Tropical Forest Community Ecology

Symmetric Neutral Theory 151

30

BCI

Log 2 species abundance

Dispersal limitation mechanism Density dependence mechanism

No. of species

20

10

0

0510

40

Korup

20

0

0 5 10 15

100

150

Lambir

50

0

0510

100

Pasoh

50

0

05 10

150

Yasuni

100

50

0

0510

10

20

Sinharaja

0

0510

Figure 9.2 Preston-style relative tree species abundance distributions for six 50 ha plots across the New and Old
World tropics, showing the equally good fits of two neutral models that have very different mechanisms. The black
bars are the data. Thexaxes of each graph are abundance categories binning species abundances into classes of log
to the base 2 individuals per species. Theyaxis is the number of species in each abundance class. The circles are the fit
of the dispersal limitation hypothesis of Hubbell (2001) and island biogeography theory. The line is the fit of the
symmetric density dependence hypothesis (Volkovet al.2005). The sites are as follows: “BCI” is Barro Colorado
Island, Panama; “Korup” is Korup National Park, Cameroon, West Africa; “Lambir” is Lambir Hills National Park,
Sarawak, Malaysian Borneo; “Pasoh” is Pasoh Research Forest, Peninsular Malaysia; “Yasuni” isYasuni National
Park, Amazonian Ecuador; “Sinharaja” is Sinharaja Forest Reserve, Sri Lanka.

Table 9.1 Parameter values for the two versions of neutral theory, dispersal limitation and density dependence,
for the six large plots of tropical forest whose distributions of relative tree species abundance are illustrated in
Figure 9.2.

Forest From

data

b/d

x

Dispersal

limitation

Density

dependence

SJ θ 1 m θ 2 c

Barro Colorado Island, Panama 225 21,457 0.9978 48.1 0.09 47.5 1.80

Yasuni, Ecuador 825 17,546 0.9883 204.2 0.43 213.2 0.51

Pasoh, Malaysia 678 26,55 40.9932 192.5 0.09 189.5 1.95

Korup, Cameroon 308 24,591 0.9979 52.9 0.54 53.0 0.24

Lambir, Sarawak 100 433,175 0.9915 288.2 0.11 301.0 2.02

Sinharaja, Sri Lanka 167 16,936 0.9982 27.3 0.55 28.3 0.38

Notes: These data are for trees>10 cm diameter at breast height. ParametersS(total number of species) andJ(total number
of individuals) are from the data and do not need to be estimated.θ 1 andθ 2 are the biodiversity numbers estimated under the
dispersal limitation and density dependence models, respectively. Parametermis the probability of immigration per birth under
the dispersal limitation model, andcis the density dependence parameter in the density dependence model (after Volkovetal.
2005).