Cave Formations

"Size is not a reliable indicator of the age of a speleothem." — National Park Service, Carlsbad Caverns visitor brochure, 2003¹

"Growth could easily be over a hundred times that of today in some caves." — Michael Oard, Journal of Creation, 2020²

"It is logical to think of sulfuric acid as a far more potent cave-former than carbonic acid.  This is not necessarily so." — Arthur and Margaret Palmer, Journal of Cave and Karst Studies, 2000³

Summary

A limestone cave is made twice.  First, acidic water dissolves the rock and carries it away, which opens the cave.  Later, after the cave drains, water dripping from the roof gives off carbon dioxide and leaves some of its dissolved calcite behind.  That calcite builds the stalactites, stalagmites and flowstone, which are called speleothems.  Both steps have been used as clocks.  Cave guides often say a stalactite grows about an inch in a thousand years, so a big one must be very old.  Young-earth writers answer that caves and their formations can grow fast, and that all of them fit in the 4,300 or so years since Noah's Flood.

This article tests both sides by the numbers.  It leaves the uranium-thorium ages of speleothems to the article on Uranium-Series Dating, and asks what growth rates, dissolution rates and counted layers can show without them.

Conclusions, stated first.

1. Size alone dates nothing, and the young-earth side is right about that. Measured growth on natural stalagmites runs from zero to a few millimetres a year, a range of more than a hundredfold. Most of the speleothems at Carlsbad are not growing at all. The National Park Service says plainly that size "is not a reliable indicator" of age.¹ Any one-rate calculation, from either side, is worthless.
2. The concrete-stalactite argument should be dropped. Stalactites under bridges and buildings grow from lime leached out of concrete. That water takes carbon dioxide in from the air, the opposite of cave water, and the deposits grow tens to hundreds of millimetres a year.⁴⁵ Creation Ministries International warned its own readers in 2021 that they "are not directly analogous."⁶ Three of the best-known young-earth examples also misreport their sources. A lemonade bottle in Jenolan Caves gained a 3 mm crust in about thirty years, about 0.1 mm a year, but is cited as a 20 cm stalagmite.⁷⁸
3. The fastest natural growth is capped by chemistry, and young-earth timing needs far more. Michael Oard, the fullest young-earth writer on the subject, proposes growth "over a hundred times" today's rates in the 300 to 500 years after the Flood.²⁹ Dreybrodt's growth equation, fed Oard's most generous conditions, gives about 2 mm a year at the top of a stalagmite. Ten times more soil carbon dioxide gives only about 2.4 times more calcium, and faster dripping makes a stalagmite wider, not faster-growing. Carlsbad's 19 m Giant Dome would need 38 mm a year in a 500-year window, twenty times that ceiling. Over the full 4,300 years an ordinary 2 m stalagmite needs only 0.5 mm a year, which is easy. The difficulty is confined to the largest formations, and it is large there.
4. Dissolving the caves is not where the young-earth timescale breaks. The standard estimate for a carbonic-acid passage is 10,000 to 100,000 years,¹⁰ and under ideal conditions 5,000 to 20,000.¹¹ That is a factor of one to twenty beyond the post-Flood window, not a factor of a thousand. Carlsbad Cavern and Lechuguilla Cave were dissolved by sulfuric acid, as young-earth writers say. But sulfuric acid is not automatically faster, it needs oxygen from water or air that came from the surface, and the dated cave levels step down 1,100 m in elevation in order.¹²³¹³
5. Banded stalagmites are good clocks over short spans, and their counts are too short to decide the age of the earth. Where bands have been checked against an independent clock they are annual in the best cases: 78 yearly chemical cycles on a stalagmite that grew on an 80-year-old tourist boardwalk.¹⁴ They also go wrong. A Chinese stalagmite was over- and under-counted by 12 to 28 years in three centuries,¹⁵ a Norwegian one lays two bands a year,¹⁶ and a Carlsbad record failed a tree-ring test.¹⁷ The checked errors are a few per cent either way. No stalagmite count read for this article runs past about 4,000 years, so the bands cannot refute a young earth on their own. What they do show is that the uranium-thorium clock agrees with a count over recent centuries, where a young-earth speed-up would have to appear.

Verdict.  The young-earth case is right that cave formations cannot be dated by their size, right that some great caves were dissolved by sulfuric acid, and right that an ordinary cave could form and decorate itself in a few thousand years.  It is wrong about concrete, it misreads several of its own examples, and its timing for the largest stalagmites asks for rates that the chemistry does not allow.  Counted bands in stalagmites agree with the decay clocks within a few per cent where both have been measured.

Part 1: What the young-earth writers claim

The challenge they answer

Arthur Strahler, a geologist who wrote a long critique of creationism in 1987, set the test that young-earth writers have answered ever since.  Andrew Snelling quotes it at the head of his chapter on caves:  "If it can be shown that either the excavation of caverns or their subsequent filling must require a vastly longer time to accomplish than the post-Flood limit, literal acceptance of the Genesis chronology is untenable."⁸  The test is fair, and it has two halves:  how long it takes to dissolve a cave, and how long it takes to fill it with formations.

