The Coconino Sandstone

"There are no modern analogs that match the precise sedimentology of the Coconino, but we believe that subaqueous sand waves may be a start in the right direction to understand how the Coconino was deposited." — John Whitmore and Paul Garner, 2018¹

"The trackways have features that seem virtually impossible to explain unless they were made with the animals completely underwater, while the sedimentary evidence, as interpreted by sedimentologists, seems to point to wind-blown sand." — Leonard Brand, who ran the trackway experiments, 2018²

"Helble is correct that the processes he outlines could not have formed the Coconino during the Flood." — John Whitmore and Paul Garner, 2025³

Summary

Three hundred feet of clean white cross-bedded sandstone sit near the top of the Grand Canyon wall.  The layer is the Coconino.  It covers about 32,000 square miles, and it is the rock young-earth geology most often has to explain away, because desert dunes are hard to fit inside a year of global flooding.

One group of creationists set out to show the dunes were never dry.  John Whitmore of Cedarville University has run the only sustained creationist field programme on the formation:  more than thirty outcrops under a Park Service permit, hundreds of thin sections, X-ray diffraction and electron microscopy.  He has published in Sedimentary Geology and the Scottish Journal of Geology.  His student Sarah Maithel published her doctoral work in Sedimentology and the Journal of Sedimentary Research.  This is real fieldwork.  The argument below is with the data, not the affiliation.

Two questions have to be kept apart.  Was the Coconino laid down by wind or under water?  And could it have formed in a one-year Flood?  The first is a live dispute with real creationist data in it.  The second is not, and the reason is arithmetic that the leading creationist Coconino researcher has now conceded in print.

Eight conclusions.

1. Wind, in a coastal dune field beside a shallow sea. The strongest single block of evidence is Edwin McKee's 1945 measurement of about fifty Coconino ripple sets against five criteria. The creationist literature has never addressed four of them.⁴
2. The creationists are right about the cross-bed dips, and the mainstream number in circulation is wrong. Whitmore's pooled Arizona mean is 20.2°, from hundreds of measurements at five named localities.⁵ Maithel's independent doctoral work, peer-reviewed in Sedimentology, agrees: "most dip angles ranging from the mid-teens to mid-twenties."⁶ Carol Hill and Stephen Moshier's uncited "29-31 deg" should not be repeated.⁷
3. The argument built on that number still fails. About 20° is not the subaqueous prediction either. And a cross-bed set preserves the lower, tangential part of the foreset, not the steep slipface, so a mean over a whole set comes out below the angle of repose on any model.
4. Two of Whitmore's eleven points fail on his own team's sources. Item 6 says avalanche tongues are missing; McKee reports dry-sand avalanche marks in the Coconino by name, and Maithel's first conference paper was titled "Morphology of avalanche beds in the Coconino Sandstone." Item 10 says the dune-diagnostic lamination types are "for the most part absent or have been difficult to recognize"; Maithel's own published conclusion is that the cross-beds "may consist of grainflow and grainfall deposits, but these remain difficult to differentiate."⁶ Present but hard to tell apart is not absent.
5. Leonard Brand's own experiment removes the underwater diagnostic, and he said so. In 1996 he published 230 newt trackways across ten substrate combinations. The distorted toe marks treated as the underwater signature appear "if the animals were walking underwater or on sloped, damp sand."⁸ His own Discussion concludes that the Coconino trackways "may have been produced in either subaqueous sand or subaerial damp sand."⁸ Sloped damp sand is what a dune face becomes after an overnight shower.
6. The eolian side answered Brand with a mechanism and a successful prediction. Lorenzo Marchetti and colleagues, including three lizard-behaviour specialists, re-read the tracks as digit sliding on an inclined dry slope. They invalidated the ichnogenera Chelichnus and Laoporus as misread drag marks. Once slope effects are subtracted, the Coconino track fauna matches the fauna of non-desert sites of the same age.⁹⁹
7. The Flood question is settled separately, by a sediment budget. Timothy Helble gave flood geology ten times its own transport rate and nine further favourable assumptions. He still got 26 cubic kilometres of sand delivered against 42,000 required, about one part in 1,600.¹⁰
8. The published creationist reply concedes it. Whitmore and Garner in 2025: "Helble is correct that the processes he outlines could not have formed the Coconino during the Flood, but the Flood model requires extrapolations beyond conditions that prevail in the present-day."³ Their own 2018 paper says "There are no modern analogs that match the precise sedimentology of the Coconino."¹ The subaqueous Coconino has no observed analogue and no quantified mechanism.

Where the creationists are right

Four corrections came from the young-earth side, and the mainstream literature should adopt them.

The dips.  About 20°, measured hundreds of times at named localities and corroborated by a Sedimentology paper.  Their answer to the obvious escape is good petrography:  high porosity, unfractured grains and undeformed dolomite ooids show the rock was never severely compacted.¹

Mica and angular potassium feldspar.  Muscovite is very soft and turns up in almost every Coconino thin section.  Potassium feldspar is softer than quartz and is often more angular than the quartz beside it.¹¹  Their abrasion experiments are published in Aeolian Research.  The mainstream literature has not answered this in print.

Frosted grains do not prove wind, and a creationist team showed it.  Whitmore's electron-microscope work found Coconino grains too small to frost by collision, with frosting that is chemical overgrowth.¹¹  That retires an argument the eolian side had been using.

The sand-filled cracks at the base are injectites, not desiccation cracks.  Whitmore and Strom established that with fieldwork at thirty sites, published in Sedimentary Geology.¹²  It corrects McKee, whose own wording had been a guess.

