The literature

This page exists so that nobody has to take the article on trust.  Every source it rests on is listed here with what it actually measured, who wrote it, and which way it cuts.

A catalogue of the literature bearing on natural rock arches and natural bridges — how they form, how fast they fail, how old they are, and whether any of that constrains the age of the earth.

It is deliberately not a reading list of things that agree with this site.  Sources are tagged by the direction they cut, and a reader who wants only the material that undercuts the argument made in the main article can filter for undercuts and read those.

How to read an entry

Measurement or model.  The most important distinction here.  A seismometer on an arch is an observation.  An exponential fitted to a count of collapses is a model, and it inherits every assumption in it.  Both are useful; they are not the same kind of thing, and conflating them is how the argument in question went wrong.

Direction.  Whether the source supports, undercuts or complicates the young-earth arch argument.  Assigned by what the source shows, not by what its author believes.  Several mainstream sources are tagged supports — the mainstream position that arches are only thousands of years old is genuinely friendly to part of the creationist case — and several creationist ones are tagged undercuts.

Verified.  Whether the text was read directly by this project (full), only its abstract or record (abstract), or only its bibliographic entry (citation).  Nothing here is quoted verbatim in the main article unless it is marked full.

Field definitions

Coverage and its limits, stated plainly.  This catalogue was built from a Crossref sweep across 45 topic queries (1,059 records, 442 passing a relevance filter), a venue-by-venue sweep of the creationist and apologetics literature, and direct retrieval from agency archives.  It is not exhaustive and cannot be.  Two known gaps: the Journal of Creation and Creation Research Society Quarterly print archives are only partly reachable online, and the Natural Arch and Bridge Society's newsletter SPAN, which has run since August 1988 and is where arch collapses are actually reported first, is members-only and was not obtained.  Anything drawn from SPAN here comes through the National Park Service administrative history that cites it.

The young-earth argument itself

How Long Have Arches Been Around?

Faulkner, Danny R.  Answers in Depth 19, Answers in Genesis (21 August 2024).

https://answersingenesis.org/geology/how-long-have-arches-been-around/

camp: creationist topic: yec-argument method: model direction: supports verified: full

Key figures: 43 collapses / 29 years = 1.48 per year; λ = 7.34 × 10⁻⁴ yr⁻¹; 2,043 arches in 1971; ~50,000 arches 4,500 years ago; 1,000,000 arches 8,400 years ago.

The flagship statement of the argument.  Fits an exponential decay to arch loss at Arches National Park and concludes the arches must postdate a 4,500-year-old Flood.  The arithmetic reproduces exactly from the stated inputs.  Both inputs are wrong: the "article published in 2000" is Eos, 13 July 2020, and the 2,043 arches in 1971 were about 90 in every contemporary record.  Contains two checkable factual errors — Wall Arch dated to 2019 (it fell in 2008) and the source misdated by twenty years.  The author has since disowned both numeric inputs (see below).  Not corrected by the publisher as of July 2026.

More About Arches, and More

Faulkner, Danny R.  Danny Faulkner Blog, Answers in Genesis (19 September 2024).

https://answersingenesis.org/blogs/danny-faulkner/more-about-arches-and-more/

camp: creationist topic: yec-argument method: commentary direction: undercuts verified: full

The author's own statement of the model's fatal assumption, made a month after publication and without prompting from anyone he names: "A good model of arch lifetimes must include both processes.  Since we have never observed the formation of an arch, in my initial analysis, I assumed that the current formation rate is zero."  Also describes Tonto Natural Bridge as a travertine arch that "probably is still growing" — an arch he believes is forming now.

My Recent Arch Trip to Utah

Faulkner, Danny R.  Danny Faulkner Blog, Answers in Genesis (25 April 2025).

https://answersingenesis.org/blogs/danny-faulkner/2025/04/25/my-recent-arch-trip-utah/

camp: creationist topic: yec-argument method: commentary direction: undercuts verified: full

The most important item in the creationist corpus.  Reports that creationist geologist Nathan Mogk could not find the source of the 43-collapse figure and concluded it was three documented collapses added to an estimated forty.  Faulkner: "the 43 arch collapses over 49 years may have been an overestimation.  I probably ought to revisit my publication from last year."  In the same post he disowns the other input: "With no real documentation of the number of arches, I suspect that someone estimated that figure."  Also contains an arch-density argument — one arch per four acres in the arch-bearing tenth of the park is "doubtful" — which turns out to be far more damaging to his own model than to the standing count.

Moose Head Arch: RIP

Faulkner, Danny R.  Danny Faulkner Blog, Answers in Genesis (18 September 2025).

https://answersingenesis.org/blogs/danny-faulkner/2025/09/18/moose-head-arch-rip/

camp: creationist topic: yec-argument method: commentary direction: mixed verified: full

Retracts the Kentucky control case: "I have often said that there has not been a documented natural arch collapse in Kentucky.  I now stand corrected."  Maintains a fair distinction — Moose Head Arch was struck by storm debris rather than failing under self-weight.  Reports Bill Patrick's Red River Gorge count at 1,175 arches with leads on a couple of hundred more.

The age of arches

Batten, Don.  Creation 40(4):23 (October 2018).

https://creation.com/en/articles/age-of-arches

camp: creationist topic: yec-argument method: model direction: supports verified: full

The precursor, and in one respect the better-sourced version: it cites the AGU blog post directly and gets the window right ("Forty-three collapsed between 1977 and 2015").  Uses a linear model and reaches ~2,000 years to total attrition.  Asserts without any citation that "no new arches are forming in the National Park."  Its reductio — "even just 100,000 years would mean an original number of arches of 50 times the number of arches still standing" — has the same zero-formation assumption buried in it.

Many arches and natural bridges likely from the Flood

Oard, Michael J.  Journal of Creation 23(1):115–118 (2009).

https://creation.com/natural-bridges

camp: creationist topic: yec-argument method: review direction: supports verified: full

The strongest young-earth treatment, and structurally different from the count-decay argument.  Oard does not count anything or fit any curve.  He argues from the localisation of erosion — why here and not next door? — that arch formation requires conditions not operating today, and assigns them to Flood runoff.  Nothing in it is falsified by correcting the 43 or the 2,000, which is exactly why it is the better argument.  It also yields no age.

A new mechanism to form freestanding arches questionable

Oard, Michael J.  Journal of Creation 31(1):4–5 (2017).

camp: creationist topic: arch-mechanics method: commentary direction: mixed verified: abstract

Oard's rebuttal of Bruthans et al. 2014, calling the stress-erosion experiments artificial.  Notable because the Institute for Creation Research cites the same paper approvingly as evidence that arches form fast.  A live internal disagreement inside young-earth geology about its own best mechanism.

Weibull Analysis of Arch Collapses

Nulk, Gabriel M.; Mogk, Nathan W.  International Conference on Creationism, New Scholars Conference (2025).  Veritas Academy of Tucson.

https://web.archive.org/web/20250909162423/https://digitalcommons.cedarville.edu/icc_new_scholars/2025/oral_presentations/17

camp: creationist topic: yec-argument method: model direction: undercuts verified: full

Internal creationist criticism, and the most technically literate item in the corpus.  Replaces Faulkner's constant hazard with a two-parameter Weibull (shape β, scale η) precisely because it "is expected to better reflect the physical changes that arches undergo than the constant failure rate model assumed by Faulkner (2024)."  Then the empirical base collapses under them: "An unsuccessful attempt to validate the total count of arch collapses in Morton (2020) was made, including scholarly articles, two online arch databases, and inquiries to Arches National Park.  Therefore, four synthetic arch collapse datasets were created."  Two young-earth researchers asked the National Park Service directly for the number their own side's argument rests on, could not get it, and modelled synthetic data instead.  They also cite the source correctly as Morton (2020), confirming from inside the creationist camp that Faulkner's "2000" is an error.  Retains the single-cohort assumption explicitly ("under the assumption that they formed around the same time").  Mogk is the same researcher Faulkner credits with questioning the 43-collapse figure.  Cedarville migrated its repository in 2025 and the live link now resolves to a relocation notice; the Internet Archive copy above is the citable one.

Stone Arch Collapse Doesn't Support Old Earth Theory

Answers in Genesis, unsigned (16 August 2008).

camp: creationist topic: yec-argument method: commentary direction: undercuts verified: abstract

Historically the most interesting item: sixteen years before Faulkner, AiG ran the count-decay argument from a Wikipedia figure and flagged it as unverifiable in the same breath — "no source is cited for the information and we could not verify the number."  The ministry knew the number was unsourced in 2008 and published arguments resting on it in 2014 and 2024.

Earth's Surface Shaped by Genesis Flood Runoff, ch. 10: The Puzzle of Large Natural Bridges and Arches

Oard, Michael J.  Self-published online book.

http://michael.oards.net/pdf/Book/Chapter10.pdf  (plain HTTP only; the HTTPS endpoint returns 401)

camp: creationist topic: yec-argument method: review direction: supports verified: full

The fullest young-earth treatment of arches anywhere.  Sets arches inside the retreating-floodwater model.  Concedes, following a published exchange with Arthur Manning, that Kolob Arch formed post-Flood by sapping from an alcove — a significant internal concession that arches can form by ordinary weathering.

