Whilst techniques such as correlation are useful, displaying two or more variables graphically can provide an insight into data which might not be obvious on a graph or from a table of numbers. Recently, I have had to look at some atmospheric data obtained from GFS (see note below), in addition to latitude and longitude, some elements of this have a vertical component defined by the pressure level (e.g. 900 mb). The example below shows relative humidity around noon for a randomly selected day in June 2015.
For lack of a better name, this graphic device has been called a dipstick. In terms of kml, each dipstick is a series of linestrings, the colour of which is determined by the value of the data value. In this example, the colour scheme has been taken from ColorBrewer. A vertical exaggeration of 5.0 has been applied, some experimentation is needed to find a value which is appropriated for the data and the geographic scope.
I'm far from being an expert on weather data, but my understanding is that clouds form when the relative humidity is high, this example suggests clear skies over the south east of England with low to medium cloud to the north.
GFS is the Global Forecast System, extensive datasets from which are made available by NOAA. I would like to express my appreciation as these are a valuable learning resource.
I believe this representation to be original, any plagiarism is unintentional.