Severe thunderstorms slammed the D.C. area Monday evening, unleashing some of the largest hailstones in memory.

Tennis-ball-sized hail, 2.5 inches in diameter, came crashing down in Rockville, Md. — probably the largest on record in Montgomery County.

A separate thunderstorm in Prince William County dropped golf-ball-sized hail, one to two inches in diameter, around Manassas, Dale City and Woodbridge. The National Weather Service received a report of an astonishing 2.75-inch hailstone, the approximate size of a baseball, in Charles County, from this same thunderstorm.

A third intense thunderstorm, which tore through eastern Fairfax County, the District and northern Prince George’s County, between 8:30 and 9:30 p.m., produced quarter- to golf-ball-sized hail and scattered wind damage.

Severe hail of this nature – occurring along three well-defined swaths – while common in the Plains, is extremely unusual in the D.C. area.

The map below shows the location of all large hail reports on May 2.

How did this happen?

A warm front that surged north through the area early Monday afternoon set the stage for this severe thunderstorm. The character of the air radically changed after the front passed. A dismally cool March-like morning abruptly turned summery. The sun burst out and temperatures sprinted up to near 80 degrees. This heating juiced the atmosphere, providing the instability to fuel the severe evening storms.

Importantly, as the air was heating up, a pocket of exceptionally strong wind shear (increase in winds with altitude) moved in from the west, spurred by a cold front and a wave of low pressure riding along it. The shear was stronger than forecast by the morning’s high-resolution computer models. In fact, the combination of strong shear and significant heating was sufficient to cook up a supercell-type thunderstorm complex across northern Virginia, between 6 p.m. and 7 p.m.

What is a supercell? It is a long-lived thunderstorm complex containing a strong, rotating updraft. Supercell is not necessarily synonymous with tornado: Only about 25 percent to 30 percent of supercells generate tornadoes. But the strong updraft effectively levitates hail embryos, allowing them to grow to a large size.

The explosive supercell development is seen in the eye-popping satellite image below:

The radar snapshot below illustrates just how intense this storm was in its organizing phase, about 6 p.m. Note two strong cores of reflectivity (returned radar power), colored red and pink. Pink colors correspond to 65 dBZ of reflectivity, which is at the very top end of the scale — meaning large hailstones mixed wind with rain (wet hail is a very effective scatterer of radar energy). The main core, south of Manassas, Va., was in the process of splitting away from a weaker core over Centreville.

The splitting phenomenon, which involves an initial storm cell that cleaves into a right-moving and left-moving part (like two cells dividing) is common during the early evolution of a supercell. The splitting is driven by rotational dynamics — the manner in which vortical (rotating) updrafts interact with the surrounding air flow.

Often the right-moving supercell continues to intensify its updraft, developing a mid-level mesocyclone and a hook echo. The left-moving supercell typically slowly weakens, although it, too, can produce a brief period of severe weather, but generally does not develop a strong mesocyclone and hook.

We have assembled four snapshots presenting the radar evolution of the splitting cell, as a mosaic, shown below. The evolution brackets 6 p.m. to 7 p.m., during which there were numerous severe thunderstorm warnings and a tornado warning issued for Charles County.

The main, right moving storm cell that cut south of D.C. had all the radar attributes of a classic supercell, including 1. longevity; 2. intense reflectivity core; 3. hook echo; and 4. V-notch. Some of these features are pointed out in the figure above. It was remarkably long-lived across the Chesapeake Bay and continued to produce severe weather over the Delmarva Peninsula.

The left-moving cell arced northeast through central Montgomery County, passing south of Olney, and did not develop a hook. However it maintained a very intense updraft, dropping two- to three-inch hail. Stones this size are truly exceptional for this region. An amazing 3D CAT-scan-like image of this cell is shown below. You can see that pink shaded “hail core” towering to nearly 20,000 feet.

Another supercell ripped through eastern Fairfax County, the District and Prince George’s County between about 8:30 p.m. and 9:30 p.m. It unloaded somewhat smaller hail then the previous two, but the constant pounding frightened witnesses. It also produced scattered reports of damaging winds (dramatic video footage here and here and, especially, here).

This same storm complex spurred flash flooding. Between 8 p.m. and 9 p.m., 0.99 and 1.37 inches of rain were reported at National and Dulles airports. Multiple swift water rescues were conducted in Montgomery County after cars became stranded in floodwaters.

Rainfall varied from about 0.5 to more than three inches, heaviest along the supercell tracks.

Forecast review

It was not a perfect forecast for Capital Weather Gang, or for other agencies.

Our early afternoon forecast was for showers and thunderstorms, some strong, becoming widespread during the evening. We did not foresee a widespread severe thunderstorm threat for the immediate D.C. region. The National Weather Service Storm Prediction Center (SPC) issued a Severe Thunderstorm Watch, initially placed well south and west of the District, during the mid-afternoon.

We reasoned that the morning and early afternoon cloud cover and low temperatures (in the wake of cold air damming) would prevent the atmosphere from fully destabilizing, with the greater severe threat over central and southwestern Virginia (to the south of the damming zone). This seemed to align with the thinking at both SPC and the National Weather Service Sterling office.

During the late afternoon, SPC extended its watch northeastward, encompassing all of Northern Virginia, D.C. and Montgomery County.

More often than not, experience suggests that the cool air is tough to erode, particularly when combined with overcast conditions. Many potential severe weather days in the past have been foiled by stubborn, cool air. High-resolution model guidance is starting to catch up, but it takes several events such as this one to inspire increased confidence in model predictions made six to eight hours in advance.

Storm multimedia

Below, find images of the some the impressive hail photos shared by readers in Virginia, Maryland and the District. We also have some photos showing some of the scattered wind damage around the region.

Virginia hail

Maryland hail

D.C. hail

Virginia wind damage

Maryland wind damage