Skip Navigation

July-August 2013

ResizeResize Text: Original Large XLarge

Weather Queries

Q: How accurate are electronic weather instruments? I have an electronic barometer right next to a mercury barometer. 26.80 inches on the mercury barometer has been a good dividing line between good weather and unsettled weather. Our elevation here is 4,960 feet. The electric barometer is set for 4,960 feet, which yields approximately 30.00 inches (range from 29.80 to 30.10 inches) when the mercury barometer reads 26.80 inches. The correlation between the two readings is weakest when the pressure is rising or falling rapidly.

The same weather station also remotely measures the temperature and dewpoint from the middle of my fields. Sometimes the dewpoint is 5°F less than the concurrent temperature, and yet there is dew everywhere. Even in the middle of a soaking, slow drizzle, the dewpoint will be 3°F less than the temperature.

Will Hutchins

Delta, Colorado


A: Your mercury barometer measures actual barometric pressure: the pressure exerted by the atmosphere at the elevation of your barometer. Your electronic barometer is trying to give you the equivalent pressure at sea level. Meteorologists call this the “pressure reduced to sea level.” There are two common ways in which the actual “station pressure,” the pressure measured by your mercury barometer, can be reduced to sea level.

One way uses what is called the standard atmosphere. In a standard atmosphere, the temperature varies smoothly with altitude, decreasing from sea level upward at a fixed rate, and the atmosphere is considered to be completely dry (no water vapor). If your electronic barometer used a standard atmosphere to reduce the pressure to sea level, then it should deliver the same answer every time your mercury barometer reads 26.80 inches of mercury (the height of the column of mercury). Your electronic barometer must use some other method.

A second method I'm familiar with assumes a certain temperature profile between the altitude of your barometer and sea level. If the hypothetical column of air between you and sea level is cold (say, in winter), the difference between the reading of your mercury barometer and an inferred sea level value will be larger than if the atmosphere in this same layer is warmer (say, in summer). If your electronic barometer is outside, it may use the temperature measurement to decide how cold the hypothetical column of air between your elevation and sea level is. If your electronic barometer is in your house, where the temperature varies little, then I cannot explain why it gives such different readings on occasions when your mercury barometer reads the same.

One close-by, reliable estimate of sea level pressure (an estimate that assumes a standard atmosphere) can be obtained from the National Weather Service office in Grand Junction, Colorado. Grand Junction's elevation is close to yours, and, on quiet weather days, the inferred sea-level atmospheric pressure there, reported in inches, should be a good estimate. By the way, pilots call this pressure the “altimeter setting.” They use it to set their altimeter so that, when they land, the altimeter will give the field elevation of the airport.

You could try an experiment in which you set the elevation on your electronic barometer to zero instead of 4,960 feet. If this instrument is outside and separable from the other electronic instruments, bring it inside so that it is at the same temperature as your mercury barometer. The electronic barometer should then be measuring the true atmospheric pressure at your elevation, and the two pressure measurements should agree more closely than before.

Dewpoint instruments are not normally very accurate (perhaps to within a few degrees), unless you spend many hundreds of dollars on them. A dewpoint sensor that relies on a chilled mirror should be accurate to within a degree. In such an instrument, a mirror is slowly cooled until vapor begins to condense on it. The temperature of the mirror when this first occurs is the dewpoint temperature. If a slow soaking drizzle or rain has been in progress for at least an hour and the wind is light, the temperature and dewpoint should be the same. That's also true in a thick fog.

The temperature and dewpoint may not be the same if there is dew on the ground. First, the temperature on a grassy surface may be a few degrees lower than the temperature as measured in a weather shelter on calm, clear mornings. In other words the grass surface, exposed to the sky, may be several degrees colder than your thermometer, which is a few feet off the ground and not exposed to the sky. Another point to keep in mind is that the temperature may rise five degrees or more after dawn before the dew completely evaporates from a grassy surface. In this case also, the temperature and dewpoint will not be the same.


