Table 3. Results of separate linear mixed models testing effects of independent variables on dependent variables using mixed models in SAS (PROC MIXED; SAS Institute, Inc., Cary, NC) with survey period as the repeated measure from winter (mid-December 2008–February 2009) or early winter (mid-December 2008–early January 2009) in eastern Mississippi, western Tennessee, and the Mississippi Alluvial Valley, USA.

Independent Variables			Dependent Variables
			Millet				Natural Seeds and Tubers				Duck Energy Days				Energy Removal
			F a	DFb b	DFw c	P	F a	DFb b	DFw c	P	F a	DFb b	DFw c	P	F a	DFb b	DFw c	P
Winter	Period		0.22	2	22	0.807	0.38	2	22	0.685					20.02	2	21.5	<0.001
		Dec × Jan													5.72		21.2	<0.001
		Dec × Feb													5.19		21.7	0.001
		Jan × Feb													-0.49		21.7	0.879
	Site		2.12	3	6	0.199	10.27	3	6	0.009
	Treatment		1.80	2	6	0.244	0.18	2	6	0.841	0.04	2	29	0.957	47.29	2	7.37	<0.001
		low × med													-3.81		7.35	0.015
		low × high													-9.66		7.37	<0.001
		med × high													-5.88		7.38	0.001
	Previous Period d										0.05	1	29.3	0.833
	EE e														1.57	1	8.5	0.243
Early Winter	Site		1.46	3	5	0.332	15.35	3	5	0.006	34.95	3	6	<0.001
	Treatment		0.76	2	5	0.514	0.24	2	5	0.792	0.37	2	6	0.701	86.4	2	8	<0.001
		low × med													-5.88		8	<0.001
		low × high													-13.1		8	<0.001
		med × high													-8.49		8	<0.001
	EE														2.27	1	8	0.17
	Water Depth		0.02	1	5	0.902	1.44	1	5	0.284

^a F-statistic from linear mixed models, or T-statistic from Tukey’s post-hoc comparison of means.

^b Degrees of freedom between groups.

^c Degrees of freedom within groups.

^d Sampling periods occurred in mid-December, early January, early February, and late February

^e Apparent Existence Energy