The standard form of an upper air station plot consists of temperature, humidity, wind speed
and direction, and geopotential height at a specific pressure : 1/

Upper air analysis on constant pressure surfaces is most meteorologically useful, and
height and temperature contours are typically included :

On the above 500 mb chart, geopotential height (black - decameters) and temperature
(red - °C) contours are subjectively analyzed.
The station plots were made using Digital
Atmosphere. A closed Low was over the Midwest, associated with snows of more than a foot over much of the region, and nearly 2
feet near the SW Lake Michigan shore.
Balloon Soundings
Similar with surface observations, upper air observations are (obviously) required for such a
chart. This is primarily accomplished with balloon
soundings at
specified locations. A radiosonde (radio-transmitted sonde) is released. As it rises and drifts
with the wind, the plotted variables are measured. Readings are obtained for mandatory and
significant pressure levels. Mandatory levels are 1000, 925, 850, 700, 500, 400, 300, 250, 200,
150, and 100 (and 50 & 10 ?) mb. Significant levels are those in between at which temperature
lapse rate or wind change significantly (more than a specific threshold - the precise values
of which I am unaware of). Thus, the temperature plot is quite accurate and smoothly varies.
A sounding is considered successful if 400 mb is reached. When sufficient soundings are made
over a region nearly simultaneously (nearly can be an hour or more different), an upper
air chart is made for each mandatory pressure, all readings plotted at the point of release
(though the balloon drifts many miles - a significant source of error for upper atmospheric
charts, actually). This radiosonde data is (of course) the source for atmospheric soundings
previously discussed.
Direction of Gradient Wind
You may recall in a previous discussion (not yet here) that the gradient wind at a
constant height such as sea level blows parallel with isobars, with low pressure to the
left/right of the wind vector in the Northern/Southern Hemisphere. This statement can be equally
made for height contours at a constant pressure as depicted on upper air charts. This can
be illustrated mathematically, rewriting the gradient wind equation previously shown :
replacing pressure gradient on a constant height surface with height gradient on a constant pressure surface :
g [
¶z/¶r]P = s2/r + f sThus, the basic interpretation of the equation does not change - horizontal gradient winds blow parallel with both isobars and height contours, and are stronger where isobars or contours are most closely-spaced. Please notice that geopotential height contours as typically plotted slightly differ from this, but a similar transformation for geopotential height can also be made. You may notice that winds on the 500 mb chart above generally follow this rule, though actual winds vary quite significantly from gradient winds for some places such as Albuquerque, NM.

Text and embedded images are copyright of Joseph Bartlo, though may be used with proper crediting.