Where the number 43 came from

An Administrative History of Arches National Park

Purcell, David E.  Museum of Northern Arizona, for the National Park Service (2019).

http://npshistory.com/publications/arch/adhi.pdf

camp: agency topic: collapse-record method: inventory direction: undercuts verified: full

The single most important source on this topic, and apparently unknown to every writer on either side.  Table 7-2, "Collapsed Arches" (pp. 377–378), prints the 43-row list in full with names, types, locations and year ranges.  Its provenance: "In 1993, a list of arches known to have fallen since 1977 was compiled by unknown persons (possibly Dale Stevens)… their inclusion on the list does not mean that the span has collapsed in every case, but that at a minimum a crack through the span has separated it from the rest of the arch."  The original list of 42 included Landscape Arch, still standing; the historian removed it and added Wall Arch (2008) and Rainbow Arch (2017–18), producing 43.  Forty-one of the 43 have windows ending 1993 or earlier; the table records no collapses at all in 1994–2007 or 2009–2016.  Also documents the arch count year by year (69 in 1938, 89 in 1967, 124 in 1973, 200 in 1982, 1,000 in 1989, 1,500 in 1990, 2,000 in 1995) and the 1940–41 Landscape Arch rockfall (95 × 55 × 14 ft, 18,287 cubic feet, 1,344 tons — the largest documented in the park).

The Delicacy of Arches

Morton, Mary Caperton.  Eos 101 (13 July 2020).  doi 10.1029/2020EO146523

https://eos.org/features/the-delicacy-of-arches

camp: mainstream topic: arch-dating method: commentary direction: mixed verified: full

Faulkner's actual source, misdated by him to 2000.  Carries the "at least 43 arches have collapsed" sentence, the "over 2,000" count, and Paul Geimer's estimate that "the smaller arches are on the order of hundreds to thousands of years old, while the largest arches may be in the range of 100,000 years old" — explicitly flagged as a "best guess" in a passage explaining that erosional landforms cannot be dated.  Also carries the statements that most contradict the argument built on it: that new arches keep forming and that fins and alcoves are visibly "on their way to becoming the next North Window Arch."

Natural arches hum their health and scientists are listening

O'Hanlon, Larry.  AGU GeoSpace blog (7 August 2015).  Blog retired; retrieved via the Internet Archive.

https://web.archive.org/web/20251213111505/https://blogs.agu.org/geospace/2015/08/07/natural-arches-hum-their-health-and-scientists-are-listening/

camp: mainstream topic: collapse-record method: commentary direction: mixed verified: full

The version Batten cited, and the earliest traced appearance of the figure in a science outlet: "at least 43 arches have collapsed since 1977, according to rangers at Arches National Park.  A few of these were destroyed by vandals."  The vandalism qualifier is dropped in every later retelling.

Agency primary sources

The Geologic Story of Arches National Park

Lohman, S. W.  U.S. Geological Survey Bulletin 1393 (1975).

https://pubs.usgs.gov/bul/1393/report.pdf

camp: agency topic: arch-mechanics method: review direction: undercuts verified: full

The official USGS guide, and a far better source than its age suggests.  States that arches will be gone "before many thousands of years" and that "many new arches will form… as the geologic clock ticks on."  Names a specific recently formed arch — "Baby Arch" — and gives the field evidence for its recency: "sharp, angular breaks in the ceiling and… the pile of freshly fallen rocks."  Records the pre-1975 catalogue history ("Nearly 90 arches have been discovered"; Stevens's 124 in 1973) and describes calcite-cement dissolution, which is the real basis for Faulkner's chemistry.  Also notes that "many of the pinnacles or piles of rock may be the broken remains of former arches."

Geologic Formations — Arches National Park

National Park Service.

https://www.nps.gov/arch/learn/nature/geologicformations.htm

camp: agency topic: arch-inventory method: commentary direction: mixed verified: full

Source of the "over 2,000 documented arches" figure and the three-foot criterion.  Describes formation as an ongoing sequence ("fins into arches").  Contains the mirror-image of the creationist argument, drawn from the same observation: "The fact that over 2,000 still stand tells us this area has been rather geologically stable for at least 50,000 years."  That 50,000-year claim is asserted without a citation and could not be traced to a source; it should be treated with the same suspicion as the 43.

Studying the Fate of Arches

National Park Service.

https://www.nps.gov/articles/studying-the-fate-of-arches.htm

camp: agency topic: arch-monitoring method: measurement direction: undercuts verified: full

Directly contradicts the argument's central premise: "Every story of the park's 2,000 arches will end at some point.  However, new arches are forming at a similar pace."  Also reports the North Window crackmeter, installed November 2015: the crack opens and closes by more than 2 cm annually and more than 0.5 mm daily, with Jeff Moore quoted that "there is no opening trend."  This is the longest continuous instrumental record of arch deterioration anywhere, and over its span the net change is zero.

GeoSights: Wall Arch, a Fallen Giant

Willis, Grant.  Survey Notes 41(2), Utah Geological Survey (May 2009).

https://geology.utah.gov/map-pub/survey-notes/geosights/wall-arch/

camp: agency topic: collapse-record method: measurement direction: mixed verified: full

A state geological survey on the best-documented arch collapse.  Corrects the widely repeated 71-foot span: the true span was 55 feet, and 71 feet is the "breadth," which "is not useful for comparing arches."  Notes the first visible warning in 1969, thirty-nine years before failure, and proposes nighttime thermal contraction as the trigger.  Contains the plainest agency statement of the dating problem: "We do not have a good geologic tool for dating arch formation, but we are sure that Wall Arch had stood nearly unchanged for hundreds, and probably thousands of years."

Geologic map of the Windows Section 7.5' quadrangle, Grand County, Utah

Doelling, Hellmut H.; Kuehne, Paul A.  Utah Geological Survey M-260dm (2013).

https://geology.utah.gov/publication-details/?pub=M-260dm

camp: agency topic: arch-mechanics method: measurement direction: undercuts verified: full

Cited by Faulkner for his cement chemistry, and it inverts his rule.  Describes the calcite-and-iron-oxide-cemented Entrada Slick Rock Member as "Well-indurated" and the Navajo Sandstone as "generally friable and weakly cemented with silica or calcite."  Attributes arch formation to structure rather than cement — jointing from folding, a soft parting beneath, and salt dissolution below — and dates it: "Arch formation is a product of the Pleistocene and Holocene."

Kachina Bridge — Natural Bridges National Monument

National Park Service.

https://www.nps.gov/nabr/planyourvisit/kachinabridge.htm

camp: agency topic: collapse-record method: measurement direction: undercuts verified: full

A dated, quantified enlargement event: "approximately 4,000 tons of sandstone fell from the inside of the Kachina bridge opening in June 1992, enlarging the opening as it has no doubt been enlarged time and time again."

Historical Rock Falls in Yosemite National Park, California (1857–2011)

Stock, G. M.; Collins, B. D.; Santaniello, D. J.; Zimmer, V. L.; Wieczorek, G. F.; Snyder, J. B.  U.S. Geological Survey Data Series 746 (2013).  doi 10.3133/ds746

https://pubs.usgs.gov/ds/746/ds746_text.pdf

camp: agency topic: rockfall-statistics method: inventory direction: background verified: full

The model of what a real rockfall inventory looks like: 154 years, more than 1,000 documented events, each with date, location, volume estimate and source.  Included here as the contrast case.  Arches National Park has nothing remotely comparable, which is why the 43 could circulate for two decades unchallenged.

El Morro National Monument Geologic Resources Inventory Report

National Park Service Geologic Resources Division.

camp: agency topic: erosion-rates method: measurement direction: undercuts verified: abstract

Key figure: 0.8 mm over five years (2000–2005) = 0.16 mm/yr.

Reference pins set into Inscription Rock and measured against the retreating face.  The only direct instrumental measurement of sandstone surface retreat located for the Colorado Plateau region.  A ceiling rather than a typical rate: the rock is friable clay-cemented Zuni Sandstone and the pins were sited where erosion was active.  Even so, 4,500 years at this rate removes 0.72 m.

An Archeological Survey of Natural Bridges National Monument, Southeastern Utah

McVickar, Janet L. (ed.), with Eininger, S. F.; Hurt, T. D.; Irwin, D. C.; Orcutt, J. D.; Spude, C. H.  National Park Service, Intermountain Cultural Resources Center Professional Paper 64 (2001).

https://npshistory.com/series/archeology/icrc/64/report.pdf

camp: agency topic: archaeology method: inventory direction: undercuts verified: abstract

Archaeology in the alcoves immediately around Sipapu, Kachina and Owachomo bridges, including Horsecollar Ruin with its original unrestored roof.  Puts securely dated human construction under Cedar Mesa Sandstone overhangs a short walk from the bridges themselves.

Datasets — what actually exists

The article's central negative claim is that no valid arch-collapse dataset exists.  This section records every candidate that was checked, so the claim can be audited rather than trusted.  Each was opened and inspected directly.

