Ecological Archives E094-035-A3

Audrey Barker Plotkin, David Foster, Joel Carlson, Alison Magill. 2013. Survivors, not invaders, control forest development following simulated hurricane. Ecology 94:414–423. http://dx.doi.org/10.1890/12-0487.1

Appendix C. Additional statistical output for ANCOVA and breakpoint analyses.

Table C1. ANCOVA analysis summary table. The linear models were fit using the lm function in R 2.9.2, and are of the form response variable = β0 + β1 × year + β2 × treatment + β3 × year × treatment. Polynomial models were of the form response variable = β0 + β1 × year + β2 × treatment + β3 × year × treatment + β4 × year2 + β5 × treatment × year2. For the polynomial functions, the 'year' variable was centered prior to analysis. Parameter estimates that are significantly different from 0 are shown in italics.

  Basal area Litterfall,
1991–1996
Litterfall,
1997–2010
Trees with
basal sprouts
Herb cover Shrub cover*
Source of variation
Year F1,6 = 67.87
P = 0.0002
F1,8 = 8.35
P = 0.0202
F1,24 = 0.32
P = 0.5754
F1,10 = 3.27
P = 0.1007
F1,8 = 4.93
P = 0.0571
F1,6 = 18.51
P = 0.0051
Treatment F1,6 = 462.58
P < 0.0001
F1,8 = 75.13
P < 0.0001
F1,24 = 0.94
P = 0.3418
F1,10 = 17.40
P = 0.0019
F1,8 = 0.36
P = 0.5633
F1,6 = 6.22
P = 0.0468
Year × Treatment F1,6 = 6.70
P = 0.0414
F1,8 = 16.48
P = 0.0036
F1,24 = 0.38
P = 0.5452
F1,10 = 12.40
P = 0.0055
F1,8 = 2.82
P = 0.1316
F1,6 = 10.47
P = 0.0178
Year2 NA NA NA F1,10 = 23.24
P = 0.0007
NA F1,6 = 10.34
P = 0.0182
Year2 × Treatment NA NA NA F1,10 = 2.53
P = 0.1429
NA F1,6 = 5.10
P = 0.0647
Parameter estimates (SE)
Intercept (β0) 0.976
(0.036)
1.099
(0.050)
1.064
(0.145)
268.142
(22.262)
0.858
(0.169)
2.338
(0.288)
Year (β1)   0.012
(0.003)
-0.011
(0.013)
 0.000
(0.010)
13.218
(3.139)
-0.006
(0.015)
0.140
(0.028)
Treatment
(pulldown) (β2)
-0.655
(0.051)
-0.531
(0.071)
-0.064
(0.205)
38.019
(31.484)
 0.406
(0.239)
 0.141
(0.407)
Year × Treatment
(pulldown) (β3)
0.011
(0.004)
0.074
(0.018)
 0.009
(0.015)
-15.597
(4.439)
-0.036
(0.021)
-0.071
(0.039)
Year24) NA NA NA -1.965
(0.434)
NA -0.003
(0.005)
Year2 × Treatment
(pulldown) (β5)
NA NA NA  0.975
(0.613)
NA -0.015
(0.007)

*The polynomial shrub model was significantly better (F2,6 = 7.72, P = 0.0219) than a linear model.

 

Table C2. Breakpoint analysis for Litterfall. Breakpoint analysis was done using the breakpoint function in strucchange package (Zeileis et al. 2002). The same analysis for the control data showed no breakpoints.   Optimal (m+1)-segment partition: Call: breakpoints.formula(formula = littere ~ 1), where 'littere' included all litter data from the experimental (pulldown) plot only.

Breakpoints at observation number: Corresponding to breakdates:  
m = 1: 6   m = 1: 1996  
m = 2: 3, 9   m = 2: 1993, 1999  
m = 3: 3, 9, 12   m = 3: 1993, 1999, 2002  
m = 4: 3, 9, 12, 15   m = 4: 1993, 1999, 2002  
RSS 0.8620209   0.4165424  0.3098453   0.2720658   0.2458000  0.2313259
BIC -0.1351537 -8.6895142 -8.6163513 -5.2254757 -1.2645268  3.5131322
Confidence intervals for breakpoints of optimal 2-segment partition: 2.5% breakpoints 97.5%
Breakpoints at observation number: 3, 6, 8 Corresponding to breakdates: 1993, 1996, 1998

Literature Cited

Zeileis, A., F. Leish, K. Hornik and C. Kleiber. 2002. strucchange: an R package for testing for structural change in linear regression models. Journal of Statistical Software 7:1–38.


[Back to E094-035]