I did a bit of research on the temperature gradients in my nozzles some years ago although not nearly as comprehensively as deckingman. I did learn in the process that some "upgrades" were not as good as hoped.
All results here are from memory as life is too short to go through handwritten and digital notes.
I have attached a drawing of my earliest hotend with a detail showing the aforementioned "upgrade" which consisted of fitting a brass nozzle instead of using a one-part heater tube/nozzle.
The results (again, from memory) is that the one part nozzle varied in temperature across the full gamut of air and filament flows by about 6°C measured on the chamfered part of the outside of the nozzle. By comparison, the replaceable brass nozzle swung by about 16°C on the nozzle surface - this was more noticeable with an external fan than with the nozzles own near field cooling flow (light purple)
I still use this arrangement on my Delta printer as I made a bucketload of nozzles of different sizes. newer printers use a more traditional nozzle design with a male threaded part.
Mike