14.3. Smoothing Data Fields

There are times you might want to smooth the raw values from your gridded output to generate contours that better reflect what you would produce through a hand analysis. There are a couple of functions that you can use to accomplish this goal in MetPy, but we’ll look at one here.

One method of smoothing involves using points surrounding a given grid point and averaging them in a thoughtful way to produce a grid of values that vary in a more gradual manner. This n-point smoothing available in MetPy can use either 5 or 9 points and can be run successively one or more times. The more times smoothing is done, the less nuance will appear in the contours, eventually not well representing reality. There are two different options to smooth the look of plotted variables, smooth_field and smooth_contour.