Riemer et al.: Food habits of Eumetopias jubatus off Oregon and northern California, 1986-2007 
377 
Table 3 
Comparison of Steller sea lion (Eumetopias jubatus) diet composition by month, after controlling for collection site and year. 
Sample size (n=number of scat analyzed) and individual number of prey types (D) are given for each diet; pooled number of unique 
prey types (D ), correlation coefficient ( R M ), and permutation-based P-value (based on 9999 replications) are given for each com- 
parison. * indicates significance at the a= 0.002 level (based on Bonferroni adjustment of a=0.05 for 26 multiple comparisons). 
Site 
Year 
Diet 1 
Diet 2 
D e 
Mantel test 
Month 
n 
D 
Month 
n 
D 
Rm 
P value 
Columbia R. 
2004 
June 
53 
25 
August 
37 
13 
27 
0.061 
0.0017* 
June 
53 
25 
September 
43 
17 
28 
0.145 
0.0001* 
August 
37 
13 
September 
43 
17 
19 
0.047 
0.0105 
Rogue Reef 
2001 
April 
45 
32 
August 
48 
32 
44 
0.089 
0.0001* 
2002 
March 
39 
27 
April 
49 
30 
36 
0.264 
0.0001* 
March 
39 
27 
July 
33 
27 
35 
0.409 
0.0001* 
March 
39 
27 
August 
37 
35 
38 
0.251 
0.0001* 
April 
49 
30 
July 
33 
27 
36 
0.173 
0.0001* 
April 
49 
30 
August 
37 
35 
39 
0.147 
0.0001* 
July 
33 
27 
August 
37 
35 
40 
0.117 
0.0002* 
2003 
April 
57 
29 
October 
53 
31 
42 
0.101 
0.0001* 
Table 4 
Comparison of Steller sea lion ( Eumetopias jubatus) diet composition by year, after controlling for collection site and month. 
Sample size (n) and individual number of prey types (D) are given for each diet; pooled number of unique prey types (D p ), cor- 
relation coefficient (R M ), and permutation-based P-value (based on 9999 replications) are given for each comparison. * indicates 
significance at the a = 0.002 level (based on Bonferroni adjustment of a=0.05 for 26 multiple comparisons). 
Site 
Month 
Diet 1 
Diet 2 
D p 
Mantel test 
Year 
n 
D 
Year 
n 
D 
P value 
Columbia R. 
August 
2004 
37 
13 
2007 
31 
16 
20 
0.105 
0.0004* 
Rogue Reef 
April 
2001 
45 
32 
2002 
49 
30 
40 
0.109 
0.0001* 
2001 
45 
32 
2003 
57 
29 
40 
0.018 
0.0682 
2002 
49 
30 
2003 
57 
29 
38 
0.126 
0.0001* 
June 
1987 
34 
14 
1993 
36 
29 
32 
0.222 
0.0001* 
July 
1990 
43 
16 
2002 
33 
27 
29 
0.199 
0.0001* 
1990 
43 
16 
2004 
33 
20 
26 
0.216 
0.0001* 
2002 
33 
27 
2004 
33 
20 
29 
0.097 
0.0001* 
August 
2001 
48 
32 
2002 
37 
35 
41 
0.123 
0.0001* 
St. George Reef 
July 
2002 
33 
13 
2004 
33 
21 
22 
0.008 
0.2515 
2002 
33 
13 
2006 
33 
21 
25 
0.122 
0.0001* 
2004 
33 
21 
2006 
33 
21 
29 
0.203 
0.0001* 
interpretable parameters. Nonetheless, it does indicate 
that researchers should be cautious about pooling sam- 
ples across space and time before investigating whether 
those samples differ. 
Athough analysis of pinniped fecal matter is a stan- 
dard technique for studying diet (e.g.. Pitcher, 1980; 
Beach et al. 3 ; Olesiuk et al., 1990; Orr et al., 2004), 
there are some limitations. For example, the use of oto- 
liths to identify prey can lead to biased diet composition 
estimates (Jobling and Breiby, 1986). We minimized 
this problem by including all bony skeletal structures 
(vertebrae, gillrakers, etc.) to identify prey. Another 
potential bias can occur when drawing inference to a 
particular population from opportunistically collected 
3 Beach, R. J., A.C. Greiger, S. J. Jeffries, S. D. Treacy, and 
B. L. Troutman. 1985. Marine mammals and their inter- 
actions with fisheries of the Columbia River and adjacent 
waters, 1980-1982: third annual report, March 1, 1980 to 
October 31, 1982. National Marine Mammal Laboratory, 
Northwest and Alaska Fisheries Center, NMFS, NOAA Proc. 
Rep. 85-03, 316 p. [Available at www.lib.noaa.gov, accessed 
May 2011.] 
