Biomass Modelling of Androstachys johnsonii Prain: A Comparison of Three Methods to Enforce Additivity

Table 5

Regression coefficients and goodness of fit statistics for the best linear and nonlinear models in (3) and (4), respectively.

Tree component	Weight function	(±SE)	(±SE)	(±SE)	Adj. (%)	(kg)	CV (%)	MRSE (kg)	MR; (kg)
Linear models
Roots		0.2522 (±0.6334)	0.0097 (±0.0002)^*		94.94	9.59	20.10	0.1520
Stem wood		0.6616 (±1.1451)	0.0251 (±0.0004)^*		97.49	19.66	15.84	0.0255
Stem bark		0.1895 (±0.3503)	0.0028 (±0.0001)^*		84.24	4.97	34.97	0.3397
Crown		−1.9106 (±1.5216)	0.0984 (±0.0044)^*		84.24	27.06	46.34	1.0292
Total tree		1.4066 (±2.1984)	0.0494 (±0.0008)^*		97.61	34.92	14.29	0.0247
Nonlinear models
Roots		0.0091 (±0.0024)^*	1.0074 (±0.0841)^*		94.94	10.40	21.79	0.1480
Stem wood		0.0197 (±0.0063)^*	1.8210 (±0.0567)^*	1.3081 (±0.1559)^*	97.71	18.83	15.18	0.0272
Stem bark		0.0022 (±0.0019)	1.7451 (±0.1564)^*	1.4084 (±0.4355)	84.84	5.70	40.11	0.3770
Crown		0.0350 (±0.0137)^*	2.1318 (±0.1385)^*	0.5290 (±0.1482)^*	84.42	26.89	46.05	0.5396
Total tree		0.0533 (±0.0093)^*	0.9920 (±0.0196)^*		97.60	38.68	15.83	0.1192

SE = standard error; “*” = significant at α ≥ 5%; “ ” = not significant at any probability level.