Abstract We explore the use of observed polar coronal holes (CHs) to constrain the flux distribution within the polar regions of global solar magnetic field maps in the absence of reliable quality polar field observations. Global magnetic maps, generated by the Air Force Data Assimilative Photospheric flux Transport (ADAPT) model, are modified to enforce field unipolarity thresholds both within and outside observed CH boundaries. The polar modified and unmodified maps are used to drive Wang–Sheeley–Arge (WSA) models of the corona and solar wind (SW). The WSA-predicted CHs are compared with the observations, and SW predictions at the WIND and Ulysses spacecraft are also used to provide context for the new polar modified maps. We find that modifications of the polar flux never worsen and typically improve both the CH and SW predictions. We also confirm the importance of the choice of the domain over which WSA generates the coronal magnetic field solution but find that solutions optimized for one location in the heliosphere can worsen predictions at other locations. Finally, we investigate the importance of low-latitude (i.e., active region) magnetic fields in setting the boundary of polar CHs, determining that they have at least as much impact as the polar fields themselves.