What is virga, and why radar shows rain that never lands

Virga is precipitation that falls from a cloud but evaporates, or sublimates if it is ice, before it reaches the ground. Radar can show it as an ordinary echo because the beam samples the air well above the surface, where the rain or snow still exists.

The how to read weather radar guide explains that the beam rises with range and that a radar image is not a photograph of rain. Virga is the everyday case where that distinction decides whether you get wet.

The definitions

The National Weather Service glossary defines virga as “streaks or wisps of precipitation falling from a cloud but evaporating before reaching the ground.” The WMO International Cloud Atlas treats it as a supplementary feature of a cloud rather than a cloud type: “vertical or inclined trails of precipitation (fallstreaks) attached to the under surface of a cloud that do not reach the Earth’s surface.” The Atlas lists it mostly with cirrocumulus, altocumulus, altostratus, nimbostratus, stratocumulus, cumulus and cumulonimbus.

Where virga is common

Virga needs dry air between the cloud base and the ground. The NWS Spotter’s Field Guide puts it plainly: the atmosphere below the clouds tends to be very dry, and rainfall evaporates before it touches the ground.

The NWS Nashville office’s downburst page describes the environment in which virga is routine: a deep, dry boundary layer of at least 3 km, and cloud bases in the western United States and over the High Plains commonly as high as 500 mb with predominantly dry layers below. It is not only a Western sight. NWS Louisville, in Kentucky, has explained why radar can show precipitation that does not arrive: with a dry air mass in the low levels, rain cannot always get through that layer before it evaporates.

Why radar shows rain that is not reaching the ground

A WSR-88D does not measure precipitation at the surface. It measures what returns energy inside a sample volume at the height of the beam, and that height grows quickly with distance. By a 4/3-earth calculation (beam centre only, height above the radar antenna, level terrain), the 0.5° lowest tilt sits at these heights:

The first two rows match the figures in the microburst guide; the rest are the same calculation carried further. Set them beside the Nashville description of a dry layer at least 3 km deep, and the problem is plain: at longer ranges, even the lowest tilt can be sampling precipitation above most of the layer that will evaporate it. The radar is reporting correctly what is at the beam. It has no way to see what happens in the air beneath.

This is also why radar colors are not intensity at the ground. The same echo can be a light shower at the surface near the radar and entirely virga farther away, and the color table will not tell you which.

Radar precipitation versus surface precipitation

The gap matters enough that NOAA’s own precipitation estimates correct for it. Zhang and colleagues at NSSL, describing the dual-polarization radar precipitation estimate they built for operational use (Journal of Hydrometeorology, 2020), apply an evaporation correction to reduce “false light precipitation in the radar QPE due to virga.” The correction depends on the rain rate, the height of the radar estimate, and the atmosphere between that height and the ground. They report that the new estimate reduced a wet bias for scattered light precipitation in the southwest and north-central United States, where boundary-layer evaporation is significant.

Evaporative cooling and dry microbursts

Evaporation takes heat from the air around the falling drops. The Spotter’s Field Guide describes the chain: the air cools, grows denser, and gravity pulls it toward the surface, so gusty winds occur in the area of the virga. The Nashville page calls evaporative cooling within the virga shaft the primary forcing of the dry microburst, which comes with little or no rain at the ground. How those outflows form, and why they are so hard to catch on radar, is covered in the microburst guide.

How to tell on radar

No single product labels an echo as virga. Several clues, read together, make it likely.

• Echo that weakens downward. Evaporation removes drops as they fall, so reflectivity decreases toward the ground. Xie and colleagues (Journal of Atmospheric and Oceanic Technology, 2016) describe a decrease of Z toward the surface as the result. Comparing tilts at the same location, or looking at the lowest tilt close to the radar where the beam is low, shows whether the echo thins out as the beam gets nearer the ground.
• Differential reflectivity that rises downward. Small drops evaporate faster than large ones. A modelling study by Kumjian and Ryzhkov (NSSL and the University of Oklahoma, 2009 AMS radar conference; full study in the Journal of Applied Meteorology and Climatology, 2010) found that this preferential loss lowers reflectivity and KDP while raising ZDR, because the drops that remain are larger on average. Xie and colleagues add that warming toward the surface can shrink that ZDR increase or even reverse it.
• Dry surface observations underneath. A station directly under the echo reporting no precipitation is the most direct evidence. Kumjian and Ryzhkov suggest pairing radar measurements aloft with the humidity of the layer below, from observations, soundings or model output, to estimate how much rain survives the fall.
• A dry environment. High cloud bases over a deep, dry low-level layer are the conditions the Nashville and Louisville offices describe. Light echoes on such a day deserve suspicion.

WeatherStory Radar decodes reflectivity and ZDR from raw Level II, gives the beam height of any gate you click, and shows surface observations on the same map.