How Fast Does Hail Fall? The Science of Terminal Velocity

Hailstones reach the ground at speeds ranging from roughly 20 mph for pea-sized pellets to over 100 mph for softball-sized monsters, with the exact figure depending mostly on the stone’s size but also on its shape, density, and the storm conditions around it. The speed a falling hailstone settles into is called its terminal velocity, the point where air resistance pushing upward balances the gravitational pull downward. That balance sounds simple, but real hailstones are lumpy, lopsided, and sometimes partially liquid, which makes predicting their speed surprisingly tricky.

Why Hailstones Don’t All Fall at the Same Speed

A hailstone accelerates the moment it begins to fall, but it does not keep accelerating forever. Air pushes back against any falling object, and that aerodynamic drag increases with speed. At some point the drag force equals the stone’s weight, and from that moment on it falls at a constant rate. That equilibrium speed is what atmospheric scientists call terminal velocity, and it is the speed most hailstones are traveling when they reach you or your roof.

Two properties dominate the equation. The first is mass: heavier stones need more drag to reach equilibrium, so they fall faster. The second is cross-sectional area, the “footprint” the stone presents to the air rushing past it. A larger footprint catches more air and slows the stone down. Because mass grows with the cube of a stone’s diameter while cross-sectional area grows with the square, bigger hailstones are disproportionately heavy for their wind resistance. That is the core reason a golf-ball hailstone can fall three or four times faster than a pea-sized one.

Speeds for Small Hail

For hailstones in the 0.5 to 1 centimeter range, high-speed photography of stones falling near the ground has produced a straightforward relationship: fall speed scales with the square root of the stone’s maximum width. A 1 cm hailstone, roughly the size of a marble, reaches about 12 to 13 meters per second at sea level, which is around 28 mph. A half-centimeter stone comes in slower, near 9 m/s (20 mph). These measurements carry an uncertainty of about 1.5 m/s in either direction, partly because no two hailstones are exactly alike and partly because turbulence inside the hailshaft jostles them around on the way down.¹

At this size, hail is mostly a nuisance. Stones under a centimeter sting if they hit bare skin and can shred delicate garden plants, but they rarely crack a windshield or dent sheet metal. Their low terminal velocity simply does not pack enough punch.

When Hail Gets Larger

Things change fast once stones grow past 2 or 3 centimeters. A hailstone the size of a golf ball (roughly 4.5 cm across) typically reaches terminal velocities in the neighborhood of 70 to 90 mph. At 7 to 8 centimeters, roughly tennis-ball size, speeds climb past 100 mph. The largest hailstones on record, wider than a softball, can theoretically exceed 110 mph, though measuring the speed of a stone that size in a live storm is extraordinarily difficult.

A large-scale field study in the U.S. Great Plains measured the physical properties of 2,295 natural hailstones, including maximum dimension, mass, and cross-sectional area, and used those measurements to build size-dependent relationships for terminal velocity and kinetic energy.² The researchers found that if you assume a hailstone is a perfect sphere, the speed estimates match older textbook numbers fairly well. But natural hailstones are not perfect spheres, and that distinction matters.

Shape and the Problem of the Imperfect Sphere

Pick up a real hailstone and you will notice it is rarely round. Many are roughly conical, others look like flattened discs, and some have protruding lobes where ice accreted unevenly as the stone cycled through the storm cloud. These irregular shapes create extra drag. A lumpy stone with the same maximum width as a smooth sphere catches more air, so it falls slower than the sphere would.

The Great Plains study confirmed this: when the researchers accounted for non-sphericity, terminal velocities were on average lower than those predicted for spheres of the same maximum diameter.² But the same dataset also showed that some irregular stones actually fell faster than their spherical equivalents. How? A few oddly shaped stones happened to be denser than average or had a shape that channeled airflow in a way that reduced drag rather than increasing it. The upshot is that a blanket correction for non-sphericity does not work; you need to know both the stone’s shape and its mass distribution.

Tumbling complicates things further. High-speed photography has shown that at least a third of hailstones in the 0.5 to 1 cm range were tumbling or oscillating as they fell.¹ A tumbling stone constantly changes the face it presents to the oncoming air, which means its drag fluctuates moment to moment. Its descent is not a smooth, predictable glide but more of a wobbly, lurching fall, sometimes drifting sideways by a surprising amount. For small hail, that turbulence matters less because the speeds are low. For larger stones with more energy, the oscillations can make the difference between a glancing hit and a direct one.

