SEED DISPERSAL AND SEEDLING RECRUITMENT OFBETULA ALLEGHANIENSIS: SPATIAL INCONSISTENCY IN TIME

Abstract

Recruitment in plants is the result of several processes operating from the time the seeds are mature and ready to be released from the parent plant to the time the seedlings emerge and become photosynthetically self-sufficient. In most species of strongly seasonal climates (e.g., trees of the cold temperate forests of North America and Japan), seed dispersal and seedling establishment are processes separated in time by at least a few weeks and more typically by a few months. This leaves ample opportunity for various factors to affect the spatial patterns of seed dispersal, survival and germination, and seedling emergence. Spatial uncoupling between these sequential processes leads to spatial independence (i.e., absence of a relationship) in the abundance patterns of life stages. Along a topographical gradient in an old-growth cold temperate forest of northeastern North America, I analyzed the spatial patterns of seed rain, seed bank, and seedling emergence of a major tree component, Betula alleghaniensis, over a 4-yr period. I examined the degree of consistency through time of the spatial patterns of each life stage and of their relationships with one another. Seed rain and seedling spatial patterns were quite variable through time, but that of the seed bank was relatively consistent despite large interyear fluctuations in abundance. Overall, two major spatial patterns emerged along the topographical gradient: one that emphasized the three peaks of density of mature Betula trees, and one with a single peak in the middle section of the gradient, where Betula tree density was higher. Correlations between the spatial patterns of seed and seedling abundance were significant in some years, but not in others. Seed survival was practically constant in space in one year, but quite variable in another; yet, it was consistently density independent. Spatial uncoupling was important between processes as reflected by the spatial independence between life-stage abundance patterns. This uncoupling significantly contributes to the spatiotemporal heterogeneity of the system and complicates fine-scale modeling of population and community dynamics.