How to approach this problem?

Hi all,

I'm trying to work out the best way to approach this problem that I have. I have a set of lat and longs, along with some data on pollution levels. I can plot the pollution on a plot ok with nice colours based on the level. I can also interpolate to fill in the bits in the middle. But what I really want is some kind of 'heatmap' that will predict the values of pollution outside the data over the whole area.

I'm not sure on the best way to approach this. Is there some kind of matlab function/toolbox that can do this easily? I'm not too fussy on the model used, probably just 1/r^3, as it is just to get an idea.

Answer 1

[Jan]

You can look for inpaintnan in the FileExchange. E.g. John D'Errico's tools might be helpful.