What the Storm Itself Does to Fall Speed

Terminal velocity describes how fast a hailstone falls through still air. Real storms are nothing close to still. A strong thunderstorm updraft can blow upward at 30 m/s (roughly 70 mph) or more, which is why hailstones exist in the first place: the updraft carries ice particles high into the cloud repeatedly, letting them collect more water and grow layer by layer. As long as the updraft exceeds a stone’s terminal velocity, the stone rises rather than falls.

A hailstone only starts its journey to the ground once it either grows heavy enough to overpower the updraft or gets ejected sideways into a region of the storm where the updraft is weaker. From that point it accelerates downward, and in calm air it would arrive at the ground moving at its terminal velocity. But the air below the cloud is not calm either. Downdrafts, the descending columns of air that produce gusty winds at the surface during a thunderstorm, can add their own speed to the stone’s descent. Research has shown that severe downdrafts accompanying hailfall can meaningfully increase the damage hailstones do when they reach buildings, effectively boosting the stones’ ground-level speed and kinetic energy beyond what terminal velocity alone would predict.³

Wind shear also tilts hailstone trajectories. In a supercell thunderstorm, the updraft and downdraft are separated, so hailstones often fall through a region where horizontal winds are strong. That is why hail frequently arrives at an angle rather than straight down, battering the windward side of a car or house while leaving the other side untouched.

What Happens to a Hailstone on the Way Down

A hailstone that forms at altitudes where temperatures are well below freezing enters warmer air as it descends. If the stone is small, it can melt entirely before reaching the ground, which is why hail is far more common from severe thunderstorms with strong updrafts and fast descent paths. Larger stones survive the trip but do not arrive unchanged. The outer surface begins to melt, creating a thin film of liquid water. Some of that water sheds off in the slipstream, reducing the stone’s mass. A lighter stone with the same shape decelerates slightly, so melting and shedding tend to slow hailstones down from the speeds you would predict based on their size at formation altitude.

At the same time, the presence of liquid water on the surface changes the stone’s aerodynamics. A wet, smooth surface has different drag characteristics than a dry, rough one. During what researchers call “wet growth,” a hailstone inside the cloud collects supercooled water faster than it can freeze, creating a spongy layer of mixed ice and liquid. Modeling of this process for hailstones in the 2 to 4 cm diameter range shows that the onset of wet growth and subsequent shedding depends on both air temperature and hailstone size, with larger stones entering the shedding regime at lower liquid water contents than smaller ones.⁴ The practical effect is that a large hailstone’s mass, density, and surface texture are all changing during its fall, making its actual ground-level speed harder to pin down than a simple formula would suggest.

What Hail Speed Means for Damage

Speed alone does not determine how destructive a hailstone is. Kinetic energy, which depends on both mass and the square of velocity, is a better predictor. A 5 cm hailstone carries far more energy than a 1 cm stone not just because it is faster but because it is dramatically heavier. Double the diameter and you roughly quadruple the cross-section but increase the mass eightfold (for a sphere). The net result is that kinetic energy climbs steeply with size, much more steeply than speed does.

Insurance and engineering researchers have built models that use radar-estimated hail sizes and the terminal velocity relationships described above to predict damage to buildings and cars. One recent open-source model evaluated against over 100 large hail events in Switzerland found it could estimate the correct order of magnitude for the number of damaged buildings in about 91 percent of events and the total cost of building damage in about 77 percent of events.⁵ The accuracy was somewhat lower for vehicles: the model estimated the number of damaged cars within the right order of magnitude about 74 percent of the time. These models rely heavily on terminal velocity estimates to convert radar-detected hail sizes into kinetic energy values, which means errors in the velocity assumptions ripple directly into damage forecasts.

The damage thresholds people care about are roughly as follows:

• Under 2 cm: Cosmetic damage to soft materials like vinyl siding and car paint, minor crop losses
• 2 to 4 cm: Cracked windshields, dented car panels, bruised roofing shingles, significant crop destruction
• Over 5 cm: Shattered windows, punctured roofing, totaled vehicles, serious injury risk for anyone caught outdoors

These thresholds shift depending on the material being hit, the angle of impact, and, as discussed, whether downdrafts are adding speed beyond terminal velocity. A 3 cm stone driven by a strong downdraft can do the work of a larger stone falling in calm air.

