A week ago Sunday, a tornado devastated the small town of La Plata, Md., killing three people and injuring more than 100. This storm, the strongest to strike east of the Appalachians in at least a century, was rated F-5 on the Fujita-Pearson Tornado Scale.

The Fujita-Pearson scale ranks storms from F-0 (light damage, estimated winds up to 72 mph) to F-6 (inconceivable damage, estimated winds of more than 318 mph). The strongest tornadoes ever recorded were F-5s (incredible damage, with winds between 261 and 318 mph.)

In the past 50 years, there have been 51 documented F-5s in the United States, and seven of those occurred during a single tornado outbreak that totaled Xenia, Ohio, on April 3, 1974.

Even though F-4 and F-5 tornadoes account for only about 2 percent of all U.S. tornadoes, they cause 70 percent of tornado deaths.

The Fujita-Pearson scale was developed by the late Ted Fujita, a professor of meteorology at the University of Chicago, and refined with the help of Allen Pearson, then director of the National Weather Service's National Severe Storms Forecast Center in Kansas City, Mo. It estimates the wind speed of tornadoes based upon the damage they inflict. Damage patterns are used because it is impossible to make wind-measuring equipment that would survive a direct hit from a tornado.

On this scale, the ``incredible damage'' inflicted by an F-5 tornado can include leveling or sweeping away strong frame houses, sending automobile-size missiles flying the length of a football field or more and blowing bark off trees.

The strongest tornadoes recorded in California have been F-2s, of which there have been 24 in the past 51 years. These have wind speeds of 113 to 157 mph and can tear off roofs, demolish mobile homes, overturn boxcars, uproot trees and lift cars.

The most recent of these was in Sunnyvale on May 4, 1998. It cut a swath 100 yards wide and more than half a mile long, causing one injury and $3.8 million in damage. According to one analyst, it was followed about 10 minutes later by a smaller tornado that hit Los Altos.

For more on the Sunnyvale tornado, see
http://tornado.sfsu.edu/geosciences/StormChasing/Cases/Sunnyvale/Sunnyvale.html

For general information about tornadoes and thunderstorms, see http://ggweather.com/tornado.htm

Q  How does the depth of snow relate to the amount of water it would contain if melted down? Ed Gravernhorst - San Jose

A  First, we need to look at the various ways that snow and its water content are measured.

The most common way is simply to measure the depth of snow that has accumulated. To do this properly, several locations are sampled, and then the average depth is determined.

However, because snowpacks have different densities, the depth of snow does not necessarily give us the amount of water that will be left behind when the snow melts.

Several techniques are used to determine water content. The one used by snow survey teams in the mountains of California involves taking ``core samples'' from nearly 300 locations.

A hollow tube is driven into the snowpack until it hits the ground below; this is the depth of the pack. The snow-filled tube is then weighed. By subtracting the weight of the empty tube from the weight of the full tube, the weight of the snow -- and thus the amount of water it contains -- can be calculated.

In the Sierra Nevada, where the snow tends to be very wet and dense and sometimes known not-too-affectionately as ``Sierra cement,'' about 6 to 8 inches of snow usually translates into about an inch of water. In the Rockies, where the snow is drier and more powdery, 10 to 15 inches of snow may equal an inch of water.

Q  On my quarterly visits to the lower desert -- the Coachella Valley -- I notice that the days are relatively calm or maybe have a slight breeze. But at dusk, the wind starts to kick in out of the west and will blow until just before sunup. This is just the opposite of the phenomenon here in the Bay Area, which is most logical -- hot air rising in the valley and drawing in the cool ocean air. Can you explain? Bruce Jenkins - Sunnyvale

A  What you are experiencing is the inland equivalent of our sea breeze/land breeze circulation. It's called a mountain/valley pattern, in which there are upslope breezes during the day and downslope breezes at night.

During the day, air in the valleys heats up more than in the adjacent mountains and it rises, creating a breeze going up the slopes. The converse is true at night as the air cools more in the mountains, becomes heavier and flows down the slopes.

In the case of the nighttime breeze you note in the Coachella Valley, the east-facing slopes of the Santa Rosa Mountains to the west of the valley cool off as the sun sets and a downslope breeze develops and flows downhill into the lowlands.