Austin, 1980: dissolution is fast, and the arithmetic is shown

Steven Austin's 1980 paper for the Institute for Creation Research is the founding young-earth treatment, and it shows its working.  Groundwater in the Mammoth Cave country of central Kentucky carries 0.16 grams of dissolved limestone and dolostone per litre.  Austin assumes a metre of rain a year soaks in, so each square kilometre loses 176 tons, or 59 cubic metres, of rock a year.  "Thus, if the dissolving power of the acid in one square kilometer of central Kentucky is carried in one conduit, a cave 1 meter square and 59 meters long could form in a year!"¹⁸  He adds that the water is only about 55 per cent saturated with calcite, so its dissolving power is not used up.¹⁸

His model has four steps.  Limestone is laid down and buried in the Flood.  Uplift and erosion as the Flood drains open joints.  Acidic groundwater flowing toward the new, lower water table dissolves horizontal passages.  Then the water table drops, the caves fill with air, and the formations grow.¹⁸

Austin also priced the formations, and he was careful.  Carlsbad's Great Dome, a stalagmite 19 m tall,¹⁸ holds perhaps 100 million cubic centimetres of calcite.  To grow it in 4,000 years at 0.5 g of calcite per litre of drip water would take "25 drops of water per second; which is a considerable flow.  Whether a stalagmite would be deposited in the above hypothetical situation is not known."¹⁸  That last sentence is the right instinct.  This article finishes the calculation he started.

Silvestru: the karst geologist

Emil Silvestru holds a doctorate in karst sedimentology and is Creation Ministries International's cave specialist.  He makes three claims that later writers repeat.

1. Carbonic acid cannot reach deep caves. "Thousands of measurements show that by dissolving limestone, the water loses its acidity within some 10 metres of the surface."¹⁹ Deep conduits must therefore have been opened by something else, which he identifies as hot "hydrothermal solutions" during the Flood that "could eat away enormous amounts of limestone in a year or even less."¹⁹
2. Drip points cannot stay put. A stalagmite hundreds of thousands of years old needs its water drop to keep "arriving at exactly the same point, with millimetre precision," even after growth stops for thousands of years. "Such explanations require common sense to take a nap."¹⁹
3. Ice Age dates are wrong. On Vancouver Island, speleothems dated 12,000 to 18,000 years old grew when the island was supposedly under two kilometres of ice. That argument rests on uranium-thorium dates and is treated, with its arithmetic, in the Uranium-Series Dating article.

Snelling and the mine stalactites

Andrew Snelling's two-volume Earth's Catastrophic Past (2009) gives caves a chapter.  Acidic hydrothermal fluids and pore water squeezed out of Flood sediments "would have rapidly dissolved out huge caverns along joints and fractures."⁸  For the formations he relies on fast growth seen in mines and tunnels, and he rejects the objection that these are not like caves:  "such rejection of those rapid growth rate observations is unwarranted, particularly where observations have been made in underground mines where conditions are very similar to those in caves."⁸  He concludes that "it is possible at these present rates to account for the growth of speleothems in cave systems within the last 4,500 years."⁸

Oard, 2020: the fullest case

Michael Oard's three-part series in the Journal of Creation is the most thorough young-earth treatment, and it engages the mainstream literature directly.

• Part 1 argues that most caves were dissolved by sulfuric acid, as Carlsbad was, and puts the work "within the middle and late Flood, and possibly a little afterwards." The acid came from organic matter buried in the Flood, and "need not ascend as believed for the Guadalupe Mountains caves, but likely descended during Flood uplift of the area."²⁰
• Part 2 lists the five controls on growth used in the mainstream literature: calcium in the drip water, cave temperature, carbon dioxide in the cave air, drip rate, and the thickness of the water film on the stalagmite.²¹
• Part 3 puts present rates at "around 0.1 to 2 mm/yr, but can vary from zero to more than 5 mm/yr," and multiplies each control by a factor for the wet, warm years after the Flood. He calls the factors "educated guesses" based on his thirty years as a weather forecaster.² The result: "Some could have easily grown over 100 times the rate of today. … It is not difficult to see how huge speleothems, especially the columns, could have grown in just 300 years following the Flood."²

A popular 2022 version makes the window "500 years after the Flood" and the rate 10 cm a year, enough for "a phenomenal growth of 50 m."⁹

Where young-earth writers accept old-sounding data

Three concessions in this literature matter for what follows.

• Oard accepts counted layers as measurements. He avoids "estimated growth rates that depend upon radiometric methods," and uses "only observed growth rates."² Among those observed rates he lists Belgian stalagmites that grew "up to 2.17 mm/yr" according to their annual layers.² If annual layers can measure a rate, they can also measure a span of years.
• David Lewis accepts the Carlsbad bands. Writing for Creation in 2010, he quotes a NASA description of "near-annual growth rings for the past 450 years from a stalagmite in Carlsbad Cavern," and contrasts their 0.1 to 0.3 mm widths with a wide ring in a Siberian museum specimen.²²
• Oard reports the Carlsbad cave ages without disputing the method. "Carlsbad Cavern (figure 6) is considered to be about 4 Ma old, … although the limestone itself is said to be about 250 Ma old."²⁰

The literature has also corrected itself, which is to its credit.  In 2021 Gavin Cox of Creation Ministries International warned that "Biblical creationists do need to be careful about how they use evidence, and as we shall see, stalactites growing under modern concrete bridges are not directly analogous to how speleothems grow in natural caverns."⁶  And in a 2020 letter, the creationist cave writer Arthur Manning wrote that "neither carbonic acid nor sulfuric acid dissolution alone, or together, would have been sufficient to produce the caves we see," and asked the question this article tries to answer:  "How long would it have taken to have formed the Big Room at Carlsbad Cavern by sulfuric acid alone, without erosion?"²³

Part 2: How fast stalagmites actually grow

The chemistry, in one equation

Rain soaking through soil picks up carbon dioxide made by roots and microbes, which turns it into weak carbonic acid.  On the way down it dissolves limestone until it holds about as much calcium as that soil's carbon dioxide allows.  When the water drips into a cave whose air holds less carbon dioxide, the gas leaves the water, the water becomes supersaturated, and calcite crystallises on the stalactite or on the stalagmite below.

