Does Wet Sand Stop Bullets Better Than Dry?

No — and the popular advice has the sign backwards. In controlled laboratory tests, a 7.62 mm ball round penetrated 98 mm of dry sand but 263 mm of wet sand: the same bullet, the same rig, water the only variable. A 12.7 mm armour-piercing round stopped at 498 mm in dry sand and shot straight through the one-metre test tube when the sand was wet. The authors state it plainly: penetration is “more than twice in wet sand than in dry sand” for those two rounds. If you are stacking sandbags, wetting them down makes the barrier worse, not better.

That claim contradicts something almost every shooter has heard, so the rest of this article does three things: shows the measurements, traces exactly where the myth came from — it has a real and traceable origin — and marks the places where the honest answer is still "nobody has measured that".

What the measurements say

The reference dataset is Børvik, Dey & Olovsson (2015), "Penetration of granular materials by small-arms bullets", published in the International Journal of Impact Engineering. Four projectile types were fired into a 320 mm-diameter steel tube packed with 0–2 mm sand, in two states: dry at 1,726 kg/m³ and wet at 1,856 kg/m³ (the same sand with 130 kg/m³ of added water). Impact velocities were measured, not assumed.

Dry-sand depths are averages over repeated firings, with standard deviations from 2 mm (7.62 Ball) to 56 mm (12.7 AP) — the spread itself is a finding, and we come back to it. The two rounds marked in the wet column are the ones the authors single out: penetration more than doubled. The 12.7 mm AP round did not produce a number at all, because it perforated the entire one-metre test tube; the researchers had to re-run it at a reduced velocity to keep the projectile inside the rig.

The 12.7 mm armour-piercing round in wet sand did not produce a number because it perforated the entire one-metre test tube. In dry sand the same round stopped at 498 mm.

How the tests were run — and why it matters

Three details of the method change how these numbers should be read, and none of them appear in the summaries that circulate online.

The 7.62 mm rounds were fired from a smooth-bore barrel. The rig used a 7.62 × 63 mm smooth-bore Mauser for the smaller rounds and a 12.7 × 99 mm McMillan for the larger ones. The Mauser imparts no spin. That is a deliberate experimental choice, but it means the 7.62 figures describe a bullet that is not gyroscopically stabilised. The authors' own simulations put a number on the difference: with spin, predicted penetration fell from 211 mm to 181 mm — about 14% shallower. A round fired from a real rifled barrel would likely penetrate somewhat less than the table shows.

Trajectory deviation, not friction alone, is what stops the bullet. This is the paper's headline finding: penetration depth in dry sand is “strongly influenced by deviation of the bullet from its original trajectory”. In the simulations, the core begins to turn after some penetration and rotates roughly 180° before coming to rest. Introducing just 3° of yaw at impact cut predicted depth by about 25%. Sand does not simply grind a bullet to a halt — it steers it off course, and the longer path through more material is what does the work.

The spread is large and it is physical. Standard deviations in dry sand run from 2 mm for the 7.62 Ball to 56 mm for the 12.7 AP. That is not sloppy measurement; it follows directly from the point above. Where the bullet happens to strike relative to individual grains changes its path, and the authors reproduce the same scatter band numerically. Treat any single figure here as the centre of a distribution, not as a constant.

Where the myth actually comes from

This is the interesting part, because the belief is not random folklore. It has a documented source, and the source is being quoted out of context.

US Army Field Manual 5-103 (Survivability) contains a line stating that damp or wet earth and sand provide better protection than dry material. That sentence is real. It appears in a section about radiation shielding — where water content genuinely helps, because hydrogen atoms are effective at moderating neutrons.

It has nothing to do with stopping bullets. In fact the same manual, in the part that actually discusses projectile penetration, states the opposite: penetration increases with increasing water content. One document, two sections, two different physics problems — and decades of range advice built on the wrong one.

Why water lets the bullet travel further

Dry sand stops projectiles mainly through inter-granular friction. Each grain has to slide, rotate and grind past its neighbours, and that grinding is where the bullet's energy goes.

