Broken symmetry density functional theory has been used to calculate g-tensor, Mn-55, N-14, and O-17 hyperfine couplings for active site models of superoxidized Mn-III/Mn-IV manganese catalase both in its native and azide-inhibited form. While a good agreement is found between the calculated and experimental g-tensor and Mn-55 hyperfine couplings for all models, the active site geometry and Mn ion oxidation state can only be readily distinguished based on a comparison of the calculated and experimental N-14 azide and O-17 HFCs. This comparison shows that only models containing a Jahn-Teller distorted 5-coordinate (Mn-III)(2) site and a 6-coordinate (Mn-IV)(1) site can satisfactorily reproduce the experimental N-14 and O-17 hyperfine couplings.