The importance of space, time and stochasticity to the demography and management of Alliaria petiolata

Authors: EVANS, J - Dartmouth College; Davis, Adam; RAGHU, S - Northern Territory Government; RAGHAVENDRAN, A - Purdue University; LANDIS, D - Michigan State University; SCHEMSKE, D - Michigan State University

Submitted to: Ecological Applications
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/23/2012
Publication Date: 9/1/2012
Citation: Evans, J.A., Davis, A.S., Raghu, S., Raghavendran, A., Landis, D., Schemske, D. 2012. The importance of space, time and stochasticity to the demography and management of Alliaria petiolata. Ecological Applications. 22:1497-1511.

Interpretive Summary: As population modeling studies are increasingly called upon to guide policy and management decisions, it is important that we understand not only the average behavior of our study systems, but their variation in space and time as well. Quantifying the sources of variability throughout the life history of an organism is the first step in this process. We studied the demography of 12 natural populations of the invasive forb Alliaria petiolata (garlic mustard) over three generations of its life cycle in Illinois and Michigan. Garlic mustard population density and climate variables played important roles both in directly affecting survival and in predicting the fates of individuals later in their life histories. Variation in germination, survival, and reproductive rates was significantly structured across sites and years. Fecundity was positively correlated with the mean temperatures during the summer and winter after germination, but negatively related to total rainfall during the spring of the second year. The frequency and amount of rainfall early in the life history were associated with increased survival for the remainder of the life cycle, but increasing summer rainfall and temperature negatively affected survival. Despite being limited to observations of populations located in the midwest U.S., the degree of variation observed in garlic mustard demographic rates encompasses the results of almost all previous studies of this species. Previous studies that have explored subsets of garlic mustard's demographic range of variability may therefore be limited in the scope and applicability of their predictions.

Technical Abstract: Variability in demographic rates among natural plant populations can have large impacts on population structure and cause populations to exhibit subtle or radical differences in behavior over time. As population modeling studies are increasingly called upon to guide policy and management decisions, it is important that we understand not only the central tendencies of our study systems, but their variation in space and time as well. Quantifying the sources of variability throughout the life history of an organism is the first step in this process. We studied the demography of 12 natural populations of the invasive forb Alliaria petiolata (garlic mustard) over three generations of its life cycle in Illinois and Michigan. Hierarchical, generalized linear mixed models (GLMMs) were used to analyze the spatial and temporal scales of structured variability in each sub-annual demographic transition in the life cycle, and the intrinsic, edaphic, and climate-driven sources of this variability. We developed statistical approaches to deal with common discontinuities, such as zero- and one-inflated data distributions, that made conventional analyses problematic. Garlic mustard population density and climate variables played important roles both in directly affecting survival and in predicting the fates of individuals later in their life histories. Variation in germination , survival, and reproductive rates was significantly structured across sites and years. Persistence of dormant seeds in the soil seed bank was high, with a mean viability of 91.9% after one year. Seedling and summer and winter rosette survival were influenced by negative density-dependent feedbacks. Overall winter rosette survival was not density-dependent, although the probability of extremely high (1) and low (0) winter survival was negatively density-dependent. Fecundity was positively correlated with the mean temperatures during the summer and winter after germination, but negatively related to total precipitation during the spring of the second year. The frequency and amount of precipitation early in the life history were associated with increased survival for the remainder of the life cycle, but increasing summer precipitation and temperature negatively affected survival. Despite being limited to observations of populations located in the midwest U.S., the degree of variation observed in A. petiolata demographic rates encompasses the results of almost all previous studies of this species. This variation is highly structured in space and time in response to biotic and abiotic conditions. Previous studies that have explored subsets of A. petiolata's demographic parameter space may therefore be limited in the scope and applicability of their predictions.