Population structure and growth‐stress relationship of Pinus taeda in rock outcrop habitats

Abstract. On the granite outcrops of the southeastern United States, soil accumulates in shallow depressions on the rock surface. A specific sequence of vascular plants characterizes the temporal development of these systems. The edaphic end point of the succession is apparently attained with a herb‐shrub‐tree stage with Pinus taeda as the dominant tree. We studied the characteristics of this stage and the population structure of P. taeda on outcrop islands in order to specify the successional status of the species in this habitat. We compared the radial growth pattern of outcrop and Piedmont populations of P. taeda , and of two outcrop sub‐populations. We checked whether trees on outcrops experience more limiting conditions than trees on the Piedmont, and studied the recent change in the relationship between growth and stress (e.g. drought and atmospheric deposition) reported for loblolly pine in the southeastern United States. We also attempted to identify the climatic variables most critical for tree growth on outcrops. On outcrop soil islands, P. taeda maintains populations that are of irregular age distribution, possibly in response to irregular recruitment and survival. There are no signs of loblolly pine replacement by hardwood species on any of the islands studied, although an understory of shrubby oak appears to characterize larger and deeper‐soil islands. Although trees on the Piedmont were growing at a higher rate than those on the outcrop in the first part of the 1950–1988 period, their growth declined at a higher rate than that of trees on the outcrop. As a result, at the end of the period considered (1950–1988), the differences in radial growth between outcrop and Piedmont populations were relatively small. On the outcrop, trees < 22 yr old in 1989 were growing at a rate somewhat lower than that of trees of similar age, 40 yr ago. Differences were apparent in the initial growth patterns between the two outcrop subpopulations (1989 and 1949 stems), and these could have resulted from differences in competition regime, stress or climatic conditions, or a combination thereof. However, a repeated‐measures ANOVA failed to reveal a significant recent decrease in the radial growth of loblolly pine in the system studied. The identification of numerous signature years (years with ≥ 80 % of the trees with similar increasing or decreasing trend in their radial growth) suggests that similar environmental variables control the growth of loblolly pine in both outcrop and Piedmont habitats. High temperature and low precipitation in the first part of the summer (June‐July) seem to limit radial growth. Those pines growing on outcrop soil islands, however, appear more sensitive to climatic fluctuations.