Appendix E. ANOVA tables from analysis of barnacle recruitment in 2002, 2003, and 2004.
2002
TABLE E1. ANOVA table of log x + 1 transformed barnacle recruitment in 2002. The ANOVA model explained 61.7% of the variation in the data (240 df total). Planned contrasts comparing the recruitment of barnacles among upwelling groups indicated that recruitment was higher in the strong (SUP) than the weak (WUP) upwelling sites (SUP vs. WUP, F ratio = 113.77, P < 0.0001, and that recruitment in the intermediate upwelling group (IUP) was higher than in the WUP group (IUP vs. WUP, F ratio = 58.09, P < 0.0001). All contrasts with 1, 407 df. 
SS 
df 
F ratio 
P 

Depth (D) 
7.751 
1 
82.163 
< 0.0001 
Upwelling (U) 
11.457 
2 
60.721 
< 0.0001 
Site[U] 
8.18 
9 
9.645 
< 0.0001 
D × U 
0.155 
2 
0.824 
0.4396 
D × Site[U] 
4.677 
9 
5.508 
< 0.0001 
Residual 
20.378 
216 
2003
TABLE E2. ANOVA of log x + 1 transformed barnacle recruitment in 2003. The ANOVA model explained 81.4% of the variation in the data (551 df total). Planned contrasts comparing the recruitment of barnacles among upwelling groups indicated that recruitment was higher in the strong upwelling group (SUP) than either the intermediate (IUP) or weak (WUP) upwelling groups (SUP vs. IUP, F ratio = 168.17, P < 0.0001; SUP vs. WUP, F ratio = 1385.39, P < 0.0001) and that recruitment in the IUP group was higher than in the WUP group (IUP vs. WUP, F ratio = 416.57, P < 0.0001). All contrasts with 1, 407 df. 
SS 
df 
F ratio 
P 

Treatment (T) 
0.0014 
2 
4.230 
0.0152 
Depth (D) 
0.0620 
1 
371.789 
< 0.0001 
Upwelling (U) 
0.2327 
2 
697.459 
< 0.0001 
Site[U] 
0.0320 
9 
21.326 
< 0.0001 
T × D 
0.0002 
2 
0.816 
0.4429 
U × T 
0.0001 
4 
0.180 
0.9484 
U × D 
0.0360 
2 
107.931 
< 0.0001 
U × D × T 
0.0007 
4 
1.152 
0.3316 
T × Site[U] 
0.0446 
18 
1.485 
0.0908 
D × Site[U] 
0.0151 
9 
10.100 
< 0.0001 
T × D × Site[U] 
0.0034 
18 
1.1589 
0.2928 
Location [TDSU] 
0.0299 
72 
2.4901 
< 0.0001 
Residual 
0.0679 
407 
2004
TABLE E3. ANOVA of log x + 1 transformed barnacle recruitment in 2004. The ANOVA model explained 85.7% of the variation in the data (475 df total). The predation treatment had 2 levels: fenced and open plates. Planned contrasts comparing barnacle recruitment of among upwelling groups indicated that recruitment was higher in the strong upwelling group (SUP) than either the IUP or weak WUP upwelling groups (SUP vs. IUP, F ratio = 186.71, P < 0.0001; SUP vs. WUP, F ratio = 1077.21, P < 0.0001) and that recruitment in the IUP group was higher the WUP group (IUP vs. WUP, F ratio = 253.41, P < 0.0001). All contrasts with 1, 379 df. 
SS 
df 
F ratio 
P 

Treatment (T) 
0.002 
1 
0.875 
0.3501 
Depth (D) 
0.083 
1 
256.533 
< 0.0001 
Upwelling (U) 
0.350 
2 
538.607 
< 0.0001 
Site[U] 
0.017 
9 
61.429 
< 0.0001 
T × D 
0.001 
1 
3.363 
0.0674 
U × T 
0.000 
2 
0.808 
0.4460 
U × D 
0.009 
2 
14.643 
< 0.0001 
U × D × T 
0.002 
2 
4.435 
0.0125 
T × Site[U] 
0.003 
9 
1.178 
0.3072 
D × Site[U] 
0.005 
9 
18.640 
< 0.0001 
T × D × Site[U] 
0.008 
9 
2.740 
0.0041 
Location [TDSU] 
0.005 
48 
3.220 
< 0.0001 
Residual 
0.1232 
379 