Table 7-2, "Collapsed Arches"

Purcell, David E.  An Administrative History of Arches National Park, pp. 377–378 (2019).

http://npshistory.com/publications/arch/adhi.pdf

camp: agency topic: collapse-record method: inventory direction: undercuts verified: full

43 rows.  Fields: year (mostly a range), name, arch type, location code.  Compiled from an anonymous manuscript of about 1993.  41 of 43 have windows ending 1993 or earlier; zero events recorded 1994–2007 and zero 2009–2016.  Most rows are last-seen-intact to first-seen-fallen brackets, not dated events, and the source note admits some entries are cracked rather than fallen.  Not usable as a rate.  It is nevertheless the best thing that exists, which is the point.

"Fallen Arches" blog category

Natural Arch and Bridge Society.

https://www.naturalarches.org/blog/category/fallen-arches/

camp: arch-society topic: collapse-record method: inventory direction: mixed verified: full

8 posts, 2015–2018, worldwide; 7 record collapses.  Enumerated in full through the site's WordPress REST API (category id 10) because the HTML category page does not filter correctly.  The seven: Ring Arch, Arches NP (partial, between 29 April and 7 October 2014); Arch Rock, Point Reyes, California (21 March 2015, killed a hiker); La Dame Blanche, Drôme, France (by April 2015); Long Island Arch, Nova Scotia (19 October 2015); Legzira Beach, Morocco (23 September 2016); the Azure Window, Malta (8 March 2017); Rainbow Arch, Arches NP (winter 2017–18).  Opportunistic reporting by whoever noticed, with no defined population and no schedule.  Not usable as a rate — but two of the seven are cases where the arch's birth was also recorded, which makes it the single most valuable qualitative source on this topic.

NABS GIS catalogue — arches flagged "(f)" for fallen

Wilbur, Jay H. (compiler).  Natural Arch and Bridge Society.

https://www.naturalarches.org/db/arches/index.html

camp: arch-society topic: collapse-record method: inventory direction: mixed verified: full

547 catalogued entries across 31 state and country lists; 17 flagged fallen.  The flag is a bare "(f)" appended to the NABSQNO coordinate string.  Of the 17, twelve carry a date or a bracket and five carry none; events run from about 1550–1750 (the Bridge of the Gods) to 2010; and two are deliberate human demolition rather than natural collapse — Fairy Arch for a road project, Namesake Natural Bridge for a Civil War-era sewer.  The index itself says the catalogue is unfinished, that most maps contain arches missing from the corresponding list, and that additions stopped when Google changed its Maps interface.

The finding that matters is the comparison with the blog list above: they barely overlap.  Wall Arch appears here but not in the blog; Rainbow Arch appears in the blog but not here.  Two records kept by the same organisation about the same national park disagree on which collapses were written down, which is a direct demonstration that neither is a sample of a defined population.

World Arch Database

Van Bebber, Tom, for the Natural Arch and Bridge Society.

https://www.naturalarches.org/blog/world-arch-database-surpasses-10000-arches/

camp: arch-society topic: arch-inventory method: inventory direction: background verified: abstract

Passed 10,000 arches in July 2020 and reportedly 17,000 by 2025.  Standing arches only; no collapse or condition field was found.  Distributed by subscription or CD-ROM, not openly.  Its growth is discovery, and it is a useful check on the intuition that a rising catalogue means a rising population.

Natural Arches of Kentucky (Ky Arch Trail)

Kentucky Geological Survey, University of Kentucky.  ArcGIS StoryMap over the Ky_Arch_Trail feature service.

https://kgs.uky.edu/arches/

camp: agency topic: arch-inventory method: inventory direction: background verified: full

Cited in the Panda's Thumb rebuttal as a state-survey arch dataset, so it was checked.  Queried directly through the ArcGIS REST API: 40 arch points, 35 trail lines, 36 parking points.  Fields are UniqueID, Name, Latitude, Longitude, County, Quadrangle, Physiographic region, map unit, Lithology, Property name and type, Public information, Source.  There is no status, condition, date or collapse field.  It is a visitor trail guide, not an inventory of the state's thousand-plus arches and not a monitoring dataset.

OpenStreetMap natural=arch

https://taginfo.openstreetmap.org/api/4/tag/stats?key=natural&value=arch

camp: unaffiliated topic: arch-inventory method: inventory direction: background verified: full

1,231 features worldwide (1,155 nodes, 73 ways), as of the taginfo snapshot of 26 July 2026.  Checked because OSM's full edit history is public and a deleted feature is in principle a detectable event.  In practice a deletion records a mapping correction, not a collapse, and the coverage is an order of magnitude below the NABS database.  Not usable, but worth knowing it was tried.

Historical Rock Falls in Yosemite National Park, California (1857–2011)

Stock, G. M.; Collins, B. D.; Santaniello, D. J.; Zimmer, V. L.; Wieczorek, G. F.; Snyder, J. B.  USGS Data Series 746 (2013).  doi 10.3133/ds746

https://pubs.usgs.gov/ds/746/ds746_text.pdf

camp: agency topic: rockfall-statistics method: inventory direction: background verified: full

Included here as the control — what a valid landform-failure dataset looks like.  154 years, more than a thousand events, each with date, location, volume estimate and source, plus an explicit discussion of completeness.  Nothing remotely comparable exists for arches anywhere in the world.  The gap between this and Table 7-2 is the whole story.

Arch mechanics and formation

Sandstone landforms shaped by negative feedback between stress and erosion

Bruthans, J.; Soukup, J.; Vaculikova, J.; Filippi, M.; Schweigstillova, J.; Mayo, A. L.; Masin, D.; Kletetschka, G.; Rihosek, J.  Nature Geoscience 7(8):597–601 (2014).  doi 10.1038/ngeo2209

https://web.natur.cuni.cz/uhigug/masin/download/Bruthans_et_al-NatureGeosci2014-pp.pdf

camp: mainstream topic: arch-mechanics method: experiment direction: mixed verified: full

Key figures: critical stress 0.5–2.6 kPa; tensile strength of saturated locked sand 0.226 kPa unconfined vs 3.88 kPa under load, a factor of 17; unconfined cubes destroyed by simulated rain in 5–7 minutes while loaded cubes survived an hour with ~70% intact.

The mechanism paper.  Shows experimentally that compressive stress locks sand grains in place, so erosion is self-limiting: as a pillar or lintel thins, stress rises and erosion slows, sculpting the rock into a load-bearing form.  Cuts both ways.  It is cited by the Institute for Creation Research as showing arches form fast in water, and rebutted by Michael Oard as artificial.  What it actually shows is a negative feedback that makes arches more durable once formed — which lengthens arch lifetimes rather than shortening them.  Underlying data are CC BY on figshare (doi 10.6084/m9.figshare.1056303).

Evolution of natural rock arches: A realistic small-scale experiment

Řihošek, Jaroslav; Slavík, Martin; Bruthans, Jiří; Filippi, Michal.  Geology 47(1):71–74 (2019).  doi 10.1130/G45421.1

camp: mainstream topic: arch-mechanics method: experiment direction: undercuts verified: citation

A laboratory experiment that grows arch forms from loaded, wetted sand.  Directly relevant to the claim that arch formation has never been observed: it has been reproduced under controlled conditions and photographed, even if not caught in the field.  Closed access; the full text could not be obtained through any legal route tried, so it is catalogued from its bibliographic record and from citations in Moore et al. 2020.

Between a beam and catenary: Influence of geometry on gravitational stresses and stability of natural rock arches

Moore, Jeffrey R.; Geimer, Paul R.; Finnegan, Riley; Bodtker, Jackson.  Geomorphology 364:107244 (2020).  doi 10.1016/j.geomorph.2020.107244

https://geohazards.earth.utah.edu/images/Moore_etal_GEOM_preprint.pdf

camp: mainstream topic: arch-mechanics method: model direction: undercuts verified: full

Nineteen arches modelled in 3-D from photogrammetry, spans 4–88 m, with material properties calibrated from measured resonance.  Two results matter.  Compressive stresses are "in all cases at least an order of magnitude lower than the reported compressive strengths" — arches do not fail by crushing.  And the mean principal stress ratio ranges from 0.7 (Two Bridge) to 32.9 (Rainbow Bridge), a factor of nearly fifty across nineteen features, which is why a single population-wide decay constant cannot be meaningful.  Contains the field's own honest concession: "Rates of arch formation and collapse are unfortunately poorly known."  Rainbow Arch, which collapsed during the study, had been flagged by a predicted tensile stress at the exact location of its visible crack.

Arcades: Products of stress-controlled and discontinuity-related weathering

Filippi, M.; Bruthans, J.; Řihošek, J.; Slavík, M.; Adamovič, J.; Mašín, D.  Earth-Science Reviews 180:159–184 (2018).  doi 10.1016/j.earscirev.2018.03.012

camp: mainstream topic: arch-mechanics method: review direction: background verified: abstract

The synthesis review of arch and arcade formation across lithologies and climates.  The standard reference for the stress-control school.

