Skip Navigation

May-June 2010

Print
Email
ResizeResize Text: Original Large XLarge

Weather Queries

Although ocean water near the West Coast of the United States is cooler than ocean water near the East Coast, why do the coastal western states stay relatively warm in the winter while the southeastern states sometimes get very cold?

Allegiance R. H. Yahweh
St. Louis, Missouri

I took the month of February as an example. I compared February mean sea-surface temperatures off the California coast with that off the East Coast south of 35° latitude. West Coast waters are about 8–10°F cooler than East Coast waters at the same latitude, south of 35°. Yet the February mean temperatures in the Southeast coastal states are a few degrees lower than in California. And it is certainly true that the Southeast is subject to cold-air outbreaks from Canada, whereas California is seldom affected.

There are two reasons for this: the prevailing winds and the topography. Before I elaborate, permit a short digression on surface warming and cooling. You can put a lot of solar energy into the ocean, but its temperature doesn't change much because solar energy is absorbed through a depth of many meters, and breaking waves mix warm surface water with cooler water below. Over land, for the same solar energy input at the surface, the ground warms much more quickly than a large body of water because soil and rock are very inefficient at conducting energy downward. All of the incoming solar energy is absorbed very near the surface.

The ocean surface doesn't cool as quickly at night as the ground does. The entire Earth's surface, land and sea, radiates energy to the sky 24 hours a day. At night, when there is no solar input, the surface usually cools. The cooling is confined to the immediate surface on land. Not so with water. At 4°C (39°F), water reaches its maximum density. As the temperature rises from there, density decreases. Thus, at temperatures above 4°C, surface cooling creates denser water on top of less dense water. The denser surface water sinks, so the energy loss at the surface quickly propagates below the surface, unlike the situation on land. This explains why the daily (and seasonal) excursions of temperature are much greater over land than over the ocean.

The prevailing wind is the first reason why the Southeast states are colder in winter than California. From the Northwest Territories to Manitoba, the snow-covered Canadian plains are a source region for cold air masses in winter. Atmospheric winds in winter normally blow from the west or northwest. In the latter case, the wind can steer cold Canadian air all the way to the Gulf and Atlantic coastlines. By contrast, prevailing winds bring air from the Pacific Ocean into California during most of the winter. Air with a long trajectory over water tends to acquire a temperature close to that of the water. That's why the temperature swings along the West Coast are much smaller than along the East Coast in winter. The nearly constant-temperature ocean supplies air to the West Coast most of the time, whereas the Canadian land mass supplies cold air to the southeast coast periodically, having the effect of lowering the mean winter temperature.

The second reason for less variability of West Coast temperatures in winter are the mountain ranges of western North America. Cold air masses are dense, usually a few thousand feet deep, rarely 10,000 feet deep. Like water, these cold air masses seek lower ground and tend to pool in topographical basins. Thus, when they move southeast from Canada into the United States, they seldom cross the Continental Divide. The mountain barriers shunt them across the northern and central Great Plains and prevent them from invading California or the Southwest United States.

What about the northern coastal states? In winter, the water off the mid-Atlantic and New England coasts is colder than off the Oregon–Washington coast. But the sea surface temperature differences don't explain the much larger differences (up to 20°F) in mean winter temperatures at both ends of the country north of 35° latitude. It's much colder in the East. Again, it's a matter of the prevailing wind bringing comparatively mild Pacific air onto the Oregon–Washington coast but dragging cold Canadian air off the mid-Atlantic and New England coastlines most of the time.

For quite some time, it has seemed to me that we have had more wind—not just separate events but more day-to-day wind. I live in northwest Kentucky. I've commented on this to others, and they agree with me. But this is not just local. I am a Ham Radio operator and find that others outside my area have the same impression. Could this be an indication of La Niña or climate change?

