Spatial Arrangement of Founders - Genetic Diversity in Establishing Plant Populations: Founder Number and Geometry - page 137

populations, setting populations on different growth and genetic
diversity trajectories.
4. The effects described in 1 and 2 may sometimes be of large signifi cance
even when differences in the spacing of founders appear to be minor
(e.g., no versus 3 average species density spaces between founders).
5. Under certain conditions, subdividing the founders into isolated
smaller groups may increase the rate of population growth, loss of
heterozygosity, F, and the number of unique alleles retained.
6. In using comparative NEWGARDEN trials to examine the effects
of varying particular input conditions, unless one is specifically
investigating the effects of differing founder geometries, it is best to
hold founder geometry constant. For example, if one wants to compare
the effects of using different numbers of founders on the percentage
of the original unique alleles retained, the founders should always be
placed in the same geometric shape (e.g., 25 versus 36 founders, both
placed in squares, identical spacing between founders).
7. Initiating NEWGARDEN trials with founders in different spatial
geometries can be used to develop best practices reintroduction
strategies, although cost-benefi t analyses will have to take into account
which is more important, for example: cost-effort for reintroduction
versus population growth versus inbreeding sensitivity versus unique
allele retention.
8. NEWGARDEN analyses can be used to develop population growth rate
and genetic diversity benchmarks to guide users as to whether, when,
and how developing populations could be manipulated to improve
population growth rate and genetic diversity retention.
9. When comparing patterns of demography and genetic diversity among
populations, one must include initial spationumeric deployment
of founders as one factor that may contribute to interpopulation
differences.
Risk Assessment
With regard to measures of population growth and genetic diversity,
NEWGARDEN analyses provide one way of comparing the risk either
among alternative introduction strategies or associated with the variation
that may occur in the development of new, naturally founded populations.
Thus far we have only reported the mean values of such measures for each
generation calculated across 30 runs for each set of input trial conditions.
NEWGARDEN also provides a standard deviation value for each mean
value. While obviously the standard deviations can be used to test whether
means are signifi cantly different, they also indicate the amount variation
about that mean and can be used to compare risk. For example, Fig. 9.13