The witness most readers will not expect

The man who ran the underwater trackway experiments does not claim what the popular literature claims for him.  Leonard Brand wrote in his 2018 textbook that the sedimentology "seems to point to wind-blown sand," that the contradiction "indicates there are some unknown pieces of the puzzle that remain to be discovered," and that "This research does not show whether the tracks were or were not made in the biblical flood."²²

Part 1: What the rock is

The Coconino Sandstone is Permian, Kungurian stage on the current time scale.  In the Grand Canyon it sits between the red Hermit Formation below and the Toroweap Formation above, and runs 20 to over 183 metres thick.⁹  Across its whole extent McKee gives 40 feet at Lees Ferry, 60 feet at Hacks Canyon, "more than 500 feet (150 m) thick along the Mogollon Rim of central Arizona" and "a maximum thickness of more than 1,000 feet (300 m) near the town of Pine."⁴  Its area is "about 32,000 square miles (51,500 km2)."⁴

Two things about it are unusual, and both sides use them.

It is almost pure quartz sand with no body fossils.  The National Park Service inventory says the Coconino "consists entirely of quartz sandstone."⁹  McKee, forty years earlier:  "No skeletal parts of the vertebrate animals responsible for Coconino footprints have yet been found."⁴  A desert with no standing water predicts that.  So does a high-energy transport regime.  On its own it decides nothing.

It is not an interior desert.  The creationists helped change this part of the picture.  Ronald Blakey, a mainstream sedimentologist, found a marine sand-wave complex in the Permian Schnebly Hill Formation of central Arizona that grades sideways into Coconino rock.  Fisher reported tongues of fossil-bearing marine limestone inside the Coconino itself.  The Hermit-Coconino contact is gradational along the Tanner Trail and in parts of Parashant Canyon.¹  Whitmore and Garner assembled that list to argue the Coconino is marine.  What it shows is a shoreline that moved back and forth beside a dune field:  a coastal erg.  McKee himself classed the Lyons and Casper Sandstones as coastal-dune deposits in the same chapter.  The popular picture of a Sahara with no sea in sight is overdrawn.

Coastal is not the same as underwater.  Marine sand waves below an erg, limestone tongues within it and a gradational contact at one trail are what the edge of a dune field beside a shallow sea looks like.  They do not put the dunes themselves under water.

The two claims that get confused

Brand's underwater hypothesis and Steven Austin's Flood model are not the same claim.  Running them together is the commonest mistake in this argument.

Brand needs quiet water:  animals walking on a submerged sand surface calm enough to hold a footprint.  Austin needs a torrent:  bedforms moving under currents of 3 to 5 metres per second beneath about 100 metres of water, fast enough to shift 42,000 cubic kilometres of sand in under two weeks.  Carol Hill and Stephen Moshier, two old-earth Christian geologists, put the collision plainly:⁷  "Incredibly, flood geologists envision these land animals walking on dunes that they propose were moving under currents of 3 m/sec (or more) beneath about 100 m of water!"⁷

Nothing in Brand's work supports Austin's, and Brand says so himself.

Part 2: The eolian case, from the primary source

Almost everyone arguing about the Coconino is arguing about Edwin McKee.  His statement of the eolian case is a chapter in the United States Geological Survey's A Study of Global Sand Seas, Professional Paper 1052, printed pages 197 to 203.⁴  It is free, it is public domain, and it is more careful than its reputation.

Cross-strata (p. 199)

"The most conspicuous primary structure preserved in the Coconino Sandstone is the large-scale cross-stratification.  It is composed of long sweeping layers, many of them 30-40 feet (9-12 m) long, and some of them as long as 80 feet (24 m).  They dip mostly at 25 deg - 30 deg, but a few reach a maximum of about 34 deg."⁴

That is a distribution with a tail reaching the angle of repose, not a mean.  McKee also records "some low-angle (< 15 deg) cross-strata that dip in the opposite direction" in a few localities, and horizontal laminae in the basal one to two metres at many.⁴

What McKee refused to claim (p. 199)

"Because sorting and texture are functions of transportation rather than deposition, and, further, may reflect processes of an earlier generation, they cannot be considered to be proof of eolian origin.  Grain characteristics of the Coconino are, however, typical in all respects of many modern dune sands."⁴

His own description of the grains is "rounded to subangular," not uniformly well rounded.⁴  Four of Whitmore's eleven points are textural.  They are aimed at a claim the founder of the eolian model disclaimed on the page where he made his case.  Whitmore is right that textbooks repeat "well-sorted, well-rounded" carelessly.  McKee is on record that texture was never the argument.

The ripple marks (p. 200), the strongest evidence anyone has

"A detailed analysis made in 1945 of ripple marks in the Coconino, in which about 50 sets were examined, shows compelling evidence, on the basis of five characteristics, that the ripple marks were wind formed (McKee, 1945, p. 317-320).  The diagnostic features are as follows:  1. Large ripple index (lowest was 17; most were considerably higher).  2. Ripple crests were definitely rounded in all specimens.  3. A concentration of coarser sand grains occurred on or about the crests in most specimens.  4. All specimens in situ had crests trending up and down surfaces of sloping laminae; none was normal to dip.  5. Crests and troughs were straight and parallel, with little change in direction where exposed for distances of 3 feet or more."⁴

Three of those five are the reverse of what moving water does.

Ripple index is wavelength divided by height.  Wind ripples run high, above about 15.  Current ripples in water run below it.  McKee's lowest value was 17 and most were higher.

Coarse grains on the crest.  Wind bounces fine grains up the windward side and leaves the coarse grains riding the crest.  Water currents do the opposite and winnow coarse material into the troughs.

Crests parallel to dip.  Wind ripples on a lee face climb along the slope, which is the geometry that produces climbing translatent strata.  Current ripples built by a flow running down a foreset would cross it.

McKee also explains why Coconino ripples are scarce, which is the obvious objection to his own evidence.  Most ripples in modern dunes form on the gentle windward slopes, "but in ancient dunes, because of differential preservation, most windward-side deposits have not been preserved."⁴

Nothing in the creationist corpus engages the ripple index, the rounded crests or the coarse-grain concentration.  Whitmore and Garner confirm the fourth criterion themselves, writing that Coconino ripple-like features "are always parallel to dip," and pass over what that means.¹  If this article names one thing the subaqueous model has never answered, it is McKee's 1945 ripple measurements.

