Solar Radiation Models and Gridded Databases to Fill Gaps in Weather Series and to Project Climate Change in Brazil

Table 3

Statistical evaluation of daily gridded databases for maximum air temperature (T_max), minimum air temperature (T_min), solar radiation (Q_g), rainfall (Rain), relative humidity (RH), and wind speed (WS_2m), considering 31 locations in Brazil.

Variable	Database	Indexes
		r^2	d	c	Classification	Bias (°C)	MAE (°C)
Tmax	AgCFSR	0.77	0.93	0.82	Very good	0.47	1.88
	AgMERRA	0.76	0.93	0.81	Very good	−0.37	1.86
	NASA/POWER	0.68	0.89	0.73	Good	−1.18	2.46
	XAVIER	1.00	1.00	1.00	Great	−0.03	0.17
		r^2	d	c	Classification	Bias (°C)	MAE (°C)
Tmin	AgCFSR	0.80	0.94	0.84	Very good	0.71	1.63
	AgMERRA	0.82	0.95	0.86	Great	0.08	1.45
	NASA/POWER	0.77	0.93	0.82	Very good	0.76	1.74
	XAVIER	0.98	0.99	0.98	Great	0.06	0.30
		r^2	d	c	Classification	Bias (MJ·m^−2·d^−1)	MAE (MJ·m^−2·d^−1)
Qg	AgCFSR	0.64	0.89	0.71	Good	−0.66	2.65
	AgMERRA	0.64	0.88	0.71	Good	−1.32	2.79
	NASA/POWER	0.76	0.93	0.81	Very good	−0.58	2.19
	XAVIER	0.94	0.97	0.94	Great	−1.19	1.57
		r^2	d	c	Classification	Bias (mm·d^−1)	MAE (mm·d^−1)
Rain	AgCFSR	0.16	0.60	0.24	Terrible	−0.03	4.88
	AgMERRA	0.25	0.67	0.33	Terrible	−0.03	4.48
	NASA/POWER	0.20	0.65	0.29	Terrible	0.12	4.86
	XAVIER	0.88	0.96	0.90	Great	−0.10	1.51
		r^2	d	c	Classification	Bias (%)	MAE (%)
RH	AgCFSR	0.52	0.79	0.57	Suffering	−2.43	14.43
	AgMERRA	0.38	0.72	0.45	Bad	−6.67	17.06
	NASA/POWER	0.55	0.82	0.61	Median	−6.75	11.14
	XAVIER	0.90	0.97	0.92	Great	0.18	3.76
		r^2	d	c	Classification	Bias (m·s^−1)	MAE (m·s^−1)
WS2m	AgCFSR	0.10	0.56	0.18	Terrible	0.17	0.83
	AgMERRA	0.08	0.53	0.15	Terrible	0.54	0.98
	NASA/POWER	0.14	0.69	0.21	Terrible	0.64	0.86
	XAVIER	0.47	0.79	0.54	Suffering	0.22	0.49