Estimation of leaf traits from reflectance measurements: comparison between methods based on vegetation indices and several versions of the PROSPECT model

Table 5 Comparison between destructive measurements of \(C_{\text{abc}}\) and \(C_{\text{w}}\) and PROSPECT or vegetation indices estimates

Variables	Metrics	PROSPECT				VIs
Variables	Metrics	P3b	P4b	P5b	PDb	Dx4	CIre	SRw	NDw
\(C_{\text{abc}}\) (µg/cm²)	ρ	0.81	0.80	0.82	0.80	0.80	0.78	–	–
	r ²	0.81	0.79	0.80	0.80	0.80	0.77	–	–
	RMSE Corr	6.54	6.72	6.70	6.71	6.63	7.08	–	–
\(C_{\text{w}}\) (mg/cm²)	ρ	0.93	0.93	0.92	0.91	–	–	0.89	0.94
	r ²	0.87	0.85	0.86	0.85	–	–	0.80	0.88
	RMSE Corr	1.14	1.08	1.05	1.11	–	–	1.28	1.29

The estimation performances from the four PROSPECT versions (including brown pigments) and vegetation indices against destructive measurements: spearman correlation coefficient (ρ), squared Pearson correlation coefficient (r²) and RMSE Corr as provided in Table 4. RMSE Corr for VIs was computed from the fitted empirical model between the biochemical contents and the VIs: linear functions for Dx4, CIre and SRw, a second order polynomial function for NDw. The numbers in italic indicate the best result for each biochemical content

ISSN: 1746-4811