Rain prints (p. 200), which this article does not use

McKee's rain-print argument has a specific, testable mechanism.  A drop landing on dry fine sand welds a thin, detached shell of grains.  On a sloping foreset the shell rotates so the pit faces upward, raising its downslope rim.  A thorough wetting destroys it:  "Only brief showers can leave such a record, for on the surface of a dune that is thoroughly wetted, the shells lose coherence."⁴  He treats the pits as "positive evidence of subaerial deposition."⁴

Part of Whitmore and Garner's attack on this lands.  Some Coconino "pits" pass right through a thin bed and appear on both faces.  Some are the moulds of pea-sized nodules that fell out.  Some are the entry and exit points of burrowing animals.  At Ash Fork and Seligman they occur in rows parallel to dip rather than scattered at random.¹

Both things can be true:  some pits are misidentified and some are rain prints.  Neither side has tested against the morphology McKee described, and the creationist follow-up exists only as a 2021 conference abstract.  This article treats rain prints as undecided and rests nothing on them.

Slump marks and contorted bedding (pp. 200-202)

McKee reports dry-sand slump and avalanche marks on Coconino surfaces, cites the experimental work that distinguishes slump marks formed in dry, damp and wet sand, and concludes that "Those described from the Coconino Sandstone seem to give further strong evidence of the formation's eolian origin."⁴

On the larger contorted beds he gives ground, and Whitmore's parabolic recumbent folds are aimed at exactly this:

"The sand apparently had some coherence at the time of contortion, but whether the strata were wet or dry is not clear, for this type of fold has been developed under both conditions."⁴

The eolian side had already marked Coconino soft-sediment folding as undecided in 1979.

The animal experiments (pp. 202-203)

This is the part of the eolian case most people have never heard, and it is experimental.

McKee piled loose Coconino-derived sand at Coconino dip angles and put live animals on it:

"On loose sand derived from the Coconino and deposited at high angles (25 deg -28 deg) comparable to that in crossbedding of the Coconino, a chuckwalla lizard was enticed to walk both up and down the surfaces of dry loose sand, damp (sprayed) sand with a crust, and wet sand.  Only in dry sand, uphill, could footprints be formed similar in size and clarity to those of the Coconino.  Damp and wet sands gave very different results, and tracks made downhill were consistently destroyed by avalanching."⁴

The last clause matters.  It is a mechanism for the uphill bias, otherwise the strangest thing about Coconino trackways:  the tracks "with only three exceptions out of many hundreds examined, were all leading uphill."⁴  If downhill tracks destroy themselves on a dry slope, the uphill bias is a preservation filter, not a behaviour.

He ran the invertebrates too.  Millipedes on low-angle dry sand made trails matching the Coconino invertebrate traces, and "These animals were too light to leave trails in wet or damp sand and were unable to effectively climb slopes steeper than about 20 deg."⁴

Two live-animal programmes, McKee's and Brand's, reached opposite conclusions.  A third, run seventy-five years after the first, is in Part 4.

Part 3: Whitmore's eleven points, one at a time

The canonical statement of the subaqueous case is a numbered list on page 282 of Whitmore's 2019 Answers Research Journal paper.⁵  The fullest version of the argument behind it is the forty-seven-page paper Whitmore and Paul Garner gave at the Eighth International Conference on Creationism in 2018, with about 500 references.¹  Here is the list in full, so a reader sees it whole before it is taken apart.

"1. The sandstone is only poorly- to moderately-sorted and is only occasionally well-sorted, 2. the grains of the Coconino can best be described as subangular to subrounded, 3. K-feldspar is a common accessory mineral and is often more angular than the quartz, despite it being softer on Mohs Scale (see Whitmore and Strom 2017, 2018), 4. muscovite, despite being extremely soft, was present in almost every thin section and our experimental results show that it disappears quickly in simulated eolian settings (Anderson, Struble, and Whitmore 2017), 5. features resembling primary current lineation (parting lineation) are commonly found on most foreset surfaces, 6. avalanche tongues, a feature common on steep slopes of eolian sand dunes, are missing in the Coconino; instead laminae can be followed in a straight line along exposed bounding surfaces, 7. cross-bed dips average about 20 deg and our hundreds of measurements agree with hundreds of other measurements made by Reiche (1938) and Maithel (2019), 8. large folds similar to parabolic recumbent folds (which are penecontemporaneous with cross-bed formation and cannot form from slumping dunes) are present in several places in the Sedona area, 9. dolomite (we believe it is primary) occurs in many places and forms in the Coconino including as beds, ooids, clasts, rhombs and cement, and 10. the expected sedimentary details commonly found in eolian sands (as far as grainfall, grainflow and ripple structures [Hunter 1977, 1981]) are for the most part absent or have been difficult to recognize in the Coconino (Maithel 2019). 11. Outside of our work, Brand (Brand 1979; Brand and Tang 1991) has reported compelling evidence from vertebrate trackways in the Coconino, whose characteristics seem to be best explained by underwater origin."⁵

Items 1 and 2: sorting and grain roundness

The observation is correct.  The Coconino is not the textbook picture.  Two things blunt the argument.

First, McKee's disclaimer on page 199, quoted in Part 2:  texture "cannot be considered to be proof of eolian origin."⁴  The claim being refuted is not the claim the eolian model rests on.

Second, Whitmore's own wording in the 2014 petrology paper is weaker than the list.¹¹  Modern fine eolian sand is well to very well sorted, and "The sorting in the Coconino falls at the upper edge of this range."¹¹  The upper edge of the range is inside the range.  Item 1 says "only poorly- to moderately-sorted."  The peer-reviewed sentence behind it says the sorting sits at the fine end of what modern deserts produce.