Wolfgang Dreybrodt worked out the rate from laboratory kinetics in the 1980s, and it has been checked against real stalagmites since.  The growth at the top of a stalagmite is proportional to how much calcium the drip water carries above what the cave air allows, multiplied by a rate constant that rises with temperature:  W = 1.168 × 10⁹ α (c − c_eq) centimetres a year.²⁴  The constant α rises about tenfold from 0 °C to 30 °C.²⁴  Dominique Genty, Andy Baker and Barbara Vokal tested the theory on 31 drip waters in six European caves, with growth measured from annual layers, dated markers and bomb carbon.  Theory and measurement agreed with an R² of 0.69.  Growth tracked temperature (R² = 0.63) and calcium (R² = 0.61) "but not drip rate (R-2 = 0.09)."²⁵

That is Oard's list of controls, and he summarises it fairly.  It also shows which controls matter.  Calcium and warmth make a stalagmite grow.  Faster dripping mostly does not.

The measured range

Most measured rates on natural stalagmites fall between a few hundredths of a millimetre and about one millimetre a year.

• A modern stalagmite from Gibraltar, Gib04a, grew "~0.9 mm a-1," dated by its annual layers, a yearly carbon-isotope cycle and the bomb-carbon spike of the 1960s.²⁶
• A Turkish stalagmite grew "~0.1 – 0.2 mm yr-1" for most of its 650 years and 0.3 to 0.7 mm a year near the top.²⁷
• A tile set under an active drip in Villars Cave, France, gained 7 mm in four years, about 1.7 mm a year, helped by the flat tile and calcium-rich water.²⁷
• Oard reports Belgian stalagmites that grew "up to 2.17 mm/yr, with one year at 4.3 mm/yr," and cites Dreybrodt for "a maximum observed about 5 mm/yr."²²

At the other end the rate is zero.  Oard himself says that "practically all the speleothems" in the Guadalupe Mountains, home of Carlsbad Cavern, are "relic or 'dead'."²  So Silvestru's claim that an average of 0.01 mm a year "is ten times slower than the slowest measured today"¹⁹ is wrong.  A stalagmite that has grown for half its life and stood dry for the other half averages half its growing rate, and dry spells are exactly what an average over a long record includes.

Cave guides who quote one slow rate are also wrong.  A range from zero to about 5 mm a year spans more than two orders of magnitude.  Dividing a stalagmite's height by any single number from that range dates nothing, as the Park Service says.¹

Concrete stalactites are a different reaction

Stalactites under bridges, car parks and basements are real and they grow fast.  Garry Smith timed one concrete straw in Australia:  "the growth of 104 mm in 237 days," and 2 mm a day at its fastest.⁵  Stalagmites in a disused railway tunnel at Peak Dale, England, averaged about 33 mm a year since 1964, and a tile there gained 12.46 mm in 32 days.⁴

They are fast because the chemistry runs backwards.  Concrete leaches calcium hydroxide, a strong base, into water at "typically pH > 9 and commonly reaches pH 13.5."  Calcite forms "when atmospheric CO2 diffuses into the drop solution, as opposed to normal cave speleothem chemistry where CO2 is degassed from solution."⁵  A cave drip has to lose its carbon dioxide slowly through a thin film.  A concrete drip grabs carbon dioxide from the air and drops calcite almost at once.  Geologists call these deposits calthemites to keep them apart from speleothems.

Young-earth writers have known this for decades and said so in their own pages.  In 1987 Creation magazine wrote that the stalactites under the Lincoln Memorial and at Jenolan's walkways come from "cement-mortar which is highly permeable, allowing these formations to develop rapidly.  The resultant formation is quite powdery and brittle however."²⁸  In 2021 Cox told readers that "drawing comparisons with modern concrete stalactites to that of cave formations should only be done in a qualified sense."⁶  The argument from concrete should be retired.

The famous examples, re-read from their sources

Several examples that young-earth books repeat do not say what they are cited for.

• The Jenolan bottle. Snelling's textbook says a drink bottle left under a drip in the early 1950s "had deposited on it within thirty years a stalagmite almost 20 cm high, a growth rate of almost 0.7 cm per year."⁸ The magazine report he cites says something else: "a coating of calcite about three millimetres thick had already formed on the bottle. That's about one millimetre thickness each decade."⁷ An earlier Creation report agrees, 3 mm in 33 years.²⁸ The 20 cm appears to be the bottle. The real rate, 0.1 mm a year, is ordinary.
• The gold-mine stalagmite. Snelling describes "a 1.6 meter high stalagmite that grew in an abandoned gold mine in central New South Wales, Australia, (where there are limestones nearby and cave systems) in less than 140 years."⁸ His source, Don Batten, describes a "shawl," a curtain of calcite on a wall, in a tunnel that "cuts through solid basalt rock." The 1.6 m is the tunnel: "The horizontal tunnel is about 1.6 metres (about 5 feet) high."²⁹ Batten gives no size for the shawl at all.
• The Mt Isa mine. Snelling's "stalactites, some as long as 5 meters" growing "up to 9 cm per year" in a Queensland lead-zinc mine come from a photograph.⁸ The two captions he relies on give no length and no mineral, and disagree on whether the photograph was taken in 1977 or 1987.³⁰³¹ Mt Isa is a sulfide ore body, where oxidising ore makes sulfuric acid and mine "stalactites" are often gypsum or metal sulfates rather than calcite. Without an analysis, the example tells nothing about cave calcite.
• The Siberian ring. One ring about 15 mm wide in a museum stalactite from the Altai Mountains rests on the museum guide's statement that "The rings show how much it has grown each year."²² No dating, cave or drip was recorded. Oard, citing it, adds that "it appears that such a rate represents very unusual circumstances."²
• The pub cellar. Gavin Cox's soda straws in an English pub cellar reached 3.5 m in at most 224 years, "an average growth rate of 16 mm (0.64 inches) per year."³² A soda straw is a hollow tube a few millimetres wide that adds only a thin ring at its tip, so it lengthens far faster than a solid stalagmite of the same calcite supply. The cellar is built of limestone blocks and lime mortar, and no chemistry was reported.
• Sequoyah Caverns. The cave's owner reported stalactites behind a plastic panel that grew "about 25 centimetres" in under ten years.²⁸ If these were straws, as the length and speed suggest, they are the same geometric case as the pub cellar.
• The Carlsbad bat. A bat cemented to the top of a Carlsbad stalagmite, photographed by National Geographic in 1953, is said to show a stalagmite that "had grown so fast it was able to preserve the bat before the creature had time to decompose."²⁸ The photograph was not traced for this article. James Baker described the same thing in the Bulletin of the National Speleological Society in 1963: "Some bats are preserved in toto within the tops of active formations (speleothems) by the rapid deposition of calcite." He also found the bones of trapped bats piled in the floors of rooms where none now live.³³ A small bat is roughly a centimetre thick, so a stalagmite growing 1 mm a year buries one in about ten years, and the skeleton outlasts that easily. The bat shows that calcite can coat a carcass, which nobody disputes. It says nothing about how long the stalagmite under it took to grow.

