When it first starts to rain, the first drops to fall always seem larger than what follows. Is this true? It also seems colder when it starts to rain. Is this true?
Bob Proctor,Lopez Island, Washington
If the rainfall is convective (showery, from tall clouds), I agree with you that the first drops to fall are often the largest. If the rainfall is stratiform (steady and gentle, from thick, layered clouds), then I disagree. Here's why.
Convective clouds—those that produce sudden and heavy showers, and sometimes thunder and lightning—have a very wide range of hydrometeor sizes, ranging from small to large droplets and from tiny ice crystals to large snowflakes. The mixture of hydrometeors may also include snow pellets and hail. In a developing thundercloud, the updraft is fairly strong. It suspends the hydrometeors growing within the cloud. Eventually, some of them become large enough to fall relative to the updraft, and these exit the base of the cloud first and can eventually reach the ground if they don't evaporate on the way down. In summertime, assuming that the temperature at cloud base is above freezing, those hydrometeors having the greatest fall speed will be first to exit the base of the cloud: hail or large raindrops. Large drops, say, three millimeters in diameter, fall at 7 to 8 m s−1 (meters per second). Hail 10 millimeters in diameter falls at about 9 m s−1. Larger hail falls much faster. Some of the largest drops come from hailstones that have melted before they reach the surface. Larger droplets are also the most likely to survive the fall between cloud base and ground without evaporating. The greater fall speed and lower susceptibility to evaporation explain why larger drops reach the ground first from convective showers.
Stratiform precipitation often begins with very small droplets reaching the ground first. In stratiform precipitation, the updrafts are steady and fairly gentle. Raindrops in stratiform clouds have a smaller range in size than in convective clouds. Those large enough to fall from the base of the cloud may not survive a long journey through relatively dry air between cloud base and the ground; initially, they may evaporate before reaching the ground, which is a phenomenon called virga. The vapor imparted to the subcloud air through evaporation increases the humidity below the cloud base, making it easier for subsequent raindrops to survive the fall. The first droplets to reach the ground have not quite evaporated, and so they tend to be very small, sometimes hardly perceptible on your cheek. Later, as the subcloud air moistens and the cloud base lowers, the drop size at the ground becomes larger.
It does get cooler when it starts to rain. The evaporation of raindrops into subsaturated air below the cloud base causes cooling, because energy is required from the air for the change in phase from liquid to vapor. In thunderstorms, this rain-cooled air spreads out as a gust front ahead of the advancing sheets of rain. These gust fronts are powered by strong downdrafts in heavy showers as well as evaporative cooling. The cooling may arrive many minutes before the rain itself. When stratiform precipitation begins, the temperature drops for the same reason—evaporating raindrops—but the cooling is usually simultaneous with the start of precipitation, unless a cold-front passage occurs before rainfall begins.
When clouds are thick and the humidity is about 80 percent, what actually determines when rain will fall?
John Baylis,Bountiful, Utah
The question is simple, but the answer is not. To avoid unnecessary complication, let's assume that the temperature throughout the cloud is above freezing. That means we don't have to deal with ice crystals—only with liquid droplets. Cloud droplets begin to form when air is lifted and cooled to the point where vapor condenses on microscopic particles in the air, called condensation nuclei. There are typically 100 to 1,000 condensation nuclei per cubic centimeter in the atmosphere, which is always enough for clouds to form. But if a greater number of nuclei compete for the available vapor, the tendency is for a larger number of smaller cloud droplets to form rather than a smaller number of larger droplets. This is important for a reason to be discussed shortly.
Droplets often grow from a micrometer in diameter to a thousand times larger before they fall to earth as raindrops. [One micrometer (μm) is one-millionth of a meter: 100 μm is the typical width of a strand of human hair.] Puffy cumulus clouds are observed to produce rain in less than 30 minutes, but condensation alone cannot grow raindrops this fast. How then do the drops grow large so quickly? The answer is collision and coalescence of raindrops.
Droplets have a natural fall speed that depends on size. Larger droplets fall faster than smaller droplets. The updraft responsible for forming the cloud in the first place initially lifts droplets that condense at the cloud base. As droplets grow within the cloud, they fall more quickly relative to the updraft. Once their fall speed exceeds the updraft speed, they fall out of the cloud as precipitation.
Suppose all cloud droplets are the same size. They will have the same fall speed relative to the updraft, and so there is very little chance for collisions. The distance between them is large relative to their size. A distribution of droplet sizes must develop within the cloud so that a larger droplet, falling faster, has a chance to overtake and collide with a smaller droplet. The mixing of drier air into the cloud near its boundaries can cause partial evaporation of cloud droplets, and turbulence within the cloud can bring these smaller droplets in proximity with larger droplets. Once a large disparity in droplet sizes develops within the cloud, collisions among droplets become likely.
Cloud physicists define a collision efficiency in terms of the radius of the collector drop R and that of the droplet collected r. The collision efficiency is equal to the fraction of those droplets with radius r in the path swept out by the collector drop that actually make contact with it. The figure shows the collision efficiency for a range of sizes of the collector drop (vertical axis) and the smaller droplets (horizontal axis). Note that collision efficiency is quite low for collector drops smaller than about 30 μm in radius, but becomes quite high for collector drops larger than 80 μm in radius. Note also that very small droplets (less than four μm in radius) seldom experience collisions; they are swept aside by the airstream around the collector drop.
From this information, we can correctly infer that a cloud consisting of a small number of larger droplets will have a better chance to deliver precipitation than a cloud consisting of a larger number of small droplets. The soaking drizzles produced by trade-wind cumulus clouds in Hawaii are a good example of this. The Pacific Ocean air is clean, so the concentration of cloud condensation nuclei is relatively low. Droplet growth by condensation is more rapid when fewer nuclei are competing for the vapor, and collision and coalescence can begin sooner.
Even if a collision occurs between two droplets, there is no guarantee that they will coalesce. They may bounce apart before their surfaces connect if a thin film of air separating them is not expelled. They may coalesce briefly, then separate into several or many droplets because of surface distortions accompanying the collision. If the two droplets do coalesce, they form a substantially larger droplet in an instant. Condensation cannot compete with this kind of droplet growth. Moreover, the fall speed of the new droplet is greater, thereby improving the odds of further collisions.
In summary, one could say that condensation makes the cloud, but collisions and coalescence grow droplets large enough to fall out of the cloud as precipitation.
Drizzle from stratus clouds, occasional sprinkles from stratocumulus clouds, and light showers from cumulus clouds (all with their tops below the altitude of freezing temperatures) form in this way. Clouds with ice in them as well as liquid can produce heavier precipitation, but additional processes, not described here, are involved.
Finally, it is good to remember that high relative humidity at the ground is not necessarily a precursor of precipitation. For example, the humidity is often high early in the morning and decreases during the afternoon as the temperature rises, but the moisture content of the air remains the same. A moist air mass in the lower atmosphere helps with cloud formation, but only if some weather disturbance can lift that air and cool it to the point of saturation.
Caption: Contours of collision efficiency, the probability that a larger drop of radius R (vertical axis) will collide with a smaller droplet of radius r (horizontal axis) when any part of the smaller droplet lies within the volume swept out by the larger droplet, which is falling faster (from A Short Course in Cloud Physics by R.R. Rogers and M.K. Yau).
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 at email@example.com, or by mail in care of Weatherwise, Taylor & Francis, 325 Chestnut St., Suite 800, Philadelphia, PA 19106.