this might be happening because I generate the host presence in ecoregion matrix solely from IUCN redlist data and this matrix is used to estimate host richness per ecoregion. it is possible that there is spatial data for host-parasite associations in gmpd that are outside of the host's IUCN redlist range. The gmpd data are used to measure actual parasite richness per ecoregion but host richness per ecoregion is given by the iucn data. So it is possible that a host-presence association from GMPD is outside of the IUCN redlist data and in an ecoregion that IUCN data would not have recorded. In this scenario, we could end up with parasite presences in ecoregions where the host is not present when looking at the actual gmpd data.
this might be happening because I generate the host presence in ecoregion matrix solely from IUCN redlist data and this matrix is used to estimate host richness per ecoregion. it is possible that there is spatial data for host-parasite associations in gmpd that are outside of the host's IUCN redlist range. The gmpd data are used to measure actual parasite richness per ecoregion but host richness per ecoregion is given by the iucn data. So it is possible that a host-presence association from GMPD is outside of the IUCN redlist data and in an ecoregion that IUCN data would not have recorded. In this scenario, we could end up with parasite presences in ecoregions where the host is not present when looking at the actual gmpd data.