Adding water changes the mechanism. Børvik and colleagues attribute the increase to reduced friction between grains, partial liquefaction along the projectile path, shock-wave effects and elevated pore pressure. The water lubricates the very process that was doing the stopping. Wet sand is denser — and still less effective, which is a useful reminder that density alone does not predict stopping power.

There is a second, counterintuitive result in the same body of work: rifle rounds often stop shallower than pistol rounds in sand. In the Box O'Truth sand tests, everything from 9 mm to a 12-gauge slug stopped inside the first 140 mm of dry sand, and 5.56 mm XM193 disintegrated outright — reaching less depth than slower handgun bullets. High velocity destroys the projectile before it can convert that velocity into distance.

Armour-piercing is the exception, and it is about twice as deep

The 7.62 mm AP round reached 202 mm in dry sand against 98 mm for the ball round — 2.06 times deeper at essentially the same velocity. The reason is structural, not energetic: the hard steel core survives the impact, while the lead-cored ball round is crushed and loses its ability to penetrate almost immediately.

An independent study using X-ray measurement (Soriano-Moranchel et al., 2020, Materials 13(22):5243) finds the same ordering in dry sand: 126.6 mm for 7.62 mm FMJ M80 against 200.2 mm for AP M61. The armour-piercing figures agree closely — 202 mm and 200.2 mm — but the ball figures do not: 98 mm against 126.6 mm, a 29% gap between two laboratories firing nominally the same class of round into nominally the same medium.

That gap is worth sitting with rather than averaging away. It is the honest scale of uncertainty in this field, and it is why we treat any single published penetration depth as a data point rather than a constant.

For anyone modelling this, that means a single penetration curve for "sand" is wrong. Projectiles that survive intact and projectiles that come apart follow different curves, and the gap between them is larger than the gap between calibres.

What this means for sandbags and backstops

• Do not wet sandbags to improve them. Rain-soaked bags are measurably worse against bullets than dry ones. For the two rounds the authors highlight, penetration more than doubled; every combination they tested went deeper in wet sand than in dry.
• Dry sand is genuinely effective. Everything from 9 mm to a 12-gauge slug stopped within about 140 mm of dry sand in field testing. As cover, that is excellent value for its thickness.
• Size the barrier for armour-piercing, not for ball. In dry sand the 7.62 AP round reached 202 mm against 98 mm for ball — twice as far. In wet sand a 12.7 mm AP round exceeded one metre and left the test rig entirely.
• Doctrine already assumes generous thickness. US Army FM 3-06.11 rates 14 inches (356 mm) of sand or 28 inches (711 mm) of packed earth against heavy-calibre AP and ball at 200 metres. That is a thickness-to-stop figure, not a measured penetration depth — a distinction that gets lost often.

What we do not know

Sand is well anchored. Soil is not.

For clay, loess or silt there is no published, chronographed penetration depth in millimetres for any ordinary bullet. What exists is doctrine, plus one environmental study documenting how 5.56 mm rounds fragment in berm soil — a lead-contamination paper, not a penetration measurement. It supports the qualitative claim that 5.56 tumbles and breaks up shallow in soil, and supplies no number.

So when a calculator gives you a depth for "earth", ask what it is anchored to. Ours currently extrapolates from sand and says so. That gap is real, and we would rather flag it than paper over it — an unmarked guess is worse than an honest blank.

Sources

• Børvik, T., Dey, S. & Olovsson, L. (2015) — "Penetration of granular materials by small-arms bullets", International Journal of Impact Engineering 75:123–139. DOI 10.1016/j.ijimpeng.2014.07.016. Author manuscript openly available. 126 tests, four projectile types, five granular media, 320 mm × 1 m confining tube, measured impact velocities. Depths quoted here are from Table 5; velocities and standard deviations from Table 4; densities from Table 1.
• Soriano-Moranchel, F. et al. (2020) — "Simulation of bullet fragmentation and penetration in granular media", Materials 13(22):5243. Open access. X-ray measurement of penetration depth in dry beach sand.
• US Army FM 5-103, Survivability — penetration increases with water content; the "wet provides better protection" line belongs to the radiation-shielding section.
• US Army FM 3-06.11, Chapter 7 — thickness-to-stop figures for sand and packed earth.
• The Box O'Truth #7 — field testing in dry sand; directional only, no chronograph.