Oard deserves credit on one point here.  The figure "0.002 to 400 mm/yr," printed in a mainstream geology paper, traces to the standard reference Cave Minerals of the World.  Oard followed it to its source and found that "Hill and Forti misquoted the growth rates from their source," whose 25 figures averaged 3.7 mm a year.²  He then dropped the 400.  The young-earth books that kept the 1 cm a year bottle and the 1.6 m stalagmite should do the same.

Silvestru's drip point

Silvestru's best argument is not about rates.  It is that a drip cannot keep landing on the same spot for 100,000 years, and certainly cannot return to it "with millimetre precision" after a pause of 10,000 years, because the soil and rock above a cave change in centuries.¹⁹  The sibling article on uranium-series dating admitted it had no answer.  There is one, and it is in the cave.

A stalagmite is fed by the stalactite above it.  The drop falls from the stalactite's tip, and the stalactite is calcite welded to the ceiling rock.  When the supply above changes, the drip slows, speeds up or stops, and the stalagmite records it as thinner layers or a hiatus.  When water returns along the same fracture, it leaves from the same stalactite tip, because the stalactite has not moved.  The precision Silvestru finds incredible is the precision of a fixed spout.  Dreybrodt's model of stalagmite shape starts from exactly this:  "The point of drop impact and the drip interval are constant in time."²⁴  When the drip interval does change, the same paper shows, the stalagmite's width changes with it:  "changes in drip interval τ and in kinetic constant α cause changes of the equilibrium radius and therefore also changes to the shape of the stalagmite."²⁴  Changes at the surface move the water's path through the soil;  they do not move the ceiling.

Part 3: Pricing the post-Flood window

What the young-earth timing asks for

The useful question is not "can a stalagmite grow fast?" but "how fast must the largest ones have grown, in the time the young-earth model allows?"  Three windows appear in the literature.  Oard's journal paper says the large formations "could have grown in just 300 years following the Flood."²  His magazine version gives 500 years.⁹  Snelling allows the whole 4,500 years since the Flood.⁸  The table divides the height of a stalagmite by each window.

The heights are from Austin (Giant Dome), Silvestru (Aven Armand) and Dreybrodt and Romanov, who show the Cuban giant, "Its height is 70 m, its diameter 20 m."¹⁸¹⁹²⁴  These are averages over the whole window, so any pause raises the rate needed for the rest.

Two results stand out at once.  An ordinary 2 m stalagmite fits easily in 4,300 years, at half a millimetre a year.  So does almost everything in an ordinary show cave.  The problem is confined to the giants, and to the short windows.

Oard's multipliers do not multiply

Oard reaches "over 100 times" by multiplying separate factors:  soil carbon dioxide "2 to 9 times," warm rock 1.5 to 3, ventilation 1.5 to 3 with another 50 per cent from dry air, drip rate "perhaps 2 to 6 times," and film thickness 1.5 to 3.²²  Taken at face value the product runs from about 14 to over 2,000.  But the growth equation does not multiply these things.

• Soil carbon dioxide enters through calcium, and calcium rises much more slowly than the gas. Calcite's solubility grows roughly with the cube root of the carbon dioxide pressure. At 10 °C, water in equilibrium with 1 per cent carbon dioxide holds 2.1 millimoles of calcium per litre; at Oard's "incredible" 10 per cent,²¹ it holds 5.0. Ten times the gas buys 2.4 times the calcium.³⁴
• Drip rate and film thickness reach a limit. Once drops arrive faster than the film can lose its excess calcium, more drops do not raise the rate at the top. They spread the water wider, and the stalagmite grows fatter instead of taller. In Dreybrodt and Romanov's words, the equilibrium radius "does not depend on the initial supersaturation," and it is set by the drop volume and drip interval.²⁴ This matches Genty's field result that growth does not track drip rate.²⁵
• Evaporation is tiny. Calcite left by evaporating water is the water lost times its calcium. Even if a cave lost 2 m of water a year, as an open pond does in a hot desert, water carrying 3.6 millimoles of calcium per litre would add only about a quarter of a millimetre a year.³⁴

The ceiling, computed

Feeding the growth equation the most generous case Oard describes gives a ceiling.  Drip water saturated at 10 per cent carbon dioxide.  Cave air ventilated down to 280 parts per million, the pre-industrial atmosphere.  Rock at 30 °C, which raises the rate constant tenfold over a cold cave.  The larger of two published fits of the rate constant.  The result is about 1.9 mm a year at the top of a stalagmite, and about 1 mm a year at 10 °C.³⁴  Measured natural rates mostly sit below this.  The best single years reported reach 4 to 5 mm, about twice the computed figure.²²  So the computed ceiling is soft by perhaps a factor of two or three, which is about how uncertain the rate constant and film thickness are (see Issues).  It is not soft by a factor of fifty.

