Annals of Forest Science

Table 6 Comparison of area-based approach and universal kriging approach for stem density (N), basal area (BA), and quadratic mean diameter (QMD), in terms of Pearson coefficient between estimated and observed values, RMSE, R², and bias

From: Integration of field sampling and LiDAR data in forest inventories: comparison of area-based approach and (lognormal) universal kriging

Response variables	Area-based approach				Universal kriging
Response variables	Bias (%)	Pearson coef.	RMSE (%)	R² adj	Bias (%)	Pearson coef.	RMSE (%)	R² adj
FMU 1
N	1.03	0.53	51	0.28	− 1.64	0.53	50	0.27
BA	0.31	0.66	25	0.42	1.90	0.72	23	0.51
QMD	− 0.03	0.66	22	0.49	− 0.04	0.66	22	0.43
FMU 2
N	0.14	0.60	39	0.33	− 0.01	0.59	38	0.33
BA	0.07	0.84	19	0.69	0.07	0.84	18	0.70
QMD	0.09	0.76	19	0.57	− 0.25	0.86	15	0.73
FMU 3
N	− 1.27	0.71	43	0.49	0.53	0.71	43	0.49
BA	0.41	0.80	20	0.64	1.49	0.81	20	0.64
QMD	− 0.04	0.68	20	0.44	− 0.07	0.69	20	0.47
FMU 4
N	− 0.41	0.43	37	0.15	− 0.03	0.50	35	0.22
BA	0.83	0.77	24	0.57	0.06	0.81	22	0.65
QMD	0.35	0.37	23	0.11	0.46	0.56	20	0.29

Back to article page

ISSN: 1297-966X

Contact us

Submission enquiries: annforsci@inrae.fr
General enquiries: info@biomedcentral.com