Ecological Archives M083-012-A5

D. M. Newbery, X. M. van der Burgt, M. Worbes, G. B. Chuyong. 2013. Transient dominance in a central African rain forest. Ecological Monographs 83:339–382.

Appendix E. Spatial statistics of large-medium tree patterns for the three dominant species.

Univariate and bivariate summaries of point process statistics for large (‘lar’, ≥ 50-cm stem diameter) trees of the three main caesalpinaceous species (Mb, Microberlinia bisulcata; Tb, Tetraberlinia bifoliolata; Tk, T. korupensis), in the P- and NW-plots at Korup, and where trees of a species were put into two size classes (‘dcl’, i.e., those < or ≥ median diameter), are shown in Fig. E1. Bivariate summaries are similarly shown for large and medium (‘med’, 10 – < 50-cm diameter) trees, either within a species or between two species, in Fig. E2. Not all combinations of species and size classes are shown, however — just those that were most significant and are mentioned in the Results.

L(t)–r is the difference between the Ripley K(t)-measure (square-root transformed) and the radius (r) as a measure of departure from expectation under a random process. The empirical data are shown as black lines, and the mean and upper and lower levels of the 95% significance bands (n = 39 simulations) are shown as red, green and blue lines respectively. Linhom(r)–r is the corresponding bivariate measure which takes into account trends in spatial heterogeneity of the two species or size classes.

These figures are largely for illustration and are at one common P-level: graphics output for other (higher) significance levels, in places referred to in the main text, are not shown here since they are largely similar. Note that the simulation results are seed-dependent, and reruns will differ slightly. Several reruns indicated the main significant patterns to be robust.


Fig. E1.



Fig. E2.

[Back to M083-012]