Geoscience Reference
In-Depth Information
been shown for the two-category case that there is an optimal number of repeat
surveys, as discussed by MacKenzie and Royle (2005), MacKenzie et al. (2006),
and Guillera-Arroita et al. (2010). Even though the optimal number depends
on how optimality is defined (Guillera-Arroita et al., 2010), from the perspec-
tive of trying to minimize the SE of the estimated occupancy probability for a
given level of effort, or minimize effort to obtain a target SE, MacKenzie and
Royle (2005) found that the optimal number of repeat surveys tended to result
in the probability of detecting the species at least once during the repeat sur-
veys being in the range 0.85-0.95. That is, the best estimate is obtained when the
survey effort at each unit is of sufficient intensity that the presence or absence
of the species is almost confirmed, at which point analyses that do and do not
account for detection would likely provide similar results. This again suggests
that the real benefit of repeat surveys is not that they enable the separation
of the sampling (detection) and biological (occupancy) processes, but that they
result in a higher-quality data set with less potential for false absences.
MacKenzie and Royle (2005) found that the optimal number of surveys
depends on the per survey detection and occupancy probability (Table 9.2).
The suggested number of repeat surveys is optimal in the sense that, for that
same level of total effort, conducting fewer repeat surveys (therefore survey-
ing more units) or more repeat surveys (hence fewer units) results in a less
precise estimate of occupancy (e.g., as shown in Table 9.3).
The idea that there is an optimal number of repeat surveys that should be
conducted each season also applies for multiseason studies. The best pre-
cision on estimates of colonization and extinction probabilities for a given
level of effort will be achieved when the optimal number of surveys are con-
ducted. On reflection, this is reasonable because, by reducing the possibility
of a false absence within any particular season to a relatively low level, there
TABLE 9.2
Optimal Number of Repeat Surveys to Conduct per Sampling Unit for Various
Levels of Detection p and Occupancy ψ
ψ
p
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.1
14
15
16
17
18
20
23
26
34
0.2
7
7
8
8
9
10
11
13
16
0.3
5
5
5
5
6
6
7
8
10
0.4
3
4
4
4
4
5
5
6
7
0.5
3
3
3
3
3
3
4
4
5
0.6
2
2
2
2
3
3
3
3
4
0.7
2
2
2
2
2
2
2
3
3
0.8
2
2
2
2
2
2
2
2
2
0.9
2
2
2
2
2
2
2
2
2
Source: MacKenzie, D.I., and Royle, J.A. (2005). Journal of Applied Ecology 42: 1105-1114.
 
Search WWH ::




Custom Search