Figure 7.23b. Figure 7.24 provides the relative frequency of these six species in each of
the sites in the chronosequence of constructed forested wetlands. The methods of
determining frequency in this study result in relative frequency, number of individuals of
a species divided by the total number of species. These methods may not be comparable,
so direct comparisons are considered tentatively. It is suggested, however, that tracking
of species dominance within a constructed wetland with time may elucidate the direction
of self-organization processes occurring within a wetland. Species may sort themselves
by environmental variables that reflect past and current site conditions.

 As with species richness, canopy tree density is at least initially dependent more
 on humans than recruitment from natural areas. Davis and others (1991) reported
 densities of trees greater than 5 cm from 1183-1809 trees ha-'. Cypress domes had the
 highest densities followed by hardwood swamps and bayheads. Constructed forested
 wetlands had tree densities (>5 cm) from 0-1102 trees ha'1. The eight-year-old site was
 the youngest to reach the densities of natural systems, while two ten-year-old sites and a
 sixteen-year-old site failed to achieve those densities. A fire in the sixteen-year-old site
 may explain low densities in this site, and the low planting densities during construction
 may explain the low density in the ten-year-old sites. '

 Shannon-Weaver canopy tree diversities for natural wetlands range from 1.08 to
2.07 (Davis and others 1991). Diversity of constructed forested wetlands in this study
ranged from 0 to 0.85. Species richness and diversity of canopy tree species in
constructed forested wetlands begins to approach that of natural systems but only in a few
of the richest and most diverse sites.

 Tree height, diameter at breast height and canopy cover are easily measured
parameters, and the variability in these parameters may be useful in assessing success,
but they provide little insight into the organization of the constructed wetland as a viable
system. Results of this study suggest that 83% of the variation in tree height is explained
by age, while only 61% of the variation in diameter at breast height is explained by age.
Variation in driving energies (environmental variation) associated with the site and the
resulting probability that a site has been created that can maintain a self sustaining
forested wetland may be more apparent from evaluating aspects of tree diameter at breast
height rather height or canopy cover since a greater percentage of the variation in dbh and
canopy cover is explained by some factor other than age.

 Frequency distributions of tree diameter at breast exhibit a hierarchical pattern in
natural systems with many smaller trees and few larger trees (Odum 1983). Odum
(1983) presents a size class distribution pattern typical of an even aged stage that slowly
moves up the size classes with age. In agroforestry, this represents a cohort of trees
established by planting at one time. This pattern of size class distribution also occurs in
nature. It can result from a stochastic event such as fire in a pineland or drawdown in a
cypress dome creating a period for recruitment and may be typical of specific species of
trees.


7-84