Set against the table, the 500-year window needs about 20 times the computed ceiling for the Giant Dome and 70 times for the Cuban stalagmite;  the 300-year window needs 33 and 120 times.  Grant instead that the best single year ever reported, 5 mm, could be kept up every year:  the 500-year window still needs 8 times that for the Giant Dome and 28 times for the Cuban giant.  Over the whole 4,300 years the Giant Dome needs about the best reported year, every year, without a pause, and the Cuban giant three times it.

Austin's own 1980 arithmetic points the same way.  To grow the Great Dome in 4,000 years he needed 25 drops a second of water carrying 0.5 g of calcite per litre.  That concentration is 5 millimoles per litre of calcium given up entirely, which is the whole calcium load of water saturated at 10 per cent carbon dioxide and more than a drip can shed by degassing.  He was right to say he did not know whether such a stalagmite would form.¹⁸

Carlsbad makes it harder

The Guadalupe caves add a twist that the young-earth writers themselves supply.  Oard says their speleothems are almost all dead because the climate is "too dry today."²  So in his model the Giant Dome must have grown during the wet early Ice Age and then stopped, which puts it in the short window, not the long one.  Lewis accepts that a Carlsbad stalagmite carries "near-annual growth rings for the past 450 years," 0.1 to 0.3 mm thick.²²  Victor Polyak and Yemane Asmerom report that "Columnar stalagmites in caves of the Guadalupe Mountains during the late Holocene record a 4000-year annually resolved climate history."³⁵  If those bands are years, the Guadalupe stalagmites were still adding thin rings slowly through most of the time since the Flood, and the bulk of each column has to fit before that.  Whether the bands are annual is disputed (Part 5), but the young-earth window shrinks either way.

What a young-earth explanation would have to show

For an ordinary cave, nothing:  the post-Flood years are enough.  For the giants, one of three things.

1. A measured natural growth rate of several centimetres a year on a stalagmite, from cave water rather than concrete or mine water, with its chemistry reported. None was found for this article, and Oard's survey cites none.
2. A drip-water chemistry that holds far more calcium than carbon dioxide can dissolve, such as water that picked up calcium hydroxide or calcium sulfate on its way down, together with evidence for it in the calcite of the giant stalagmites themselves.
3. A longer window. Even Snelling's full 4,500 years needs the giants to grow every year at or above the best single years ever measured.

Part 4: How fast caves dissolve

Carbonic acid: the "10 metres" claim

Silvestru and Oard say carbonic acid is spent "within about 10 m of the surface," so it cannot carve caves hundreds of metres down.¹⁹²⁰  The observation behind it is real.  Most of the limestone that water dissolves is taken near the surface, where the water is most aggressive.  But the conclusion does not follow, and karst scientists explained why in the 1970s and 1980s.

Calcite dissolves fast while water is far from saturation and very slowly as it gets close.  Near saturation the rate falls off steeply, by roughly the fourth power of the remaining undersaturation in Dreybrodt's models.³⁶  So water that is 90 per cent saturated within metres of the surface keeps a small dissolving power for kilometres.  It carries that power deep and spends it slowly along the whole length of a fracture.  Once the fracture widens enough for water to rush through, fresh water reaches the far end before it saturates, and the fracture becomes a conduit.  Austin himself used the nonlinear solubility to show that mixing two saturated waters makes an undersaturated one.¹⁸

Dreybrodt, Gabrovšek and Romanov's book on modelling cave origins gives the times this takes.  A standard fracture 1 km long, 0.2 mm wide, with an ordinary hydraulic gradient, reaches "breakthrough" to fast flow in about 20,000 years;  a 10 m fracture in about 300;  a 10 km one in 467,000.  Carbon dioxide made by bacteria inside the fracture cuts the standard time to 7,500 years.³⁶³⁶  After breakthrough the conduit enlarges.  Arthur Palmer, whose 1991 paper is the standard reference, puts the maximum rate of wall retreat at "typically about 0.01-0.1 cm/yr" and says "Most caves require 10⁴-10⁵ yr to reach traversable size."¹⁰  William White's textbook gives the fast end:  "Under optimum conditions, the conduit could enlarge from the centimeter size to a diameter of 10 m in from 5000 to 20,000 years," and notes conduits that formed since the last ice sheet retreated.¹¹

This is the honest result:  dissolving an ordinary carbonic-acid cave passage takes thousands to tens of thousands of years, not millions.  The fastest mainstream figures are within a factor of one to five of the post-Flood window.  Cave formation by carbonic acid is not a strong argument for an old earth on its own, and an atheist reader who has heard that caves take millions of years to form has heard wrong.

Austin's mass budget is also basically sound.  Kentucky groundwater really does remove tens of cubic metres of rock per square kilometre each year.  His figure is about twice too high, because he assumed a metre of the rain soaks in while giving the runoff that actually leaves the area as 51 cm,¹⁸ but the order of magnitude stands.  The limit on cave-making is not how much rock the water can carry.  It is how fast a fracture can open, and how much of the dissolving is done in conduits rather than near the surface.

