves tectonics, but not the kind of tectonics we’re familiar with on Earth, Martin pointed out. No other plate tectonics like Earth’s exists in the solar system (except maybe on Europa), so the cracks and faults on Dione and Rhea result from the pushing and pulling of gravity as the moons orbit Saturn.

However, Martin is currently exploring an alternative theory involving impacts: Maybe an impact’s scour created the bright streaks, or perhaps a cloud of orbiting debris rained down on the moons.

—JoAnna Wendel (@JoAnnaScience), Staff Writer

Citation: Wendel, J. (2017), Four planetary landscapes that scientists can’t explain, Eos, 98, https://doi.org/10.1029/2017EO088617. Published on 11 December 2017.
© 2017. The authors. CC BY-NC-ND 3.0
  • Also, the allochthonous nature of the ~1600 cubic kilometers of sand in this unit geologic formation adds further doubt as to any locally derived origin. The high purity quartz grains of the Carolina bays sand are angular to sub-angular, shown no signs of typical terrestrial transport and are often reported to be highly fractured. Sometimes noted with a white paste appearing to be of pisolitic(?) nature, the bulk volume of the sand is commonly reported as monotonously uniform and completely void of absolutely any biotic detritus when examined at any distance from horizontal margins or upper/lower contacts of the unit. Lower contact is sometimes reported as a red/orange clay layer. According to the diagrams within the Zgonnik et al reference (Springer 2015), the concentration of the detected molecular hydrogen seems to vary directly (geometrically?) with thickness of the unit. The bays are observed to persist from a few meters below sea level in the Chesapeake bay to roughly 600 meters ASL in the Nebraska portion of the formation.
    Notably, the Carolina bays are only co-aligned in axial orientation when observed locally. Overall, the orientation of the 50,000 bays within the formation is remarkably systematic in radial alignment, all pointing back to the western margin of the Great Lakes, or slightly west of Lake Michigan. Those radial alignments tend to converge around Lakes Michigan, Huron and Superior when adjusted for Earth’s rotation during time of suborbital flight from that area to observed emplacement, for suborbital transport speed of approximately 3.4 km/s. The odds of co-alignment converging on one area for a given suborbital VEL being coincidence is on roughly the same order as 1 over 2 to the n-th power where n is the number of samples (50,000 and counting as more continue to be rigorously documented and mapped using modern LiDAR data and geographic tools such as Global Mapper). Likelihood of coincidence = vanishingly small. Try to calculate it. Hand held calculators typically fail beyond exponents of roughly 35,000.
    Also of interest, results of that massive mapping effort indicate that all 50,000+ cases conform to just 6 archetype shapes, a very strange (revolutionary?) finding indeed. The Zgonnik et al finding is both a blessing and a curse. The situation of quartz sand seeping hydrogen is strange indeed, and needs to be further examined over the entire region of the Carolina bays and through the depth of the unit formation, including (especially) beneath it. In the reported literature, the blanket averages only 2 to 10 meters thick, just thick enough for the lower contact to be out of reach for ground penetrating probes. Unfortunately that work (Zgonnik et al) took no measurements beneath the lower contact of the depositional sand blanket in which the bays are expressed at the surface. The work assumes a hydrogen migration pathway from deep in the geologic column, yet measures only to a few meters in depth, not enough to determine if the sand itself may be the source of the hydrogen seepage.
    The reason it is important to determine if the sand itself is the source of the hydrogen becomes clear when the KE required for suborbital transport to observed emplacement from the Great Lakes region is considered. If a Mid Pleistocene impact into the North American continental ice sheet shocked enough ice to transport the sand, the indicated KE would be 10 to 1000 time the KE of the Chixulub impact that finished off the dinosaurs, depending on the efficiency of comminuted sandstone & limestone target mass aggregate becoming entrained in expanding shocked ice. Ongoing suborbital analysis shows this may be the case as the distribution and transport ranges of the observed bays fits the Schultz & Gault model of oblique impact to ice overburdened targets (Prolonged Global Catastrophe From Oblique Impact… 1990). Such large KE scale event at such a recent date is “problematic” for any/all contemporary impact frequency models.
    In terms of dating the Carolina bays formation, no work has yet to put a successful and repeatable date on the geologic unit. Carbon or OSL techniques fail before the age indicated by river terrace date constraints where the bays do and don’t exist in the Carolinas. The organic matter or overlying sand or detrital units of various regions across the larger Carolina bay unit formation are naturally not representative of the unit formation itself. This explains why odd melt product found in a railroad cut through one of the bays was initially proposed as Chesapeake Bay impact remnant (35 Ma) and later proposed as possible Younger Dryas impact remnant (~13 ka). The river terraces where the Carolina bays are found are roughly constrained from 200 ka to 1.2 Ma, bracketing the mid Pleistocene. Those river terrace dating efforts assume constant sedimentation rate which would have been seriously violated if this formation is an impulsive depositional event (i.e. ejecta blanket). Those river terrace dating works also assume a constant uplift rate which also may have been violated if the energy scale of an ice sheet impact is as indicated by the KE required for transport of the peculiar sand from the Great Lakes region. Such transport would, however, give solid explanation for energetic steam plasma infusion into the highly fractured quartz grains.
