The mean loss of occupied 20 m cells did not persist at 100 individuals for 130 Barro Colorado woody species, 1985–2015

We tested whether common woody species occupied fewer cells after the number of individuals was held constant in the Barro Colorado Island 50 ha forest plot, Panama. We used 130 named species with at least 100 living, measured individuals of diameter at least 1 cm in both the 1985 and 2015 censuses. Mean occupancy in 1,250 fixed 20 m cells fell from 460.07 to 440.25 cells per species. The exact expected occupancy of a random sample of 100 individuals per species rose from 88.19 to 88.38 cells. The mean change was +0.19 cells, or +0.21%. Thus the raw mean loss did not persist at this fixed count. There were 68 positive and 62 negative species changes. The middle 80% ranged from -2.36 to +3.41 cells. Checks with 10 m, 50 m and 100 m cells also gave small positive mean changes. A 10 cm diameter cutoff gave a small negative mean change for 40 eligible species. The result describes persistent common species in one mapped plot. It does not establish unchanged distributions or a cause. The package has source extracts, provenance, code, all species results, figures, independent numerical checks, and an offline reproduction record.

spatial occupancy · forest dynamics · Barro Colorado Island · rarefaction · species distributions · sample size · open data · reproducibility