Sulfuric acid: Carlsbad and Lechuguilla

The large caves of the Guadalupe Mountains in New Mexico, Carlsbad Cavern and Lechuguilla Cave among them, were dissolved by sulfuric acid, and the young-earth writers who say so are right.  Carol Hill summarised it in 2000:  "It is now nearly universally accepted that the caves of the Guadalupe Mountains are of "hypogene" origin, having developed where ascending H2S, derived from hydrocarbons, became oxidized to sulfuric acid, which in turn dissolved out huge chambers at approximately the level (or levels) of the paleo-water table."³⁷

The evidence is in the caves.  Massive gypsum blocks up to 10 m high and native sulfur were left behind by the reaction.³⁷  The sulfur in them is isotopically very light, down to −25.6 per mil, unlike the sulfur of the local pyrite, which points to hydrogen sulfide from the oil and gas of the Delaware Basin next door.³⁷  The chemistry has two steps.  Hydrogen sulfide rising from the basin "reacted with meteoric, oxygenated, ground water to form sulfuric acid.  The sulfuric acid attacked the limestone and dissolved out the large caves."¹²  The same process is running today in Cueva de Villa Luz in Mexico, where drops of sulfuric acid hang from the walls.³⁸

Young-earth writers have adopted this enthusiastically.  Oard thinks "more than 50%" of caves, and perhaps all of them, were dissolved this way, during the uplift at the end of the Flood.²³  Snelling suggests "speleogenesis possibly only taking centuries."⁸

Is sulfuric acid fast?

Not necessarily, say the geologists who modelled the Guadalupe caves.  Arthur and Margaret Palmer:  "It is logical to think of sulfuric acid as a far more potent cave-former than carbonic acid.  This is not necessarily so."  Of the two hydrogen ions sulfuric acid supplies, usually only one attacks limestone, and "under totally closed conditions oxygen is unable to enter the system to produce sulfuric acid."³  The only measured rate found for limestone in sulfidic cave water, from tablets hung in the Frasassi caves of Italy, is 24 mm per thousand years in still water and 68 to 119 mm per thousand years where the water moves and mixes with oxygen-rich seepage.³⁹  That is 0.02 to 0.12 mm a year, slower than Palmer's ceiling for carbonic acid.  At those rates a wall retreats one metre in about 8,000 to 42,000 years.

The Palmers' model of a flowing supply is faster.  With 10 litres a second of water carrying 100 mg of hydrogen sulfide per litre, "this amounts to ~23 m3 of limestone per year.  At that rate, even the largest cave rooms in the Guadalupes could form in less than 100,000 years."  They conclude that "the main episodes probably lasted at least 10^5 years."³

Manning's question: the Big Room by sulfuric acid

The Big Room of Carlsbad Cavern has a volume of about 1,343,000 cubic metres.  Hill computed that dissolving it took 3 × 10¹² g of limestone and "6.7 x 10^{silvestru-2003b:bigroom][}ide gas," and compared that gas with "the amount of natural gas removed from Eddy County in 1978," which was ten times larger.¹²  Silvestru and Snelling retell this as needing "only about 10% of the H2S of the annual commercial production of the neighbouring gas fields."¹⁹⁸  That is a slip:  Hill compared the hydrogen sulfide needed with all the natural gas produced, not with the hydrogen sulfide in it.  The point survives in a weaker form.  The Delaware Basin holds plenty of sulfur and gas, and supply of hydrogen sulfide is not obviously the limit.

Oxygen is.  Each molecule of hydrogen sulfide needs two of oxygen to become sulfuric acid, and the oxygen must come from the surface, from air or from rain water that has soaked down.  The Park Service's own brochure says the rising water "combined with oxygen in the water table."¹  Dissolving the Big Room needs about 6.2 × 10¹⁰ moles of oxygen.  Water saturated with air holds about 11 mg of oxygen per litre,³⁶ so it takes about 1.7 × 10¹¹ cubic metres of oxygenated water to supply it.⁴⁰

• In one year, that is about 5,500 cubic metres a second through one room, a large river.
• In 300 years, about 18 cubic metres a second.
• Hill measured scallops, the ripple marks dissolving water leaves on a wall, in Carlsbad's Bat Cave. They record "a past flow velocity of about 0.5-2 cm/sec, or, for a passage cross section of 30 x 20 m, a flow rate of 3-12 m3/sec."¹² At that flow the oxygen for the Big Room alone takes about 460 to 1,850 years.

So the chemistry does not forbid dissolving the Big Room in a few centuries, if a river of oxygen-rich water flowed through it the whole time, or if much of the oxygen came from air in a partly drained cave.  It does rule out the version Oard proposes, in which acid made from Flood-buried organic matter "descended during Flood uplift" and "I cannot see any meteoritic water affecting the origin of caves."²⁰²³  Without meteoric water or air there is no oxygen, and without oxygen hydrogen sulfide does not become sulfuric acid.  Hill's own pairing of numbers shows that faster water leaves shorter scallops:  1.5 to 5 m scallops go with 0.5 to 2 cm a second.  On that relation, water racing through at metres a second would have cut scallops a few centimetres long, not the metre-scale ones Hill found.

The levels and their ages

The Guadalupe caves were not dissolved at one time.  They lie at several levels, and the mineral alunite, which forms during sulfuric-acid attack, gives argon ages that fall with elevation:  about 12 million years in caves near 2,000 m, then "~6 Ma (1230-1250 m), ~5 Ma (1180 m), and ~4 Ma (1100 m)" in Lechuguilla and Carlsbad.¹³  Victor Polyak and colleagues read this as the water table falling 1,100 m as the mountains rose, with "the caves at the higher elevations" the oldest.⁴¹  Oard reports the "about 4 Ma" for Carlsbad without disputing the method,²⁰ and in 1998 he reported the age-elevation correlation accurately.³¹

The absolute ages belong to the radiometric argument.  But the order does not depend on decay rates.  Even if every argon clock ran fast by the same factor, the oldest levels would still be the highest, and four or more separate episodes of sulfuric-acid attack would still have to happen, each at its own water table, as the land rose.  A young-earth model has to fit uplift, gas generation, several descending acid episodes, draining, and then all the speleothem growth of Part 3 into the end of the Flood and the centuries after it.  None of the young-earth sources read for this article attempts that sequence step by step.