    We do know that the the Australasian (AA) tektites, between 30 to 60 billion tons of impact melt glass, are known to have been formed at ~789 ka, and the more recently discovered Belize tektite strewn field is co-eval with but of different target mass composition than the AA tektites, implying multiple tektite forming events at the same epoch within error of measurement (Schwarz et al, “Coeval ages of Australasian, Central American and Western Canadian tektites reveal multiple impacts 790 ka ago”, Geochimica et Cosmochimica Acta, http://www.sciencedirect.com/science/article/pii/S0016703716300059). Still, after 50+ years of searching for the parent impact structure, in the Indochina region, no luck.
    Seemingly there were multiple “foreign objects” flying around the inner solar system during the mid Pleistocene, an idea also reinforced by the benthic mass extinctions leading up to, peaking at and tapering off after the mid Pleistocene (Hayward et al 2012, Cushman Foundation for Foraminiferal Research Special Pub # 43). What could cause such a time-distributed perturbation to the global assemblage of our benthic comrades, living where they do in the most stable environment on Earth? Hayward et al (2012) is a good synopsis of what went on in the mid Pleistocene. It isn’t an explanation, it is simply the measured truth of the imprint. For an explanation, we must be creative and astute. Our own imagination is the only limit of what we may conceive. Matching the imprint, the truth, is the only mandate. Hypotheses come and go. For the longest of stalled mysteries such as the Carolina bays or the Australasian tektites, we may even have to think the unthinkable. Whatever the new hypothesis, if it matches more elements of the imprint even while violating consensus, then we are left with a clear choice: embrace the truth of the imprint while abandoning some degree of consensus thinking, or remain stalled scientifically as a species. Each choice involves risk.
    The AA tektites being co-eval with those newly discovered tektites of Belize is a very unusual scenario indeed, since tektites are only formed by 1 or 2% of all known impacts on Earth. Is the type of projectile, or the target mass composition, or the impact condition, critically important for the formation of tektites? We are still not sure. The Australasian tektite imprint in particular contains a wealth of seemingly conflicting elements….
    We do know that a majority of the (gigantic) mass of the AA tektite strewn field rained down at velocities exceeding 90% of Earth escape velocity. We also know they typical quartz-rich minerals are vaporized at somewhat less than half of that post shock KE. So what was majority contributor of the AA tektite transport KE? Volatiles at the target would help explain this strange finding. Curiously the AA tektites don’t seem to have any odd volatile signature upon first look. They are largely de-volatilized and not significantly fractionated (not vapor condensate). The lack of volatile signature changes,however, with further consideration. Coming from a commonly accepted target mass assemblage of greywacke, sandstone and shale, the AA tektites are indicated as originating in the upper 1/2 meter of sedimentary target based on Al/Be nuclide ratio data within that substantial mass of melt.
    The strange iron oxidation state of the AA tektite melt is most illuminating. As one samples further and further south in the AA tektite strewn field, the composition and iron ox state become increasingly uniform. The observation that ever greater mixing was applied to the melt that landed further South suggests an impact to the North of the strewn field. Also of great significance, the re-entry KE yields 5 to 15 hrs of loft time depending on launch elevation. This is a direct result of the governing suborbital mechanics and is not open to debate. Averaging 10 or 12 hours of loft puts the Earth’s rotation beneath at ~1/2 day or on the order of 150 to 180 degrees of longitude from launch at the parent impact structure to observed emplacement. Inter-hemispheric transport of the AA tektites is indicated, without any wiggle room. Not just a good idea, “its the law”. More on the AA tektite iron ox state later.
    John A. O’Keefe wisely summarized the exceedingly large reentry velocity (bracketing Earth escape VEL of ~11.2 km/s) and concluded that AA tektites must have come from the moon. He may not have been too far off from an suborbital mechanics standpoint, an area of science relatively young in the 1960s when he first proposed Lunar Origin of the AA tektites. At that time, suborbital math was largely applied to ballistic “payload delivery” during the Cold War, with KE limited to what we could produce with man-made rockets, and solutions typically favoring minimum time of flight or minimum KE for tactical and practical considerations, respectively. J.A. O’Keefe was eventually proven wrong by improved geochem analysis tools that showed the AA tektite composition to be representative of Earth’s upper continental crust (“UCC”).
