[Purdue-pm] followup to last night's meeting

Mark Senn mark at senn.us
Thu Dec 15 13:49:43 PST 2016

This is	a followup to last night's meeting regarding
Jupyter	worksheets (similar to Mathematica notebooks),
Mathematica, and Mathematica's built-in data---in particular,
WeatherData and WeatherForecastData.


Even if I could get Jupyter to run locally, instead of over the web I'd
still want to use Mathematica.  The breadth and depth of the Mathematica
language plus all the curated data that's available conveniently for
Mathematica makes it too nice and too productive not to use for
science-related programming.


Regarding working with weather data, you may be interested in:

     typing "lafayette indiana weather" into wolframalpha.com

An explanation of how to use numerical weather data in Mathematica:




And a simple example:

     # Type a semicolon followed by Enter after each line
     # if you're writing a program---type Shift-Enter if
     # you are entering the commands one-by-one and want
     # to see results for each line immediately.

     # Get mean monthly temperatures for 1600	through	2016.
     wd = WeatherData["KLAF", "Temperature", {{1600}, {2016}, "Month"}]

     # 528 data points from 1973-01-01 to 2016-12-01 are returned.

     # Plot the time series.
     p = DateListPlot[wd]

     # Export the time series plot to a "z.png" file.
     Export["z.png", p]

(To see if the globe is warming, one could loop over weather stations,
get temperature data, and test (perhaps do a best fit of the data
to a linear equation and plot a	dot at each weather station that
varies from blue to red depending on the slope of the best fit).
Mathematica also has routines for contour lines and confidence
ellipses built-in.)


More information about the Purdue-pm mailing list