There is a considerable body of work around spatial interpolation, which, putting all the surrounding tech to one side, seems to me to be the critical methodolo

There is a considerable body of work around spatial interpolation, which, putting all the surrounding tech to one side, seems to me to be the critical methodological issue in this work. Getting valid results out of spatial interpolation, i.e. a result that predicts reasonably well at areas between sample points, is as much of an art as a science, and requires familiarity with the various techniques available, and even reliance on intuition. Moreover, it is generally accepted that no technique will ever produce a perfect result. However, one way of testing a chosen technique is to sample at points between initial sample points and compare to your interpolation output. I don't see that this was done. I would be concerned about relying too heavily on a result without proper analysis of its validity, and if I were a customer I would be asking questions. http://pro.arcgis.com/en/pro-app/help/analysis/geostatistical-analyst/deterministic-methods-for-spatial-interpolation.htm http://pro.arcgis.com/en/pro-app/help/analysis/geostatistica... https://www.gisresources.com/types-interpolation-methods_3/ https://www.gisresources.com/types-interpolation-methods_3/ This is the very latest in a series of articles I've seen involving MapBox that appear to claim novelty in some way but IMHO don't show anything genuinely new. Edit: looking at this some more I wonder whether you can even see the poor result of the method visually directly in the output - each sample point has a heat island around it, whereas I suspect that heat does not drop off between sample points, that is unless a sample point is mounted on something that might cause the spike/drop-off, like a server. Also I am puzzled by this in particular: 'Normally, when working with mapping libraries a “heatmap” is a gradient displaying “hot spots” based on the density of points plotted on the map. In our case, we are actually trying to create a HEATmap, where we’re showing a gradient based on varying values of those individual points on the maps. We were actually a bit surprised there isn’t a way to do this already, since it would seem like a fairly common use case.' How is this not the exact same thing, already available in R? http://rspatial.org/analysis/rst/4-interpolation.html http://rspatial.org/analysis/rst/4-interpolation.html