    During the historic period that UCC composition of the AA tektites became increasingly clear, the size of the strewn field was expanded by further finds, with AA tektites located in N.W. Canada, Antarctica and across the Indian Ocean Basin to Madagascar. O’Keefe never publicly addressed the UCC findings (to the best of my understanding, based on my readings from lots of various papers throughout that period), so there was no replacement of his former leadership on the topic. No one else with deep understanding of the suborbital implications seems to have addressed. S.C. Lin wrote a brilliant paper “Cometary impact and the origin of tektites” (1966, Journal of Geophysical Research) which was promptly blasted by more mainstream workers, and the consensus opinion continued to be that the impact structure would eventually be located in or near the centroid of the AA tektite strewn field, Indochina, where layered or Moung Nong-type tektites of the AA family are found. The layered Muong Nong-types are thermally fuzed but not liquified, with the occasional unmelted mineral grain such as zircons. Those zircons, coincidently, have a provenance matching the Michigan Basin. The 3 target rock types of greywacke, sandstone and shale have not yet been found together at the surface anywhere in Indochina. They are commonly co-located around the Michigan Basin, specifically on the shore line of the geologically anomalous Saginaw Bay of Lake Huron.
    Although many planetary scientists have been trying to rationalize a smaller and smaller parent impact structure for the AA tektites, some of those efforts ignore various seemingly conflicting elements of the imprint, such as the re-entry VEL above 90% of Earth’s escape velocity. Such conjecture is not helpful, and clouds the overall effort to get at the truth. We can not simply ignore inconvenient elements of the imprint and still expect to solve such longstanding mysteries as the Carolina bays or the Australasian tektites. The expansive size of today’s known boundaries place the the AA strewn field as the largest know tektite strewn field, by 2 orders of magnitude. This is an important fact for suborbital considerations of tektite transport, which are still lacking even in today’s AA tektite literature. This all boils back down to the applicable “laws and regulations”….
    Tektite strewn fields are not ejecta blankets. They are distal ejecta, not proximal ejecta. They do not follow the rules, formulas, distribution patterns or thickness predictions of ejecta blankets, because of that fact that tektite strewn fields are not ejecta blankets. Embarrassing as it is to be so trite and repetitive on this point, the emphasis sadly is completely necessary, due to the trend in the literature of various scientific workers (typically not orbit analysts, ever) to mistakenly assume that a parent impact structure of any tektite strewn field would be co-located with the centroid of that strewn field. This is especially true for the AA tektite strewn field because of it incredibly large scale, now known to cover 1/4 to 1/3 of Earth’s surface.
    Suborbital mechanics is the governing law of tektite transport. It mandates that for any given tektite strewn field, the distance from its centroid to its parent impact structure increases with the strewn field area. The AA tektite strewn field in particular, being so large, most likely has an impact structure located inter-hemispheric from its centroid. This is doubly the case for the re-entry velocity being near Earth escape VEL for this case, as explained earlier. No wiggle room. Its the law.
    There are very few cases of these tektite strewn fields to compare to, but the defining feature of a tektite is that it has been shock melted and vacuum quenched due to suborbital transport. So we know any ejecta blanket equation, distribution, depth or thickness profile or whatever other description is not applicable, and therefore represents a critical error when used to describe this rare species of ejecta. It is no surprise to me as an orbit analyst that the AA tektite parent impact structure remains unfound. Drawing straight lines across flat maps for any sub-global or global scale tektite strewn field is as obsolete as the Flat Earth Model itself. To review, the Flat Earth Model went out before the Channeled Scablands were identified as flood ravines, which has been accepted since before Plate Tectonics has been largely verified. So as modern scientists, we should be thinking at least 3 scientific paradigms, 3 modern academic epochs, beyond straight lines drawn on flat maps, thinking well beyond ejecta blanket models, to describe tektite distributions and transport distances. I don’t know how to make this any more clear.
    The same goes for any advanced hydrodynamic modeling of hypervelocity impact claiming to match the melt volume estimation of the AA tektite case while at once merely accounting for less than half of the known transport KE required to explain re-entry velocity of the majority mass fraction of that strewn field. That thinking is clearly obsolete, regardless of how well refined the equation of state of that hydrocode, or what the definition of the target, projectile or impact condition. Wrong model gives wrong answer, consistently. Five decades of failed search for the impact structure is a strong message. We must accept these realities, embrace what we do know to be truth within the imprint, and move on with seemingly conflicting elements of the imprint as our primary focus. If we don’t address the most difficult and stubborn facts facing us, we will continue to fail, endlessly. This brings us back to the Carolina bays….
