Table 4. Top models (Δ AIC_c < 8) predicting the average edge weight in seasonal networks of female white-tailed deer (Odocoileus virginianus) association rate^a in Lake Shelbyville, Illinois (2007–2009).

Seasonb	Modelc	k	AICc	Δ AICc	Weight	R2 d
Gestation	HR overlap [+]	3	-15.27	0	0.60	0.68
	Ag (prop) + HR overlap	4	-11.29	3.99	0.08	0.70
	Ag (conn) + HR overlap	4	-10.89	4.39	0.07	0.69
	Forest (conn) + HR overlap	4	-10.70	4.58	0.06	0.68
	Forest (prop) + HR overlap	4	-10.66	4.61	0.06	0.68
	Edge (prop) + HR overlap	4	-10.57	4.70	0.06	0.68
	Edge (conn) + HR overlap	4	-10.56	4.71	0.06	0.68
Fawning	Null	2	-3.88	0	0.41
	HR overlap [+]	3	-2.45	1.43	0.20	0.25
	Ag (conn)	3	0.09	3.97	0.06	0.03
	Ag (prop)	3	0.25	4.13	0.05	0.02
	Edge (prop)	3	0.30	4.18	0.05	0.01
	Forest (prop)	3	0.30	4.18	0.05	0.01
	Edge (conn)	3	0.37	4.25	0.05	3.5 x10-3
	Forest (conn)	3	0.40	4.28	0.05	8.5 x10-4
	Forest (prop) + HR overlap	4	2.04	5.92	0.02	0.35
	Ag (prop) + HR overlap	4	2.66	6.53	0.02	0.31
	Forest (conn) + HR overlap	4	2.67	6.55	0.02	0.31
	Edge (conn) + HR overlap	4	2.70	6.58	0.02	0.31
	Ag (conn) + HR overlap	4	2.94	6.82	0.01	0.29
	Edge (prop) + HR overlap	4	3.09	6.97	0.01	0.28
Rut	Ag (prop) [+]	3	-4.98	0	0.40	0.82
	Null	2	-4.63	0.36	0.33
	Ag (conn) [+]	3	-3.86	1.12	0.23	0.78
	Edge (conn)	3	1.55	6.54	0.02	0.47
	Edge (prop)	3	2.85	7.84	0.01	0.34
	Forest (conn)	3	3.02	8.00	0.01	0.32

Models with Δ AIC_c < 2 are in bold font.

^a We weighted network edges by association rate; the number of times two deer were within 25m of one another at the same time divided by the total number of simultaneous locations.

^b Gestation (1 Jan– 14 May; n = 12), fawning (15 May– 31 Aug; n = 10), rut (1 Sep– 31 Dec; n = 6).

^c [+] and [–] indicate the direction of each variable’s effect. Variables are described in the footnote of Table 3.

^d All models with Δ AICc <2 had F-statistics >95% of randomized values except the home range overlap model during fawning.