Environmental Engineering Reference
In-Depth Information
environmental weeds. Panetta and Timmins (2004) have suggested that the
eff ort required to achieve eradication should be a function of both gross infest-
ation area (
=
the area over which the weed is distributed and must be searched
in repeat visits following control treatments which may or may not be applied
over the whole area at each event), and what they call 'impedance' (
=
groups of
constraints such as accessibility, detectability, weed biological characteristics,
control e
cacy).
Two regression approaches have been used to model the costs of eradication
against scale of infestation and other attributes of weeds. Cunningham
et al
. (2003)
surveyed experts in weed eradication for their views on the probability of success for
each of 15 hypothetical weeds with diff erent life histories and status, within eight
broad cost ranges. Regressions of the relationships between eradication cost and
weed incursion/attributes showed that only four variables (as categorical values)
were signifi cant: the total gross area of infestation (a), the number of infestations (b),
the ease of access for control (c), and seed longevity (d):
Cost
=
1000 exp[9.43
+
(
-
0.5a)
+
(
-
0.63b)
+
(
-
0.36c)
+
(
-
0.42d)]
Woldendorp and Bomford (2004) modelled costs (estimated from actual
eff ort) for 20 weed eradication projects in various stages of completeness from
around the world against a variety of attributes. h is study showed that the
cost of eradication was largely determined by the total area of infestation that
was treated, the net area (c.f., the gross area of infestation in Cunningham
et al
.
(2003)) with no improvement to the model (Cost
e
9.89
Area
0.66
) by inclusion
of other variables, although these may have been implicit within the original
decisions that selected the projects as being feasible. Interestingly, the actual
costs for those projects successfully completed were almost always lower than the
costs modelled (Table 4.2), which hopefully is not an indication of false claims
of success!
Plant eradications generally take longer to achieve than animal eradications,
but application of some of the concepts discussed above has suggested that the
view of Rejmánek and Pitcairn (2002) about the feasibility of larger-scale eradi-
cations might have been too pessimistic, or conversely that agencies have been
prepared to invest more resources than considered possible by Rejmánek and
Pitcairn. h e tumbleweed bush (
Bassia scoparia
) has been eradicated from over
2480 ha in Western Australia, although, since it was introduced deliberately and
planted as a forage plant, its total extent was known, making the task easier than it
otherwise might have been (Panetta and Lawes 2005). An eradication campaign
in the USA against
Striga asiatica
, an obligate root parasite of cereals and legumes,
has since the early 1970s reduced the infestation from c. 200 000 ha to less than
1600 ha (Eplee 2001; R.E. Eplee, unpubl. data). Eradication may be possible
for this plant because there is an eff ective means of eliminating soil seed banks,
which are proving a major impediment to eradicating another annual parasitic
weed, branched broomrape (
Orobanche ramosa
), over 7000 ha in South Australia
(Panetta and Lawes 2007).
=