    The ovoid shapes of the Carolina bays are identical to ballistic targeting diagrams. They are identically reproducible by a suborbital obstruction shadowing model applied to the ascent phase of sand as comminuted target mass during suborbital transport. The pattern is so incredibly clear that I am embarrassed to admit it took me a few years to realize it. The 6 shapes of the 50,000+ Carolina bay shallow depressions in a massively distributed sand blanket are highly representative of gas dynamics and variable impedance during entrained outflow from a colossal impact into volatile target mass. The patterns of the Carolina bay shapes, their distribution and their co-aligned pointing back to nearly the Great Lakes fall out of suborbital mechanics so clearly that no accounting of re-entry effects is necessary to the first order. This is understandable since the blanket mass is 3 to 5 times that of the atmospheric column, and the indicated launch/reentry VEL of ~3.5 km/s.
    Regarding the uniform iron ox state of the 30 to 60 billion tons of AA tektite melt (just on the 3+ side of 2+), it is match by steam plasma (i.e. shocked ice sheet) at an astoundingly high temp of roughly 45,000 K. This is only explainable by a highly oblique and large scale cosmic impact into an ice sheet, where the forcing length along the oblique direction is extended to many times (hundreds of times) the thickness of the ice sheet. The black body radiant wavelength at that temp is squarely in the far UV, which coincidently also lies within the 100% absorption band of quartz. The radiant power at that temp is roughly one million watts per square centimeter. Those words are typed out, not as numbers, to avoid any confusion. It is enough power to melt a 1 centimeter (one cm) layer of quartz every 7 (seven) milliseconds assuming unity as the transfer coefficient, not necessarily a bad assumption when considering a few km/s relative velocity between astronomically shocked ice sheet and substrate sedimentary bedrock. Tektite formative environment? This scenario qualifies.
    Lastly at least for the moment, a comparison to the Ries impact and its central European tektites is informative. The ratio of excavated mass from the Ries and partner binary crater to estimated Moldavite tektite mass gives about 10 to the seventh. Applying this ratio to the estimated mass of the Australasian tektite melt gives 10 to the 21 or 10 to the 22 grams of excavated mass from whatever the impact structure. These values straddle the mass of earth’s entire atmosphere. It is too big to understand or explain if the excavation was bedrock. If that excavated mass was largely ice sheet however, the answer suddenly makes sense. This would be the volatile mass that delivered a majority of the transport KE to the AA tektites (otherwise unexplained by shock alone to the un-vaporized tektite melt), and that entrained the Carolina bay mass to its huge indicated KE to emplace that massive formation from the Great Lakes region. The mass ratio and distance ratio of “medial” Carolina bay sand ejecta and distal AA tektite ejecta also fall neatly into power law decay when both are considered as originating from the Great Lakes region, so general consensus of ejecta transport are not required to be abandoned.
    As additional comment(s), the AA tektite strewn field is easily populated along the NE-to-SW axis of Lake Huron along the Saginaw Bay centerline of that lake using only a variation in elevation angle, and less than 10% VEL variation thorough said elevation. The outflow time constant of that amount of shocked ice in a fire ball is roughly 40 minutes, based on the same “Effects of Nuclear Weapons” reference used in the “Earth Impact Effects” work of recent Planetary Science. The tangential impulse to Earth’s crust may have been enough, according to this scenario, to throw its rotational rate out of sync with Earth’s solid core, disrupting the convective cellular structure of the geomagnetic dynamo. This would explain not only Earth’s most recent Geomagnetic reversal 12 to 16 ka after the AA tektite event, but also the precursor spike of that geomag collapse that was coincident with the tektite event.
    The risk of thinking the unthinkable with respect to these long standing Earth Science mysteries is that we may have to revise our conventional paradigm. The risk of ignoring these roughly 50,017 coincidences is that large impacts take place on Earth in contemporary geologic time scales (AA and Belize tektites), and we are failing to explain how they could have happened for lack of any parent impact structure as the demonstrated causal mechanism. It is a potentially large and as-of-yet unquantified risk. Can we live with that? Maybe….
    I’m not even going to address the catastrophic remnant in the column surrounding and beneath the AA tektite layer in N.E. Thailand and other places in S.E. Asia, for example. Or the mid Pleistocene climate cycle shift that took place over the same time scale as the benthic extinctions, or the disappearance of Java Man from Indonesia, or a few other associated mysteries of that period. They may all be completely unrelated. We will never, ever know for sure if we don’t find the AA tektite impact structure or resolve the Carolina bays causal mechanics. The clock is running. 5 to 7 decades for these mysteries to persist does not reflect well in our report card as scientists, as a species, and as the inhabitants of a single planet. Time to get busy.
    Its time to be creative and take some intellectual risks. That is the message of the AA tektites and the amazing Carolina bays unit geologic formation that seems to be nearly purely quartz sand, with no biotic detritus, that exists over a range of altitudes from sea level to 600 meters, and that is apparently seeping hydrogen (?!?).