Items 3 and 4: mica and angular potassium feldspar, the best argument on the list

Muscovite is soft enough to be destroyed quickly by wind transport, and their laboratory work in Aeolian Research shows it disappearing in simulated eolian settings.  It appears in almost every Coconino thin section.  Potassium feldspar is softer than quartz, so it should be rounder.  In the Coconino it is often more angular.¹¹

The eolian answer is that abrasion depends on transport distance within the dune system, not distance from the ultimate source, and that a coastal erg fed from an adjacent shoreface has very short wind paths.  That answer is plausible.  Nobody has quantified it against the Cedarville experiments, and no mainstream paper has taken the mica result on.  This is the point where the creationist side has done work the other side has not answered.

Item 5: parting lineation on foreset surfaces

Parting lineation is a current-flow structure, and finding it on most foreset surfaces would be a real problem for a dune interpretation.  This article cannot evaluate the claim.  The observation appears only in the creationist corpus, no mainstream worker has published a check, and "features resembling" is doing work in the sentence.  It is an open claim, not a rebutted one.

Item 6: "avalanche tongues are missing," which fails on the movement's own sources

McKee reports dry-sand slump and avalanche marks in the Coconino by name, with figures.⁴  And the first conference paper out of Whitmore's own group, by Maithel, Brand and Whitmore in 2013, was titled "Morphology of avalanche beds in the Coconino Sandstone at Chino Wash, Seligman, Arizona."  Their own team named the beds avalanche beds.  Item 6 and the 2013 abstract cannot both be right, and no published creationist source reconciles them.

Item 7: the dips, right on the number and wrong on the conclusion

This is the point the creationists have earned.  Whitmore's Figure 6 gives measured means at five Arizona localities:  Bunker Trail 21.4°, Aubrey Cliffs 19.3°, Clear Creek 22.8°, Holbrook 17.3°, pooled 20.2°.  Comparable numbers come out of the De Chelly, Cedar Mesa, Casper and Tensleep Sandstones.⁵  Maithel measured the dips again at different outcrops for her dissertation and got an average of 19.8° from 135 measurements.¹³  The Sedimentology version reports "most dip angles ranging from the mid-teens to mid-twenties."⁶  Hill and Moshier's "29-31 deg" carries no citation and matches no modern dataset.⁷

Whitmore also closes the obvious escape.  Burial compaction rotates foresets to shallower dips.  But Coconino thin sections "show high porosities, an abundance of unfractured grains and undeformed ooids, which would not be present if the rock had been severely compacted," and "Compaction can probably account for only a few degrees of dip reduction at the most."¹  That is a petrographic test, not an assertion.

Three things stop the argument anyway.

McKee's figure was a distribution, not a mean.  "Mostly at 25 deg - 30 deg, but a few reach a maximum of about 34 deg" is compatible with a mean of 20° once the shallow tail is included.⁴  Nobody on either side has published the maxima from their hundreds of measurements, and that is where the argument is decided.  If the Coconino contains foresets at 32° to 34°, the low mean is a preservation statistic.  If its steepest measured foreset anywhere is 25°, the creationist case gets much stronger.

A preserved foreset is not a slipface.  Cross-strata curve "to form a tangent with the underlying bounding plane," in McKee's words.⁴  The next dune erodes the steep upper slipface.  What survives is the lower, tangential toeset.  A mean over a whole preserved set therefore has to come out below the angle of repose.  Gary Kocurek and Robert Dott wrote this out in 1981, and Maithel quotes them on page 36 of her dissertation:¹³

"Foresets formed by [eolian] processes other than avalanching (grainfalls and wind ripple migration) are necessarily below the angle of repose.  Much of the cross-stratification formed on the lower portions of a dune – that portion of the dune most likely to be preserved – is at angles from less than the angle of repose to near horizontal."

Maithel's own field data show it.  Dip is "fairly consistent along the cross-beds, but decreases rapidly upon approaching the lower bounding surface," so her 19.8° average "does not account for significant changes in dip along a single cross-bed."¹³

A sub-repose dip is not the subaqueous prediction either.  Both models predict it.  Maithel's own literature review, page 47:  "Smith (1972) proposed that cross-bed dip angle decreases with increasing flow velocity and water depth, so beds deposited by continuous subaqueous avalanching should dip at angles below the angle of repose."¹³  Eolian grainfall and wind-ripple foresets are below the angle of repose.  Subaqueous avalanche foresets are below it.  The preserved part of any foreset is below it.  A measured mean of 20° is consistent with all three.  The dip number is a good measurement that cannot decide the question, and Maithel's own dissertation is where that is set out most clearly.

Item 8: parabolic recumbent folds

Real structures, and McKee had already conceded the class was ambiguous:  such folds have "been developed under both conditions," wet and dry.⁴  There is also a split inside the creationist camp.  Whitmore, Forsythe and Garner argue against a seismic origin for the large Sedona folds, because seismic slumping leaves small thrust faults, compaction evidence and convolute beds that these folds lack;  they prefer a change of flow regime.¹⁴  Brand and Maithel, in Frontiers in Earth Science in 2021, attribute small-scale Coconino soft-sediment deformation to seismic shaking, a mechanism at home in deep time.  Two overlapping creationist teams explain deformation in the same formation by different mechanisms.

Item 9: dolomite, including ooids in cross-bedded units

The observation is good fieldwork:  dolomite over about half the Arizona outcrop area, as beds, ooids, clasts, rhombs and cement, concentrated near the base and top of the formation, and read as marine and primary rather than a replacement of calcite.¹¹

The distribution is the answer.  Dolomite at the base and top of a coastal erg is where the shoreline was closest, and interdune sabkha carbonate and early marine cement are ordinary there.