Part 5: Counting years in stalagmites

Why some stalagmites have annual bands

Where the seasons differ, so does the drip water.  A wet season flushes organic matter and trace elements out of the soil, and a dry season does not.  Many stalagmites record this as alternating light and dark layers, or as layers that glow under ultraviolet light, or as yearly cycles in phosphorus, strontium or barium.  In a strongly seasonal cave each pair should be a year, and the bands can be counted like tree rings.⁴²  Oard uses such counts as observed growth rates,² and Lewis accepts the Carlsbad rings as "near-annual."²²

Where the count has been checked, and passed

The useful cases are the ones with an independent check.

• A boardwalk. Stalagmite MND-S1 in Moondyne Cave, Western Australia, grew on a tourist boardwalk that was in place from 1911 to 1992. Its yearly trace-element cycles gave "approx. 78 peaks, close to the actual age constraint of 80 years," which the authors call "a 2% uncertainty."¹⁴
• Bomb carbon. Nuclear tests in the early 1960s more than doubled the carbon-14 in the air, and the spike reaches stalagmites through the soil.⁴² The Gibraltar stalagmite Gib04a has an annual-layer age model "substantiated by annual carbon isotope cyclicity and correct identification of the atmospheric 14C activity 'bomb spike'."²⁶
• Layer-by-layer uranium-thorium. Chuan-Chou Shen and colleagues dated single layers of a Chinese stalagmite to ±0.5 to 2 years. "For the most recent 52 years, agreement between the number of bands and 230Th age (Fig. 2a) indicates that the laminae formed annually at this period."¹⁵
• A long count. In Scottish stalagmites with "thousands of years of regular luminescent laminae," occasional double bands gave "a counting error of <20 years over 1000 yrs (2%)."⁴³

Where it failed

The same literature reports failures plainly, and young-earth readers should know them.

• Shen's stalagmite, in a monsoon cave where annual bands are expected, was under-counted by 8 bands in each of two stretches of about 80 years and over-counted by 10 in another of 27 years. "We find that the layers do not always deposit annually. Annual bands can be under- or over-counted by several years during different multi-decadal intervals."¹⁵
• In Arctic Norway "a typical 'year' in a stalagmite from Rana, Norway, comprises two laminae, one from spring snowmelt and one from autumn flushing of soil organic matter."¹⁶
• In a Turkish stalagmite, uranium-thorium and radiocarbon dates showed the visible bands were "not annual but supra-annual, occurring every ~69 years."²⁷
• A Scottish stalagmite lost 95 years of layers to a pause that was found only by matching it against its neighbours and by uranium-thorium dating.⁴³
• Julio Betancourt and three tree-ring scientists compared band widths in a Carlsbad stalagmite with three tree-ring chronologies from the region. "We found no correspondence."¹⁷ Baker and Genty replied that the stalagmite and the trees answer to different seasons, while agreeing "that the annual nature (or otherwise) of laminae in stalagmites needs to be confirmed for each individual sample."¹⁶

So the young-earth answer, that a stalagmite can lay down more than one band a year, is correct.  It happens, and mainstream workers say so.  The checked errors, though, are a few per cent and run both ways:  Shen's count was short in some stretches and long in others.  Where a band is not annual it is usually caught, because the same papers date the bands independently.

How far the counts reach, and what they show

No stalagmite count read for this article reaches past the post-Flood window.  The checked counts cover decades to about a thousand years.  The longest records mentioned are a Chinese stalagmite with about 2,650 years of bands and the Guadalupe stalagmites' "4000-year annually resolved climate history."³⁵⁴⁴  Even if every one of those bands is a year, they fit inside 4,300 years.  So annual bands in stalagmites, on their own, do not refute a young earth.  An old-earth writer who says otherwise should be corrected.  The long counted clocks are elsewhere:  Greenland ice and lake sediments, treated under Ice-Core Layer Counting and Lake Varves.

The bands still bear on the young-earth case in two ways.

They test the uranium-thorium clock inside the post-Flood window.  Silvestru's reading of Vancouver Island has speleothems that date to 12,000 to 18,000 years growing only in the last 4,000, so that clock would have to run three to four and a half times fast.⁴⁵⁴⁵  Where single layers of a stalagmite have been dated and counted over the last three centuries, the two agree to within a few years at the top and a few tens of years at worst.¹⁵¹⁵  A clock running three times fast would make a 300-year-old layer read 900 years.  That is not seen.  The check covers centuries, not millennia, but it is the stretch of time a young-earth model says the speed-up affected.

They tie into other clocks through events.  The Laacher See volcano in Germany erupted at the end of the last ice age.  Its sulphur appears in a German stalagmite at a layer that uranium-thorium dates to "13,049 ± 30 years B.P. 1950."  Radiocarbon on trees the eruption buried, calibrated by tree rings, gives "13,006 ± 9 calibrated years B.P.," and the matching event in the Greenland ice count falls at "12,994 years B.P. GICC05."⁴⁶  A decay clock, a counted tree-ring calibration and a counted ice-layer clock agree to within about 55 years at 13,000 years.  That agreement is beyond the reach of any argument about stalagmite growth rates.  It is the uranium-thorium and ice-core arguments, and it is weighed in those articles.

