Estimation of Summer Air Temperature over China Using Himawari-8 AHI and Numerical Weather Prediction Data

Table 4

Statistical results for T_air estimation by linear and NN models.

Model	Predictive factors	Linear model			NN model
		R	RMSE (°C)	Bias (°C)	R	RMSE (°C)	Bias (°C)
1	BT11μm	0.75	5.53	0.07	0.77	4.46	0.06
2	BT11μm, PWV	0.85	4.79	−0.05	0.89	3.20	0.01
3	BT11μm, PWV, RH	0.84	4.78	−0.06	0.95	2.21	0.09
4	BT11μm, PWV, RH, EL	0.84	4.50	−0.03	0.95	2.15	0.12
5	BT11μm, PWV, RH, EL, BT12μm	0.84	4.36	−0.05	0.95	2.09	0.10
6	BT11μm, PWV, RH, EL, BT12μm, LAT	0.84	4.27	−0.05	0.96	2.06	0.13
7	BT11μm, PWV, RH, EL, BT12μm, LAT, hour	0.84	4.25	−0.07	0.97	1.81	0.06
8	BT11μm, PWV, RH, EL, BT12μm, LAT, hour, LON	0.85	4.03	−0.06	0.97	1.76	0.04
9	BT11μm, PWV, RH, EL, BT12μm, LAT, hour, LON, VZA	0.89	3.28	0.07	0.97	1.72	0.04