Wilbur knows nothing of rainfall. The "Special MDR" file format is "Special" because it encodes the whole world in the 2:1 aspect Equirectangular projection suitable for wrapping on a sphere and "MDR" because that's a Wilbur-unique format I made back in the mid-1990s to solve my problem of image size and data precision information attached to a binary raster array.
On the rainfall topic, I wouldn't put much faith in FT's model. Like the temperature model, it was developed purely to give the users a place to start other than 0cm/0 kelvins.
The original ProFatnasy spec didn't include any of this rainfall/temperature/climate nonsense - it was added in sort of at the last minute. When I added it I always assumed that users would use the editing tools more than the generation tools. It seems that my assumption was incorrect.
Perhaps it was the poor quality of the editing tools or something else, but users seem to want to take the randomly-generated information as gospel. Some of them become incensed and flame the software publicly for having "such a terrible waste of time" built into it. Ah well, live and learn, I suppose.