Density dependence and habitat selection affect overwintering abundance of monarch butterflies at regional and site scales in California

Abstract The monarch butterfly ( Danaus plexippus ) is a species of iconic cultural interest. Thanks to annual overwintering monarch counts at hundreds of locations in coastal California, we are able to track fluctuations with high temporal and spatial resolution. Between 1997 and 2024, monarch populations at overwintering sites in the western United States experienced severe dips, at times (2018–2020, 2023–2024) giving the appearance of a population collapse. From 2018 to present, the Pismo State Beach Overwintering Monarch Grove has conducted multiple counts during overwintering and geolocated counts of individual monarch clusters to specific trees within the site. This study determined how annual monarch population variability is influenced by both climate and prior year population density at the state, region, and overwintering‐site scale. Furthermore, through a machine‐learning process, we investigated how overwintering site configuration and structure drive monarch winter space‐use dynamics within the Pismo Beach site. Our approach found monarchs exhibit a preference for specific overwintering sites in California, and that 64% of annual variability of counts across sites can be explained by climate and density dependence, with density dependence explaining 50% of total variability. Within the site we found very little regional climate effect, but individual trees, tree size, distance to boundary, and the amount of shade were all strong indicators of monarch presence. Additionally, only 11 out of 320 trees at the Pismo Beach site accounted for 83.6% of all counts over 6 years, highlighting how monarchs use specific trees and how tree structure may create preferred microclimates for clustering.