Longleaf pine proximity effects on air temperatures and hardwood top-kill from prescribed fire

Fire Ecology

Table 5 Comparison of models of hardwood top-kill probability after prescribed burns in upland longleaf pine forest in southwest Georgia, USA, in May 2016. Models express hardwood top-kill probability as Michaelis-Menten (MM) or logistic (LG) functions of density of neighboring pine trees (neighborhood density index: NDI), hardwood height (Ht), or both. Hardwood species are treated as a group or separated by guild (denoted by subscripted g) representing fleshy-fruited trees, mesic oaks, upland oaks, and xeric oaks. Bold font shows the best model (i.e., with lowest AICc), in which differences among hardwood guilds are expressed by an exponent that modifies height

Model	K^a	L	R²	AICc	Δ_i
Intercept-only	1	−1049.68	0.00	2101.36	865.69
MM = f (NDI)	3	−861.73	0.23	1729.47	493.80
LG = f (NDI)	3	−816.15	0.28	1638.32	402.69
MM = f (Ht)	3	−761.84	0.34	1529.68	294.01
LG = f (Ht)	3	−771.29	0.33	1548.60	312.93
MM = f (NDI, Ht)	4	−686.47	0.43	1380.95	145.28
LG = f (NDI, Ht)	4	−695.94	0.42	1399.89	164.22
MM = f (NDI, Ht _g )	7	−610.82	0.50	1235.67	0.00
LG = f (NDI, Ht_g)	7	−646.46	0.47	1306.95	71.28
MM_g = f (NDI, Ht_g)	13	−607.82	0.50	1241.76	6.09
LG_g = f (NDI, Ht_g)	13	−621.07	0.49	1268.27	32.60

^aK = number of parameters, L = log-likelihood, R² = Nagelkerke’s R² for discrete models, AICc = sample-size-corrected Akaike’s Information Criterion, and Δ_i indicates strength of support from the data in comparison to the best model (Δ_i < 2 = substantial support, 4 < Δ_i < 7 = considerably less support, and Δ_i > 10 = essentially no support; Burnham and Anderson 2002)