Biology Reference
In-Depth Information
occur, and in-water disposal areas for sediment. The survey must be completed not
earlier than 90 days prior and not later than 30 days prior. If any
C. taxifolia
is
found, activity must be halted until the infestation was isolated, treated, and any
risk eliminated. If work is to be undertaken in
Caulerpa
-infected waters, two sur-
veys not less than 60 days apart must occur, one at the high intensity level (50% of
bottom covered) and one at the eradication level (100% of bottom covered). If the
project extends over 90 calendar days, additional surveys at the high-intensity level
will be required. Additional surveys of dredged materials may also be required.
15.5.9 Modeling
Models allow managers to focus limited resources to survey only the most suitable
sites for an invasive species, such as
C. taxifolia
. Glardon et al. (2008) have devel-
oped such a model for the genus
Caulerpa
in Florida waters. Glardon et al. (2008)
conducted field surveys of 24 coastal areas around Florida in each of six zones
chosen in a stratified manner and evaluated the association of potential indicators
for the presence of
Caulerpa
. In total, 14 species of
Caulerpa
, but not
C. taxifolia
,
were found at 31 of the 132 sites. Latitude, presence of seagrass beds, human popu-
lation density, and proximity to marinas were simultaneously considered. A posi-
tive correlation between
Caulerpa
spp. presence and seagrass beds and proximity
to marinas was documented while a negative correlation with latitude and human
population density was also noted. The parameters in the logistic regression model
assessing the association of
Caulerpa
occurrence with the measured variables then
were used to predict current and future probabilities of
Caulerpa
spp. presence
throughout the state. Percent correct for this model was 61.5% for presence and
98.1% for absence. While aquarium dumping provides an explanation for the posi-
tive correlation with marinas, the human population density results were surprising.
This may be because, in Florida waters, high population densities enhance pollutant
loads, freshwater inputs, and nutrient runoff, and these factors may decrease mac-
roalgal growth.
A second type of useful model is one that accurately predicts the pace of an
invasion once it has begun so that managers know how best to undertake eradica-
tion. Ruesink and Collado-Vides (2006) found that the model that best describes
actual field distributions of
C. taxifolia
invokes local growth via rhizome expansion
plus low levels of fragment dispersal and attachment (increases of 4-14-fold annu-
ally). The model goes on to suggest that the most effective plan for maximizing
eradication is removal of established patches before summer and removal of
fragments in the fall (Ruesink and Collado-Vides 2006). The times corresponded to
just before maximum growth and just after maximum fragment production, respec-
tively. Only a mixed strategy that combined 99% removal of all fragments and
annual removal of 99% of established patches was predicted to entirely eliminate
C. taxifolia
(Ruesink and Collado-Vides 2006). This level of effort is only likely to
be possible early in an invasion.
