What I will say is that as I was watching Ghost Story, there's this part
where Krige, as the ghost Eva, promises her lover that she «will show [him] things [he's] never seen» in a supernaturally menacing sort of way that will be echoed exactly by Clive Barker's Cenobites in the novella The Hellbound Heart from 1986.
Not exact matches
But in the data gaps, it consists of satellite data that have been converted to near - surface temperatures,
where the difference between the two is determined by a
kriging interpolation from the edges.
Plenty of scope for making stuff up
Kriging and Mannipulating
where there are hardly any reliable temperature measuring facilities, right?
Where stations exhibit an objective change in regime they are broken into separate records and then the
kriging process estimates the field accordingly.
Now if you want to create data
where there is none,
kriging like to like is probably the best.
Cowtan & Way (2013) evaluated global surface warming across the globe using a statistical method known as «
kriging» and using satellite data to fill in the gaps
where there are no temperature stations.
An alternative approach to the same problem would be regression
kriging where the trends are removed externally resulting in a trend surface (a polynomial function of latitude and elevation), the residuals are
kriged using ordinary
kriging, and the predictions from the trend surface and
kriging of the residuals recombined at locations of interest.
While I've had a good response to what I'm doing, because it's contrarian and I'm not at a point
where I could write a paper and get it through a review process past people who think homogenization,
kriging, infilling is required, I just don't have the time, so I've in general stopped making updates to the code.
BEST uses a form of statistical extrapolation called
kriging to extrapolate to
where they have no thermometers.