Numerical modeling of the evolution of arcades and rock pillars

Safonov, A.; Filippi, M.; Mašín, D.; Bruthans, J.  Geomorphology 365:107260 (2020).  doi 10.1016/j.geomorph.2020.107260

camp: mainstream topic: arch-mechanics method: model direction: background verified: abstract

Numerical implementation of the stress-erosion feedback, reproducing arcade and pillar forms.

Natural erosion of sandstone as shape optimisation

Ostanin, I.; Safonov, A.; Oseledets, I.  Scientific Reports 7 (2017).

https://pmc.ncbi.nlm.nih.gov/articles/PMC5725602/

camp: mainstream topic: arch-mechanics method: model direction: background verified: full

Treats arch and pillar formation as a shape-optimisation process driven by stress-dependent erosion.  Open access.

Role of fracture localization in arch formation, Arches National Park, Utah

Cruikshank, K. M.; Aydin, A.  Geological Society of America Bulletin 106(7):879–891 (1994).

camp: mainstream topic: arch-mechanics method: measurement direction: background verified: citation

The structural-geology account of why arches occur where they do at Arches: joint localisation in the Entrada fins.  Cited by both the Utah Geological Survey and by Michael Oard, who reads its description of localised erosion as "enigmatic."

The igloo and the natural bridge as ultimate structures

Handy, R. L.  Arctic 26(4):276–281 (1973).

camp: mainstream topic: arch-mechanics method: model direction: background verified: citation

The original inverted-catenary treatment of natural bridges as optimal compression structures.  Arctic is open access.

Mechanical weathering and rock erosion by climate-dependent subcritical cracking

Eppes, M.-C.; Keanini, R.  Reviews of Geophysics 55:470–508 (2017).

camp: mainstream topic: erosion-rates method: review direction: background verified: abstract

The review establishing subcritical crack growth as the dominant mechanical weathering process — the mechanism by which arches actually age between collapses.

Rates of subcritical cracking and long-term rock erosion

Eppes, M.-C.; Hancock, G. S.; Chen, X.; et al.  Geology 46(11):951–954 (2018).  doi 10.1130/G45256.1

https://www.osti.gov/servlets/purl/1477884

camp: mainstream topic: erosion-rates method: measurement direction: background verified: abstract

Measured subcritical cracking rates tied to long-term erosion rates.

Thermal influences on spontaneous rock dome exfoliation

Collins, B. D.; Stock, G. M.; Eppes, M.-C.; Lewis, S. W.; Corbett, S. C.; Smith, J. B.  Nature Communications 9:762 (2018).  doi 10.1038/s41467-017-02728-1

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

Instrumented documentation that daily thermal cycling alone drives exfoliation and rockfall — the same mechanism the Utah Geological Survey proposed for Wall Arch.

Arch monitoring

The University of Utah Geohazards group has instrumented more arches than everyone else combined.  Their publication list, with author PDFs, is at https://geohazards.earth.utah.edu/publications.html.

Ambient resonance of Mesa Arch, Canyonlands National Park, Utah

Starr, Alison M.; Moore, Jeffrey R.; Thorne, Michael S.  Geophysical Research Letters 42(16):6696–6702 (2015).  doi 10.1002/2015GL064917

camp: mainstream topic: arch-monitoring method: measurement direction: undercuts verified: abstract

Key figures: four resonant modes at 2.9, 6.0, 6.9 and 8.5 Hz; ~3% daily frequency variation with rock temperature; nine measurements over eighteen months.

The first arch health-monitoring study.  Its negative result is the important one: "We observed no permanent resonant frequency shifts related to irreversible damage of Mesa Arch during our study period."

Sparse Ambient Resonance Measurements Reveal Dynamic Properties of Freestanding Rock Arches

Geimer, Paul R.; Finnegan, Riley; Moore, Jeffrey R.  Geophysical Research Letters 47(9) (2020).  doi 10.1029/2020GL087239

https://geohazards.earth.utah.edu/images/grl60517_accepted.pdf

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: full

Key figures: 17 arches, spans 3–88 m, three resonant modes each between 1 and 40 Hz; outcrop-scale elastic moduli 0.8–8.0 GPa; low damping at all sites.

The dataset underlying the stress models.  Establishes that arches can be assessed non-invasively.

Anthropogenic Sources Stimulate Resonance of a Natural Rock Bridge

Moore, J. R.; Thorne, M. S.; Koper, K. D.; Wood, J. R.; Goddard, K.; Burlacu, R.; Doyle, S.; Stanfield, E.; White, B.  Geophysical Research Letters 43(18):9669–9676 (2016).  doi 10.1002/2016GL070088

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: abstract

Rainbow Bridge, and the effect of Glen Canyon Dam operations and other human sources on its vibration.

Use of Seismic Resonance Measurements to Determine the Elastic Modulus of Freestanding Rock Masses

Moore, J. R.; Geimer, P. R.; Finnegan, R.; Thorne, M. S.  Rock Mechanics and Rock Engineering 51(12):3937–3944 (2018).  doi 10.1007/s00603-018-1554-6

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: abstract

Method paper; source of the density and modulus values used in the later stress models.

Vibration of natural rock arches and towers excited by helicopter-sourced infrasound

Finnegan, Riley; Moore, Jeffrey R.; Geimer, Paul R.  Earth Surface Dynamics 9(6):1459–1479 (2021).  doi 10.5194/esurf-9-1459-2021

https://esurf.copernicus.org/articles/9/1459/2021/esurf-9-1459-2021.pdf

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: full

Key figures: seven arches and towers; vibration velocities up to 1000× ambient during close helicopter passes; peak 0.13 mm/s; admittance up to 0.11 mm s⁻¹ Pa⁻¹ — "lower than likely instantaneously damaging values."

Open access.  Relevant because it establishes that even a deliberately adverse anthropogenic loading does not instantaneously damage an arch.

Contribution of anthropogenic vibration sources to crack growth in natural rock arches

Finnegan, Riley; Moore, Jeffrey R.; Geimer, Paul R.  Frontiers in Earth Science 10:1035652 (2022).  doi 10.3389/feart.2022.1035652

camp: mainstream topic: arch-monitoring method: model direction: background verified: full

12–15 months of continuous vibration data, converted through subcritical fracture mechanics into predicted annual crack growth.  Result: cracks grow about 1% longer per year in anthropogenically noisy settings than quiet ones.  The absolute year figures in the paper (200 ky, 20 ky, 2 ky…) are input crack lengths for a sensitivity sweep, not lifetime estimates — a point worth knowing, because they are easy to misread as arch ages.

Meteorological Controls on Reversible Resonance Changes in Natural Rock Arches

Geimer, P. R.; Finnegan, R.; Moore, J. R.  Journal of Geophysical Research: Earth Surface 127(10) (2022).

camp: mainstream topic: arch-monitoring method: measurement direction: undercuts verified: citation

Establishes that most observed frequency variation is reversible and weather-driven — which is what makes an irreversible shift diagnostic of real damage, and why the absence of such shifts matters.

An Update on Techniques to Assess Normal-Mode Behavior of Rock Arches by Ambient Vibrations

Häusler, M.; Geimer, P. R.; Finnegan, R.; Fäh, D.; Moore, J. R.  Earth Surface Dynamics 9(6):1441–1457 (2021).

https://esurf.copernicus.org/articles/9/1441/2021/esurf-9-1441-2021.pdf

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: full

Open access methods paper.

Identifying fracture-controlled resonance modes for structural health monitoring: insights from Hunter Canyon Arch (Utah, USA)

Grechi, G.; Moore, J. R.; McCreary, M. E.; Jensen, E. K.; Martino, S.  Earth Surface Dynamics 13(1):81–95 (2025).

https://esurf.copernicus.org/articles/13/81/2025/esurf-13-81-2025.pdf

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: full

The current state of the art: array measurements resolving which internal fractures control which resonance modes.

Meteorological Controls on Reversible Resonance Changes in Natural Rock Arches

Geimer, Paul R.; Finnegan, Riley; Moore, Jeffrey R.  Journal of Geophysical Research: Earth Surface 127(10), 2022.  doi 10.1029/2022JF006734

camp: mainstream topic: arch-monitoring method: measurement direction: mixed verified: abstract

Key figures: 17 arches monitored, one for 15 months; resonant frequency sensitivity 0.5–6% per 10 °C; jumps of ~17% per −10 °C when pore water freezes; effects confined to "the outermost centimeters of the structure."

Establishes the detection floor.  Most measured frequency change is reversible weather, which is what makes an irreversible shift diagnostic — and also what makes small irreversible shifts invisible.

Combined Ambient Vibration and Surface Displacement Measurements for Improved Progressive Failure Monitoring at a Toppling Rock Slab in Utah, USA

Jensen, Erin K.; Moore, Jeffrey R.; Geimer, Paul R.; Finnegan, Riley.  Frontiers in Earth Science 12:1364653, 2024.  doi 10.3389/feart.2024.1364653

camp: mainstream topic: arch-monitoring method: measurement direction: mixed verified: abstract

Key figures: three continuous years at the Courthouse Mesa slab; crack aperture opening 2–4 mm/yr, primarily temperature-driven; seasonal frequency swing 29% annually and 16% maximum daily; "no clearly detectable irreversible changes in modal parameters, including frequency."