    • Patrick Mangou

      I am familiar with Zgonnik’s article. Other geochemical theories have also been suggested such as karstic solution. It’s true that even if hydro-eolian processes are responsible for the shape of the bays, the initiation process has to be defined. Geochemical and structural (faults) explanations are seducing, but I have doubts whether they can be satisfactory for huge areas (from NJ to Fla. even if the largest numbers are in the Carolinas). Perhaps the initiation of the bays are just related to random slight irregularities in the surface of a recently emerged coastal plain, made of loose, unconsolidated material. Once more, the location of the bays in badly-drained interfluves has be considered in any explanation.

  • Several interesting and relevant facts regarding the Carolina bays are not mentioned in the article or comments to date. Zgonnik et al (Springer, Progress in Earth and Planetary Science, 2015) measured molecular hydrogen in the depositional sand blanket in which the bay depressions and their rims is expressed, with “the hypothesis that Carolina bays are the result of local collapses caused by the alteration of rock along the deep pathways of H2 migrating towards the surface. The present H2 seepages are comparable to those in similar structures previously observed in the East European craton.”
    The roughly ovoid shapes, presence of hydrogen and location over a cratonic margin of the Carolina bays are indeed similar to the East European case, yet the similarities end there. There are no local collapses of the bedrock column beneath the Carolina bays reported in the literature. Generally the Carolina bays depositional sand blanket overlies several ancient antecedent formations across its roughly 400,000 square kilometer geographic expanse.