How Scientists and Engineers Simulate Hail Impact

Testing whether a product can survive hail requires launching ice at it under controlled conditions. Solar panel manufacturers, roofing companies, and automotive engineers all need standardized hail-impact tests. The standard approach uses frozen ice spheres fired from a pneumatic gun at speeds matching the terminal velocity for the sphere’s diameter. Early testing of photovoltaic solar panels by the U.S. Department of Energy, for example, used simulated hailstones projected at terminal velocity, along with steel ball substitutes and gravity-drop methods, to evaluate panel resistance.⁶

There is a catch: these tests typically use perfect ice spheres, which are easy to manufacture consistently. As the field data show, real hailstones are not spheres. A conical stone concentrates its energy on a smaller contact area, potentially doing more localized damage than a sphere of equivalent mass and speed. This mismatch between lab conditions and field reality is a known limitation. Some researchers have argued that current hail-resistance standards are either too lenient for irregular stones or too conservative for the average stone, depending on the shape distribution in a given storm. No consensus has emerged on how to fix this, partly because the shape of natural hailstones varies so much from event to event.

Measuring Hail Speed in the Field

Getting hard data on how fast real hailstones travel when they hit the ground is surprisingly difficult. You cannot exactly point a radar gun at a hailshaft and expect clean readings. Most speed estimates come indirectly: researchers measure a stone’s size and mass after it lands, then calculate what its terminal velocity should have been. This method works reasonably well but assumes the stone was falling at terminal velocity when it hit, which ignores the effects of updrafts, downdrafts, and melting.

Automated hail sensors deployed in places like Switzerland offer a more direct approach. These sensors measure the kinetic energy of each impact and then work backward to estimate the stone’s diameter, assuming a spherical shape, a standard ice density of about 870 kg/m³, and a drag coefficient of 0.5.⁷ The approach is practical and gives continuous data through an entire hail event, but the spherical assumption means the estimated diameters are only approximate for irregular stones. Still, networks of these sensors are producing the most detailed records of ground-level hail characteristics that exist, including information about how long hailfall lasts at a single point and how stone sizes vary minute to minute during a storm.

High-speed photography, the technique that gave the earliest direct speed measurements for small hail, remains useful but hard to deploy. You need a camera in exactly the right place at the right time, with enough contrast and lighting to track individual stones frame by frame. Doppler radar can estimate fall speeds from above, but radar sees a mix of rain, hail, and sometimes debris, and distinguishing hail echoes from rain echoes requires careful processing.

Graupel and the Fuzzy Boundary With Hail

Not every chunk of ice that falls from a storm qualifies as hail. Graupel, sometimes called soft hail or snow pellets, forms when supercooled water droplets freeze onto a snowflake, creating a small, opaque, crumbly ball of rime ice. Graupel is typically under 5 mm across and much less dense than true hailstones. Because of its lower density and rougher surface texture, graupel falls considerably slower than a hailstone of the same width.

Characterizing where graupel ends and hail begins is not just a naming exercise. Terminal velocity relationships for heavily rimed ice crystals, graupel, and hail each follow different curves because their bulk densities and drag coefficients differ.⁸ Weather models and radar algorithms that treat all frozen precipitation the same will misestimate both fall speeds and accumulation at the surface. For a person on the ground, the practical distinction is clear enough: graupel bounces harmlessly off your jacket, while a true hailstone at the same diameter would leave a welt.

Why Altitude and Air Density Matter

Terminal velocity is not a fixed property of a hailstone; it changes with altitude. The higher you go, the thinner the air, and thinner air provides less drag. A hailstone at 5,000 meters above sea level has a higher terminal velocity than the same stone at sea level because there is less air resistance to counterbalance gravity. As the stone descends into denser air, its terminal velocity drops. In practice, a hailstone is continually adjusting, falling a bit faster than its current-altitude terminal velocity as it enters thicker air, then gradually decelerating.

For most practical purposes, people care about the speed at ground level, and ground-level air density does not vary dramatically across typical inhabited elevations. But cities at high altitude, like Denver at about 1,600 meters, sit in noticeably thinner air than sea-level locations. Hailstones reaching Denver arrive somewhat faster than identical stones reaching, say, Houston. The difference is on the order of 5 to 10 percent, enough to matter for damage modeling but not enough to change the fundamental picture of which stone sizes are dangerous.

Temperature also plays a secondary role. Warm air is less dense than cold air at the same pressure, so on a hot summer afternoon, the air near the surface provides marginally less drag than on a cool day. Humidity matters too, though its effect is smaller. None of these factors rival size as the dominant variable, but they add uncertainty to any attempt to convert a radar-observed hail diameter into a precise ground-level speed.