This one isn't on Wikipedia's list (at time of writing) unlike "Play it, Sam. Play 'As Time Goes By'" or "Why don't you come up sometime and see me?"
For context that above is processed and reduced data, for those that care.
The raw data (which was once online (and may still be, having shifted)) is full 256 channel counts + height + distance to ground + raw GPS packets + post processed GPS position + etc... straight out of the airframes (and additional "meta" data - the concurrent calibration runs, stripping weights, raw GPS base station packets, etc).
The data linked above is three values per grid cell post process - the full spectrums are normalised, then U-K-Th (Uranium, Potassium, Thorium) decay product peaks are extracted to reconstruct estimated U-K-Th values for a grid cell (and a pseudo colour image created from those three values)
Great for background environmental radiation mapping - useless for spotting out of band "stray" radiation events not of a natural origin.
For that you want to subtract a NASVD kernel from the data and look to the leftover peaks between the big three daughter decay peaks ( and similarly for removing radon smear, etc ).
The entirety of Fiji,and most of Mali should ideally also be online / available - but can be a tricky find (ditto many other places about the globe).
Before and after Pokhran-II India, (1998) should also be an interesting study that might be tricky to hoard the data for.
I apologize if this is answered in the PDF, but I couldn't find an explanation there - any idea why fairly large chunks were omitted from the map? Some of them seem like population centers, others look like nature preserves.
Crikey