I have a data series (weather radar) that I am trying to plot with basemap in python. Plotting the raw data is easy; I use meshgrid with the lat/lon points, feed these 2-d data arrays into the first two parameters of contourf and put the data matrix (same dimension as the lat/lon) into the third parameter.
My problem; I then want to do some manipulation with the data. I flatten the lat/lon/data arrays, take null values out of all three arrays, and end up feeding the data back into three vectors- lat, lon, and data. While the rows line up (eg, each index of lat/lon line up with that index in the data vector), they are in random order.
I want to feed this back into contourf, but I'm not sure how to unpack the data back into the initial arrays. Is there any way to plot pure lat/lon/data pairs with contourf? If I have the initial 2-d lat/lon meshgrid arrays, what would be the easiest way (using Numpy) to populate a new 2-d array with my data based on the three vectors my algorithm spits out?
eg:
lat = [44.1, 44.3, 44.2]
lon = [-67.2, -67.4, -67.3]
data = [3, 7, 2]
Original lat/lon meshgrid:
lat = [
[44.1, 44.1, 44.1],
[44.2,44.2,44.2],
[44.3,44.3,44.3]
]
lon = [
[-67.2,-67.3,-67.4],
[-67.2,-67.3,-67.4],
[-67.2,-67.3,-67.4]
]
Data that I need to feed into contourf:
data = [
[3,nan,nan],
[nan,2,nan],
[nan,nan,7]
]
Any ideas? This has to be easier than I'm thinking.