Harold Gillim
Lewisport, Kentucky

A real upward trend in surface wind speed would be related to increasing average pressure gradients, caused by greater differences in pressure between the centers of highs and lows on the weather map. In the mid-latitude United States, this occurs every winter as the jet stream strengthens, the storm track invigorates, and the alternation between highs and lows becomes more frequent. The opposite occurs each summer. High winds in summer are usually short-lived, caused by thunderstorms and, once in awhile, a land-falling hurricane, but this wouldn't affect the annual average wind speed as much as strong low-pressure systems that spread windy weather over thousands of miles as they cross the country.

I'm not aware of any study that links increased windiness to climate change or La Niña (lower than normal sea-surface temperatures persisting for many months in the tropical Pacific Ocean), though some climate models predict increased storminess if global warming continues.

I found a recent paper, “Wind Speed Trends over the Contiguous United States,” by S. C. Pryor and eight coauthors (Journal of Geophysical Research, Vol. 114, D14105, 2009) that presents evidence for a decrease in annual average wind speeds over much of the lower 48 states, most pronounced in the eastern United States, from the early 1970s to the early 2000s. If real, such a trend has major implications for wind power generation, because small changes in wind speed at the turbine hub height cause large changes in power generation.

The authors examined two sources of surface wind observations, both archived by the National Climate Data Center, for 00:00 GMT (evening) and 12:00 GMT (morning). The observations from some sites appear in both databases. In order to qualify for the study, a station had to report reliably throughout the year and for all years included in the study: 1973 through 2000 for the first database, and 1973 to 2005 for the second. All observations were adjusted to a common height above the surface, 10 meters, which is the international standard. The adjustment is necessary because wind speed increases from zero right at the ground and is almost always greater at 10 meters than at any lower height. All observations were subject to quality control.

Many aspects of the observation record were beyond the control of the investigators. (1) For the purpose of climatological observations, the NWS recommends that a wind sensor “must be at least 15 feet above the height of any obstruction within 500 feet.” It defines an obstruction as any object subtending a horizontal angle of more than 10° as measured from the sensor. Obstructions block the wind, and that is why wind measurements should be taken in open areas. Information about obstructions and their distance from the anemometer is seldom available. (2) In a span of 30 years, the measuring site often changes; sites used in this study were not allowed to move more than 5 km. (3) Even for fixed sites, trees grow where none were before, and buildings are constructed, thus changing the exposure to the wind. (4) The wind sensor itself changes with advancing technology. It is important to compare new and old sensors side-by-side so as to calibrate one with respect to the other, but this does not always occur.

The authors focused their attention on the 50th and 90th percentiles of the wind speed distribution at each station. That is, they arranged all the wind speed reports for the year at a given station in decreasing order, and picked out the value in the middle of the distribution (the 50th percentile, also called the median) and the value just 10 percent from the top of the list (10 percent of all wind speeds were higher than the 90th percentile value). A plot of the 50th and 90th percentile values year by year indicates a downward trend in wind speed at most sites in the lower 48 states but especially in the eastern half. Wisely, I think, the authors do not speculate about the cause of this trend.

The authors also examined 10-meter wind data from four different reanalyses, spanning anywhere from 27 to 58 years. In a reanalysis project, a modern data assimilation system is used to process historical weather observations and create initial conditions for a modern computer prediction model. In reanalyses, wind speeds at 10 meters depend upon (1) the prediction model, which carries the atmospheric state forward in time, and (2) multiple sources of observations in the general vicinity of the wind estimate, not just surface wind observations. It is significant that none of the four reanalyses showed a clear downward trend in wind speed over the past 30 years or so. One showed an upward trend, but the others showed no clear trend at all.

One concludes from this study that the discernment of long-term trends in surface winds is fraught with difficulties, not the least of which are nonuniformity in the wind sensors, their height above ground, and their location with respect to surrounding obstacles. It is easier to establish long-term trends in temperature than in wind speed. The trend toward decreasing wind speed exhibited by the observations may be real, but it is not corroborated by the other data sources.

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 St. Suite 800, Philadelphia, PA 19106.

In this Issue

On this Topic

© 2010 Taylor & Francis Group · 325 Chestnut Street, Suite 800, Philadelphia, PA · 19106 · heldref@taylorandfrancis.com