The ooids look harder, because ooids form in agitated shallow marine water and not in a dune.  But that says where the grains formed, not where they were deposited.  Wind can pick a marine ooid off a beach and build a dune out of it, and carbonate eolianites made of wind-reworked marine grains are a recognised rock type.  No primary source for that step was checked for this article, and it is listed below as unverified.

One point from the same paper is not answered this easily.  Whitmore's team found that "grain size tends to increase from the northern part of the formation to the southern part; opposite to what usually occurs in modern deserts."¹¹  In a wind system fed from the north, grains should fine downwind.  A coastal erg fed from a southern shoreface would not have that problem, but nobody has published that as a Coconino model with the transport directions worked out.

Item 10: the stratification types said to be "for the most part absent," which misstates its own source

This is the strongest single point against the creationist synthesis, and it comes from inside the team.

Item 10 cites Maithel 2019, her 434-page Loma Linda dissertation, whose advisers included Leonard Brand and John Whitmore.¹³  The peer-reviewed version is Maithel, Brand and Whitmore, Sedimentology 68(1):238-265, 2021.  Both reach the same conclusion in slightly different words.  The journal paper:

"Observed textures and sedimentary structures suggest that the cross-beds may consist of grainflow and grainfall deposits, but these remain difficult to differentiate at outcrop and thin-section scales."⁶

The dissertation, in its results chapter:

"Considering the vertical grain-size trends, bedding geometry, and lack of clear ripple lamination, we suggest that the Coconino Sandstone cross-beds consist of grainflow and grainfall deposits, but these are not readily differentiated at outcrop or thin-section scales."¹³

Grainflow and grainfall are the two processes that build a dune foreset.  Maithel's finding is that the Coconino cross-beds look like both and that telling them apart is hard, because laminae contacts are indistinct and diagenesis has degraded the primary textures.  Whitmore's list renders that as the structures being "for the most part absent or have been difficult to recognize."  The paper is titled "Characterization of hard-to-differentiate dune stratification types in the Permian Coconino Sandstone."  It treats the rock throughout as a dune-foreset problem.

One third of item 10 survives, and it is a real point.  Maithel does not confirm wind-ripple lamination.  Her dissertation lists "lack of clear ripple lamination" among the observations behind the grainflow-plus-grainfall reading.¹³  That matters, because climbing translatent strata made by migrating wind ripples are about as close to a wind-only structure as sedimentology has, and David Loope built his 1992 Geology comment on them.

The literature disagrees with her.  Her own review, page 23 of the dissertation, reports that Middleton, Elliott and Morales in 2003 "suggested that the majority of the cross-bedding formed as wind ripple deposits," and that while they published no photograph of a grainflow bed, "they did show lamination interpreted as wind ripple deposits."¹³  Blakey and Middleton reported climbing translatent strata in the lower Coconino near Sedona in 1983.  Maithel's answer is that those outcrops sit in the Schnebly Hill transition zone and may not be typical.¹³

So the count is a published photograph of wind-ripple lamination on one side and a doctoral study that did not find it on the other, at different outcrops.  That is a genuine open question.  See the unanswered questions below.

Item 11: the trackways

That is Part 4.

Part 4: The trackways

The Coconino has no bones.  It has footprints, thousands of them, on the surfaces of dune foresets.  One slab from Ash Fork in the Raymond Alf Museum carries more than 500 footprints and "the longest trackway of the Paleozoic for number of footprints (almost 200)."⁹  The formation is, in the National Park Service's words, "a reference unit for describing tetrapod tracks in Paleozoic eolian environments."⁹

Leonard Brand of Loma Linda University spent decades on these tracks and concluded they were made under water.  His four anomalies, in the creationists' own words:  some trackways show only toe marks and others only sole marks;  some prints are oriented obliquely to the direction of travel, with the front-foot impression faint or missing;  some trackways begin and end abruptly with no disturbance of the bedding;  and almost all of them run uphill.¹

Those are real observations and they are strange.  Four things bear on them.

First, Brand's own experiment removes the diagnostic

In 1996 Brand published a controlled study in the Journal of Paleontology:  230 trackways of the western newt Taricha torosa across ten combinations of substrate.⁸  Mud or fine sand.  Level or sloped at 25°.  Dry, damp, wet or submerged.

The result, from his abstract on page 1004:

"Trackways made on sloped, submerged mud or sand, sloped, dry sand, and sloped, damp sand rarely included the full complement of toes.  The positions and orientations of the toe marks were distorted if the animals were walking underwater or on sloped, damp sand."⁸

The distorted, obliquely oriented toe marks that the underwater hypothesis treats as its signature come from two conditions, not one:  submerged sediment, and sloped damp sand.  His Results give the mechanism for each.  The toe marks were distorted "if the animals were walking underwater and consequently were partially buoyed up by the water, or if they were walking on loose, damp sand on a slope."⁸  Buoyancy and slipping on loose damp sand leave the same mark.

A dune lee face becomes sloped damp sand after any dew or light shower.  McKee's own account requires that:  dry sand needs a little moisture to hold a surface structure through burial at all.⁴  Brand's own procedure is more specific.  Sand walked on straight after wetting recorded nothing.  What worked was sand "moderately dampened (four to 10 sprays) and then left to dry overnight or for at least eight hours."⁸  A lee face wetted by a night shower and walked on the next morning is that condition exactly.

The strongest sentence is Brand's own, in his Discussion on page 1010.  It is what he concluded in the journal:

"The data do suggest that the Coconino Sandstone fossil trackways may have been produced in either subaqueous sand or subaerial damp sand."⁸

He leaves the choice to two future lines of work:  better criteria for detecting buoyancy in a trackmaker, and "continued study of the sedimentological evidence."⁸  On the second, his own 2018 textbook says the sedimentology points to wind.  He also warns that "Trackways on the slopes of cross-stratified deposits make reliable identification of the trackmaking animals exceptionally difficult."⁸

Popular young-earth accounts that cite this work as proof of an underwater Coconino are citing a paper whose author left the question open on the page.

