Though it seems excessive, it is exactly what you are likely thinking: Pi x Horz.Radius^2
This appears to increase the overturning moment for most tank sizes, and can produce strange results depending upon tank diameter to height ratios. The key thing to note, I believe, is that there is a reduction taken in equation #1 of section 5.11.2 to somewhat offset this additional moment.
One point of note also: the dead load resisting moment of the shell and roof supported by the shell should be the corroded weight (if a C.A. is designed into the plate).