The most important qualifier on the whole monitoring literature.  A landform that is demonstrably and measurably failing produced a null vibration result, and the authors say detection "may be limited due to the magnitude of seasonal changes."  Cite this whenever citing the null results, or the null results are overclaimed.

Dynamic response of the Chamousset rock column (Western Alps, France)

Lévy, C.; Baillet, L.; Jongmans, D.; Mourot, P.; Hantz, D.  Journal of Geophysical Research: Earth Surface 115(F4), 2010.  doi 10.1029/2009JF001606

camp: mainstream topic: arch-monitoring method: measurement direction: mixed verified: abstract

Key figures: 21,000 m³ limestone column; ~20% drop in first resonance frequency in under two weeks before collapse in November 2007; four months of monitoring.

The positive control.  Ambient-noise monitoring can detect impending collapse — but what it detected was the terminal phase, days to weeks out, not a slow decline over the landform's life.

Thermomechanical behavior of Pravčická Brána Rock Arch (Czech Republic)

Greif, V.; Brček, M.; Vlčko, J.; Vařilová, Z.; Zvelebil, J.  Landslides 14(4):1441–1455 (2017).  doi 10.1007/s10346-016-0784-5

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: abstract

Long-term deformation monitoring of the largest natural arch in Europe, showing seasonal thermally driven movement.  The European counterpart to the Utah programme.

The application of non-destructive methods to assess the stability of the national nature monument of the Pravčická Brána rock arch

Vařilová, Z.; Zvelebil, J.; Hubatka, F.; Beneš, V.; Frolka, J.  AUC Geographica 49(1):49–59 (2014).  doi 10.14712/23361980.2014.5

https://karolinum.cz/data/clanek/24/Geogr_1_2014_04_Varilova.pdf

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: abstract

Open access.  Geophysical assessment of the same arch.

Microstructural analysis of sandstone from Pravčická brána, the largest rock arch in Europe

Maľa, M.; Greif, V.; Slavík, M.; Lanzendörfer, M.  Geomorphology 473:109617 (2025).  doi 10.1016/j.geomorph.2025.109617

camp: mainstream topic: arch-mechanics method: measurement direction: background verified: citation

Arch inventory and definitions

Natural Arch Information (FAQ, classification, taxonomy, formation, measurement)

Wilbur, Jay H.  Natural Arch and Bridge Society.

https://www.naturalarches.org/archinfo/index.htm

camp: arch-society topic: arch-inventory method: review direction: mixed verified: full

The definitional authority, produced by a Standards and Definitions Working Group that sat from 1990 to 1994 and included geologists, physical geographers, a physicist and a mathematician.  Two passages matter most.  On age: "Except for the handful of natural arches whose formation has been observed and recorded, there is currently no way to determine the age of a specific arch… it is safe to say that no natural arch is older than about 30 thousand years.  Most are probably between 5 and 15 thousand years old… Sea natural arches have typical ages of a few centuries, or even decades."  On count: about 2,000 significant natural arches in the United States (significant = two orthogonal opening dimensions whose product exceeds 10 m²) and perhaps 30,000 worldwide.  That 2,000 is a different number from the 2,000 in Arches National Park and must not be confused with it.  The society's own arch-formation page independently derives the same catenary stress-hardening mechanism that Bruthans et al. demonstrated experimentally twenty years later.

The World's Longest Natural Spans

Brandt-Erichsen, David; Wilbur, Jay H.  Natural Arch and Bridge Society.

https://www.naturalarches.org/big_arches/index.htm

camp: arch-society topic: arch-inventory method: inventory direction: background verified: full

The 19 known arches with spans over 200 feet: nine in China, nine on the Colorado Plateau.  Fairy Bridge (Buliu River, Guangxi) at 400 ft ranks first; Landscape Arch at 290 ft is fifth.  Notes the ranking has changed thirteen times.  Useful as a corrective: the Utah Geological Survey calls Landscape Arch the world's longest, which was true only before the Chinese arches were surveyed.

The Arches of Arches National Park: A Comprehensive Study

Stevens, Dale J.; McCarrick, J. Ed.  Mainstay Publishing, Moab (1988).

camp: arch-society topic: arch-inventory method: inventory direction: background verified: citation

The survey that produced the modern catalogue and the three-foot criterion.  Cited by Moore et al. 2020 and by the NPS administrative history; the book itself was not obtained.

World Arch Database Surpasses 10,000 Arches

Natural Arch and Bridge Society; database compiled by Van Bebber, Tom (6 July 2020).

https://www.naturalarches.org/blog/world-arch-database-surpasses-10000-arches/

camp: arch-society topic: arch-inventory method: inventory direction: undercuts verified: abstract

The global catalogue passed 10,000 arches in 2020 and is reported to have passed 17,000 by 2025 — growth entirely by discovery, and a useful check on the intuition that a rising catalogue means a rising population.

Arch dating

Dating the landscape evolution around the Chauvet-Pont d'Arc cave

Genuite, Kim; Delannoy, Jean-Jacques; Bahain, Jean-Jacques; Gresse, Marceau; Jaillet, Stéphane; Philippe, Anne; Pons-Branchu, Edwige; Revil, André; Voinchet, Pierre.  Scientific Reports 11:8944 (2021).  doi 10.1038/s41598-021-88240-5

https://www.nature.com/articles/s41598-021-88240-5

camp: mainstream topic: arch-dating method: measurement direction: undercuts verified: full

The only natural arch anywhere with a radiometric formation age.  Uranium-series and electron spin resonance on the abandonment surface of the Combe d'Arc meander: cut off between 108 and 138 ka (95%), mean 124 ka, with the opening itself older but narrow.  Caveat that must always accompany it: limestone, river meander cutoff, France — the method and the age do not transfer to Utah's eolian sandstone fins.  It nonetheless falsifies the Natural Arch and Bridge Society's own ≤30,000-year ceiling as a universal claim.

Age of Barrier Canyon-style rock art constrained by cross-cutting relations and luminescence dating techniques

Pederson, Joel L.; Chapot, M. S.; Simms, S. R.; Sohbati, R.; Rittenour, T. M.; Murray, A. S.; Cox, G.  PNAS 111(36):12986–12991 (2014).  doi 10.1073/pnas.1405402111

https://pmc.ncbi.nlm.nih.gov/articles/PMC4246979/

camp: mainstream topic: archaeology method: measurement direction: undercuts verified: abstract

Dates the rockfall that clipped the Great Gallery panel, bracketing the art itself.  Directly relevant twice over: it constrains how long a painted sandstone alcove wall survives, and it is one of very few cases where a rockfall event on the Colorado Plateau has been dated numerically.  Published with a critical comment and reply, both catalogued below — a model of how a contested date should be handled in public.

Incorrect representation of Barrier Canyon rock art site's history and other factors invalidate reported dates

Simon, N.; Reed, R.  PNAS 111(52):E5602–E5603 (2014).  doi 10.1073/pnas.1420818112

https://pmc.ncbi.nlm.nih.gov/articles/PMC4284594/

camp: mainstream topic: archaeology method: commentary direction: mixed verified: abstract

Reply to Simon and Reed: Independent and converging results rule out historic disturbance

Pederson, J. L.; et al.  PNAS 111(52):E5604 (2014).  doi 10.1073/pnas.1421319112

https://pmc.ncbi.nlm.nih.gov/articles/PMC4284606/

camp: mainstream topic: archaeology method: commentary direction: mixed verified: abstract

Constraining the age of rock art by dating a rockfall event using sediment and rock-surface luminescence dating

Chapot, M. S.; Sohbati, R.; Murray, A. S.; Pederson, J. L.; Rittenour, T. M.  Quaternary Geochronology 13:18–25 (2012).  doi 10.1016/j.quageo.2012.08.005

camp: mainstream topic: arch-dating method: measurement direction: undercuts verified: abstract

The methods paper behind the above.  This is the technique that could date a Utah arch and has not been applied to one.  Rock-surface luminescence dates when a rock face was last exposed to light — exactly the quantity an arch's formation age consists of.

Tree-Ring Dates From Arizona E: Chinle – De Chelly – Red Rock Area

Bannister, Bryant; Dean, Jeffrey S.; Gell, Elizabeth A. M.  Laboratory of Tree-Ring Research, University of Arizona (1966).

No stable public URL; repository.arizona.edu returns HTTP 403 to automated requests.  The Laboratory of Tree-Ring Research series is held at https://ltrr.arizona.edu/ and in the University of Arizona Campus Repository.

camp: mainstream topic: archaeology method: measurement direction: undercuts verified: abstract

Key figures: Mummy Cave, Canyon del Muerto — all specimens from the east and west alcoves date between A.D. 306 and 703, with cutting dates at 306, 430, 470, 474, 666, 680, 684, 691, 702 and 703.