Second, the eolian side produced a mechanism, not a denial

The 1992 replies in Geology were weak.  Martin Lockley and David Loope proposed that the odd trackways were made by animals galloping, loping, trotting, jumping or walking sideways on a dune.  Whitmore and Garner answer that fairly:  in modern animals that lope sideways "the toes are only slightly displaced from the angle of the trackway, unlike the sharply oblique angles of displacement observed in many of the fossil tracks."¹  That objection stands against the 1992 comments.

It does not stand against what came next.  Lorenzo Marchetti, Sebastian Voigt, Spencer Lucas and colleagues, including three herpetologists from a Milan lizard-behaviour group, re-examined the material with anatomy-based ichnotaxonomy.  The conclusion, as stated in the National Park Service inventory chapter that summarises it:⁹

"In this context, the track morphology is strongly influenced by gravity and thus by sliding of the digits during the trackmaker's locomotion.  So, most of the previous ichnotaxonomic assignments were incorrect because they interpreted digit drag marks as actual digit impressions."⁹

On that reading, the marks Brand read as current drift are the animal's own toes slipping downslope on a steep dry face.  The same chapter records the verdict on the underwater hypothesis:  it "was rejected by Loope (1992), Lockley (1992) and Marchetti et al. (2019b), also because of the clear eolian origin of the Coconino Sandstone lithofacies."⁹

The revision was not cosmetic.  It invalidated the ichnogenera Chelichnus and Laoporus, because "the material assigned to Chelichnus and Laoporus is characterized by short and parallel 'digit impressions,' which are actually digit drag marks of different ichnotaxa."⁹  That took down the standard "eolian tetrapod ichnofacies" concept with them.  A model that costs its own side two ichnogenera and a widely used ichnofacies is not a face-saving move.

Third, it made a prediction and the prediction came true

If the odd Coconino track shapes are slope artefacts, subtracting the slope effects should reveal ordinary Permian animals.  If they are a distinct desert fauna, or a drowned assemblage, it should not.

Subtract the slope effects and the Coconino ichnoassociation comes out as "cf. Amphisauropus isp., cf. Dromopus isp., Erpetopus isp., Ichniotherium sphaerodactylum, cf. Tambachichnium isp. and Varanopus curvidactylus."⁹  That is the fauna known from non-desert sites of the same age elsewhere, and it falls inside the Erpetopus footprint biochron, which gives the Coconino a Kungurian age from the tracks alone.⁹

A correction derived from lizard locomotion recovers the taxa and the age that other evidence already assigned.  That is a successful prediction, and neither the underwater nor the Flood model has produced one.

Fourth, the phenomenon is not a Coconino peculiarity

The same oblique and asymmetric trackways turn up in the De Chelly Sandstone of Arizona and the Corncockle Sandstone of Scotland.⁹  Whitmore and Garner themselves report that the Coconino's low-diversity track fauna "is now recognized as part of a widely distributed ichnofacies characterizing Permian 'eolianites' in Scotland, Germany, Argentina and across the southwestern USA."¹

Every Permian cross-bedded sandstone anyone has examined has these trackways.  A hypothesis that needs a global Flood to explain a feature found wherever a Permian erg is looked at does less work than a locomotion model that predicts where the feature will be found.

And Brand himself is the most candid witness

From his 2018 textbook, on the page where he states the problem:²

"The trackways have features that seem virtually impossible to explain unless they were made with the animals completely underwater, while the sedimentary evidence, as interpreted by sedimentologists, seems to point to wind-blown sand.  This seeming contradiction indicates there are some unknown pieces of the puzzle that remain to be discovered."²

And on p. 475:

"It is important not to go beyond our evidence in reaching conclusions.  This research does not show whether the tracks were or were not made in the biblical flood.  It simply points to the underwater origin of these trackways."²

That is more careful than the way the argument reaches the public.  Quoting it is fair to Brand and fatal to the popular version, which presents him as having proved an underwater Coconino and a Flood Coconino.

One thing the mainstream has not done

Nobody has repeated the invertebrate-trace experiments with marine or freshwater animals.  Whitmore and Garner make the point fairly:  "studies of the invertebrate traces have employed terrestrial animals, such as spiders and scorpions.  Experiments with a broader range of invertebrates, including marine and freshwater forms, would be instructive."¹  That experiment is cheap, it has not been done, and it could change part of the picture.

Part 5: The injectites at the base

At the bottom of the Coconino, sand-filled cracks run down into the red mudstone of the Hermit Formation.  McKee's reading was tentative and wrong.  The Hermit surface, he wrote, "is trenched with large shrinkage cracks, probably the result of desiccation, which are filled with white sandstone of Coconino type that apparently filtered down from above."⁴

Whitmore and Strom went and looked.  They examined cracks at thirty locations, sampled two under a Grand Canyon research permit, and published in Sedimentary Geology.  Their field pattern is convincing.  The deepest cracks, at least 15 metres, sit beside the Bright Angel Fault where its offset is greatest at 61 metres.  Average crack depth falls off east and west, "disappearing altogether along Tanner and South Bass Trails."  Sixty-eight cracks at nine locations come out oriented rather than random on two statistical tests.  And the fills never show the horizontal bedding that top-down infiltration would produce.¹²  The sand was injected upward under pressure, not washed down.

Give them the credit.  Then read their own dating.  Whitmore and Strom put the mechanism in their own conclusion:  "evidence suggests the basal Coconino was water saturated and underwent liquefaction during an ancient seismic episode, perhaps along the Bright Angel Fault."¹²  Brand's textbook restates it:  "The researchers believe the sand-filled cracks were injected because of earthquake energy released during the movement of the Bright Angel Fault."²  Laramide movement is conventionally 80 to 55 million years ago, about 200 million years after the Coconino was deposited.  On the authors' own reading the injectites say nothing about how fast the Coconino accumulated.  They say the basal sand was still mobile when the fault moved.