Objections and responses

"Stalactites under bridges grow centimetres a year, so cave ones could too."  Concrete drips take carbon dioxide in from the air at a pH above 9;  cave drips give it off at a pH near 8.  The reactions run in opposite directions, and the concrete rate says nothing about the cave rate.⁵  Creation Ministries International has told its readers the same thing.⁶

"Mines are just like caves, so mine stalactites count."  A mine can be like a cave in air and darkness and still differ in its water.  Mine water passes through broken rock, cement and oxidising ore, and a sulfide mine like Mt Isa makes sulfuric acid.  A mine example counts when someone reports what the stalactite is made of and what the water carried.  None of the young-earth examples does.

"The Ice Age was wetter and warmer underground, so growth was a hundred times faster."  Wetter climates do raise growth, mainly by raising calcium and temperature.  But calcium rises with the cube root of soil carbon dioxide, faster dripping widens a stalagmite rather than raising it, and the equation at its most generous gives about 2 mm a year.  Oard's factors are, in his words, "educated guesses."²  A measured rate of centimetres a year on a natural cave stalagmite would settle it, and none has been published.

"Uniformitarian geologists assume today's rates."  The growth equation is chemistry, measured in laboratories and checked on stalagmites with known ages.  It is the same equation Oard uses to list his controls.  This article applied it to Oard's own proposed Ice Age conditions, not to today's.

"Annual bands are not annual;  some stalagmites lay two a year."  True, and the mainstream literature says so.¹⁶¹⁵  The checked errors are a few per cent either way, and no stalagmite count reviewed here is long enough to conflict with a young earth anyway.  The bands matter as a check on the uranium-thorium clock over recent centuries, where they agree with it.

"Carlsbad formed by sulfuric acid, which is fast, so the Flood could have made it."  The sulfuric-acid origin is right.  Speed is not automatic:  the geologists who modelled it say sulfuric acid is "not necessarily" more potent than carbonic acid, and the one measured rate in sulfidic water is 0.02 to 0.12 mm a year.³³⁹  The acid needs oxygen from the surface, about 1.7 × 10¹¹ cubic metres of oxygenated water for the Big Room alone.  Delivered in the Flood year, that is thousands of cubic metres a second through one room;  delivered over centuries afterward, it is plausible.  A post-Flood version is the one to defend, and it must still fit several descending acid episodes before the speleothems start.

"The alunite ages are radiometric and therefore wrong."  That is an argument for the radiometric articles.  But the ages fall steadily with elevation, which is what a falling water table predicts.  Any uniform error in the argon clock keeps that order, so the sequence of separate dissolution episodes stands even if the absolute ages do not.

"Cave guides say stalactites grow an inch in a thousand years, which proves an old earth."  They do not prove it, and young-earth writers are right to say so.  Rates vary by more than a hundredfold, and the Park Service itself says size is not a reliable guide to age.¹  The case for old cave formations rests on dating and on the largest formations, not on dividing height by a rate.

Issues and unanswered questions

1. The rate constant is uncertain. Two published fits of Dreybrodt's rate constant differ by up to 2.7 times at 0 °C and rise by 2.5 times or 10 times between 0 and 30 °C.²⁴²⁶ The ceiling in Part 3 uses the larger at each temperature. Even doubling it again leaves the 500-year window ten times short for the Giant Dome. Baker and others 1998, which both fits cite, was not obtained and would settle it.
2. Film thickness is poorly known. Baker, Mattey and Baldini note that "few constraints on δ exist."²⁶ A much thicker, turbulent film could raise growth. No measured natural stalagmite has shown rates above a few millimetres a year, which suggests this is not a large hidden factor.
3. No young-earth source attempts the Guadalupe sequence. None of the sources read works through uplift, hydrogen sulfide generation, four or more acid levels, draining and speleothem growth in order and in years. This article could not find a creationist treatment of Lechuguilla at all.
4. The Carlsbad bands are unresolved. Polyak and Asmerom's 4,000-year banded record and Betancourt's failed tree-ring test were read only in summary. Whether the Carlsbad bands are annual matters for the young-earth window, and both full papers are needed.
5. Measured sulfuric-acid rates are few. The Frasassi tablets are one site and one method. Rates from Villa Luz or the Kane caves of Wyoming, where the process is active and fast-flowing, were not found.
6. The longest stalagmite counts were not read. The 2,650-year Shihua record (Tan et al. 2003) and the global review of annually laminated stalagmites by Baker and colleagues in 2021 would show how long the checked counts run. If any reaches well beyond 4,300 years with independent checks, Part 5 should be revised.
7. The Silvestru drip-point argument has been answered by reasoning from the stalactite, not by a study. A paper documenting how drip points behave over long hiatuses, with dated stalagmites before and after, would test it directly.

Verdict

Two young-earth arguments about caves hold.  Size alone dates nothing, because growth rates span more than two orders of magnitude.  And sulfuric acid, not only carbonic acid, dissolved some of the largest caves in the world, Carlsbad and Lechuguilla among them.  An old-earth writer who quotes a single growth rate, or who says caves take millions of years to dissolve, is wrong.

Three arguments should be dropped.  Concrete stalactites are a different chemical reaction and say nothing about caves.  The Jenolan bottle and the New South Wales gold-mine "stalagmite" are misreadings of their own sources, and the Mt Isa stalactites have no reported mineralogy.  And the claim that carbonic acid is spent within 10 m of the surface ignores the slow kinetics near saturation that let it work deep underground.

Where the young-earth case has a real gap, it is the largest formations.  Fitting a 19 m to 70 m stalagmite into 300 to 500 years needs growth of 4 to 23 cm a year:  twenty to 120 times what the chemistry allows under the most generous Ice Age conditions Oard proposes, and eight to 47 times the best single year ever measured.  Even across the full 4,300 years the giants need sustained rates at or above the best single years ever measured.  To close the gap, a young-earth model would have to show a natural cave stalagmite growing centimetres a year from cave water, or a drip chemistry that carries far more calcium than carbon dioxide allows.  Neither has been shown.

Sources