The oldest securely dated standing structure under a sandstone overhang located in this search: roughly 1,700 years of continuous survival of both the alcove roof and the masonry beneath it.

Tree-Ring Dates From Colorado V: Mesa Verde Area

Robinson, William J.; Harrill, Bruce G.  Laboratory of Tree-Ring Research, University of Arizona (1974).

No stable public URL; repository.arizona.edu returns HTTP 403 to automated requests.  The Laboratory of Tree-Ring Research series is held at https://ltrr.arizona.edu/ and in the University of Arizona Campus Repository.

camp: mainstream topic: archaeology method: measurement direction: undercuts verified: abstract

Cliff Palace cutting dates 1229–1279; Balcony House 1250–1269.  Sandstone alcoves holding intact masonry for 750 years.

Erosion rates on the Colorado Plateau

Colorado Plateau uplift and erosion evaluated using GIS

Pederson, J. L.; Mackley, R. D.; Eddleman, J. L.  GSA Today 12(8):4–10 (2002).

https://rock.geosociety.org/net/gsatoday/archive/12/8/pdf/i1052-5173-12-8-4.pdf

camp: mainstream topic: erosion-rates method: measurement direction: undercuts verified: abstract

Open access.  Regional erosion budget for the plateau.

Dates and rates of arid region geomorphic processes

Nichols, K. K.; Bierman, P. R.; Foniri, W. R.; Gillespie, A. R.; Caffee, M.; Finkel, R.  GSA Today 16(8):4–11 (2006).  doi 10.1130/GSAT01608.1

https://www.geosociety.org/gsatoday/archive/16/8/pdf/i1052-5173-16-8-4.pdf

camp: mainstream topic: erosion-rates method: measurement direction: undercuts verified: abstract

Open access.  Cosmogenic-nuclide erosion rates in desert settings — the closest general reference to the rates that apply at Arches.

Rapid incision of the Colorado River in Glen Canyon

Cook, K. L.; Whipple, K. X.; Heimsath, A. M.; Hanks, T. C.  Earth Surface Processes and Landforms 34(7):994–1010 (2009).  doi 10.1002/esp.1790

http://geomorphology.sese.asu.edu/Papers/cook_etal_glen_canyon_incision_espl_09.pdf

camp: mainstream topic: erosion-rates method: measurement direction: undercuts verified: abstract

Early Pleistocene incision of the San Juan River, Utah, dated with 26Al and 10Be

Wolkowinsky, A. J.; Granger, D. E.  Geology 32(9):749–752 (2004).  doi 10.1130/G20541.1

camp: mainstream topic: erosion-rates method: measurement direction: undercuts verified: abstract

Cosmogenic burial dating near Natural Bridges National Monument — the closest numerical chronology to the three bridges.

Erosion rates and patterns in a transient landscape, Grand Staircase, southern Utah

Riley, K. E.; Rittenour, T. M.; Pederson, J. L.; Belmont, P.  Geology 47(9):811–814 (2019).  doi 10.1130/G45993.1

https://digitalcommons.usu.edu/geology_facpub/510

camp: mainstream topic: erosion-rates method: measurement direction: undercuts verified: abstract

Evolution and degradation of flat-top mesas in the hyper-arid Negev, Israel, revealed from 10Be cosmogenic nuclides

Boroda, R.; Matmon, A.; Amit, R.; Haviv, I.; et al.  Earth Surface Processes and Landforms 39(12):1611–1621 (2014).  doi 10.1002/esp.3551

camp: mainstream topic: erosion-rates method: measurement direction: undercuts verified: abstract

A desert analogue with directly measured cliff-retreat and surface-lowering rates.

Sediment dynamics below retreating cliffs

Ward, D. J.; Berlin, M. M.; Anderson, R. S.  Earth Surface Processes and Landforms 36(8):1023–1043 (2011).  doi 10.1002/esp.2129

http://geomorphology.sese.asu.edu/Papers/Ward_2011_ESPL.pdf

camp: mainstream topic: erosion-rates method: model direction: background verified: abstract

Relevant to the talus question: how debris below a retreating cliff is produced, stored and removed.  The mainstream answer to the creationist "missing talus" argument.

Quantitative study of a rapidly weathering overhang developed in an artificially wetted sandstone cliff

Bruthans, J.; Filippi, M.; Schweigstillová, J.; Řihošek, J.  Earth Surface Processes and Landforms 42(5):711–723 (2017).  doi 10.1002/esp.4016

camp: mainstream topic: erosion-rates method: measurement direction: supports verified: abstract

Measured overhang development under artificial wetting — a case where sandstone landform change was fast and quantified.  Tagged supports because it is the best mainstream evidence that these features can develop quickly when water supply is high.

Accelerated disintegration of in situ disconnected portions of sandstone outcrops

Filippi, M.; Slavík, M.; Bruthans, J.; Weiss, T.; Řihošek, J.  Geomorphology 391:107897 (2021).  doi 10.1016/j.geomorph.2021.107897

camp: mainstream topic: erosion-rates method: measurement direction: background verified: abstract

The converse of the stress-hardening result: once a block loses its load path it disintegrates far faster.  Explains why collapse debris disappears more quickly than the standing rock it came from — relevant to the missing-talus question.

Case hardening of sandstone

Conca, J. L.; Rossman, G. R.  Geology 10(10):520–523 (1982).

camp: mainstream topic: erosion-rates method: measurement direction: background verified: abstract

The classic statement of the competing (pre-stress-control) explanation for sandstone landform durability.

Origin of honeycombs: Testing the hydraulic and case hardening hypotheses

Bruthans, J.; Filippi, M.; Slavík, M.; Svobodová, E.  Geomorphology 303:68–83 (2018).  doi 10.1016/j.geomorph.2017.11.013

camp: mainstream topic: erosion-rates method: experiment direction: background verified: abstract

Rockfall statistics and the observation-window problem

This section is the statistical core, and it is where the mainstream literature has already solved — for rockfall — exactly the problem the arch argument gets wrong.

The Importance of Monitoring Interval for Rockfall Magnitude-Frequency Estimation

Williams, J. G.; Rosser, N. J.; Hardy, R. J.; Brain, M. J.  Journal of Geophysical Research: Earth Surface 124(12):2841–2853 (2019).  doi 10.1029/2019JF005225

camp: mainstream topic: rockfall-statistics method: measurement direction: undercuts verified: abstract

The most directly relevant methodological paper in the whole catalogue.  Demonstrates with high-frequency survey data that the apparent magnitude-frequency distribution of rockfall depends strongly on how often you look, because events coalesce and small events are missed between surveys.  A collapse rate inferred from an infrequent survey is not the true rate, and the direction of the bias is predictable.  This is precisely the defect in the 1993 fallen-arch list.

Influence of Monitoring Time on Rockfall Magnitude-Frequency Uncertainty

Phillips, et al.  58th U.S. Rock Mechanics/Geomechanics Symposium (2024).

camp: mainstream topic: rockfall-statistics method: model direction: undercuts verified: citation

The companion result on how long you must observe before a magnitude-frequency estimate stabilises.

Statistical analysis of rockfall volume distributions: Implications for rockfall dynamics

Dussauge, C.; Grasso, J.-R.; Helmstetter, A.  Journal of Geophysical Research: Solid Earth 108(B6):2286 (2003).  doi 10.1029/2001JB000650

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

Power-law volume distributions for rockfall, across several orders of magnitude.  The reason a population-averaged failure rate cannot be applied to the large members of a population.

Probabilistic approach to rock fall hazard assessment: potential of historical data analysis

Dussauge-Peisser, C.; Helmstetter, A.; Grasso, J.-R.; Hantz, D.; Desvarreux, P.; Jeannin, M.; Giraud, A.  Natural Hazards and Earth System Sciences 2(1–2):15–26 (2002).  doi 10.5194/nhess-2-15-2002

https://nhess.copernicus.org/articles/2/15/2002/nhess-2-15-2002.pdf

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

Open access.  How to extract a rate from an incomplete historical record — the discipline's own treatment of the problem the arch argument attempts by inspection.

Rockfall hazard and risk assessment in the Yosemite Valley, California

Guzzetti, F.; Reichenbach, P.; Wieczorek, G. F.  Natural Hazards and Earth System Sciences 3(6):491–503 (2003).  doi 10.5194/nhess-3-491-2003

https://nhess.copernicus.org/articles/3/491/2003/nhess-3-491-2003.pdf

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

Magnitude and frequency relations: are there geological constraints to the rockfall size?

Corominas, J.; Mavrouli, O.; Ruiz-Carulla, R.  Landslides 15(5):829–845 (2018).  doi 10.1007/s10346-017-0910-z

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

Emergent characteristics of rockfall inventories captured at a regional scale

Benjamin, J.; Rosser, N. J.; Brain, M. J.  Earth Surface Processes and Landforms 45(12):2773–2787 (2020).  doi 10.1002/esp.4929

camp: mainstream topic: rockfall-statistics method: inventory direction: background verified: abstract

Quantifying 40 years of rockfall activity in Yosemite Valley with historical Structure-from-Motion photogrammetry

Guérin, A.; Stock, G. M.; Radue, M. J.; Jaboyedoff, M.; Collins, B. D.; Matasci, B.; Avdievitch, N.; Derron, M.-H.  Geomorphology 356:107069 (2020).  doi 10.1016/j.geomorph.2020.107069

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

Demonstrates that archival photographs can be turned into quantitative change measurements over decades.  The method that could settle the arch question at Arches and has not been applied there — the park has photographs going back to the 1890s.

