[S II]/Hα as a function of [N II]/Hα for pixels across the map (left). Considering the line emissivity models of Madsen et al. (2006) (navy lines), we see how line ratios are expected to change as a function of S⁺/S (dashed). We color-code this 2D histogram distribution by the median extinction corrected Hα intensity (center) and the median extinction corrected [S III]/[S II] line ratio (right). We see that regions predicting lower S⁺ abundances are well matched to regions with high S⁺⁺ abundances, as traced by [S III]/[S II]. The Flame Nebula stands out, with high [N II]/Hα but low [S II]/Hα line ratios, and high Hα intensities. Integrated Milky Way H II regions (IC 434: triangle, IC 432: square, Flame Nebula: circle, WHAM: orange points) exhibit low S⁺/S values as much of the sulfur is ionized to S⁺⁺. Extragalactic H II regions at matched Hα luminosity (PHANGS: purple points) illustrate the challenges of separating H II regions from the DIG on ~100 pc scales, biasing diagnostic line ratios.