The part that is a real problem for deep time

Whitmore and Strom say so themselves, in the Elsevier journal:

"A problem with no readily apparent solution is that it appears the basal Coconino was water saturated and unlithified (or only partially lithified) at the time of Laramide faulting which we suspect caused dike intrusion.  It is unlikely the Coconino could have remained uncemented in excess of 250 million years."¹²

That is a fair objection.  On the conventional timescale, sand deposited in the Permian has to be still soft and saturated when a Cretaceous-to-Paleogene fault moves through it.  The authors offer two conventional escapes:  the basal cement was removed shortly before Laramide faulting, or the Bright Angel Fault moved more than anyone has recorded soon after the Coconino was laid down.¹²

Their second escape runs against the literature they cite in the same paragraph, which has the Bright Angel Fault "relatively inactive between the Precambrian and the Laramide deformation."¹²  Their first is the live one, and it is not exotic.  A buried sandstone's cement is dissolved and reprecipitated over its history rather than set once and kept.  The reference work Whitmore and Strom cite on injectites is built on cases of injection from sand bodies buried for a long time, and Whitmore and Garner quote it approvingly:  "It is not unusual for clastic dikes, injectites, and sand volcanoes to occur coincident with faulting;  fine-grained water-saturated sands are especially mobile."¹  The central cases in that volume, the Paleogene injectite complexes of the North Sea, came from reservoirs buried for millions of years.

Notice which way this argument points.  It is an argument against a long gap between deposition and injection.  It is not an argument for a Flood.  The injection is dated by the fault, and the fault is Laramide on the authors' own reading.  Compressing the Coconino into a Flood year does nothing to shorten the interval that troubles them unless the Laramide is compressed with it, and that is a separate claim needing separate evidence, which this paper does not supply.

The injectite paper is a genuine correction to McKee.  It is not a clock.

Part 6: The sediment budget, which decides the Flood question

Grant the subaqueous reading in full.  Suppose every one of Whitmore's eleven points is right and the Coconino is a marine sand-wave deposit.  The Flood model still fails on arithmetic, and its own side has conceded the arithmetic.

Timothy Helble is a retired National Weather Service hydrologist.  His 2011 paper in Perspectives on Science and Christian Faith is a mass balance run on Steven Austin's own inputs.¹⁰

What he granted the Flood side.  Austin's own volume for the Coconino plus its correlatives:  10,000 cubic miles, or 42,000 cubic kilometres.  Austin's own 1,600-kilometre northern boundary that the sand has to cross.  Austin's own accounting for the time available, which gives twelve days.  And then Austin's own sediment transport rate of 3.0 kilograms per second per metre of front, multiplied by ten as a gift.  Helble lists nine assumptions made in the young-earth side's favour, including a deliberately low sand density of 1,900 kg/m³.

The result, on page 34:

"To form the Coconino in Austin's Flood scenario, 42,000 km3 of sand would need to be transported across the 1,600 km long northern boundary in twelve days.  However, the volume just calculated—26 km3—is only about 1/1,600 of the required amount."¹⁰

At that ten-times-generous rate the Coconino takes fifty-two years, not twelve days.

Run it backwards and the requirement becomes physical:

"These calculations indicate a slab of sand 25 m high, 1,600 km wide, and 1,000 km long would have to be continuously sliding southward across the boundary at one meter per second to form the Coconino Sandstone in twelve days."¹⁰

Austin's own procedure gives a water depth of 54 metres, so the moving sand slab is nearly half the depth of the water carrying it.  Helble's comment bites:  "If it were possible for such a regional-scale sand slab to slide southward … formation of sand waves and cross-beds would obviously be precluded."¹⁰  The mechanism destroys the structures it was invoked to explain.

Helble also checks the rescues.  Tsunami currents are not unidirectional in Austin's own cited source, and 500 cm/sec is far outside the 90 to 155 cm/sec window Austin's procedure requires for sand waves.  On fast surface currents over slow bottom ones:  "near-surface and near-bottom velocities differing by a factor of 50 are not hydraulically realistic."¹⁰

The published creationist reply concedes it, in an abstract at the 2025 New Scholars conference:

"Helble is correct that the processes he outlines could not have formed the Coconino during the Flood, but the Flood model requires extrapolations beyond conditions that prevail in the present-day.  If a Flood model is correct, we will never be able to observe the depositional processes that made the Coconino."³

Take that at face value.  It is an honest statement of a position.  But it grants that no observable process deposits the Coconino in the time available, and appeals instead to conditions unobservable in principle.  Set it beside Whitmore's 2019 hedge that "Sand waves may not be an exact analogue of how the 'blanket' of the Coconino and other sandstones were deposited, but we envisage something similar,"⁵ and the 2018 admission that "There are no modern analogs that match the precise sedimentology of the Coconino."¹

The subaqueous Coconino has no observed analogue, no measured rate, and no quantified mechanism.  The eolian Coconino has hundreds of modern ergs to compare against.

The sediment-supply problem is not an outsider's objection.  In 1994 Austin, Baumgardner, Humphreys, Snelling, Vardiman and Wise listed the difficulties facing Flood sedimentology, and the third was:  "It may not be possible to derive all the Flood sediments from igneous and/or metamorphic precursors by physical and chemical processes in the course of a single, year-long Flood."¹⁰  It was raised inside the movement thirty years ago, and it has not been solved.

Objections and responses

"Even the geologists disagree about how it formed, so why trust either of them?"  They disagree about wind versus water.  They do not disagree about time.  McKee's dune field, Blakey's coastal erg, Maithel's grainflow foresets and Whitmore's marine sand waves are all Permian.  The one position that requires twelve days is the one whose principal researcher says in print that no observable process could do it.

