Western monarchs do not always use what is most abundant: Site-level analyses reveal context-dependent hostplant use

Use versus availability analyses provide a foundational framework for understanding animal resource use across complex landscapes. For mobile herbivores like monarch butterflies (Danaus plexippus), hostplant use may vary spatially depending not only on plant availability but also on site-level habitat characteristics. To examine habitat use, we conducted 26,000 surveys across five western U.S. states (2017–2019, 2023–2024). At the site level, we calculated expected use per hostplant species (milkweed; Asclepias spp.) based on the proportion of available stems and compared this to observed use (eggs and larvae) using a binomial generalized linear model with a logit-transformed expected proportion offset. We found roughly half of sites had one milkweed species, the rest had two; three or more was rare. Monarchs did not use milkweed species in proportion to their availability. A significant species × site interaction suggests oviposition decisions were not shaped hostplant species identity alone. Relative use varied among sites where the same milkweed species pairs co-occurred. For example, at two refuges where Asclepias speciosa and A. fascicularis co-occurred, monarch juveniles were observed more often on A. speciosa than expected at a site in Oregon, whereas at a site in Nevada, monarchs used A. fascicularis disproportionately. Half of sites had one milkweed species, the rest had two; three or more was rare. Our findings highlight that monarchs do not use milkweed species consistently across sites, even when the same species co-occur. This suggests that milkweed use is influenced by site-specific context, which may be missing from lab-based preference studies.