PARENT SESSION
Poster Session 36: Genetics / Population Ecology
Thursday, August 11, 5:00 PM - 6:30 PM, Exhibit Hall 220 A-E, Level 2, Palais des congrès de Montréal

Demography of fragmented populations of a long-lived common plant.

Tomimatsu, Hiroshi^*,1, Ohara, Masashi², ¹ Tokyo Metropolitan University, Hachioji, Japan² Hokkaido University, Sapporo, Japan

ABSTRACT- In fragmented plant populations, decreased fecundity has been considered a primary cause of local extinctions. However, the consequences of fragmentation on population viability can be controversial particularly for perennial plants, because density-dependent mortality may buffer the effects on recruitment and variable demographic terms such as fecundity may contribute little to population growth (i.e., low elasticity values). We investigated demographic variability in the common, long-lived nonclonal herb Trillium camschatcense that occurs in the understory of deciduous forests within the agricultural matrix of northern Japan. Five years of census data from four fragmented populations with different sizes were analyzed using stage-structured matrix models to see: (1) How large are spatial and temporal variations in population growth rate, ? (2) How do life-history processes contribute to the variance of ? The results indicate that was variable across populations and census periods (0.874–1.113), and in some cases it differed significantly from =1. Fecundity, defined as the number of seedlings produced per flowering plant, was also variable (1.03–12.86). Although elasticity analysis showed that the fecundity contributed only proportionately little to , it was positively associated with survival of flowering plants which made the greatest proportional contribution. As a result, life-table response experiment analysis revealed that variations in the fecundity substantially contributed to the variance in . The contribution of fecundity was diminished, however, by negative covariances between fecundity and survival of juveniles probably caused by density-dependent mortality. Our results suggest that the significance of decreased fecundity for long-term dynamics of fragmented plant populations depends not only on variance and elasticity values of fecundity but also on the patterns of covariances between fecundity and other demographic terms.

Key words: habitat fragmentation, recruitment, environmental stochasticity, life-table response experiment