"Underwater dunes are a mainstream idea too.  Blakey found marine sand waves right next to the Coconino."  He did, and that is why the eolian interpretation here is coastal, not Saharan.  A sand-wave complex beside the Coconino shows a sea was nearby, which a coastal dune field requires.  Austin's own mainstream support for underwater Coconino dunes was a comparison with sands in the Altamaha River estuary in Georgia, and Hill and Moshier's response holds:  "that mainstream geologist never envisioned the catastrophic conditions of high-velocity flow prescribed by Austin."⁷  Subaqueous dunes are ordinary.  Subaqueous dunes built at 3 to 5 metres per second under 100 metres of water in twelve days, with reptiles walking on them, are not.

"You are quoting creationists as though they were doing science."  On the Coconino, some of them are, in Sedimentary Geology, Sedimentology, the Journal of Sedimentary Research and Aeolian Research.  Refusing to read the data because of the authors' beliefs is how Hill and Moshier ended up printing a dip figure with no source that no modern measurement supports.

"Brand is a trained paleontologist who spent decades on the tracks.  Who is anyone else to overrule him?"  Nobody is overruling him.  His own 1996 experiment is the strongest single piece of evidence against his conclusion, and his own 2018 book says the sedimentology points to wind.  The article agrees with Brand about Brand, and disagrees with the people who quote him more confidently than he quotes himself.

"McKee wrote in 1979, and you concede four creationist points.  Isn't the eolian model in trouble?"  The popular version of it is.  The four concessions are about texture, frosting, dip statistics and one basal structure.  None touches the ripple index, the coarse grains on the crests, the dip-parallel crests, the animal experiments or the ichnological revision, and nobody has redone those.  A forty-seven-year-old measurement that no one has repeated is not refuted by its age.

"The cross-beds average 20°, which is a sand-wave dip.  Why is that not decisive?"  Because a preserved cross-bed set is not a slipface, as Part 3 shows.  The number that would settle it is the maximum dip in Whitmore's dataset, and it has not been published.

"If it was a desert, where are the fossil soils, the wind-polished stones and the root traces?"  The eolian answer is that an actively aggrading dune field is where soils, deflation lags and roots do not form or survive.  That answer is sound and it is close to unfalsifiable.  Worse for the objection:  the claimed absence of paleosols, ventifacts and root traces could not be found asserted with a source anywhere in the creationist corpus read for this article.  The objection may be true.  Nobody appears to have documented it.

Issues and unanswered questions

The standard modern technical statement could not be obtained.  Middleton, Elliott and Morales, "Coconino Sandstone," in Grand Canyon Geology, 2nd edition, pages 163 to 179, is what Whitmore and Garner call "the most often cited technical summary of the Coconino."  No scan exists on archive.org.  Everything this article knows about the chapter is at second hand,¹⁵ from Whitmore's polemic and Maithel's literature review, which agree on what it says.

The wind-ripple question is the last live diagnostic and it is unresolved.  Middleton, Elliott and Morales assign most Coconino cross-bedding to wind-ripple deposits and illustrate the lamination.  Maithel's fieldwork at other outcrops found none.  A study that revisits the localities Middleton illustrated would settle it.  Until then, item 10 of Whitmore's list is one-third alive.

The maximum cross-bed dip has never been published.  Both teams report means and neither states the steepest foreset it measured, and that number decides the argument.  Maithel's 135 measurements exist, and extracting the maximum would take an afternoon.  The compaction analysis exists only as a 2011 conference abstract, so "a few degrees at the most" has no published calculation behind it.

One step in the ooid reply was not checked against a source.  The claim that wind can build dunes out of marine ooids picked off a beach is standard sedimentology, and no primary reference for it was read.  It should be cited or dropped.

The mica result has no published answer.  The short-wind-path reply is plausible and nobody has quantified it.  This is the point on which the creationist side is ahead.

Parting lineation on foreset surfaces has never been independently checked.  Item 5 is a current-flow structure reported only by one group.

Rain prints are undecided and both sides should stop claiming them.

Two sources are still read at one remove.  Maithel's 2021 Sedimentology paper is closed at Wiley;  its abstract is confirmed by three publisher-deposited records and its content by the dissertation, which was read in full.  Loope's 1992 Geology comment, the strongest published sedimentological reply to Brand, has not been read in the original.

Two arguments cut against the eolian side, and the creationists have not pressed them.  McKee reports that "flat-bedded interdune deposits are sparse," and explains it by "an abundant sand source with resulting rapid dune migration and burial of interdune spaces."⁴  That is an inference from an absence.  An erg of 51,500 square kilometres should preserve interdune facies.  And the grain-size gradient runs the wrong way for a north-fed desert.¹¹  Both are better arguments than several the creationist literature does make.

Verdict

On wind versus water, the weight is with wind, in a coastal dune field beside a shallow sea.  The creationists measured the dips, and the mainstream figure in circulation is wrong.  They retired mechanical grain frosting.  They corrected McKee on the basal cracks.  Their mica result is unanswered.  Against that stand McKee's five ripple criteria, four of which nothing in the creationist corpus addresses;  the chuckwalla and millipede experiments;  Maithel's own peer-reviewed finding that the cross-beds look like grainflow and grainfall;  and an ichnological revision that explains Brand's anomalies, invalidates two of its own side's ichnogenera, and predicts the right track fauna.  Two of Whitmore's eleven points fail on his own team's sources.  This is a live dispute, and it is not close to even.

On the Flood, there is no dispute left.  The sand budget falls 1,600-fold short at ten times the transport rate flood geology asks for.  The mechanism that would close the gap would destroy the cross-beds it exists to explain.  The model has no modern analogue by its authors' own statement.  And the reply from its principal researcher grants the arithmetic and appeals to conditions that cannot be observed.

The creationists have improved the description of the Coconino Sandstone.  They have not moved it into a one-year Flood.

Every source this article rests on is catalogued, with what it measured and which way it cuts, on a companion page:  the literature.

Sources