Will climate-mediated phenological shifts affect population viability? A test with monarch butterflies on Department of Defense lands

Diane Debinski, Department of Ecology, Montana State University, Bozeman, MT, Elizabeth Crone, Department of Biology, Tufts University, Medford, MA, Norah Warchola, Biology, Tufts University, Medford, MA and Sonia Altizer, Center for the Ecology of Infectious Diseases, University of Georgia, Athens, GA

Background/Question/Methods
Ecologists now widely recognize that the timing of life cycle events is shifting in response to directional environmental change. However, it is not clear whether these shifts generally benefit or reduce population viability. Here, we combined historical data and experimental manipulations to determine how non-stationary environments and shifting interactions affect population viability of monarch butterflies. We investigated (1) the extent of recent changes in monarch abundance and phenology and (2) the importance of these changes for population viability. Both were conducted at Camp Dodge in Johnston, Iowa, in the core of the monarch’s breeding range. Butterfly surveys were conducted during the summers of 2003-2019 using a Pollard‐transect‐like survey. We analyzed monarch abundance and phenology to determine if there were trends over time. The experimental study, conducted in 2020, simulated how shifting monarch spring arrival times might affect development and survival. We placed three cohorts of 300-400 first instar larvae on common milkweed (Asclepias syriaca) plants timed to simulate early monarch arrival, average time of spring monarch arrival, and late arrival. Caterpillars were placed on caged plants to allow for semi-natural levels of insect predation and parasitism. We monitored the survival and developmental stage of monarchs until they eclosed as adults.
Results/Conclusions
Transect survey data of monarch adults showed evidence of a shift in abundance patterns over time from 2003-2019, particularly with lower abundances at the end of the summer season in later years. However, there was no advance in breeding phenology. Arrival times of adult monarch butterflies to Camp Dodge, IA for 17 years were remarkably consistent. However, our analysis did show that monarchs are staying a little later into fall. Survival of monarch larvae, as measured by percent survival to eclosion, was highest when timing of release matched the average time of the first natural cohort (10%). Survival was lower (5%), but not significantly different when released 10 days later than the first natural cohort. Notably, survival was lowest (1%) when released 10 days earlier than the first natural cohort. Thus, our field experiment provides evidence that demography depended strongly on phenology. The results of our observational and experimental research are broadly consistent: In a situation where there would be high fitness costs to phenological shifts, phenological shifts do not seem to be occurring.