Drought and Survival in a Self-Perpetuating Pinus pungens Population: Equilibrium or Nonequilibrium?

Abstract

A 10-year follow-up study of recruitment and survival of table mountain pine (Pinus pungens) growing in shallow soil on a granite dome in western North Carolina showed that the population, which perpetuates itself without fire or other disturbance, has experienced long intervals of high survival and recruitment interrupted by brief episodes of low survival and recruitment probably caused by extreme drought. For these pines growing in shallow soil over bedrock, the Palmer Drought Severity Index was poorly correlated with episodes of low survival. A more appropriate drought index for the edaphic situation was the maximum number of consecutive rainless days in each 5-year age-class germination interval. Analysis of age-specific survivorship for the 1977-1986 decade showed that table mountain pines less than 30 years of age experienced lower survival rates during drought than older pines. INTRODUCTION In 1976 an unusual population of table mountain pine (Pinus pungens Lam.) was found growing on the rounded SW shoulder of Looking Glass Rock, a 1-km2 granite dome near Brevard, in western North Carolina (Barden, 1977). The age structure of these pines, derived from increment cores, showed that this population was an edaphic climax which had perpetuated itself since the most recent fire in 1889 (Barden, 1977). Zobel (1969) had predicted that such a stand of Pinus pungens might be found on rock outcrops or shale slopes where hardwood species grow poorly. McIntosh (1950) found pine populations (Pinus strobus, P resinosa, and P banksiana) which, on the basis of size structure of the stands, appeared to be perpetuating themselves on sandstone and limestone bluffs in southwestern Wisconsin. However, the table mountain pine population on Looking Glass Rock is the first reported pine climax, based on known age structure and disturbance history, in the eastern United States (Barden, 1977). In 1976 I interpreted the close fit of a log-linear curve to the age-structure data as evidence that the population was approaching equilibrium, that is, having relatively constant and approximately equal mortality and recruitment rates. Harper and White (1974) suggested an expanding population as an alternative interpretation for such data, but this explanation seemed unlikely because of the number of dead pines found in varying stages of decomposition (Barden, 1977). The carrying capacity of the site appeared to be set by the availability of pockets of soil deep enough to support droughttolerant pines but too shallow to support hardwoods during summer droughts (Barden, 1977). In 1986 I recensused the same trees, using detailed maps and fixed-point photographs from the 1976 study to relocate all members of the population. The purpose of the new study was to determine whether the population continued to exhibit the loglinear age structure after 10 additional years of recruitment and mortality. My reinterpretation of the cohort data, based on 5-year rather than 20-year age classes and a more critical analysis of weather records, differed sharply from conclusions of the earlier study, reflecting a new understanding of the importance of sporadic high morality in the population dynamics of the stand. STUDY AREA AND METHODS Looking Glass Rock is an oblong granite dome, 2.0 km by 0.5 km, which rises 300 m above the surrounding terrain to a maximum elevation of 1200 m. It is located 10 km NW of the nearest permanent weather station at Brevard, North Carolina, where