  • Patrick Mangou

    About the Carolina Bays:
    Some of them were present in the area of South Carolina I studied for my PhD in the 1980s, hence my interest.
    I am always surprised to hear that some credit is still given to the meteorite theory (and other even crazier speculations as fish digging holes when mating). Not only the Bays (I capitalize the word Bay to avoid any confusion with any other kind of bay) have not the shape of meteorite craters, but absolutely no meteorite material has ever been found in them.
    The whole thing has been debunked by Kaczorowski in his overlooked PhD dissertation (KACZOROWSKI, Raymond T., THE CAROLINA BAYS AND THEIR RELATIONSHIP TO MODERN ORIENTED LAKES. University of South Carolina, Ph.D., 1977, available at the U. of SC library and on ProQuest). He showed the action of prevailing winds in the digging of the Bays and the building of sandy rims, especially on the NE corner.
    But, in my opinion, Kaczorowski didn’t go far enough. The carbon-dating of the peat filling the Bays dates the Bays themselves from about 50 Ka to 6 Ka. That means they formed in a periglacial climate, cold in the winter and dry at times. Wind erosion was very efficient in lakes or maybe depressions without much vegetation. The sandy rims represents the sorting by the wind of heavier grains which remained on the rims, while the finer, lighter, clay-sized grains where swept much farther.
    The most overlooked feature of the Bays, which definitely eliminates any fancy theory, is their exclusive location in interfluves. No stream ever crosses a Bay, which proves they are a phenomenon of bad drainage in a very flat coastal plain (still badly drained at this time: swamps, flooded areas).
    The only thing special about the Bays is their regular shape, hence the fascination. But once one understands the role of paleowinds, there’s no mystery any more.

    • In his defense of the gradualistic position on the geomorphology of the Carolina bays, Patrick Mangou has provided a concise and competent summary of the “wind & wave” proposal that is widely accepted in the geological community. If I might add to his fine summary, the most damming finding against an impact-related phenomenon is the total absence of the requisite impact structure.

      Among the challenging aspects for the Carolina bay researchers is the diverse dates assigned to their contents. Patrick Mangou has chosen to focus on an unfortunate one. Carbon dating, having been applied on bay fill across decades, can only reach the 50ka age he noted. In every case I have seen, the report of that date has been properly caveated as “no younger than”, because the deep organic sediments are typically carbon-dead. Of course, the in-filling by organics and silioclastic eolian particles is ongoing, so there are copious contemporary dates applicable. OSL dating, capable of reaching back as far as 120 ka, has offered proof of bay fill at least as old as that constraint.

      Dr. Kaczorowski’s thesis paper has provided the most recent (1977) thesis-level scholarly effort to explain the bays, and although it was never published as a peer-reviewed paper in the literature, it stands as a fine example of research. His university did make copies of the double-spaced thesis text and addendum graphics available, an inscribed original of which is in my personal collection. His experiment to produce a “Carolina bay” planform needs a bit of qualification, in my view. First, (as in all wind & wave explanations) he mandated that a circular depression, water filled, be provided as the starting point. His regime of wind was provided by an oscillating fan, sweeping back and forth over the incipient bay (i.e., not a planar wind field). In addition, his protocol worked only when he shifted that oscillating wind field 180º on a 50-50 duty cycle at regular intervals. When the commonly-invoked glacial era “katabatic winds” are applied by the gradualist community, they conveniently fail to even consider a 50-50 duty cycle of predominant wind – because that is quite unrealistic for winds pouring off the Laurentide ice sheet hundreds of km to the north. Kaczorowski commented in his thesis that the helical nature of the artificial wind field needed to be corrected in future work, but there is nothing in the literature to suggest that the experiment has been replicated nor successfully attempted with improved protocol. Finally, our survey has shown that 50,000 measured bays robustly conform to a handful of elliptical ovoid planforms, none of which have the shape arrived at in Kaczorowski’s experiment – which had tight radiuses at both ends of the primary axis.

      Patrick Mangou suggested that all Carolina bays lie on undissected interfluves, and that cannot be debated. What is important to note is that when Carolina bay becomes dissected by headward fluvial erosion, it remains visible only in the high-resolution LiDAR-derived elevation data (and they do in prodigious quantities) because it is no longer hydraulically closed and often does not provide the wetland so commonly viewed as a “Carolina bay”. Prouty observed that the few water-filled Carolina bays that do exist are set in areas of high water table, and Wells noted that peat is not endemic to open water-basins, but only grows where moist conditions exist. While it has been postulated by the gradualist community that bays currently without open water (the clear majority) necessary to the wind & wave model necessitated open-water conditions only during the glacial-age katabatic wind regime. It is an inconvenient truth that glacial eras are well known to have been far drier than the interstadials.