Q: In summer 2012, a dome of hot air formed over the central United States. This air mass, characterized by high pressure aloft, was presumably sinking toward the surface and warming and compressing. Normally, a hot air mass rises by convection and expands. Its temperature decreases dry adiabatically until condensation occurs, then decreases moist adiabatically until the rising air is no longer warmer than its surroundings. Can you explain this conundrum?

Thomas R. Kuhns, M.D.

New York, New York


A: The key to resolving this conundrum is recognizing the importance of the earth's rotation as a constraint on the motion of the earth's atmosphere, and how this constraint operates. With respect to the fixed stars, the earth rotates about its axis once per sidereal day, that is, once every 23 hours 56 minutes. (For purposes of this argument, the difference between the sidereal day and the conventional 24-hour day is negligible.) If we are talking about weather events that evolve over several hours or more, we must account for the earth's rotation by means of the Coriolis force. This force acts at right angles to the wind direction, tending to accelerate the air to the right in the Northern Hemisphere and to the left south of the equator. Over the United States, that means, for example, that air flowing toward the east will experience an acceleration toward the south. In fact, the Coriolis force is responsible for the tendency of the wind to blow more or less parallel to the height contours on a constant pressure map, such as in Figure 1.

Figure 1. 500-mb chart for 0000 Greenwich Mean Time (GMT) (1800 CDT), August 1, 2012. A large anticyclone covers the Southern Great Plains. See text for plotting details.
Figure 1. 500-mb chart for 0000 Greenwich Mean Time (GMT) (1800 CDT), August 1, 2012. A large anticyclone covers the Southern Great Plains. See text for plotting details.

Figure 1, the 500-mb chart for late afternoon on August 1, 2012, illustrates conditions typical during last summer's heat wave (“mb” means millibars, a measure of pressure). Note the closed anticyclone over Texas, with weak clockwise circulation around its periphery. The solid black contours, drawn at 60-meter intervals, give the height of the 500-mb pressure surface above sea level. These heights are plotted in tens of meters to the upper right of the station locations. The wind staffs point into the wind. Long barbs on the staff indicate 10 knots each; short barbs count for five knots. Temperatures (°C) are plotted in red. Near the center of the anticyclone, they are no lower than −5°C, indicating very warm air at this level. Dewpoints (°C) are plotted in green. For the most part, they are much lower than the temperature, indicating dry air at 500 mb. Note how the winds follow the height contours around the anticyclone. The Coriolis force is at work here.

Meteorologists often treat small volumes of air, say, 100 meters across, as if they were isolated from their surroundings, that is, no mixing with surrounding air. The word adiabatic implies as much. This approximation is adequate for the present discussion. If a volume of air rises, it cools at the dry adiabatic lapse rate, which is about 1°C for each 100 meters of lift. As the temperature drops, the humidity rises. If lifting is sufficient, the humidity reaches 100 percent and a cloud forms. Condensation releases energy into the rising volume of air so that the cooling rate upon further lifting is no longer 1°C/100 meters but instead a lesser rate, called the moist adiabatic lapse rate.

Now back to the specific question at hand: If we have volume of air that is warmer than its surroundings, it will tend to rise because warmer air is less dense than cooler air at the same pressure. However, the size of this volume of warm air is critical. Suppose it is only 100 meters or so across, for example, a recently plowed field with dark soil surrounded by fields of corn. The radiant energy in sunlight reaching the plowed field acts primarily to heat the ground, whereas it goes primarily toward transpiration in the corn crop. Thus the plowed field becomes warmer than its surroundings, and so does the air overlying the plowed field. A thermal rises over the plowed field, cooling at the dry adiabatic rate. If a cumulus cloud forms within the thermal, it will begin cooling at the moist adiabatic rate. As long as it remains warmer than its surroundings, it will continue rising. As the thermal rises, air from over the adjacent cooler fields moves in laterally to replace the air in the thermal. The lifetime of a thermal is from a few minutes (a small thermal that buoys a sailplane) to a few tens of minutes (a large thermal that builds a towering cumulus cloud). Either way, the process is essentially local and short-lived. The effect of the earth's rotation is negligible because the Coriolis force doesn't have time to alter the wind field.

