Weather Forecasting

Because the many variables that affect weather are constantly changing, meteorologists have used high-speed computers to devise atmospheric models that describe the present state of the atmosphere. These are not physical models that paint a picture of a developing storm; they are, rather, mathematical models consisting of dozens of mathematical equations that describe how atmospheric pressure, winds, and moisture will change with time. Actually, the models do not fully represent the real atmosphere but are approximations formulated to retain the most important aspects of the atmosphere’ s behavior.

?

Computer-generated forecast charts using these models to interpret recorded weather data are known as prognostic charts, or simply progs. At present, there are a variety of models, and hence progs, from which to choose. One model may work best in predicting the position of troughs (elongated areas of low barometric pressure) in the upper atmosphere. Another may forecast the position of surface lows quite well. And some models even forecast the state of the atmosphere 384 hours (16 days) into the future.

?

A good forecaster knows the idiosyncrasies of each model and carefully scrutinizes all the progs. The forecaster then makes a prediction based on the guidance of the computer. This prediction is a personalized practical interpretation of the weather situation given the local geographic features that influence the weather within the specific forecast area. Since the forecaster has access to hundreds of weather maps each day, why is it that forecasts are sometimes wrong?

?

Unfortunately, there are flaws inherent in the computer models that limit the accuracy of weather forecasts. Computer-forecast models idealize the real atmosphere, meaning that each model makes certain assumptions about the atmosphere that may be accurate for some weather situations and be way off for others, so a forecaster who bases a prediction on an inaccurate prog may find a forecast of rain and wind turning out to be a day of clear and colder weather.

?

Another forecasting problem arises because the majority of models are not global in their coverage, and errors are able to creep in along a regional model’s boundaries. A model that predicts the weather for North America may not accurately treat weather systems that move in along its boundary from the Pacific Ocean. Obviously, a good global model of similar sophistication would require an incredible number of computations.

?

Even though many thousands of weather observations are taken worldwide each day, there are still regions where observations are sparse, particularly over the oceans and at higher latitudes. Temperature data recorded by satellites have helped to alleviate this problem, as the newest satellites are able to provide a more accurate profile of temperature for the computer models. Since the computer’s forecast is only as good as the data fed into it and since observing stations may be so far apart that they miss certain weather features, a denser network of observations is needed, especially in remote areas, to ensure better forecasts in the future.

?

Another forecasting problem is that computer models cannot adequately interpret many of the factors that influence surface weather, such as the interactions of water, ice, and local terrain on weather systems. Some models do take large geographic features such as mountain chains and oceans into account, while ignoring smaller features such as hills and lakes. These smaller features can have a marked influence on the local weather. Given the effect of local terrain, as well as the impact of some of the other problems previously mentioned, computer forecasts presently do an inadequate job of predicting local weather conditions, such as surface temperatures, winds, and precipitation.

?

Even with a denser network of observing stations and near-perfect computer models, there are countless small, unpredictable atmospheric fluctuations that fall under the heading of chaos. These small disturbances, as well as small errors (uncertainties) in the data, generally amplify with time as the computer tries to project the weather further and further into the future. After a number of days, these initial imperfections (errors) tend to dominate, and the forecast shows little or no skill in predicting the behavior of the real atmosphere.

?

?

1.

?Because the many variables that affect weather are constantly changing, meteorologists have used high-speed computers to?devise?atmospheric models that describe the present state of the atmosphere. These are not physical models that paint a picture of a developing storm; they are, rather, mathematical models consisting of dozens of mathematical equations that describe how atmospheric pressure, winds, and moisture will change with time. Actually, the models do not fully represent the real atmosphere but are approximations formulated to retain the most important aspects of the atmosphere’ s behavior.