Ecological Archives E096-069-A4
Byju N. Govindan, Zhilan Feng, Yssa D. DeWoody, and Robert K. Swihart. 2015. Intermediate disturbance in experimental landscapes improves persistence of beetle metapopulations. Ecology 96:728–736. http://dx.doi.org/10.1890/14-0044.1
Appendix D. Model-averaged estimates of selected predictor variables for extinction-colonization rates, and pairwise comparisons of the eight treatments for selected parameters.
Table D1. AICc based model averaged estimates (± adjusted standard error) of predictor variables for landscape-level colonization and extinction rates. An asterisk indicates that the 95% credible interval did not contain zero.
Response Variable |
Intercept |
Predictors (Coefficient ± adjusted SE)1 |
||||
RS |
PC |
PD |
RS X PC |
RS X PC*PD |
||
c |
0.47 ± 0.04 * |
0.07 ± 0.05 |
0.28 ± 0.06* |
0.06 ± 0.04 |
-0.14 ± 0.06* |
|
e |
0.11 ± 0.01* |
0.01 ± 0.01 |
0.08 ± 0.01* |
0.04 ± 0.01* |
|
|
Table D2. Mean (and SD) of landscape-level metapopulation attributes for the eight treatments. Treatment means without shared superscripts are different from each other (Tukey sandwich HSD, P < 0.05).
Trt # |
RS |
PC |
PD |
N |
e |
c |
|
1 |
Homogeneous |
High |
Fast |
3.20bc (0.32) |
0.19c |
0.75d |
1.99ef (0.13) |
2 |
Homogeneous |
High |
Slow |
2.87b (0.16) |
0.12b |
0.75d (0.05) |
0.75 (0.04) |
3 * |
Homogeneous |
Low |
Fast |
3.56c |
0.10ab (0.02) |
0.52c (0.04) |
2.65g (0.27) |
4 |
Homogeneous |
Low |
Slow |
3.69abc (0.49) |
0.07a (0.03) |
0.37a (0.08) |
1.39bc (0.34) |
5 ** |
Heterogeneous |
High |
Fast |
3.61c |
0.18c (0.01) |
0.69bd (0.01) |
2.13de (0.02) |
6 |
Heterogeneous |
High |
Slow |
3.49c |
0.14bc (0.02) |
0.66bd (0.04) |
0.88a |
7 ** |
Heterogeneous |
Low |
Fast |
4.27a |
0.10ab (0.00) |
0.50abc (0.15) |
2.86bcdfg (0.85) |
8 ** |
Heterogeneous |
Low |
Slow |
4.53a |
0.08a (0.02) |
0.49ac (0.09) |
1.06a (0.15) |
*or ** Mean ± SD is based on two replicates [* One replicate for treatment #3 suffered fungal infestation; ** 95% credible interval for the dispersal rate for one replicate each for treatments #5, #7, and #8 contained zero]. RS, PC, and PD denote the predictor variables resource structure, patch connectivity and patch dynamics respectively. The response variables N, e, c and denote mean abundance of beetles, extinction rate, colonization rate, and extinction threshold respectively, all estimated at the landscape-level.