Transient Behavior and Life History Analysis of Teasel (Dipsacus Sylvestris Huds.)

Abstract

A transient behavior analysis of the population dynamics of an herbaceous plant species (teasel, Dipsacus sylvestris Huds.) was developed to contrast with the long—term limiting response analysis of an earlier study. Following the arrival of a cohort of seeds at a site, there is typically an immediate increase in the rate of growth of rosette cover and the rate of seed production, followed by a decline several years later. This decline, caused by shifts in the stage structure of the population, may be critical in determining success or failure of a colonization episode. Simple calculations of the rate at which a population of Dipsacus might expand in space show that density effects on numbers of adult plants will become important in 15—20 yr. This conclusion supports the use of transient analysis in addition to the standard limiting behavior analysis, which assumes a constant environment over long time intervals. The reproductive value of Dipsacus increases exponentially from one stage to the next, and as a power function of accumulated biomass. A hint of a sigmoid response of reproductive value to biomass suggests that the transition from the vegetative rosette to the flowering stage may have evolved in response to the rate of accumulation of reproductive value as plant biomass increases. The response of Dipsacus to a variety of hypothetical changes in life history parameters was examined. A general method of calculating the sensitivity of population growth rate (lamda)_m was developed; when applied to the teasel model it showed that (lamda)_m is most sensitive to changes in the growth and reproduction of early stages. Delayed seed germination is not important to long—term population growth; the complete elimination of delayed germination from the model caused no more than a 4% decrease in (lamda)_m. It may be much more important in terms of transient behavior; elimination of delayed germination depressed the initial decline in rosette cover by up to 72%. If this decline delays the colonization process, the most important populational effect of delayed germination would be an increase in the probability that an attempted colonization will be successful. We also examined the optimality of the current biennial life history of Dipsacus, relative to both annual and perennial alternatives. The shift to an annual strategy is rendered suboptimal by the low probability of germination and survival of seeds relative to the probability of survival of rosettes, and by the high growth potential of rosettes. This growth potential increases the energy available for seed production by an order of magnitude over the hypothetical annual strategy, greatly increasing the reproductive value of the biennial flowering plants. The switch to a perennial habit is suboptimal because of the cost, in annual reproductive output, of maintaining a perennial flowering plant. This effect is greatest in rapidly growing populations. There seem to be no important differences between the transient and the limiting behavior analyses of life history optimality. Our direct, rather than inferred, demonstration of the optimality of teasel's life history supports the development of current theory based on optimality assumptions