Environmental Engineering Reference
In-Depth Information
Table 1
Long-term eagle collision rate, with confi dence intervals (CI), based on data collected up
to October 2012
Tasmanian wedge-tailed eagle White-bellied sea eagle
BPWF SBWF BPWF SBWF
Mortality (per year) 1.54
0.95 0.36 0.00
Lower 95 % CI 0.82 0.31 0.07 0.00
Upper 95 % CI 2.63 2.22 1.04 0.7
Note that these data cover the monitoring periods 2002-2012 for the Bluff Point Wind Farm, and
2007-2012 for the Studland Bay Wind Farm
X
2
= 3.5 <
X
2
0.95, DF = 4, p > 0.4. The Laplace U test, found that an increase in the
mortality rate over time was strongly rejected (BPWF: U = −0.662, p ~0.5). At the
SBWF there was a possible decrease in the collision rate over time (U = −2.759,
p < 0.01), but the small sample size involved means this pattern may be due to a false
positive result.
There was no evidence of a pattern in the interval between collisions at either
wind farm (Mann test: BPWF: Z = −0.305, p > 0.75; SBWF: Z = 1.225, p > 0.24).
The Wald-Wolfowitz (WW) test also found no support for a pattern (BPWF: W-W
Z = 0.605, p > 0.5; SBWF: W-W Z = 0.109, p > 0.9). Note that when there are less
than 10 “runs” in a data set, the WW test is conservative (i.e. has the possibility of
indicating a deviation when there is none, or making a Type I error). The WTE data
contained eight runs at BPWF and fi ve at SBWF. All these tests indicated there was
no evidence to support the assertion that events were clustered in time, and instead
there was strong support for the null hypothesis that the incidence of collisions at
these sites was random.
Eagle Breeding Success
As stated in the methods section, no formal patterns could be inferred from the data
as analysis was not possible. However, the survey results are presented in Tables
2
and
3
. Each line represents a separate nest, and each column has the results from the
two surveys each year. Active nests (where evidence of attendance at the nest was
observed, including indications such as refurbishing of the nest with material such
as green leaves) are denoted by a “1” and inactive by “0”. Note that successful
breeding in this case is defi ned as the presence of a chick at the second nest check
(later in the breeding season). The second last row shows the proportion of active
nests (in the fi rst survey each year). The bottom row is the proportion of success
(active at second survey), given that the nest was active the fi rst time.
The time series for both the activity and success were fl at (no evidence of a trend
up or down), as shown in the lower two lines of the tables. The two WTE nests at
the BPWF (522 and 854) were amongst the most likely to be active, and showed no
sign that their likelihood of breeding success was any different from other nests.
Table
2
also indicated that 2006 was a bad year for all nests, with a low chance of
