Population Dynamics of an Herbaceous Perennial Danthonia spicata During Secondary Forest Succession

Abstract

The dynamics of the grass Danthonia spicata were studied in five sites in a pine-hardwood forest undergoing secondary succession in northern lower Michigan. The five sites had last been burned in 1980, 1954, 1948, 1936 and 1911, respectively. Total population size, number of seedlings, adult survivorship and seedling survivorship were monitored by mapping quadrats from 1982-1985. Adult survivorship was also measured by marking individuals and observing their fates from 1980-1985. Fifty individuals in each site were measured for size, relative reproductive effort and survival. Population size rose rapidly following clear-cutting and fire, and slowly declined between 30 and 70 years following disturbance. One-half of each population, on average, dies every 2.2 years. The number of seedlings and total population size varied significantly among census years. Seedling survivorship and adult survivorship did not vary significantly. Number of seedlings, total population size and adult survivorship varied significantly among sites of different successional age. Seedling survivorship did not vary significantly. Adult survivorship increased with increasing size and decreased with increasing relative reproductive effort. Changes in population size in these sites were controlled by genetic and environmental changes, especially the size and number of patches with high light levels. The long-term dynamics of these populations cannot be predicted by extrapolating from short-term measurements. INTRODUCTION Can successional trends be predicted by the initial state of the system combined with information on birth and death rates of the component species? Conversely, are population changes caused by environmental shifts mediated by succession so that information on present birth and death schedules cannot predict the future state of the system? Can genetic changes in each species during succession alter the demographic parameters of the populations again invalidating attempts at extrapolation? The answers to these questions are important if we are to create a cohesive theory about succession. Additionally, this information will help guide future studies by indicating which components, environmental changes, population dynamics or genetic shifts, should be the focus of investigation. There is also a larger issue of the coordination of the successional process. If succession occurs primarily because of the independent demographic and genetic properties of each species then any interactions among species are a chance phenomenon. On the other hand, if succession is driven primarily by environmental changes controlled by the abundances of the dominant species, then consistent interactions among species are to be expected as one property of the system potentially setting the stage for coevolution. Demographic studies of successional species have focused on the dominant species, either trees in temperate and tropical forests or herbaceous plants during the early years of old-field succession (e.g, Spring et al., 1974; Pickett, 1982; Knowles and Grant, 1983; Sakai et al., 1985). Studies of dominant species have shown a complex feedback mechanism by which changes in abundance cause environmental changes which in turn effect demographic parameters (e.g, Woods and Whittaker, 1981; Peet, 1981). For an herbaceous perennial in a forest succession sequence the separation of those factors is more straightforward. Environmental changes are primarily mediated by the trees and thus are external to the dynamics of the herbaceous population. Herbaceous species that persist for long periods account for a significant proportion of the total species in a temperate forest system (Scheiner and Teeri, 1981; Collins et al., 1985). Thus, a study