Table 4.
Mean MOTU richness estimates, standard deviation, and confidence intervals for each sampling strategy
| Strategy | Observed individuals (genetic) | Singletons mean | Doubletons mean | ICE mean | ICE SD | Chao2 mean | Chao2 SD | Chao2 95% CI | Jack1 mean | Jack1 SD |
|---|---|---|---|---|---|---|---|---|---|---|
| Time-based* |
18 |
10 |
5 |
32.0 |
0 |
24.7 |
5.7 |
19.6 – 46.6 |
26.9 |
4.5 |
| Rapid-blitz* |
18 |
12 |
1 |
41.3 |
0 |
90 |
83.5 |
29.8 – 458.2 |
29.1 |
4.3 |
| Liberal re-sort* |
29 |
18 |
5 |
58.2 |
4.9 |
60.4 |
1.8 |
37.8 – 135.4 |
44.7 |
7.4 |
| Fixed-protocol▪ |
17 |
7.2 |
2.6 |
24.0 |
10.7 |
25.7 |
15.3 |
14.0 – 91.2 |
18.1 |
2.1 |
| Flexible-protocol▪ |
29 |
13.3 |
5.5 |
42.6 |
15.1 |
40 |
13.9 |
26.4 – 90.4 |
33.6 |
3.8 |
| Fixed-protocol♦ |
17 |
5.4 |
1.3 |
21.5 |
13.0 |
14.7 |
7.6 |
8.6 – 46.0 |
11.9 |
2.1 |
| Flexible- protocol♦ | 29 | 12.8 | 3.5 | 56.6 | 47.5 | 41.0 | 16.2 | 24.6 – 97.6 | 29.1 | 4.3 |
To enable a comparison of strategies despite differences in number of sites sampled, we show here the diversity estimators for a mean of 14 sites sampled for all strategies except for time-based, for which the total number of sites sampled was nine. Fourteen was selected due to being the maximum number of sites available for two of the strategies, and nine sites (total for time-based) was recognized as being too few sites to accurately estimate richness for several of the strategies. By comparison, Chao2 and ICE stabilized at approximately 15 sites sampled for most strategies (Figure 3). Values are shown to one decimal place.
SD refers to standard deviation. CI refers to confidence interval.
*refers to strategies where the number of total sites sampled corresponds to the total number of sites where ostracods were found.
■ refers to results where only the total number of sites where ostracods were found was included.
♦ refers to results where the total number of sites sampled was included regardless of the presence of ostracods.