      In South Carolina, fully 200 meters above sea level, and 200 km inland from the coast, lies a continuous drainage divide that runs from Augusta, GA to Columbia, SC. LiDAR-based DEMS show there to be over a hundred crisply described Carolina bay basins on that ridge, which has been identified by other workers as a surviving remnant of the once-extensive Cretaceous terrace. In the far future that slender divide will be compromised and dissected, and at that point the bays will be eroded away – but for now they stand as excellent examples of bay existence far away from the high-water table interfluvial realms closer to the Atlantic coast.

      All this is provided to explain why some feel the gradualistic solution needs to be questioned.

      Certainly, the bays are not primary or secondary impact structures, as their rim relief over their vast spatial extents belies any connection to an excavated structure. Our goal with the Survey is to visually and statistically demonstrate their robust conformance to a very specific set of planform shapes, suggesting (for an out-of-the-box example) a cosmic agent such as a large-scale mass flow away from an ice-sheet encased target. The publically-accessible Survey orientation data has been used by us to ascertain that a locus exists using a network triangulation that recognizes mass transport over a rotating planetary surface. As for there being no identified impact structure that could be implicated in the creation of the bays, it is imperative to acknowledge that the impact structure has yet to be located for an event that is widely accepted as being the most energetic over the past tens of millions of years. The vast Australasian tektite strewn field was generated by an impact 800 ka (just yesterday geologically), and excavated freshwater sandstone and greywacke sediments from a continental setting that should be easy to locate, yet has failed to be in over 50 years of research.

    • Joan Lederman

      The inquiry is compelling and the welcoming attitude is an important step to gathering insights from any/everywhere but I object to superimposing images by naming phenomena like “taffy” and “brains” a hindrance to fresh-view thinking. In the last decade I’ve observed titles of scientific talks and articles tend to be misleading because of the need (or fad?) to attract interest. This is the art of articulating…….and where discerning is a vigilant practice.

    • In his defense of the gradualistic position on the geomorphology of the Carolina bays, Patrick Mangou has provided a concise and competent summary of the “wind & wave” proposal that is widely accepted in the geological community. If I might add to his fine summary, the most damming finding against an impact-related phenomenon is the total absence of the requisite impact structure.

      Certainly, the bays are not primary or secondary impact structures, as their rim relief over their vast spatial extents belies any connection to an excavated structure. Our goal with the Survey is to visually and statistically demonstrate their robust conformance to a very specific set of planform shapes, suggesting perhaps (for an out-of-the-box example) a cosmic agent such as a large-scale mass flow away from an ice-sheet encased target. The publically-accessible Survey orientation data has been used by us to ascertain that a locus exists using a network triangulation that recognizes mass transport over a rotating planetary surface. As for there being no identified impact structure that could be implicated in the creation of the bays, it is imperative to acknowledge that the impact structure has yet to be located for an event that is widely accepted as being the most energetic over the past tens of millions of years. The vast Australasian tektite strewn field was generated by an impact 800 ka (just yesterday geologically), and excavated freshwater sandstone and greywacke sediments from a continental setting that should be easy to locate, yet has failed to be in over 50 years of research.

      Among the challenging aspects for the Carolina bay researchers is the diverse dates assigned to their contents. Patrick Mangou has chosen to focus on an unfortunate one. Carbon dating, having been applied on bay fill across decades, can only reach the 50ka age he noted. In every case I have seen, the report of that date has been properly caveated as “no younger than”, because the deep organic sediments are typically carbon-dead. Of course, the in-filling by organics and silioclastic eolian particles is ongoing, so there are copious contemporary dates applicable. OSL dating, capable of reaching back as far as 120 ka, has offered proof of bay fill at least as old as that constraint.

      Dr. Kaczorowski’s thesis paper has provided the most recent (1977) thesis-level scholarly effort to explain the bays, and although it was never published as a peer-reviewed paper in the literature, it stands as a fine example of research. His university did make copies of the double-spaced thesis text and addendum graphics available, an inscribed original of which is in my personal collection. His experiment to produce a “Carolina bay” planform needs a bit of qualification, in my view. First, (as in all wind & wave explanations) he mandated that a circular depression, water filled, be provided as the starting point. His regime of wind was provided by an oscillating fan, sweeping back and forth over the incipient bay (i.e., not a planar wind field). In addition, his protocol worked only when he shifted that oscillating wind field 180º on a 50-50 duty cycle at regular intervals. When the commonly-invoked glacial era “katabatic winds” are applied by the gradualist community, they conveniently fail to even consider a 50-50 duty cycle of predominant wind – because that is quite unrealistic for winds pouring off the Laurentide ice sheet hundreds of km to the north. Kaczorowski commented in his thesis that the helical nature of the artificial wind field needed to be corrected in future work, but there is nothing in the literature to suggest that the experiment has been replicated nor successfully attempted with improved protocol. Finally, our survey has shown that 50,000 measured bays robustly conform to a handful of elliptical ovoid planforms, none of which have the shape arrived at in Kaczorowski’s experiment – which had tight radiuses at both ends of the primary axis.