But what of the situation where the hot-air mass covers several states, as in summer 2012? Figure 2 shows surface conditions for the same time as Figure 1. Temperatures (red) and dewpoints (green) are plotted in degrees Fahrenheit (°F). Late afternoon temperatures are above 100°F under the anticyclone. Sea-level pressure contours are drawn at 4-mb increments. There are very few contours, indicating that the pressure field is nearly flat. If the sea-level pressure is relatively uniform where the hottest weather prevails, it is easy to explain why there is an anticyclone aloft. Consider a column of air bounded by two pressure surfaces, say 1008 and 500 mb. (1008 mb is the average sea-level pressure under the hot air mass.) The vertical distance between these two pressure surfaces is determined by the average temperature of the air column (with a slight contribution by atmospheric water vapor). Warmer columns stand higher than cooler ones. Thus, if the sea-level pressure is uniform, near 1008 mb, the height of the 500-mb pressure surface must be a maximum over the hottest part of the air mass. Figure 1 bears this out.

Figure 2. Surface weather map for 0000 GMT (1800 CDT), August 1, 2012. Note widespread temperatures above 100°F in Texas, Oklahoma, and Kansas. Gray contours of surface pressure are drawn at 4-mb increments. The pressure field is flat and nondescript. See text for plotting information.
Figure 2. Surface weather map for 0000 GMT (1800 CDT), August 1, 2012. Note widespread temperatures above 100°F in Texas, Oklahoma, and Kansas. Gray contours of surface pressure are drawn at 4-mb increments. The pressure field is flat and nondescript. See text for plotting information.

The extensive hot-air mass is warmer than its surroundings (outside the Southern Great Plains). Why doesn't it rise like some giant thermal? The buoyancy of a small thermal relied on a reasonably sharp temperature difference between itself and its immediate surroundings. Any horizontal temperature gradients across the hot-air mass, and even along its periphery, are 100 times smaller and not enough to impart buoyancy. Even if the entire air mass began to ascend slowly (by what means I cannot conceive), the cooler air around its periphery would take so long to replace it that the Coriolis force would have time to act. Inflowing air would be deflected to the right, forming a weak cyclonic circulation around the edges, thus inhibiting further ascent toward the middle.

Thunderstorms are few and far between in the hot-air mass. Why don't more of them form? Refer again to Figure 2. The station circles are filled in according to the fraction of sky covered by clouds. Note that most station circles in Texas, under the anticyclone, are open, indicating clear sky, even near the Gulf Coast where dewpoints are high. Hot air and low-level moisture are supposed to be good for thunderstorms. Correct? Yes, but conditions in mid-troposphere must favor deep convection before a thunderstorm can form, and at least two mid-tropospheric features in the temperature profile work against this: (1) Regions south of the jet stream (where the upper-air anticyclone persists) tend to have slowly subsiding air aloft. The sinking air warms at about 1°C for each 100 meters of descent, and a stable layer of air, sometimes an inversion, forms at the base of the sinking air layer. (2) Moist air flowing into Texas from the Gulf, on light winds in the lowest 1,000 meters or so, is cooler than the air above. This, too, results in a stable layer at the top of the moist air. Stable layers inhibit convection trying to punch through from below, and thus thunderstorms are few and far between.

This discussion begs the question of what led to the formation of this warm air mass in the first place, but it does help explain why such warm air masses can be difficult to dislodge, once they form. This is particularly true in mid-summer over the continental United States, when weather systems are weak and the storm track and jet stream are displaced well to the north. Under such conditions, there are no weather systems to dislodge the warm air, and so the sun continues to beat down on the earth, mostly unshaded by clouds, prolonging the heat.

I thank John M. Brown of NOAA's Earth System Research Laboratory, who helped me answer this question.

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, or by mail in care of Weatherwise, Taylor & Francis, 325 Chestnut St., Suite 800, Philadelphia, PA 19106.

In this Issue

On this Topic

© 2010 Taylor & Francis Group · 325 Chestnut Street, Suite 800, Philadelphia, PA · 19106 ·