Progressive failure of sheeted rock slopes: the 2009–2010 Rhombus Wall rock falls in Yosemite Valley

Stock, G. M.; Martel, S. J.; Collins, B. D.; Harp, E. L.  Earth Surface Processes and Landforms 37(5):546–561 (2012).  doi 10.1002/esp.3192

camp: mainstream topic: rockfall-statistics method: measurement direction: undercuts verified: abstract

Documented progressive, accelerating failure — evidence that rock failure hazard rises with damage, which is exactly what an exponential (constant-hazard) model denies.

Optimising 4-D surface change detection: an approach for capturing rockfall magnitude-frequency

Williams, J. G.; Rosser, N. J.; Hardy, R. J.; Brain, M. J.; Afana, A. A.  Earth Surface Dynamics 6(1):101–119 (2018).  doi 10.5194/esurf-6-101-2018

https://esurf.copernicus.org/articles/6/101/2018/esurf-6-101-2018.pdf

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

A 6-year lidar survey reveals enhanced rockwall retreat and modified rockfall magnitudes/frequencies in deglaciating cirques

Hartmeyer, I.; Keuschnig, M.; Delleske, R.; Krautblatter, M.; Lang, A.; Schrott, L.; Prasicek, G.; Otto, J.-C.  Earth Surface Dynamics 8(3):753–768 (2020).  doi 10.5194/esurf-8-753-2020

https://esurf.copernicus.org/articles/8/753/2020/esurf-8-753-2020.pdf

camp: mainstream topic: rockfall-statistics method: measurement direction: background verified: abstract

Probability distributions of landslide volumes

Hergarten, S.  Natural Hazards and Earth System Sciences 4(2):309–313 (2004).  doi 10.5194/nhess-4-309-2004

https://nhess.copernicus.org/articles/4/309/2004/nhess-4-309-2004.pdf

camp: mainstream topic: rockfall-statistics method: model direction: background verified: abstract

Modeling cliff erosion using negative power law scaling of rockfalls

Barlow, J.; Lim, M.; Rosser, N.; Petley, D.; Brain, M.; Norman, E.; Geer, M.  Geomorphology 139–140:416–424 (2012).  doi 10.1016/j.geomorph.2011.11.006

camp: mainstream topic: rockfall-statistics method: model direction: background verified: abstract

Coastal arches — the fast end of the spectrum

Sea arches fail orders of magnitude faster than desert arches, and they are the most visible examples.  Any argument that treats "arches" as one population has to reckon with the spread documented here.

GlobR2C2 (Global Recession Rates of Coastal Cliffs): a global relational database

Premaillon, M.; Regard, V.; Dewez, T. J. B.; Auda, Y.  Earth Surface Dynamics 6(3):651–668 (2018).  doi 10.5194/esurf-6-651-2018

https://esurf.copernicus.org/articles/6/651/2018/esurf-6-651-2018.pdf

camp: mainstream topic: coastal method: inventory direction: undercuts verified: abstract

Open access.  The global compilation of measured coastal cliff recession rates by lithology — the single best source for showing how far apart chalk and hard sandstone are.

Coastal chalk cliff retreat rates during the Holocene, inferred from submarine platform morphology and cosmogenic dating

Duguet, T.; Duperret, A.; Costa, S.; Regard, V.; Maillet, G.  Marine Geology 433:106405 (2021).  doi 10.1016/j.margeo.2020.106405

camp: mainstream topic: coastal method: measurement direction: background verified: abstract

Reconstruction of the collapse of the 'Azure Window' natural arch via photogrammetry

Caruana, J.; Wood, J.; Nocerino, E.; Menna, F.  Geomorphology 408:108250 (2022).  doi 10.1016/j.geomorph.2022.108250

camp: mainstream topic: collapse-record method: measurement direction: background verified: abstract

Quantitative reconstruction of a documented arch collapse from tourist photographs — a template for what could be done at Arches.

Seismic Signature of the Azure Window Collapse, Gozo, Central Mediterranean

Galea, P.; Bozionelos, G.; D'Amico, S.; Drago, A.; Colica, E.  Seismological Research Letters 89(3):1108–1117 (2018).  doi 10.1785/0220170115

camp: mainstream topic: collapse-record method: measurement direction: background verified: abstract

The collapse was recorded on seismometers, giving the exact time and an energy estimate.  One of very few arch collapses anywhere with instrumental data.

Stability Assessment and Geomorphological Evolution of Sea Natural Arches by Geophysical Measurement

Leucci, G.; Persico, R.; De Giorgi, L.; Lazzari, M.; Colica, E.; Martino, S.; Iannucci, R.; Galone, L.; D'Amico, S.  Sustainability 13(22):12538 (2021).  doi 10.3390/su132212538

camp: mainstream topic: arch-monitoring method: measurement direction: background verified: abstract

Open access.  Geophysical stability assessment of a sea arch.

The Drowned Apostles: The Longevity of Sea Stacks over Eustatic Cycles

Bezore, R.; Kennedy, D. M.; Ierodiaconou, D.  Journal of Coastal Research SI 75:592–596 (2016).  doi 10.2112/SI75-119.1

camp: mainstream topic: coastal method: measurement direction: undercuts verified: abstract

Drowned sea stacks offshore of the Twelve Apostles, Victoria, surviving across glacial–interglacial sea-level cycles.  A direct measurement of sea-stack longevity far exceeding the "few centuries" usually quoted, in the very lithology creationist writers cite as fast-eroding.

Intermittent but Rapid Changes to Coastal Landscapes: The Tsunami and El Niño Wave-Formed Sea Arch at Laie Point, Oahu

Jordan, Benjamin R.  Geosciences 11(3):147 (2021).  doi 10.3390/geosciences11030147

camp: mainstream topic: arch-mechanics method: measurement direction: undercuts verified: abstract

Open access.  The best published case of an arch forming in historical time: witnesses report the Kukuihoʻolua Island arch did not exist before the tsunami of 1 April 1946.  A 135-tonne boulder was subsequently washed out from under it on the night of 24–25 February 2016, enlarging the opening further.  Rests on eyewitness testimony plus wave modelling rather than photographs, and the author hedges accordingly.

The Effect of Rock Collapse on Coastal Cliff Retreat along the Chalk Cliffs of Northern France

El Khattabi, J.; Carlier, E.; Louche, B.  Journal of Coastal Research 34(1) (2018).  doi 10.2112/JCOASTRES-D-16-00116.1

camp: mainstream topic: coastal method: measurement direction: background verified: abstract

Precariously balanced rocks — the best-dated fragile landforms

Nobody has dated an arch, but a whole seismological subfield exists to date fragile rock features, because their survival bounds past ground motion.  This is the strongest independent evidence on how long delicate rock landforms persist.

Precariously balanced rocks and ground-motion maps for Southern California

Brune, J. N.  Bulletin of the Seismological Society of America 86(1A):43–54 (1996).  doi 10.1785/BSSA08601A0043

camp: mainstream topic: pbr method: measurement direction: undercuts verified: abstract

The founding paper of the field.

Dating precariously balanced rocks in seismically active parts of California and Nevada

Bell, J. W.; Brune, J. N.; Liu, T.; Zreda, M.; Yount, J. C.  Geology 26(6):495–498 (1998).

camp: mainstream topic: pbr method: measurement direction: undercuts verified: citation

The first dating study: varnish microlamination and cosmogenic methods applied to balanced rocks, yielding ages in the tens of thousands of years.

Exposure dating of precariously balanced rocks

Balco, Greg; Purvance, Matthew D.; Rood, Dylan H.  Quaternary Geochronology 6(3–4):295–303 (2011).  doi 10.1016/j.quageo.2011.03.007  — with Corrigendum, Quaternary Geochronology 9 (2012), doi 10.1016/j.quageo.2012.02.011

camp: mainstream topic: pbr method: measurement direction: undercuts verified: abstract

The methodological standard: how to correct a cosmogenic exposure age for the exhumation history of a balanced rock.  The corrigendum should be read with it.

Earthquake Hazard Uncertainties Improved Using Precariously Balanced Rocks

Rood, A. H.; Rood, D. H.; Stirling, M. W.; Madugo, C. M.; Abrahamson, N. A.; Wilcken, K. M.; Gonzalez, T.; Kottke, A.; et al.  AGU Advances 1(4):e2020AV000182 (2020).  doi 10.1029/2020AV000182

https://spiral.imperial.ac.uk/server/api/core/bitstreams/e8c409bb-d165-4134-b5d2-869616329b81/content

camp: mainstream topic: pbr method: measurement direction: undercuts verified: abstract

Open access.  The modern treatment, with 10Be ages and fragility modelling used to constrain a real seismic hazard model.

