A different technique is available for dealing with changing weather conditions and changing winds. It assumes that the weather conditions are closely tied to the surface pressure and frontal patterns, and that these patterns are changing only slowly. If you have reason to doubt either of these assumptions, don't try this method.
Here's how it works: suppose you're forecasting for Cincinnati. According to the numerical model, a high pressure center will be just southeast of Cincinnati tomorrow. What will the weather be? You could try figuring out an air trajectory, but you'd probably find that the wind was first from the north, then calm, then from the south, and all the while heating and subsidence were warming the air mass. A better, and simpler, method would be to just look on the weather map today and locate that high pressure system. Is it moving slowly? Is it roughly the same intensity? If so, check the current (or high) temperatures just northwest of the high pressure center today. That's your forecast for Cincinnati tomorrow.
FORECASTING TECHNIQUE: Weather pattern extrapolation
On the left is an analysis from Sept. 30, 1996. On the right is the forecast for the next morning. Suppose you were forecasting for Concord, New Hampshire. The forecast calls for New Hampshire to be directly underneath the center of a large area of high pressure. Presumably, with the weak pressure gradients and high pressure, there will be clear skies and light winds. What will the minimum temperature be?
Look at the analysis map. Western Pennsylvania was under the same high pressure center this morning. Using the above forecasting technique, you would look at the minimum temperatures this morning in western Pennsylvania to estimate the minimum temperatures tomorrow in New Hampshire. Perhaps you would add a few degrees for the decrease in elevation from western Pennsylvania to New Hampshire, but you get the idea.