      Patrick Mangou suggested that all Carolina bays lie on undissected interfluves, and that cannot be debated. What is important to note is that when a Carolina bay becomes dissected by headward fluvial erosion, it remains visible only in the high-resolution LiDAR-derived elevation data (and they do in prodigious quantities) because it is no longer hydraulically closed and often does not provide the wetland so commonly viewed as a “Carolina bay”. Prouty observed that the few water-filled Carolina bays that do exist are set in areas of high water table, and Wells noted that peat is not endemic to open water-basins, but only grows where moist conditions exist. It has been postulated by the gradualist community that bays currently without open water (the clear majority) necessary to the wind & wave model only required open-water conditions during the glacial-age katabatic wind regime. It is an inconvenient truth that glacial eras are well known to have been far drier than the interstadials.

      In South Carolina, fully 200 meters above sea level, and 200 km inland from the coast, lies a continuous drainage divide that runs from Augusta, GA to Columbia, SC. LiDAR-based DEMS show there to be over a hundred crisply described Carolina bay basins on that ridge, which has been identified by other workers as a surviving remnant of the once-extensive Cretaceous terrace. In the far future that slender divide will be compromised and dissected, and at that point the bays will be eroded away – but for now they stand as excellent examples of bay existence far away from the high-water table interfluvial realms closer to the Atlantic coast.

      All this is provided to explain why some feel the gradualistic solution needs to be questioned.

    • In his review of the gradualistic position on the geomorphology of the Carolina bays, Patrick Mangou has provided a concise and competent summary of the “wind & wave” proposal that is accepted in the geological community. If I might add to his fine summary, the most damming finding against an impact-related phenomenon is the total absence of the requisite impact structure. Certainly, the bays themselves are not primary or secondary impact structures, as their rim relief over their vast spatial extents belies any connection to an excavated structure.

      The LiDAR DEM maps of the bays, such as the examples provided along Carroll Road here, present these landforms in a light not available until recently. They provide incontrovertible evidence the bays are well described along 360° of their rims, not only along a specific quadrant. The vast percentage of bays appear as basins sunken into elevated platforms, as seen here. While some bays show well-described dunes attendant to their south-east rims, examples such as these are devoid of prevailing wind dune structures.

      Our goal with the Survey is to visually and statistically document their robust conformance to a very specific set of ovoid planform shapes, to accurately document their orientations, and to better understand their persistence in the LiDAR even when subjected to headward erosion that breaches their rims and drains their basins.

      • Patrick Mangou

        Thanks to Michael for his insights. I acquired the LIDAR DEMs for Woods Bay, SC,which I studied in the 1980s (and which is preserved as a State Park, unfortunately with a marker perpetuating the meteoritic origin) and they confirm the exixtence of a rim all around, although the highest dunes are on the NE and SE corners.

    • Richard Wagener

      For an idea of what these might have looked like during the ice-ages, have a look at some landsat images of the lakes on permafrost around Barrow, Alaska.
      Here is an old one I happen to know about, but I am sure you can find better quality examples.
      https://www.xdc.arm.gov/data_viewers/nsa_surfchar/LandsatNSACart.html

      • Patrick Mangou

        Thanks for the link; Kaczorowski mentioned these oriented lakes, along with similar ones in Chile (Tierra del Fuego) in his dissertation. Carolina Bays could have been initiated as thaw lakes or collapsed pingos. I think there’s some literature about pollen studies in the bays’ paleosols (at least Frey’s 1953 article). They should provide some indication about the bays’ paleoclimates to verify if the North Slope lakes can be considered as a modern equivalent of the bays. Kaczorowski didn’t address the pollen question.