Estimating upper stem diameters and volume of Douglas-fir and Western hemlock trees in the Pacific northwest

Forest Ecosystems

Table 7 Mean prediction bias, root mean squared error (RMSE) in estimating inside bark cubic volume using different models. Statistics were obtained by using leave-one-out cross validation method i.e. bias and RMSE for dataset I were obtained by applying the models fitted using datasets II and III

Model	Dataset	Douglas-fir				Western hemlock
Model	Dataset	Bias (cm)	Bias percent	RMSE (cm)	RMSE percent	Bias (cm)	Bias percent	RMSE (cm)	RMSE percent
M1	I	0.22	12.56	0.35	20.36	−0.06	−4.29	0.58	42.51
	II	0.02	11.03	0.04	20.51
	III	−0.16	−6.09	0.85	31.55	0.34	15.98	0.45	21.19
M2	I	0.13	7.21	0.31	17.77	−0.16	−11.63	0.35	25.99
	II	−0.01	−5.01	0.03	16.27
	III	−0.06	−2.07	0.36	13.52	0.23	10.80	0.42	19.67
M3	I	−0.04	−2.31	0.29	16.79	−0.17	−12.26	0.43	31.66
	II	−0.02	−9.20	0.04	18.82
	III	−0.15	−5.51	0.43	15.96	0.23	10.70	0.42	19.75
M4	I	0.05	2.75	0.24	13.95	−0.18	−13.16	0.46	33.65
	II	−0.01	−4.28	0.03	14.64
	III	−0.13	−4.89	0.47	17.52	0.22	10.40	0.40	18.83
M5	I	0.07	4.24	0.29	16.45	−0.13	−9.41	0.61	45.14
	II	0.00	−2.07	0.03	15.94
	III	−0.44	−16.13	1.18	43.91	0.16	7.41	0.28	13.33