You are not currently logged in.
Access JSTOR through your library or other institution:
Fine-Scale Predictability of Forest Community Dynamics
Mark R. Fulton and Paul A. Harcombe
Vol. 83, No. 5 (May, 2002), pp. 1204-1208
Stable URL: http://www.jstor.org/stable/3071935
Page Count: 5
You can always find the topics here!Topics: Mortality, Forest ecology, Mathematical vectors, Matrices, Plant ecology, State vectors, Predictability, Forest communities, Coastal ecology, Fire ecology
Were these topics helpful?See something inaccurate? Let us know!
Select the topics that are inaccurate.
Preview not available
We assayed the predictability of forest community dynamics by examining the internal structure of a 10-12-yr record of forest change for three sites in east Texas, USA. Within each site, the growth and fate of all adult tree stems within many small (0.04 ha) patches have been monitored. These data were reduced to community composition (basal area by species) for each of 28-30 spatially independent patches within each site. We measured predictability of compositional change by the rank correlation of (1) distances between vectors of initial composition vs. (2) distances between vectors of compositional change. Vectors of compositional change were predictable at two of the three sites; the third site was an old forest with somewhat high mortality of large adult stems. The data also allowed decomposition of the change vectors into components due to growth, mortality, and adult tree recruitment. Growth vectors were highly predictable from the initial composition, but mortality and adult tree recruitment were not. The results support the idea that growth processes in forests are closely tied to interactions among adjacent trees, whereas recruitment and mortality have components that operate at different spatial or temporal scales and so are empirically decoupled from growth.
Ecology © 2002 Wiley