During thunderstorms, I've noticed that the number of lightning flashes seems to peak shortly before the storm breaks. Once rainfall has begun, the frequency of lightning strokes seems to decrease significantly. I cannot prove this with actual lightning frequency statistics, but observations by others seem to corroborate that this happens. Why?
David A. James
The U.S. National Lightning Detection Network has been in operation since 1989. Sensors at over 100 U.S. locations detect and count cloud-to-ground lightning strokes. Satellites can detect in-cloud lightning strokes from space, and, in the past few years, so-called Lightning Mapping Arrays have delivered the capability to detect and count all lightning strokes within small geographic areas measuring tens of kilometers on a side. (They can even map the pathways taken by the strokes.) And so it is now possible to detect most lightning discharges, whether they strike the ground or not. However, the number of strokes detected by sensors is significantly greater than the number an observer would see as a storm approaches, breaks overhead, and moves on.
Caption: Rain and fragments of cloud obscure parts of a lightning stroke.
The flash rate of a given storm depends upon the atmospheric conditions in which it develops, for example, atmospheric instability and how the wind changes direction and speed with altitude. Moreover, the flash rate depends upon where the storm is in its life cycle. Let's assume that a mature storm with a constant flash rate passes by. What are some of the reasons why an observer wouldn't see all the flashes and why the flash rate would appear to change?
It is much easier to see lightning flashes at night than during the day, particularly in-cloud flashes.
- From a distance, approaching thunderstorms can often be seen from the side. Their cauliflower-like walls and spreading anvils are visible, as is the lightning within them. As the thunderstorm draws closer, however, lower clouds obscure this view, and lightning high in the cumulonimbus cloud may no longer be perceived.
- Heavy rain lowers visibility. I recall driving down I-25 at dusk in Colorado while watching an impressive rain shaft. I didn't notice much lightning until torrential rain engulfed me, and then I saw the stabbing strokes of lightning not visible from the outside. Even in this case, one wouldn't see all the flashes, only those nearby. The figure shows how a curtain of falling rain and perhaps a few wisps of cloud can obscure parts of a cloud-to-ground flash.
In summary, the perception of decreasing flash rate just before a thunderstorm breaks overhead is often due to the changing vantage point of the observer with respect to the storm, even when the true flash rate changes little.
I'm an observer for the Community Collaborative Rain, Hail, and Snow (CoCoRaHS) Network. I live about ten miles west of Redway, in northern California, near the coast. We're at an elevation of 1,600 feet, about 200 feet below a ridge top and on the west (ocean) side. I can see three ridges between me and the King Range, which is right on the coast. The King Range and the adjacent ridge to the east are both higher than we are. The third ridge is about even with our elevation, and the fourth and nearest one is not as high. We watch the low clouds coming in from the coast, dragging a sheet of rain across the ridges to the west. It always appears that we'll be drenched, but sometimes Redway, in the valley to the east, gets more rain than we do, even though we're near a ridge top on the windward side. In general, I'm surprised at the substantial differences in precipitation between us and our neighbors, even those who live very close to us. Can you explain?
(near) Redway, California
I found you on a map of California, by the “X” opposite the black arrow. You live in very rugged country about 20 miles from the Pacific Ocean, in one of the wettest parts of California. The highest peaks of the King Range are between 2,000 and 3,000 feet elevation west and northwest of you. This range receives about 80 inches of precipitation per year (dark green shading). Because you are in a weak “rain shadow” of the King Range, you get a little less. Local variations in rainfall do not show on this large-scale map of California, but they are substantial, depending upon the orientation of the ridges and valleys, and the wind direction. A difference of only 10 degrees in wind direction can make a big difference in the precipitation from individual storms.
Caption: Color-coded annual precipiation (inches) in California.
This still does not explain why valleys sometimes get more precipitation than the ridges directly upwind by just a few miles. Traditional meteorological wisdom says that the windward sides of ridges should receive the most precipitation. Clifford Mass and colleagues at the University of Washington in Seattle have run computer prediction models that can resolve terrain features as small as a mile or two. They have indeed simulated in the Pacific Northwest what you have observed: Sometimes valleys a short distance downwind from ridge tops receive more precipitation than the ridge tops themselves. The reason has to do with cloud physics.
As moist air approaches the ridge top, forced ascent causes cooling and condensation, but the precipitation isn't immediate. Cloud droplets take some time to grow by colliding with each other and coalescing to form larger drops. Eventually, they become large enough to fall out of the clouds as precipitation. While the raindrops are growing, they may be carried laterally a number of miles, particularly when ridge-top winds are strong. Condensation occurs on the upwind side of the ridge tops. Droplets about 0.5 millimeters in diameter fall at roughly 2 meters per second (4.5 mph). A fall of 1,000 meters (3280 feet) from the genesis region inside the cloud to the ground will take 500 seconds. In a wind of 20 meters per second (45 mph—not out of the ordinary in many Pacific storms), these droplets will be carried laterally 10 kilometers (6.2 miles) before reaching the ground, quite possibly the difference between landing near a ridge top and an adjacent valley. For snow, whose fall speed is generally 1 meter per second or less, the horizontal transport can be much greater.
To sum up, in complex terrain, precipitation is often highly variable, depending upon the wind direction and speed, the location of mountains, and the orientation of ridge lines. To explain the small-scale variability, one must resort to computer models capable of describing flow over terrain features as small as 1 kilometer in horizontal size. These models must take into account how precipitation grows inside of clouds, the fall speed of raindrops or snowflakes, and the possibility of substantial horizontal transport before the precipitation reaches the ground.
Weatherwise Contributing Editor THOMAS SCHLATTER is a retired meteorologist and volunteer at NOAA's Earth System Research Laboratory in Boulder, Colorado. Submit queries to the author in care of Weatherwise, Taylor & Francis, 325 Chestnut Street, Suite 800, Philadelphia, PA 19106.