Validation and Improvement of Probabilistic Seismic Hazard Analysis Models Using Precariously Balanced Rocks

Rood, Anna Hope.  PhD thesis, Department of Civil and Environmental Engineering, Imperial College London (2022).  Licensed CC BY-NC.

camp: mainstream topic: pbr method: measurement direction: undercuts verified: full

Key figures: 20 precariously balanced rocks dated; fragility ages 1.5–50 ka; oldest (LJ5) 49.8 ka (48.3–51.2 ka, 16th–84th percentile); 10 of 20 younger than 10 ka; a site-wide fragility age of 21 ± 2.4 ka for a group of seven.

The best single source of hard numbers on how long fragile rock landforms actually survive.  "Fragility age" is the time since the rock became precarious — the direct analogue of an arch's formation age, and measured by cosmogenic ¹⁰Be rather than estimated.  Cuts both ways and should be reported both ways: half the population is younger than 10,000 years, which supports the general point that these are young features, while the oldest has stood in a precarious state for about 49,800 years, which is eleven times the young-earth ceiling.

First Use of Fragile Geologic Features to Set the Design Motions for a Major Existing Engineered Structure

Stirling, M. W.; Abbott, E. R.; Rood, D. H.; McVerry, G. H.; Abrahamson, N. A.; Barrell, D. J. A.; Huso, R.; Litchfield, N. J.; et al.  Bulletin of the Seismological Society of America (2021).

camp: mainstream topic: pbr method: measurement direction: undercuts verified: abstract

Fragile rock features used as engineering evidence.  A useful marker of how much weight the profession puts on the survival ages of delicate rock landforms.

Toppling Analysis of the Echo Cliffs Precariously Balanced Rock

Veeraraghavan, S.; Hall, J. F.; Krishnan, S.; Hudnut, K. W.  Bulletin of the Seismological Society of America 107(1):72–84 (2017).  doi 10.1785/0120160169

https://authors.library.caltech.edu/records/na4ma-37d69/files/72.full.pdf?download=1

camp: mainstream topic: pbr method: model direction: undercuts verified: abstract

Stability Analysis of Fragile Rock Pillars and Insights on Fault Activity in the Negev, Israel

Finzi, Y.; Ganz, N.; Dor, O.; Davis, M.; Volk, O.; Langer, S.; Arrowsmith, J. R.; Tsesarsky, M.  Journal of Geophysical Research: Solid Earth 125(12):e2019JB019269 (2020).

camp: mainstream topic: pbr method: model direction: undercuts verified: abstract

Fragile sandstone pillars, which is closer to the arch case than the granitic balanced rocks of California.

Cosmogenic exposure and erosion history of Australian bedrock landforms

Bierman, P. R.; Caffee, M.  Geological Society of America Bulletin (2002).

camp: mainstream topic: erosion-rates method: measurement direction: undercuts verified: citation

Extremely low measured bedrock erosion rates on ancient cratonic surfaces — the end-member showing how long exposed rock can persist.

Rebuttals and critiques

Creationist astronomer calculates age of the Flood from Utah arch collapse

Phelps, Dan; Nuttall, Brandon.  The Panda's Thumb (24 August 2024).

https://pandasthumb.org/archives/2024/08/creationist-astronomer.html

camp: mainstream topic: rebuttal method: commentary direction: undercuts verified: full

The only dedicated published rebuttal of the arch argument that could be found anywhere.  Written by a Kentucky Paleontological Society officer and a retired Kentucky Geological Survey geologist.  It makes several real points — that no cumulative-loss plot was ever made, that climate change over the period is ignored, that arch types are heterogeneous, and (best of all) that "the discovery/documentation rate of arches in Kentucky has no bearing on their rate of formation or collapse (and the same holds for Utah)."  But it is a dismissal rather than a refutation: it corrects none of the numbers, computes nothing, does not notice the 2000/2020 date error, does not cite the mainstream arch ages, and misstates Faulkner's assumption as "a uniform rate of arch formation" when he assumed zero.  Its tone ("laughably bad") will persuade nobody who is not already persuaded.  This article is more critical of Faulkner than the Panda's Thumb piece is, and considerably more so of its sources.

Index to Creationist Claims

Isaak, Mark, ed.  TalkOrigins Archive.

https://www.talkorigins.org/indexcc/

camp: mainstream topic: rebuttal method: review direction: undercuts verified: abstract

Carries the erosion-rate claim family (continental denudation, Niagara, coastal retreat) but has no entry for the arch-collapse argument, which is itself a finding: the argument is recent enough and marginal enough that the standard counter-apologetics reference has never addressed it.

Related young-earth erosion arguments

The arch argument is a member of a family, and it inherits the family's characteristic defect — a rate measured over a short window, extrapolated over millennia, with the replenishing process omitted.  These are catalogued in full on the companion page for erosion rates and the age of the earth.

The Mystery of Missing Talus

Thomas, Brian.  Institute for Creation Research (2015).

https://www.icr.org/article/mystery-missing-talus

camp: creationist topic: yec-argument method: commentary direction: supports verified: abstract

The logical parent of the "where is the debris" question, and the better-argued version of it: talus piles are too small for millions of years of cliff retreat.  The mainstream answer is that talus is itself consumed — see Ward et al. 2011 and Filippi et al. 2021 above, the latter showing that disconnected blocks disintegrate faster than the rock they fell from.

Missing Talus on the Colorado Plateau

Oard, Michael J.  Proceedings of the Second International Conference on Creationism (1990).

No stable URL.  Cedarville University migrated its repository in 2025 and every DigitalCommons path for the ICC proceedings now resolves to a relocation notice; no Internet Archive snapshot of the PDF could be confirmed.  Obtained during research from a since-broken link.

camp: creationist topic: yec-argument method: measurement direction: supports verified: full

Eight measured sites in Colorado, New Mexico and Arizona, with photogrammetry.  The only ICC paper attempting a quantitative erosion argument on the Colorado Plateau, and the closest thing to original field data in the creationist arch literature.  Names the arches in its opening paragraph.

Discrepancies found in the sources themselves

Recorded because anyone re-checking this material will hit them.

• Rainbow Arch's collapse date is given three ways by the same research group. Moore et al. 2020 (Geomorphology) and Finnegan et al. 2022 (Frontiers) both say the winter of 2017–18; Geimer et al. 2020 (GRL) says 2018–19. The Natural Arch and Bridge Society and the National Park Service both support 2017–18, with a ranger noticing in February 2018. Do not cite a specific winter without flagging this.
• Geimer et al. 2020 says both 17 and 18 arches. The abstract says "17 natural arches"; the Study Sites section opens "The 18 natural arches in this study." Both are right in context — 17 sparse-sensor sites plus Musselman Arch characterised by dense array — but they read as a contradiction and the difference propagates into any arch-years calculation.
• Dzubay et al. 2022 is The Seismic Record 2(4):260–268, not 2(3), which is where several secondary listings put it.
• Wall Arch's span is 55 feet, not 71. The Utah Geological Survey says the widely repeated 71 feet is the "breadth," which "is not useful for comparing arches."
• Landscape Arch's span has been published as 291, 306, 290.4 and 290.1 feet by different surveyors between the 1950s and 2004. The 1940–41 rockfall is often said to have enlarged it from 291 to 306 feet, but Wilbur's 2004 laser measurement of 290.1 ± 0.8 ft suggests the 306 figure was a measurement artefact rather than a real enlargement. Do not use Landscape Arch as an example of documented arch growth; use Skyline Arch (November 1940) instead.
• The two "2,000 arches" figures are different numbers. Arches National Park has "over 2,000 documented arches" on a three-foot criterion; the Natural Arch and Bridge Society estimates "about 2000 significant natural arches" in the entire United States on a ten-square-metre criterion. They are routinely conflated.

What is missing

Six gaps, listed because an honest catalogue should name what it could not find.

1. No arch anywhere in Utah has been dated by any numerical method. Rock-surface luminescence (Chapot et al. 2012) is the obvious tool and has never been applied. This is a real, tractable, unfunded research question.
2. No inventory of former arches exists. Collapsed arches are recognisable — the NPS geologic report identifies them in the Fiery Furnace — but nobody has counted them anywhere. Both the young-earth and steady-state models make different, testable predictions about how many there should be.
3. No modern collapse monitoring at Arches National Park. There is no equivalent of the Yosemite rockfall database. The park's last systematic arch survey was in the early 1990s, and the "over 2,000" figure has not been revised since 1995.
4. The historical photograph archive has never been analysed quantitatively. Arches has been photographed since the 1890s and Guérin et al. 2020 showed that archival photographs can be converted into measured change. Nobody has done it here.
5. The SPAN newsletter archive (1988–present) was not obtained. It is members-only and is where arch collapses are first reported. Any serious collapse chronology has to go through it.
6. Řihošek et al. 2019 (Geology) could not be obtained in full. It is the laboratory arch-formation experiment and it is the single most relevant paywalled item in this catalogue.

Coverage statistics