Decomposition of the total per-altitude uncertainty into its three physical components, for the 20 profiles in Figure 6: the thermal noise floor σ_ne,thermal (dashed gray, scalar; Equation (5)), the baseline-slope contribution σ_ne,slope(z) (orange; Equation (4)), the FIDES spherical-symmetry contribution ﹩{\sigma }_{{\rm{ne}},{\rm{FIDES}}}(z)=\sqrt{{\rm{\max }}(0,\,{\sigma }_{{\rm{total}}}^{2}(z)-{\sigma }_{{\rm{ne}},{\rm{thermal}}}^{2}-{\sigma }_{{\rm{ne}},{\rm{slope}}}^{2}(z))}﹩ (purple), and the resulting σ_total(z) (black). This figure quantifies how the FIDES-derived model uncertainty compares with the measurement-only uncertainty across SZA: near the high SZA dayside M2 peak the FIDES contribution dominates, whereas on the nightside the slope and thermal terms dominate.